1

SCIENTIFIC PAPER

Magnesium salts as compounds of the preparation of magnesium

oxide from Tunisian natural brines

Souheil Behij; Halim Hammi*; Ahmed Hichem Hamzaoui; Adel M’nif

Technological park of Borj Cedria

National Center of Research in Materials Sciences

Valorization Laboratory of Useful Materials

Tourist road of Soliman B. P. 73 - Soliman 8027, Tunisia

Received 7.12.2011.

Revised 8.6.2012.

Accepted 8.6.2012.

* Corresponding author : Halim Hammi, Technological park of Borj Cedria, National Center of

Research in Materials Sciences, Valorization Laboratory of Useful Materials, Tourist road of Soliman

B. P. 73 - Soliman 8027, Tunisia; Tel: +216 79 325 470; Fax: +216 79 325 314; E-mail :

halim.hammi@inrst.rnrt.tn

2

Abstract

Magnesium oxide is one of the most important magnesium compounds used in industry. The

production of MgO is often done from calcined magnesium carbonate or from natural

magnesium saline solutions (sea water and brines). In the case of these solutions, magnesium

oxide is precipitated after the addition of a strong base (eg. Ammonia). Magnesium hydroxide

is calcined after its separation from the excess resulting from the strong base through

filtration. Thus, magnesia qualities may differ depending on several physical parameters and

particularly on the nature of the compound. Consequently, two different compounds were

selected: magnesium chloride and magnesium sulphate which can be recovered from Tunisian

natural brines. Three physical factors were considered: calcination temperatures, precipitation

temperatures and calcination time of Mg(OH)2. The decomposition of Mg(OH)2 was

investigated by DTA/TGA. Mass losses vary in the range (23.0%-29.9%). Starting

decomposition temperatures are between 362°C and 385°C. The MgO produced from MgSO4

under 1000°C within 48 hours of calcination time and with 40°C as a reaction temperature for

Mg(OH)2 shows a good crystallinity and is of a cristallyte size of 86.3 nm and has a specific

surface area equal to 16.87 m2g

-1. Finally, morphological differences between MgO

agglomerates at different temperatures were observed by SEM. Consequently, magnesium

sulphate as precursor for preparing MgO is selected.

Keywords: Cristallinity; Crystallite size; Mg(OH)2; MgO or periclase; Specific surface area.

3

INTRODUCTION

Magnesium oxide (MgO or periclase) is among the most important industrial magnesium

compounds. Approximately 20% of the world’s production comes from seawater, brines and

desalination of reject brine [1]. Magnesium oxide is used as an exceptionally important

material in catalysis [2,3], toxic waste remediation [4], or as additives in refractory products,

paint, manufacture of fertilizers, animal foodstuff, building materials (Sorel cement,

lightweight building panels) and superconductor products [5-7]. A panel of fundamental and

applied studies exists in literature reviews [7-11]. It particularly shows that magnesium

hydroxide production from seawater or brines is precipitated by the addition of a strong base

(eg. Ammonia). Later on, separation is calcined to produce MgO. It is important to note that

magnesia qualities may differ depending on the physico-chemical conditions of preparation

and on the compound type which is why two compounds are selected to be tested: magnesium

chloride and magnesium sulphate. The respective final products are compared on the basis of

their X-ray diffractogram, specific surface area and microstructural differences.

To accomplich this task we first make a DTA/DTG decomposition study of Mg(OH)2

(brucite). After that, XRD Results are used to study the impact of calcination temperature, the

impact of calcination time and finally the impact of Mg(OH)2 reaction temperature.

Additionally, X-ray diffractograms are used to calculate MgO crystallite size depending on

some physical parameters such as calcination temperatures and reaction times. At a later

stage, the specific surface area is determined in order to confirm the particles’ cristallyte size

variations. Finally, MgSO47H2O is selected as the compound used for MgO preparation.

EXPERIMENTAL

Methodology

The industrial exploitation of the abundant natural brines in the Sebkhas and Chotts of the

Tunisian South is conditioned by the development of specific processes, the technique and

economic viabilities of which have to be demonstrated beforehand. For this purpose, we

agreed to study this kind of saline solutions previously studied by many authors[12,13],

assimilated to a quinary system represented by Na+, K

+, Mg

2+/ Cl

-, SO

2- // H2O.

