1.1- Background

Chapter One - Introduction

2

1.1- Background In our previous work (1) different phases of (vanadium- phosphorus-

oxygen) (VPO) catalyst were prepared, characterized and evaluated by

catalytic test in order to produce Maleic anhydride (MA) via selective

oxidation of n-butane.

Maleic anhydride (MA), and its derivatives malic acid and fumaric

acid, are produced with a worldwide capacity of about (2.8 X 10 6) metric

tons per year. While the market demand for (MA), increased from

(0.616 X 10 6) metric tons in 2000 to almost (1.27 X 10 6) metric tons in

2004(2-4).

These multifunctional chemical intermediates have found applications

in almost any field of industrial chemistry. The principal use of (MA) is in

manufacturing of unsaturated polyester resins (UPR) ( 63 %), lubricating

oil additives (10 %) , copolymers (9 %) ,alkenyl succinic anhydride (5 %) ,

malic acid (3 %) ,fumaric acid (2 %) agricultural chemicals (1%),

miscellaneous, including reactive plasticizers , sulfosuccinic acid esters ,

and alkyd resins (% 7). Furthermore, due to its double bond and anhydride

function, (MA) is a versatile intermediate for the production of co-

polymers of (MA), for example, ethylene glycol and vinyl monomer.

Recently, potential new uses of (MA) have been found in its conversion to

(1-4) butanediol and the manufacturing of tetrahydrofuran (THF) and

butyrolactone via hydrogenation (5).

Maleic anhydride (MA) and the two di-acid isomers were first prepared

in the 1830's, but it took about 100 years before commercial manufacture

was performed in 1933. The National Aniline and Chemical Company Inc.

started a process for the production of (MA) based on benzene oxidation

using a vanadium oxide catalyst. Smaller amounts of maleic acid were also

formed as by product in the production of phthalic anhydride. The use of


3

benzene as a feedstock for the production of (MA) was dominant in the

world until the late 1980's (6).

Currently, worldwide production of (MA) is based on the major feed

stocks benzene, butenes and n-butane (1-3,6-14). Most of the capacity is

produced via fixed–bed oxidation of benzene, though benzene is being

displaced by butane as a feed stock (all production in United States is

butane based) (1-3,6-14), because butane is a lower cost and environmentally

more desirable raw material and because butane oxidation produces a

clean product stream, forming mainly (MA) and carbon oxides as shown

below:

C4H10 + 3.5 O2 C4H2O3 + 4 H2O ∆ H = - 1236 KJ/ mole (1.1)

C4H10 + 4.5 O2 4 CO + 5 H2O ∆ H = - 1521 KJ/ mole (1.2)

C4H10 + 6.5 O2 4 CO2 + 5 H2O ∆ H = - 2656 KJ/ mole (1.3)

It is obvious that CO and CO2 are thermodynamically more favored

products. Only kinetic control by a catalyst will enhance the formation of

(MA). In practice, the process is operating at a yield of approximately

(60%) to the desired product. CO and CO2 are the sole carbon containing

by–products in a ratio of about unity. Suppression of the unselective and

very exothermic oxidation to carbon oxides requires sufficient heat transfer

capacities of the reactor. Nonetheless, hot spots are frequently met in (MA)

production plants.

Processes for the production of MA From n-Butane

In general, three different types of process can be distinguished in

commercial production of (MA) from n-butane; fixed-bed processes


4

(Scheme 1.1), fluidized–bed processes (Scheme 1.2) and the re-circulating-

solids process (Scheme 1.3).

The fixed-bed reactor consists of a number of tubes that are packed

with coarse catalyst bodies. The reactants flow through these tubes. As a

result of the obstruction of the gas flow by the catalyst bodies, a pressure

drop across the bed is exhibited. Therefore, pressure has to be applied at

the inlet to ensure an adequate flow rate. The magnitude of the pressure

drop is depending on the flow rate, the length of the catalyst bed and the

size to the catalyst bodies. Since the selective oxidation of n-butane to

(MA) is highly exothermic, fixed-bed reactors must containing a facility to

remove the reaction heat. This can be done in various ways: the bed can be

split into different sections, with provision for cooling the gas between the

sections ,or using a reactor containing a large number of tubes, along

which a cooling gas or the liquid is recirculated. However, hot spots can

occur easily in fixed-bed reactors. These can be prevented by using larger

catalyst bodies, a less active catalyst, or by dilution of the catalyst with an

inert solid (support). In view of the explosion limits and the flammability

of mixtures of n-butane and air, only low concentrations of n- butane can

be applied (2 - 4 %).Furthermore, the gases must be mixed and pre-heated

before entering the reaction zone. In a fixed-bed reactor ,the concentration

of n-butane will decrease when moving to the end of the tube .To maintain

a sufficiently high selectivity at the exit of the reactor a less active catalyst

is installed at the entrance and a very active catalyst at the end of the

reactor (4,5).

A single passed fixed - bed reactor was used in our previous work (1) ,

(1 m long and 0.019 m in diameter) mounted in four heating zones vertical

furnace. The flow of n-butane was controlled before entering the reactor

using gasometer. The total gas flow after leaving the reactor was also

measured and fixed at industrial conditions (space velocity 2000 h -1).The


5

gas mixture (1.5 Vol. % n-butane in air ) was preheated before passing on

the catalyst using inert alumina granules in the upper half of the reactor .

The lower half of the reactor (0.1 L) was filled with (2.3 mm) granules by

(4) temperature indicators controllers (TIC).Maleic produced was passed

through scrubber and was dissolved in water to produce Maleic acid .this

acid was titrated with (0.01 N) NaOH using phenolphthalein as indicator.

Scheme 1.1: Huntsman fixed-bed reactor for MA production (4).

In a fluidized-bed process, reaction gases flow upward through a bed

of catalyst particles .When the force of the gas flow on the catalyst bed is

equal to the weight of the bed, the catalyst bed expands significantly and

the catalyst bodies are brought in continuous motion. Because of this

motion ,better heat transfer characteristics are established and ,hence, hot

spots cannot occur in a fluidized-bed reactor comprise the fact that reaction

gases can be used without pre-mixing and pre-heating before entering the

reactor. Furthermore, higher n-butane concentrations can be used due to a

decreased explosion risk compared to the fixed-bed process (4). The flow

diagram for fluidized-bed (Scheme 1.2) process is comparable to that of

the fixed-bed process (Scheme 1.1).Several different ways have been


6

developed to produce bulk (VPO) catalyst particles of the right size and the

attrition resistance for fluidized-bed purposes (16) .When the catalyst bodies

are too small, they will be blown out of the reactor .Large bodies, on the

other hand, call for extremely high linear gas flow rates in the reactor,

therefore, the particle size of fluidized-bed catalysts usually ranges from

(10-150 microns). Development of fluid-bed butane system has been

reported (17,18).These systems offer the advantage of operation at higher

butane concentration (up to 4%) and thus lower original costs. However,

less selectivity at the higher butane loadings and the intermediate problems

with attrition of (VPO) materials need to be overcome before this process

becomes economically preferred (17,18).

Scheme 1.2: ALMA Fluidized bed reactor for MA (4).

DuPont commercially operates the third type of process in their

recirculation-solids reactor (17,18).In this case, oxidation of n-butane and

regeneration of the catalyst are carried out in two separate reaction zones

(Scheme 1.3) .The selectivity to (MA) is increased, because the oxidation

of n-butane is carried out in absence of oxygen (19,20) . In the first step, n-

butane reacts with lattice oxygen from the catalyst. In this stage, the


7

catalyst is reduced by n-butane resulting in the selective formation of MA,

which is removed in a stripper. Regeneration of the reduced catalyst with

oxygen takes place in the second reactor zone. In principle, the oxidation

state of the catalyst can be controlled optimally in this way (17,18).This

process operates at lower conversion per pass and with higher selectivities

than normally encountered in fixed or fluid bed systems. It also uses a

unique attrition resistance system (19-21).

Scheme 1.3: DUPONT re-circulating – solids process for the production of THF

from n-butane via MA (4).

Table 1.1 shows the 1993 and 1995 world-wide (MA) production

capacity for the different processes (4).As can be seen from the table, both

fixed-bed and fluidized-bed butane based processes have been growing at

the expense of benzene-based processes. Furthermore, it seems that

production of (MA) with fluidized –bed technology would not surpass

production with fixed-bed technology which was still operated at higher

yields to (MA). Table 1.2 represents the long – term (MA) production –

consumption balance (4) .The average of the annual demand growth which

was estimated to be about (4%). By the year 2004, total world –wide

consumption will be increased by about (76.5%) as compared to 1998.


8

Table 1.1: World MA capacity (in metric tons) by reactor type (4).

Reactor (feed) 1993 ١٩٩٥

10 3 t/y % 10 3 t/y %

Fixed - bed

n-Butane 369 42.0 704 51.8

Benzene 325 37.9 388 28.5

Fluidized – bed

n-Butane 127 14.8 217 16.0

Phthalic anhydride co-product 37 4.3 50 3.7

Total 858 100.0 1359 100.0

Table 1.2: Total world production & consumption of MA (X 103 metric tons) (4).

Capacity to produce 1994 1995 1996 1997 1998 2001 2004

From n-Butane 551.3 647.6 657.1 702.6 767.6 1019.2 1088.7

From Benzene 301.0 302.0 300.0 300.0 311.0 289.5 289.5

From PA recovery 23.5 20.5 22.5 24.5 24.5 28.5 28.5

Total 875.8 970.1 979.6 1027.1 1103.1 1337.2 1406.7

Utilization (%) 81.9 79.4 82.1 83.4 83.8 79.6 85.9

Production 717.1 770.4 803.9 856.9 924.8 1064.1 1207.7

Consumption

UPE resins 355.3 372.3 383.4 399.7 416.7 471.0 532.9

BDO Chemicals 27.0 36.1 10.0 54.2 86.8 120.4 145.4

Other Uses 330.2 361.2 380.5 402.5 421.3 472.7 529.4

Total 712.5 769.6 803.9 856.4 924.8 1064.1 1207.7

All catalysts used industrially for the production of (MA) from

n-butane are (VPO) catalyst based system (8,12,22).The numbers in (Tables

1.1 and 1.2) ,together with the fact that the current yield (only amounts to

about 58%) to (MA) of an equilibrated catalyst, indicate that improvement

of the process is of great economic and environmental interest. To this end,


9

several developments can be considered. First, the bulk (VPO) catalyst

should exhibit a higher attrition resistance (mechanical strength) in order to

be more suitable for fluidized-bed process. Secondly, the activation period

for the catalyst should be shortened. This will result in an earlier

achievement of optimum performance. However, the catalyst formulation

could be changed; resulting in better properties and an improved catalytic

performance. Furthermore, there is also a comprehensive demand for a

cheaper and more reproducible preparation procedure for the currently

applied (VPO) catalyst. In any case, these improvements can never be

achieved without thoroughly investigations of the catalytic and structural

properties of the active (VPO) phase.

1.2- Scope of the Literature Survey

The n-butane to (MA) reaction is a fascinating complex system. This

catalytic system performance a (14-electron) oxidation involving the

abstraction of (8 hydrogen atoms) and insertion of (3 oxygen atoms) as

described in (eq. 1.4) (21).

CH3

CH3

+ 3.5 O2

O

O

O + 4 H2O ... (1.4)

Compared to other industrially practiced hydrocarbon selective

oxidation reactions (23-26), it is the most complex one (Table 1.3). It is the

only example of an industrially practiced selective oxidation reaction

involving alkane activation. Knowledge gained through study and

understanding of this system may contribute to advances in alkane

activation in general.


10

Table 1.3: Number of Electrons and Oxygen Molecular Involved in some Principal

Reactions of Industrial Interest in Selective Oxidation (9).

Reaction Electrons

Involved

Moles

Oxygen

+ 0.5 H2O

2 0.5

O + H2O

4 1

C

O

H+ H2O

4 1

O

+ 3 H2O

O

O

12 3

One characteristic of the selective oxidation of n-butane involves series

of oxidation steps (utilizing) different kinds and reactivates of oxygen (27)

as described in (Scheme 1.4). This scheme was built up according to

experimental work which determined the form of selective and non-

selective oxygen in the reaction of n-butane oxidation.

O O

M M

(a)

M

O-

(b)

Split double bond


11

O M O

O

O

O O

O M O

(c) (d) (e)

O M O

OO

Radical Pi adsorbed

Scheme 1.4: Kinds of oxygen adsorbed species (28)

1.3 -Structure of the catalyst

1.3.1- Crystalline (VPO) phases

The (VPO) system is characterized by the facile formation of a number

of crystalline phases of vanadium (III), vanadium (IV) and of vanadium

(V).In the more interesting (P/V) ratio near (1.0), the different crystalline

(α, β) (29-35), polymorphic (α1, α11, γ, δ) ( 29,35-37) or hydrated phases of

(VOPO4) (34-41) have been extensively studied. In general, the vanadyl

orthophosphate crystal structure consists of (VO6) and (PO4) groups

arranged in layers [VOPO4] ∞, held together by long (V-O) bonds or by

hydrogen bonds (Figure 1.1-A). The layered structure leads to rich

intercalation chemistry, with formation of layered solids consisting of

alternating inorganic (42-48) and organic (49-56) layer or the formation of

solvated inorganic intercalation compounds (43).


12

Figure 1.1-A: Structure of VOPO4.2H2O showing infinite layers of PO4

tetrahedral linked to VO6 octahedral (40).

In dihydrated (VOPO4) the layer lattice is built up of neutral (VOPO4)

layers and interlayer water molecules (40, 41) .The vanadium atom lies on a

fourfold axis and is surrounded by six oxygen atoms to give distorted

octahedral .The four equatorial oxygens are provided by four different

phosphate tetrahedral .One of the axial vanadium–oxygen bond is very

short corresponding to a double bond (V=O) as shown in Figure 1.1-B.

Figure 1.1-b: Structure of VOPO4.2H2O showing bond distances (°A) and angles

(degree) (40).


13

The structure of (α-VOPO4) is composed of chains of highly distorted

(VO6) octahedral sharing four oxygen atoms with four different (PO4)

groups (30-35). These groups are arranged to form layers (Figure1.2-a).

α- VOPO4 β- VOPO4

Figure 1.2: Structure of α and β- VOPO4 (34).

A distortion of the (VO6) octahedral occurs along the (c) axis,

generating a short (V=O) bond and a very long (V-O) bond (V----O). Thus

the oxyvanadium units can be approximated as (VO5) pyramids and (PO4)

tetrahedral in its crystal structure, analogous to α-VOPO4 (Figure 1.3).

However, the primitive unit cell of (β-VOPO4) contains twice the number

of such structural groups (Figs.1.2-b&1.3), resulting in a network structure.

The γ–and δ -polymorphic forms of (β-VOPO4) are suggested (36-38) to

contain a different framework, in which pairs of edge-sharing octahedral


14

with trans vanadyl oxygens are alternately unshared or shared with

phosphate tetrahedral. Such pairs do not exist in the (α- and β-VOPO4)

forms. During thermal treatment at high temperature the consecutive

transition has been observed (36).

δ- VOPO4 γ-VOPO4 β-VOPO4 …. 1.5

α-VOPO4

β-VOPO4

Figure 1.3: Comparison between structure of α and β-VOPO4 (34).


15

Vanadium (IV) hydrogen phosphate (57-59) which is the most widely

used precursor of the active phase of n-butane oxidation possesses (21,57-62) a

structure made up of atom arrangements in the (ac) plane which are very

similar to that proposed for γ- VOPO4 (Figure 1.4 and 1.5) .It may be

obtained from the dihydrated V(V) phosphate by reduction with organic

alcohols (57) or from the reduction of V2O5 followed by the addition of

o-H3PO4 (61,62).

(A)

(010) plane

(B)

(100) plane

Figure 1.4: Crystal structure of γ -VOPO4 (36).


16

Figure 1.5: Crystal structure of VOHPO4.0.5 H2O (010) plane (59).

Both the hemi hydrate and tetrahydrate of vanadium hydrogen

phosphate are known (58) and their structures are closely related, showing a

layered structure and pairs of face sharing vanadium (IV) octahedral

hydrogen phosphate (OH) groups are directed into the interlayer space.

The coordination geometry around both vanadium and phosphorus is

similar in the two structures (Figure 1.5 and 1.6) .The possible formation

of the tri-hydrate VOHPO4 has also been reported (63).

Figure 1.6: View of the infinite double chains of (VO6) octahedra and (PO4)

tetrhedra running along the (100) direction in VOHPO4.4H2O (75).


17

The intercalation compounds of vanadium hydrogen phosphate with

organic molecules or inorganic cations and anions such as (V n+) of HPO42-

are reported (23-33,36,37,63-72) .This intercalation chemistry is of particular

importance in the description of the structurally and catalytically related

chemistry of VOHPO4.0.5 H2O and of its derived phases.

Two main effects observed in the preparation of (VOHPO4.0.5 H2O)

may be, in fact, strictly connected to intercalation, properties. First, non

stoichiometry is easily accommodated as evidenced by the preparation of

compounds with (0.9-1.2) P: V ratio without any apparent modification of

structural and morphological properties (62).Second, the preparation

conditions have a pronounced effect on the microstructure, i.e., on the

morphology, solid - state reactivity, and the presence of disorder in the

stacking fold of crystalline planes of its layered structure.

In fact, reduction of the starting V (V) compound may be realized (62) by

using, for example, aqueous HCl or isobutyl alcohol. In both cases, almost

pure vanadyl hydrogen phosphate is obtained, but with different

microstructure (61,62,73). The layers of vanadyl hydrogen phosphate (010)

plane (Figure 1.5) are interconnected in tri dimensional structure by weak

hydrogen bonding of phosphates and of water molecules. The organic

alcohol competes with this effect, reducing the bonding between the planes

and allowing the formation of crystals in which these (010) planes are

predominantly exposed (plate like morphology) (21,56,59,62,74) .This effect, in

addition to the increase in surface area modifies the surface properties due

to a change in the relative ratio of crystalline planes at the surface. The

alcohol also can remain partially intercalated between layers (62) .This

effect induces some local modification of the vanadyl hydrogen phosphate

structure, which can strongly modify its solid-state reactivity (61,73).

Ball et al.(75) proposed that the known mono hydrogen phosphate phases

can be classified in three different structural types, namely type I,


18

(VOHPO4.0.5H2O) ;type II (α-VOHPO4.2H2O); type III, (VOHPO4.4H2O)

and (β-VOHPO4.2H2O) .practically ,the β-dihydrate appears as

intermediate in the thermal treatment of the tetrahydrate from both

thermogravimetric and thermodiffractometric experiments .Moreover , the

structure of (β-VOHPO4.2H2O) is closely related with that of

(VOHPO4.4H2O) (Figure 1.6, 1.7-a and 1.7-b). Both compounds present a

similar arrangement of (VO6) octahedra and (PO4) tetrahedral forming

alternating chains which lie along the (c) direction in the β-dihydrate and

along the a direction in the tetrahydrate and along the (a) direction in the

tetrahydrate. The coordination geometry around both vanadium and

phosphorus atom is similar in both structures. Each phosphate group

contains three oxygen atoms (shared with three different vanadium atoms)

and a hydroxyl group. The (V) atoms show very similar coordination

polyhedra, having a water molecule trans-coordinated to the axial (V=O)

group and a second coordinated water molecule in the equatorial plane of

the (VO6) octahedra. The similarity of these structures suggests that the

dehydration of the tetrahydrate into the dihydrate may precede

topotacitically. However, several ways for the reorganization of the infinite

double chains that lie parallel to the (a) axis (6.379 °A ) of the tetrahydrate

may be imagined to give the interconnected single chains running in the (c)

direction (12.623 °A = 2 X 6.379 °A) of the β-dihydrate . All the possible

models that imagined need to break some bonds (at least 2 or 4 per unit

cell), to rotate some polyhedra and to reconstruct the bonds.

Figure 1.7- a: Projection of the structure of β-VOHPO4. 2H2O along (100) (75).


19

Figure 1.7-b: Projection of the structure of β-VOHPO4. 2H2O along (010) (75).

The transformation of V (IV) and V (V) phases into vanadyl

pyrophosphate is an important step in forming active catalysts, by thermal

treatment (Calcinations) at 400 °C, the VOHPO4.0.5 H2O dehydrates to

(VO)2P2O7 .Alternatively, the vanadyl pyrophosphate may be obtained by

reduction from VOPO4 or by thermal treatment of different crystalline or

amorphous (V-P-O-X) phases (where X indicates a generic thermally

decomposable anion or cation such as (C2O4 2- or NH4

+ ) (63, 76).

The vanadyl pyrophosphate is built of chains of (V) polyhedral linked

by pyrophosphate groups (60,77) .The (V) atoms are linked through the

oxygen atoms of the vanadyl in (V-O-V) chains in the (c) direction, and

the (V) atom octahedra are linked in pairs through a common edge,

forming double chains in this direction. The vanadyl groups in the paired

vanadium octahedra are oriented trans to one another (Figs. 1.8 and 1.9).


20

Figure 1.8: Idealized structure of (VO)2P2O7 (65).

Figure 1.9: crystal structure of (VO)2P2O7 (020) plane (58).


21

The unit cell of (VO)2P2O7 is orthorhombic (77,78) and is topological

similar in its (bc plane) to that of (VOHPO4.0.5H2O) in the (ab)

plane (56,60).The change from face-shared to edge-shard vanadium

octahedral in converting the hemihydrate to the pyrophosphate results in a

small expansion of one axis, but the other in plane dimension shows little

change according the pseudomorphic relations between the two crystalline

structures (59) as shown in (Figure1.10) this indicates a topotactic

mechanism of transformation, and this result has been confirmed with

Scanning Electron Microscopy. (21,56,59,62) .The layer spacing decreases from

(5.69°A) in the hemihydrate to (3.91°A) in the pyrophosphate. This is

consistent with removing the water molecules shared by the vanadium

pairs and filling the resulting vanadium coordination site with the oxygen

atoms of vanadyl groups from the layers above. This transformation

requires only very small displacements of the atoms. Importantly, since the

conversion of the hemihydrate to pyrophosphate can take place without

breaking any (V-O-P) bonds, the structural order/disorder and morphology

of the precursor phase are maintained during the transformation to vanadyl

pyrophosphate. This means that it is possible to control some of the

structural /morphological characteristics of vanadyl pyrophosphate by

controlling the specific nature of the vanadyl hydrogen phosphate

hemihydrate precursor phase (62,73). Furthermore the terminal vanadyl

oxygen atoms in the face – shared octahedral pairs of vanadyl hydrogen

phosphate have a syn arrangement; while in vanadyl pyrophosphate they

are in anti positions. These arrangements in the layer stacking direction

result in the initial formation of (VO)2P2O7 crystalline with many defects

(36,37,65,79). Alternatively, Bordes and Courtine (70,74) discussed the possible

presence of disorder in the crystalline structure of (VO)2P2O7 in terms of

different crystalline phases (β- and γ- VO)2P2O7) in which former

possesses a network structure versus a layered structure for the γ- phase.


22

X-ray diffraction patterns, however, are very similar in these two phases.

According to these authors, β-(VO)2P2O7) and γ-(VO)2P2O7 forms by

dehydration of VOHPO4.0.5H2O or reduction from γ-VOPO4.

VOHPO4.0.5H2O

α = 104.6 °

β= 127.82 °

(VO)2P2O7

α = 108.2 °

Figure1.10: Pseudomorphic relations between the crystal of VOHPO4.05H2O

and (VO)2P2O7 (36).

Vanadyl acid phosphate VO (H2PO4)2 is another phase of (VPO)

system (80,81) , this phase is made of infinite chains of corner sharing (VO6)

octahedra and isolated PO2(OH)2 tetrhedra (80) .The (V) atoms are displaced


23

(0.364) °A from the equatorial plane along the fourfold axis. As a result,

one short bond (1.600 °A) which characterizes the vanadyl (VO2+) ion,

forms in almost regular oxygen octahedra .The bond between the (V atom)

and the oxygen in the trans position is significantly longer(2.382 °A).

Furthermore ,the equatorial planes of the octahedra are alternatively

rotated of ( +18° and - 18°) around the chain axis (Fig. 1.11-a ).Phosphate

tetrahedra act as bidentate bridges via oxygen atoms for (V) atoms

belonging to adjacent chains, (OH) corners established a contact between

tetrahedra to form hydrogen bonds (Figs. 1.11- a &1.11-b ).

(A)

Figure 1.11-a: Crystal structure of VO (H2PO4)2 along XY axis (80).


24

(B)

Figure 1.11-b: Crystal structure of VO (H2PO4)2 along XZ (80).

1.3.2 - Active phase in n-Butane Selective Conversion to MA

Because the (VPO) system is characterized by the facile formation of a

number of crystalline phases, the structure of the active phase must be

discussed in term of factors such as, oxidation state, (P/V) ratio, and

crystal phase transformations under reaction atmosphere.

The various crystal phases can interconvert as a function of the reducing or

oxidizing properties of the reactants, the time on stream, and the reaction

temperature (29,37,61,82,83).

The orthophosphate (VOPO4) phases are transformed to (VO)2P2O7 by

reaction with the hydrocarbon mixture. In this reduction process single

(VO6) octahedral form pairs by loss of oxide anions. However, the

different (VOPO4) phases previously discussed possess different

reducibility to vanadyl pyrophosphate, depending on the structure or

morphology (36).


25

The Complex solid-state chemistry of (VPO) system has led to some

confusion and contradictions in the literature concerning the nature of the

active phase in the n-butane oxidation and the identification of the active

site involved in the different steps of the reaction.

Bordes and Courtine (37) suggested that the active sites in n-butane

oxidation to (MA) were associated with coherent interfaces between slabs

of (100) VOPO4 and of (010) (VO)2P2O7 along the (100) and (201) planes,

respectively. On the contrary Volta et al. (84,85) believed that the active sites

are not associated with interfacial effects between two crystalline phases.

On the basis of comparison between X-ray diffraction, they suggested that

the active phase for selective oxidation of n-butane consists of a mixture of

well-crystallized (VO)2P2O7 (V4+) and an amorphous surface (VPO) phase

of (V5+) involving many corner-sharing VO6 octahedra. This amorphous

phase may be interpreted as a precursor of (β-VOPO4), which forms at

higher reaction temperatures.

Hodnett and Delmon (67-69) used a prereduction treatment with hydrogen

in order to improve selectivity to maleic anhydride and suggest that the

best catalyst consists of an oxidized surface layer built upon a reduced core

of a V (IV) phase. The selectivity is not related to the presence of a

specific well-crystallized phase, but only to the distribution of vanadium

oxidation states between the bulk and the surface.

The active catalyst must possess an optimal [V (IV)/V (V)] ratio for

selectivity in n-butane oxidation according to Zozhigalov et al., (86) from

studies of V-P compounds with variable V (IV) /V (V) ratios. These

compounds preserve the layer structure of the initial (α-V5+OPO4)

compound. Optimal performance properties are associated with catalysts

containing [4-9 V (V) ions per V (IV) ion]. n-butane oxidation and (MA)

take place at the expense of the catalyst surface oxygen and are

accompanied by its reduction to (VO)2P2O7. On (VO)2P2O7 however, the


26

rate of reduction of the catalyst is lower than that of butane to (MA)

reaction, and therefore these authors conclude that the V(V) phase is

reduced and that the V(IV) phase is oxidized under the dynamic conditions

of catalytic reaction (86).

Weing and Schrader (74) claimed that only (VO)2P2O7 is the active and

selective phase in n-butane oxidation to (MA). A slight excess of catalyst

phosphorus (P/ V=1.1 catalysts) is necessary to stabilize the active phase

The excess phosphorus creates a distortion of (P2O7)4- crystal

environment. The α-VOPO4 is considered by these authors and others (21)

as active, but non selective phase.

Many authors proposed that the vanadyl pyrophosphate (VO)2P2O7 as

the active phase (1,9,21,56,62,73,77,80,87) .Trifiro et al. (62,88-90) attributed the

activity of the catalyst to the V(IV) phase (the vanadyl pyrophosphate),

whereas the selectivity to maleic anhydride(MA) was connected to the

presence of a very limited and controlled amount of V (V) sites .

The same discordance as to the nature of the active phase is present in

the literature concerning the optimal (P/V) ratio of the catalysts, even

though there is a general agreement that phosphorus stabilizes the (+4)

valance state of vanadium and limits its oxidation (8,74,86,90).

Garbassi et al. (91) have found that the specific conversion of

n-butane increases by an order of magnitude for a (P/V) ratio just

exceeding unity, but extended X-ray absorption measurements do not show

any structural effect of the phosphorus.

The (P/V) ratio is a key parameter in determining catalyst selectivity

and activity according to Weing and Schrader (74). Selectivity for (MA)

increase with catalyst phosphorus loading, whereas specific activity of

both selective and nonselective oxidation decreases on increase of

phosphorus content in the range (0.9-1.2) P/V range. Best catalytic


27

performances are exhibited with a catalyst with (P/V = 1.1) according to

Buchanan and Sunderesan (92).

Similar results were observed by Pepera et al. (93). According to Bosh

and co-workers (94) who studied the selective oxidation of n-butane to (MA)

under oxygen deficient over (VPO) catalyst, they found that the selectivity

was strongly influenced by the actual surface V (V)/V (IV) ratio).

Garbassi et al. (91) found a value of (P/V) surface ratio in the

(2.0-2.8) range for P/V bulk ratio in the (1.0-1.4) range.

Hodnett and Delmon (67-69) reported that the surface (P/V) ratio is (1.0)

for bulk stoichiometric (P/V) values of 1.0 or higher. They concluded

therefore that the reactivity of near-surface layers is hardly affected by the

(P/V) ratio, but bulk reactivity is drastically curtailed.

Selectivity to (MA) from n-butane maximizes for (P/V) =1.0 according

to Ai (70), who associated this catalytic effect with the presence of strong-

acid sites to activate n-butane. Finally, an optimal value of (P/V) around

(1.0) was suggested by Trifiro et al. (73,88-90) and Contractor et al. (21).

The active sites has been proposed as being the (200) plane of

(VO)2P2O7 crystallites that preferentially expose the (200) plane have been

found to exhibit higher selectivities for n-butane oxidation (95).

Transformation Electron Microscopy has shown that catalyst selectivity

correlated with the disappearance of amorphous platelets (96). Seven

different active sites have been suggested for the (200) plane of (VO)2P2O7

catalysts as shown in (Figure 1.12) whereas:

a- Bronsted acid sites, probably –POH groups.

b- Lewis acid sites, probably V IV and V V.

c- One electron redox couples, (VV / V IV), (V IV / VIII ).

d- Two electron redox couples (V V / VIII ).

e- Bridging oxygen, V-O-V, or V-O-P.

f- Terminal oxygen VV= O.


28

g- Activated molecular oxygen peroxo and superoxo species (96).

Figure.1.12: Termination of the (200) plane of (VO)2P2O7 and proposed active

sites for oxidation (s1) = Lewis acid, (s2) = Lewis acid site, (s3) = terminal

oxygen, (s4) = bridging oxygen, (s5) = superoxo and peroxo site, (s6) = V v/ viv

redox couple (97).

1.4- Kinetics of n-Butane Oxidation

A central question in analyzing the problem of bridging the gap

between surface science and applied catalysis approaches is the

verification of the possibility of describing the macro kinetic behavior

using rate equations and constants derived from the analysis of the kinetics

of the single elementary steps. Impressive results have been obtained in

this direction (98) .The fitting of macro kinetic data on the basis of a

microkinetic model is usually considered the best demonstration of the

applicability of the suggested reaction mechanism under real working

conditions (99). Is it this true also for more complex and multifunctional

reactions? In order to replay to this question it is good to briefly recall the

principles of the Langmuir description of the catalytic reaction on an


29

"ideal" surface, because this is the basis of most of the macrokinetic

models for deriving the reaction rates over catalytic surfaces. On the ideal

Langmuir surface there is one type only of active sites, the energetic of

chemisorption is independent on the coverage, and there is no interaction

between substrate (Figure 1.13-a). Real surfaces are far from the ideal

Langmuir surface, as schematically represented in (Figure 1.13- b), due to

the presence of surface heterogeneities the simplest is the presence of

steps and kinks, but as mentioned In the introduction the situation is far

more complex on oxides) which imply a distribution in the energetic of

interaction, and the presence of interaction between the adsorbates .A

chemical bond of a molecule with a surface site implies the donation of

electrons to or from the surface with thus a modification of the catalyst

conduction band, band gap, etc.; therefore, the energetic of interaction of

molecule with the surface is not independent from the surface coverage,

and the adsorption of " spectator" species, i. e. species which do not play

a direct role in the reaction mechanism, should be also considered. This

problem is even more accentuated in selective oxidation reactions where a

multielectron transfer occurs, as mentioned before.

It may be thus expected that kinetic models based on Langmuir

approach do not correctly fit macrokinetic data, but on the contrary it is a

common experience that these models are well suited to describe the

macrokinetic behavior. How can this dilemma be solved? One possible

interpretation, which is schematically shown in (Figure 1.13-c), is that

during catalytic reaction the largest part of the catalyst surface is covered

by "strongly chemisorbed" species which are characterized species

characterized by a surface lifetimes much longer than of reaction

intermediates and therefore reactant or intermediate seen a local situation

similar to "ideal" surface. This pointed out that the concept of " clean"

catalyst surface, i. e. of reaction of a molecule at single specific active sites


30

without considering the modification of the surface reactivity induced by

the presence of other co-adsorbates (reactants, intermediates, " spectator"

species), may not lead to correct description of the " real" working catalyst

surface and reaction mechanism, especially when complex, multi steps

reactions (selective oxidation reactions, for example) are considered.


31

A- A simple model of the ideal Langmuir surface

B- A simple model of real surface with kinks and steps, surface interaction between

adsorbates and electron donation of a chemisorbed molecule to catalyst conduction

band (strongly chemisorbed species)

C-Working catalyst surface with the largest part of the surface occupied from strongly

chemisorbed (reactant intermediate)

Figure 1.13- Model of the ideal Langmuir surface (99).


32

Several studies have been conducted in the eighties to derive kinetic

expressions to describe the reaction sequence for n-butane oxidation to

Maleic Anhydride in steady- state conditions on (VPO) system (86,87,91,100,101).

Escardino et al. (100) studied the kinetic of n-butane oxidation in fluidized -

bed reactor over (VPO) catalyst with (P/V= 0.8) at (676-753) K.

A triangular reaction network was proposed as shown in Scheme 1.5.

CH3

CH3

O

O

O

CO , CO2

r1

r2 r3

Scheme1.5: A triangular reaction network on n-butane to MA (100).

Maleic Anhydride (MA) and Carbon oxides (COx) were formed directly

from n-butane (at rates r1 and r2 respectively), and MA was also oxidized

to carbon oxides (at rate r3). At n-butane concentrations typical of

industrial reactors, the rate of n-butane oxidation was controlled by the

reaction between n-butane gas and surface oxygen.


33

Wohlfahrt and Hofmann (101) investigated n-butane oxidation kinetics

over a wide range of n-butane and oxygen concentration at (719-777K)

over (VPO) catalyst.

Centi, et al. (87) used a very active catalyst prepared in an organic

medium which permitted low reaction temperatures (573-613K) to be

employed. Under these conditions it was found that the rate of reaction of

n-butane to carbon dioxide did not depend on the hydrocarbon

concentration, but only on the concentration of oxygen.

The order of reaction maintained in the literature depended on the

method of analysis of available data from experimental results of

hydrocarbon depletion and oxygen partial pressure. The comparison of

these kinetics results stressed the importance of surface catalyst behavior

(selectivity to MA) as well as the role of the redox properties of vanadium

in determining the activity of the catalyst (87,101).

1.5- Surface Modifications by Interaction of n-Butane with

Catalyst Surface

Centi et al. (87) used fresh catalyst for their kinetics study and suggested

that, a critical factor governing the selectivity at very high butane

conversion is the instability of the formed (MA) in the back end of the

catalytic bed. It is thus worthwhile to analyze the variations in the catalyst

surface as a function of position in the catalytic bed (86). These experiments

showed formation of V (V) at the end of catalyst bed which was attributed

to formation the more oxidizing atmosphere present.

Trifiro et al. (89) studied the oxidation of n-butane at low and high

hydrocarbon concentrations on vanadium (IV)-phosphorus (1:1) mixed

oxide in relation to the surface modifications induced by the reaction

medium. The results showed the presence, in the catalyst, of high amounts


34

of V (III) together with V (IV) and the absence of V (V), whereas lower

amounts of V (III) together with both V (IV) and V (V) are formed, when

maleic anhydride is formed. It is suggested that two redox couples operate

in (MA) synthesis:

a- V (IV)-V (III) in the synthesis of olefins from n-butane.

b- V (V)-V (IV) in the synthesis of (MA) from the olefins is formed.

Mori and co-workers (102) studied the oxidation of n-butane on various

unsupported or supported V(V) oxides, they confirmed that (V(V)=O)

species are very active in total combustion of butane, these experiments

results indicate the important relationship between catalytic behavior and

the surface modification induced by the medium itself. The redox

properties of the fed influence the surface oxidation state of the catalyst,

which in turn profoundly affected the nature of the products formed in the

reaction.

Centi and Perathoner (99) referred to (VO)2P2O7 as the active phase in

the industrial catalysts for the selective oxidation of n- butane to (MA).

This catalyst is very selective in n-butane oxidation, but when propane is

fed Instead of n-butane, only carbon oxides and traces of other products

(propene mainly) are detected.

A first question is thus why the decrease in length of the carbon chain

produces a so drastic change in the selectivity on the same catalyst? The

answer can be that from n-butane a stable product against consecutive

oxidation (MA) forms, but not in the case of propane oxidation. However,

this answer responds to only part of the problem. The data reported in

(Figure 1.14) have already pointed that the sensitivity of the reaction

product against consecutive oxidation is not the only factor that determines

the rate of consecutive oxidation to carbon oxides.


35

Figure 1.14- Comparison of the selectivity dependence on acrylonitrile in propane

and propane ammoxidation at 480 °C on VSb4 + Sb2O4 (99).

Data in (Figure 1.15) further evidence this problem. In fact, it is expected

that a decrease in the oxygen to alkane ratio may increase the selectivity to

the partial oxidation product (propene in the case of propane), because the

formation of carbon oxides requires a larger number of oxygen atoms

(propane oxidation to (CO2 + H2O) requires (5O2 molecules), whereas

propene formation from propane requires ten time less oxygen). However,

data in (Figure 1.15) show that this is possible only for high initial

concentrations of propane in the feed, whereas for lower initial

concentrations of propane the increase in propene selectivity decreasing

the (O2/ propane) inlet ratio is much less remarkable. A comparison

between the two cases points out that the differences arise from the

different formation of alkene product, which is about ten times larger using

the higher inlet concentration of propane in the feed. This indicates that

when the formation of propene is higher, the rate of its consecutive

oxidation to carbon oxides also decreases and thus the selectivity


36

increases. The same type of phenomenon has also been observed by

changing the (O2/ hydrocarbon ratio) in n-butane oxidation on (VO)2P2O7,

although in this case a wider range of intermediate products was

detected (100).

The formation of alkenes thus induces a self- modification of the

surface reactivity. This in confirmed from the analysis of the kinetics of

but- 1-ene oxidation on (VO)2P2O7, the analysis of the transient reactivity

as well as spectroscopic studies (95). This evidences that without

considering this self modification of the surface reactivity induced by the

reaction intermediates itself it is not possible to correctly describe the

kinetic of the surface reactions.

Figure 1.15- Selectivity in propane formation from propane on (VO)2P2O7 as a

function of the inlet ration between oxygen and propane concentrations, for two

entail concentrations of propane in the feed, reaction temperature 322 °C (99).

Both (VO)2P2O7 and (VSbO4 + Sb2O4) catalysts are characterized from

the presence of coordinatively unsaturated (V4+=O) surface sites which act

as strong Lewis acid sites and which play a relevant role in the mechanism

of alkane activation (103). Alkenes through their π- bond system can


37

chemisorb on these sites forming relatively stable chemisorbed species,

although they may be considered "spectator" species, because they are not

directly involved in the mechanism of further selective oxidation of these

alkene intermediates. Oxygen also strongly chemisorbs on the surface

Lewis acid sites forming thermal stable species (they desorb above 450-

500°C), but which play a relevant role in the mechanism of oxidation.

When the surface concentration of the intermediate alkenes in alkane

oxidation is as high as to limit the amount of chemisorbed oxygen, due to

this competitive chemisorption, it is thus possible to control the population

of oxygen adspecies by this mechanism. This explains the considerable

promotion of selectivity to Partial oxidation products by increasing alkane

inlet concentration (Figure 1.15) and, on the other hand, explains also the

apparent contradiction of the different rate of acrylonitrile consecutive

oxidation when forms from propene instead of that from propane

(Figure 1.14).

Although propene forms from propane as a reaction intermediate, its

concentration is clearly higher when it is fed directly and thus its effect in

limiting the concentration of Surface oxygen species is present even for

higher conversions of the hydrocarbon. Therefore, the maximum in the

formation of acrylonitrile is observed at higher hydrocarbon conversions

when the alkene is fed instead of the alkane on the same catalyst

(Figure 1.14).

In the oxidation of n-butane ،partial oxidation product (butenes and

butadiene) are observed at low oxygen concentrations when a very limited

number of vanadium (V) species are present on the surface of the catalyst.

The change of valance state of vanadium on the catalytic surface upon a

balance of three factors:

a- The redox potential of the feed.

b- The rate of oxidation of the catalyst at the temperature of reaction.


38

c- The rate of reduction of the catalyst at the temperature of reaction.

1.6- Relationship between Redox Properties & Catalytic

Behavior

As pointed out in the section on catalyst structure ( section 1.3), the

binary (VPO) system was rather complicated because of the great variety

of observed phases and the difficulties encountered in the development of

successful and reproducible syntheses of pure single phases stable under

reaction conditions. Over the past (20) years improved understanding of

the factors determining the structure of (VPO) has allowed the

development of suitable methods of preparation of (VO)2P2O7(65,89).In

particular is the preparation of vanadyl pyrophosphate active phase with

(0.95-1.2) P/V ratio and variable mean valance states of vanadium (up to

about 20% of V (V) without significant changes of X-ray diffraction

patterns (104) (Figure 1.16).

V(III) V(IV) V(V)

0 85 15

0 100 0

……. 11 89 0

Figure 1.16: Effect of the presence of different valence states of vanadium on the

principal X-ray diffraction lines of vanadyl pyroph osphate (P/V=1.0) (104) .


39

The variation of the content of (P) in the composition modifies the

catalytic properties in n-butane oxidation (86) and the redox properties of the

catalysts (104), a slight deficiency of phosphorus, does not change the rate of

V (IV) oxidation to V (V). On the other hand, an excess of phosphorus

only slightly influences the rate of oxidation, but strongly affects the rate

of reduction. This effect is in part attributed to a decrease of the number of

the active site, but also reflects a particular kind of interaction of

phosphorus upon the vanadium ions (61,104). In order to correlate the

observed variations in the redox properties with the catalytic behavior in n-

butane oxidation on the fresh catalyst used by these authors, it is necessary

to distinguish between tests at low conversion and high conversion.

At low conversion, the catalysts with (P/V ratio = 0.95 and 1.01) show

the same activity, selectivity, and kinetic behavior (86). However, at high

conversion of n-butane (80%) the catalyst deficient in phosphorus and with

the observed higher rate of vanadium oxidation forms primarily carbon

oxides. The catalyst with more phosphorus with respect to the

stoichiometric ratio of (1.0) are less active and less selective (90), but do not

show the strong decline of (MA) selectivity at the highest conversion.

It is reasonable to correlate these catalytic effects with the redox

properties of the catalyst (105). The strong increase of the rate of vanadium

oxidation in the phosphorus-deficient catalyst leads to an enhancement of

the rate of consecutive oxidation of ( MA) (effect on the selectivity),

whereas the decreased rate of reduction of vanadium in the catalyst with

higher (P) content than a stoichiometric one leads to the reduced rate of

hydrocarbon depletion (effect on the activity)

Nakamura et al. (105) also suggested that some V (V) ions were

necessary for (MA) synthesis, but the optimum mean of valance state of

catalyst is close to four. Similar conclusions were presented by Weing and

Schrader (74). They reported catalyst phosphorus loading is a key parameter


40

in determining catalyst selectivity and activity. An increase of (P) content

in the (0.9-1.2 P/V) ratio decreases specific activity in n-butane depletion

and increases the selectivity to (MA). The same selectivity was reported by

Pytnitskaya et al. (71), Buchanan and Sundarsan (92), also found a correlation

in vanadium phosphorus oxides between the concentration of V (IV) ions

in the discharged catalyst and the activity in n-butane oxidation.

The results of surface modifications induced by the interaction of C-4

hydrocarbons as well as the results of (P/V) ratio effect on redox properties

and on (selectivity/activity) are interpreted on the basis of two different

types of reaction in going from the alkane to (MA): a first step is oxidative

dehydrogenation up to adsorbed butadiene and second step is further

oxidation of this intermediate. The first step is suggested to be controlled

by the rate of V (IV) reduction, whereas the second step is controlled by

the amount of V (V) available. However, when the content of V (V) is too

high, the oxidation proceeds further to carbon oxides.

These data suggest a strong relationship between selectivity to (MA) at

high conversion and rate of formation of V (V) as reported by Trifiro and

co-worker (62) (Figure 1.17).

(A)

Conversion

3 >2 >1 (in disorder arrangement) (62)

Figure 1.17-a: Maleic Anhydride yield VS conversion at 300°C in n-butane

oxidation for different catalyst with P/V= 1.0, feed = 0.6% n-butane, 12% O2 (62).

MA

%


41

(B)

Time (minutes)

Figure 1.17- b: Rate of V (IV) oxidation to V (V) in air for the same catalyst of

(A) (62).

1. ٧ - Mechanism of n-Butane Activation

Kinetics data agree that the hydrocarbon activation is the first step in

n-butane oxidation, the alkane is rather inert and the rate determining step

for saturated hydrocarbons the is the dissociation of the (C-H)

bond in a manner similar to the production of hydrocarbon free

radicals(87, 91,100,101,106) . This would be requiring very reactive oxygen to

break (C-H) bond which is bound weakly to the solid for it to be

thermodynamically favorable. It is known from the literature (107-110) that

(O) - and (O2) - species (one oxygen molecule may be activated by

accepting electrons from the solid) are very oxidative dehydrogenation

and activation of light alkane (108-113) i.e., highly reactive surface oxygen

(weakly adsorbed oxygen and or localized surface lattice defects) are the

active sites for alkane activation. They suggested that the oxygen

chemisorption occurs via a dissociative pathway on vanadium dimers-

V (

V)

/ V to

t %


42

leading to a (V (V)-O*) type surface species capable of activating the

alkane.