Natural Brine is a highly concentrated complex aqueous solution. In the case of the brine

called Sebkha El Melah, magnesium, chloride and sulphate ions seem to be sufficiently high

to recover magnesium salts like epsomite (MgSO4.7H2O) and bischofite (MgCl2

.6H2O) useful

in a wide range of applications (such as fertilizers, Sorel cement, and refractory compounds).

4

Epsomite is obtained by polythermal crystallization conducted in two steps. The first one is

completed at 35°C, temperature during which the brine is evaporated to reach a density of

1.29 g.cm-3

and the maximum concentrations of magnesium and sulphate ions. The second

step consists in cooling the above mentioned pretreated solution at 0°C in order to crystallize

MgSO4.7H2O.

Purification of the product, realized by washing it with a saturated solution of Epsom salts,

provides 99% Epsomite purity [14].

For bischofite synthesis, the concluded process is mainly composed of six stages based on

two main unit operations, firstly isothermal evaporation and crystallization and then chemical

precipitation.

The resulting brine, composed mainly of magnesium chloride, is concentrated by isothermal

evaporation at 35°C to precipitate magnesium chloride. The recovered product, collected from

the final stage, is mainly composed of bischofite the purity of which exceeds 90% [15].

MgO powders preparation is done in two steps starting from two different precursors:

bishofite (MgCl2.6H2O) and epsomite (MgSO4

.7H2O) successfully produced from Tunisian

natrual brines.

At first, each reactant was dissolved in deionized water at room temperature to produce 0.8 M

Mg2+

solution. After that, an excess of ammonium hydroxide was added and vigorously

stirred at 50°C with pH = 10. A white precipitate of Mg(OH)2 was obtained which was

thoroughly washed with distilled water and dried at 100°C during 2 hours. At this stage, the

following reaction takes place:

Mg2+

+ 2OH- Mg(OH)2 (1)

In a second step, magnesium hydroxide powders were calcined to produce MgO. The heating

treatment was carried out in an electric furnace by raising the temperature 10°C per minute.

Therefore, the below thermal reaction takes place:

Mg(OH)2 t MgO + H2O (2)

Materials

The X-Ray Diffractometer with a PHILIPS PW 3040 generator, goniometer PW 3050/60 θ/2θ

and a cathode of copper PW 3373/00 is used to distinguish MgO from two different

precursors. Microstructural differences between MgO agglomerates were examined using

SEM FEI Quanta 200. The decomposition of precipitated Mg(OH)2 was analysed by

5

DTA/TGA (DSC) SETARAM SETSYS EVOLUTION. The specific surface area was

determined by a QuantaChrome Autosorb-1 apparatus.

RESULTS AND DISCUSSION

Starting from the two compounds prepared from the natural brine Sebkha El Melah as

described above (MgCl2.6H2O & MgSO4

.7H2O), the two successive steps of brucite and

magnesia preparation are performed as previously indicated. The thermogravimetric

decomposition (DTA/DTG) of Mg(OH)2 was studied. The recovered products are

characterized by XRD and SEM.

Mg(OH)2 characterization

Mg(OH)2 decomposition (DTA/DTG) study

The start of Mg(OH)2 decomposition is studied using a DTA/DTG experiments. Mg(OH)2

experimental diagrams related to the compounds magnesium chloride and magnesium

sulphate are respectively presented in Fig. 1(a) and Fig. 1(b) It becomes possible to compare

masses’ loss of the two obtained powders to the theoretical ones from these two diagrams.

Figure 1(a)

Figure 1(b)

The above two DTA-DTG analyses show two endothermic peaks respectively related to

dehydration process and to magnesium hydroxide decomposition. The second phenomenon

occurs at 362°C and 385°C respectively for the compounds magnesium chloride and

magnesium sulphate. However, in terms of brucite (Mg(OH)2) theoretical decomposition,

30.864% mass loss should be achieved, which is slightly larger than the observed one

29.883% and 23.022% respectively. This is due to the incompletion of the decomposition

reaction within this temperature range. These results are in good agreement with others

presented elsewhere [16-18].

As a preliminary conclusion it comes into view that Mg(OH)2 prepared from the two studied

compounds is of good chemical purity but its morphology varies from one compound to

another. Furthermore starting decomposition temperature differs depending on the

compound’s nature. In order to clarify the compound’s impact on the final product, we

studied MgO prepared under different physical conditions.

Mg(OH)2 X ray diffraction description

The magnesium hydroxide prepared from the two above compounds was characterized by X-

rays, it was found that the produced compounds correspond to Mg(OH)2 regardless of the

considered compound. Figure 2 shows a marked diffractogramme of the compound obtained

from MgSO4.7H2O; an identical one is produced when MgCl2 is used as precursor.

6

Figure 2

Mg(OH)2 SEM description

The morphology of Mg(OH)2 from different precursors is explored by SEM analysis. Brucite

was dried accordingly at 100°C and after cooling was observed. Microstructural differences

were detected for the two compounds (Fig. 3). This observation indicates that magnesium

oxide products may present different physical properties related to the compound.

Magnesium hydroxide derived from magnesium sulphate shows a plate-like shape, the

particles were joined to each other forming spherical particles around 15 m which tend to

form large agglomerates (30 m).

Figure 3

MgO characterization

MgO X-ray diffraction study

X-ray diffraction will allow, in a first step, to qualitatively compare the crystallinity of the

obtained materials; such a parameter is related to the peaks’ intensity and their affinities. In a

second step, the diffractograms will be used to calculate the average crystallite size of the

prepared MgO in a more quantitative term. Regarding these two parameters, the following

effects are investigated: calcination temperature, calcination time and reaction temperature.

crystallinity

o Calcination temperature effect

As previously concluded, Mg(OH)2 decomposition starts at the range varying from 362 to

385°C, thus a calcination temperature ranging from 500°C to 1000°C for a duration of 2 hours

was operated in order to study the effect of such a factor on the different obtained products

from the two studied compounds. Figure 4a and b illustrate X-ray diffractograms

respectively for the precusors magnesium chloride and magnesium sulphate.

Figure 4

Figure 5 shows that all samples are magnesium oxide formed with CFC structure (NaCl type).

The best crystallinity corresponds to the calcination temperature of 1000°C with regard to the

two compounds. Nevertheless, it is important to note that MgO, obtained from MgSO4.7H2O,

presents the highest intensity and the sharpest peak at the same temperature. In addition,

unlike the MgCl2 compound, MgSO4.7H2O shows a gradual increase of intensity parallel to

that of temperature. These remarks clearly indicate the influence of the compound on the

properties of the final product (MgO).

7

o Calcination time effect

MgO calcination time is studied in the range 2h to 48h. Fig. 5 illustrates XR diffractogramms

for MgO obtained from magnesium sulphate at 1000°C (a similar result is concluded for the

other compound). Therefore, increasing the calcination time improves the crystallinity. As a

result, 48 hours of calcination time give the best results. CFC magnesium oxide is always

obtained.

Figure 5

o Reaction temperature effect

Mg(OH)2 reaction temperature parameter was studied in the range 25°C – 75°C. Fig.6

illustrates only XRD results for MgO obtained from the compound magnesium sulphate at

1000°C and with 48 h of calcination time. The variations of the peak intensities, which play

the function of temperature, are irregular. However, in terms of crystallinity, the below

diffractograms enable us to deduce that 40°C is the best reaction temperature for the two

compounds.

Figure 6

average crystallite size

The above diffractogramms are handled to determine the average crystallite size, L, of the

prepared powders. For this pupose, we applied Scherrer formula:

cos

9.0L (3)

where:

is the wavelength of the X-ray,

is the diffraction angle associated with a Bragg peak,

2

1

22 )( sm is the corrected full widh at half maximum (FWHM),

m being the FWHM of the well defined (200) Bragg peak,

s being that of a standard crystallized sample of MgO (Across sample).

Considering the two tested reactants, the crystallite sizes with calcination temperature,

calcination time and reaction temperatures are shown respectively in Fig. 7, Fig. 8 and Fig.9.

Figure 7

Figure 8

Figure 9

8

Fig. 7 shows that for the two precursors, the crystallite size increases in accordance with the

increase of calcination temperature. Crystallite sizes are within the ranges: (64.7 nm to 115.4

nm) and (18.1 nm to 86.3 nm) respectively for the compounds magnesium chloride and

magnesium sulphate. Therefore, the MgO obtained from magnesium sulphate has the smallest

particle size. This result enhances the relation between the full width at half maximum of

Bragg peaks and crystallite size. In fact, the FWHM is the result of the convolution of an

instrumental contribution and a size effect.