Due to the topotactic transformation that occurs without the breaking

of (VPO) connections and due to the higher electronegativity of (P) with

respect to (V), the presence of defects in the structure of vanadyl

pyrophosphate is reflected on the surface in an enhancement of Lewis

acidity of surface unsaturated vanadium ions (Figure 1.18). Strong acid

sites on (VO)2P2O7 were observed also by Puttock and Rochester (111,112) by

infrared spectroscopy.

(A)

A- Medium –strong Lewis acid sites, well crystalline structure

(B)

B- Very strong Lewis acid sites, disordered crystalline structure

Figure 1.18: Model of Lewis acid sites in (VO)2P2O7 with different degrees of

disorder in the stacking fold of (020) planes (62).


43

All above observations may be summarized in the following model:

The presence of defects in the crystalline structure creates strain in the

V-P-O bonds and would create a surface-activated reactive couple.

V O Pδ δ+

− − − − −−

→

……….. (1.5)

This reactive couple was proposed by Centi et al. (73) as being

responsible for the concerted mechanism of the removal of two hydrogen

atoms, according to the (Figure 1.19-a, 1.19-b and Figure 1.20) which show

the redox cycle proposed to this mechanism of activation of n-butane.The

model is based on the idea that (V (IV)–oxygen) are the active sites of butane

dehydrogenation (62,99).

Munson and et al. (113) had developed an experimental protocol to study

the mechanism of this reaction in which 13C-isotopically labeled n-butane

is flowed over a catalyst bed and the reaction products are analyzed using 13C NMR spectroscopy. This strongly suggests that the total oxidation of

n-butane on (VPO) catalysts involves the oxidation and abstraction of the

two methyl groups of n-butane, and the two methylene groups of n-butane

form ethylene. An organometallic mechanism is proposed to explain these

results.

In conclusion, it is necessary to try to develop new models of the

surface reactivity which include new evidence on aspects such as (i) the

role of chemisorbed species on the surface reactivity, (ii) the presence of

multiple pathways of reaction (iii) the dynamics of catalyst reconstruction

(iv) the mobility of surface adspecies, etc. The consideration of all these

possible effects in analyzing the surface reactivity will make possible the

design of new catalysts as well as the understanding of surface

reactivity at oxide surfaces (99).


44

HC

CMe

Me

H

V O

V

O

OP

H

(A) A- Detail of the concerted mechanism of (2H) abstraction.

VV

O

O

O

CCH

CH3

H3C

H

H

H

V VO

O

O

O

O

O

O

H

HO

-H2O

VV

O

O O

O

O2

O

O

O

O

3+

O

CH

HC

H3C

CH3

4+

(B) B- Redox Cycle involved in the mechanism of activation.

Figure 1.19: Proposed mechanism of n-butane activation on VO)2P2O7(73).


45

O O

R

O O

+

O

O OO

OO O

COOHHOOC

HOOC

O O

O O

CH3COOH

O

CHO

O O

COX

Furan

MA

Dihydrofuran

Butadiene

n-Butane

Methyl vinyl Ketone

But-3-ene-oxide

Crotonaldehyde

Maleic acid

Butirrolactone

But-1-ene-3-oxide

2-Butoxide

2-Butanone

Butenes

H2O

Figure 1.20: Reaction network in n-butane oxidation on (VO)2P2O7(99).


46

1.8 - Role of Adsorbed & Lattice Oxygen Species

In the selective oxidation of n-butane to (MA) on vanadyl

pyrophosphate (VO)2P2O7 catalysts, many results have been reported on

the respective roles of lattice and gas-phase oxygen in the formation of

partial and complete oxidation products (113).

Three types of lattice oxygen species with different reactivites are

present at the surface of vanadyl pyrophosphate (Figure 1.21) (24).

OVO

VV

P

O

VV

(a) (b) (c)

Figure 1.21: The Three Types of Oxygen Species (24).

The participation of lattice oxygen is a general characteristic of metal

oxide catalyzed reaction in selective hydrocarbon oxidation (114-117).

Pepera et al. (93) concluded that the lattice oxygen ions located in the

top few surface layers are responsible for the oxidation of n-butane to

(MA, CO and CO2). Another authors (106 , 117, 118) obtained experimental

evidence to support this conclusion. Abon and co-workers (119) claimed on

the basis of isotopic labeling results that only lattice oxygen is active for

the formation of (MA) and other products. The circulating fluidized-bed

riser reactor technology for (MA) production described in the literature


47

(18,21,119) is based upon the fact that, under anaerobic conditions, the lattice

oxygen of (VPO) can selectively oxidize butane to (MA) (18,21,119).

This problem becomes more complex when oxygen is co-fed with

butane under steady state reaction conditions employed in industrial fixed-

bed reactor processes. In addition to the conventional Mars-Van Krevelen

mechanism, where lattice oxygen is the active agent for butane oxidation,

gas-phase oxygen, surface lattice oxygen and / or activated chemisorbed

oxygen have all been proposed as important oxidants in the formation of

Maleic anhydride or in the formation of unselective products, (CO and

CO2)(96,107,120-122). For example, Trifiro et al. (95) proposed that adsorbed

oxygen is responsible for selective oxidation and that it is involved in the

oxygen insertion steps required for the formation of (MA). Ebner and co-

workers (95,106) concluded that adsorbed oxygen is selective only in the

(MA) formation step. They proposed that two types of oxygen are involved

in butane oxidation: surface lattice oxygen that is responsible for ring

closure, and activated chemisorbed oxygen, (O*), that is involved in the

further step of (MA) formation. Rodemerck et al. (121) reported that

adsorbed oxygen is active but not selective, i. e.; it can only produce

(CO2). In contrast to all of these studies, Zazhigalov et al. (122) concluded

that (MA) formation over (VO)2P2O7 is mainly due to gas- phase oxygen.

The controversy in the literature about the roles of lattice and

chemisorbed oxygen calls for further clarification of the puzzle. Wang and

co-workers (123) used a novel microbalance reactor to carry out kinetic

analyses of butane oxidation by (VPO) catalysts and of the oxidation of

partially reduced (VPO) with oxygen .These authors conclude from there

experiments that both lattice oxygen and adsorbed oxygen on (VPO)

catalyst can selectively oxidize butane to (MA). Under aerobic conditions,

the oxidation of butane by adsorbed oxygen species is much faster than

by lattice oxygen.


48

The first attempt to rationalize the problem of types of oxygen species

involved in the mechanism of C-4 alkene transformation to (MA) was

reported by Weiss et al. (124) , vanadium-oxygen double bonds (V=O) were

suggested to be the active sites both in:

The oxydehydrogenation step from butenes to butadiene, the

consecutive steps of oxygen insertion to (MA).The first step involves a

homolytic (C-H) dissociation of the adsorbed olefine and formation of the

allyl radical coordinated to a metallic transition ion as a surface π-allyl

complex. The allyl carbcation obtained can form the diene by loss of a

proton through reaction with a nucleophilic agent. The allyl ester that

forms can undergo a rapid reversible rearrangement in which each end of

the carbon of the skeleton is alternatively bonded to the lattice oxygen.

This lattice oxygen possesses a weak electrophilic reactivity. On the

contrary, electrophilic oxygen must be involved in the attack on the diene.

The proposed mechanism is shown in (Figure 1.22-a). The product of

attack on the diene can occur according to two-selective pathways, the first

one involves 1, 4- cyclization to 2, 5-dihydrofuran which is easily

dehydrogenated to furan; the second pathway is deprotonization of the

allyl carbcation to a dienolate structure.


49

n-Butane

- 2HCH2=CH-CH2-CH3 CH2=CH-CH=CH2

CH2=CH-CH=CH2

O

M

O

MH

HO

M3

CH3-CH-CH-CHO

HO OM

O

M2

-2H

-2 e a

O

b

H2O -2e

-H

-H

(A) Figure 1.22 -a: Proposed reaction of n-butane to MA (62,124).

Hetrocyclization of this structure by a new intermolecular electrophilic

attack leads to furan. The second path, however, can also form

crotonaldehyde by hydration and retroaldolization, leading to a form

crotonaldehyde by possible reduction in selectivity .An analogous

mechanism may be assumed for the further oxidation to (MA),i.e.,

electrophilic attack ,deprotonization , etc…….(Figure 1.22- b) .


50

O

-2e

O

M

O

O

M

-H

O

H2O

O

M

-2e-H

O

OH

OH

-2H

OOO

M2+

(B)

Figure 1.22- b: Proposed reaction of n-butane to MA (62,124).

1.19 - Nature & Mobility of Adsorbed Species

Puttock and Rochester (125)

studied the adsorption of n-butane, 1-

butene, 1, 3-butadiene, furan and (MA) on vanadyl pyrophosphate by

infrared spectroscopy. Contact between butane vapor and vanadyl

pyrophosphate at temperatures up to (450°C) did not give rise to infrared

bands due to adsorbed species. 1-Butene, on the contrary, gave rise to the

formation of adsorbed species by contact at room temperature. Higher

temperatures (200°C) catalyzed the transformation of 1-butene to 2-butene,

confirming the high surface acidity of these catalysts (104,125,126)

.

After removal of the gas phase an adsorbed species, characterized by a

strong band observed near (1600cm-1) and bands appeared in the

(2800-3000cm-1) region, was observed. The latter bands, which attributed

to saturated hydrocarbon groups, disappeared at higher temperatures

(350°C), whereas the band near (1600cm-1) unchanged. This band was also

observed by Rozhkova et al. (127)

and was attributed to a surface carbonyl.


51

A similar spectrum of adsorbed species was found in the interaction of

furan with the catalyst. The primary interaction of furan and vanadyl

pyrophosphate was suggested to involve coordination between the oxygen

atoms of associatively adsorbed furan molecules and coordinatively

unsaturated exposed vanadium cations with Lewis acid properties. Furan

showed a slight oxidation to maleic anhydride in the absence of oxygen but

the presence of oxygen promoted maleic anhydride was primarily non

dissociatively adsorbed on (VO)2P

2O

7, but two further bands at (1560 and

1450cm-1), attributed to carboxylate anions, suggested a partial oxidation

of the adsorbed (MA).

Weing and Schrader (74)

studied by infrared spectroscopy the n-butane

interaction with (VPO) catalysts with variable (P/V= 0.9, 1.0 and 1.1)

using an in situ FTIR cell. These authors presented evidence for the

presence of reactant (n-butane), partially oxidized product (MA),

combustion products (CO, CO2, and H2O) and reactive surface species

(Maleic acid and olefins) on the vanadyl pyrophosphate. Another

observations reported by Pepera and co-workers (93)

agree with the

hypothesis that the catalytic process in this complex reaction involves the

shuttling of hydrogen a way from and oxygen toward the intermediate

adsorbed on the surface, and the intermediate forms a stable surface

species (which does not adsorbed or react) to (MA). This mechanism is a

key to the selectivity and the absence of intermediate products in the

reaction of n-butane conversion to (MA).

1.10- Promoted (VPO) Catalysts

Numerous attempts have been made to synthesize improved (VPO)

catalysts by adding various amounts of other metallic elements. Although


52

reports of some of these attempts have appeared in literature (128-146), the

vast majority were found in the form of patents (64,128-130,133-136)

.

The effect of these promoters was studied by many authors; Ai (131,132)

studied the effect of alkali metal addition on activity and selectivity. It was

reported that the addition of less than (10% Li) showed no remarkable

effect, Ai also studied the effect of the methods of preparing V2O

5-P

2O

5-

ZrO2 catalysts on their activity and selectivity in the oxidation of n-butane

by changing the procedures of ZrO2addition. Ai found that the best

performance was obtained with the catalyst prepared by adding

simultaneously two solutions of ethyleneglycol in which ZnOCl2 and o-

H3PO

4 have been dissolved to a powder of the precursor of the (V

2O

5-

P2O

5) catalyst prepared in an organic medium. The presence of the vanadyl

pyrophosphate crystalline phase was detected in all of the catalysts. (131,132)

The addition of suitable activators to (VPO) catalyst was reported to

give an improved yield of (MA) by several patent claims (6,133-136)

using a

catalyst containing (P-V-Mo) to oxidation n-butane to (MA), this catalyst

gave a maximum (MA) yield of (88%).

Zazhigalov and co-workers (128,137) studied the effect of added alkali and

alkaline-earth metals (Li, Na, K, Mg, Ca or Ba) on the properties of (VPO)

catalysts. The results showed that the addition of alkali and alkaline earth

metals to these catalysts produced two effects, on the one hand, there was

an increase in the content of oxygen on the surface and hence an increase

in the number of acid centers, on the other hand, the presence of these

promoters increase the (P/V) ratio on the surface of the catalyst and this

explains their stabilizing effect. It should be noted that the greatest increase

in the surface content of phosphorus was observed when (Li) is added, and

(Li) was considered to be one of the best stabilizing additives


53

A new catalyst belonging to the (V-P-Mo-O) system selective in the

mild oxidation of n-butane to (MA) were prepared and characterized by

Courtine et al. (139)

These attempts led to the conclusion that (Mo)could be

substituted up to (7%) in (VOPO4) phases. Thermal analysis of the

hydrated precursor, XRD and IR spectroscopies of both hydrated and

anhydrous solid phases obtained showed that the solid solutions

isostructural with (VOPO4.2H2O) and (α-VOPO4) receptively could be

formulated.

Zazhigalov and co-workers (140) studied the properties of cobalt-promot

(VO) ٢ P2O7 in the oxidation of n-butane. This study focused on the

influence of cobalt additives on the composition of the vanadium-

containing catalyst, and on cobalt's other properties which were important

for the production of industrial (VPO) catalysts. Fresh catalysts were

composed of (VOHPO4.0.5H2O) phase. After reaction the catalysts

contained (VO)2P2O7. Cobalt was uniformly distributed in the pellets. Its

presence increased the content of phosphorus at the surface, which

modified the surface acidity and in turn improved the selectivity for n-

butane oxidation. No changes of the profile of phosphorus with depth were

observed, even after (500h) on stream, the surface composition of the

catalyst remained unchanged. Cobalt stabilized the catalyst performance

by forming cobalt phosphate which reduced phosphorus losses, improves

its catalytic properties and prolonged its lifetime.

A series of (VPO) catalysts with different (P/V) ratios and with or

without (indium and TEOS) additives have been characterized by

controlled-environment (XRD, ICP, ESCA, TEM, SEM, BET) and

chemical titration used for n-butane oxidation to (MA) (141) .The best

catalyst contained slight excess (P /V) with indium and tetra ethyl ortho-

silicate (TEOS) promoters. It was found that excess P increased the


54

resistance of catalyst precursors toward oxidation and resulted in (VPO)

catalyst with a large exposed platelet face of layer morphology. The

promoters reduce the thickness of the platelet face of layer morphology.

The promoters reduce the thickness of the platelet and facilitate the

oxidation of the precursor which contains disordered (VOHPO4.0.5HO)

.The combination of the promoters and excess P results in a (VPO) catalyst

with appropriate oxidizability and morphology and gave high yields of

MA.

Guliants et al. (142) investigated the oxidation of n-butane to (MA)

over a model (Nb, Si, Ti, V and Zr) promoted bulk (VPO) and supported

vanadia catalyst , The promoters were concentrated in the surface region of

the bulk (VPO) catalysts .For the supported vanadia catalyst ,the vanadia

phase was present as a two-dimensional metal oxide over layer on the

different oxide supports (TiO2,ZrO2,Nb2O5,Al2O3 and SiO2) .No

correlation was found between the electronegativity of the promoter or

oxide support cation and the catalytic properties. Both promoted bulk

(VPO) and supported vanadia catalysts containing surface niobia species

were the most active and selective to (MA). These data suggested that the

activation of n-butane on both the bulk and supported vanadia catalysts

probably required both surface redox and acid sites, and that the acidity

also played an important role in controlling further kinetic steps of n-

butane oxidation.

The presence of (Na) in technical grade (V2O5) leads to solid solution

formation, this phenomenon lead to prepare VOPO4.2H2O with a new (P/V

=1.1). (1)

Sajip and co-workers (143) described and discussed the effect of Co and

Fe doping on (VPO) catalyst, prepared by organic method (with

isobutanol). At low levels, both Co and Fe dopants significantly enhanced

the selectivity and the intrinsic activity to MA. A combination of powder


55

X-ray diffraction, P NMR spin-echo mapping spectroscopy and

transmission electron microscopy ,together with catalysts tests data, was

utilized to analyze the origin of the effects of Co doping. Co appears to be

essentially insoluble in crystalline (VO)2P2O7 and was preferentially

distributed in and stabilized an amorphous (VPO) material .It is suggested

that the origin of the promotional effect of (Co) was associated with its

interaction with the disordered (VPO) phase. The same techniques have

been used to analyze the Fe-doped catalyst, but at present it is not possible

to be definitive concerning the specific location of the Fe-dopant within

the phases present .Previous studies have indicated that (Fe) can form a

solid solution within (VO)2P2O7 and therefore it is probable that the (Fe)

amorphous vanadium phosphate phases are formed in the catalyst system.

Other authors (144) studied the effect of (Bi) on (VPO) catalyst, they

found that the incorporation of (Bi) into the (VPO) lattice lowered the

overall vanadium oxidation state from (4.24 to 4.08) .It also lowered both

the peak maximum temperature for the desorption of oxygen from the

lattice from (1001 K) (undoped) to (964 K) with shoulder at (912 K). The

total oxygen desorbed from the Bi-doped catalyst was only one-fourth that

of the undoped catalyst, while the amount of oxygen removed by TPR was

roughly the same for both catalysts. These results suggested that in

anaerobic oxidation, the Bi-doped catalyst will have roughly the same

activity as in undoped catalyst in C-4 hydrocarbon oxidation but would

have a higher selectivity to products such as olefins and (MA).

The barothermal and mechanochemical treatment of (VPO-Bi)

precursor was investigated (145), the barothermal treatment led to increase of

the relative ratio (001) plane of precursor without changes of the phase

composition. The (P/V) surface ratio increased more than two times and

the phosphorus surplus forms the islands on catalyst surface which

decrease the available active surface fragments. The change of Bronsted /


56

Lewis acidity ratio of surface as result of treatment also was observed. The

activity of the catalyst in n-butane oxidation less change up to several

value of P/V ratio and decreased with its growth. The selectivity to (MA)

increase (more than 10% mol. %) practically in all interval of the (P/V)

ratio changes. Good correlation was observed between the selectivity to

Maleic Anhydride (MA) and Bronsted acidity of the catalyst.

The mechanochemical treatment of precursor showed an enhancement of

n-butane conversion and an improvement in (MA) selectivity and yield.

The sample milled in water exhibited a rise in conversion to (91%) but

selectivity increased only some percents. The maximal increase of the

selectivity after treatment in ethanol (more than 15 mol. %) and growth of

activity (but only 5-6 %) was observed. The correlation between the

changes of selectivity and Bronsted acidity was established (145).

1.11- Preparation of the catalysts

In accordance with the growing importance of the C-4 partial oxidation

route to (MA), numerous patents (16,19,20,146,147) have been awarded in the

past years for process which involved the preparation of (VPO) catalyst.

Various routes are available to prepare Vanadyl Hydrogen Phosphate

Hemihydrate VOHPO4.05H2O precursor, as outlined in Figure 1.23:

V2O5 + H3PO4 + i-BuOH

Organic Route

V2O5 + NH2OHHCl

VOHPO4.0.5H2O

(VO)2P2O7

Aqueous Route

+ H3PO4

Figure 1.23: Synthesis routes for vanadyl pyrophosphate catalysts (97).

The aqueous route entails the reduction ( V5+ ) by NH2OH , HCl or


57

other reducing agents followed by the addition of H3PO4, VOHPO4.0.5H2O

is recovered by crystallization or evaporation.

The organic route involves the reduction of V2O5 by an organic solvent

(an alcohol), followed by the reaction with o-H3PO4 and recovery of the solid.

After recovery of the precursor, it is washed to remove trace amounts of

water soluble (V5+) compound and then calcined in nitrogen at (773) K

followed by final activation in air or n-butane/air at (673) K. It is critical that

calcination and activation be carefully controlled or the catalyst can be over

oxidized resulting in the formation of VOPO4 which is non selective

catalyst for n-butane (97).

The synthesis route and the reaction conditions during synthesis, affect the

morphology of and ultimately the catalyst performance. The precursor

prepared by the aqueous synthesis route is generally more crystalline than the

catalyst prepared by the organic route (95) .The organic synthesis route results

in platelet crystalline morphology; the size of the platelets and the way they

pack are determined by the choice of organic solvent. Isobutanol produces

rosette morphology where the platelets agglomerate. With sec. butyl or t-butyl

alcohol well formed platelets forms that do not agglomerate (148). The crystal

ordering of the precursor is also affected by the solvent. Large alcoholic

molecules, such as benzyl alcohol, appear to produced platelets with stacking

faults, deduced from the broadening of the (200) reflection in the XRD (149).

Stacking fault defects in the precursor have been correlated with improved

catalyst performance (148).

In recent years Hutchings and co-workers (149) prepared the precursor

by using V2O4 as a starting material with water as solvent. They also used

V2O4 with either H3PO4 99% or H4P2O7; reacted in autoclave at 145 ˚C to

form well crystalline VOHPO4.0.5H2O.The surface area of the precursor is

significantly enhanced when water is added as a solvent. On activation in

(n-butane / air) the catalyst surface area is increased from (4 m2/gm) for the


58

precursor to (10-13 m2/gm). The selectivity to (MA) observed to be very

similar to other non- promoted (VPO) catalyst reported in the litterateur.

Although this study considered to unlikely that V2O4 can be used as a

starting material for the synthesis of commercial vanadium phosphate

catalyst, this study does show that relatively high area catalysts can be

achieved using water as solvent.

A dihydrate phases of (VPO) catalyst VOPO4.2H2O was prepared for

the first time in 1965 by Ladwig (33) by suspend V2O5 with rapped stirring

in distilled water and H3PO4 with (P/V = 7.3).The mixture was refluxed for

24 hours. The resulting precipitate was filtered, washed with distilled water

until a red color appeared then it was washed for several times with

acetone and left to dry.

No sign in the litterateur to prepare this phase with (P/V ratio =1.1).In

2000, several attempts were unsuccessfully exerted to prepare this phase in

the above ratio starting from V2O5 produced by Fluka and BDH company,

but all of these trials were not successful, on the other hand this phase was

successfully prepared starting from technical grade V2O5 with shorter

reaction time (16h), due to the intercalation effect of (Na ion) which led to

form solid solution of this phase. (1)

A new method through intercalation of VOPO4.2H2O crystallites in

primary alcohol (2- propanol or 1-butanol), followed by reduction with the

alcohol, have been investigated for the preparation of catalyst precursor.

Lamellar compounds were formed consisting of (V 4+, P +5 and alkyl

group) (150).

According to Bordes and Courtine (29), the calcination of VOPO4.2H2O

at 700 °C leads to obtain α-VOPO4.

Recently Bartley et al. (151) prepare vanadium (V) phosphate using

solvent-free method, by reacting (V2O5) with (H3PO4) in the absence of

water at 150 °C ,the resulting material was designed as anhydrous VOPO4.


59

This material is readily hydrate to formVOPO4.2H2O. Activation of this

phase with n-butane/air leads to formation of (α1-VOPO4), while the

reaction of anhydrous (VOPO4) with alcohol leads to the exclusive

formation of VO (H2PO4)2.

Finally, there are three important parameters affect in determining

the final characteristics and activity of these catalysts which can be

summarized as:

a- The reducing agent and solvent (8, 70, 61,152-157).

b- The (P/ V) ratio

(8,146,148,155).

c- Activation conditions (29, 71)

.


60

Conclusion of Bulk (VPO) catalyst

Oxidation of n-butane on the unsupported or bulk (VPO) catalysts is

the only known commercial process for an alkane oxidation.

A number of V (IV) and V (V) phosphates exist in the (VPO) system

and the correlation of catalytic performance with crystalline structure has

been reviewed (95,158).

Vanadyl (IV) pyrophosphate, (VO)2P2O7, has been identified as

critical for active and selective industrial catalysts (158). Some argue that the

(V5+/V4+) dimeric species in the top most oxidized layer of (VO)2P2O7 are

the active sites (8,159) , while others believe that the active sites lie within

the micro domains of crystalline vanadyl (V) orthophosphates, (β -, αII- ,

γ-, and δ-VOPO4), or (vanadyl (IV) meta phosphate) VO(PO3)2, formed

on the (100) faces of vanadyl pyrophosphate under the catalytic reaction

conditions (106). A recent kinetic study demonstrated that the best bulk

(VPO) catalysts contained only crystalline vanadyl (IV) pyrophosphate

after reaching the steady state (142). The vanadyl (IV) pyrophosphate phase

displayed preferential exposure of the (100) planes which contain vanadyl

dimers associated with active sites for n-butane oxidation (142). A high-

resolution Electron Microscopy study demonstrated that the surface (100)

planes in the fresh catalysts are covered with (1.5 nm) amorphous layered

which completely disappears within (23 days) of n-butane oxidation (96).

The kinetic studies of the bulk (VPO) catalysts further demonstrated the

similarity between the bulk (VPO) catalysts and supported catalytic

systems.

The catalytic activity of the bulk (VPO) catalysts was confined to a

very thin surface region of the (100) crystalline planes of vanadyl (IV)

pyrophosphate (1-2 atomic layers) (92).These findings suggested that the

crystalline vanadyl (IV) pyrophosphate phase in the bulk (VPO) catalysts

functioned as a support for the active surface. Such (VO)2P2O7 support


61

stabilized some specific surface termination of the (100) planes with

(V4+/V5+) species and phosphate groups without contributing its lattice

oxygen to n-butane oxidation. Unfortunately, various spectroscopic

techniques, such as (Raman, ESR, UV-Vis, IR and EXAFS/XANES), are

unable to provide information about the molecular structure of the surface

present in the bulk (VPO) catalysts because of the much stronger signals

from the catalyst bulk than the catalyst surface. Limitations of both the

bulk and surface characterization techniques currently available, coupled

with the complex solid-state chemistry of vanadium phosphates, have led

to considerable confusion and contradictions in the literature concerning

the identity of the active sites involved in different steps of n-butane

oxidation to (MA) over the bulk (VPO) catalysts.

1.12- Supported (VPO) Catalyst:

Supported metal oxide catalysts consist of an active metal oxide

component, e.g. vanadia, deposited at the surface of an oxide support, such

as SiO2, TiO2 and Al2O3.

Currently, considerable interest is devoted to study selective oxidation

of alkanes on supported vanadia catalysts. The selective oxidation of

alkanes is highly desirable due to their potentially low environmental

impact and the relatively low cost of raw materials. Some typical systems

being studied are partial oxidation of n-butane to (MA) on supported

(VPO) catalysts (105,160) and oxidative dehydrogenation of n-butane (161).

Many fundamentals questions still remain unanswered about supported

vanadia catalysts despite the importance of this catalytic system.

This section concerns with structure-property relationships in C4-

hydrocarbon oxidation on supported vanadia catalysts.


62

1.12.1-Molecular Structure of Vanadia Species in Fresh Catalysts

Structure of supported vanadia catalysts is a strong function of the

surface coverage. Surface coverage of vanadia over layer supported on

metal oxides has been determined experimentally using several

spectroscopic techniques, such as (Raman, IR, XPS, 51V NMR and UV-

Vis), as well as temperature-programmed reduction (TPR) and redox

reactions. Among these techniques, Raman spectroscopy is particularly

discriminating between the surface vanadia species and the crystalline

(V2O5). Characteristic Raman bands of polymeric vanadia species

corresponding to the (V-O-V) stretches, appear below (1000 cm-1) at high

coverages, while the isolated vanadate species present at sub monolayer

coverages produce only one Raman band of the terminal (V=O) bond at

(1020-1040cm-1). The finger print Raman features allow detecting

monolayer formation immediately prior to appearance of crystalline

(V2O5). Based on Raman spectroscopy measurements, monolayer surface

coverage on a number of oxide supports (Al2O3, TiO2, ZrO2, Nb2O5 and

CeO2) was found to be (7-8 VOx/nm2) (162) where nm is nanometer.

Vanadia supported on silica exhibited unusually low monolayer surface

coverage of only (0.7VOx/nm2). Lower density and reactivity of the

hydroxyls explain such low monolayer surface coverage on the silica

surface, which anchor vanadia species to the support. However,

commercial grade supports, such as pigment grade Titania, typically

contain monolayer quantities of (P, Na, K, Ca, etc.), which interact with

surface vanadia to form an amorphous phase (163). In such systems, the

surface vanadia titrates both the oxide support and the surface impurities,

delaying formation of crystalline V2O5 until several vanadia monolayers

are formed. Thus, the observed "monolayer'' vanadia coverage in such

impurity-containing systems can be (2-4) times greater than that found

above for the well defined supported catalysts.


63

A number of synthesis methods were used to prepare supported

vanadia catalysts, non aqueous impregnation with vanadium alkoxides (169,170), such as incipient impregnation of support powder like (TiO2) with

aqueous solution of NH4VO3 in oxalic acid (164). Aqueous impregnation

with vanadium oxalate is favored in commercial preparation due to its high

solubility in water and the absence of undesirable volatile organic solvents.

Under ambient conditions the supported catalysts contain multilayers of

adsorbed water. Under such conditions the bridging (V-O-support) bonds

are hydrolyzed and the hydrated vanadia species are dissolved in the thin

aqueous layer.

A new generation of supports based on ceramics with high thermal

conductivity and without oxygen atoms were prepared and tested to

disperse the active (VO)2P2O7 phase (165). The best example is a new

relatively large specific surface area (β-SiC < 20 m2/g) prepared via the

shape memory synthesis. The (β-SiC-supported VPO) approved to be a

much better catalyst in terms of (MA) selectivity and yield than its bulk

equivalent, in fixed-bed configuration of reactor. The ability of this

ceramic to well disperse (VPO) was explained by the presence of an inter

phase (a glue) containing the (V, P, Si, C and O) elements, which did not

interfere with the formation of the active and selective (VPO) phase from

the (VOHPO4.0.5H2O).The real nature of this inter phase still needs to be

clarified. The high (MA) selectivity obtained on (VPO/β-SiC) was

attributed to both the absence of microporosity on the support and more

importantly, to the heat sink role of this support avoiding the formation of

hot spots on the catalytic sites and protecting the product, MA ,from

further oxidation into CO and CO2.

The uses of other exotic heat conductive materials as support such as

(Si3N4) and (BN) ,proves the validity of the concept of heat transfer from


64

the surface to the bulk of the catalyst, without taking into account the

macroscopic transfer of heat outside the reactor (165).

It was finally demonstrated, from an industrial point of view ,that the

presence of the (β-SiC) support could provide high volume yield of (MA/

volume of catalyst) in a fixed-bed reactor because it was possible to work

with relatively high concentrations of n-butane ,contrary to what is

normally possible with the conventional bulk catalyst .In addition because

of the good dispersion, a large fraction of the expensive unused internal

part of the bulk active phase was replaced by cheap and light (β-SiC)

support .The optimal amount of active phase was found at (30wt %) of the

total composite. Finally, also demonstrated was the protective role of the

support on the thermal stability of the reactive phase which was much

more resistant to overheating accidents than the bulk material.

This new concept of heat control is currently being studied and used

for other similar problems where nonoxidic conductive materials are

potentially the answer to the difficulties that have been encountered in

many applications.

1.12.2- Molecular structure in dehydrated state

The moisture adsorbed in supported vanadia catalysts under ambient

conditions may be removed by heating supported catalysts in non reducing

oxygen containing atmosphere at elevated temperature, (573-973 K). Such

treatment also helps maintain the (5+) oxidation state of the surface

vanadia species. At temperatures below (473 K), the Raman stretch of the

terminal (V=O) bond occurring at (1020-1040 cm-1) is shifted significantly

downward due to extensive hydration of the surface vanadia. At

temperatures above (473 K) a small amount of moisture hydrogen bonded

to surface vanadia results in a few cm-1 downward shift of the Raman


65

(V=O) band. At these temperatures (18O) labeling experiments showed that

the reversibly adsorbed moisture is also capable of rapidly undergoing

(18O) exchange with the terminal (V=O) bonds.

The structural studies suggested that the dehydrated vanadia species

on (Al2O3,TiO2, ZrO2, Nb2O5 and CeO2) was predominantly present as

isolated and polymeric tetrahedral VO4 units (Figure 1.24).The (18O)

labeling experiments of supported (VOx/ Al2O3, TiO2, ZrO2) showed that

these surface species possessed only one (terminal V = O) bond and three

bridging (Vـ�O-support) bonds for the isolated species, while the polymeric

surface species has only one (Vـ�O-support) bond per vanadium atom ,the

other two are being bridging V-O-V bonds (Figure 1.24). (166)

Spectroscopic techniques (X-ray absorption, Raman and UV-Vis.)

were utilized to monitor the effect of dehydration for vanadia supported on

(SiO2, Al2O3, TiO2, ZrO2 and HfO2) prepared with (VOx) surface densities

ranging from (0.46- 11.1 VOx/nm2) (166). UV-Vis. and near –edge spectra

indicate that supported (VOx) species exist in larger domains and have

higher – coordinate centers when hydrated. Dehydration consistently leads

to a breakup of these domains. Raman spectra suggest that (V=O) bonds in

mono and polyvanadate species are most susceptible to hydration. On all

supports, dehydration leads to the development of monovanadate species.

Polyvanadate on (TiO2, ZrO2, and HfO2) also undergo structural changes

when hydrated. Interpretation of this is more difficult because of the broad

Raman bands which allow only qualitative assignments. Support material

plays an important role in determining the extent of hydration, but this role

might arise from the ability to support polymeric vanadia species.


66

Figure 1.24: Molecular structure of:

(A) Dehydrated isolated

(B) Polymerized vanadia species on oxide supports (166).

1.12.3- Structural changes during Hydrocarbon Oxidation

Wachs et al (161) investigated the oxidation of n-butane to (MA) over a

series of model-supported vanadia catalysts where the vanadia phase was

present as a two- dimensional metal oxide overlayer on the different oxide

supports ( TiO2, ZrO2, CeO2, Nb2O5, Al2O3, and SiO2). No correlation was

found between the properties of the terminal (V = O) bond and the butane

oxidation turnover frequency (TOF) which is mean the number molecules

converted per vanadium per second , during in situ Raman spectroscopy

study. Furthermore, neither the n-butane oxidation (TOF) nor (MA)

selectivity was related to the extent of reduction of the surface vanadia

species. The n-butane oxidation (TOF) was essentially independent of the

surface vanadia coverage, suggesting that the n-butane activation requires

only one surface vanadia site. The (MA) TOF, however, increased by a

factor of (2–3) as the surface vanadia coverage was increased to monolayer

coverage. The higher (MA) (TOF) at near monolayer coverages suggests

that a pair of adjacent vanadia sites may efficiently oxidize n-butane to

Maleic Anhydride. Other factors may also play a contributing role such as

(increase in surface Bronsted acidity and decrease in the number of


67

exposed support cation sites. Varying the specific oxide support changed

the n- butane oxidation (TOF) by 50:

Ti > Ce > Zr ~ Nb > Al > Si

As well as the (MA) selectivity.

The (MA) selectivity closely followed the Lewis acid strength of the

oxide support cations:

Al > Nb > Ti > Si > Zr > Ce

The addition of acidic surface metal oxides (W, Nb, and P) to the surface

vanadia layer was found to have a beneficial effect on the n- butane

oxidation (TOF) and the (MA). The creation of bridging (V–O–P) bonds

had an especially positive effect on the (MA) selectivity .

1.12.4–Structure-Activity Relationships in Oxidation of

C4-hydrocarbons

Several studies on C4-hydrocarbon oxidation have been reported in the

literature (102,106,160).

Nakamura et al. (105) studied oxidation of 1-butene on alumina-

supported catalysts and reported reasonably high selectivities to (MA). The

results of their study showed that supported (VPO) catalysts could

selectively oxidize C4- hydrocarbons to (MA). They concluded that the

activity and selectivity of their supported catalysts for the oxidation of 1-

butene to (MA) were closely related to the oxidation state and the degree

of aggregation of vanadium ions. They implicated the (V4+/V5+) redox

mechanism of oxidation in which (V = O) bonds played an important role.

Mori et al. (102) studied oxidation of n-butane on titania-supported

vanadia catalysts at high temperature (693-763 K) and high n-butane

conversion. Such severe experimental conditions led to an over oxidation

of n-butane to (COx) and, likely, to combustion of the partial oxidation

products, including (MA). They found that the reaction rate under such


68

conditions was proportional to the amount of (V 5+= O) species in the

catalyst and concluded that the reaction proceeded via the reduction-

oxidation or Mars-van Krevelen mechanism according to which the

surface (V = O) species contained the active oxygen.

Busca et al. (167) employed titania-supported vanadia catalysts in n-

butane oxidation and observed over oxidation of the hydrocarbon on very

active high surface area catalyst (117m2/g) at (723 K). When they

supported (10 wt% V2O5) on the low surface area Titania (18.4 m2/g),

vanadia formed micro crystals which were detrimental to selective

oxidation and led to n-butane combustion.

Owens and Kung (160) studied n-butane dehydrogenation on silica-

supported vanadia catalysts at (793) K. They observed that the isolated

vanadia species present at low surface coverage (0.53 and 0.58 wt% V)

were responsible for the high total dehydrogenation selectivity (1-butene,

cis-/trans-2-butene, and 1, 3-butadiene as shown in (Figure 1.25) (160).

1 2 3 4

Figure 1.25: Oxidation of butane on dispersed vanadia species (160).

The presence of the crystalline V2O5 species at high surface coverage

(998, 703, 526, 480, 404, 304 and 284 cm-1) Raman bands for the (6.4 wt%

V) catalyst contributed to the production of total oxidation products.

C4H9


69

1.12.5- Role of Surface Oxygen Species in Supported Catalysts

The terminal (V = O)oxygen has been proposed by investigators (161,162 )

to be the active oxygen involved in hydrocarbon oxidation over supported

vanadia catalysts. However, the combination of (in situ Raman and

hydrocarbon oxidation reactivity studies) have recently suggested that the

reaction properties are not related to the characteristics of the terminal

(V = O) bonds in supported vanadia catalysts. Butane oxidation over a

series of supported vanadia catalysts was found to vary by over an order of

magnitude in the butane turn over frequency (TOF) (Table1.4) (161), but

identical (V=O) Raman features were observed for these catalysts (V=O

stretch at (1025-1032cm-1) (Figure 1.26 A-D ) .

Table 1.4: the effect of the metal oxide support on n-Butane oxidation on

supported Vanadia Catalysts at 494 K in 1.2 Vol. % n-Butane (161).

Catalyst Wt.

(g)

Flow

cm3/min.

Butane

Conversion

mol%

MA

Selectivity

mol%

Butane

(TOF)

10-5 s-1

MA

(TOF)

10-5 s-1

7% V2O5/SiO2 0.577 7.4 2.8 91.8 0.4 0.3

17.5% V2O5/Al 2O3 0.625 17.8 7.2 44.5 0.9 0.4

6% V2O5/Nb2O5 0.885 13.9 17.3 36.7 3.6 1.3

4% V2O5/ZrO2 0.794 11.3 16.0 9.3 4.5 0.4

3% V2O5/CeO2 0.622 14.2 10.6 12.6 6.3 0.8

5% V2O5/TiO2 0.566 25.5 27.8 30.5 19.6 6.0


70

Figure 1.26 A- In situ Raman spectra of 1% V2O5/TiO 2 catalyst during n-butane

oxidation (a) O2100cm3/min.,503K; (b) C4H10/O2/He, 100cm3

/min.,503K;(c)

C4H10/O2/He, 100cm3/min.573 K ; (d) C4H10/O2/He, 100cm3

/min. 623 K; (e)

C4H10/O2/He, 50cm3/min. 623 K; (f) O2 623K (161).

Figure 1.26 B- In situ Raman spectra of 7% V2O5/SiO2 catalyst during n-butane

oxidation (a) O2100cm3/min.,503K; (b) C4H10/O2/He,100cm3

/min.,503K;(c)

C4H10/O2/He, 100cm3/min. 573 K ; (d) C4H10/O2/He, 100cm3

/min. 623 K; (e)

C4H10/O2/He, 50cm3/min. 623 K; (f) O2 623K (161).


71

Figure 1.26 C- In situ Raman spectra of 7% V2O5/Nb2O5 catalyst during n-butane


/min.,503K;(c)


/min. 623 K; (e)

C4H10/O2/He, 50cm3/min. 623 K; (f) O2 623K (161) .

Figure 1.26 D- In situ Raman spectra of 4% V2O5/ZrO 2 catalyst during n-butane


/min.,503K;(c)


/min. 623 K; (e)

C4H10/O2/He, 50cm3/min. 623 K; (f) O2 623K (161) .


72

Furthermore, 18 O – labeling of the terminal (V = O) bond, during the

n-butane oxidation revealed that this bond is very stable and has an

exchange time that is approximately (20) times longer than the

characteristic reaction time. The (18O-labeled 4 wt% V2O5/ZrO2) catalyst

exhibited the (V = 18O) Raman stretch at (983 cm-1) which disappeared

only after (25min.) of n-butane oxidation (Figure 1.26-d (161)). Therefore,

the available data suggests that the terminal (V = O) oxygen is not involved

in kinetically significant reaction steps of hydrocarbon oxidation over

supported vanadia catalysts.

The surface concentration of bridging (V-O-V) bonds increases with

surface vanadia coverage due to the increase in the ratio of polymerized to

isolated surface vanadia species (Figure 1.24). The (V2O5/SiO2) system

represents one notable exception to this rule, since microcrystalline V2O5 is

formed at above monolayer coverage (Figure 1.26-b) (161). The (TOF) for

the oxidation of butane to (MA) over (V2O5/TiO2) was found to slightly

increase, by a factor of (2-3), with surface vanadia coverage (Figure 1.27)

because of the requirement of several surface vanadia sites for this

hydrocarbon.

Figure 1.27: performance of Titania support in n-butane oxidation on supported

vanadia catalyst at 494 K in 1.2 vol. % n-butane in air (161).


73

During n-butane oxidation to (MA) the Raman signal of the surface

V(V) species decreased (10-35%) reflecting partial reduction of the

vanadia species under reaction conditions (V =O and V-O-V stretches)(163).

The average oxidation state of vanadium in the selective (VOx /TiO2)

catalyst for partial oxidation of n-butane was found to be near (+ 4.5). The

surface vanadia coverage was also found to be a critical variable as the

polymeric surface vanadia species present at high coverage are more easily

reducible than the isolated vanadia species. The reducibility of the surface

vanadia species during butane oxidation on various oxide supports

followed the order:

TiO2 >CeO2 >ZrO2 >Al2O3 >SiO2

However, the butane oxidation (TOF) was not found to correlate with the

extent of reduction of the surface vanadia species (Table 1.5).

Table 1.5: The effect of acidic promoters on n-Butane Oxidation on the

1% V2O5 /TiO2 catalyst at 494 K in n-Butane in air (161).

Catalyst Wt.

(g)

Flow

cm3/min

Butane

Conversion

mol%

MA

Selectivity

mol%

Butane

(TOF)

10-5 s-1

MA

(TOF)

10-5 s-1

1% V2O5/5%P2O5/ TiO2 0.172 5.0 12.1 56.2 27.0 15.2

6% WO3/1%V2O5/ TiO2 0.708 8.9 23.6 26.2 34.1 8.9

9% WO3/1%V2O5/ TiO2 0.465 11.1 14.8 6.8 40.5 2.8

6% Nb2O5/1%V2O5/ TiO2 0.181 7.5 10.7 35.1 50.8 17.8

The (MA) selectivity did not appear to directly correlate with the extent of

reduction either, since the selectivity pattern was:

Al 2O3 >Nb2O5 >TiO2 >SiO2 >ZrO2 ~ CeO2

Varying the specific oxide support or oxide support ligands without

hanging the structure of the surface vanadia species can alter the

characteristics of the bridging (V-O-support) bond. The bridging (V-O-


74

support) bond appears to be associated with the critical oxygen required

for hydrocarbon oxidation reactions since changing the specific oxide

support dramatically affects the (TOF) approximately two orders of

magnitude for n-butane oxidation to (MA) (Table1.4). The general trend

appears to be:

CeO2 >ZrO2 ~TiO2 >Nb2O5 >Al2O3 >SiO2

This suggests that bridging oxygens in the (V-O-support bonds) that are

more electronegative or basic, corresponding to oxide support cations with

lower electronegativity, are associated with the critical oxygen required

for hydrocarbon oxidation reactions over supported vanadia catalysts. The

formation of the (V-O-P) bond has a particularly positive effect on the

butane oxidation (TOF) and (MA) selectivity (Table 1.6). The

(1%V2O5/5% P2O5/TiO2) catalyst displayed the highest selectivity to (MA)

among all systems studied (Table 1.5). At (494 K in 1.2 vol. % n-butane in

air) and (12.1 mol% n-butane conversion), the selectivity to (MA) reached

(56.2 mol %). Such catalytic behavior is consistent with the above

observation that (bridging V-O) support bonds are critical in the oxidation

of n-butane to (MA).

The insight into the number of critical surface vanadia sites required in

hydrocarbon oxidation reactions can be gained by examination of the

variation of the (TOF) with surface vanadia coverage. Those reactions that

require only one surface site will exhibit a (TOF) that is independent of

surface vanadia coverage .The oxidation of butane to (MA) over titania-

supported vanadia catalysts (Table 1.6 and Figure 1.27) exhibited an

increase in (TOF) with surface vanadia coverage. This may reflect the

requirement of multiple surface vanadia sites or the influence of other

factors, such as surface acidity influence of (bridging V-O-V bonds),

structural changes, etc...


75

Table 1.6: Performance of Titania support in n-Butane oxidation at 494 K in 1.2

Vol. % n-butane in air (161).

Catalyst Wt.

(g)

Flow

m3/min

Butane

Conversion

mol %

MA

Selectivity

mol %

Butane

(TOF)

10-5 s-1

MA

(TOF)

10-5 s-1

1%V2O5/TiO2 0.890 8.7 16.0 22.8 12.4 2.8

2% V2O5/TiO2 0.612 17.6 24.2 9.9 27.7 2.7

3% V2O5/TiO2 0.603 19.9 27.0 17.0 23.4 4.0

4% V2O5/TiO2 0.700 30.7 30.7 29.9 25.5 6.6

5% V2O5/TiO2 0.566 0.566 27.8 30.5 19.6 6.0

7% V2O5/TiO2 0.572 0572 18.5 24.7 8.0 2.0

Information about the number of surface sites required for hydrocarbon

oxidation reactions can be also probed by the addition of non interacting

surface metal oxides, which preferentially coordinate with the oxide

support rather than the surface vanadia species under dehydrated

conditions .Typical noniteracting oxides are surface oxides of W, Nb, S,

Si, Mo, Ni, Co and Fe .Noniteracting metal oxides only indirectly affect

the molecular structure of the surface vanadia species via lateral

interactions .Such lateral interactions influence the relative concentration

of polymerized and isolated vanadia species in supported metal oxides.

Interacting acidic oxides, such as P2O5 significantly increase the (TOF) for

butane oxidation to (MA) (161). The enhancement in n-butane oxidation

(TOF) and (MA) selectivity upon introduction of acidic oxides further

confirms the positive role of acidity in this reaction. The butane oxidation

(TOF) was increased by a factor of (2 and 3) when acidic metal oxides,

such as WO3 and Nb2O5, were introduced to the (V2O5/TiO2) system (Table

1.5). For example, introduction of surface niobia species to (V2O5/TiO2)

catalysts led to a threefold increase in the (TOF) for the oxidation of

butane to (MA) (161). This increase in (TOF) also reflects the requirement of


76

multiple surface metal oxide sites for this hydrocarbon oxidation. In

contrast to non interacting additives that mainly affect oxidation reactions

requiring multiple surface sites, interacting additives affect all hydrocarbon

oxidation reactions since they directly alter the structure and reactivity of

the surface vanadia sites. Interacting metal oxides, especially basic oxides,

retard the reduction of surface vanadia species. Consequently, the (TOF)

for all hydrocarbon oxidation reactions are lowered when basic additives

are introduced (163,164).

1.12.6 - Role of Acidity of Supported Vanadia Catalysts

The oxide supports possess only surface Lewis acid sites, with the

exception of silica where no Lewis acid sites exist on the surface. The

relative Lewis acid strength of these sites follows the order

Al 2O3 >Nb2O5 >TiO2 >ZrO2

In contrast to the oxide supports, unsupported V2O5 crystalline powders

displays both surface Bronsted and Lewis acid sites (167) and an increase in

the number of surface Bronsted acid sites.

Supported (VPO) catalysts typically exhibit surface enrichment in

phosphate similar to the conventional unsupported or bulk (VPO)

catalysts (106,156,161). In these catalysts the concentration of the surface

Bronsted acid sites increases with the surface (P/V) ratio. The surface

enrichment in phosphorus stabilizes the V (IV) oxidation state, which

results in decreased activity in n-butane oxidation. On the other hand, the

increased surface Bronsted acidity facilitates desorption of partially

oxidized products with acidic properties, such as (MA) leading to

significant improvement in (MA) selectivity (162).According to Grasselli st

al. (168), slight excess phosphate (P/V = 1.1-1.2) forms a protective

pyrophosphate "fence" around active surface sites at the surface preventing

overoxidation of the reactive intermediates by the active oxygen diffusing


77

fast at the surface. However, the changes in the activity and selectivity of

the (VPO) catalysts at much higher phosphorus enrichment (P/V ratio near

2.0) were associated with formation of microcrystalline VO (PO3)2 (169).

Reports in the literature have demonstrated that the VO (PO3)2phase is

inactive in n-butane oxidation (169). Therefore, the catalytic behavior of the

VO (PO3)2 system is explained by the presence of other (VPO) impurity

phases, such as (VO)2P2O7 and various (VOPO4) phases (169). The oxidation

of n-butane on well-defined supported vanadia catalysts demonstrated that

the substitution of (V-O-Ti) for less reducible and more acidic (V-O-P)

bonds has a positive effect on the n-butane oxidation (TOF) and (MA)

selectivity . This observation indicates that the oxidation of n-butane

to (MA) requires both a redox site and some acidic functionality (170 ) .

1.12.7: The Role of Support

We can conclude that although supports are essentially diluents, they

play important multifunctional roles which can be summarized as

following:

1- Economic, to reduce cost by extending expensive phase in catalyst.

2- Mechanical, to give mechanical strength, to optimize bulk density, to

prove a heat sink or a heat source, and to dilute the overactive phase.

3- Geometric, to increase the surface area, to optimize the porosity of a

catalyst, to optimize crystal and particle size, to allow the catalyst

particles adopt the most favorable configuration of a catalyst.

4- Chemical, to react with the catalyst either to improve specific activity

or to minimize sintering and to accept or donate chemical entities,

porosity via a spillover mechanism.

5- Deactivation, to stabilize the catalyst against sintering and to minimize

poisoning.


78

Conclusions

Significant progress had been achieved in recent years in studying

molecular structures of the surface vanadia species present in supported

metal oxide catalysts. The detailed structural information on well-defined

systems provided a foundation for developing structure-reactivity

relationships required to molecularly engineer supported vanadia catalysts

for oxidation reactions. The nature of the metal oxide support was found to

play a crucial role in defining catalytic properties of vanadia monolayer in

n-butane oxidation to (MA). The terminal vanadyl oxygen does not appear

to critically influence the reactivity properties of the surface vanadia

species during hydrocarbon oxidation reactions. The bridging (V-O-V)

oxygen plays only a minor role in enhancing the n-butane oxidation (TOF),

primarily due to the preference of multiple active sites in this oxidation

reaction. The bridging (V-O-support oxygen), however, appears to be the

most critical oxygen since its properties can change the (TOF) for

hydrocarbon oxidation reactions by as much as four orders of magnitude.

The specific phase of the oxide support as well as the specific preparation

method does not appear to influence the molecular structure or reactivity

of the surface vanadia species. The number of surface vanadia sites

required for a hydrocarbon oxidation reaction is dependent on the specific

reactant molecule. Oxidation reactions requiring only one surface vanadia

site are generally not sensitive to the surface vanadia coverage and the

presence of non interacting metal oxides. Oxidation reactions requiring

multiple surface vanadia sites are very sensitive to surface vanadia

coverage and the presence of non interacting metal oxides. However,

interacting metal oxides influence all hydrocarbon oxidation reactions

since they modify the surface vanadia sites. Acidic and basic metal oxides

also influence the selectivity of hydrocarbon oxidation reactions, but the

effect appears to be reaction-specific.


79

Unlike bulk (VPO) catalysts, detailed surface structural information on

a molecular level can be obtained from model supported vanadia catalysts

containing two-dimensional over layers of surface vanadia species (172).

Raman spectroscopy provides direct fundamental surface information

about:

1- The ratio of isolated and polymerized surface vanadia species

(Fig.1.24).

2- Terminal (V=O) and bridging (V-O-V) bonds.

3- Extent of reduction of the surface vanadia species during catalysis.

4- Influence of the oxide support ligands.

5- Influence of (acidic/basic) metal oxide additives (promoters/poisons).

6- Participation of specific (V- O) bonds in catalysis (with the aid of

oxygen- 18O labeled isotope experiments.

Thus, the in situ Raman studies during oxidation of reactions over model

supported vanadia catalysts can provide new insights into the surface

properties of oxide catalysts which are not attainable with bulk metal oxide

catalysts.

The studies of the model supported catalysts provided several important

insights about the origins of the catalytic activity of the (VPO) system that

were not possible with the bulk (VPO) catalysts. Firstly, it was shown that

the (V=O) oxygen was not involved in n-butane oxidation to (MA) and

that the (V-O-support) bonds contained active oxygen for n-butane

oxidation. The (V-O-P) bonds were particularly beneficial for both the

activity and selectivity of the supported vanadia catalysts (Table 1.6 (163)),

suggesting that this oxygen species may be responsible for selective

oxidation of n-butane. Secondly, it was demonstrated that the oxidation of

n-butane to (MA) could occur at a single vanadia site, although adjacent

sites were more efficient (Figure 1.27).


80

The results of n-butane oxidation on titania-supported vanadia

catalysts suggested that isolated surface vanadia species are capable of

n-butane oxidation to (MA), although multiple vanadia sites are more

efficient in this oxidation. Microcrystalline vanadia was found to be

detrimental for the process of (MA) formation. The kinetic studies of the

supported vanadia system provided more direct evidence that the multiple

vanadia sites were better at oxidizing n-butane to (MA).

The selectivity of the supported catalysts to (MA) is correlated with

the Lewis acid strength of the metal oxide promoters. Especially high

selectivity to (MA) was found when the (V-O-P) bonds formed after

addition of phosphorus oxide in accordance with previous observations in

supported and unsupported bulk (VPO) catalysts. These findings indicate

that the supported vanadia catalysts represent a suitable model system

capable of providing insights into the mechanism of n-butane oxidation on

bulk (VPO) catalysts.

Lastly, the use of acidic additives with supported (VPO) catalysts

further demonstrated their similarity to the bulk (VPO) system. Moreover,

it shed some light on the possible role of promoters that improve the

performance of the bulk (VPO) catalysts. The acidic additives used in the

supported vanadia catalysts enhance those catalytic oxidations which

require a combination of a redox and acidic site for selective catalysis. The

acidic additives promoted both the rate of n-butane oxidation and the (MA)

formation on supported vanadia catalysts (Table 1.6).

The supported (VPO) system proved to be a good model for the bulk,

i .e. unsupported (VPO) system. The in situ Raman studies of the

supported vanadia catalysts provided several important insights into the

nature of the selective n-butane oxidation that were not possible with

unsupported vanadium-phosphorus-oxygen (VPO) catalysts due to the

experimental limitations. The many discovered similarities between the


81

bulk and supported (VPO) systems have important implications for the

design of new selective (VPO) catalysts. The nature of the oxide support,

the surface coverage of the active vanadia species and the acidity of metal

oxide additive are the most important determinants of the catalytic activity

and selectivity in n-butane oxidation. These structure-reactivity

relationships pave the way for molecular engineering of selective active

sites for hydrocarbon oxidation in supported (VPO) catalysts.

The aim of work

The aim of our previous work was precisely to use technical grade

vanadium pentoxide locally produced to prepare a catalyst to convert

n-butane to MA.

Our aim in this work is to study the effect of Titania (anatase) on the

physical, chemical and structural properties of our previous already

tested catalyst in order to prepare an industrial catalyst to increase the

feasibility of vanadium pentoxide production in our country and supply

further prospects to use it.

Chapter Two - Experimental

83

2.1: Samples Analysis

Ten techniques are used for analysis were employed in this work.

Eight of them were carried in international laboratories in four countries

(United States of America, Germany, United Arab Emirates, and

Ukraine) and two of them were performed locally as shown below:

1-The determination of vanadium, nickel, iron, sodium, calcium,

potassium, magnesium, molybdenum, silicon and aluminum were

carried out by atomic absorption spectrophotometer Model AA-680

Shimadzu, vanadium, sodium, silicon, aluminum, molybdenum and

calcium were determined in an N2O- acetylene flame at (318.4) nm,

(589.0) nm, (251.6) nm, (309.3) nm, (313.3) nm, and (422.7) nm

Principle lines. Nickel, iron, potassium, magnesium were determined in

Air- Acetylene flame at (232.0) nm, (248.3) nm, (766.5) nm, and

(285.2) nm Principle lines respectively. This measurement was carried

out in Ibn Sena State Company-Baghdad- Iraq.

2-The qualitative identification of the samples were performed on

STOE STADIP X-ray diffractometer with Cu Kα-radiation in Fritz-

Haber Institute, Berlin, Germany.

3-Thermogravimetric analysis (TG) and differential thermo-

gravimetric analysis (DTA) were carried out by MOM, Hungary, Q-

derivatograph , system F.Paulik, J.Paulik, L.Erday , in flowing air at

(10°C/min.). Determinations were carried in Institute for Sorption and

Problems of Endoecology National Academy of Sciences of Ukraine,

Ukraine


84

4- Fourier Transform Infra - Red analysis using Fluka, technical

gradeV2O5, and prepared phases were run using KBr disk by FT-IR-

8300 Fourier transform infrared spectrophotometer, Shimadzu between

4000-400 cm-1 in Al-Nahrain University-Collage of Science-Chemistry

Department- Baghdad – Iraq.

5- Laser Raman spectra were obtained in Jovin Yvon – Horiba Lab

Raman equipped with a CCD detector and with three different

excitation sources having wavelengths using (He-Ne laser) to give

632.8 nm (Red laser) and 514 nm (green laser). Backscattering

geometry of the micro-Raman technique (X 50) was used to record

spectra of the powder samples. Typical laser powers ranged from 3.1 to

5.0 mW was used; integration time was around 30 s for each spectrum;

and up to 10 Raman spectra were averaged for each sample. The

analysis was carried out in School of Chemical Engineering and

Materials Science- University of Oklahoma-U.S.A.

6- Specific Surface Area was determined by method of argon

temperature desorption on measured by GASOCHROM-1 equipment

(Mosneftekip).The mixture containing 8 vol. % of argon in helium (Ar

in He) was used. The standard – Al2O3 was with SSA = 4.2 m2/g (BET

measurement). The obtained value of specific surface area was the

average value from 5 measurements for one sample. The determination

error is 4-5 %.The specific surface area was calculated according to the

following equation:

SSA (sample) = (SAr sample/SAr standard) X SSA st X (g st / g sample)


85

This analysis was performed in Institute for Sorption and Problems of

Endoecology National Academy of Sciences of Ukraine, Ukraine.

7- The composition and the oxidation state of the surface of prepared

samples were obtained by X- Ray Photo Electron Spectroscopy (XPS).

Data were recorded on a physical electronics PHI 5800 ESCA system

with a background pressure of an approximately 2.0 X 10-9 Torr. The

electron take- off angle was 45° with respect to the sample surface. An

800-µm spot size and 23 eV pass energy were typically used for the

analysis. The binding energies were corrected by reference to the

carbon 1s line C/s line at 284.8 eV for hydrocarbon. A nonlinear

Shirley-type background was used for the area analysis of each peak.

The fitting of the XPS spectra was carried out with asymmetric peaks,

using the MulitPak software from Physical Electronics in School of

Chemical Engineering and Materials Science - University of Oklahoma-

U.S.A.

8- The surface acidic properties of the samples were studied by

adsorption of pyridine (py) and 2, 6 Dimethyl pyridine (dmpy). Using

(CHROM-5) chromatograph with flame ionization detector (FID) was

used. The sample (0.5 ml with specified weight) was loaded in

chromatographic column (short column with l=15 cm).The Argon

(carrier gas) was bubbled in special glass pot with “py” or “dmpy” and

was introduced by a valve pulses on column at (T=100 °C).The

difference between initial peak without sample and that passed through

sample of (“py”) or (“dmpy”) was determining the amount of adsorbed

in case in pulses the outlet peak was equal to initial peak the full


86

quantity of adsorbed base was determined .It was known that “py”

adsorbed on Lewis (LA) and Bronsted (BA) acid centers (total acidity

but “dmpy” adsorbed only on (BA) centers. The amount of Lewis acid

centers could be determined by difference (A py – A dmpy). This

measurement was carried out in Institute for Sorption and Problems of

Endoecology National Academy of Sciences of Ukraine.

9- The morphology of the surface of the prepared samples was

examined by Scanning Electron micrographs (SEM) using Scanning

Electron Microscope type JEOL HSM- 6400 (0.2- 45 KV accelerating

voltage , X 10- 300000 magnification ), which is a high resolution

Scanning Electron Microscope with a modern digital image processing

systems. The samples were coated with a thin layer of gold using

coating unit to prevent electrostatic charging on the surface of the

sample. This test was made by School of Mechanical Engineering -

American University of Sharjah – U.A.E.

10- A Hitachi S-4700 scanning Electron Microscope (SEM) equipped

with operating at 25 Kv on energy dispersive X-ray (EDX)

spectrometer (YAG) was used for examination of surface morphology

and composition, in Institute for Sorption and Problems of Endoecology

National Academy of Sciences of Ukraine, Ukraine.


٨٧

2.2 - Chemicals:

The chemicals used in this work are listed in Table 2.1 below

Table 2.1: Chemicals used in this work

No. Chemical Materials Molecular formula Purity % Co.

1 Acetone (CH3)2 CO 99 Merck

2 Benzyl alcohol C6H5 CH2OH 99 BDH

3 Diphenyl amine (C6H5)2 NH 99 Fluka

4 Isobutyl alcohol C4H10OH 99 Fluka

5 Iron (II) ammonium sulfate Fe (SO4)2 (NH4)2. 6H2O 99 Fluka

6 Oxalic acid C2O4H2 >99.99 BDH

7 o-Phosphoric acid O-H3PO4 89-92 BDH

8 Potassium bromide KBr 99 Merck

9 Potassium permanganate KMnO4 99 BDH

10 Sulfuric acid H2SO4 >95.97 Fluka

11 Titanium dioxide (anatase) TiO2 > 99 Fluka

12 Vanadium pentoxide V2O5 >99.99 Fluka

13 Vanadium pentoxide V2O5 (Technical Grade) 93.87

Ibn-Sena

State Co.

Iraq


٨٨

2.3: Preparation:

Two preparation routs are used in this work in order to prepare catalyst

precursor phases (Scheme 2 – 1). The (P/V) ratio used in all preparation

routes is mole ratio.

2.3.1: Route (A):

Route (A) is designated to prepare supported and unsupported of two

phases: intermediate phase (VOPO4.2H2O P/V = 1.1) and its derivative

(VO (H2PO4)2 P/V = 1.1).The intermediate phase was prepared and

isolated in order to study the correlation between these two phases as

separate phase. The reduction step is carried out after the preparation of

intermediate

2.3.1.1: Intermediate phase (VOPO4.2H2O P/V = 1.1)

2.3.1.1.1: Sample No.1 – un (unsupported)

According to the method which had been conducted for the first time in

our previous work (1), the quantity of (24) g of locally produced (technical

grade) V2O5 was suspended by rapid stirring in (96) ml distilled Water and

(17) ml of O- H3PO4 in order to obtain a molar ratio (P/V = 1.1). The

mixture was refluxed for 16 hrs. The color of mixture changed form brown

to yellow during (30) min. The resulting yellow precipitate was filtered ,

washed with distill water until a red color appeared in washing water, then

washed with acetone for several times and left to dry.

Several attempts were unsuccessfully exerted to prepare this phase with

molar ratio (P/V =1.1) from vanadium pentoxide (laboratory grade)


٨٩

produced by both Fluka and BDH Companies; the reaction did not proceed

even after 48 hours of continuous reflux.

2.3.1.1.2: Sample No. 1 – supp (supported)

Technical grade V2O5 (24 g) was mixed and ground with (1.92) g of

Titania (anatase) in order to obtain (TiO2 /V2O5 = 8% weight ratio). The

same procedure (as described in above section 2.3.1.1.1) was conducted to

prepare this supported phase.

2.3.1.2: Derivative (VO (H2PO4)2)

2.3.1.2.1: Sample No.3 – un (un supported)

To a suspension of (VOPO4.2H2O P/V = 1.1), 24 ml of benzyl alcohol

and (48) ml isobutanol were added and refluxed with stirring for 7 hrs.

During this time the color was changed from yellow to dark green. The

sample was filtered and dried at 125 °C for overnight.

2.3.1.2.2: Sample No. 3 – supp (supported Sample)

Technical grade V2O5 (24 g) was mixed and ground with (1.92) g of

Titania (anatase) in order to obtain (TiO2 /V2O5 = 8% weight ratio).

The same method which described above was conducted to prepare

supported sample with longer time reaction (13) hrs. The resulting

precursor was light green.


٩٠

2.3.2: Route B (VOHPO4.0.5H2O)

The reduction of V (V) step has taken place at the beginning of this

preparation route.

2.3.2.1: Sample No. 2 (Reference Sample)

According to Trifiro (171), V2O5 (Fluka) (24 g) was suspended by rapid

stirring with a mixture of (24 ml) benzyl alcohol and (48) ml isobutyl

alcohol.

The vanadium oxide-alcohol mixture was refluxed for 3 hours at 120

°C. The color of the mixture changed from brown to black. The apparently

reduced vanadium suspension was cooled to 40 °C prior to the addition of

22 ml O- H3PO4 in order to obtain a molar ratio for ( P/V = 1.1) . The

suspension was again heated to 120°C and refluxed for additional 3 hrs.

During this time reduced vanadium (IV) - phosphate was formed as

indicated by a change in color from black to blue. The resulting precursor

was separated by filtration and dried at 125°C for overnight. The

preparation of this sample shown in the Scheme 2-1.

2.3.2.2: Sample No. 2- un (unsupported)

The method described above (Section 2.3.2.1) was followed to prepare this

sample from technical grade V2O5. A green precursor was obtained.


٩١

2.3.2.3: Sample No. 2- supp (Supported)

V2O5 (24) g of technical grade, was mixed and ground with (1.92) g of

Titania (anatase) in order to obtain (TiO2 /V2O5 = 8% weight ratio). The

same procedure was conducted in order to prepare the supported phase (as

described in above section 2.3.2.1).


2

1.1- Background In our previous work (1) different phases of (vanadium- phosphorus-

oxygen) (VPO) catalyst were prepared, characterized and evaluated by

catalytic test in order to produce Maleic anhydride (MA) via selective

oxidation of n-butane.

Maleic anhydride (MA), and its derivatives malic acid and fumaric

acid, are produced with a worldwide capacity of about (2.8 X 10 6) metric

tons per year. While the market demand for (MA), increased from

(0.616 X 10 6) metric tons in 2000 to almost (1.27 X 10 6) metric tons in

2004(2-4).

These multifunctional chemical intermediates have found applications

in almost any field of industrial chemistry. The principal use of (MA) is in

manufacturing of unsaturated polyester resins (UPR) ( 63 %), lubricating

oil additives (10 %) , copolymers (9 %) ,alkenyl succinic anhydride (5 %) ,

malic acid (3 %) ,fumaric acid (2 %) agricultural chemicals (1%),

miscellaneous, including reactive plasticizers , sulfosuccinic acid esters ,

and alkyd resins (% 7). Furthermore, due to its double bond and anhydride

function, (MA) is a versatile intermediate for the production of co-

polymers of (MA), for example, ethylene glycol and vinyl monomer.

Recently, potential new uses of (MA) have been found in its conversion to

(1-4) butanediol and the manufacturing of tetrahydrofuran (THF) and

butyrolactone via hydrogenation (5).

Maleic anhydride (MA) and the two di-acid isomers were first prepared

in the 1830's, but it took about 100 years before commercial manufacture

was performed in 1933. The National Aniline and Chemical Company Inc.

started a process for the production of (MA) based on benzene oxidation

using a vanadium oxide catalyst. Smaller amounts of maleic acid were also

formed as by product in the production of phthalic anhydride. The use of


3

benzene as a feedstock for the production of (MA) was dominant in the

world until the late 1980's (6).

Currently, worldwide production of (MA) is based on the major feed

stocks benzene, butenes and n-butane (1-3,6-14). Most of the capacity is

produced via fixed–bed oxidation of benzene, though benzene is being

displaced by butane as a feed stock (all production in United States is

butane based) (1-3,6-14), because butane is a lower cost and environmentally

more desirable raw material and because butane oxidation produces a

clean product stream, forming mainly (MA) and carbon oxides as shown

below:

C4H10 + 3.5 O2 C4H2O3 + 4 H2O ∆ H = - 1236 KJ/ mole (1.1)

C4H10 + 4.5 O2 4 CO + 5 H2O ∆ H = - 1521 KJ/ mole (1.2)

C4H10 + 6.5 O2 4 CO2 + 5 H2O ∆ H = - 2656 KJ/ mole (1.3)

It is obvious that CO and CO2 are thermodynamically more favored

products. Only kinetic control by a catalyst will enhance the formation of

(MA). In practice, the process is operating at a yield of approximately

(60%) to the desired product. CO and CO2 are the sole carbon containing

by–products in a ratio of about unity. Suppression of the unselective and

very exothermic oxidation to carbon oxides requires sufficient heat transfer

capacities of the reactor. Nonetheless, hot spots are frequently met in (MA)

production plants.

Processes for the production of MA From n-Butane

In general, three different types of process can be distinguished in

commercial production of (MA) from n-butane; fixed-bed processes


4

(Scheme 1.1), fluidized–bed processes (Scheme 1.2) and the re-circulating-

solids process (Scheme 1.3).

The fixed-bed reactor consists of a number of tubes that are packed

with coarse catalyst bodies. The reactants flow through these tubes. As a

result of the obstruction of the gas flow by the catalyst bodies, a pressure

drop across the bed is exhibited. Therefore, pressure has to be applied at

the inlet to ensure an adequate flow rate. The magnitude of the pressure

drop is depending on the flow rate, the length of the catalyst bed and the

size to the catalyst bodies. Since the selective oxidation of n-butane to

(MA) is highly exothermic, fixed-bed reactors must containing a facility to

remove the reaction heat. This can be done in various ways: the bed can be

split into different sections, with provision for cooling the gas between the

sections ,or using a reactor containing a large number of tubes, along

which a cooling gas or the liquid is recirculated. However, hot spots can

occur easily in fixed-bed reactors. These can be prevented by using larger

catalyst bodies, a less active catalyst, or by dilution of the catalyst with an

inert solid (support). In view of the explosion limits and the flammability

of mixtures of n-butane and air, only low concentrations of n- butane can

be applied (2 - 4 %).Furthermore, the gases must be mixed and pre-heated

before entering the reaction zone. In a fixed-bed reactor ,the concentration

of n-butane will decrease when moving to the end of the tube .To maintain

a sufficiently high selectivity at the exit of the reactor a less active catalyst

is installed at the entrance and a very active catalyst at the end of the

reactor (4,5).

A single passed fixed - bed reactor was used in our previous work (1) ,

(1 m long and 0.019 m in diameter) mounted in four heating zones vertical

furnace. The flow of n-butane was controlled before entering the reactor

using gasometer. The total gas flow after leaving the reactor was also

measured and fixed at industrial conditions (space velocity 2000 h -1).The


5

gas mixture (1.5 Vol. % n-butane in air ) was preheated before passing on

the catalyst using inert alumina granules in the upper half of the reactor .

The lower half of the reactor (0.1 L) was filled with (2.3 mm) granules by

(4) temperature indicators controllers (TIC).Maleic produced was passed

through scrubber and was dissolved in water to produce Maleic acid .this

acid was titrated with (0.01 N) NaOH using phenolphthalein as indicator.

Scheme 1.1: Huntsman fixed-bed reactor for MA production (4).

In a fluidized-bed process, reaction gases flow upward through a bed

of catalyst particles .When the force of the gas flow on the catalyst bed is

equal to the weight of the bed, the catalyst bed expands significantly and

the catalyst bodies are brought in continuous motion. Because of this

motion ,better heat transfer characteristics are established and ,hence, hot

spots cannot occur in a fluidized-bed reactor comprise the fact that reaction

gases can be used without pre-mixing and pre-heating before entering the

reactor. Furthermore, higher n-butane concentrations can be used due to a

decreased explosion risk compared to the fixed-bed process (4). The flow

diagram for fluidized-bed (Scheme 1.2) process is comparable to that of

the fixed-bed process (Scheme 1.1).Several different ways have been


6

developed to produce bulk (VPO) catalyst particles of the right size and the

attrition resistance for fluidized-bed purposes (16) .When the catalyst bodies

are too small, they will be blown out of the reactor .Large bodies, on the

other hand, call for extremely high linear gas flow rates in the reactor,

therefore, the particle size of fluidized-bed catalysts usually ranges from

(10-150 microns). Development of fluid-bed butane system has been

reported (17,18).These systems offer the advantage of operation at higher

butane concentration (up to 4%) and thus lower original costs. However,

less selectivity at the higher butane loadings and the intermediate problems

with attrition of (VPO) materials need to be overcome before this process

becomes economically preferred (17,18).

Scheme 1.2: ALMA Fluidized bed reactor for MA (4).

DuPont commercially operates the third type of process in their

recirculation-solids reactor (17,18).In this case, oxidation of n-butane and

regeneration of the catalyst are carried out in two separate reaction zones

(Scheme 1.3) .The selectivity to (MA) is increased, because the oxidation

of n-butane is carried out in absence of oxygen (19,20) . In the first step, n-

butane reacts with lattice oxygen from the catalyst. In this stage, the


7

catalyst is reduced by n-butane resulting in the selective formation of MA,

which is removed in a stripper. Regeneration of the reduced catalyst with

oxygen takes place in the second reactor zone. In principle, the oxidation

state of the catalyst can be controlled optimally in this way (17,18).This

process operates at lower conversion per pass and with higher selectivities

than normally encountered in fixed or fluid bed systems. It also uses a

unique attrition resistance system (19-21).

Scheme 1.3: DUPONT re-circulating – solids process for the production of THF

from n-butane via MA (4).

Table 1.1 shows the 1993 and 1995 world-wide (MA) production

capacity for the different processes (4).As can be seen from the table, both

fixed-bed and fluidized-bed butane based processes have been growing at

the expense of benzene-based processes. Furthermore, it seems that

production of (MA) with fluidized –bed technology would not surpass

production with fixed-bed technology which was still operated at higher

yields to (MA). Table 1.2 represents the long – term (MA) production –

consumption balance (4) .The average of the annual demand growth which

was estimated to be about (4%). By the year 2004, total world –wide

consumption will be increased by about (76.5%) as compared to 1998.


8

Table 1.1: World MA capacity (in metric tons) by reactor type (4).

Reactor (feed) 1993 ١٩٩٥

10 3 t/y % 10 3 t/y %

Fixed - bed

n-Butane 369 42.0 704 51.8

Benzene 325 37.9 388 28.5

Fluidized – bed

n-Butane 127 14.8 217 16.0

Phthalic anhydride co-product 37 4.3 50 3.7

Total 858 100.0 1359 100.0

Table 1.2: Total world production & consumption of MA (X 103 metric tons) (4).

Capacity to produce 1994 1995 1996 1997 1998 2001 2004

From n-Butane 551.3 647.6 657.1 702.6 767.6 1019.2 1088.7

From Benzene 301.0 302.0 300.0 300.0 311.0 289.5 289.5

From PA recovery 23.5 20.5 22.5 24.5 24.5 28.5 28.5

Total 875.8 970.1 979.6 1027.1 1103.1 1337.2 1406.7

Utilization (%) 81.9 79.4 82.1 83.4 83.8 79.6 85.9

Production 717.1 770.4 803.9 856.9 924.8 1064.1 1207.7

Consumption

UPE resins 355.3 372.3 383.4 399.7 416.7 471.0 532.9

BDO Chemicals 27.0 36.1 10.0 54.2 86.8 120.4 145.4

Other Uses 330.2 361.2 380.5 402.5 421.3 472.7 529.4

Total 712.5 769.6 803.9 856.4 924.8 1064.1 1207.7

All catalysts used industrially for the production of (MA) from

n-butane are (VPO) catalyst based system (8,12,22).The numbers in (Tables

1.1 and 1.2) ,together with the fact that the current yield (only amounts to

about 58%) to (MA) of an equilibrated catalyst, indicate that improvement

of the process is of great economic and environmental interest. To this end,


9

several developments can be considered. First, the bulk (VPO) catalyst

should exhibit a higher attrition resistance (mechanical strength) in order to

be more suitable for fluidized-bed process. Secondly, the activation period

for the catalyst should be shortened. This will result in an earlier

achievement of optimum performance. However, the catalyst formulation

could be changed; resulting in better properties and an improved catalytic

performance. Furthermore, there is also a comprehensive demand for a

cheaper and more reproducible preparation procedure for the currently

applied (VPO) catalyst. In any case, these improvements can never be

achieved without thoroughly investigations of the catalytic and structural

properties of the active (VPO) phase.

1.2- Scope of the Literature Survey

The n-butane to (MA) reaction is a fascinating complex system. This

catalytic system performance a (14-electron) oxidation involving the

abstraction of (8 hydrogen atoms) and insertion of (3 oxygen atoms) as

described in (eq. 1.4) (21).

CH3

CH3

+ 3.5 O2

O

O

O + 4 H2O ... (1.4)

Compared to other industrially practiced hydrocarbon selective

oxidation reactions (23-26), it is the most complex one (Table 1.3). It is the

only example of an industrially practiced selective oxidation reaction

involving alkane activation. Knowledge gained through study and

understanding of this system may contribute to advances in alkane

activation in general.


10

Table 1.3: Number of Electrons and Oxygen Molecular Involved in some Principal

Reactions of Industrial Interest in Selective Oxidation (9).

Reaction Electrons

Involved

Moles

Oxygen

+ 0.5 H2O

2 0.5

O + H2O

4 1

C

O

H+ H2O

4 1

O

+ 3 H2O

O

O

12 3

One characteristic of the selective oxidation of n-butane involves series

of oxidation steps (utilizing) different kinds and reactivates of oxygen (27)

as described in (Scheme 1.4). This scheme was built up according to

experimental work which determined the form of selective and non-

selective oxygen in the reaction of n-butane oxidation.

O O

M M

(a)

M

O-

(b)

Split double bond


11

O M O

O

O

O O

O M O

(c) (d) (e)

O M O

OO

Radical Pi adsorbed

Scheme 1.4: Kinds of oxygen adsorbed species (28)

1.3 -Structure of the catalyst

1.3.1- Crystalline (VPO) phases

The (VPO) system is characterized by the facile formation of a number

of crystalline phases of vanadium (III), vanadium (IV) and of vanadium

(V).In the more interesting (P/V) ratio near (1.0), the different crystalline

(α, β) (29-35), polymorphic (α1, α11, γ, δ) ( 29,35-37) or hydrated phases of

(VOPO4) (34-41) have been extensively studied. In general, the vanadyl

orthophosphate crystal structure consists of (VO6) and (PO4) groups

arranged in layers [VOPO4] ∞, held together by long (V-O) bonds or by

hydrogen bonds (Figure 1.1-A). The layered structure leads to rich

intercalation chemistry, with formation of layered solids consisting of

alternating inorganic (42-48) and organic (49-56) layer or the formation of

solvated inorganic intercalation compounds (43).


12

Figure 1.1-A: Structure of VOPO4.2H2O showing infinite layers of PO4

tetrahedral linked to VO6 octahedral (40).

In dihydrated (VOPO4) the layer lattice is built up of neutral (VOPO4)

layers and interlayer water molecules (40, 41) .The vanadium atom lies on a

fourfold axis and is surrounded by six oxygen atoms to give distorted

octahedral .The four equatorial oxygens are provided by four different

phosphate tetrahedral .One of the axial vanadium–oxygen bond is very

short corresponding to a double bond (V=O) as shown in Figure 1.1-B.

Figure 1.1-b: Structure of VOPO4.2H2O showing bond distances (°A) and angles

(degree) (40).


13

The structure of (α-VOPO4) is composed of chains of highly distorted

(VO6) octahedral sharing four oxygen atoms with four different (PO4)

groups (30-35). These groups are arranged to form layers (Figure1.2-a).

α- VOPO4 β- VOPO4

Figure 1.2: Structure of α and β- VOPO4 (34).

A distortion of the (VO6) octahedral occurs along the (c) axis,

generating a short (V=O) bond and a very long (V-O) bond (V----O). Thus

the oxyvanadium units can be approximated as (VO5) pyramids and (PO4)

tetrahedral in its crystal structure, analogous to α-VOPO4 (Figure 1.3).

However, the primitive unit cell of (β-VOPO4) contains twice the number

of such structural groups (Figs.1.2-b&1.3), resulting in a network structure.

The γ–and δ -polymorphic forms of (β-VOPO4) are suggested (36-38) to

contain a different framework, in which pairs of edge-sharing octahedral


14

with trans vanadyl oxygens are alternately unshared or shared with

phosphate tetrahedral. Such pairs do not exist in the (α- and β-VOPO4)

forms. During thermal treatment at high temperature the consecutive

transition has been observed (36).

δ- VOPO4 γ-VOPO4 β-VOPO4 …. 1.5

α-VOPO4

β-VOPO4

Figure 1.3: Comparison between structure of α and β-VOPO4 (34).


15

Vanadium (IV) hydrogen phosphate (57-59) which is the most widely

used precursor of the active phase of n-butane oxidation possesses (21,57-62) a

structure made up of atom arrangements in the (ac) plane which are very

similar to that proposed for γ- VOPO4 (Figure 1.4 and 1.5) .It may be

obtained from the dihydrated V(V) phosphate by reduction with organic

alcohols (57) or from the reduction of V2O5 followed by the addition of

o-H3PO4 (61,62).

(A)

(010) plane

(B)

(100) plane

Figure 1.4: Crystal structure of γ -VOPO4 (36).


16

Figure 1.5: Crystal structure of VOHPO4.0.5 H2O (010) plane (59).

Both the hemi hydrate and tetrahydrate of vanadium hydrogen

phosphate are known (58) and their structures are closely related, showing a

layered structure and pairs of face sharing vanadium (IV) octahedral

hydrogen phosphate (OH) groups are directed into the interlayer space.

The coordination geometry around both vanadium and phosphorus is

similar in the two structures (Figure 1.5 and 1.6) .The possible formation

of the tri-hydrate VOHPO4 has also been reported (63).

Figure 1.6: View of the infinite double chains of (VO6) octahedra and (PO4)

tetrhedra running along the (100) direction in VOHPO4.4H2O (75).


17

The intercalation compounds of vanadium hydrogen phosphate with

organic molecules or inorganic cations and anions such as (V n+) of HPO42-

are reported (23-33,36,37,63-72) .This intercalation chemistry is of particular

importance in the description of the structurally and catalytically related

chemistry of VOHPO4.0.5 H2O and of its derived phases.

Two main effects observed in the preparation of (VOHPO4.0.5 H2O)

may be, in fact, strictly connected to intercalation, properties. First, non

stoichiometry is easily accommodated as evidenced by the preparation of

compounds with (0.9-1.2) P: V ratio without any apparent modification of

structural and morphological properties (62).Second, the preparation

conditions have a pronounced effect on the microstructure, i.e., on the

morphology, solid - state reactivity, and the presence of disorder in the

stacking fold of crystalline planes of its layered structure.

In fact, reduction of the starting V (V) compound may be realized (62) by

using, for example, aqueous HCl or isobutyl alcohol. In both cases, almost

pure vanadyl hydrogen phosphate is obtained, but with different

microstructure (61,62,73). The layers of vanadyl hydrogen phosphate (010)

plane (Figure 1.5) are interconnected in tri dimensional structure by weak

hydrogen bonding of phosphates and of water molecules. The organic

alcohol competes with this effect, reducing the bonding between the planes

and allowing the formation of crystals in which these (010) planes are

predominantly exposed (plate like morphology) (21,56,59,62,74) .This effect, in

addition to the increase in surface area modifies the surface properties due

to a change in the relative ratio of crystalline planes at the surface. The

alcohol also can remain partially intercalated between layers (62) .This

effect induces some local modification of the vanadyl hydrogen phosphate

structure, which can strongly modify its solid-state reactivity (61,73).

Ball et al.(75) proposed that the known mono hydrogen phosphate phases

can be classified in three different structural types, namely type I,


18

(VOHPO4.0.5H2O) ;type II (α-VOHPO4.2H2O); type III, (VOHPO4.4H2O)

and (β-VOHPO4.2H2O) .practically ,the β-dihydrate appears as

intermediate in the thermal treatment of the tetrahydrate from both

thermogravimetric and thermodiffractometric experiments .Moreover , the

structure of (β-VOHPO4.2H2O) is closely related with that of

(VOHPO4.4H2O) (Figure 1.6, 1.7-a and 1.7-b). Both compounds present a

similar arrangement of (VO6) octahedra and (PO4) tetrahedral forming

alternating chains which lie along the (c) direction in the β-dihydrate and

along the a direction in the tetrahydrate and along the (a) direction in the

tetrahydrate. The coordination geometry around both vanadium and

phosphorus atom is similar in both structures. Each phosphate group

contains three oxygen atoms (shared with three different vanadium atoms)

and a hydroxyl group. The (V) atoms show very similar coordination

polyhedra, having a water molecule trans-coordinated to the axial (V=O)

group and a second coordinated water molecule in the equatorial plane of

the (VO6) octahedra. The similarity of these structures suggests that the

dehydration of the tetrahydrate into the dihydrate may precede

topotacitically. However, several ways for the reorganization of the infinite

double chains that lie parallel to the (a) axis (6.379 °A ) of the tetrahydrate

may be imagined to give the interconnected single chains running in the (c)

direction (12.623 °A = 2 X 6.379 °A) of the β-dihydrate . All the possible

models that imagined need to break some bonds (at least 2 or 4 per unit

cell), to rotate some polyhedra and to reconstruct the bonds.

Figure 1.7- a: Projection of the structure of β-VOHPO4. 2H2O along (100) (75).


19

Figure 1.7-b: Projection of the structure of β-VOHPO4. 2H2O along (010) (75).

The transformation of V (IV) and V (V) phases into vanadyl

pyrophosphate is an important step in forming active catalysts, by thermal

treatment (Calcinations) at 400 °C, the VOHPO4.0.5 H2O dehydrates to

(VO)2P2O7 .Alternatively, the vanadyl pyrophosphate may be obtained by

reduction from VOPO4 or by thermal treatment of different crystalline or

amorphous (V-P-O-X) phases (where X indicates a generic thermally

decomposable anion or cation such as (C2O4 2- or NH4

+ ) (63, 76).

The vanadyl pyrophosphate is built of chains of (V) polyhedral linked

by pyrophosphate groups (60,77) .The (V) atoms are linked through the

oxygen atoms of the vanadyl in (V-O-V) chains in the (c) direction, and

the (V) atom octahedra are linked in pairs through a common edge,

forming double chains in this direction. The vanadyl groups in the paired

vanadium octahedra are oriented trans to one another (Figs. 1.8 and 1.9).


20

Figure 1.8: Idealized structure of (VO)2P2O7 (65).

Figure 1.9: crystal structure of (VO)2P2O7 (020) plane (58).


21

The unit cell of (VO)2P2O7 is orthorhombic (77,78) and is topological

similar in its (bc plane) to that of (VOHPO4.0.5H2O) in the (ab)

plane (56,60).The change from face-shared to edge-shard vanadium

octahedral in converting the hemihydrate to the pyrophosphate results in a

small expansion of one axis, but the other in plane dimension shows little

change according the pseudomorphic relations between the two crystalline

structures (59) as shown in (Figure1.10) this indicates a topotactic

mechanism of transformation, and this result has been confirmed with

Scanning Electron Microscopy. (21,56,59,62) .The layer spacing decreases from

(5.69°A) in the hemihydrate to (3.91°A) in the pyrophosphate. This is

consistent with removing the water molecules shared by the vanadium

pairs and filling the resulting vanadium coordination site with the oxygen

atoms of vanadyl groups from the layers above. This transformation

requires only very small displacements of the atoms. Importantly, since the

conversion of the hemihydrate to pyrophosphate can take place without

breaking any (V-O-P) bonds, the structural order/disorder and morphology

of the precursor phase are maintained during the transformation to vanadyl

pyrophosphate. This means that it is possible to control some of the

structural /morphological characteristics of vanadyl pyrophosphate by

controlling the specific nature of the vanadyl hydrogen phosphate

hemihydrate precursor phase (62,73). Furthermore the terminal vanadyl

oxygen atoms in the face – shared octahedral pairs of vanadyl hydrogen

phosphate have a syn arrangement; while in vanadyl pyrophosphate they

are in anti positions. These arrangements in the layer stacking direction

result in the initial formation of (VO)2P2O7 crystalline with many defects

(36,37,65,79). Alternatively, Bordes and Courtine (70,74) discussed the possible

presence of disorder in the crystalline structure of (VO)2P2O7 in terms of

different crystalline phases (β- and γ- VO)2P2O7) in which former

possesses a network structure versus a layered structure for the γ- phase.


22

X-ray diffraction patterns, however, are very similar in these two phases.

According to these authors, β-(VO)2P2O7) and γ-(VO)2P2O7 forms by

dehydration of VOHPO4.0.5H2O or reduction from γ-VOPO4.

VOHPO4.0.5H2O

α = 104.6 °

β= 127.82 °

(VO)2P2O7

α = 108.2 °

Figure1.10: Pseudomorphic relations between the crystal of VOHPO4.05H2O

and (VO)2P2O7 (36).

Vanadyl acid phosphate VO (H2PO4)2 is another phase of (VPO)

system (80,81) , this phase is made of infinite chains of corner sharing (VO6)

octahedra and isolated PO2(OH)2 tetrhedra (80) .The (V) atoms are displaced


23

(0.364) °A from the equatorial plane along the fourfold axis. As a result,

one short bond (1.600 °A) which characterizes the vanadyl (VO2+) ion,

forms in almost regular oxygen octahedra .The bond between the (V atom)

and the oxygen in the trans position is significantly longer(2.382 °A).

Furthermore ,the equatorial planes of the octahedra are alternatively

rotated of ( +18° and - 18°) around the chain axis (Fig. 1.11-a ).Phosphate

tetrahedra act as bidentate bridges via oxygen atoms for (V) atoms

belonging to adjacent chains, (OH) corners established a contact between

tetrahedra to form hydrogen bonds (Figs. 1.11- a &1.11-b ).

(A)

Figure 1.11-a: Crystal structure of VO (H2PO4)2 along XY axis (80).


24

(B)

Figure 1.11-b: Crystal structure of VO (H2PO4)2 along XZ (80).

1.3.2 - Active phase in n-Butane Selective Conversion to MA

Because the (VPO) system is characterized by the facile formation of a

number of crystalline phases, the structure of the active phase must be

discussed in term of factors such as, oxidation state, (P/V) ratio, and

crystal phase transformations under reaction atmosphere.

The various crystal phases can interconvert as a function of the reducing or

oxidizing properties of the reactants, the time on stream, and the reaction

temperature (29,37,61,82,83).

The orthophosphate (VOPO4) phases are transformed to (VO)2P2O7 by

reaction with the hydrocarbon mixture. In this reduction process single

(VO6) octahedral form pairs by loss of oxide anions. However, the

different (VOPO4) phases previously discussed possess different

reducibility to vanadyl pyrophosphate, depending on the structure or

morphology (36).