For the compound magnesium sulphate, the crystallite size increases with calcination time

(Fig. 8). For the other one (magnesium chloride), the evolution of crystallite is irregular. The

final sizes are lower than those obtained from magnesium sulphate after 48 hours of

calcination time. Calcination time effect on crystallite size seems to be more important than

that of the temperature effect (150 nm in 48 h and 80 nm at 1000°C).

Fig. 9 shows that MgO produced from magnesium chloride has an irregular evolution of

crystallite sizes. It is therefore difficult to confirm the previous conclusion, but for energy

considerations 40°C will be retained as the reaction temperature.

BET Specific surface

In order to confirm the above results, the specific surface area by BET for MgO powders

burnt-off at 1000°C during 24 hours are set up. The obtained results are 9.31 m2g

-1 and 16.87

m2g

-1 for magnesium chloride and magnesium sulphate respectively. These results confirm

that the sample obtained from the magnesium sulphate has the highest specific surface area

which is in concordance with the data [19-22] indicating that small particles have the highest

surface areas.

SEM study

The morphology of MgO powders obtained from different compounds is explored by SEM

analysis. Periclase powders (MgO) obtained at 1000°C during 48 hours from the two

compounds after cooling and SEM examination show the presence of fine particles, less than

1 m, forming agglomerates (less than 5 m) with homogeneous distribution (Fig.10).

At 1000°C, a well-defined plate-like morphology was observed which can be related to

periclase phase, a similar morphology has been reported elsewhere [23] .

CONCLUSION

The results of this study demonstrate that the MgO obtained from the two compounds has

different physical proprieties such as the crystallite size, shape, and structure. A set of

9

parameters like calcination temperature, calcination time and temperature of brucite

precipitation has to be controlled to produce a high quality product.

Intermediate and final products were characterized by X-ray diffraction, DTA/DTG analysis,

BET and SEM.

This study shows that 40°C is an appropriate temperature to obtain Mg(OH)2 from the

compounds used in the experiments.

The calcination temperature for MgO production with high crystallinity is set to be within the

range 800-1000°C. Within this interval, the maximum brucite decomposition into periclase is

achieved. Calcination time of 48h confirm the above result with a very pure product. Finally,

magnesium oxide obtained from magnesium sulphate is the best approach to obtain the ideal

product, due to its higher surface area, and eventually, to its smallest primary particle size at

the limits of calcination temperature. Besides, the specific area value can be related to the

microstructure of the sections disposed around a central nucleus. According to these results,

we select magnesium sulphate as the precursor to obtain magnesium oxide.

ACKNOWLEDGMENTS

This work was supported by the National Centre of Material Science Research. The authors

would like to thank the technical team for helpful assistance with the XRD, SEM and

DTA/DTG studies.

10

REFERENCES

[1] K.H. Bûchel, H.-H. Moretto, P. Woditch ; Industrial Inorganic chemistry. Wiley-Vch. (2000) 235.

[2] S.H.C. Liang, I.D. Gay, J. Catal. 10 (1986) 293. 295- .

[3] H. Tsuji, F. Yagi, H. Hattori, H. Kita, J. Catal. 148 (1994) 759-770. [4] A.N. Copp, Am. Ceram. Soc. Bull. 74 (1995) 135-137.

[5] W. Wang, X. Qiao, J. Chen, J. .Am. Ceram. Soc. 91 (2008) 1697-1699.

[6] Y.S. Yuan, M.S. Wong, S.S. Wang, J. Mater. Res. 11 (1996) 8-17.

[7] P. Yang, C. M. Lieber, Science. 273 (1996) 1836 1840- .

[8] G.W. Wagner, P.W. Bartram, O. Koper, K.J. Klabunde, J. Phys. Chem. B. 103 (1999) 3225-3228.

[9] A.I. Boldyrev, J. Simons, J. Am. Chem. Soc. 119(20), (1997) 4618–4621.

[10] M. Zhou, H. Zhu, Y. Jiao, Y. Rao, S. Hark, Y. Liu, L. Peng, Q. Li, J. Phys. Chem. C. 113 (20), (2009)

8945–8947.

[11] M. Sterrer, O. Diwald, E. Knozinger, J. Phys. Chem. B. 104 (2000) pp 3601-3607.