25

The Complex solid-state chemistry of (VPO) system has led to some

confusion and contradictions in the literature concerning the nature of the

active phase in the n-butane oxidation and the identification of the active

site involved in the different steps of the reaction.

Bordes and Courtine (37) suggested that the active sites in n-butane

oxidation to (MA) were associated with coherent interfaces between slabs

of (100) VOPO4 and of (010) (VO)2P2O7 along the (100) and (201) planes,

respectively. On the contrary Volta et al. (84,85) believed that the active sites

are not associated with interfacial effects between two crystalline phases.

On the basis of comparison between X-ray diffraction, they suggested that

the active phase for selective oxidation of n-butane consists of a mixture of

well-crystallized (VO)2P2O7 (V4+) and an amorphous surface (VPO) phase

of (V5+) involving many corner-sharing VO6 octahedra. This amorphous

phase may be interpreted as a precursor of (β-VOPO4), which forms at

higher reaction temperatures.

Hodnett and Delmon (67-69) used a prereduction treatment with hydrogen

in order to improve selectivity to maleic anhydride and suggest that the

best catalyst consists of an oxidized surface layer built upon a reduced core

of a V (IV) phase. The selectivity is not related to the presence of a

specific well-crystallized phase, but only to the distribution of vanadium

oxidation states between the bulk and the surface.

The active catalyst must possess an optimal [V (IV)/V (V)] ratio for

selectivity in n-butane oxidation according to Zozhigalov et al., (86) from

studies of V-P compounds with variable V (IV) /V (V) ratios. These

compounds preserve the layer structure of the initial (α-V5+OPO4)

compound. Optimal performance properties are associated with catalysts

containing [4-9 V (V) ions per V (IV) ion]. n-butane oxidation and (MA)

take place at the expense of the catalyst surface oxygen and are

accompanied by its reduction to (VO)2P2O7. On (VO)2P2O7 however, the


26

rate of reduction of the catalyst is lower than that of butane to (MA)

reaction, and therefore these authors conclude that the V(V) phase is

reduced and that the V(IV) phase is oxidized under the dynamic conditions

of catalytic reaction (86).

Weing and Schrader (74) claimed that only (VO)2P2O7 is the active and

selective phase in n-butane oxidation to (MA). A slight excess of catalyst

phosphorus (P/ V=1.1 catalysts) is necessary to stabilize the active phase

The excess phosphorus creates a distortion of (P2O7)4- crystal

environment. The α-VOPO4 is considered by these authors and others (21)

as active, but non selective phase.

Many authors proposed that the vanadyl pyrophosphate (VO)2P2O7 as

the active phase (1,9,21,56,62,73,77,80,87) .Trifiro et al. (62,88-90) attributed the

activity of the catalyst to the V(IV) phase (the vanadyl pyrophosphate),

whereas the selectivity to maleic anhydride(MA) was connected to the

presence of a very limited and controlled amount of V (V) sites .

The same discordance as to the nature of the active phase is present in

the literature concerning the optimal (P/V) ratio of the catalysts, even

though there is a general agreement that phosphorus stabilizes the (+4)

valance state of vanadium and limits its oxidation (8,74,86,90).

Garbassi et al. (91) have found that the specific conversion of

n-butane increases by an order of magnitude for a (P/V) ratio just

exceeding unity, but extended X-ray absorption measurements do not show

any structural effect of the phosphorus.

The (P/V) ratio is a key parameter in determining catalyst selectivity

and activity according to Weing and Schrader (74). Selectivity for (MA)

increase with catalyst phosphorus loading, whereas specific activity of

both selective and nonselective oxidation decreases on increase of

phosphorus content in the range (0.9-1.2) P/V range. Best catalytic


27

performances are exhibited with a catalyst with (P/V = 1.1) according to

Buchanan and Sunderesan (92).

Similar results were observed by Pepera et al. (93). According to Bosh

and co-workers (94) who studied the selective oxidation of n-butane to (MA)

under oxygen deficient over (VPO) catalyst, they found that the selectivity

was strongly influenced by the actual surface V (V)/V (IV) ratio).

Garbassi et al. (91) found a value of (P/V) surface ratio in the

(2.0-2.8) range for P/V bulk ratio in the (1.0-1.4) range.

Hodnett and Delmon (67-69) reported that the surface (P/V) ratio is (1.0)

for bulk stoichiometric (P/V) values of 1.0 or higher. They concluded

therefore that the reactivity of near-surface layers is hardly affected by the

(P/V) ratio, but bulk reactivity is drastically curtailed.

Selectivity to (MA) from n-butane maximizes for (P/V) =1.0 according

to Ai (70), who associated this catalytic effect with the presence of strong-

acid sites to activate n-butane. Finally, an optimal value of (P/V) around

(1.0) was suggested by Trifiro et al. (73,88-90) and Contractor et al. (21).

The active sites has been proposed as being the (200) plane of

(VO)2P2O7 crystallites that preferentially expose the (200) plane have been

found to exhibit higher selectivities for n-butane oxidation (95).

Transformation Electron Microscopy has shown that catalyst selectivity

correlated with the disappearance of amorphous platelets (96). Seven

different active sites have been suggested for the (200) plane of (VO)2P2O7

catalysts as shown in (Figure 1.12) whereas:

a- Bronsted acid sites, probably –POH groups.

b- Lewis acid sites, probably V IV and V V.

c- One electron redox couples, (VV / V IV), (V IV / VIII ).

d- Two electron redox couples (V V / VIII ).

e- Bridging oxygen, V-O-V, or V-O-P.

f- Terminal oxygen VV= O.


28

g- Activated molecular oxygen peroxo and superoxo species (96).

Figure.1.12: Termination of the (200) plane of (VO)2P2O7 and proposed active

sites for oxidation (s1) = Lewis acid, (s2) = Lewis acid site, (s3) = terminal

oxygen, (s4) = bridging oxygen, (s5) = superoxo and peroxo site, (s6) = V v/ viv

redox couple (97).

1.4- Kinetics of n-Butane Oxidation

A central question in analyzing the problem of bridging the gap

between surface science and applied catalysis approaches is the

verification of the possibility of describing the macro kinetic behavior

using rate equations and constants derived from the analysis of the kinetics

of the single elementary steps. Impressive results have been obtained in

this direction (98) .The fitting of macro kinetic data on the basis of a

microkinetic model is usually considered the best demonstration of the

applicability of the suggested reaction mechanism under real working

conditions (99). Is it this true also for more complex and multifunctional

reactions? In order to replay to this question it is good to briefly recall the

principles of the Langmuir description of the catalytic reaction on an


29

"ideal" surface, because this is the basis of most of the macrokinetic

models for deriving the reaction rates over catalytic surfaces. On the ideal

Langmuir surface there is one type only of active sites, the energetic of

chemisorption is independent on the coverage, and there is no interaction

between substrate (Figure 1.13-a). Real surfaces are far from the ideal

Langmuir surface, as schematically represented in (Figure 1.13- b), due to

the presence of surface heterogeneities the simplest is the presence of

steps and kinks, but as mentioned In the introduction the situation is far

more complex on oxides) which imply a distribution in the energetic of

interaction, and the presence of interaction between the adsorbates .A

chemical bond of a molecule with a surface site implies the donation of

electrons to or from the surface with thus a modification of the catalyst

conduction band, band gap, etc.; therefore, the energetic of interaction of

molecule with the surface is not independent from the surface coverage,

and the adsorption of " spectator" species, i. e. species which do not play

a direct role in the reaction mechanism, should be also considered. This

problem is even more accentuated in selective oxidation reactions where a

multielectron transfer occurs, as mentioned before.

It may be thus expected that kinetic models based on Langmuir

approach do not correctly fit macrokinetic data, but on the contrary it is a

common experience that these models are well suited to describe the

macrokinetic behavior. How can this dilemma be solved? One possible

interpretation, which is schematically shown in (Figure 1.13-c), is that

during catalytic reaction the largest part of the catalyst surface is covered

by "strongly chemisorbed" species which are characterized species

characterized by a surface lifetimes much longer than of reaction

intermediates and therefore reactant or intermediate seen a local situation

similar to "ideal" surface. This pointed out that the concept of " clean"

catalyst surface, i. e. of reaction of a molecule at single specific active sites


30

without considering the modification of the surface reactivity induced by

the presence of other co-adsorbates (reactants, intermediates, " spectator"

species), may not lead to correct description of the " real" working catalyst

surface and reaction mechanism, especially when complex, multi steps

reactions (selective oxidation reactions, for example) are considered.


31

A- A simple model of the ideal Langmuir surface

B- A simple model of real surface with kinks and steps, surface interaction between

adsorbates and electron donation of a chemisorbed molecule to catalyst conduction

band (strongly chemisorbed species)

C-Working catalyst surface with the largest part of the surface occupied from strongly

chemisorbed (reactant intermediate)

Figure 1.13- Model of the ideal Langmuir surface (99).


32

Several studies have been conducted in the eighties to derive kinetic

expressions to describe the reaction sequence for n-butane oxidation to

Maleic Anhydride in steady- state conditions on (VPO) system (86,87,91,100,101).

Escardino et al. (100) studied the kinetic of n-butane oxidation in fluidized -

bed reactor over (VPO) catalyst with (P/V= 0.8) at (676-753) K.

A triangular reaction network was proposed as shown in Scheme 1.5.

CH3

CH3

O

O

O

CO , CO2

r1

r2 r3

Scheme1.5: A triangular reaction network on n-butane to MA (100).

Maleic Anhydride (MA) and Carbon oxides (COx) were formed directly

from n-butane (at rates r1 and r2 respectively), and MA was also oxidized

to carbon oxides (at rate r3). At n-butane concentrations typical of

industrial reactors, the rate of n-butane oxidation was controlled by the

reaction between n-butane gas and surface oxygen.


33

Wohlfahrt and Hofmann (101) investigated n-butane oxidation kinetics

over a wide range of n-butane and oxygen concentration at (719-777K)

over (VPO) catalyst.

Centi, et al. (87) used a very active catalyst prepared in an organic

medium which permitted low reaction temperatures (573-613K) to be

employed. Under these conditions it was found that the rate of reaction of

n-butane to carbon dioxide did not depend on the hydrocarbon

concentration, but only on the concentration of oxygen.

The order of reaction maintained in the literature depended on the

method of analysis of available data from experimental results of

hydrocarbon depletion and oxygen partial pressure. The comparison of

these kinetics results stressed the importance of surface catalyst behavior

(selectivity to MA) as well as the role of the redox properties of vanadium

in determining the activity of the catalyst (87,101).

1.5- Surface Modifications by Interaction of n-Butane with

Catalyst Surface

Centi et al. (87) used fresh catalyst for their kinetics study and suggested

that, a critical factor governing the selectivity at very high butane

conversion is the instability of the formed (MA) in the back end of the

catalytic bed. It is thus worthwhile to analyze the variations in the catalyst

surface as a function of position in the catalytic bed (86). These experiments

showed formation of V (V) at the end of catalyst bed which was attributed

to formation the more oxidizing atmosphere present.

Trifiro et al. (89) studied the oxidation of n-butane at low and high

hydrocarbon concentrations on vanadium (IV)-phosphorus (1:1) mixed

oxide in relation to the surface modifications induced by the reaction

medium. The results showed the presence, in the catalyst, of high amounts


34

of V (III) together with V (IV) and the absence of V (V), whereas lower

amounts of V (III) together with both V (IV) and V (V) are formed, when

maleic anhydride is formed. It is suggested that two redox couples operate

in (MA) synthesis:

a- V (IV)-V (III) in the synthesis of olefins from n-butane.

b- V (V)-V (IV) in the synthesis of (MA) from the olefins is formed.

Mori and co-workers (102) studied the oxidation of n-butane on various

unsupported or supported V(V) oxides, they confirmed that (V(V)=O)

species are very active in total combustion of butane, these experiments

results indicate the important relationship between catalytic behavior and

the surface modification induced by the medium itself. The redox

properties of the fed influence the surface oxidation state of the catalyst,

which in turn profoundly affected the nature of the products formed in the

reaction.

Centi and Perathoner (99) referred to (VO)2P2O7 as the active phase in

the industrial catalysts for the selective oxidation of n- butane to (MA).

This catalyst is very selective in n-butane oxidation, but when propane is

fed Instead of n-butane, only carbon oxides and traces of other products

(propene mainly) are detected.

A first question is thus why the decrease in length of the carbon chain

produces a so drastic change in the selectivity on the same catalyst? The

answer can be that from n-butane a stable product against consecutive

oxidation (MA) forms, but not in the case of propane oxidation. However,

this answer responds to only part of the problem. The data reported in

(Figure 1.14) have already pointed that the sensitivity of the reaction

product against consecutive oxidation is not the only factor that determines

the rate of consecutive oxidation to carbon oxides.


35

Figure 1.14- Comparison of the selectivity dependence on acrylonitrile in propane

and propane ammoxidation at 480 °C on VSb4 + Sb2O4 (99).

Data in (Figure 1.15) further evidence this problem. In fact, it is expected

that a decrease in the oxygen to alkane ratio may increase the selectivity to

the partial oxidation product (propene in the case of propane), because the

formation of carbon oxides requires a larger number of oxygen atoms

(propane oxidation to (CO2 + H2O) requires (5O2 molecules), whereas

propene formation from propane requires ten time less oxygen). However,

data in (Figure 1.15) show that this is possible only for high initial

concentrations of propane in the feed, whereas for lower initial

concentrations of propane the increase in propene selectivity decreasing

the (O2/ propane) inlet ratio is much less remarkable. A comparison

between the two cases points out that the differences arise from the

different formation of alkene product, which is about ten times larger using

the higher inlet concentration of propane in the feed. This indicates that

when the formation of propene is higher, the rate of its consecutive

oxidation to carbon oxides also decreases and thus the selectivity


36

increases. The same type of phenomenon has also been observed by

changing the (O2/ hydrocarbon ratio) in n-butane oxidation on (VO)2P2O7,

although in this case a wider range of intermediate products was

detected (100).

The formation of alkenes thus induces a self- modification of the

surface reactivity. This in confirmed from the analysis of the kinetics of

but- 1-ene oxidation on (VO)2P2O7, the analysis of the transient reactivity

as well as spectroscopic studies (95). This evidences that without

considering this self modification of the surface reactivity induced by the

reaction intermediates itself it is not possible to correctly describe the

kinetic of the surface reactions.

Figure 1.15- Selectivity in propane formation from propane on (VO)2P2O7 as a

function of the inlet ration between oxygen and propane concentrations, for two

entail concentrations of propane in the feed, reaction temperature 322 °C (99).

Both (VO)2P2O7 and (VSbO4 + Sb2O4) catalysts are characterized from

the presence of coordinatively unsaturated (V4+=O) surface sites which act

as strong Lewis acid sites and which play a relevant role in the mechanism

of alkane activation (103). Alkenes through their π- bond system can


37

chemisorb on these sites forming relatively stable chemisorbed species,

although they may be considered "spectator" species, because they are not

directly involved in the mechanism of further selective oxidation of these

alkene intermediates. Oxygen also strongly chemisorbs on the surface

Lewis acid sites forming thermal stable species (they desorb above 450-

500°C), but which play a relevant role in the mechanism of oxidation.

When the surface concentration of the intermediate alkenes in alkane

oxidation is as high as to limit the amount of chemisorbed oxygen, due to

this competitive chemisorption, it is thus possible to control the population

of oxygen adspecies by this mechanism. This explains the considerable

promotion of selectivity to Partial oxidation products by increasing alkane

inlet concentration (Figure 1.15) and, on the other hand, explains also the

apparent contradiction of the different rate of acrylonitrile consecutive

oxidation when forms from propene instead of that from propane

(Figure 1.14).

Although propene forms from propane as a reaction intermediate, its

concentration is clearly higher when it is fed directly and thus its effect in

limiting the concentration of Surface oxygen species is present even for

higher conversions of the hydrocarbon. Therefore, the maximum in the

formation of acrylonitrile is observed at higher hydrocarbon conversions

when the alkene is fed instead of the alkane on the same catalyst

(Figure 1.14).

In the oxidation of n-butane ،partial oxidation product (butenes and

butadiene) are observed at low oxygen concentrations when a very limited

number of vanadium (V) species are present on the surface of the catalyst.

The change of valance state of vanadium on the catalytic surface upon a

balance of three factors:

a- The redox potential of the feed.

b- The rate of oxidation of the catalyst at the temperature of reaction.


38

c- The rate of reduction of the catalyst at the temperature of reaction.

1.6- Relationship between Redox Properties & Catalytic

Behavior

As pointed out in the section on catalyst structure ( section 1.3), the

binary (VPO) system was rather complicated because of the great variety

of observed phases and the difficulties encountered in the development of

successful and reproducible syntheses of pure single phases stable under

reaction conditions. Over the past (20) years improved understanding of

the factors determining the structure of (VPO) has allowed the

development of suitable methods of preparation of (VO)2P2O7(65,89).In

particular is the preparation of vanadyl pyrophosphate active phase with

(0.95-1.2) P/V ratio and variable mean valance states of vanadium (up to

about 20% of V (V) without significant changes of X-ray diffraction

patterns (104) (Figure 1.16).

V(III) V(IV) V(V)

0 85 15

0 100 0

……. 11 89 0

Figure 1.16: Effect of the presence of different valence states of vanadium on the

principal X-ray diffraction lines of vanadyl pyroph osphate (P/V=1.0) (104) .


39

The variation of the content of (P) in the composition modifies the

catalytic properties in n-butane oxidation (86) and the redox properties of the

catalysts (104), a slight deficiency of phosphorus, does not change the rate of

V (IV) oxidation to V (V). On the other hand, an excess of phosphorus

only slightly influences the rate of oxidation, but strongly affects the rate

of reduction. This effect is in part attributed to a decrease of the number of

the active site, but also reflects a particular kind of interaction of

phosphorus upon the vanadium ions (61,104). In order to correlate the

observed variations in the redox properties with the catalytic behavior in n-

butane oxidation on the fresh catalyst used by these authors, it is necessary

to distinguish between tests at low conversion and high conversion.

At low conversion, the catalysts with (P/V ratio = 0.95 and 1.01) show

the same activity, selectivity, and kinetic behavior (86). However, at high

conversion of n-butane (80%) the catalyst deficient in phosphorus and with

the observed higher rate of vanadium oxidation forms primarily carbon

oxides. The catalyst with more phosphorus with respect to the

stoichiometric ratio of (1.0) are less active and less selective (90), but do not

show the strong decline of (MA) selectivity at the highest conversion.

It is reasonable to correlate these catalytic effects with the redox

properties of the catalyst (105). The strong increase of the rate of vanadium

oxidation in the phosphorus-deficient catalyst leads to an enhancement of

the rate of consecutive oxidation of ( MA) (effect on the selectivity),

whereas the decreased rate of reduction of vanadium in the catalyst with

higher (P) content than a stoichiometric one leads to the reduced rate of

hydrocarbon depletion (effect on the activity)

Nakamura et al. (105) also suggested that some V (V) ions were

necessary for (MA) synthesis, but the optimum mean of valance state of

catalyst is close to four. Similar conclusions were presented by Weing and

Schrader (74). They reported catalyst phosphorus loading is a key parameter


40

in determining catalyst selectivity and activity. An increase of (P) content

in the (0.9-1.2 P/V) ratio decreases specific activity in n-butane depletion

and increases the selectivity to (MA). The same selectivity was reported by

Pytnitskaya et al. (71), Buchanan and Sundarsan (92), also found a correlation

in vanadium phosphorus oxides between the concentration of V (IV) ions

in the discharged catalyst and the activity in n-butane oxidation.

The results of surface modifications induced by the interaction of C-4

hydrocarbons as well as the results of (P/V) ratio effect on redox properties

and on (selectivity/activity) are interpreted on the basis of two different

types of reaction in going from the alkane to (MA): a first step is oxidative

dehydrogenation up to adsorbed butadiene and second step is further

oxidation of this intermediate. The first step is suggested to be controlled

by the rate of V (IV) reduction, whereas the second step is controlled by

the amount of V (V) available. However, when the content of V (V) is too

high, the oxidation proceeds further to carbon oxides.

These data suggest a strong relationship between selectivity to (MA) at

high conversion and rate of formation of V (V) as reported by Trifiro and

co-worker (62) (Figure 1.17).

(A)

Conversion

3 >2 >1 (in disorder arrangement) (62)

Figure 1.17-a: Maleic Anhydride yield VS conversion at 300°C in n-butane

oxidation for different catalyst with P/V= 1.0, feed = 0.6% n-butane, 12% O2 (62).

MA

%


41

(B)

Time (minutes)

Figure 1.17- b: Rate of V (IV) oxidation to V (V) in air for the same catalyst of

(A) (62).

1. ٧ - Mechanism of n-Butane Activation

Kinetics data agree that the hydrocarbon activation is the first step in

n-butane oxidation, the alkane is rather inert and the rate determining step

for saturated hydrocarbons the is the dissociation of the (C-H)

bond in a manner similar to the production of hydrocarbon free

radicals(87, 91,100,101,106) . This would be requiring very reactive oxygen to

break (C-H) bond which is bound weakly to the solid for it to be

thermodynamically favorable. It is known from the literature (107-110) that

(O) - and (O2) - species (one oxygen molecule may be activated by

accepting electrons from the solid) are very oxidative dehydrogenation

and activation of light alkane (108-113) i.e., highly reactive surface oxygen

(weakly adsorbed oxygen and or localized surface lattice defects) are the

active sites for alkane activation. They suggested that the oxygen

chemisorption occurs via a dissociative pathway on vanadium dimers-

V (

V)

/ V to

t %


42

leading to a (V (V)-O*) type surface species capable of activating the

alkane.

Due to the topotactic transformation that occurs without the breaking

of (VPO) connections and due to the higher electronegativity of (P) with

respect to (V), the presence of defects in the structure of vanadyl

pyrophosphate is reflected on the surface in an enhancement of Lewis

acidity of surface unsaturated vanadium ions (Figure 1.18). Strong acid

sites on (VO)2P2O7 were observed also by Puttock and Rochester (111,112) by

infrared spectroscopy.

(A)

A- Medium –strong Lewis acid sites, well crystalline structure

(B)

B- Very strong Lewis acid sites, disordered crystalline structure

Figure 1.18: Model of Lewis acid sites in (VO)2P2O7 with different degrees of

disorder in the stacking fold of (020) planes (62).


43

All above observations may be summarized in the following model:

The presence of defects in the crystalline structure creates strain in the

V-P-O bonds and would create a surface-activated reactive couple.

V O Pδ δ+

− − − − −−

→

……….. (1.5)

This reactive couple was proposed by Centi et al. (73) as being

responsible for the concerted mechanism of the removal of two hydrogen

atoms, according to the (Figure 1.19-a, 1.19-b and Figure 1.20) which show

the redox cycle proposed to this mechanism of activation of n-butane.The

model is based on the idea that (V (IV)–oxygen) are the active sites of butane

dehydrogenation (62,99).

Munson and et al. (113) had developed an experimental protocol to study

the mechanism of this reaction in which 13C-isotopically labeled n-butane

is flowed over a catalyst bed and the reaction products are analyzed using 13C NMR spectroscopy. This strongly suggests that the total oxidation of

n-butane on (VPO) catalysts involves the oxidation and abstraction of the

two methyl groups of n-butane, and the two methylene groups of n-butane

form ethylene. An organometallic mechanism is proposed to explain these

results.

In conclusion, it is necessary to try to develop new models of the

surface reactivity which include new evidence on aspects such as (i) the

role of chemisorbed species on the surface reactivity, (ii) the presence of

multiple pathways of reaction (iii) the dynamics of catalyst reconstruction

(iv) the mobility of surface adspecies, etc. The consideration of all these

possible effects in analyzing the surface reactivity will make possible the

design of new catalysts as well as the understanding of surface

reactivity at oxide surfaces (99).


44

HC

CMe

Me

H

V O

V

O

OP

H

(A) A- Detail of the concerted mechanism of (2H) abstraction.

VV

O

O

O

CCH

CH3

H3C

H

H

H

V VO

O

O

O

O

O

O

H

HO

-H2O

VV

O

O O

O

O2

O

O

O

O

3+

O

CH

HC

H3C

CH3

4+

(B) B- Redox Cycle involved in the mechanism of activation.

Figure 1.19: Proposed mechanism of n-butane activation on VO)2P2O7(73).


45

O O

R

O O

+

O

O OO

OO O

COOHHOOC

HOOC

O O

O O

CH3COOH

O

CHO

O O

COX

Furan

MA

Dihydrofuran

Butadiene

n-Butane

Methyl vinyl Ketone

But-3-ene-oxide

Crotonaldehyde

Maleic acid

Butirrolactone

But-1-ene-3-oxide

2-Butoxide

2-Butanone

Butenes

H2O

Figure 1.20: Reaction network in n-butane oxidation on (VO)2P2O7(99).


46

1.8 - Role of Adsorbed & Lattice Oxygen Species

In the selective oxidation of n-butane to (MA) on vanadyl

pyrophosphate (VO)2P2O7 catalysts, many results have been reported on

the respective roles of lattice and gas-phase oxygen in the formation of

partial and complete oxidation products (113).

Three types of lattice oxygen species with different reactivites are

present at the surface of vanadyl pyrophosphate (Figure 1.21) (24).

OVO

VV

P

O

VV

(a) (b) (c)

Figure 1.21: The Three Types of Oxygen Species (24).

The participation of lattice oxygen is a general characteristic of metal

oxide catalyzed reaction in selective hydrocarbon oxidation (114-117).

Pepera et al. (93) concluded that the lattice oxygen ions located in the

top few surface layers are responsible for the oxidation of n-butane to

(MA, CO and CO2). Another authors (106 , 117, 118) obtained experimental

evidence to support this conclusion. Abon and co-workers (119) claimed on

the basis of isotopic labeling results that only lattice oxygen is active for

the formation of (MA) and other products. The circulating fluidized-bed

riser reactor technology for (MA) production described in the literature


47

(18,21,119) is based upon the fact that, under anaerobic conditions, the lattice

oxygen of (VPO) can selectively oxidize butane to (MA) (18,21,119).

This problem becomes more complex when oxygen is co-fed with

butane under steady state reaction conditions employed in industrial fixed-

bed reactor processes. In addition to the conventional Mars-Van Krevelen

mechanism, where lattice oxygen is the active agent for butane oxidation,

gas-phase oxygen, surface lattice oxygen and / or activated chemisorbed

oxygen have all been proposed as important oxidants in the formation of

Maleic anhydride or in the formation of unselective products, (CO and

CO2)(96,107,120-122). For example, Trifiro et al. (95) proposed that adsorbed

oxygen is responsible for selective oxidation and that it is involved in the

oxygen insertion steps required for the formation of (MA). Ebner and co-

workers (95,106) concluded that adsorbed oxygen is selective only in the

(MA) formation step. They proposed that two types of oxygen are involved

in butane oxidation: surface lattice oxygen that is responsible for ring

closure, and activated chemisorbed oxygen, (O*), that is involved in the

further step of (MA) formation. Rodemerck et al. (121) reported that

adsorbed oxygen is active but not selective, i. e.; it can only produce

(CO2). In contrast to all of these studies, Zazhigalov et al. (122) concluded

that (MA) formation over (VO)2P2O7 is mainly due to gas- phase oxygen.

The controversy in the literature about the roles of lattice and

chemisorbed oxygen calls for further clarification of the puzzle. Wang and

co-workers (123) used a novel microbalance reactor to carry out kinetic

analyses of butane oxidation by (VPO) catalysts and of the oxidation of

partially reduced (VPO) with oxygen .These authors conclude from there

experiments that both lattice oxygen and adsorbed oxygen on (VPO)

catalyst can selectively oxidize butane to (MA). Under aerobic conditions,

the oxidation of butane by adsorbed oxygen species is much faster than

by lattice oxygen.


48

The first attempt to rationalize the problem of types of oxygen species

involved in the mechanism of C-4 alkene transformation to (MA) was

reported by Weiss et al. (124) , vanadium-oxygen double bonds (V=O) were

suggested to be the active sites both in:

The oxydehydrogenation step from butenes to butadiene, the

consecutive steps of oxygen insertion to (MA).The first step involves a

homolytic (C-H) dissociation of the adsorbed olefine and formation of the

allyl radical coordinated to a metallic transition ion as a surface π-allyl

complex. The allyl carbcation obtained can form the diene by loss of a

proton through reaction with a nucleophilic agent. The allyl ester that

forms can undergo a rapid reversible rearrangement in which each end of

the carbon of the skeleton is alternatively bonded to the lattice oxygen.

This lattice oxygen possesses a weak electrophilic reactivity. On the

contrary, electrophilic oxygen must be involved in the attack on the diene.

The proposed mechanism is shown in (Figure 1.22-a). The product of

attack on the diene can occur according to two-selective pathways, the first

one involves 1, 4- cyclization to 2, 5-dihydrofuran which is easily

dehydrogenated to furan; the second pathway is deprotonization of the

allyl carbcation to a dienolate structure.


49

n-Butane

- 2HCH2=CH-CH2-CH3 CH2=CH-CH=CH2

CH2=CH-CH=CH2

O

M

O

MH

HO

M3

CH3-CH-CH-CHO

HO OM

O

M2

-2H

-2 e a

O

b

H2O -2e

-H

-H

(A) Figure 1.22 -a: Proposed reaction of n-butane to MA (62,124).

Hetrocyclization of this structure by a new intermolecular electrophilic

attack leads to furan. The second path, however, can also form

crotonaldehyde by hydration and retroaldolization, leading to a form

crotonaldehyde by possible reduction in selectivity .An analogous

mechanism may be assumed for the further oxidation to (MA),i.e.,

electrophilic attack ,deprotonization , etc…….(Figure 1.22- b) .


50

O

-2e

O

M

O

O

M

-H

O

H2O

O

M

-2e-H

O

OH

OH

-2H

OOO

M2+

(B)

Figure 1.22- b: Proposed reaction of n-butane to MA (62,124).

1.19 - Nature & Mobility of Adsorbed Species

Puttock and Rochester (125)

studied the adsorption of n-butane, 1-

butene, 1, 3-butadiene, furan and (MA) on vanadyl pyrophosphate by

infrared spectroscopy. Contact between butane vapor and vanadyl

pyrophosphate at temperatures up to (450°C) did not give rise to infrared

bands due to adsorbed species. 1-Butene, on the contrary, gave rise to the

formation of adsorbed species by contact at room temperature. Higher

temperatures (200°C) catalyzed the transformation of 1-butene to 2-butene,

confirming the high surface acidity of these catalysts (104,125,126)

.

After removal of the gas phase an adsorbed species, characterized by a

strong band observed near (1600cm-1) and bands appeared in the

(2800-3000cm-1) region, was observed. The latter bands, which attributed

to saturated hydrocarbon groups, disappeared at higher temperatures

(350°C), whereas the band near (1600cm-1) unchanged. This band was also

observed by Rozhkova et al. (127)

and was attributed to a surface carbonyl.


51

A similar spectrum of adsorbed species was found in the interaction of

furan with the catalyst. The primary interaction of furan and vanadyl

pyrophosphate was suggested to involve coordination between the oxygen

atoms of associatively adsorbed furan molecules and coordinatively

unsaturated exposed vanadium cations with Lewis acid properties. Furan

showed a slight oxidation to maleic anhydride in the absence of oxygen but

the presence of oxygen promoted maleic anhydride was primarily non

dissociatively adsorbed on (VO)2P

2O

7, but two further bands at (1560 and

1450cm-1), attributed to carboxylate anions, suggested a partial oxidation

of the adsorbed (MA).

Weing and Schrader (74)

studied by infrared spectroscopy the n-butane

interaction with (VPO) catalysts with variable (P/V= 0.9, 1.0 and 1.1)

using an in situ FTIR cell. These authors presented evidence for the

presence of reactant (n-butane), partially oxidized product (MA),

combustion products (CO, CO2, and H2O) and reactive surface species

(Maleic acid and olefins) on the vanadyl pyrophosphate. Another

observations reported by Pepera and co-workers (93)

agree with the

hypothesis that the catalytic process in this complex reaction involves the

shuttling of hydrogen a way from and oxygen toward the intermediate

adsorbed on the surface, and the intermediate forms a stable surface

species (which does not adsorbed or react) to (MA). This mechanism is a

key to the selectivity and the absence of intermediate products in the

reaction of n-butane conversion to (MA).

1.10- Promoted (VPO) Catalysts

Numerous attempts have been made to synthesize improved (VPO)

catalysts by adding various amounts of other metallic elements. Although


52

reports of some of these attempts have appeared in literature (128-146), the

vast majority were found in the form of patents (64,128-130,133-136)

.

The effect of these promoters was studied by many authors; Ai (131,132)

studied the effect of alkali metal addition on activity and selectivity. It was

reported that the addition of less than (10% Li) showed no remarkable

effect, Ai also studied the effect of the methods of preparing V2O

5-P

2O

5-

ZrO2 catalysts on their activity and selectivity in the oxidation of n-butane

by changing the procedures of ZrO2addition. Ai found that the best

performance was obtained with the catalyst prepared by adding

simultaneously two solutions of ethyleneglycol in which ZnOCl2 and o-

H3PO

4 have been dissolved to a powder of the precursor of the (V

2O

5-

P2O

5) catalyst prepared in an organic medium. The presence of the vanadyl

pyrophosphate crystalline phase was detected in all of the catalysts. (131,132)

The addition of suitable activators to (VPO) catalyst was reported to

give an improved yield of (MA) by several patent claims (6,133-136)

using a

catalyst containing (P-V-Mo) to oxidation n-butane to (MA), this catalyst

gave a maximum (MA) yield of (88%).

Zazhigalov and co-workers (128,137) studied the effect of added alkali and

alkaline-earth metals (Li, Na, K, Mg, Ca or Ba) on the properties of (VPO)

catalysts. The results showed that the addition of alkali and alkaline earth

metals to these catalysts produced two effects, on the one hand, there was

an increase in the content of oxygen on the surface and hence an increase

in the number of acid centers, on the other hand, the presence of these

promoters increase the (P/V) ratio on the surface of the catalyst and this

explains their stabilizing effect. It should be noted that the greatest increase

in the surface content of phosphorus was observed when (Li) is added, and

(Li) was considered to be one of the best stabilizing additives


53

A new catalyst belonging to the (V-P-Mo-O) system selective in the

mild oxidation of n-butane to (MA) were prepared and characterized by

Courtine et al. (139)

These attempts led to the conclusion that (Mo)could be

substituted up to (7%) in (VOPO4) phases. Thermal analysis of the

hydrated precursor, XRD and IR spectroscopies of both hydrated and

anhydrous solid phases obtained showed that the solid solutions

isostructural with (VOPO4.2H2O) and (α-VOPO4) receptively could be

formulated.

Zazhigalov and co-workers (140) studied the properties of cobalt-promot

(VO) ٢ P2O7 in the oxidation of n-butane. This study focused on the

influence of cobalt additives on the composition of the vanadium-

containing catalyst, and on cobalt's other properties which were important

for the production of industrial (VPO) catalysts. Fresh catalysts were

composed of (VOHPO4.0.5H2O) phase. After reaction the catalysts

contained (VO)2P2O7. Cobalt was uniformly distributed in the pellets. Its

presence increased the content of phosphorus at the surface, which

modified the surface acidity and in turn improved the selectivity for n-

butane oxidation. No changes of the profile of phosphorus with depth were

observed, even after (500h) on stream, the surface composition of the

catalyst remained unchanged. Cobalt stabilized the catalyst performance

by forming cobalt phosphate which reduced phosphorus losses, improves

its catalytic properties and prolonged its lifetime.

A series of (VPO) catalysts with different (P/V) ratios and with or

without (indium and TEOS) additives have been characterized by

controlled-environment (XRD, ICP, ESCA, TEM, SEM, BET) and

chemical titration used for n-butane oxidation to (MA) (141) .The best

catalyst contained slight excess (P /V) with indium and tetra ethyl ortho-

silicate (TEOS) promoters. It was found that excess P increased the


54

resistance of catalyst precursors toward oxidation and resulted in (VPO)

catalyst with a large exposed platelet face of layer morphology. The

promoters reduce the thickness of the platelet face of layer morphology.

The promoters reduce the thickness of the platelet and facilitate the

oxidation of the precursor which contains disordered (VOHPO4.0.5HO)

.The combination of the promoters and excess P results in a (VPO) catalyst

with appropriate oxidizability and morphology and gave high yields of

MA.

Guliants et al. (142) investigated the oxidation of n-butane to (MA)

over a model (Nb, Si, Ti, V and Zr) promoted bulk (VPO) and supported

vanadia catalyst , The promoters were concentrated in the surface region of

the bulk (VPO) catalysts .For the supported vanadia catalyst ,the vanadia

phase was present as a two-dimensional metal oxide over layer on the

different oxide supports (TiO2,ZrO2,Nb2O5,Al2O3 and SiO2) .No

correlation was found between the electronegativity of the promoter or

oxide support cation and the catalytic properties. Both promoted bulk

(VPO) and supported vanadia catalysts containing surface niobia species

were the most active and selective to (MA). These data suggested that the

activation of n-butane on both the bulk and supported vanadia catalysts

probably required both surface redox and acid sites, and that the acidity

also played an important role in controlling further kinetic steps of n-

butane oxidation.

The presence of (Na) in technical grade (V2O5) leads to solid solution

formation, this phenomenon lead to prepare VOPO4.2H2O with a new (P/V

=1.1). (1)

Sajip and co-workers (143) described and discussed the effect of Co and

Fe doping on (VPO) catalyst, prepared by organic method (with

isobutanol). At low levels, both Co and Fe dopants significantly enhanced

the selectivity and the intrinsic activity to MA. A combination of powder


55

X-ray diffraction, P NMR spin-echo mapping spectroscopy and

transmission electron microscopy ,together with catalysts tests data, was

utilized to analyze the origin of the effects of Co doping. Co appears to be

essentially insoluble in crystalline (VO)2P2O7 and was preferentially

distributed in and stabilized an amorphous (VPO) material .It is suggested

that the origin of the promotional effect of (Co) was associated with its

interaction with the disordered (VPO) phase. The same techniques have

been used to analyze the Fe-doped catalyst, but at present it is not possible

to be definitive concerning the specific location of the Fe-dopant within

the phases present .Previous studies have indicated that (Fe) can form a

solid solution within (VO)2P2O7 and therefore it is probable that the (Fe)

amorphous vanadium phosphate phases are formed in the catalyst system.

Other authors (144) studied the effect of (Bi) on (VPO) catalyst, they

found that the incorporation of (Bi) into the (VPO) lattice lowered the

overall vanadium oxidation state from (4.24 to 4.08) .It also lowered both

the peak maximum temperature for the desorption of oxygen from the

lattice from (1001 K) (undoped) to (964 K) with shoulder at (912 K). The

total oxygen desorbed from the Bi-doped catalyst was only one-fourth that

of the undoped catalyst, while the amount of oxygen removed by TPR was

roughly the same for both catalysts. These results suggested that in

anaerobic oxidation, the Bi-doped catalyst will have roughly the same

activity as in undoped catalyst in C-4 hydrocarbon oxidation but would

have a higher selectivity to products such as olefins and (MA).

The barothermal and mechanochemical treatment of (VPO-Bi)

precursor was investigated (145), the barothermal treatment led to increase of

the relative ratio (001) plane of precursor without changes of the phase

composition. The (P/V) surface ratio increased more than two times and

the phosphorus surplus forms the islands on catalyst surface which

decrease the available active surface fragments. The change of Bronsted /


56

Lewis acidity ratio of surface as result of treatment also was observed. The

activity of the catalyst in n-butane oxidation less change up to several

value of P/V ratio and decreased with its growth. The selectivity to (MA)

increase (more than 10% mol. %) practically in all interval of the (P/V)

ratio changes. Good correlation was observed between the selectivity to

Maleic Anhydride (MA) and Bronsted acidity of the catalyst.

The mechanochemical treatment of precursor showed an enhancement of

n-butane conversion and an improvement in (MA) selectivity and yield.

The sample milled in water exhibited a rise in conversion to (91%) but

selectivity increased only some percents. The maximal increase of the

selectivity after treatment in ethanol (more than 15 mol. %) and growth of

activity (but only 5-6 %) was observed. The correlation between the

changes of selectivity and Bronsted acidity was established (145).

1.11- Preparation of the catalysts

In accordance with the growing importance of the C-4 partial oxidation

route to (MA), numerous patents (16,19,20,146,147) have been awarded in the

past years for process which involved the preparation of (VPO) catalyst.

Various routes are available to prepare Vanadyl Hydrogen Phosphate

Hemihydrate VOHPO4.05H2O precursor, as outlined in Figure 1.23:

V2O5 + H3PO4 + i-BuOH

Organic Route

V2O5 + NH2OHHCl

VOHPO4.0.5H2O

(VO)2P2O7

Aqueous Route

+ H3PO4

Figure 1.23: Synthesis routes for vanadyl pyrophosphate catalysts (97).

The aqueous route entails the reduction ( V5+ ) by NH2OH , HCl or


57

other reducing agents followed by the addition of H3PO4, VOHPO4.0.5H2O

is recovered by crystallization or evaporation.

The organic route involves the reduction of V2O5 by an organic solvent

(an alcohol), followed by the reaction with o-H3PO4 and recovery of the solid.

After recovery of the precursor, it is washed to remove trace amounts of

water soluble (V5+) compound and then calcined in nitrogen at (773) K

followed by final activation in air or n-butane/air at (673) K. It is critical that

calcination and activation be carefully controlled or the catalyst can be over

oxidized resulting in the formation of VOPO4 which is non selective

catalyst for n-butane (97).

The synthesis route and the reaction conditions during synthesis, affect the

morphology of and ultimately the catalyst performance. The precursor

prepared by the aqueous synthesis route is generally more crystalline than the

catalyst prepared by the organic route (95) .The organic synthesis route results

in platelet crystalline morphology; the size of the platelets and the way they

pack are determined by the choice of organic solvent. Isobutanol produces

rosette morphology where the platelets agglomerate. With sec. butyl or t-butyl

alcohol well formed platelets forms that do not agglomerate (148). The crystal

ordering of the precursor is also affected by the solvent. Large alcoholic

molecules, such as benzyl alcohol, appear to produced platelets with stacking

faults, deduced from the broadening of the (200) reflection in the XRD (149).

Stacking fault defects in the precursor have been correlated with improved

catalyst performance (148).

In recent years Hutchings and co-workers (149) prepared the precursor

by using V2O4 as a starting material with water as solvent. They also used

V2O4 with either H3PO4 99% or H4P2O7; reacted in autoclave at 145 ˚C to

form well crystalline VOHPO4.0.5H2O.The surface area of the precursor is

significantly enhanced when water is added as a solvent. On activation in

(n-butane / air) the catalyst surface area is increased from (4 m2/gm) for the


58

precursor to (10-13 m2/gm). The selectivity to (MA) observed to be very

similar to other non- promoted (VPO) catalyst reported in the litterateur.

Although this study considered to unlikely that V2O4 can be used as a

starting material for the synthesis of commercial vanadium phosphate

catalyst, this study does show that relatively high area catalysts can be

achieved using water as solvent.

A dihydrate phases of (VPO) catalyst VOPO4.2H2O was prepared for

the first time in 1965 by Ladwig (33) by suspend V2O5 with rapped stirring

in distilled water and H3PO4 with (P/V = 7.3).The mixture was refluxed for

24 hours. The resulting precipitate was filtered, washed with distilled water

until a red color appeared then it was washed for several times with

acetone and left to dry.

No sign in the litterateur to prepare this phase with (P/V ratio =1.1).In

2000, several attempts were unsuccessfully exerted to prepare this phase in

the above ratio starting from V2O5 produced by Fluka and BDH company,

but all of these trials were not successful, on the other hand this phase was

successfully prepared starting from technical grade V2O5 with shorter

reaction time (16h), due to the intercalation effect of (Na ion) which led to

form solid solution of this phase. (1)

A new method through intercalation of VOPO4.2H2O crystallites in

primary alcohol (2- propanol or 1-butanol), followed by reduction with the

alcohol, have been investigated for the preparation of catalyst precursor.

Lamellar compounds were formed consisting of (V 4+, P +5 and alkyl

group) (150).

According to Bordes and Courtine (29), the calcination of VOPO4.2H2O

at 700 °C leads to obtain α-VOPO4.

Recently Bartley et al. (151) prepare vanadium (V) phosphate using

solvent-free method, by reacting (V2O5) with (H3PO4) in the absence of

water at 150 °C ,the resulting material was designed as anhydrous VOPO4.


59

This material is readily hydrate to formVOPO4.2H2O. Activation of this

phase with n-butane/air leads to formation of (α1-VOPO4), while the

reaction of anhydrous (VOPO4) with alcohol leads to the exclusive

formation of VO (H2PO4)2.

Finally, there are three important parameters affect in determining

the final characteristics and activity of these catalysts which can be

summarized as:

a- The reducing agent and solvent (8, 70, 61,152-157).

b- The (P/ V) ratio

(8,146,148,155).

c- Activation conditions (29, 71)

.


60

Conclusion of Bulk (VPO) catalyst

Oxidation of n-butane on the unsupported or bulk (VPO) catalysts is

the only known commercial process for an alkane oxidation.

A number of V (IV) and V (V) phosphates exist in the (VPO) system

and the correlation of catalytic performance with crystalline structure has

been reviewed (95,158).

Vanadyl (IV) pyrophosphate, (VO)2P2O7, has been identified as

critical for active and selective industrial catalysts (158). Some argue that the

(V5+/V4+) dimeric species in the top most oxidized layer of (VO)2P2O7 are

the active sites (8,159) , while others believe that the active sites lie within

the micro domains of crystalline vanadyl (V) orthophosphates, (β -, αII- ,

γ-, and δ-VOPO4), or (vanadyl (IV) meta phosphate) VO(PO3)2, formed

on the (100) faces of vanadyl pyrophosphate under the catalytic reaction

conditions (106). A recent kinetic study demonstrated that the best bulk

(VPO) catalysts contained only crystalline vanadyl (IV) pyrophosphate

after reaching the steady state (142). The vanadyl (IV) pyrophosphate phase

displayed preferential exposure of the (100) planes which contain vanadyl

dimers associated with active sites for n-butane oxidation (142). A high-

resolution Electron Microscopy study demonstrated that the surface (100)

planes in the fresh catalysts are covered with (1.5 nm) amorphous layered

which completely disappears within (23 days) of n-butane oxidation (96).

The kinetic studies of the bulk (VPO) catalysts further demonstrated the

similarity between the bulk (VPO) catalysts and supported catalytic

systems.

The catalytic activity of the bulk (VPO) catalysts was confined to a

very thin surface region of the (100) crystalline planes of vanadyl (IV)

pyrophosphate (1-2 atomic layers) (92).These findings suggested that the

crystalline vanadyl (IV) pyrophosphate phase in the bulk (VPO) catalysts

functioned as a support for the active surface. Such (VO)2P2O7 support


61

stabilized some specific surface termination of the (100) planes with

(V4+/V5+) species and phosphate groups without contributing its lattice

oxygen to n-butane oxidation. Unfortunately, various spectroscopic

techniques, such as (Raman, ESR, UV-Vis, IR and EXAFS/XANES), are

unable to provide information about the molecular structure of the surface

present in the bulk (VPO) catalysts because of the much stronger signals

from the catalyst bulk than the catalyst surface. Limitations of both the

bulk and surface characterization techniques currently available, coupled

with the complex solid-state chemistry of vanadium phosphates, have led

to considerable confusion and contradictions in the literature concerning

the identity of the active sites involved in different steps of n-butane

oxidation to (MA) over the bulk (VPO) catalysts.

1.12- Supported (VPO) Catalyst:

Supported metal oxide catalysts consist of an active metal oxide

component, e.g. vanadia, deposited at the surface of an oxide support, such

as SiO2, TiO2 and Al2O3.

Currently, considerable interest is devoted to study selective oxidation

of alkanes on supported vanadia catalysts. The selective oxidation of

alkanes is highly desirable due to their potentially low environmental

impact and the relatively low cost of raw materials. Some typical systems

being studied are partial oxidation of n-butane to (MA) on supported

(VPO) catalysts (105,160) and oxidative dehydrogenation of n-butane (161).

Many fundamentals questions still remain unanswered about supported

vanadia catalysts despite the importance of this catalytic system.

This section concerns with structure-property relationships in C4-

hydrocarbon oxidation on supported vanadia catalysts.


62

1.12.1-Molecular Structure of Vanadia Species in Fresh Catalysts

Structure of supported vanadia catalysts is a strong function of the

surface coverage. Surface coverage of vanadia over layer supported on

metal oxides has been determined experimentally using several

spectroscopic techniques, such as (Raman, IR, XPS, 51V NMR and UV-

Vis), as well as temperature-programmed reduction (TPR) and redox

reactions. Among these techniques, Raman spectroscopy is particularly

discriminating between the surface vanadia species and the crystalline

(V2O5). Characteristic Raman bands of polymeric vanadia species

corresponding to the (V-O-V) stretches, appear below (1000 cm-1) at high

coverages, while the isolated vanadate species present at sub monolayer

coverages produce only one Raman band of the terminal (V=O) bond at

(1020-1040cm-1). The finger print Raman features allow detecting

monolayer formation immediately prior to appearance of crystalline

(V2O5). Based on Raman spectroscopy measurements, monolayer surface

coverage on a number of oxide supports (Al2O3, TiO2, ZrO2, Nb2O5 and

CeO2) was found to be (7-8 VOx/nm2) (162) where nm is nanometer.

Vanadia supported on silica exhibited unusually low monolayer surface

coverage of only (0.7VOx/nm2). Lower density and reactivity of the

hydroxyls explain such low monolayer surface coverage on the silica

surface, which anchor vanadia species to the support. However,

commercial grade supports, such as pigment grade Titania, typically

contain monolayer quantities of (P, Na, K, Ca, etc.), which interact with

surface vanadia to form an amorphous phase (163). In such systems, the

surface vanadia titrates both the oxide support and the surface impurities,

delaying formation of crystalline V2O5 until several vanadia monolayers

are formed. Thus, the observed "monolayer'' vanadia coverage in such

impurity-containing systems can be (2-4) times greater than that found

above for the well defined supported catalysts.


63

A number of synthesis methods were used to prepare supported

vanadia catalysts, non aqueous impregnation with vanadium alkoxides (169,170), such as incipient impregnation of support powder like (TiO2) with

aqueous solution of NH4VO3 in oxalic acid (164). Aqueous impregnation

with vanadium oxalate is favored in commercial preparation due to its high

solubility in water and the absence of undesirable volatile organic solvents.

Under ambient conditions the supported catalysts contain multilayers of

adsorbed water. Under such conditions the bridging (V-O-support) bonds

are hydrolyzed and the hydrated vanadia species are dissolved in the thin

aqueous layer.

A new generation of supports based on ceramics with high thermal

conductivity and without oxygen atoms were prepared and tested to

disperse the active (VO)2P2O7 phase (165). The best example is a new

relatively large specific surface area (β-SiC < 20 m2/g) prepared via the

shape memory synthesis. The (β-SiC-supported VPO) approved to be a

much better catalyst in terms of (MA) selectivity and yield than its bulk

equivalent, in fixed-bed configuration of reactor. The ability of this

ceramic to well disperse (VPO) was explained by the presence of an inter

phase (a glue) containing the (V, P, Si, C and O) elements, which did not

interfere with the formation of the active and selective (VPO) phase from

the (VOHPO4.0.5H2O).The real nature of this inter phase still needs to be

clarified. The high (MA) selectivity obtained on (VPO/β-SiC) was

attributed to both the absence of microporosity on the support and more

importantly, to the heat sink role of this support avoiding the formation of

hot spots on the catalytic sites and protecting the product, MA ,from

further oxidation into CO and CO2.

The uses of other exotic heat conductive materials as support such as

(Si3N4) and (BN) ,proves the validity of the concept of heat transfer from


64

the surface to the bulk of the catalyst, without taking into account the

macroscopic transfer of heat outside the reactor (165).

It was finally demonstrated, from an industrial point of view ,that the

presence of the (β-SiC) support could provide high volume yield of (MA/

volume of catalyst) in a fixed-bed reactor because it was possible to work

with relatively high concentrations of n-butane ,contrary to what is

normally possible with the conventional bulk catalyst .In addition because

of the good dispersion, a large fraction of the expensive unused internal

part of the bulk active phase was replaced by cheap and light (β-SiC)

support .The optimal amount of active phase was found at (30wt %) of the

total composite. Finally, also demonstrated was the protective role of the

support on the thermal stability of the reactive phase which was much

more resistant to overheating accidents than the bulk material.

This new concept of heat control is currently being studied and used

for other similar problems where nonoxidic conductive materials are

potentially the answer to the difficulties that have been encountered in

many applications.

1.12.2- Molecular structure in dehydrated state

The moisture adsorbed in supported vanadia catalysts under ambient

conditions may be removed by heating supported catalysts in non reducing

oxygen containing atmosphere at elevated temperature, (573-973 K). Such

treatment also helps maintain the (5+) oxidation state of the surface

vanadia species. At temperatures below (473 K), the Raman stretch of the

terminal (V=O) bond occurring at (1020-1040 cm-1) is shifted significantly

downward due to extensive hydration of the surface vanadia. At

temperatures above (473 K) a small amount of moisture hydrogen bonded

to surface vanadia results in a few cm-1 downward shift of the Raman


65

(V=O) band. At these temperatures (18O) labeling experiments showed that

the reversibly adsorbed moisture is also capable of rapidly undergoing

(18O) exchange with the terminal (V=O) bonds.

The structural studies suggested that the dehydrated vanadia species

on (Al2O3,TiO2, ZrO2, Nb2O5 and CeO2) was predominantly present as

isolated and polymeric tetrahedral VO4 units (Figure 1.24).The (18O)

labeling experiments of supported (VOx/ Al2O3, TiO2, ZrO2) showed that

these surface species possessed only one (terminal V = O) bond and three

bridging (Vـ�O-support) bonds for the isolated species, while the polymeric

surface species has only one (Vـ�O-support) bond per vanadium atom ,the

other two are being bridging V-O-V bonds (Figure 1.24). (166)

Spectroscopic techniques (X-ray absorption, Raman and UV-Vis.)

were utilized to monitor the effect of dehydration for vanadia supported on

(SiO2, Al2O3, TiO2, ZrO2 and HfO2) prepared with (VOx) surface densities

ranging from (0.46- 11.1 VOx/nm2) (166). UV-Vis. and near –edge spectra

indicate that supported (VOx) species exist in larger domains and have

higher – coordinate centers when hydrated. Dehydration consistently leads

to a breakup of these domains. Raman spectra suggest that (V=O) bonds in

mono and polyvanadate species are most susceptible to hydration. On all

supports, dehydration leads to the development of monovanadate species.

Polyvanadate on (TiO2, ZrO2, and HfO2) also undergo structural changes

when hydrated. Interpretation of this is more difficult because of the broad

Raman bands which allow only qualitative assignments. Support material

plays an important role in determining the extent of hydration, but this role

might arise from the ability to support polymeric vanadia species.


66

Figure 1.24: Molecular structure of:

(A) Dehydrated isolated

(B) Polymerized vanadia species on oxide supports (166).

1.12.3- Structural changes during Hydrocarbon Oxidation

Wachs et al (161) investigated the oxidation of n-butane to (MA) over a

series of model-supported vanadia catalysts where the vanadia phase was

present as a two- dimensional metal oxide overlayer on the different oxide

supports ( TiO2, ZrO2, CeO2, Nb2O5, Al2O3, and SiO2). No correlation was

found between the properties of the terminal (V = O) bond and the butane

oxidation turnover frequency (TOF) which is mean the number molecules

converted per vanadium per second , during in situ Raman spectroscopy

study. Furthermore, neither the n-butane oxidation (TOF) nor (MA)

selectivity was related to the extent of reduction of the surface vanadia

species. The n-butane oxidation (TOF) was essentially independent of the

surface vanadia coverage, suggesting that the n-butane activation requires

only one surface vanadia site. The (MA) TOF, however, increased by a

factor of (2–3) as the surface vanadia coverage was increased to monolayer

coverage. The higher (MA) (TOF) at near monolayer coverages suggests

that a pair of adjacent vanadia sites may efficiently oxidize n-butane to

Maleic Anhydride. Other factors may also play a contributing role such as

(increase in surface Bronsted acidity and decrease in the number of


67

exposed support cation sites. Varying the specific oxide support changed

the n- butane oxidation (TOF) by 50:

Ti > Ce > Zr ~ Nb > Al > Si

As well as the (MA) selectivity.

The (MA) selectivity closely followed the Lewis acid strength of the

oxide support cations:

Al > Nb > Ti > Si > Zr > Ce

The addition of acidic surface metal oxides (W, Nb, and P) to the surface

vanadia layer was found to have a beneficial effect on the n- butane

oxidation (TOF) and the (MA). The creation of bridging (V–O–P) bonds

had an especially positive effect on the (MA) selectivity .

1.12.4–Structure-Activity Relationships in Oxidation of

C4-hydrocarbons

Several studies on C4-hydrocarbon oxidation have been reported in the

literature (102,106,160).

Nakamura et al. (105) studied oxidation of 1-butene on alumina-

supported catalysts and reported reasonably high selectivities to (MA). The

results of their study showed that supported (VPO) catalysts could

selectively oxidize C4- hydrocarbons to (MA). They concluded that the

activity and selectivity of their supported catalysts for the oxidation of 1-

butene to (MA) were closely related to the oxidation state and the degree

of aggregation of vanadium ions. They implicated the (V4+/V5+) redox

mechanism of oxidation in which (V = O) bonds played an important role.

Mori et al. (102) studied oxidation of n-butane on titania-supported

vanadia catalysts at high temperature (693-763 K) and high n-butane

conversion. Such severe experimental conditions led to an over oxidation

of n-butane to (COx) and, likely, to combustion of the partial oxidation

products, including (MA). They found that the reaction rate under such


68

conditions was proportional to the amount of (V 5+= O) species in the

catalyst and concluded that the reaction proceeded via the reduction-

oxidation or Mars-van Krevelen mechanism according to which the

surface (V = O) species contained the active oxygen.

Busca et al. (167) employed titania-supported vanadia catalysts in n-

butane oxidation and observed over oxidation of the hydrocarbon on very

active high surface area catalyst (117m2/g) at (723 K). When they

supported (10 wt% V2O5) on the low surface area Titania (18.4 m2/g),

vanadia formed micro crystals which were detrimental to selective

oxidation and led to n-butane combustion.

Owens and Kung (160) studied n-butane dehydrogenation on silica-

supported vanadia catalysts at (793) K. They observed that the isolated

vanadia species present at low surface coverage (0.53 and 0.58 wt% V)

were responsible for the high total dehydrogenation selectivity (1-butene,

cis-/trans-2-butene, and 1, 3-butadiene as shown in (Figure 1.25) (160).

1 2 3 4

Figure 1.25: Oxidation of butane on dispersed vanadia species (160).

The presence of the crystalline V2O5 species at high surface coverage

(998, 703, 526, 480, 404, 304 and 284 cm-1) Raman bands for the (6.4 wt%

V) catalyst contributed to the production of total oxidation products.

C4H9


69

1.12.5- Role of Surface Oxygen Species in Supported Catalysts

The terminal (V = O)oxygen has been proposed by investigators (161,162 )

to be the active oxygen involved in hydrocarbon oxidation over supported

vanadia catalysts. However, the combination of (in situ Raman and

hydrocarbon oxidation reactivity studies) have recently suggested that the

reaction properties are not related to the characteristics of the terminal

(V = O) bonds in supported vanadia catalysts. Butane oxidation over a

series of supported vanadia catalysts was found to vary by over an order of

magnitude in the butane turn over frequency (TOF) (Table1.4) (161), but

identical (V=O) Raman features were observed for these catalysts (V=O

stretch at (1025-1032cm-1) (Figure 1.26 A-D ) .

Table 1.4: the effect of the metal oxide support on n-Butane oxidation on

supported Vanadia Catalysts at 494 K in 1.2 Vol. % n-Butane (161).

Catalyst Wt.

(g)

Flow

cm3/min.

Butane

Conversion

mol%

MA

Selectivity

mol%

Butane

(TOF)

10-5 s-1

MA

(TOF)

10-5 s-1

7% V2O5/SiO2 0.577 7.4 2.8 91.8 0.4 0.3

17.5% V2O5/Al 2O3 0.625 17.8 7.2 44.5 0.9 0.4

6% V2O5/Nb2O5 0.885 13.9 17.3 36.7 3.6 1.3

4% V2O5/ZrO2 0.794 11.3 16.0 9.3 4.5 0.4

3% V2O5/CeO2 0.622 14.2 10.6 12.6 6.3 0.8

5% V2O5/TiO2 0.566 25.5 27.8 30.5 19.6 6.0


70

Figure 1.26 A- In situ Raman spectra of 1% V2O5/TiO 2 catalyst during n-butane


/min.,503K;(c)

C4H10/O2/He, 100cm3/min.573 K ; (d) C4H10/O2/He, 100cm3

/min. 623 K; (e)

C4H10/O2/He, 50cm3/min. 623 K; (f) O2 623K (161).

Figure 1.26 B- In situ Raman spectra of 7% V2O5/SiO2 catalyst during n-butane


/min.,503K;(c)


/min. 623 K; (e)

C4H10/O2/He, 50cm3/min. 623 K; (f) O2 623K (161).


71

Figure 1.26 C- In situ Raman spectra of 7% V2O5/Nb2O5 catalyst during n-butane


/min.,503K;(c)


/min. 623 K; (e)

C4H10/O2/He, 50cm3/min. 623 K; (f) O2 623K (161) .

Figure 1.26 D- In situ Raman spectra of 4% V2O5/ZrO 2 catalyst during n-butane


/min.,503K;(c)


/min. 623 K; (e)

C4H10/O2/He, 50cm3/min. 623 K; (f) O2 623K (161) .


72

Furthermore, 18 O – labeling of the terminal (V = O) bond, during the

n-butane oxidation revealed that this bond is very stable and has an

exchange time that is approximately (20) times longer than the

characteristic reaction time. The (18O-labeled 4 wt% V2O5/ZrO2) catalyst

exhibited the (V = 18O) Raman stretch at (983 cm-1) which disappeared

only after (25min.) of n-butane oxidation (Figure 1.26-d (161)). Therefore,

the available data suggests that the terminal (V = O) oxygen is not involved

in kinetically significant reaction steps of hydrocarbon oxidation over

supported vanadia catalysts.

The surface concentration of bridging (V-O-V) bonds increases with

surface vanadia coverage due to the increase in the ratio of polymerized to

isolated surface vanadia species (Figure 1.24). The (V2O5/SiO2) system

represents one notable exception to this rule, since microcrystalline V2O5 is

formed at above monolayer coverage (Figure 1.26-b) (161). The (TOF) for

the oxidation of butane to (MA) over (V2O5/TiO2) was found to slightly

increase, by a factor of (2-3), with surface vanadia coverage (Figure 1.27)

because of the requirement of several surface vanadia sites for this

hydrocarbon.

Figure 1.27: performance of Titania support in n-butane oxidation on supported

vanadia catalyst at 494 K in 1.2 vol. % n-butane in air (161).


73

During n-butane oxidation to (MA) the Raman signal of the surface

V(V) species decreased (10-35%) reflecting partial reduction of the

vanadia species under reaction conditions (V =O and V-O-V stretches)(163).

The average oxidation state of vanadium in the selective (VOx /TiO2)

catalyst for partial oxidation of n-butane was found to be near (+ 4.5). The

surface vanadia coverage was also found to be a critical variable as the

polymeric surface vanadia species present at high coverage are more easily

reducible than the isolated vanadia species. The reducibility of the surface

vanadia species during butane oxidation on various oxide supports

followed the order:

TiO2 >CeO2 >ZrO2 >Al2O3 >SiO2

However, the butane oxidation (TOF) was not found to correlate with the

extent of reduction of the surface vanadia species (Table 1.5).

Table 1.5: The effect of acidic promoters on n-Butane Oxidation on the

1% V2O5 /TiO2 catalyst at 494 K in n-Butane in air (161).

Catalyst Wt.

(g)

Flow

cm3/min

Butane

Conversion

mol%

MA

Selectivity

mol%

Butane

(TOF)

10-5 s-1

MA

(TOF)

10-5 s-1

1% V2O5/5%P2O5/ TiO2 0.172 5.0 12.1 56.2 27.0 15.2

6% WO3/1%V2O5/ TiO2 0.708 8.9 23.6 26.2 34.1 8.9

9% WO3/1%V2O5/ TiO2 0.465 11.1 14.8 6.8 40.5 2.8

6% Nb2O5/1%V2O5/ TiO2 0.181 7.5 10.7 35.1 50.8 17.8

The (MA) selectivity did not appear to directly correlate with the extent of

reduction either, since the selectivity pattern was:

Al 2O3 >Nb2O5 >TiO2 >SiO2 >ZrO2 ~ CeO2

Varying the specific oxide support or oxide support ligands without

hanging the structure of the surface vanadia species can alter the

characteristics of the bridging (V-O-support) bond. The bridging (V-O-


74

support) bond appears to be associated with the critical oxygen required

for hydrocarbon oxidation reactions since changing the specific oxide

support dramatically affects the (TOF) approximately two orders of

magnitude for n-butane oxidation to (MA) (Table1.4). The general trend

appears to be:

CeO2 >ZrO2 ~TiO2 >Nb2O5 >Al2O3 >SiO2

This suggests that bridging oxygens in the (V-O-support bonds) that are

more electronegative or basic, corresponding to oxide support cations with

lower electronegativity, are associated with the critical oxygen required

for hydrocarbon oxidation reactions over supported vanadia catalysts. The

formation of the (V-O-P) bond has a particularly positive effect on the

butane oxidation (TOF) and (MA) selectivity (Table 1.6). The

(1%V2O5/5% P2O5/TiO2) catalyst displayed the highest selectivity to (MA)

among all systems studied (Table 1.5). At (494 K in 1.2 vol. % n-butane in

air) and (12.1 mol% n-butane conversion), the selectivity to (MA) reached

(56.2 mol %). Such catalytic behavior is consistent with the above

observation that (bridging V-O) support bonds are critical in the oxidation

of n-butane to (MA).

The insight into the number of critical surface vanadia sites required in

hydrocarbon oxidation reactions can be gained by examination of the

variation of the (TOF) with surface vanadia coverage. Those reactions that

require only one surface site will exhibit a (TOF) that is independent of

surface vanadia coverage .The oxidation of butane to (MA) over titania-

supported vanadia catalysts (Table 1.6 and Figure 1.27) exhibited an

increase in (TOF) with surface vanadia coverage. This may reflect the

requirement of multiple surface vanadia sites or the influence of other

factors, such as surface acidity influence of (bridging V-O-V bonds),

structural changes, etc...


75

Table 1.6: Performance of Titania support in n-Butane oxidation at 494 K in 1.2

Vol. % n-butane in air (161).

Catalyst Wt.

(g)

Flow

m3/min

Butane

Conversion

mol %

MA

Selectivity

mol %

Butane

(TOF)

10-5 s-1

MA

(TOF)

10-5 s-1

1%V2O5/TiO2 0.890 8.7 16.0 22.8 12.4 2.8

2% V2O5/TiO2 0.612 17.6 24.2 9.9 27.7 2.7

3% V2O5/TiO2 0.603 19.9 27.0 17.0 23.4 4.0

4% V2O5/TiO2 0.700 30.7 30.7 29.9 25.5 6.6

5% V2O5/TiO2 0.566 0.566 27.8 30.5 19.6 6.0

7% V2O5/TiO2 0.572 0572 18.5 24.7 8.0 2.0

Information about the number of surface sites required for hydrocarbon

oxidation reactions can be also probed by the addition of non interacting

surface metal oxides, which preferentially coordinate with the oxide

support rather than the surface vanadia species under dehydrated

conditions .Typical noniteracting oxides are surface oxides of W, Nb, S,

Si, Mo, Ni, Co and Fe .Noniteracting metal oxides only indirectly affect

the molecular structure of the surface vanadia species via lateral

interactions .Such lateral interactions influence the relative concentration

of polymerized and isolated vanadia species in supported metal oxides.

Interacting acidic oxides, such as P2O5 significantly increase the (TOF) for

butane oxidation to (MA) (161). The enhancement in n-butane oxidation

(TOF) and (MA) selectivity upon introduction of acidic oxides further

confirms the positive role of acidity in this reaction. The butane oxidation

(TOF) was increased by a factor of (2 and 3) when acidic metal oxides,

such as WO3 and Nb2O5, were introduced to the (V2O5/TiO2) system (Table

1.5). For example, introduction of surface niobia species to (V2O5/TiO2)

catalysts led to a threefold increase in the (TOF) for the oxidation of

butane to (MA) (161). This increase in (TOF) also reflects the requirement of


76

multiple surface metal oxide sites for this hydrocarbon oxidation. In

contrast to non interacting additives that mainly affect oxidation reactions

requiring multiple surface sites, interacting additives affect all hydrocarbon

oxidation reactions since they directly alter the structure and reactivity of

the surface vanadia sites. Interacting metal oxides, especially basic oxides,

retard the reduction of surface vanadia species. Consequently, the (TOF)

for all hydrocarbon oxidation reactions are lowered when basic additives

are introduced (163,164).

1.12.6 - Role of Acidity of Supported Vanadia Catalysts

The oxide supports possess only surface Lewis acid sites, with the

exception of silica where no Lewis acid sites exist on the surface. The

relative Lewis acid strength of these sites follows the order

Al 2O3 >Nb2O5 >TiO2 >ZrO2

In contrast to the oxide supports, unsupported V2O5 crystalline powders

displays both surface Bronsted and Lewis acid sites (167) and an increase in

the number of surface Bronsted acid sites.

Supported (VPO) catalysts typically exhibit surface enrichment in

phosphate similar to the conventional unsupported or bulk (VPO)

catalysts (106,156,161). In these catalysts the concentration of the surface

Bronsted acid sites increases with the surface (P/V) ratio. The surface

enrichment in phosphorus stabilizes the V (IV) oxidation state, which

results in decreased activity in n-butane oxidation. On the other hand, the

increased surface Bronsted acidity facilitates desorption of partially

oxidized products with acidic properties, such as (MA) leading to

significant improvement in (MA) selectivity (162).According to Grasselli st

al. (168), slight excess phosphate (P/V = 1.1-1.2) forms a protective

pyrophosphate "fence" around active surface sites at the surface preventing

overoxidation of the reactive intermediates by the active oxygen diffusing


77

fast at the surface. However, the changes in the activity and selectivity of

the (VPO) catalysts at much higher phosphorus enrichment (P/V ratio near

2.0) were associated with formation of microcrystalline VO (PO3)2 (169).

Reports in the literature have demonstrated that the VO (PO3)2phase is

inactive in n-butane oxidation (169). Therefore, the catalytic behavior of the

VO (PO3)2 system is explained by the presence of other (VPO) impurity

phases, such as (VO)2P2O7 and various (VOPO4) phases (169). The oxidation

of n-butane on well-defined supported vanadia catalysts demonstrated that

the substitution of (V-O-Ti) for less reducible and more acidic (V-O-P)

bonds has a positive effect on the n-butane oxidation (TOF) and (MA)

selectivity . This observation indicates that the oxidation of n-butane

to (MA) requires both a redox site and some acidic functionality (170 ) .

1.12.7: The Role of Support

We can conclude that although supports are essentially diluents, they

play important multifunctional roles which can be summarized as

following:

1- Economic, to reduce cost by extending expensive phase in catalyst.

2- Mechanical, to give mechanical strength, to optimize bulk density, to

prove a heat sink or a heat source, and to dilute the overactive phase.

3- Geometric, to increase the surface area, to optimize the porosity of a

catalyst, to optimize crystal and particle size, to allow the catalyst

particles adopt the most favorable configuration of a catalyst.

4- Chemical, to react with the catalyst either to improve specific activity

or to minimize sintering and to accept or donate chemical entities,

porosity via a spillover mechanism.

5- Deactivation, to stabilize the catalyst against sintering and to minimize

poisoning.


78

Conclusions

Significant progress had been achieved in recent years in studying

molecular structures of the surface vanadia species present in supported

metal oxide catalysts. The detailed structural information on well-defined

systems provided a foundation for developing structure-reactivity

relationships required to molecularly engineer supported vanadia catalysts

for oxidation reactions. The nature of the metal oxide support was found to

play a crucial role in defining catalytic properties of vanadia monolayer in

n-butane oxidation to (MA). The terminal vanadyl oxygen does not appear

to critically influence the reactivity properties of the surface vanadia

species during hydrocarbon oxidation reactions. The bridging (V-O-V)

oxygen plays only a minor role in enhancing the n-butane oxidation (TOF),

primarily due to the preference of multiple active sites in this oxidation

reaction. The bridging (V-O-support oxygen), however, appears to be the

most critical oxygen since its properties can change the (TOF) for

hydrocarbon oxidation reactions by as much as four orders of magnitude.

The specific phase of the oxide support as well as the specific preparation

method does not appear to influence the molecular structure or reactivity

of the surface vanadia species. The number of surface vanadia sites

required for a hydrocarbon oxidation reaction is dependent on the specific

reactant molecule. Oxidation reactions requiring only one surface vanadia

site are generally not sensitive to the surface vanadia coverage and the

presence of non interacting metal oxides. Oxidation reactions requiring

multiple surface vanadia sites are very sensitive to surface vanadia

coverage and the presence of non interacting metal oxides. However,

interacting metal oxides influence all hydrocarbon oxidation reactions

since they modify the surface vanadia sites. Acidic and basic metal oxides

also influence the selectivity of hydrocarbon oxidation reactions, but the

effect appears to be reaction-specific.


79

Unlike bulk (VPO) catalysts, detailed surface structural information on

a molecular level can be obtained from model supported vanadia catalysts

containing two-dimensional over layers of surface vanadia species (172).

Raman spectroscopy provides direct fundamental surface information

about:

1- The ratio of isolated and polymerized surface vanadia species

(Fig.1.24).

2- Terminal (V=O) and bridging (V-O-V) bonds.

3- Extent of reduction of the surface vanadia species during catalysis.

4- Influence of the oxide support ligands.

5- Influence of (acidic/basic) metal oxide additives (promoters/poisons).

6- Participation of specific (V- O) bonds in catalysis (with the aid of

oxygen- 18O labeled isotope experiments.

Thus, the in situ Raman studies during oxidation of reactions over model

supported vanadia catalysts can provide new insights into the surface

properties of oxide catalysts which are not attainable with bulk metal oxide

catalysts.

The studies of the model supported catalysts provided several important

insights about the origins of the catalytic activity of the (VPO) system that

were not possible with the bulk (VPO) catalysts. Firstly, it was shown that

the (V=O) oxygen was not involved in n-butane oxidation to (MA) and

that the (V-O-support) bonds contained active oxygen for n-butane

oxidation. The (V-O-P) bonds were particularly beneficial for both the

activity and selectivity of the supported vanadia catalysts (Table 1.6 (163)),

suggesting that this oxygen species may be responsible for selective

oxidation of n-butane. Secondly, it was demonstrated that the oxidation of

n-butane to (MA) could occur at a single vanadia site, although adjacent

sites were more efficient (Figure 1.27).


80

The results of n-butane oxidation on titania-supported vanadia

catalysts suggested that isolated surface vanadia species are capable of

n-butane oxidation to (MA), although multiple vanadia sites are more

efficient in this oxidation. Microcrystalline vanadia was found to be

detrimental for the process of (MA) formation. The kinetic studies of the

supported vanadia system provided more direct evidence that the multiple

vanadia sites were better at oxidizing n-butane to (MA).

The selectivity of the supported catalysts to (MA) is correlated with

the Lewis acid strength of the metal oxide promoters. Especially high

selectivity to (MA) was found when the (V-O-P) bonds formed after

addition of phosphorus oxide in accordance with previous observations in

supported and unsupported bulk (VPO) catalysts. These findings indicate

that the supported vanadia catalysts represent a suitable model system

capable of providing insights into the mechanism of n-butane oxidation on

bulk (VPO) catalysts.

Lastly, the use of acidic additives with supported (VPO) catalysts

further demonstrated their similarity to the bulk (VPO) system. Moreover,

it shed some light on the possible role of promoters that improve the

performance of the bulk (VPO) catalysts. The acidic additives used in the

supported vanadia catalysts enhance those catalytic oxidations which

require a combination of a redox and acidic site for selective catalysis. The

acidic additives promoted both the rate of n-butane oxidation and the (MA)

formation on supported vanadia catalysts (Table 1.6).

The supported (VPO) system proved to be a good model for the bulk,

i .e. unsupported (VPO) system. The in situ Raman studies of the

supported vanadia catalysts provided several important insights into the

nature of the selective n-butane oxidation that were not possible with

unsupported vanadium-phosphorus-oxygen (VPO) catalysts due to the

experimental limitations. The many discovered similarities between the


81

bulk and supported (VPO) systems have important implications for the

design of new selective (VPO) catalysts. The nature of the oxide support,

the surface coverage of the active vanadia species and the acidity of metal

oxide additive are the most important determinants of the catalytic activity

and selectivity in n-butane oxidation. These structure-reactivity

relationships pave the way for molecular engineering of selective active

sites for hydrocarbon oxidation in supported (VPO) catalysts.

The aim of work

The aim of our previous work was precisely to use technical grade

vanadium pentoxide locally produced to prepare a catalyst to convert

n-butane to MA.

Our aim in this work is to study the effect of Titania (anatase) on the

physical, chemical and structural properties of our previous already

tested catalyst in order to prepare an industrial catalyst to increase the

feasibility of vanadium pentoxide production in our country and supply

further prospects to use it.

Chapter Two - Experimental Part

٩٣

2.4: REDOX TITRATION

2.4.1: Redox Titration experiments

The redox titration for the technical grade V2O5 and prepared catalytic

phases was carried out as described below:

A- Preparation of the samples for Redox Titration

According to Nakamura (105), (0.1) g of each solid sample was dissolved

in (17 ml) o-H3PO4 and boiled until a clear solution was obtained for the

unsupported phases and a dissolved supported samples. This solution was

added to a solution containing (10) ml H2SO4 which was diluted with

(250) ml distilled water. This solution was divided into two parts (100 ml)

for each part.

B- Method of titration

After standardization of potassium permanganate KMnO4 solution with

oxalic acid (0.05N), one of the above two parts of solution was titrated

with Potassium Permanganate KMnO4 to the end point. Then the

titration was continued with (0.05N) iron (II) ammonium sulphate

Fe (SO4)2(NH4)2.6H2O using diphenyl amine as indicator.

The other part of solution in section (A) was titrated with (0.05N) of iron

(II) ammonium sulphate Fe (SO4)2(NH

4)2.6H

2O after the addition of one

drop of diphenyl amine as indicator.


٩٤

C- Method of Calculation

According to Buchanan (92) the average oxidation number of Vanadium

was calculated according to this equation:

OHNHSOFeofVol

mlKMnOofVolnV 26.)4()4(.

)(.5

22

4−= .............. 2. 1

The amount of V (III), V (IV) and V (V) was calculated according to

the following equations:-

T1 = V (IV) + 2V (III) ………………………… (2.2)

T2 = V (V) + V (IV) + V (III) ………………... (2.3)

T3 = V (V) …………………………………. (2.4)

Where:

T1 = Vol. of KMnO4 added to the first part of solution represents

the amount of V (IV) and V (III).

T2 = Vol. of Fe (SO4)2(NH4)2.6H2O added to the first part

solution represents the amount of V (V), V (IV) and V (III).

T3 = Vol. of Fe (SO4)2(NH4)2.6H2O added to the second part of

solution represents the amount of V (V).


٩٥

2.4.2: Redox Titration of reaction evolution:

A redox titration was conducted for supported VO (H2PO4)2 during the

reaction evolution (i.e. during the preparation of the phase), with interval

of reaction time of one hour, up to 7 hours. The investigated samples were

filtered, dried at 125 °C and then treated in same method described in

section 2.4.1.

Chapter Three

Results and Discussion

Part One

Results

Chapter Three – Part One – Results

٩٨

3.1: Results

The results obtained from using the techniques mentioned in chapter

two are highlighted in this section in reserving the sequences of analysis in

order to follow systematic and logic presentation of data, this approach can

facilitate results prospective prior to effective and global discussion.

3.1.1: The Elemental Analysis of technical grade V2O5

Table 3.1 represents the elemental analysis of technical grade V2O5 for

vanadium and impurities percentage which carried out by atomic

absorption.

It is obviously from the table that the higher concentration of the

impurities is represented by Sodium.

Table 3.1: Elemental Analysis of technical grade V2O5 used in this work.

V% Na% Al% Si% Ca% Fe% Mg% K% Ni% Mo% Remarks

52.520 0.770 0.170 0.060 0.160 0.060 0.050 0.041 0.006 0.017 Sample *

52.520 0.780 0.170 0.070 0.170 0.070 0.050 0.039 0.008 0.017 General

Sample**

* Sample: is the main source of material used in this research

**General Sample: random sample from V2O5 production line inventory


٩٩

3.1.2: Identification of prepared Phases by XRD

The X- ray diffraction pattern of VOHPO4.0.5H2O, VHPO4.H2O/ TiO2,

VO (H2PO4)2 and VO (H2PO4)2 /TiO2 (samples No.2-un, 2-supp, 3-un and

3-supp) are shown in Figure 3.1. The initial sample was used for the

calibration of the X- ray Diffractometer this sample is known (VPO)

system catalyst which is used for Propane Oxidation. The identification of

these phases is given in (Table 3.5), while the X-ray diffraction data of the

above samples (unsupported, supported and reported) are given in tables

3.2, 3.3and 3.4 respectively.

100

2Theta12.0 16.0 20.0 24.0 28.0 32.0 36.0

0

4000

8000

12000

16000

20000

24000

28000

Abs

olut

e In

tens

ityC:\Documents and Settings\ushkalov\REZULTATI\XRD\vpo\040824D1R_VP1_SolX_2_STOE.raw / VP1_ (Range 1)

C:\Documents and Settings\ushkalov\REZULTATI\XRD\vpo\zazhigalov\041027D1R_1z_STOE.raw / 1z (Range 1)C:\Documents and Settings\ushkalov\REZULTATI\XRD\vpo\zazhigalov\041027D2R_2z_STOE.raw / 2z (Range 1)C:\Documents and Settings\ushkalov\REZULTATI\XRD\vpo\zazhigalov\041027D3R_3z_STOE.raw / 3z (Range 1)C:\Documents and Settings\ushkalov\REZULTATI\XRD\vpo\zazhigalov\041027D4R_4z_STOE.raw / 4z (Range 1)

Initial

2- un

2-Supp

3 - un

Figure 3.1: XRD Spectrum of (sample No.2 un , sample No.2-Supp, sample No.3-un and sample No. 3-supp).

3-supp

Chapter Three – Part One - Results

١٠١

Table 3.2: VOPO4.2H2O; sample (No.1-un) X-Ray Diffraction Data.

Sample No.

Reported reference (172) 1- un

2θ d (A°) I / I ° d (A°) I / I °

12.00 7.38 100 7.450 W

18.65 4.77 23 4.760 S

20.20 4.37 8 4.370 Vw

24.05 3.70 74 3.700 Vvw

28.05 3.18 6 3.180 Vvw

28.80 3.10 30 3.105 M

31.30 2.87 7 2.863 Vw

37.75 2.38 4 2.833 Vw

2.397 Vw

2.197 W

2.108 W

1.995 M

1.965 M

1.900 W

1.802 Vw

1.554 S

1.519 S

1.461 M

1.385 M


102

Table 3.3: VOHPO4.0.5H2O; samples No. (2 – un)& (2- supp) X- Ray Diffraction Data.

Sample No. Reported ref. (57)

2 – un 2- supp **

2θ d (A°) I / I ° 2θ d (A°) I / I ° d (A°)* I / I °*

11.5 7.48 8 11.5 7.485 100 5.901 3

13.1 6.574 100 13.00 6.625 64 5.719 100

13.7 6.288 38 13.6 6.334 9 4.818 2.0

16.42 5.252 8 16.56 5.204 9 4.535 40

19.00 4.544 8 4.260 18 4.099 5

21.30 4.058 11 21.34 4.050 9 3.684 23

23.71 3.625 8 23.72 3.649 9 3.300 32

26.35 3.290 15 25.34 3.649 27 3.116 18

28.42 3.055 8 26.50 3.419 9 2.944 34

28.57 3.039 15 28.40 3.272 18 2.796 10

28.92 3.003 15 28.56 3.057 36 2.663 27

29.28 2.967 23 29.00 2.995 36 2.615 7

31.70 2.746 23 29.32 2.963 13 2.567 5

2.734 11 11 2.454 6

2.405 4

2.263 6

2.234 5

2.212 2

2.131 5

2.055 4

2.047 6

1.960 1

1.903 7

1.864 3

1.854 9

1.836 8

1.808 1

* = By Johnson et al. (57) for single crystal.

** = Data; not reported in the literature.

Chapter three – Part One - Results

103

Table 3.4: VO (H2PO4) 2, samples (No.3 – un & 3- supp) X- Ray Diffraction Data.

Sample No. Reported

reference (172) Reported

reference ( 80) 3- un 3- supp ** 2θ d (A°) I / I ° 2θ d (A°) I / I ° d (A°) I / I ° d (A°) I / I °

11.50 7.485 18 13.28 6.483 100 6.34 M 6.327 100 13.28 6.486 100 13.72 6.279 7 3.99 S 4.482 4 13.72 6.279 3 14.28 6.031 10 3.586 S 3.988 12 14.28 6.031 2 14.57 5.914 17 3.378 M 3.637 1 14.71 5.858 13 16.57 5.204 20 3.173 F 3.581 29 15.28 5.641 2 21.42 4.035 2 2.983 M 3.375 8 16.57 5.204 3 24.90 3.478 7 2.837 F 3.167 75 18.64 4.630 1 26.71 3.246 8 2.483 M 2.977 5 21.42 4.034 3 29.00 2.995 10 2.375 W 2.835 45 22.57 3.832 7 29.57 2.938 5 2.312 Vw 2.480 9 23.00 3.761 2 30.00 2.969 7 2.215 Vw 2.371 4 25.42 3.407 1 31.71 2.744 2 2.098 M 2.213 1 26.78 3.230 3 33.85 2.576 7 2.007 W 2.111 5 27.42 3.163 5 1.995 W 2.096 19 27.57 3.147 5 1.955 Vvw 2.003 9 29.00 2.995 3 1.945 Vw 1.954 2 29.57 2.938 3 1.902 W 1.945 1 30.14 2.884 3 1.867 Vw 1.900 1 30.71 2.832 1 1.785 W 1.887 3 31.71 2.742 1 1.758 W 1.758 6 32.71 2.663 1 1.688 W - - 33.85 2.575 1 1.683 M 34.71 2.514 4 1.584 W

** = Data; are not reported in the literature.


104

Table 3.5: Identification of prepared phases by XRD

Sample No. Identified Phases

2 –un VOHPO4.0.5H2O▼+ VOHPO4.H2O ■ + VO(H2PO3)2

■

2 – supp VOHPO4.H2O ▼ + VO(H2PO4)2■ +TiO2 (a)

3 – un VO (H2PO4)2 ▼ + VO(H2PO3)2

■

3 – supp VO(H2PO4)2 ▼ + VO(H2PO3)2

■ VH2P3O10.2H2O ■ + TiO2 (a)

▼ = Main phase. ■ = Trace amount.

a = Anatase.

3.1.3: Thermal Analysis:

The results of Thermogravimetric and differential thermal analysis for the

supported and unsupported catalytic phases are presented in (Figs. 3.2- 3.5)

and (Tables 3.6 and 3.7.

Reported date in the literatures is numerous and some of them are in

contradictory due to the analysis environments like inert atmosphere, air,

nitrogen, oxygen, helium and argon. Our results are performed in air and are

in accordance with related reported data and the rate of heating (36, 57, 81, 80-83).


١٠٥

Figure 3.2: TG & DTA of VOHPO 4.0.5H2O (sample No. 2-un).


١٠٦

Figure 3.3: TG & DTA of VOHPO 4. H2O / TiO2 (sample No. 2-supp).


١٠٧

Figure 3.4: TG & DTA of VO (H 2PO4)2 (sample No. 3- un).


١٠٨

Figure 3.5: TG & DTA of (H 2PO4)2 / TiO2 (sample No. 3- supp).


١٠٩

Table 3.6: DTA Analysis Results*

Sample NO.

Initial weight ( mg)

Temperature range °C

Change of the sample weight Effect mg %

2- un 98

50-237 -8 -8.16 Little endo(205-230°C) 237-340 -1.7 -1.73 Endo (240-310 °C) 340-500 -4.8 -4.89 Endo (310-510 °C)

Total -14.5 -14.79

500-580 Without the change

of weight Endo (500-580 °C)

580-700 1 1.02 Exo- (580-630 °C)

2- supp 100

30-230 -5.7 -5.7 Not (180-220) 230-330 -2.0 -2.0 Endo- (200-320°С) 380-510 -6.2 -6.2 Endo- (320-510°С)

Total -13.9 -13.9


of weight Endo -(510-600°С

625-700 1 1 Exo- (600-660°С)

3-un

100

47-277 -6 -6 Two little endo (50-270 °С)

277-350 -0.7 -0.7 Not (277-350 °С) 350-470 -5.3 -5.3 Endo (370-490°С)

Total -12 -12


of weight Endo (490-560°С

570-640 1 1 Exo (560-620°С)

3-supp 101

45-255 -8.8 -8.71 Endo (110-180°С) 255-340 -0.5 -0.50 Not (255-340) 340-460 -4.7 -4.65 Endo (350-440°С)

Total -14 -13.86


of weight Endo (470-560°С)

560-700 1 0.99 Little exo (560-620°С)

* Analysis is carried in air; rate of heating is 10 oC / min.


١١٠

Table 3.7: The comparison between theoretical and experimental loss of weight % for catalytic phase.

Sample No.

Theoretical loss of

weight %

Experimental loss of

weight %

Starting Identified phase

Theoretical loss of weight %

Ref.

2-un 10.5 * 14.79

VOHPO4.0.5H2O▼

+ VOHPO4.H2O ■

+ VO(H2PO3)2

■

Much higher than theoretical loss of

weight

36 59 82

9.18 – 12.23 *** 57

2- supp 11.5 13.90

VOHPO4.H2O ▼

+ VO(H2PO4)2

■ +

TiO2 (a)

Not reported

3-un

13.7 **

12.00 VO (H2PO4)2

▼ +

VO(H2PO3)2 ■

13.86 80

3-supp 13.56

VO(H2PO4)2 ▼

+ VO(H2PO3)2

■ +

VH2P3O10.2H2O ■

+ TiO2 (a)

Not reported

*** = A* = VOHPO4.0.5H2O to (VO)2P2O7 (57, 83).

** = VO (H 2PO4)2 to VO (PO3)2 (80).

According to the preparation method, time in flow, and the range of

temperature.


١١١

3.1.3- Fourier Transform Infrared (FTIR) Spectrum

Figures 3.6 and 3.7 show the FTIR spectrum of laboratory grade

(Fluka) and technical grade V2O5.The observed bands frequencies and

their vibrational assignments for V2O5 are shown in (Table 3.8).

Figures 3.8 and 3.9 represent the FTIR spectrum of laboratory grade

(Fluka) and technical grade V2O5 supported on Titania (anatase), all bands

were listed in (Table 3.9).

The spectrum of VOPO4.2H2O sample (No.1-un and 1-supp) is given in

Figures 3.10 and 3.11.The phase’s bands, for the above samples are

reported in (Table 3.10).

Figures 3.12, 3.13 and 3.14 can show the FTIR spectra recorded for

VOHPO4.0.5H2O and VOHPO4.H2O/TiO2 phase, reference sample

(sample No.2 , No. 2- un and sample No.2-supp) the bands positions are

tabulated in (Table 3.11).

The FTIR of spectrum of VO (H2PO4)2, samples (No. 3- un and No. 3-

supp) are shown in (Figures 3.15 & 3.16) respectively. The positions and

the assignment of the FTIR bands for this phase are given in (Table 3. 12).

Chapter T

hree – Part O

ne – Results

Figure3.6: FTIR Spectrum of Fluka V2O5

112

Chapter T

hree – Part O

ne - Results

Figure 3.7: FT IR Spectrum of Technical Grade V2O5.

113


114

Table 3.8: FT-IR Data of V2O5

Band Frequencies (cm -1) Attribution Reported Data Ref.

Fluka V2O5 T.G. V2O5

N.B 3450 (Sh) ν O-H (H2O) 3400 172

N.B 3198 (M) δ N-H ( NH4)+

3400-3180 172 3300-3030 174

N.B 2941(Sh) δ O-H (H2O)

3020-2270 172 N.B 2797 (Vw) δ O-H (H2O)

2366 (W) 2380 (w) δ O-H (H2O)

1649 (Vw) 1632 (Vw) δ O-H (H2O) 1640 62

1554 (Vw) N.B δ O-H (H2O) 1590 - 1500 172

1520 (Vw) 1520 (Vw) δ O-H (H2O) 1590 - 1500 172

N.B 1413 (Vs) δ N-H ( NH4)+

1300 - 1430 172 Near 1429 174 Near 1450 189

1016 (Vs) 1001 (M) δ V=O * 1024 1018

1050-800

176 189

N.B 966 (S) ν V=O 172 175 176

N.B 941 (W) δ M=O * 1000-850 176 Near 1000 189

N.B 895 (M) δ M=O * 1000-850 176 Near 1000 189

825 (Vs) 849 (S) ν V-O-V

818 - 850

172 175 176

N.B 731 (S) δ N-H 850-750 176

540 530 (W) δ V - O 520 - 540 172

453 467 (M) δ V - O 470 172

N.B 413 (W) δ V - O 416 62

*M = Na, Si, Fe, Ti ……etc

Chapter T

hree – Part O

ne - Results

Figure 3.8: FTIR Spectrum of Fluka V2O5 / TiO2 (anatase).

115

Chapter T

hree – Part O

ne – Results

Figure 3.9: FTIR Spectrum of Technical Grade V2O5 / TiO2 (anatase).

116


117

Table 3.9: FTIR Data of Supported V2O5.

Band frequencies Attribution

Band frequencies in Reported Data

Ref. Fluka V2O5 / TiO2(a) T.G. V2O5/ TiO2(a)

N.B 3180 (S) ν N-H ( NH4)+ 3400-3180 172

3300-3030 174

1649.0 (Vw) 1632 (W) δ O-H (H2O) 1645 57 1640 62

1520 (Vw) 1520 (Vw) δ O-H (H2O) 1625-1591 172

N.B 1414 (Vs) δ N-H ( NH4)+

1430-1395 172 Near 1429 174

1016 (Vs) 1001 (m) ν V=O 1024 1018

1050-800

176 189

N.B 966 (S) δ V = O 172 173 176

N.B 941 (W) δ M=O 1000-850 176 Near 1000 189

N.B 895 (m) δ M=O 1000-850 176 Near 1000 189

825 (S) 849 (S) ν V-O-V 818-850 172 176

N.B 731(S) δ N-H 850-750 176

N.B 683 (Sh) ω O-H (H2O) 686 62

478 (Sh) 467 (m) δ V=O 470-430 172

N.B 413 (W)

M= Na, Si, Fe, Ti ……………etc

Chapter T

hree – Part O

ne - Results

Figure 3.10: FTIR Spectrum of VOPO4.2H2O sample (No. 1- un).

118

Chapter T

hree – Part O

ne - Results

Figure 3.11: FTIR Spectrum of VOPO4.2H2O / TiO2 sample (No. 1- supp).

119


١٢٠

Table 3.10: FT-IR Data of VOPO4.2H2O

Band Frequencies (cm-1 )

Attribution

Reported Data

References Sample

(No. 1- un) Sample *

(No. 1- supp)

3747 (M) 3747 (M) ν O-H (H2O) ٣٧٠٠ 174

3531 (m) 3531.4 (m) ν O-H (H2O) ٣٥٤٠ 57

3182 (m) 3199.7 (S) ν N-H (NH4)+

3400-3180 ١٧٢ ٣٠٣٠- ٣٣٠٠ ١٧٤

2400 (m) N.B δ O-H (H2O) ٣٠٢٠-2270 ١٧٥

1624 (S) 1625.9 (S) δ O-H (H2O) ١٦٢٠ ٥٧ ١٧٢

1521 1521 δ O-H (H2O)

1419.5 (Vs) 1421.5 (Vs) δ N-H (NH4)+

1430-1300 ١٧٢ Near 1429 ١٧٤ Near 1450 ١٨٩

1082 (m) 1087 (m) ν as P- O ١٧٢ ١٠٧٢

976 974 ν V= O ١٧٢ ٩٨٠

881(S) 885(S) ν V- OH ٨٦٠

١٧٢ ν P- O ٩٠٠

800 (Vs) 806 (Vs) δ P-OH

670.0 (w) N.B δ V- OH ١٧٢ ٦٧٢

599 (m) 601.7 (m) δ s O-P-O ١٧٢ ٥٦٠

449 (S) 453 δ V-O ١٧٢ ٤٢٠

* = Data not reported.

Chapter T

hree – Part O

ne - Results

Figure 3.12: FTIR Spectrum of VOHPO4.0.5H2O☼ sample No. 2).

121

Chapter T

hree – Part O

ne - Results

Figure 3.13: FTIR Spectrum of VOHPO4.0.5H2O sample (No.2 – un).

122

Chapter T

hree- Part O

ne - Results

Figure 3.14: FT IR Spectrum of VOHPO4.0.5 H2O/ TiO2 sample (No.2 supp).

123


124

Table3.11: The FTIR Data of VOHPO4.0.5H2O

Sample No. Attribution

Reported Data

Reported reference Sample

No.2 2-un 2-supp

3747 (W) 3747 (W ) 3747 (W ) υ O-H (alcohol) 3700 29,172,174 3590 (W ) N.B 3550 (W ) υ O-H 3590 62

3371 (Vs ) 3537 (M ) N.B υ O-H 3375 57 3370 62

N.B 3174 ( M) 3201 (M) υ N-H 3400-3180 173 3300-3030 174

N.B 3050 (Sh) 3050 (Sh) υ P- OH 3050 62 2345 (Vw) 2345 (m) 2345 (Sh) C- O (CO2) 2345 62 2320 (Vw ) 2320 (Sh) 2320 (Vw) δ + ω O-H (H2O) 2320 62 2243 (Sh ) N.B 2243 (Sh ) 2X υ PO3 2240 62

1643 (M ) 1641 (S ) 1641 (M) δ O-H (H2O) 1645 62 1641 57

1640-1591 172

N.B 1620 (Sh ) 1620 (Vw ) O-H (H2O) 1640 62

1640-1591 172 1556 (Vw ) 1556 (Vw ) 1556 (Vw) O-H (alcohol) 1554 57,93

N.B 1409 (S) 1409 (S ) υ N-H 1430-1395 172 Near 1429 174 Near 1450 189

N.B 1340 (M) N.B δ N-H 1310 - 1430

172

1199 (Vs ) N.B N.B υ PO3 1194 62 N.B 1150 (Sh) 1150 (Sh ) δ ip P- OH 1132 62

1107 ( M) 1078 (Vs ) 1083 (Vs ) υ PO3 1103 62 1045 (S) 1037 (S ) 1075 (W ) υ PO3 1050 62 977 (S ) 966 (M ) 975 (M) υ V=O 976 62

929 (W) 906 (M) 908 (W) υ P- OH 930 62

1040-910 174 N.B 748 (Vw) 732 (S ) δ N-H 850-750 176

680 (Vw ) 683 (M ) 690 (Vw ) ω O-H (H2O) 686 62 644 ( Vs) N.B N.B δ P-OH 641 62 548 (W ) 550 (Sh) 543 (W )

δ O-P-O

548 , 531 62 484 (M ) N.B 490 (Vw ) 483

62 418 (S) N.B N.B 416

* On the basis also of the IR spectrum of VODPO4.0.5D2O (57)

Chapter T

hree – Part O

ne – Results

Figure 3.15: FTIR Spectra of VO (H2PO4)2 sample (No. 3-un).

125

Chapter T

hree – Part O

ne – Results

Figure 3.16: FTIR Spectrum of VO (H2PO4)2 /TiO 2 sample (No. 3- supp).

126


127

Table3.12: The FT-IR Data of VO (H2PO4)2

Sample No. Attribution

Reported Data

Reference 3-un 3-supp

N.B 3747(Vw) υ O-H 3700 57,91,173

3550 (W ) 3550 (m) υ O-H (H2O) N.B 172

3400 (W) 3400 (Sh) υ O-H (H2O) 3400 172 υ P-OH 3420 80

3213 (M) 3223 (M) υ N-H (NH4) +

3400-3180 172 3300-3030 174

N.B N.B

υ O-H 3020-2270 172 N.B 2860 (Sh)

2293 (Vw) 2289 (W)

1622 (M) 1632 (M) δ O-H 1620 172

1418 (Vs) 1416 (Vs) υ N-H (NH4) +

1430-1395 172 Near 1429 174 Near 1450 189

1340 (W) N.B δ N-H

1300 172

δ ip P- OH 80

N.B N.B υ O-P-O 1175 172 1185 80

N.B N.B υ O-P-O 1155 172

1082 (Vs) 1084 (M) υ O-P-O 1100 172 1110 80

1020 (W) 1019 (M) υ V = O 1015 172

962 (M) 964 (S) υ P- O 930 172 υ V = O 935 80

N.B N.B υ V-OH 910 172

879 (M) 883 (M) υ V-OH 890 172 υ P-OH 885 80

816 (M) 812 (M) υ P-OH 800 172

746 (S) 741 (S) δ N-H 850-750 176

684 (W) 685 W) ω O-H (H2O) 686 62

640 642 (M)

δ O- P- O

645

172

N.B N.B 610

598 (M) 594 (M) 600

N.B N.B δ O- P- O

+ δ V- O

495

461 (Vw) 450 (Vw) 470 -430


١٢٨

3.1.5- Laser Raman Spectroscopy

Laser Raman spectroscopy analysis was carried out using red and green

laser for VOHPO4.0.5H2O and VO (H2PO4)2. Raman bands were shown in

Figures 3.17-A and 3.17-B. The obtained data were tabulated in (Tables

3.16 & 3.17) respectively.

In red laser (excitation line = 632.8 nm), all samples show the main

bands correspond to V= O, P-O-P and crystalline V2O5.

Different bands were obtained in green laser (excitation line = 514 nm)

some bands were appeared and others were disappeared as shown in the

below tables.


١٢٩

Figures 3.17-A & 3.1 7-B: Raman Shift for prepared phases carried out by Red & Green Laser.

200 400 600 800 1000 1200 1400 1600 18000

500

1000

1500 200 400 600 800 1000 1200 1400 1600 18000

200

400

600

800

1000

3-supp

2-un

Green laser a (514) nm

Raman Shift cm-1

3-supp

BOrigin Demo Origin Demo Origin Demo

Origin Demo Origin Demo Origin Demo


C/S

C

/S

700

728

810

873

984

421

360

316

203

1023

569

638

922

295

256

2-un

3-supp

3-un

Red laserA(632.8nm)Origin Demo Origin Demo Origin Demo




١٣٠

Table 3.13: Raman Shift carried out by Red Laser

Band Shift cm-1 Attribution

Reported shift

cm-1 Ref. Sample No.

2-un 3-un 3- supp

203 203 203

Crystalline V2O5

203

161

166

177

178

N.B 256 256 295

N.B 295 295 304

316 316 316 316

N.B 360 360 394

N.B 421 421 404

N.B N.B N.B 480

N.B N.B N.B 512

569 N.B N.B 526

N.B N.B 638 TiO2 (anatase) 636 167

700 N.B N.B

Crystalline V2O5

703

166

180

N.B 728 728

N.B 810 810 poly vanadate layer 800 166

N.B 873 873 P- OH 870 166

922 N.B N.B ν P-O-P 920- 930 181

169

970 984 984 ν V=O 984

141

179

180


١٣١

Table 3.14: Raman Shift carried out by Green Laser

Band shift cm -1 Attribution

Reported shift

cm-1 Ref.

2-un 3-un 3- supp

N.B 203 203

Crystalline V2O5

203 161

166

177

178

N.B 360 360 394

569 569 569 526

N.B - 638 ν Ti-O (anatase) 636 167

700 728 728 Crystalline V2O5 703 166

180

N.B 810 810 Poly vanadate layer 800 166

N.B 873 873 P- OH 870 166

922 N.B N.B ν P-O-P 920 -930 181

182

N.B 984 984 ν V=O 984

141

179

180

1023 N.B 1023 Mono vanadate layer 1023

179

180

1022 166

1105 N.B N.B ν V=O 1095 149


١٣٣

3.1.6- Redox Titration's Results

Table3.15 shows the results of redox titration data of the vanadium ion

from Fluka V2O5, Technical grade V2O5 and for the catalytic phases.

The table gives the average oxidation number of vanadium and the

percentage of V (III), V (IV) and V (V) in each case.

Table 3.15: Results of Redox Titration

Sample No. AV.OX.NO.

of V V(III)% V(IV)% V(V)% Identified phase

V2O5 T.G 4.97 0 5.30 94.7 V2O5 V2O5T.G/ TiO2 4.90 0 27.39 72.60 V2O5 + TiO2(a)

1- un 4.93 0 12.50 87.50 VOPO4.2H2O (S.S)

1- supp 4.92 0 10.20 89.79 VOPO4.2H2O +TiO2(a)

(S.S)

2- un 4.39 0 87.00 13.00

VOHPO4.0.5H2O▼

+ VOHPO4.H2O ■

+ VO(H2PO3)2

■

2- supp 4.29 0 90.00 10.00 VOHPO4.H2O ▼

+ VO(H2PO4)2

■+TiO2 (a)

3- un 4.55 0 58.9 43.1 VO(H2PO4)2

▼ +

VO(H2PO3)2 ■

3- supp

4.50

1.75

63.10

35.15

VO(H2PO4)2 ▼

+ VO(H2PO3)2

■ +

VH2P3O10.2H2O +

TiO2(a) a = Anatase.

S.S = Solid Solution.

▼ = Main Phase.

■ = Trace amount.


١٣٤

There is a good correlation between average oxidation number of

vanadium and the V (IV) % as shown in Figure 3.18

60 70 80 90 100

4.0

4.1

4.2

4.3

4.4

4.5

4.6 B Linear Fit of Data1_B

AV

.Ox.N

o. V

V (IV) %

Figure 3.18: The relationship between the Av.Ox.No.V & the V (IV) %

The results of redox titration during time of the reaction of vanadyl acid

phosphate VO (H2PO4)2/TiO2 sample (No 3- supp) are given in table 3.16

and the correlation between the average oxidation number of vanadium ion

and V (IV) % are shown in Figure 3.19.

There is an increase of V (IV) % with increase of time of reaction.


١٣٥

Table 3.16: Results of Redox Titration during time of reaction of VO (H2PO4)2 sample (No. 3-supp).

Time (hour)

Av.Ox.No. V V(III) % V(IV) % V(V) %

0 4.90 0 27.39 72.60 1 4.89 5.45 7.27 87.20 2 4.82 0 27.10 72.90 3 4.86 0 38.70 61.00 4 4.85 1.01 55.00 43.99 5 4.75 2.15 63.11 34.74 6 4.60 2.75 65.00 32.25 7 4.55 2.79 65.13 32.08 8 4.55 2.79 65.13 32.08 9 4.55 2.79 65.13 32.08

0 2 4 6 8 100

10

20

30

40

50

60

70

V(I

V)

%

Time (h)

O r i g i n D e m o O r i g i n D e m o O r i g i n D e m o






Figure3.19: The relationship between the V (IV) % with time of reaction o f VO (H2PO4)2 / TiO2 sample (No.3-supp).


١٣٧

3.1.7- Surface Acidity Measurement

This measurement was carried out by adsorption of Pyridine (py) and

2, 6- Dimethylpyridine (dmpy).

The results are shown in Table3.18, an increase in total acidity due to

increase in LA (Lewis acidity) in sample (No. 3 – supp) (The value is

doubled) was shown, this indicates the effect of support on the acidity.

Table3.18: Data of acidity centers for catalytic phases

Sample No (P / V)s *

Acidity X 107 mol/m2

Lewis Acidity

(A Py-A dmpy)

Bronsted Acidity

A Dmpy

Total Acidity

A Py

2-un 2.05 4.923 0.287 5.211

2- supp 2.36 1.428 0.308 1.737

3- un 0.98 2.050 0.893 2.943

3- supp 1.10 4.420 0.423 4.844

* = From XPS measurements.

A = Adsorption.


١٤١

XPS1U N 1.s pe : SAMPLE 1 U N SU PPOR TED , H IN D , 6-8-05 U nive rs ity o f O

05 Jun 8 Al m ono 350 .0 W 0 .0 45 .0° 187 .85 eV 2 .8518e+004 m ax 4 .59 m in

Sur1 /Fu ll/1 (SG7 Shft)

100200300400500600700800900100011000

0.5

1

1.5

2

2.5

3

3.5x 10

4 XP S 1UN1.spe

B inding E nergy (eV)

c/s

-O

KLL

-O

KLL

-O

1s

-N

1s

-C

1s

-P2

s

-P2

p

-V

2p3

-S

i2s

-S

i2p

-A

l2s

-A

l2p

-Fe

LM

M

-Fe

3p

-Fe

2p3

-C

a2s

-C

a2p3

Figure 3.21: XPS Spectrum of VOPO4.2H2O (sample No.1-un).


١٤٢

XPS1SUPP1.s pe: SAMPLE 1 SUPPORTED, HIND, 6-8-05 Univers ity of O

3.0198e+004 m ax 4.59 m in

Sur1/Full/1 (SG7 Shft)

100200300400500600700800900100011000

0.5

1

1.5

2

2.5

3

3.5x 10

4 XPS1SUPP1.spe

Binding Energy (eV)

c/s

-O

KLL

-O

KLL

-O

KLL

-O

1s

-N

1s -C

1s

-P2

s

-S

i2s -

P2p

-S

i2p

-N

a K

LL

-N

a1s

-C

a2s

-C

a2p3

-V

2p3

-A

l2s

-A

l2p

-Fe

LM

M

-Fe

3p

-Fe

2p3

Figure 3.22: XPS Spectrum of VOPO4.2H2O / TiO2 sample (No.1-supp).


143

IR AQ1 H 1 .s p e : IR AQ # 1 , VO H P O 4 .0 5 H 2 O , H IN D 1 0 /1 5 /2 0 0 5 U n ive rs ity o f O

1 .6 1 0 8 e +0 0 4 m a x 4 .5 9 m in

S u r1 /Fu ll /1 (S G 7 S h ft)

100200300400500600700800900100011000

2000

4000

6000

8000

10000

12000

14000

16000

18000IRA Q 1H1.s pe

B ind ing E nergy (eV )

c/s

-O

KLL

-O

KLL

-O

1s

-N

1s

-C

1s

-P2

s

-P2

p

-V

2p3

-N

a K

LL

-N

a1s

Figure3.23: The XPS Spectrum of VOHPO4.0.5H2O Sample (No. 2- un).


144

XPS2SUPP1.s pe: SAMPLE 2 SUPPORTED, H IND, 6-8-05 Univers ity of O

05 Jun 8 Al m ono 350.0 W 0.0 45.0° 187.85 eV 3.3885e+004 m ax 4.59 m in


100200300400500600700800900100011000

0.5

1

1.5

2

2.5

3

3.5

4x 10

4 XPS2SUPP1.spe

Binding Energy (eV)

c/s

-O

KLL

-O

KLL

-O

1s

-N

1s

-C

1s

-P2

s

-P2

p

-N

a K

LL

-N

a1s

-C

a2s

-C

a2p3

-V

2p3

-Fe

LM

M

-Fe

3p

-Fe

2p3

-A

l2s

-A

l2p

-S

i2s

-S

i2p

Figure3.24: XPS spectrum of VOHPO4.H2O / TiO2 sample (No.2-supp).


145

IR AQ2H 1.s pe: IR AQ #2, VOH PO4.05H 2O, H IN D 10/15 /2005 U n ivers ity o f O

04 Oct 15 Al m ono 350.0 W 0 .0 45 .0° 93 .90 eV 1.9090e+004 m ax 4.59 m in

Sur1 /Fu ll/1 (SG7 Shft)

100200300400500600700800900100011000

0.5

1

1.5

2

2.5x 10

4 IRAQ2H1.spe

B inding Energy (eV)

c/s

-O

KLL

-O

KLL

-O

KLL

-O

1s

-N

1s

-C

1s

-P2

s

-P2

p

-V

2p3

-N

a K

LL

-N

a1s

Figure3. 25: The XPS Spectrum of VO (H2PO4)2 Sample (No. 3 un).


146

IRAQ3H1.spe: IRAQ #3, VOHPO4.05H2O/TIO2, HIND 10/15/2005 University of O

1.6582e+004 max 4.59 min


100200300400500600700800900100011000

2000

4000

6000

8000

10000

12000

14000

16000

18000IRAQ3H1.spe

Binding Energy (eV)

c/s

-O

KLL

-O

KLL

-O

1s

-Ti

2p3

-N

1s

-C

1s

-P2

s

-P2

p

-N

a K

LL

-N

a1s

Figure3.26: The XPS Spectrum of VO (H2PO4)2 Sample (No. 3- supp).


147

Table 3.20: Atomic concentration of the elements on the surface of the catalytic

phases.

Atomic concentration

Sample

No. C 1s P2p V2p N1s Ti2p Na1s Al2p Ca 2p Si 2p Fe2p

1-un 16.51 8.99 12.40 N.B - 0.00 4.00 0.00 0.06 0.00

1-supp 14.12 9.06 11.78 N.B 0.00 0.00 4.57 0.09 0.46 0.00

2-un 16.72 13.29 6.47 3.95 - 0.15 N.B N.B N.B N.B

2-supp 18.74 13.32 5.64 N.B 0.00 0.13 2.20 0.04 0.06 0.05

3-un 21.38 9.79 10.03 4.21 - 0.005 N.B N.B N.B N.B

3-supp 23.39 9.30 8.43 0.46 0.83 0.00 N.B N.B N.B N.B


148

Table 3.21: The atomic ratio of the elements on the surface of the catalytic phases.

Sample No.

Identified phases in bulk* P/V(s) O/V(s) O/P(s) Ti/V(s) C/V(s) Na/V(s) Al/V (s) Si/V (s) Ca/V (s) Fe/V (s) N/V (s)

1-un VOPO4.2H2O 0.725 4.680 6.456 - 1..334 0.00 0.322 0.004 0.00 0.00 N.B

1-supp VOPO4.2H2O

+ TiO2 (a)

0.769 5.086 6.613 0.00 1.198 0.00 0.387 0.039 0.007 0.00 N.B

2-un

VOHPO4.0.5H2O▼

+ VOHPO4.H2O ■

+ VO(H2PO3)2

■

2.05 9.18 4.47 - 2.584 0.15 N.B N.B N.B N.B 0.610

2-supp

VOHPO4.H2O ▼

+ VO(H2PO4)2

■ +

TiO2 (a)

2.36 10.602 4.489 0.00 3.322 0.023 0.390 0.010 0.007 0.008 N.B

3-un VO (H2PO4)2

▼ +

VO(H2PO3)2 ■

0.98 5.44 5.57 - 2.131 0.005 N.B N.B N.B N.B 0.045

3-supp

VO(H2PO4)2 ▼

+ VO(H2PO3)2

■ +

VH2P3O10.2H2O ■ +

TiO2 (a)

1.10 3.20 5.80 0.09 2.774 0.00 N.B N.B N.B N.B 0.054


١٤٩

3.1.10- Scanning Electron Microscope (SEM) Analysis

Scanning electron images clearly show that the morphology of

VOPO4.2H2O with designed (P/V = 1.1) (Fig. 3.27 A and B ) is dominated

by small polygene-like particles with few small particles, while the

supported phase (Figure 3.28 A-D) is compose of larger particles.

The morphology of VOHPO4.0.5H2O (Figure 3.29 A-D) consist of

small lamellar morphology .It is obviously that using the organic medium

as reducing agent leads to obtain fine particles.

Scanning Electron Micrographs, shown in (Figure 3.30 and 3.31) for

VO(H2PO4)2 confirm the similarity in crystalline size between the

unsupported ( Fig. 3.30) and supported ( Fig. 3.31) phases, both of them

composed of well – formed plate- like crystals.

Figure 3.32 is representing the comparison of unsupported with

supported phases

Chapter Three - Part One - Results

١٥٠

(A) X 1000 (B) X 2500

Figure3.27: SEM Microstructure of VOPO4.2H2O (sample No. 1- un).

(A) X 740 (B) X 2219

(C) X 2969 (D) X 3699

Figure3.28: SEM Microstructure of VOPO4.2H2O / TiO2 (sample No. 1- supp).


١٥١

(A) X 400 (B) X 1000

(C) X 5000 (D) X 10000

Figure 3.29: SEM Microstructure of VOHPO4.0.5H2O (sample (No. 2- un).


152

(A) X 740 (B) X 1400

(C) X 2219 (D) X 2969

(E) 2959X (F) X 3693

Figure 3.30: SEM; Microstructure of VO (H2PO4)2 (sample (No. 3- un).


153

(A) X 740 (B) X 2219

(C) X 2953 (D) X 3693

Figure 3.31: SEM; Microstructure of VO (H2PO4)2/TiO2 (sample No. 3- supp).


154

(A) X 740 (un) (B) X 740 (supp)

(A) X 2962 (un) (B) X 2962 (supp)

(A) X 3693 (un) (B) X 3693 (supp)

Figure 3.32: A comparison between the morphology of unsupported & supported VO (H2PO4)2 (sample No.3-un and 3-supp).


١٥٥

3.1.11- Energy Dispersive X- Ray (EDX) Analysis

Besides the Scanning Electron Microscopy (SEM) for the information

of morphology, the energy dispersive X-Ray analysis (EDX) for

qualitative and quantities elemental analysis were applied in this study,

using the same instrument, in analyzing the surface layer up to 1 µm.

The qualitative information for each sample is given in the tables which

are appending to the figures.

The surface of VOHPO4.0.5H2O sample (No.2- un) is shown in

(Figures 3.33 and 3.34); this sample has a plate- like morphology. All

selected spots of this specimen indicate the presence of (O, V, and P) and

indicate the shell (K, L, M….etc) of these atoms. The weight ratio (P/V)

on the surface varied between (1.032 and 1.708) while atomic ratio for the

same components are ranged between (1.700- 2.811) .knowing that the

absolute designed atomic ratio for all prepared phases was (1.8).

The supported phase sample (No.2- supp) has also a plate –like

morphology with small rhombus of Titania (anatase) (figure 3.35). The

weight ratio of (P/V) is between (1.632 – 1.887) .The minimum atomic

ratio is (2.684) while the maximum value is (3.108).

For VO (H2PO4)2 sample (No.3- un), it has blocky and plate-like

morphology with regular face (Figures 3.36-3.38).

Finally VO (H2PO4)2/ TiO2, sample (No.3- supp), has flat plate with

different shape dispersed of Titania particles as appeared in (figures

3.39- 3.41).

The elemental analyses on the surface of above samples are listed in

(Table 3.22).


156

Element Wt % At %

O K 47.15 68.14

P K 26.91 20.09

V K 25.94 11.77

Total 100 100

Figure 3.33: The EDX results of VOHPO4.0.5H2O (sample No.2- un).

Element Wt % At %

O K 58.31 76.86

P K 22.05 15.01

V K 19.64 8.13

Total 100 100

Element Wt % At %

O K 64.64 81.18

P K 19.14 12.42

V K 16.22 6.4

Total 100 100

O K 52.92 72.56

P K 25.79 18.27

V K 21.29 9.17

Total 100 100


157

Figure 3.34: The EDX results of VOHPO4.0.5H2O (sample No. 2- un).

Element Wt % At %

O K 76.99 88.37

P K 14.38 8.52

V K 8.63 3.11

Total 100 100

Element Wt % At %

O K 64.02 80.16

P K 22.48 14.54

V K 13.5 5.31

Total 100 100

Element Wt % At %

O K 71.24 84.87

P K 18.14 11.16

V K 10.62 3.97

Total 100 100


158

Figure 3.35: The EDX results of VOHPO4.H2O / TiO2 (sample No.2- supp).

Element Wt % At %

O K 69.65 83.92

P K 18.82 11.71

V K 11.53 4.36

Total 100 100

Element Wt % At %

O K 68.92 83.54

P K 18.78 11.76

Ti K 0.79 0.32

V K 11.5 4.38

Total 100 100

Element Wt % At %

O K 68.92 83.54

P K 18.78 11.76

Ti K 0.79 0.32

V K 11.5 4.38

Total 100 100

Element Wt % At % O K 75.2 87.32

P K 15.33 9.2

Ti K 1.34 0.52

V K 8.12 2.96

Total 100 100


159

Figure 3.36: The EDX result of VO (H2PO4)2 (sample No. 3- un).

Element Wt % At %

O K 69.69 84.84

P K 14.51 9.12

V K 15.8 6.04

Total 100 100

Element Wt % At % O K 65.57 81.12 P K 21.95 14.03 V K 12.48 4.85 Total 100 100

Element Wt % At %

O K 65.37 81.18

P K 21.17 13.58

V K 13.46 5.25

Total 100 100


160

Figure 3.37: The EDX result of VO (H2PO4)2 (sample No.3 -un).

Element Wt % At %

O K 37.18 58.47

P K 32.94 26.76

V K 29.88 14.76

Total 100 100


161

Figure 3.38: The EDX result of VO (H2PO4)2 (sample No. 3- un).

Element Wt % At %

O K 65.66 81.17

P K 21.93 14.01

V K 12.41 4.82

Total 100 100

Element Wt % At %

O K 73.62 87.08

P K 13.02 7.96

V K 13.36 4.96

Total 100 100


163

Figure 3.40: The EDX results of VO (H2PO4)2 / TiO2 (sample No.3- supp).

Element Wt % At %

O K 56.42 75.92

P K 20.53 14.27

Ti K 2.66 1.2

V K 20.38 8.61

Total 100 100

Element Wt % At %

O K 59.87 77.87

P K 21.61 14.52

Ti K 1.84 0.8

V K 16.68 6.81

Total 100 100


164

Figure 3.41: The EDX results of VO (H2PO4)2 / TiO2 (sample No.3- supp).

Element Wt % At %

O K 74.24 87.65

P K 11.45 6.98

Ti K 2.62 1.03

V K 11.7 4.34

Total 100 100

Chapter Three – Part Two - Discussion

١٦٧

3.2.1- Effects of Preparation method

Unlike route (A), the essential steps of the preparation of the precursor

vanadyl hydrogen phosphate hemihydrate (VOHPO4.0.5H2O) are listed

below for rout (B):

1- Reduction of V2O5 to (VO 2+) species under reflux in an organic

medium using (benzyl alcohol and isobutanol as a reducing agent.

2- Reaction with o- H3PO4 to form the VOHPO4 .0.5H2O precursor

phase.

3- Separation of this phase by vacuum filtration.

4- Drying overnight at 125°C.

While, the preparation of route (A) depends on the reaction with

Phosphoric acid to produce VOPO4.2H2O before the reduction step with

the mixture of (benzyl alcohol and isobutanol); In some instances already

prepared VOPO4.2H2O (sample No.1un and 1-supp) or a suspension of this

phase was reacted in the manner described to yield and hold the (P/V) ratio

constant at near 1.1

The selection of these routs is related to our previous work (1), since the

product of route (A) (new innovation) exhibited good catalytic activity and

the product of route (B) are recognized as official catalyst for n-butane

oxidation in the literature (21, 57 - 62) . The addition of support material

(TiO2) anatase to the materials of both routes and determine its effects on

their products, is the core of this study

Several effects were observed in the preparation steps of catalytic

precursor phases it may be summarized by these notes:

1 -In rout (A) Scheme 2.1, depending on the observed change in product

color, there is an increase of five hours of required reaction time for

VOPO4.2H2O / TiO2 preparation (16 h for unsupported sample and 21h for

supported one), and there is an increase of 13 hours of required reaction


١٦٨

time necessary to prepare VO (H2PO4)2/TiO2 (21 hours for unsupported

VO (H2PO4)2 and 34 hours for supported phase).

In these two cases (supported and unsupported) the final main phases

are the same (vanadyl acid phosphate), but the only difference is in the

accompanying trace phase which is VO (H2PO3)2 .There is also an

additional phase which is VOH2P3O10.2H2O (where V Is in (3+) oxidation

state), exists in the supported phases .These anomaly phenomena can be

attributed to several factors:

A- The dispersion of variant solid quantity in fixed volume quantity of

liquid.

B- The intercalation of TiO2 within the structure of VOPO4.2H2O i.e.

TiO2 intercalate it self within the lamellar layers of VOPO4.2H2O,

prevents the process of organic material intercalation

C- In general the formation of VO (H2PO4)2 was due to conversion of a

mixture of alcoholic reducing agent (isobutanol and benzyl alcohol) to

the corresponding aldehyde and ketone. According to Bartley (81) who

found that the presence of such mixture lead to the formation of

benzoic acid and benzaldehyde which also behave as reducing agent

in place of alcohol, would result in different oxidation products. They

proposed that the involvement of an aldehyde or ketone as a reducing

agent is via the enol tautomer. For most aldehydes and ketones only a

small proportion of the enol tautomer is present. But in the presence of

a Bronsted acid, the equilibrium is more favorable and more enol is

formed. As demonstrated that the ketone and aldehyde operates as

reducing agent in the enol form, VOPO4.2H2O is reacted with

aldehyde and such reaction led to obtain vanadyl acid phosphate

VO (H2PO4)2.


١٦٩

The preparation method in route A (this work) is a new method to prepare

vanadyl acid phosphate VO (H2PO4)2, which was previously prepared by

others in different methods (80- 82).

2- In route (B) Scheme 2.1, depending on the observed change in

product color, the reaction produced hydrogen phosphate hemihydrate

(VOHPO4.0.5H2O) in unsupported phase, and produced vanadyl hydrogen

phosphate mono hydrate (VOHPO4.H2O) in supported phase .The effect of

support is directed toward the addition of one single water molecule in the

final product instead of hemihydrate normally expected in this case.

When supporting on TiO2 ,the effect of formation of VOHPO4.H2O

instead of VOHPO4.0.5H2O is related to the competition between H2O

molecules and organic reducing agents, during the reaction, to be

intercalated in the layers formed (in suspended solid already formed,

containing TiO2).Although the X-ray diffraction pattern of these two

phases are identical (Fig.3.1 and Table 3.3) (with slight shift when

containing TiO2) but in thermogravimetric measurement there is a

significant loss of weight ) is observed of unsupported phase (Table 3.7) .

The extra loss of weight indicated that the presence of organic species is

accumulated in such extend in VOHPO4.0.5H2O phase, that they are being

easily expulsed ,this explanation is fortified by the data presented in (Table

3.6) for these two supported and unsupported phases in the range of (50-

230°) C .the loss of weight in the unsupported phase is related to extra

amount of organic molecules intercalated between layers ,there is no such

study of thermogravimetric analysis in the literature concerning the

presence of TiO2 in prepared phases.

The huge departure of organic molecules from unsupported phase is

observed in our previous work (1) even under investigation of this phase in

situ by Scanning Electron Microscope (SEM).The change of the structure


١٧٠

is obvious during the inspection due to the evolved heat from the contact

of electronic beam and the sample.

Table 3.5, in rout (B) highlighted the formation of vanadyl acid

phosphate VO (H2PO4)2 in trace quantity accompanying the formation of

VOHPO4.H2O / TiO2.

Table 3.5 reveals that VOHPO4.0.5H2O which is formed without

support contains traces of other phases of (VPO) system. The explanation

of the presence of such phases is given by Bartley (81), as stated before, it is

due to the reaction of the products (aldehyde and ketone) involved from

alcohol oxidation in the precense of small amount of o- H3PO4.The organic

phase (route B) permit the reaction between (benzyl alcohol + isobutanol)

to obtain the reduced phase V2O4 once the H3PO4 is added, the majority of

V2O4 is converted to VOHPO4.0.5H2O in huge quantity. Some traces of

resultant product of alcohol oxidation which present in reaction

environment will convert small quantity of V2O4 to other phases of (VPO)

system. The presence of TiO2 is selective to produce vanadyl acid

phosphate VO (H2PO4)2. In contrary in (route A) the phosphoric acid is

added just in the beginning of the reaction. The small quantity of

phosphoric acid arises in the suspension media direct the reaction quickly

(when adding alcohol as reducing agent) toward the formation of vanadyl

acid phosphate VO (H2PO4)2.

The (P/V) ratio in VOHPO4.0.5H2O is higher than designed ratio

(1.8 atomic ratio) because of presence more than one phase, impurities,

and support which have high electronegativity which led to increase

(P/V) ratio (148 ) .

3.2.2- Promoters, impurities, and Solid Solution effects

The presence of support (TiO2) in 8 % wt percent in the crystal phase

network structure led to cancel the combined effects of promoters (or


١٧١

impurities) which already exist in the unsupported phases; like

intercalation effect. Unlike the intercalation, the solid solution effect is still

present. The selection of this support and its percentage is suggested by

published literature (183-186)

The XRD and Raman spectra of the bulk (VPO) precursors which

demonstrate the presence of only vanadyl hydrogen phosphate

hemihydrate VOHPO4.0.5H2O, vanadyl hydrogen phosphate monohydrate

VOHPO4.H2O and vanadyl acid phosphate VO (H2PO4)2 .These bulk

characterization techniques did not detect promoter phases in the bulk

(VPO) since these techniques are blind for traces of impurities (116,187).

The intercalation of the promoters into the layered structure of the

VOHPO4.0.5H2O precursor and the cleavage of its (010) plane should

result in an increase of the surface area and solid solution

formation (1, 172).

Furthermore, the promoter elements in the most active and selective

(Na promoted catalyst) were concentrated in the surface region despite the

low promoter content, which probably indicates higher distribution of the

Na promoter in the (VPO) surface matrix. Therefore, it appears that the

promoter elements may partially form a solid solution in the bulk in the

absence of the support.

However, this process is primarily limited to the surface region and

affects the catalytic properties of the bulk (VPO) catalysts (172).

Thus the obtained results from this work determine the pathway for the

increase of the activity and selectivity of the catalyst for n-butane

oxidation. The increase of activity could be obtained by the introduction in

base (VPO) composition of the element donor, which increases the

effective negative charge on oxygen of catalyst surface, Ti

(electronegativity = 1.6) belong to this type of additives. However, the

content of other additives in the catalysts has been not significant. The


١٧٢

introduction of the additives which increase the surface acidity favors to an

increases of the selectivity to products of partial n-butane oxidation, in this

case the influence of additives origin on acidity type did not determine

(and as a result it was very difficult to predict ) the group effective

promoters, which increase the selectivity to definite anhydride.

3.2.3 -Thermal Behaviors (DTA and TGA)

Generally, in all cases; the thermal behaviors included four

distinguished combined elements which correspond to (a) the loss of

adsorbed organic molecules which attributed to the residue of benzyl

alcohol and isobutanol, (b) decomposition of intercalated species, (c) the

loss of water molecules and (d) finally the stage of topotactic

transformation of the precursor to (VO)2P2O7 in route (B) and to

VO (PO3)2 in route (A).

In vanadyl hydrogen phosphate hemihydrate VOHPO4.0.5H2O (sample

(No.2-un), four endothermic phenomena were occurred (Figure 3.2). The

first one is a little endothermic, and it was occurred between (205- 230 °C)

with a loss of weight equal to 8.16 %. The second endothermic

phenomenon happened between (240- 310) °C with a loss of weight 1.73

%.Third endothermic peak is extended from (310 to 510 °C) with loss of

weight 4.90 %. The last one is laid between (500- 580) °C without any

change in the weight. These endothermic peaks followed by an exothermic

peak between (580- 630) °C with gain of weight.

The total loss of weight was (14.79 %), this value is greater than

theoretical loss (10.52%) to obtain vanadyl pyrophosphate, and this

difference could be attributed to intercalated species and to the presence of

inorganic impurities in technical grade V2O5.


١٧٣

The vanadyl hydrogen phosphate monohydrate VOHPO4.H2O sample

(No. 2- supp) has different behavior, it has five stages the first one is

neither endothermic, nor exothermic which occurred between (180 – 220)

°C with loss of weight (5.7 %), followed by three endothermic phenomena:

the first is laid between (200- 320) °C occupied with loss in weight

equaled to 2 %. The second endothermic step happened in temperatures

rang between (320 – 510) °C with loss of weight 6.2 %. The last one was

occurred in rang between 510 and 600 °C with no change in final weight.

The total loss was 13.9%.

The gain in weight in the last exothermic peak between (600- 660) °C

was seen .It is obviously that the decomposition of supported sample was

extending to a higher range of temperatures which reflects the grater

resistance of such sample to the heat treatment.

Vanadyl acid phosphate VO (H2PO4)2 sample (No.3- un) shows another

different behavior (Figure 3.4), It started with two little endothermic peak

extended between (80 to 110) °C for the first and between (180 to 245) °C

for the second with loss 6 %; then another loss (0.7 %) was happened in

the range of temperature (277- 350) °C by neither endo nor exothermic

changes. The forth loss (5.3 %) was occurred in (370- 490) °C range, by an

endothermic phenomena. An additional endothermic peak was laid

between (490 – 560) °C without any change in the weight. Finally a little

exothermic peak was laid between (560 – 620) °C with again in weight

(0.99 %).

The theoretical loss of this phase is (13.7 %) (80), which equaled to loss

of two water molecule. The difference in the loss weight between our

sample and the literature could be explained by the presence of trace

amount of other species.

The supported sample showed a different behavior; it dehydrate in the

first endo stage between (110 -180 0 °C with loss of weight 8.71%


١٧٤

followed by (neither endo or exothermic) at (255- 340) °C, the loss in this

stage was (0.5 %). The third stage was endothermic stage (350 – 440 °C)

occupied with loss of weight (4.65 %). in the range (470 – 560 °C) the last

endothermic peak was seen it without any change in the weight and finally

the last stage; which was a little exothermic between ( 560 and 620)°C

with gain in weight (0.99 %).The total loss in weight for this sample is

(13.8 %), this result is in agreement with the literature for our product (81) .

The presence of TiO2 in these sample and its behavior affect the presence

of other species, this confirm our suggestion above related to the

competition of intercalated species

The gain in weight in all samples at higher temperature with an

exothermic peak is presumably due to reoxidation of the samples that is

partially reduced on pyrolesis of the adsorbed or intercalated organics (57).

From above presentation ; two ranges of temperature have to be

considered .The first ,from room temperature up to 340 °C corresponds to

the departure of adsorbed species ,and the second stage ,from (340 to

510)°C, to the transformation of the solid .while the first range is

approximately always the same ,with a mass loss occurring in the two

steps at around (50 and 237) °C (Table 3.6) ,the second range differs from

one precursor to another .The loss which may be attributed to the departure

of physisorbed (mixture of isobutanol and benzyl alcohol ) and water are

reported in (Table 3.6) . These results showed that the dopant have a great

effect on the physisorption of water (172). This agrees with our results when

considering the effect of TiO2 on the selective adsorption of both water and

organic molecule to produce vanadyl hydrogen phosphate monohydrate

VOHPO4.H2O.

The topotactic mechanism of dehydration of vanadyl hydrogen

phosphate monohydrate VOHPO4.H2O, vanadyl hydrogen phosphate

hemihydrate VOHPO4.0.5H2O and vanadyl acid phosphate VO (H2PO4)2


١٧٥

to (VO)2P2O7 and VO (PO3)2 has been described (57,81,174). The

VOHPO4.0.5H2O retains its water of hydration up to 350 °C. The

departure of these molecules results in vacancy formation at the apex of

vanadium oxygen octahedral. In this stage an electronic rearrangement

proceed with respect to the promoters or TiO2 which corresponds to the

formation of VOHPO4 layers .These layers will join expelling the

constitutional water molecule to form (P2O7) group form (2 HPO4) groups.

This second water departure was observed for VOHPO4.0.5H2O at 430 °C.

Doping with a low percentage 8 % of Titania delays the temperature of

dehydration. Doping with Titania at low percentage also promotes

vanadium oxidation this can be explained by better dispersion of Titania as

evidenced by energy depressive x-ray EDX result (Table 3.22).

The promoting effect of Titania on the oxidation state of vanadium

which increases V (V)/ V (IV) balance on the n-butane / air atmosphere is

highly important to explain the catalytic activity of this dopant in n-butane

oxidation to (MA). This is made possible by the formation of very limited

solid solution of type ((VO) 1-x Ti x). 2 P2O7) (186). This conform the

specificity of this support added in such amount as concluded

previously(183-187).

3.2.4 - FTIR and Raman structural properties

In FTIR spectra, there are large differences between laboratory grade

V2O5 (Fluka) and technical grade V2O5 (Fig. 3.6, Fig. 3.7 and table 3.8).

The FTIR spectra of the derived phases from the technical grade are also

different (Figures 3.10-3.16) these differences are due to:

a- The impurities of original material (Table 3.1).

b- The additional material used in the preparation methods.


١٧٦

c- The formation of other phases as previously shown in XRD (section

3.1.2).

Due to the complexity of these above three effects and due to the

absence of the request of single pure phase capable to affect the activity of

the catalyst. These effects are governed by the concentration and quality of

these impurities added, although they are very little. Each element of these

impurities repeats the same effects with the main phases exist.

In addition the presence of TiO2 (8%) which is remarkable percentage

that could cause additional relational effects, now and then; they are

difficult to be clarified by using spectrum of FTIR .The matter is still open

to use several other techniques which can be applied or in accordance with

the main idea exposed.

The effect of these impurities especially (Na) was studied in our

previous work (1), the presence of this element leads to formation a solid

solution and this phenomenon leads, in X-ray, to shift the (2 θ) in a

comparison with the standard samples. The formation of solid solution still

exists in our present work; it is obvious from the comparison of the d-

spacing value which is reported in the literature (table 3.2, 3.3 and 3.4). It

is worthy to say that the effect of impurities in XRD spectrum is differing

from the effect on FTIR.

It is obviously that the technical grade (V2O5) contains amount of

ammonium meta vanadate (NH4VO3) which was indicated according to the

literatures (172,174) by the presence of band at 3197 cm-1 and band

between 1300 and 1430 cm-1 which refer to the stretching frequency of

N-H bond of (NH4) +.

It is clear that technical grade have several different bands which are

not similar to those which shown in Fluka V2O5, these bands are refer to

impurities which already exist in technical grade V2O5 (section 3.1.1).and

it also contained shifted bands in comparison with Fluka V2O5.


١٧٧

The obtained data of FTIR of VOPO4.2H2O are in good agreement with

Bordes et al. (29,172).

The spectra of the precursor vanadyl hydrogen phosphate hemihydrate

VOHPO4 0.5H2O prepared in route (B) (Figures 3.13 and 3.14 and table

3.11) characterized by bands near (3174) cm-1 and (3201) cm-1 which

attributed to (υ N-H bond) however the standard sample (prepared from

pure Fluka V2O5. can't exhibit these bands (Figure 3.12) .This indicates the

presence of ammonium metavanadate since the frequency of N-H bond

belongs to ammonium group which is certainly intercalated between

lamellar layers. XPS results also indicates the presence of Nitrogen atom

(fig 3.21-3.24) which is related to of nitrogenous trace compound

This phase exhibited a band centered at (1645) cm -1, such frequency is

typical of water coordination ((in plane) deformation) in agreement with

the crystal structure.

Two bands are appeared at (1409) cm-1 (strong) and (1340) cm-1 in

the measured unsupported sample (No.2-un) the last one does not appear in

the supported sample ,the modification of the structure could lead to the

vanishing of one band of N-H bond .

Some bands (1199, 1107, 1045 and 929) cm-1 appeared in the spectrum

of standard sample and also appear in (1078, 1037and 906 cm-1) for

unsupported sample (No.2-un.) and near (1083, 1075 and 908 cm-1) for

supported sample No.2-supp).All these bands belong to the υ P-O bond

which is being extended between (1200- 900) cm-1 (57).

Meanwhile the υ V=O appeared at frequencies (975, 966, 972) cm-1 in

spectrum of sample (No.2) [reference sample of vanadyl hydrogen

phosphate hemihydrate prepared from V2O5 Fluka], sample No.2-un and

sample No.2-supp) respectively, This band normally appear at 1016 cm-1

in pure Fluka V2O5 and at 966 cm-1 in technical grade V2O5.


١٧٨

The spectrum of this phase is also remarked by presence of a week band

centered at (748) cm-1 in unsupported sample (No.2-un), and strong band in

the region of (732) cm-1 in supported sample (No.2-supp) (which exist

originally in the technical grade V2O5 belongs to general bond (M = O)

which refers to the presence of impurities present in technical grade V2O5

(Table 3.1).

The spectrum of vanadyl acid phosphate VO (H2PO4)2 is characterized

by a week band near (3430) cm-1 which corresponding to (υ P-OH) bond

this is in agreement with Villeneuve (80).

The band of N-H bond appears near (3213 and 1418) cm-1 for

unsupported phase and near (3222.8 and 1415.8) cm-1 in the supported

phase, the positions of υ V= O is in good agreement with Bordes et.al.

(1015) cm-1 (172), (1020.3, 1018.9) cm-1 for unsupported and supported

phase respectively

The literature mentioned several bands corresponding to (υ P-O) cm-1

near (1185, 1110 and 910) cm-1 (81,175) in our case only one band appeared

near (1082) cm-1 for unsupported and (1084) cm-1 for supported phase.

This phenomenon indicates the presence of the deformation.

There are two bands centered at (879.5, and 815.5) cm-1 and (838.3 and

812) cm-1 in unsupported and supported phases respectively, these bands

are attributed to δ P-OH and / or δ V-OH. These results are in accordance

with published literature (80,172).

In conclusion the effect of support and intercalation between phases and

the effect of ammonium metavanadate are summarized in the following

points:

1- Slightly shifted bands are resultant of support effect presence.

2- All prepared phases are deformed.

3- The effect of Ammonium ion result from original ammonium


١٧٩

metavanadate is clear in the spectrum of technical grade V2O5 and its

derived phases.

4- The impurities existent in technical grade V2O5 has certain influence

on the spectrum of the derived phases.

The dependence only on FTIR technique in explanation of the

unknown bands lead to some risk without the use of XRD, Raman

technique, and XPS in order to identify the prepared phases.

The data [table (3.13, 3.14) Fig (3.17, 3.18)] which rose from laser

Raman spectroscopy is approximated with the literature. Raman

spectroscopy is a useful technique for the identification of a wide range of

substances , solids, liquids, and gases. It is a straightforward, non-

destructive technique requiring no sample preparation. Raman

spectroscopy involves illuminating a sample with monochromatic light and

using a spectrometer to examine light scattered by the sample. Due to the

nature of this technique which is based on regrouping the similar and

different effects on the same bonds behind one single band obtained.

From our Data offered by University of Oklahoma, there are two

different types of vanadate layer, the first at (1023) cm -1 which pointed out

to monovanadate layer, this band are shown in both vanadyl hydrogen

phosphate hemihydrate phase, and vanadyl acid phosphate phase

(supported and unsupported). The second type is a polyvanadate layer at

(810) cm-1, which appeared only in VO (H2PO4)2 (supported and

unsupported).

Generally the Raman spectra of catalysts (81) have a strong broad band

with Raman shift of (935) cm -1, which is characteristic of VO (H2PO4)2,

this band is absent in our data. However, other bands for the same phase

(575, and 900) cm -1 are also missing from our spectra. Iglesia (166) suggest

that the characteristic band of hydrated (5VOx/TiO2) produces too poorly

resolved bands centered at about (800 and 900) cm-1, indicating the


١٨٠

presence of polyvanadate species. Raman intensity in the region of (800)

cm-1 may also be due to second- order scattering from TiO2. Upon

dehydration, the broad polyvanadate bands remain, but the band centered

near (900) cm-1 becomes less intense. The appearance of a band at (1022)

cm-1 indicates the formation of monovanadate species. Four bands at (149,

395, 511, and 639) cm-1 are attributable to anatase TiO2, do not change

throughout hydration- dehydration cycles. These suggestions are in

accordance with our results above.

Since the green laser is higher energy than red laser, the distribution of

sodium and oxygen in the space of V2O5 crystalline network structure in

the moment of photon applying will cause the arises of atoms space

distribution more than in red laser , the matter which can be explained

significantly in the disappearance of number of bands (cystallinaty bands)

and the appearance of other bands which are related to vanadium and

phosphors these bands nominated as previously indicated as polyvanadate

layers

The appearance of V=O bond near (984) cm-1 for VO (H2PO4)2 and

near (970) cm-1 in VOHPO4.0.5H2O in week state (in red laser) and at

(1590) cm-1 for VOHPO4.0.5H2O (in green laser) is noted in relation to the

literature which gives higher intensity for this bands. This phenomena

indicate that Raman shows regroup several effects of impurities and

collect these effects in single (may be less sharp) bands. When using green

laser and its higher energy the masked effect (can not be seen with red

laser) would be highlighted. This supports the crystalline structure and

reflected to single sharp band.

Based on a combination of the powder XRD and Raman spectroscopy

studies, it can be concluded that the precursor phases of catalyst comprise

disordered materials.


١٨١

2.3.5 - Bulk Oxidation States (Redox Titration)

Our results in section (3.1.6) are in agreement with the literature (68, 92,141)

which indicate that the best catalyst is the one which has an oxidation

number varied between (4.3- 4.6).

It is very clear that the support TiO2 has a positive effect toward the

average oxidation number of vanadium ion and the V (IV) % content i.e. it

decrease the average oxidation number (approached to +4) and increase the

percentage content of V (IV) .

It is clear from the results of redox titration of reaction evolution that

the time of reaction for VO (H2PO4)2 syntheses sample (No.3-supp) which

is (13h for reduction step) is enough for preparation reaction time

2.3.6- Effect of support on the specific surface area

No changes in the values of specific surface area (10 m2 /gm) are

observed in VOHPO4.0.5H2O (unsupported and supported) (Table 3.17A );

this is probably due to the presence of sodium ion (Na) in this phase.

According to Riva et al. (190) the presence of this ion with vanadium

pentoxide (V2O5) leads to the formation of new bond (Na- V); and the

presence of such species decreases the specific surface area. It also may be

seen according to Volta et al.(173) that whereas any increase with the

amount of dopant, at the same time, the surface areas of the solid will

decrease. The effect of support here is to equalize the effect of sodium.

For VO (H2PO4)2 phase, where (Na ion) was dissolved while

preparation (i.e. separated from the suspended solid) in water, which used

in the first stage of the preparation (Scheme 2.1), this lead to increase the

specific surface area for the supported phase (Table 3.17-A).No figures are

reported in open literatures about the supported phase surface area. The


١٨٢

result given by this work of supported vanadyl acid phosphate surface area

(12 m2 / gm) is considered as the highest surface area found in the

literature for this phase, the increase of specific surface area has (with

other combined factors ) a great influence on catalytic activity

However, the specific surface area of VOPO4.2H2O/TiO2 as separated

phase (sample No.1- supp) is high (16 m2 /gm) in comparison with the

unsupported sample (11 m2 /gm) (sample No.1-un); In this case, there is

good indication, here, for the effect of support and, the effect washing out

the Sodium ion and the effect of obscenity of adsorbed species (which thy

will be added later on to the reduced phases when using reducing agents).

These results were given by method of (Argon Temperature

Desorption) offered by the National Academy of Science in Ukraine. This

method is unlike BET, did not permit to determine the pore shape or other

characteristics of morphology. It is well known that the pore shape for

unsupported sample by using BET method is cylindrical with two opened

ends. (1)

2.3.7 Acidity Effect

According to Zazhigalove (145) the Lewis centers can effect on the

activity of the catalyst i.e. higher conversion of n-butane to product.

However, the Bronsted centers (BA) can contribute more precisely on the

selective production of Maleic Anhydride (MA).

The acidic properties of phases were defined by adsorption of pyridine

(py) and 2, 6-dimethylpyridine (dmpy) in chromatic variant. This method

permits to determine the total acidity and amount of Bronsted and Lewis

acidity centers.

In comparison between the two types of phases selected in this work

route (A) and route (B), we can remake the jump of Lewis acidity centers


١٨٣

(LA) value in the route (A) for supported phase (Table 3. 18) .This is good

indicator and presumption of the catalytic activity while the decrease in

Bronsted acidity centers (BA) will probably affect the selectivity.

The increase in activity (due to the increase in Lewis acidity (LA)

centers) is attributed to the deformation of crystalline network according to

Busca (189). These deformations play great role in the renewal of the

surface while the reaction proceeds, more surfaces is exposed and hence

more reaction could be occurred. i.e., the more surface exposed more

pyridine could be adsorbed and the more acidity exhibition.

According to Busca (189) who discussed this matter thoroughly. These

results can be confirmed by using FT-IR in such distinguished bands of

absorption.

2.3.8 - Surface Chemistry by XPS analysis

X-ray photoelectron spectroscopy (XPS), a surfaces nondestructive

measurement technique, is well suited to study a surface chemistry by

using appropriate depth profiling methods, can yield useful "chemical

depth profiles" by virtue of the strong "chemical shift" in the binding

energy of core – electron levels due to chemical bonding . XPS has been

applied to many different areas of applied and basic research.

Several important results were obtained from XPS measurement offered

by Oklahoma University in this work, maintained in section (3.1.9):

1- Identification of elements already exists in the surface of precursors

like (V, P, O, Ti, and Na). (Table 3.19)

2- Identification of several types of Oxygen species on the surface in

coordination with other element in situ (Fig 3.22 – 3.24).

3- Determination of oxidation state of vanadium ion on the surface (table

3.19). According to Watches (162), Jerry Li (180), Ruitenbeek (186), and Moser


١٨٤

and Schrader (188), this value indicates that the average oxidation state of

vanadium on the surface, for all measured samples, is (5+).

4- Determination of atomic concentration of each element on the surface

(Table 3.20). This can lead to the determination of surface P/V ratio. The

surface enrichment in phosphorus which was determined by XPS

measurement; has been also reported on (VPO) catalysts (95).This surface

excess of Phosphorus and Oxygen has been tentatively explained by the

presence of other phases like VO (H2PO3)2 and VH2P3O10.2H2O.

XPS data appeared to be inadequate because surface and bulk

contributions are averaged in the spectra .Furthermore, the atomic

sensitivity factors for vanadium and phosphorus have proven to be not

satisfactory to enable one to quantify the XPS results. However, with

experiments at different incident angles will be able to establish that

surface phosphate enrichment which has occurred in all unsupported

(VPO) (186) .

Our XPS results exhibit no trace of Vanadium (IV) on the surface in

contradictory with the results of Bartley (81) for the catalyst derived from

Vanadyl Acid Phosphate VO (H2PO4)2, who suggest that although the

surface comprises Vanadium predominately in the (V) oxidation state, he

can not completely roll out a contribution from a small surface

concentration of V (IV).

The percentage of Na ion (as well other impurities)is varied between

the measured samples due to the preparation method, it is higher in

VOHPO4.0.5H2O (sample no.2-un) in relation to VO (H2PO4)2 sample

(no.3- un) .It is obvious that Na ion is dissolved in water which used

during preparation of the later sample. The presence of Titania leads to

neglected role of Na ion on the surface (sample no.3-supp).


١٨٥

2.3.9- Morphology measurements by SEM

There are several parameters affecting on the morphology of the

prepared phases: P/V ratio, preparation method, and reducing agent.

It is worthy to note that the same phase prepared from technical grade of

V2O5 but with (P/V = 7.3), has a plate-like crystallites (1), which means that

the preparation method and the (P/V ratio) affect on the morphology of the

catalytic phases.

In conclusion there is no effect (in specified picture magnification) is

observed on the morphology for support (Titania anatase) on the

morphology of prepared phases (Figure 3.32).

Liskowski (191) has reported blocky morphology with regular sided of

long crystalline ,the ratio of the sides being 1:1: 2 regardless of the

dimension of crystal .However it is clear that there are not single crystal

phase since the surfaces are marked with many indentations. In contras ,

Vanadyl Acid Phosphate VO (H2PO4)2 prepared by Hutchings (149) using

aldehydes and ketones as reducing agents had needle shaped crystal which

at higher magnification did not reveal the presence of surface indentations.

Hence the solvent used to prepare VO (H2PO4)2 can control the

morphology of the crystallite. Our crystallites morphology which obtained

by alternative method of preparation and also the vanadyl hydrogen

phosphate hemihydrate VOHPO4.0.5H2O obtained by this work the using a

mixture of (benzyl alcohol and isobutanol) were not similar to crystallites

prepared by using aldehyde and ketone prepared by Hutchings (149) .

Scanning Electron images clearly show that the morphology of

VOPO4.2H2O (Figure 3.28) is dominated by large flat platelets with small

particles on the surface these results are in agreement with Bordes (172) and

Liskowski (191).


١٨٦

2.3.10 – Identification of elemental composition by EDX

Analysis

EDX analysis stands for energy dispersive X-ray analysis .it is a

technique used for identifying the elemental composition of the specimen

or an area of interested thereof.

The distribution of the atoms on the surface are varied between the

selected spots, this point out the presence of more than one phase on the

surface in each prepared samples.

The inspection of the selected spot leads to conclude that the atoms were

randomly distributed. This observation was concluded from the

examination of closely different spots. This technique, however, has

failed in identification of impurities (192).

The presence of Titania as support facilitates the redistribution of

vanadium on the surface to be more homogenous in sample (No.2-supp),

this opinion must be taken with precaution when Titania percentage in the

selected spot is increased as it led to decrease the vanadium ratio.

The relationship between EDX and SEM:

It is obviously from these images that there are differences in the

crystal type which selected from different spots, but it is difficult to specify

that such differences were related to different phases, this technique is

incapable to identify the phases, so the SEM technique could be helpful

when using X-ray in determining different positions of atoms in crystallites

The distribution of Titania in VO (H2PO4)2 sample No.3-supp is better than

VOHPO4.0.5H2O

One of the uses of EDX is to identify the distribution and coordination

of atoms on the surface in depth of 1 µm i.e. the distribution of atoms

related to the identified phases on the bulk.


١٨٧

This can be achieved by a hypothetic method which links the

distribution of atoms on the surface with these on the bulk by a percentage

of related atoms. The method of calculation consists of determining the

average value of P/V, V/O and P/O ratio for each sample and a adopting a

model which can exhibit the fractionate value for each atom. The results

show that the assumed phases on the surface could be something like:

V 1-x P 1-y O 1-z

This assumed method of calculation actually neglects the presence of

water molecule (OH group) and the presence of impurities in the proposed

structure and on the surface .However, for the supported phase the

distribution of titania is non- uniform and arbitrary depending on the

position of examination spot, so can be assumed as adjacent atom not

linked to the above proposed surface structure.

It is obvious that all samples have a higher (P/V) ratio. The increase in

this ratio is attributed to presence of different trace identified phases (95);

the presence of such phases is a normal phenomenon in (VPO) system (171).

2.3.11- Resumption of the catalytic activity

The unsupported phase vanadyl acid phosphate VO (H2PO4)2 had

been examined in terms of catalytic activity in our previous work in

February 2000 (1).

The addition of TiO2 (anatase) to this phase is mainly to determine the

chemical and physical effects (Structural and superficial) on this phase.


١٨٨

From above points of discussion of these effects, presumption could be

conducted and catalytic properties could be assumed.

The reactivity of Vanadyl Acid Phosphate VO (H2PO4)2 was

determined in literature by Bartley (81) in June 2000. A detailed study of

n-butane oxidation over (VPO) catalyst derived from in situ activation of

Vanadyl Acid Phosphate VO (H2PO4)2 is described thoroughly .This

study suggests by using detailed powder X-ray diffraction and Laser

Raman Spectroscopy that , the catalyst derived from vanadyl acid

phosphate VO (H2PO4)2 is highly disordered , containing some

transformed (intermediate species) .

However, our method of preparation is quite alternative method to

prepare this phase in regard to the work of Hutching. The above points of

discussion also conclude the formation of disordered material.

According to these common points between these two comparable lines

of work, we can assume certain extra activity and selectivity due to the

presence of support TiO2 (anatase) well dispersed on the surface of the

catalyst as it was indicted by EDX (2.3.10).

Conclusions &

Future Prospects

١٩٠

Conclusions

According to this study we can conclude the following:

1- The possibility of preparing of Vanadyl Acid Phosphate VO (H2PO4)2

from technical grade Vanadium pentoxide, and with designed molar

ratio (P/V= 1.1), as an alternative method never mentioned in the

literature.

2- Trace amounts of co-exist phases are accompanying the main

prepared phase whatever the origin of V2O5 is.

3- The addition of (TiO2 anatase) could change the resultant phases

prepared in route (B), since a monohydrate composition

(VOHPO4.H2O) is obtained, instead of Vanadyl Hydrogen Phosphate

Hemihydrate (VOHPO4.0.5H2O) normally prepared without the use

of support .

4- Vanadyl Acid Phosphate VO (H2PO4)2 is always the same phase

prepared with or without the addition of TiO2. No effect of support is

observed on the preparation of this phase by route (A).

5- The effect of support on the thermal behaviors is obvious on the

prepared phases by both methods (A & B).

6- The presence of wide shift in the bands positions in the FTIR. These

shifts reflect the deformation of the angles and lengths of bonds in all

derived phases from these in technical grade V2O5.

7- The presence of mono vanadate layer in all prepared phases. However

the poly vanadate layers are identified only in the prepared Vanadyl

acid phosphate. This phenomenon is very useful to support the

employment of this phase, as a catalyst in the reaction of n-butane to

Maleic Anhydride. It may indicate a presence of catalytic activity.

8- The oxidation state of vanadium ion changes toward (4+) In the

١٩١

presence of the support.

9- There is a significant increase in specific surface area when adding

support to of intermediate VOPO4.2H2O preparation.

10- A specific surface area values of VO (H2PO4)2 and VO (H2PO4)2 /

TiO2 (10, 11 m2/g) are measured in this work for the first time.

11- The results of (XPS) point out the presence of structural elements

(V, P, O) and detectable percentage of inorganic impurities on the

surface of the prepared phases. They also imply the presence of

Nitrogen (N 1s) which could be attributed to Ammonium

metavanadate (NH4VO3).

12- Presence of several oxygen species in the surface of the prepared

phases.

13- The (P/V) atomic ratio on the surface of both Vanadyl Acid

Phosphate and Vanadyl Hydrogen Phosphate Hemihydrate in the

presence of support is near (1).

14- The Lewis acidity centers value of VO (H2PO4)2 in the presence of

support is doubled; a fact that reflects the good catalytic activity.

15- All phases have a lamellar morphology. Using a mixture of (Benzyl

Alcohol and Isobutanol) led to obtain a fine crystal in Vanadyl

Hydrogen Phosphate Hemihydrate VOHPO4.0.5H2O.

16- Presence of TiO2 in selected spots goes down to the depth of 1 µm

from the surface.

17- The intercalation of organic and /or inorganic species between the

lamellar layers of the catalyst precursor, thus, resulting in low

temperature dehydration transformation to the active phase

(VO)2P2O7 .All the distortion phenomena accompanying this

transformation are due to either the reaction of lattice oxygen with

organic solvents intercalated to produce vacancies, or the

١٩٢

incorporation of inorganic impurities inside the lattice structure

modification intramolcular bonds to produce solid solution.

18- The ability of solid solution formation phenomenon while preparing

the intermediate phase and the derived supported phase in the method

of preparation.

19- Many signs in these conclusions lead to predict a higher catalytic

activity for Vanadyl Acid Phosphate VO (H2PO4)2 prepared in new

method. This work is in accordance with the reported catalytic

activity in the literature.

١٩٣

Future Prospects

1- Studying the effect of catalytic test parameters like flow rate, contact

time, reaction temperature, air/butane mixing ratio on the conversion

and selectivity of n- butane reaction using supported Vanadyl Acid

Phosphate VO (H2PO4)2 in order to perform a feasibility study.

2- Determining the catalyst life time and studying the catalyst

regeneration.

3- Studying the poisoning and sintering of the catalyst.

4- Determining of crystallographic parameters including Millar Indices of

the phases prepared in this work.

5- Drawing a map of the distribution of atoms on the surface of precursor

and the derived catalyst. And localize the exact active spots

responsible for higher activity and selectivity in n –butane reaction to

produce Maleic anhydride.

6- Studying the catalytic reactor design required to support the indicated

chemical, physical, and engineering parameters to develop economical

industrial process.

References

١٩٥

References 1. Al-TA'EE, H.A. M.Sc Thesis, University of Baghdad, Iraq, 2000.

3. Malow, M. Hydrocarbon Process, 1980, 11, 149.

2. Ponder, T. Hydrocarbon Process, 1992,3,33.

4. Felthouse. T.R.; Burnet, J.C.; Michael, S.F; Mummy, M.J; Kuo, Y.

http://jws-edck.interscience.wiley.com.:809/kirk_articles _fs.html.Jan 2001.

5. http://www.the-innovation–group.com/ChemProfiles/MaleicAnhydride,

htm Oct.2003.

6. Robinson, W.D .Ency. Chem. Tech. 1979, 14, 770.

7. John, J.M.; William, A.C. Ency. Chem. Proc. Des. 1988,29,35.

8. Hodnett, B.K. Catal. Rev. Sci. Eng. 1985, 27, 373.

9. Wohlfahrt, k.; Emig, G. Hydrocarbon Process. 1980, 6, 83.

10. Budi, F.; Neri, A.; Stefani, G. Hydrocarbon Process. 1982, 1, 159.

11. Ponder .T. Hydrocarbon Process. 1979, 11, 88.

12. Arnold, S.C.; Sucire, G.D.; Verde, V.; Neri, L. Hydrocarbon Process.1985,

9,123.

13. Varma, R.A.; Saraf, D.A. Ind. Eng. Chem. Prod. Res. Dev. 1979, 18, 7.

14. Obenaus, F. Droste, W. Ulman's Ency. Ind. Chem. 1985, A-4, 486.

15. Palmer, D.A.; Holzhauer, J.K. U.S. Patent 1982,4 342 699,.

16. McDermott, J.X. U.S. Patent 1979, 4 151 116,.

17. Wellauer , T.P.; Gresswell, D.L. ; Newson , E.J. Chem. Eng. Sci. 1986, 41,

765

18. Heylin, M. Chem. Eng. News 1987, 2, 24.

19. Contractor, R.M. U.S. Patent 1991, 5 021 588.

20. Contractor, R.M. U.S. Patent 1987, 4 668 802.

21. Contractor, R.M.; Bergna, H.E.; Horowitz, H.S. ; Blackstone , C.M ;

Malon, B. ; Torardi , C.C.; Griffiths ; B.; Chowdhry , U.; Sleight, A. Catal.

Today, 1987, 1, 49.

References

١٩٦

22. Hucknall, D.J. Selective Oxidation of Hydrocarbon, Academic Press.

London, 1974.

23. Dadyburjor, D.B.; Jewar, S.S.; Ruckenstein, R. Catal. Rev. Sci. Eng. 1979,

19, 293.

24. Centi, G.; Trifiro, F. Appl. Catal. 1984, 12, 1.

25. Kung, H.H. Ind. Eng. Chem. Prod. Res. Dev. 1986, 25,171.

26. Grasselli, R.K.; Burrington, J.D.; Brazdil, J.F. Faraday Discuss. Chem. Soc.

1981, 72, 203.

27. Centi, G.; Trifiro, F.; Vaccari, A.; Fajonk, G.M.; Teichner, S.J. Bull. Soc.

Chim. Fr. 1981, 7-8, 290.

28. Trimm, D.L. Design of Industrial Catalysis, Elsevier, 1980, 81.

29. Bordes, E.; Courtine, P.J. Catal. 1979, 57, 235.

30. Jordan, B.; Calvo, C. Can. J. Chem. 1973, 51, 2621.

31. Jordan, B.; Calvo, C. Acta Crystallogr. Sect. B. 1979, B 32, 2899.

32. Bahargava, R.N.; Condrate, R.A. Appl. Spectrosc. 1977, 31, 230.

33. Ladwig, G.Z. Anorg. Allg. Chem. 1965, 338, 266.

34. Bordes, E. ; Courtine. P.; Pannetier, G. Ann. Chim. (Paris) 1973, 8,105.

35. Ballutaud. D.; Bordes, E.; Courtine. P.; Mater. Res. Bull. 1982, 17, 519.

36. Bordes, E.; Courtine. P.; Johnson, J.W.; Raminsona, A. Mater. Sci.

Monogr. 1985, 28, 887.

37. Bordes, E.; Courtine. P.; J. Chem. Soc. Chem. Commun. 1985, 294.

38. Tim, S.C. ; Baugherg, J.T. ;Harrison, W.T. ; Dussack ,L.L. ; Jacobson ,

A.J.; Johnson , J.W. Bordes, E. Ionics 1996, 84, 219.

39. Techez, M.; Theobald, F.; Bordes, E. J. Solid State .Chem.1981, 40, 280.

40. Tietiz, H.R. Aust. J. Chem. 1981, 34, 2035.

41. Chernorukov, N.G.; Egorov, N.P.; Korshunov, I.A. Russ. J. Inorg. Chem.

(Engl. Transl.) 1978, 23, 1479.

42. Johnson, J.W.; Jacobson, A.J.; Brody, J.F.; Rich, S.M. Inorg. Chem. 1982,

References

١٩٧

21, 3820.

43. Jacobson, A.J.; Johnson, J.W.; Brody, J.F.; Scanlon, J.C.; Lewansowski,

J.T.Inorg. Chem. 1985, 24, 1782.

44. Johnson, J.W.; Johnston , D.C. ; King , H.E.; Halbert, T.R,; Brody, J.F.;

Goshorn, D.P. Inorg. Chem. 1988, 27, 1646.

45. Huan, G.; Johnson, D.P. J. Solid State. Chem. 1991, 93, 514.

46. Voughey, J.R.; Harrison, W.T.A.; Jacobson, A.J.; Inorg. Chem. 1994, 33,

2481.

47. Harrison , W.A. ; Lim , S.C,; Voughey, J. T. ;Jacobson, A.J.; Goshorn,

D.P.; Johnson, J.W.; . J. Solid State. Chem. 1994, 113, 444.

48. Harrison , W.A. ; Voughey, J. T. ; Jacobson, A.J.; Goshorn, D.P.; Johnson,

J.W.; J. Solid State. Chem. 1995, 116, 77.

49. Harrison , W.A. ; Lim, S.C. ; Dussack, A.L. ; Jacobson, A.J.; Goshorn,

D.P.; Johnson, J.W.; J. Solid State. Chem. 1995, 118, 241.

50. Johnson, J.W.; Brody, J.F.; Alexander, R.M. Chem. Mater. 1990, 2, 198.

51. Huan, G.; Jacobson, A.J.; Johnson, J.W.; Gorcoran, E.W.Chem. Mater.

1990, 2, 91.

52. Huan, G.; Johnson, J.W.; Jacobson, A.J. Chem. Mater. 1990, 2, 719.

53. Huan, G.; Johnson, J.W.; Jacobson, A.J. J. Solid State. Chem. 1990, 89,

220.

54. Huan, G.; Johnson, J.W.; Jacobson, A.J. J. Solid State. Chem. 1991, 91,

385.

55. Huan, G.; Jacobson, A.J.; Johnson, J.W.; Goshorn, D.P.; Chem. Mater.

1992, 4, 631.

56. Huan, G.; Johnson, J.W.; Brody, J.F.; Goshorn, D.P.; Mater. Chem. Phys.

1993, 35, 199.

57. Johnson, J.W.; Jacobson, A.J.; Brody, J.F. J. Am. Chem. Soc. 1984, 106,

8123.

References

١٩٨

58. Leonowicz, M.E.; Johnson, J.W.; Brody, J.F.; Shannon, H.F.; New Sam, J.

M. J. Solid State. Chem. 1985, 56, 370.

59. Bordes, E.; Courtine. P.; J. Solid State. Chem. 1984, 55, 270.

60. Torardi, C.C.; Calabress, J.C. Inorg. Chem. 1984,23,1308.

61. Centi, G.; Trifiro, F.; Poli, G. Appl. Catal. 1985, 19, 225.

62. Busca, G.; Cavani, F.; Centi, G.; Trifiro, F. J. Catal. 1986,99,400.

63. Garbassi, F.; Bart, J.C.; Monitino, F. Petrini, Appl. Catal. 1985, 16, 271.

64. Udovich, C.A.; Edwards, R.C.; U.S. Patent 1986, 4 564 688.

65. Cavani, F.; Centi, G.; Trifiro, F.J. Chem. Soc. Chem. Commun. 1985, 492.

66. Martini, G.; Morselli, L.; Riva, A; Trifiro, F. React. Kinet. Catal. Lette.

1978, 8, 431.

67. Hodnett, B.K.; Delmon. B. J. Catal. 1984, 88, 43.

68. Hodnett, B.K.; Delmon. B. Ind. Eng. Chem. Fundam. 1984, 23, 465.

69. Hodnett, B.K.; Delmon. B. Appl. Catal. 1985, 15, 141.

70. Ai, M. J. Catal. 1986, 100, 336.

71. Pyatnitskaya, A.I.; Komashko, G.A.; Zazhigalov, V.A.; Gorokhovatskii,

Ya. B. React. Kinet. Catal. Lette. 1977, 6, 341.

72. Zhang, Y.; Martin, A.; Wolf, G.; Robs, U.; Worzala, S. Luck, H. B. Chem.

Mater. 1996, 8, 1135.

73. Trifiro, F.; Busca, G.; Centi, G.; Appl. Catal. 1986, 25, 265.

74. Weing, R.W.; Schrader, G.L. Ind. Eng. Chem. Fundam. 1986. 25, 612.

75. Ball, A.; Ferey, G.; Amoros, P.; Villeneuve, G. J. Solid State. Chem. 1989,

79, 169.

76. Shimoda, T.; Okuhara, T.; Misono, M. Bull. Chem. Soc. Jpn. 1985, 58,

2163.

77. Gorbunova, Yu. E.; Linde, S.A. Sov. Phys. Dokl. 1979, 24, 138.

78. Johnson, D.C.; Johnson, J.W.; J. Chem. Soc. Chem. Commun. 1985, 1720.

79. Johnson, J.W.; Johnson, D.C.; Jacobson, A.J. Preprints 4th.International

References

١٩٩

Symposium on the Scientific Vases for the Preparation of Heterogeneous

Catalysts (Louvain – Ia- Neuve, Belgium, Sep. 1986, P 85).

80. Villeneuve, G.; Erragh, A.; Beltran, D.; Drillon, M.; Hagenmuller, P. Matt.

Res. Bull, 1986, 21, 62.

81. Bartly , J.K.; Rhodes, C. ; Kiely ,C.J.; Carley, A.F.; Hutchings , G.J. Phys.

Chem. Chem. Phys. 2000, 2, 4999.

82. Cavani, F; Centi, G.; Trifiro, F.; Poli, G.J. Thermal. Anal. 1985, 30, 1241.

83. Cavani, F; Centi, G.; Riva, A. Trifiro, F. Catal. Today, 1987, 1, 17.

84. Volta, J.C. ; Bergeret, G.; Brover, J.P.; David , M.; Gallezot, P.; Hecquet,

G.J. Chem. Soc. Chem. Commun. 1986, 825.

85. Volta, J.C.; Bergeret, G.; David, M.; Broyer, J.P.; Hecquet, G. Catal.

Today, 1987, 1, 37.

86. Zazhigalov, V.; Pyatnitskaya, A.I.; Bechetokova, I.V.; Komashkoo, G.A.;

Ladwig, G.; Belousov, V.M. React. Kinet. Catal. Lette. 1983, 23, 119.

87. Centi, G.; Trifiro, F.; Fornasari, G. Ind. Eng. Chem. Prod. Res. Dev. 1985,

24, 32.

88. Trifiro, F.; Cavani, F; Centi, G. Appl. Catal. 1985, 15, 151.

89. Trifiro, F.; Morselli, L.; Urban, L. J. Catal. 1982, 75, 112.

90. Trifiro, F.; Centi, G.; Fornasari, G. J. Catal. 1984, 89, 44.

91. Garbassi, F.; Bart, J.C.; Tassinari, R.; Vlaic, G.; Lagard, P. J. Catal. 1986,

98, 317.

92. Buchanan, J.S.; Sundaresan, S. Appl. Catal. 1986, 26, 211.

93. Pepera , M. A.; Callahan, J.L.; Desmond, M.J. ; Milberger, E.C. ; Blum,

P.R.; Bremer, N.J. J. Am. Chem. Soc. 1985, 107, 4883.

94. Bosch, H.; Bruggink, A.A.; Ross. J.R. Appl. Catal. 1987, 31, 323.

95. Centi, G.; Trifiro, F.; Abner , J.R.; Franchetti , V.M. Chem. Rev. 1988,

88,55.

96. Guliants, V.V.; Benziger, J.B.; Sundaresan, S.; Wachse, I.E. Catal. Lette.

References

٢٠٠

1995, 32,379.

97. Benziger, J.B.; Guliants, V.V.; Sundaresan, S. Catal. Today. 1997, 33, 49.

98. Topso, H. Studies in Surface Science and Catal. 2000, 103, 1.

99. Centi, G.; Perathoner, S. Int. J. Mol. Sci. 2001, 2,183.

100. Escardino, A.; Sola, C.; Ruiz, F. An. Quim. 1973, 69, 1157, Summarized in

Ind. Eng. Chem. Prod. Res. Dev. 1979, 18, 7.

101. Wohlfahrt, K.; Hofmann, H. Chem. Ing. Tech. 1980, 52,811.

102 Mori, M.; Miyamoto, A.; Murakami, Y. J. Phys. Chem. 1985, 89, 4265.

103. Busca, G. Centi, G.; Trifiro, F. J. Am. Chem. Soc. 1985, 107, 7757.

104. Cavani, F.; Centi, G.; Manenti, I.; Trifiro, F. Ind. Eng. Chem. Prod. Res.

Dev. 1985, 24, 221.

105. Nakamura, M.; Kawi, K.; Fujiwara, Y. J .Catal. 1974, 34, 345.

106. Gleaves, J.T.; Ebner, J.R.; Kuechler .T.C. Catal. Rev. Sci. Eng. 1988, 30,

49.

107. Yang, T.J.; Lunsford, J.H. J. Catal. 1980, 63, 506.

108. Che, M.; Tench, A. J. Adv. Catal. 1982, 31, 77.

109. Nechiporuk, P.P.; Mishchenko, Yu. A.; Avetisov, A.K.; Dulin, D.A.;

Alinovski, I.O.; Gel'bashtein, A.I. Kinet. Catal. 1986, 27, 1304.

110. Khan, M. M.; Somorjai, G. A. J. Catal. 1985, 91, 263.

111. Puttock, S.J.; Rochester, C.H .J. Chem. Soc. Faraday Trans 1,

1986,82,2773.

112. Puttock, S.J.; Rochester, C.H .J. Chem. Soc. Faraday Trans 1,

1986,82,3013.

113. Munson, E.; Chen, B.; J. Am. Chem. Soc. 2002,124, 1638.

114. Ueda, W; Moro-Oka, Y.; Ikawa, T. J. Chem. Soc. Faraday Trans. 1, 1982,

78, 495.

115. Glaeser, L. C; Brazdil, J.F.; Hazle, M.A.; Mehicic, M.; Grasselli, R.K. J.

Chem. Soc. Faraday Trans. 1, 1982, 78, 495.

References

٢٠١

116. Askander, J. Ph.D Thesis, University of Technology of Compiegne, France,

1982.

117. Taufiq- Yap, Y.H.; Sakakini, B.H.; Waugh, K.C. Catal. Lette. 1997, 48,

105.

118. Beg, S.K.; Rao, M.S.; Delichere, P. Catal.Today 1991, 30, 1819.

119. Abon, M.; Bere, K.E.; Delichere, P. Catal. Today 1997, 33, 15.

120. Contractor, R.M.; Horowitz, H.S.; Sisler, G.M.; Bordes, E.; Catal. Today

1997, 37, 51.

121. Rodemerk, B.; Kubias, H.; Zanthoff, H.W.; Baerns, M. Appl. Catal. 1997,

53,203.

122. Zazhigalov, V.A.; Zaitsev, P.Yu. Belousov, N.; Wyustnek, W.H. React.

Kinet. Catal. Lette. 1984, 24, 375.

123. Wang, D.; Barteau, M.A. Catal. Lette. 2003, 90, 7.

124. Weiss, F.; Marion, J.; Metzger, J.; Coginon, J.M. Kinet. Catal. 1973, 14, 32.

125. Puttock, S.J.; Rochester, C.H. J. Chem. Soc. Faraday Trans. 1, 1986, 82,

3023.

126. Busca, G.; Centi, G.; Trifiro, F.; Lorenzelli, V. J. Phys. Chem. 1986, 90,

1337.

127. Rozhkova, E.V.; Gerei, S.V.; Gorokhovatskii, Ya. B. Kinet. Catal. 1975,

15, 618.

128. Baron, B.J.U.S. Patent 1981, 4 251 390.

129. Schineder, R.A. U.S. Patent 1975, 3 864 280.

130. Schineder, R.A. U.S. Patent 1977 , 4 043 943.

131. Ai. M. Chem. Abstr. 1971, 75, 117918 V.

132. Ai. M. J. Appl. Catal. 1986, 28, 223.

133. Lemanski , M . F .; Bremer, N . J.; Milberger, E.C. U. S. Patent 1980,

4 293 498.

134. Udovich, C.A. Bertolacini, R.J. U.S. Patent 1982,4 328 126.

References

٢٠٢

135. Ebner, J.R.; Gleaves, J.T. U.S. Patent 1986, 4 626 412.

136. Edwards, R.C.; Meyers, B.L. U.S. Patent 1987,4 652 543.

137. Zazhigalov, V.A Kinet. Catal. 1986, 23, 364.

138. Zazhigalov, V.A; Konovalova, N.D.; Stock, E. Kinet. Catal. 1988, 29,364

139. Raminsona, A.; Bordes, E.; Courtine, P. J.Solid State Chem. 1987, 68, 1.

140. Zazhigalov, V.A; Haber, J.; Stoch ,J.; Pyatnitzkaya ,A.I.; Komashko, G.A.;

Belousov, V.M. Appl. Catal. 1993, 96,135.

141. Cheng, W. Appl. Catal. 1996, 147, 55.

142. Guliants, V.V.; Benziger, J.B.; Sundaresan, S.; Wachse, I.E.; Jehng, M.

Catal. Lette. 1996, 28,275.

143. Sajip, S.; Bartley, J.K.; Burrows, A.; Sananés – Schulz, M.; Tuel ,A.; Volta,

J.C. ; Kiely, C.J.; Hutching, G.J. New J. Chem. 2001, 25,125.

144. Taufiq- Yap, Y.H.; Tan, K.C.; Waugh,K.C.; Hussein, M.Z. ; Ramil, I. ;

Abdul Rahman, M.B. Catal. Lette. 2003,98,87.

145. Zazhigalov, V.A; Yaremenko, V.E.; Ayub, I.; Su, D.; Willinger, ;

Kharlamov.; Ushkalov.; Schlogle, R.DGMK- Conference , Munich 2004.

146. Stefani, G. Fontana, P. U.S. Patent 1978, 4 085 122.

147. Katsumoto, K.; Marquis, M. U.S. Patent 1979, 4 132 670.

148. Horowitz, H.S.; Blackstone, C.M.; Sleight, A.W.; Tufer, G. Appl. Catal.

1988, 38,193.

149. Bartley, J.K.; Sanchez, J.A.L.; Hutching, G.J. Catal. Today 2003, 81, 197.

150. Kamiya, Y.; Ueki. S.; Hiysohi, N.; Yamamoto, N.; Okuhara, T. Catal.

Today 2003, 78, 281.

151. Bartley, J.K.; Kiely, C.J.; Wells, R.P.K.; Hutching, G.J. Catal. Lette. 2001,

72, 99.

152. Buchanan, J.S.; Apostolakis, J.; Sundaresan, S. Appl. Catal. 1985, 19, 65.

153. Centi, G.; Trifiro, F. Catal. Rev. Sci. Eng. 1986, 28,165.

154. Martine, G.; Trifiro, F.; Vaccari, A.J. Phys. Chem.1982, 86, 1573.

References

٢٠٣

155. Brkic, D.; Trifiro, F. Ind. Eng. Chem. Prod. Res. Dev. 1979, 18, 333.

156. Varma, R.L.; Saraf, D.M. J. Catal. 1978, 55,361.

157. Cavani, F.; Centi, G.; Trifiro, F. Ind. Eng. Chem. Prod. Res. Dev. 1983, 22,

570.

158. Centi, G. Catal. Today 1993, 16, 5.

159. Hodnett, B.K. Catal. Today 1987, 1, 477.

160. Owens, L.; Kung, H.H. J. Catal. 1993, 144, 202.

161. Guliants, V.V. Catal. Today 1999, 51, 255.

162. Watchs , I.E.; Jehng, J.M.; Deo, G.; Weckhuysen, B.M.; Guliants, V.V.;

Benziger, J.B. ; Sundaresan, S.J. Catal. 1997, 170, 75.

163. Deo, G.; Watchs, I.E. J. Catal. 1994, 146, 323.

164. Deo, G.; Watchs, I.E. J. Catal. 1994, 146, 335.

165. Ledoux, M.J.; Crouzet, C.; Pham- Huu, C.; Turines, V.; Kourtakis, K.;

Mills, P.L.; Lerou, J. Catal. 2001, 203, 495.

166. Olthof, B.; Khodakov, A.; Bell, A.T.; Iglesia, E. J. Phys. Chem. B 2000,

104, 1516.

167. Busca, G.; Lorenzelli, V.J. Mol. Catal. 1989, 50, 231.

168. Agaskar, P.A.; DeCaul, L.; Grasslli, R.K. 1994, 23, 339.

169. Guliants, V.V.; Benziger, J.B.; Sundaresan, S.; Wachse, I.E.; Jehang, J.M.

Catal. Today. 1996, 28,275.

170. Zazhigalov, V.A; Haber, J.; Stoch ,J.; Bacherikova, I.V.; Komashko , G.A.;

Pyatnitzkaya ,A.I.; Appl. Catal. 1996, 134,225.

171. Centi, G.; Nieto, J.L.; Ungarelli, F.; Trifiro, F. Catal. Lette. 1990, 4,309.

172. Bordes, E. Thesis, University of Technology of Compiegne, France, 1980.

173. Martin, V.; Miller, J.M.; Volta, J.C.J. Thermal Anal. 1998, 86, 731.

174. Silverstin, M.R; Webster, X.F. Spectrometric Identification of Organic

Compounds, John Wiley and Sons Inc. 1998.

175. Machei, T.; Ruiz, P.; Delmon, B. J. J. Chem. Faraday Trans 1990, 86, 731.

References

٢٠٤

176. Nakamoto, K. Infra- Red and Raman Spectra of Inorganic and Coordination

Compounds, P.B John Wiley and Sons Inc. 5 th.1997.

177. Scharf, U.; Schneider, M.; Baiker, A.; Wokaun, A.J. Catal. 1994, 149, 344.

178. Abelo, Li; Husson, E.; Replin, Y.; Lucazau. U.G.; Spectrochim. Acta. 1983,

39,641

179. Went, G.T.; Oyama, St.; Bell, A. J. Phys. Chem. 1990, 94, 240.

180. Dr. Jerry, Li, University of Oklahoma, U.S.A.; Personal Communication

2005.

181. Abdelouahab, F.B.; Olier, R.; Guilhanme, N.; Lefebrre, F.; Volta, J.C.J.

Catal. 1992, 134, 151.

182. Overbeek, R.A.; Warringa, P.A.; Crombag, M.J.D.; Visser L.M.; Van

Dellen, A.J.; Geus, J.W. Appl. Catal. 1996, 135, 209.

183. Overbeek, R.A.; Pekelharing, A.R.C.J.; Van Dellen, A.J.; Geus, J.W. Appl.

Catal. 1996, 135, 231.

184. Ruitenbeek, M.; Van Dellen, A.J.; Koningsberger D.C.; Geus, J.W. Studies

in Surface Science and Catalysis 1998, 118, 549.

185. Ruitenbeek, M.; Van Dellen, A.J.; Barbon, B.; Van Faassen, E.E.;

Koningsberger D.C.; Geus, J.W. Catal. Lette. 1998, 55,133.

186. Ruitenbeek, M. Ph.D Thesis, University of Utrecht, Holland, 1999.

187. Askander, J. DEA Report, University of Technology of Compiegne, France,

1978.

188. Moser, T.P.; Schrader, G.L. J. Catal. 1985, 92, 216.

189. Busca, G.; Lavalley, J. Spectrochim Acta.1986, 42A, 443.

190. Riva, V.; Martítn, C. J. Catal. 1988, 114, 473.

191. Liskowski, A.; Su, D.C.; Schlogl, R.; Bartly, J.K.; Hutchings, G.H; Macros.

Micronal 2003, 9, 316.

192. www.semiconfast.com/edx.htm. June 2005.

References

.ھــــ�� ـــ�هللا �ـــ�� ــ�ــ� ا�ـــ��ـــ� : ط�� ا��راه

�� : ��ان ا�ط�و�ـ��ـ ) ا��"�ز ("! ��ات ا��ـ�( ا)ـ'��ـ ص ا%ــ�$�ــ وا��ــ ).2%�6ر أو�%4ـ�3 –�2�د�0م ( �ـ/�م ا-ـ�ا,ـ+ ا�ـ%��ة

. ٢٠٠٥/ ٨/ ٣: "�ر90 ا��78

.٣٧د – ٣٥ز – ٨٥٣م –�ــ� ا�!,ـــ�A– Bــ@�اد : ا%?3 ��ان

D"�Eا B8٥٥٩٢٠٤٠: ر

٠٧٩٠١٧٣٠٧٥٥: رB8 ا�4ال

�� [email protected]: ا��0 ا�?��و

Effects of Titania (Anatase) on

Structural & Superficial Properties

of

(VPO) Catalytic System

A thesis

Submitted to the College of Science

Al-Nahrain University

In partial fulfillment of the requirements for the

Degree of Doctor of Philosophy of Science in

Chemistry

By

Hind Abdullah Ali Hammo AL-TA'EE

B.Sc. Baghdad University 1987

M.Sc. Baghdad University 2000

June 2005

Special Dedication

To

My country, My Family,

And To My supervisors

III

We certify that this thesis was prepared under our supervision at Al-

Nahrain University / Collage of Science as a partial fulfillment of the

requirement for the degree of Ph.D in Chemistry.

Signature : Signature :

Name : Dr. Jalal A. Askander Name : Dr. Shahbaz A. Maki Address : ALFRED ENGINEERING Address : Chemistry Department Date : / /2005 Date : / /2005

Supervisor Supervisor

In view of the available recommendations, I forward this thesis for

debate by the examination committee.

Signature : Name : Dr. Shahbaz A. Maki

Address : Chairman of Chemistry Department Date : / /2005

IV

We certify that we have read this thesis and as examining

committee examined the student (Hind A. Ali AL-TA'EE) in its

content and that in our opinion it is adequate for the partial fulfillment

of the requirement for the degree of Ph.D in chemistry, with ( ) .

Signature : Signature :

Name : Dr. Moayyed G .Jalhoom Name : Prof. Falih H. Musa Title :Chief of Scientific Researchers Title : Professor

(Member)

(Member)

Signature

:

Signature

:

Name :Dr. Samir M. Jasim Name : Dr. Ramzi R. Ali Al-Ani

Title : Assistant Professor Title : Assistant Professor

(Member)

(Member)

Signature:

Signature:

Name: Dr. Jalal A. Askander Name: Dr .Shahbaz A. Maki Title: Senior Scientific Researcher Title: Assistant Professor

(Supervisor) (Supervisor)

Approved by the council of Collage of Science

Signature : Name : Laith A. Al-Ani

Address : Dean of the College of Science – Al-Nahrain University Date :

Signature :

Name : Prof. Saadoon A. Isa

Title : Vice president of Scientific affairs – Al-Nahrain University

(Chairman)

V

Acknowledgements

I would like to express my deep appreciation and sincere gratitude to my supervisor

Dr. Jalal A. Askander for his suggestions of this work, for his valuable guidance

constructive criticisms, and encouragement during the development of this work and in

the preparation of this thesis.

I wish to express my gratitude and thanks to Dr. Shahbaz A. Maki for his assistance

and supervision during this work and during the preparation of this thesis.

I also would like to thanks Dr. Ameer A. Ameer for his supervision assistance at the

first period of this work.

My thanks go to the Chairman and the staff of the department of Chemistry, College

of Science (Al- Nahrain University) for their encouragement. My deep thank to Dr.

Sawsan Harith for her explanations for FTIR results. My thanks to Mrs. Nasreen and

Mrs. Yassmen for their help in performed FTIR analysis

Deep thanks to Prof. Valery Zazhigalov the head department of heterogeneous

catalysis at The National Academy of Science of Ukraine for his help by making several

analysis (DTA, SSA, and Acidity miscreants, EDX and his generosity by sending the

samples to Germany in order to perform XRD there), I also would like to appreciate his

help by sending some of his published papers.

I would like to express my deep thanks to Prof. Danielle Resasco and Dr. Jerry Li at

University of Oklahoma – U.S.A, for there help in making XPS and Raman analysis and

sending some important papers.

My deep thanks for Mr. Dahfar S. Al-Ani and Dr. Hani Al-Kadi and Dr. Hussein Al-

Hindawi at American University of Sharjah for there help in making SEM analysis.

I would like to thanks Prof. E. Bordes (CNRS laboratories –University of Lille France)

and Prof. Jozef Kowalewski (Stockholm University - Sweden) for their supports and

invitations to perform part of this work in their laboratories in future

My appreciates to Prof. G. Hutchings (Chairman of Chemistry Department at

Cardiff University- England), Dr. M. Ruitenbeek (DSM Centre – Holland) and Prof.

Sleight, A. (U.S.A), Prof. V. V. Guliants (Cincinnati University – U.S.A), and Prof. G.

Clarizia (research Institute on Membrane Technology, ITM - CNR ,former IRMERC-CNR-

Italy) for their assistance by sending many useful published papers.

Finally my deep thanks and appreciate to everyone who helps me in this work.

Hind

VI

Symbols and Abbreviations

VPO = Vanadium – Phosphorus – Oxygen. MA = Maleic Anhydride. XRD = X-ray diffraction. ICP = Inductively Coupled Plasma. ESCA = Electron Spectroscopy Chemical Analysis. SEM = Scanning Electron Microscope. BET = Bruner – Emit – Teller. TEOS = Tetra ethyl o- silicate. EXAFS = Energy X-ray absorption fine structure. TOF = Turnover frequency = number of molecules

converted per Vanadium per second. ▼ = Main phase. ■ = Trace amount. SSA = Specific surface area. LA = Lewis acidity centers. BA = Bronsted acidity centers. XPS = X-ray photoelectron spectroscopy. Py = Pyridine. Dmpy = 2, 6- Dimethyl Pyridine. un = Unsupported. supp = Supported. A = Anatase. TG = Thermogravimetric analysis. DTA = Differential thermogravimetric analysis. Endo = Endothermic. Exo = Exothermic. T.G = Technical grade. S.S = Solid solution. AV.OX.No.Of V = Average oxidation number of Vanadium ion. eV = Electron volt. (s) = On surface. EDX = Energy dispersive X-ray analysis.

XI

Contents Page No. Certificate III Acknowledgement V Symbols and Abbreviations VI Summary VII Contents XI List of tables XIII List of figures XV List of scheme XV

VIII Chapter One

Introduction 1.1 Background 2 1.1.1 Processes for the production of MA from n-butane 3 1.2 Scope of the literature survey 9 1.3 Structure of the catalyst 11 1.3.1 Crystalline (VPO) phases 11 1.3.2 Active phase of n–butane selective conversion of MA 24 1.4 Kinetic of n-butane oxidation 28 1.5 Surface modifications by interaction of n-butane with

catalyst surface 33

1.6 Relationship between redox properties of catalytic behavior 38 1.7 Mechanism of n-butane activation 41 1.8 Role of adsorbed and lattice oxygen species 46 1.9 Nature and mobility of adsorbed species 50 1.10 Promoted (VPO) catalyst 51 1.11 Preparation of the catalyst 56 Conclusion of bulk (VPO) catalyst 60 1.12 Supported (VPO) catalyst 61 1. 12.1 Molecular structure of Vanadia species in fresh catalyst

molecular 62

1.12.2 Molecular structure in dehydrated state 64 1.12.3 Structural changes during hydrocarbon oxidation 66 1.12.4 Structure–activity relationships in oxidation of C-4

hydrocarbon 67

1.12.5 Role of surface oxygen species in supported catalysts 69 1.12.6 Role of acidity of supported vanadia catalyst 76 1.12.7 Role of Support 77

XII

Contents Page No.

Conclusions 78 Aim of work 81

Chapter two Experimental Part

2.1 Samples analysis 83 2.2 Chemicals 87 2.3 Preparation 88 2.3.1 Route A 88 2.3.2 Route B (VOHPO4.0.5H2O) 90 2.4 Redox titration 93

Chapter three Results and Discussion Part One ( The results)

3.1 Results 98 3.1.1 The elemental analysis of technical grade V2O5 98 3.1.2 Identification of prepared phases by XRD 199 3.1.3 Thermal analysis 104 3.1.4 Fourier Transform Infra-Red (FT-IR) spectrum 111 3.1.5 Laser Raman Spectroscopy 128 3.1.6 Redox titration 132 3.1.7 Specific surface area measurement 135 3.1.8 Surface acidity measurement 136 3.1.9 X-ray photoelectron spectroscopy (XPS) analysis 138 3.1.10 Scanning Electron Microscopes (SEM) analysis 149 3.1.11 Energy dispersive X-ray (EDX) analysis 155

Part Two (Discussion) 3.2.1 Effect of preparation method 167 3.2.2 Promoters, impurities, and Solid Solution effects 170 3.2.3 Thermal behaviors (DTA and (TG) 172 3.2.4 FT-IR and Raman structural properties 175 3.2.5 Bulk oxidation state (Redox titration) 181 3.2.6 Effect of support on the specific surface area 181 3.2.7 Acidity effect 182 3.2.8 Surface chemistry by XPS analysis 183 3.2.9 Morphology measurement by SEM 185 3.2.10 Identify the elemental composition by using EDX 196 3.2.11 Resumption of the catalytic activity 197 Conclusions 190 Future prospects 193 References 195

XIII

XII

List of Tables Fig. No. Title Page No. 1.1 World MA capacity (in metric tons) by reactor type. 6 1.2 Total world production and consumption of MA (X10 3

metric tons). 6

1.3 Number of electrons and oxygen molecules involved in some principle reactions of industrial intersect in selective oxidation

8

1.4 The effect of metal oxide support in n-butane oxidation on supported Vanadia catalysts at 494 K in 1.2 Vol.% n-butane.

58

1.5 Performance of titania support in n-butane oxidation at 479 K in 1.2 Vol.% n-butane in air.

63

1.6 The effect of acidic promoters on n-butane oxidation on the 1% V2O5 /TiO2 catalyst at 497 K in n-butane in air.

63

2.1 Chemical used in this work. 73 3.1 Elemental analysis of technical grade V2O5 used in this

work 82

3.2 VOPO4.2H2O (sample No.1-un) X-ray diffraction data. 84 3.3 VOHPO4.0.5H2O and VOHPO4.H2O samples (No.2-un and

2-supp) X-ray diffraction data. 85

3.4 VO (H2PO4)2 (sample No.3-un and 3-supp) X-ray diffraction data.

86

3.5 Identification of prepared phases by XRD. 87 3.6 DTA analysis results of precursors. 92 3.7 The comparison between theoretical and experimental loss of

weight % for catalytic phase. 93

3.8 Fourier Transform Infra-Red data of V2O5. 97 3.9 Fourier Transform Infra-Red data of supported V2O5. 100 3.10 Fourier Transform Infra-Red data of VOPO4.2H2O. 103 3.11 Fourier Transform Infra-Red data of VOHPO4.0.5H2O and

VOHPO4.H2O. 107

3.12 Fourier Transform Infra-Red data of VO (H2PO4)2. 110 3.13 Raman shift carried out by Red laser. 113 3.14 Raman shift carried out by Green laser. 114 3.15 Results of redox titration 115 3.16 Results of redox titration during time of reaction for

supported VO (H2PO4)2. 116

3.17-A Specific surface area for the catalytic phases. 118 3.17-B Specific surface area for VOPO4.2H2O 119 3.18 Data of acidity centers on the surface of catalytic phases 120 3.19 The binding energy of elements on the surface of catalytic

phases. 122

3.20 Atomic concentration of the elements on the surface of the catalytic phases.

129

3.21 The atomic ratio of the elements on the surface of the catalytic phases.

130

3.22 EDX results of the catalytic phases. 147

XIII

List of Figures Figure No. Title Page No.

1.1-A Structure of VOPO4.2H2O showing infinite layers of PO4 tetrahedral linked to VO6 octahedral

10

1.1-B Structure of VOPO4.2H2O showing bond distance °(A) and angles (degree)

11

1.2 Structure of α- and β- VOPO4 12 1.3 Comparison between structure of α- and β- VOPO4 13

1.4 Crystal structure γ-VOPO4

A- 010 plane B- 100 plane

14

1.5 Crystal structure of VOHPO4.0.5H2O (010) plane 15

1.6 View of the infinite double chains of (VO6) octahedra and (PO4) tetrahedra running along (100) direction in VOHPO4.0.5H2O

15

1.7-A Projection of the structure of β- VOHPO4.2H2O along (100)

17

1.7-B Projection of the structure of β- VOHPO4.2H2O along (010)

17

1.8 Idealized structure of (VO)2P2O7 (020) pane 18 1.9 Crystal structure of (VO)2P2O7 (020) plane. 18

1.10 Pseudomorphizm relations between the crystal structure of VOHPO4.0.5H2O and (VO)2P2O7.

20

1.11 Crystal structure of VO (H2PO4)2 21

1.12 Termination of the (200) plane of (VO)2P2O7 and proposed active sites for oxidation

24

1.13 Models of the ideal Langmuir surface 26

1.14 Comparison of the selectivity dependence on acrylonitrile in propane and propane ammoxidation at 480°C on VSb4 + Sb2O4.

29

1.15

Selectivity in propane formation from propane on (VO)2P2O7 as a function of the inlet between oxygen and propane concentration for two entail concentration of propane in the feed reaction temperature 322 °C

30

1.16 Effect of the presence of different valance states of Vanadium on the principle X-ray diffraction lines of Vanadyl pyrophosphate (P/V = 1.0)

32

1.17-A Maleic Anhydride yield Vs conversion at 300 °C in n-butane oxidation for different catalysts with (P/V = 1.0) feed: 0.6 n-butane, 12 % O2.

33

1.17-B Rate of V (IV) oxidation to V (V) in air for the same catalysts (A).

34

1.18 Model of Lewis acid sites (VO)2P2O7 with different degrees of disorder in the stacking fold of (020) planes.

38

1.19 Proposed mechanism of n-butane activation on (VO)2P2O7

39

1.20 Reaction network in n-butane oxidation on (VO)2P2O7 40

Figure No. Title Page No.

XIV

1.21 The three types of oxygen species 41 1.22-A Proposed reaction of n-butane to MA 43 1.22-B Proposed reaction of n-butane to MA 49 1.23 Synthesis routes for Vanadyl Pyrophosphate catalysts 51

1.24 Molecular structure of dehydrated isolated and polymerized vanadia species on oxide supports.

54

1.25 Oxidation of n-butane on dispersed Vanadia species. 57

1.26-A In situ Raman spectra of 1% V2O5 / TiO2 catalysts during n-butane oxidation.

60

1.26-B In situ Raman spectra of 7% V2O5/SiO2 catalysts during n-butane oxidation.

60

1.26-C In situ Raman spectra of 7% V2O5/Nb2O5 catalysts during n-butane oxidation.

61

1.26-D In situ Raman spectra of 7% V2O5/ZrO2 catalysts during n-butane oxidation.

61

1.27 Performance of Titania support in n-butane oxidation on supported Vanadia catalysts at 494 K in 1.2 Vol. % n-butane in air.

63

3.1 XRD spectrum of samples (2-un, 2-supp,3-un and 3-supp).

83

3.2 TG and DTA of VOHPO4.0.5H2O (sample No.2-un). 88 3.3 TG and DTA of VOHPO4. H2O (sample No.2-supp). 89 3.4 TG and DTA of VO (H2PO4)2 (sample No.3-un). 90 3.5 TG and DTA of VO (H2PO4)2 / TiO2 (sample No.3-supp). 91 3.6 FT-IR spectrum of Fluka V2O5. 95 3.7 FT-IR spectrum of technical grade V2O5 95 3.8 FT-IR spectrum of Fluka V2O5 / TiO2. 98 3.9 FT-IR spectrum of technical grade V2O5 / TiO2. 99 3.10 FT-IR spectrum of VOPO4.2H2O (sample No.1-un). 101

3.11 FT-IR spectrum of VOPO4.2H2O/ TiO2 (sample No.1-supp).

102

3.12 FT-IR spectrum of VOHPO4.0.5H2O ◊. 104

3.13 FT-IR spectrum of VOHPO4.0.5H2O (sample No.2-un) 105

3.14 FT-IR spectrum of VOHPO4.H2O/TiO2 (sample No.2-supp)

106

3.15 FT-IR spectrum of VO (H2PO4)2 (sample No.3-un). 108

3.16 FT-IR spectrum of VO (H2PO4)2/TiO2 (sample No.3-supp).

109

3.17A,3.17B Raman shift for prepared phases carried out by Red Laser and Green Laser.

112

3.18 The relationship between the AV.OX.No.of V ion and V (IV) %

116

3.19 The relation between the V (IV) percent with time. 117

3.20 The binding energy spectrum of the elements on the surface.

121

3.21 The XPS spectrum of VOPO4.2H2O (sample No.1-un). 123

3.22 The XPS spectrum of VOPO4.2H2O/TiO2 (sample No.1-supp).

124

Figure No. Title Page No. 3.23 The XPS spectrum of VOHPO4.0.5H2O (sample No.2-un) 125

XV

3.24 The XPS spectrum of VOHPO4.H2O/ TiO2 (sample No.2-supp)

126

3.25 The XPS spectrum of VO (H2PO4)2 (sample No.3-un). 127

3.26 The XPS spectrum of VO (H2PO4)2/TiO2 (sample No.3-supp).

128

3.27 SEM; Microstructure of VOPO4.2H2O (sample No.1-un). 132

3.28 SEM; Microstructure of VOPO4.2H2O/TiO2 (sample

No.1-supp). 132

3.29 SEM ; Microstructure of VOHPO4.0.5H2O (sample No.2-

un) 133

3.30 SEM; Microstructure of VO (H2PO4)2 (sample No.3-un). 134

3.31 SEM; Microstructure of VO (H2PO4)2/TiO2 (sample No.3-

supp). 135

3.32 A comparison between the morphology of unsupported

and supported 136

3.33 The EDX result of VOHPO4.0.5H2O (sample No.2-un) 138 3.34 The EDX result of VOHPO4.0.5H2O (sample No.2-un) 139

3.35 The EDX result of VOHPO4. H2O/ TiO2 (sample No.2-supp)

140

3.36 The EDX result of VO (H2PO4)2 (sample No.3-un). 141 3.37 The EDX result of VO (H2PO4)2 (sample No.3-un). 142 3.38 The EDX result of VO (H2PO4)2 (sample No.3-un). 143

3.39 The EDX result of VO (H2PO4)2/TiO2 (sample No.3-

supp). 144


supp). 145


supp). 146

XVI

List of Schemes Scheme No. Title Page No.

1.1 Huntsman fixed-bed reactor for MA production.

4

1.2 ALMA Fluidized-bed reactor for MA. 5

1.3 DUPANT recalculating – solid process for the production of THF from n-butane via MA.

5

1.4 Kinds of oxygen adsorbed species 9

1.5 A triangular reaction network on n-butane to MA.

27

2.1 Phase's preparation diagram from technical grade V2O5.

77

VII

Summary

The catalytic activity of Vanadyl Acid Phosphate towards the reaction

of n-butane oxidation to Maleic Anhydride ( MA) was achieved in our

previous work .So it is useful to study the effect of addition on (TiO2)

anatase as a support on structural properties and on its surface chemistry.

This phase is prepared in alternative method never maintained in the

literature. Route (A) depends on the reaction of the technical grade (V2O5)

with (o-H3PO4 ) in order to prepare the intermediate phase (VOPO4.2H2O)

which is then being reduced by using a mixture of ( Benzyl Alcohol and

Isobutanol) to reach the assigned phase VO (H2PO4)2 . While the

traditional route (B) depends on the reduction of Vanadium pentoxide

using alcohol mixture and then adding phosphoric acid to the reduced

phase.

The structural properties and surface chemistry of Vanadyl Acid

Phosphate and the intermediate phase prepared by route (A) and Vanadyl

Hydrogen Phosphate Hemihydrate prepared by route (B) are studied by

having confidence in number of physical and chemical analysis performed

in international known laboratories in U.S.A, Germany, Republic of

Ukraine and U.A.E. by using special techniques devoted in catalytic

structural and surfaces studies.

The identification of prepared phases is carried out by using X- ray

diffraction (XRD) in Fritz – Haber institute. The results of phase

identification show the co- existence of traces of other phases like VO

(H2PO3)2 with the dominated main phase prepared by rout (A). These

results also indicate that never reach very pure prepared phase which is

VIII

known phenomena in the field of preparation of (VPO) system. It was

appeared through these identification examination that the addition of TiO2

anatase could affect the resultant phases prepared in rout (B), since a

monohydrate composition (VOHPO4.H2O) is obtained in sated of Vanadyl

Hydrogen Phosphate Hemihydrate (VOHPO4.0.5H2O) normally prepared

without the use of support . No effect of support is observed on the main

prepared phase by rout (A).

The differential thermogravimetric behaviors for the supported and

unsupported precursors are studied.. The percentage of total loss of weight

for Vanadyl Acid Phosphate VO (H2PO4)2 / TiO2 is determined for the first

time.

The comparison of FT- IR spectrum of prepared supported and

unsupported phases with laboratory grade and technical grade of

Vanadium pentoxide are achieved. The obtained results emphasis on the

presence of wide shift in the band positions, these shifts reflects the

deformation of the angles and length of bonds in these phases. It is well

known that FT-IR is not adequate technique to study solid state structural

properties. However laser Raman spectroscopy could be helpful in this

aspect. Numbers of samples using this technique were carried out in

Oklahoma University laboratories in U.S.A. The presence of (TiO2

anatase) is accorded in the structure and layer of mono vanadate layer are

identified in all phases prepared, however the poly vanadate layer are

identified only in the prepared Vanadyl acid phosphate, this phenomenon

is very useful to support the employment of this phase, as a catalyst in the

reaction of n-butane to Maleic Anhydride and it is very good sign of

expected catalytic activity.

Vanadium ion oxidation stare and the percentage distribution for V (5+)

, V (4+) and V (3+) oxidation state are studied for supported and

unsupported phases by using redox titration technique , the results indicate

IX

that in the presence of the support the vanadium ion oxidation state

decreases toward (4+) oxidation state.

Desorption at Argon temperature technique was used for specific

surface area determination. The results state that there are significant

increases in specific surface area when adding support to intermediate

VOPO4.2H2O preparation. A specific surface area value for the first time is

assigned for the phase prepared by rout (A).

The X- ray photoelectron spectroscopy (XPS) technique was used on

samples in University of Oklahoma – U.S.A in order to study the surface

chemistry of the precursor VOHPO4.0.5H2O , VO (H2PO4)2 and

VOPO4.2H2O , the results point out the presence of structural elements (V,

P ,O), and detectable percentage of inorganic impurities on the surface and

also imply the presence of Nitrogen (N 1s) which is attributed to

Ammonium metavanadate, This compound was used as intermediate in

technical grade vanadium pentoxide preparation. The results also suggest

the presence of several oxygen species on the phases prepared. By using

this technique which can offer the (P/ V) atomic ratio, the prediction of

catalytic activity of supported phase can be assumed.

The Lewis and Bronsted acidity centers on the surface of the samples

were effectuated in National Academy of Science laboratories in republic

of Ukraine by using pyridine and 2, 6- Dimethyl pyridine adsorption. The

results indicated the doubled value of Lewis acidity centers in the

presence of support in phase VO (H2PO4)2. This is another signal to

predict activity as maintained in the literatures.

The investigation of surface of prepared phases samples were

performed in the American University of Sharjah. The results indicate

obtaining of lamellar crystalline characterized which is identical to these

proposed by literature.

X

Finally the combination between energy dispersive X-ray (EDX)

technique with Scanning Electron Microscope (SEM) technique was

carried out on samples in National Academy of Science laboratories in

republic of Ukraine in order to reach the elementally atomic percentage

down to depth of 1 µm from the surface. The results indicate the presence

of TiO2 in selected spots.

In general this work emphasize on the intercalation phenomenon

between structure layers and solid solution formation through out several

analysis which are made. This work can predict the catalytic activity of

Vanadyl Acid Phosphate VO (H2PO4)2 prepared phase enhancement by the

presence of TiO2 as support in the n-butane oxidation to Maleic Anhydride

ىـعل) ازـاتـنأ( الـتـيـتـانـيـايرات ـتأث

ةـيـة والسطحـبيـص التركيـالخصائ

) أوكسجين –فسفور –ناديوم ڤ( العوامل المـساعـدة لنظام

ةــالـرس

نـريـهـة النـعـامـج –وم ـلـعـة الـيـلـى كـة الـدمـقـم

اءـيـمـيـفي الك ةـسفـفل توراهـدك ةـيل درجـبات نـطلـزئي لمتـال جـكأستكم

من قبل

يـائـو الطـمـي حـلـداهللا عـد عبـنـه

١٩٨٧ –جامعة بغداد –بكلوريوس

٢٠٠٠ -جامعة بغداد –ماجستير

٢٠٠٥ زيرانح ـه١٤٢٦ جمادي االولى

ي ومـا ـبـرَ ِ◌ أمـر مـنْ وحُ ل الـرُ ح قـُـوُ الـرٌ ِ◌ عـن كَ ونَ سئلُ يَ وَ ﴿ ﴾ االقـليال ً ِ العلمَ◌ م منأوتيتُ

صدق اهللا العظيم )سورة االسراء ( ٨٤ا��

VII

Summary

This work is a continuous of our previous work in M.Sc study. Our

previous work was concerned with the use of locally produced V2O5

(technical grade) in the preparation of catalyst successfully employed in

the oxidation of n-butane to Malice Anhydride. The catalyst was

evaluated using geomechanic system (France) specially built for catalytic

test and modified to our proposes. The results had indicated to a functional

catalyst. The catalytic properties were attributed to several factors (highly

disordered, P/V around 1.1, adequate specific surface area, preparation

method and other factors) (1).

Since the catalytic activity of Vanadyl Acid Phosphate towards the

reaction of n-butane oxidation to maleic anhydride (MA) was achieved in

our previous work. It is useful to study the effect of addition (TiO2) anatase

as a support on structural properties and surface chemistry of the catalyst.

This precursor is prepared in an alternative method never mentioned in the

literature. Route (A) depends on the reaction of the technical grade (V2O5)

with (o- H3PO4 ) in order to prepare the intermediate phase (VOPO4.2H2O)

which is then reduced by a mixture of (benzyl alcohol and isobutanol) to

reach the assigned phase VO (H2PO4)2 . While the traditional route (B)

depends on the reduction of vanadium pentoxide by alcohol mixture and

then adding phosphoric acid to the reduced phase.

The structural properties and surface chemistry of vanadyl acid

phosphate and the intermediate phase prepared by route (A) and vanadyl

hydrogen phosphate hemihydrate prepared by route (B) are studied after

carrying out physical and chemical analysis in international known

laboratories in U.S.A, Germany, Republic of Ukraine and U.A.E. by using

special techniques devoted in catalytic structural and surfaces studies.

VIII

The identification of the prepared phases was carried out by using X-

ray diffraction (XRD). The results of phases identification show the co-

existence of traces of other phases like VO (H2PO3)2 with the dominated

main phase prepared by rout (A). These results also indicate that a very

pure phase was never reached phase which is known phenomenon in the

field of preparation of (VPO) system. It was found through these

identification examination that the addition of TiO2 (anatase) could affect

the resultant phases prepared in rout (B), since a monohydrate composition

(VOHPO4.H2O) is obtained instead of Vanadyl Hydrogen Phosphate

Hemihydrate (VOHPO4.0.5H2O) normally prepared without the use of

support . No effect of support is observed on the main phase prepared by

route (A).

The differential thermogravimetric behaviors for the precursors were

studied. This study highlights only the supported and unsupported

precursors in order to assign the effect of TiO2 (anatase) on the thermal

behavior, two effects were observed; the first one is increase the

percentage of total loss of weight for supported Vanadyl Acid Phosphate

(13.86 %), the second effect is the types of stage which in the precursor

started it dehydration; the unsupported phase started the dehydration in (2

endothermic peak between 50-27º C) while the supported phase started its

dehydration in one endothermic peak between ( 110-180 º C ). The total

loss in weight of supported VO (H2PO4)2 is reported here for the first time.

The comparison of FTIR spectrum of prepared supported and

unsupported phases with laboratory grade and technical grade of vanadium

pentoxide are achieved. The obtained results emphasis on the presence of

wide shifts in the band positions. These shifts reflect the deformation of

the angles and lengths of bonds in these phases. Laser Raman spectroscopy

could be helpful in this aspect. Structural properties were further

confirmed by studying the effect of presence of TiO2 (anatase) is accorded

IX

in the structure, and layers of mono vanadate layer are identified in all

prepared phases. However the poly vanadate layers are identified only in

the prepared Vanadyl acid phosphate. This phenomenon is very useful to

support the employment of this phase, as a catalyst in the conversion of n-

butane to maleic anhydride and it is very good sign of expected catalytic

activity.

The oxidation state of Vanadium ion and the percentage distribution for

V (5+), V (4+) and V (3+) in the bulk of precursors are studied for

supported and unsupported phases by using redox titration technique. The

results indicated that in the presence of the support the vanadium ion

oxidation states change towards (4+) oxidation state.

Desorption at Argon temperature technique was used for specific

surface area determination. The results state that there are significant

increases in the specific surface area when adding support to the

intermediate phase VOPO4.2H2O (16 m2.g -1) and the result show an

increase in this value for supported VO (H2PO4)2 (11 m2. g -1). A specific

surface area value for supported phase is reported here for the first time.

The X- ray photoelectron spectroscopy (XPS) technique was used in

order to study the surface chemistry of the two precursors

(VOHPO4.0.5H2O), and VO (H2PO4)2 and the intermediate phase

VOPO4.2H2O. The results pointed out the presence of structural elements

(V, P ,O), and detectable percentage of inorganic impurities on the surface

and also imply the presence of Nitrogen (N 1s) which is attributed to

ammonium metavanadate, This compound was used as intermediate in

technical grade vanadium pentoxide preparation. The results also suggest

the presence of several oxygen species on the prepared phases. The results

also indicate that the average oxidation state for prepared precursors on the

surface is (5 +). By using this technique which can offer the (P/ V) atomic

X

ratio, the prediction of catalytic activity of supported phase can be

assumed.

The Lewis and Bronsted acidity centers on the surface of the samples

were effectuated by adsorption of Pyridine and 2, 6- Dimethyl pyridine

adsorption. The results indicated the doubled value of Lewis acidity

centers in the presence of the support in VO (H2PO4)2 phase (2.050 X107

mol. m2) for unsupported sample and (4.420050 X107mol.m2) for

supported sample). This is another signal to predict the activity as

mentioned in the literature.

The investigation of the surface of the prepared phase's samples has

indicated the formation of lamellar crystalline character which is identical

to those proposed by literature.

Finally, the combination between energy dispersive X-ray (EDX)

techniques with Scanning Electron Microscope (SEM) technique was

carried out on samples in order to reach the elementally atomic percentage

down to depth of 1 µm from the surface. The results indicated the presence

of TiO2 in selected spots.

In general this work emphasize on the intercalation phenomenon

between structure layers and solid solution formation through out several

analysis which are made. This work can predict enhancement of Vanadyl

Acid Phosphate VO (H2PO4)2 prepared phase by the presence of TiO2 as a

support in the n-butane oxidation to Maleic Anhydride.

الصــــــــــةــــالخ

هــتم عملنــا الســابق باســتخدام أ. بق فــي دراســة الماجســتيراالســ لبحثنــاهــذا البحــث هــو اســتمرار

مســاعد اســتخدم بنجــاح فــي عامــلالمنــتج محليــا فــي تحضــير ، خــامس اوكســيد الفنــاديوم الصــناعي

اعد باسـتخدام منظومـة فرنسـية تـم تقيـيم العامـل المسـ. تان الطبيعي الى انهدريد المالييـكبيو أكسدة ال

اشـارت .بنيت لغرض تقييم العوامل المساعدة وقد تم تحويرهـا الغـراض فحـص هـذا العامـل المسـاعد

. ي الــى انهدريــد المالييــكعــالنتــائج الــى فعاليــة العامــل المســاعد المحضــر فــي تحويــل البيوتــان الطبي

ديوم العــالي ، نســبة الفســفور الــى الفنــالالنتظــام ا( الــى عــدة عوامــل منهــا ةعزيــت الفعاليــة التحفيزيــ

).، المساحة السطحية المناسبة ، طريقة التحضير وعوامل أخرى١,١المساوية الى

اكسـدة تفاعـل تجـاها ) Vanadyl Acid Phosphate(بعـد تاكيـد الفعاليـة التحفيزيـة للطـور

ثير اضـافة الطـور السـاند ثـاني من المفيد دراسة تـأ صبحأ ،الطبيعي الى انهدريد المالييك البيوتان

امـل عللعلـى الخصـائص التركيبيـة وكيميـاء السـطح ) TiO2 Anatase ( أوكسـيد التيتـانيوم

) أ ( يعتمــد الطريــق اذ ، االدبيــاتبطريقــة لــم يســبق االشــارة اليهــا فــي حضــر هــذا الطــور. المحفــز

لتحضــير الطـور الوســطي يكعلـى تفاعـل خــامس اوكسـيد الفنــاديوم الصـناعي مـع حــامض الفوسـفور

(VOPO4.2H2O) وتانول ـيـــــول البنزايــــل وااليزوبـن كحـــــج مـــــدام مزيـــــخـــــالــــذي يجــــري اختزالــــه باست

، بينمــا تعتمــد الطريقــة التقليديــة) Vanadyl Acid Phosphate( وصــوال الــى الطــور المعــين

ت واضــافة حــامض بواســطة مــزيج الكحــوال الفنــاديوم دعلــى اختــزال خــامس اوكســي ))ب (الطريــق (

. الفسفوريك على الطور المختزل

))ا ( الطريق (ور الوسطي ـوالط ورـذا الطـهـح لـاء السطـيـدراسة الخواص التركيبية وكيم تتم

ويـة اميياسـتنادا الـى عـدد مـن التحاليـل الفيزياويـة والك) )ب(الطريق ( بموجبايضا طور المحضر لوا

ميـة المعروفـة فـي الواليـات المتحـدة والمانيـا وأوكرانيـا واالمـارات لت العافي عدد مـن المختبـرا المنجزة

اذ ، وباســــتخدام تقنيـــات متخصصــــة فــــي دراســـات الســــطوح والعوامــــل المســــاعدة ، العربيـــة المتحــــدة

ـــود االشـــعة الســـينية للمس ـــة حي ــــشخصـــت االطـــوار المحضـــرة باســـتخدام تقني X-ray ). احيقـ

Diffraction )ن ـار مــود اثــوجـ ترافق ))ا ( الطريق (في وارـلالط ئج التشخيصرت نتاـأظهلقد و

كمـا . (Main Phase) ي الغالـبــــمـع الطـور الرئيس) VO (H2PO3)2)( اطوار اخرى مثل

ير االطـوار ـروفة فـي مجـال تحضــمعـ ةرـ ي وهـي ظاهــور نقــول الى طــكن الوصـارت الى انه اليمـأش

المـادة اضـافة كمـا تبـين مـن خـالل هـذا الفحـص ان ،)اوكسـجين –ر فسفو –فناديوم ( ام ـظنفي

اذ) ب (علـى نـواتج التفاعـل للطريـق اثـر قـد ) TiO2 Anatase التيتانيـا (Support) اندةـــالس

ة تحتــوي علــى ـبـــكير ت الحصــول علــىحيــث تــم انــه قــاد الــى عــدم الحصــول علــى الطــور المطلــوب

في . (VOHPO4.0.5H2O) بدال من نصف جزيئة VOHPO4.H2O)( دةـاء واحـزيئة مـج

حيـث تـم )) أ( الطريـق ( اندة على الطور الرئيسي المحضر مـنــــحين لم يؤثر وجود المادة الس

.كطور رئيسي في كلتا الحالتين ) VO(H2PO4)2 ( الحصول على

و تناولت الدراسة تسليط ، ) Precursors( وك الحراري التفاضلي لالطوار البادئةلدرس الس

) Support(لتحديد تأثير المادة الساندة )المسندة وغير المسندة( بادئة فقط لالضوء على االطوار ا

وقد اشارت النتائج الى ارتفاع نسبة الفقـدان فـي الـوزن فـي حالـة االطـوار . اوك الحراري لهلعلى الس

ــــــث بلغــــــت المســــــندة فــــــي حــــــين بلغــــــت TiO2 /(VO(H2PO4)2 للطــــــور % ١٣,٨٦(حي

كـذلك اشـارت النتـائج الـى ان فقـدان المـاء لهـذا الطـور فـي .للطور غير المسـند ) % ١٢,٠٠(النسبة

في المدى endothermic)(حالة كونه غير مسند في المرحلة االولى تتم بقمتين ماصتين للحرارة

اعثــة للحــرارة فــي المــدى فــي حــين ان الطــور المســند يبـدأ بقمــة واحــدة ب) درجـة مئويــة ٢٧٠- ٥٠(

TiO2 /VO( للطــور الــوزن دانـقـــوية لفـة المئـــتــم تحديــد النسبــ لقــد .)درجــة مئويــة ١٨٠-١١٠(

(H2PO4)2 ( بالطريق المحضر )الول مرة) ا.

Fourier Transform Infrared spectroscopy) الحمـراء اطيـاف االشـعة تحـت درست

FTIR) مقارنتهــا مــع خــامس أوكســيد توتمــ ) المســندة وغيــر المســندة( لكافــة االطــوار المحضــرة

مـع ذلـك المشـتق القياسـي ) VOHPO4.0.5H2O(وكـذلك تمـت مقارنـة الطـور ،الفنـاديوم النقـي

دلــت النتــائج علــى وجــود انحــراف كبيــر فــي مواقــع الحــزم و . الصــناعي مــن خــامس اوكســيد الفنــاديوم

فـي) deformation( بقية االطـوار وهـذا مؤشـر علـى وجـود التشـوه علىوانسحب هذا االنحراف

.هذه االطوارفي زوايا واطوال االواصر

( Solid state) التعويل على هذه التقنية فـي دراسـة المـادة الصـلبة ه اليمكنمن المعروف ان

، )Laser Raman Spectroscopy(ةيـــــلـــــذا تـــــم االســـــتعانة بتقنيـــــة مطيـــــاف رامـــــان الليزر

رف علـى ـعــتـكمـا تـم ال ) TiO2 Anatase( التيتانيـاجـود مـن خـالل هـذه التقنيـة تـم التحقـق مـن و

وتبـين مـن خـالل Poly) ( أو متعـددة )(Mono ت منفـردةـما أذا كانــات الفناديـت فيــة طبقــنوعيـ

ود ـت علـى وجــفيما دل ،ميع االطوار المحضرة بالطريقتينـجردة متكونة في ـفـنـالفحص ان الطبقة الم

جــدا حيــث تــدعم ةظــاهرة مفيــد وهــي ) VO (H2PO4)2( ي الطــور ـددة فـــتعـــردة والمـنفـــة المـالطبقــ

مهمـة اشـارة يوهـ ،الـى المالييـك انهدريـدسـدة البيوتـان الطبيعـي أك عـلهذا الطور في تفال نااستخدام

.له على الفعالية المتوقعة

لحـاالت االكسـدة حددت حالة االكسدة اليون الفناديوم وكذلك النسبة المئوية اليونـات الفنـاديوم

سندة باستخدام تسحيح للالطوار المسندة وغير الم (bulk)في الكتلة الخماسية والرباعية والثالثية

حالـة االكسـدة اليــون تحـول واشـارت النتـائج الــى ، (Redox Titration) أختـزال –االكسـدة

. لجميع االطوار المحضرة بوجود المادة الساندة+ ) ٤( الفناديوم باتجاه حالة االكسدة الرباعية

Desorption at Argon( االركــــون حــــرارة عنــــد درجــــة االمتــــزاز أســــتخدمت طريقــــة

temperature ( لقيـــاس المســـاحة الســـطحية للطـــور ) (VOHPO4.0.5H2O . وقـــد اشـــارت

ـــى حـــدوث ارتفـــاع ملحـــوظ فـــي المســـاحة الســـطحية للطـــو ـــائج ال (VOPO4.2H2O)ر الوســـطيالنت

للطـور غيـر الواحـد متـر مربـع للغـرام ١١( حيث ارتفعـت مـن )Support (بوجود المادة الساندة

لمســــاحة اكــــذلك تــــم الول مــــرة تحديــــد قيمة ) متــــر مربــــع للغــــرام للطــــور المســــند ١٦ المســــند الــــى

متــر مربــع للغــرام ١٠(حيــث كانــت )VO (H2PO4)2 ) (ا (ضــر بالطريقــةحالســطحية للطــور الم

. )للطور المسند الواحد متر مربع للغرام ١١للطور غير المسند وارتفعت الى

) VO (H2PO4)2و VOHPO4.0.5H2O) (لبادئـــة كيميـــاء الســـطح لالطـــوار ا تدرســـ

X-ray photoelectron)باســــــتخدام تقنيــــــة ) VOPO4.2H2O (وللطــــــور الوســــــطي

spectroscopy XPS) الـــى وجـــود اشـــارت النتـــائج لقـــد . ألطيـــاف االشـــعة الســـينية الضـــوئية

رات الشـوائب ذنسـبة البـاس بهـا مـن علـى السـطح والـى وجـود (P,V,O)عناصر التركيبة البلورية

وهـي التـي يحتويهـا علـى سـطوح االطـوار المحضـرة ) Fe, Ca , Na, Al , Si( مثـل الالعضـوية

ان مصـدر النتـروجين علـى .وجـود النتـروجين الـى كمـا اشـارت ،) (V2O5خامس أوكسيد الفنـاديوم

Ammonium metavanadate)( االمونيــوم فنــاداتميتا ود الــىـيعــ ة ـقاســـوار المـســطوح االطــ

NH4VO3 واكــدت وجــود .الصــناعي خــامس أوكســيد الفنــاديوم هــو طــور وســطي فــي انتــاج و

على تأكيد حصـول ظـاهرة المحلـول ايضادلت هذه التقنية و فصائل مختلفة اليونات االوكسجين،

اشــارت النتــائج . فــي بعــض االطــوار بســبب التيتــانيوم والصــوديوم (solid solution) الصــلب

الل ـن خــكمـا تـم مـ+) ٥(الـة االكسـدة اليـون الفنـاديوم فـي االطـار المحضـرة هـو كذلك الـن معـدل ح

فـــي وهـــي مرتفعـــة (P/V ratio) ب الذريـــة للفســـفور والفنـــاديومـســــهـــذه التقنيـــة التوصـــل الـــى الن

VOحيـــث بلغـــت هـــذه النســـبة فـــي الطـــور االطـــوار المســـندة عنهـــا فـــي االطـــوار غيـــر المســـندة

(H2PO4)2 / TiO2 )تقائيــة العامــل ننســبة مهمــة فــي الــتحكم بفعاليــة واهــذه ال تعــدو ، )١,١٠

. المساعد

ام طريقــــة دـتخـــــباس نمــــاذجللقيســــت حامضــــية مراكــــز لــــويس وبرونشــــد علــــى ســــطوح االطــــوار

Dimethyl -2,6) )وثنــائي مثيــل البــردين -٦،٢(و ) (Pyridineامدصــاص البــردين

Pyridine) د تضـاعفت بوجـود المـادة السـاندة ة حامضـية مراكـز لـويس قـقيمـ ان واظهرت النتائج)

Support ( في الطور) VO (H2PO4)2 ( حيث بلغـت) X 10 7 مـول للمتـر المربـع ٢,٠٥٠

للطـور المسـند )مول للمتر المربع الواحد 4.420X 107( في حين بلغت للطورغير المسند )الواحد

.لك المشار اليها في ادبيات الموضوعوهذه اشارة اخرى الى فعالية افضل من ت ،

Scanning Electron)ةـيــــنـقـدام تـخـــتـضرة باسـوح االطـــوار المحــــطـــس ةـاينــــمع تـتمــ

Microscope SEM) ودلـت النتـائج علـى الحصـول علـى سـطوح ، المجهـر االلكترونـي الماسـح

لضـمان فـي ادبيـات الموضـوع،مشـابهة لمـا موجـود وهي صـفة ا،له ) Lamellar( مستوية رقائقية

. فعالية أفضل للعامل المساعد

نيـة مـع تق ) Energy Dispersive X-ray( وأخيـرا تـم موالفـة تقنيـة تشـتت االشـعة السـينية

للوصـول الـى ) (Scanning Electron Microscope SEMالمجهـر االلكترونـي الماسـح

واشـارت تلـك النتـائج بوضـوح ، نمـاذج عـدة فـي عن السطحµ 1) ( لسمكالنسب الذرية للعناصر

.افة في البقع المختارة سالى وجود التيتانيوم على هذه الم

بــين الطبقــات الوكســيد التيتــانيوم (Intercalation) لقــد أكــد هــذا البحــث وجــود ظــاهرة اقحــام

اليــــل تحمــــن خــــالل ال . (Solid solution) حصــــول ظــــاهرة المحلــــول الصــــلب و ،البلوريــــة

.ذا البحثـي هـها فـم اجرائـعددة التي تـالمت

التنبــؤ بفعاليــة ،) Vanadyl Acid Phosphate( ة التــي حضــر فيهــا الطــور ـبالطريقــ كنـمـــي

سيؤدي الى تحسين الصـفات (Support)ان استخدام المادة الساندة كما ،مساعد عامل ك له جيدة

. صناعيا الييكانهدريد الم النتاجالطبيعي اكسدة البيوتان تفاعل فيالتحفيزية له