[12] R. Cohen-Adad, Chr. Balarew, S. Tepavitcharova, D. Rabadjieva, Pure Appl. Chem. 74 (2002) 1811-1821

[13] H. Hammi, J. Musso, A. M’nif, R. Rokbani, J. Calphad 27 (2003) 71-77

[14] H. Hammi, A. Mnif, R. Rokbani, J. Phys IV. Pr10 (2001) 157-163. [15] R. Fezei, H. Hammi, A. Mnif, LAAR. 39 (2009) 375-380.

[16] E. Alvarado, L. M. Torres-Martinez, A. F. Fuentes, P. Quintana, Polyhedron. 19 (2000) 2345-2351.

[17] J. A. Wang, O. Novaro, X. Bokhimi, T. Lopez, R. Gomez, J. Navarrete, M. E. Llanos, E. Lopez-Salinas,

Mater. Lett. 35 (1998) 317-323.

[18] S. Ardizzone, C. L. Bianchi, M. Fadoni, B. Vercelli, Appl. Surf. Sci. 119 (1997), 253-259.

[19] V.R. Choudhary, V.H.Rane, R.V.Gadre, J. Catal., 145 (1994), 300-311.

[20] A.Pilarska, T.Jesionowski, Physicochem. Probl. Miner. Process, 46(2011) 83-94

[21] A.Pilarska, E.Markiewicz, F.Ciesielczyk, T. Jesionowski, Drying Technology, 29 (2011)1210–1218

[22] W. Wang, X. Qiao, J. Chen, H. Li, Mater. Lett. 61 (2007) 3218-3220.

[23] J. Staron, S. Palco, Am. Ceram. Soc. Bull. 72 (1993) 83-87.

11

Figure captions

Fig. 1. DTA/DTG of Mg(OH)2 obtained from two different precursors : (a) MgO from

chlordie, (b) MgO from sulphate

Fig.2. Diffractogram of Mg (OH)2 prepared from MgSO4.7H2O.

Fig.3. Microstructure of Mg(OH)2 powders obtained from different precursors (2500x)

Fig.4. Calcination temperature effect on MgO obtained from different compounds :

(a) Magnesium chloride, (b) Magnesium sulphate.

Fig.5. Diffractogram showing the calcination time on MgO obtained from magnesium

sulphate precursor.

Fig.6. MgO obtained from magnesium sulphate diffractograms vs reaction temperature

Fig.7. Development of crystallite size with temperature on the crystallisation of MgO.

Fig.8. Development of crystallite size with calcination time on the crystallisation of MgO.

Fig.9. Development of crystallite size with reaction temperature on the crystallisation of

MgO.

Fig.10. Microstructure of MgO powders obtained from different precursors (5000x)

12

Figure 1 (a)

Figure 1 (b)

13

Position [°2Theta]

20 30 40 50 60 70 80 90 100 110

Counts

0

2000

4000

1 1

1;

bru

cite

3 1

1;

bru

cite

2 1

2;

bru

cite

2 2

2;

bru

cite

5 1

0;

bru

cite

6 0

1;

bru

cite

3 3

3;

bru

cite

6 2

1;

bru

cite

6 1

3;

bru

cite

3 0

5;

bru

cite

4 4

4;

bru

cite

6 4

4;

bru

cite

5 1

6;

bru

cite

5 5

5;

bru

cite

8 6

1;

bru

cite

Mg(OH)2

Search Unit Cell Result 1

Figure 2

Mg(OH)2 sulfate Mg(OH)2 chloride

Figure 3

2

Counts

14

Figure 4

Figure 5

(b)

(a)

2

2

2

Counts

Counts

Counts

15

Figure 6

Figure 7

Figure 8

0

20

40

60

80

100

120

140

400 500 600 700 800 900 1000 1100

Temperature (°C)

Cry

sta

llit

e s

ize (

nm

)

Chloride

Sulfate

0

20

40

60

80

100

120

140

160

2 6 12 24 48 Calcination time ( hours )

Chloride

Sulfate

Cry

sta

llit

e s

ize (

nm

)

Counts

2

16

Figure 9

MgO sulfate MgO chloride

Figure 10

0 20 40 60 80

100 120 140 160 180

25 35 40 50 75

Reaction temperature (°C)

Chloride

Sulfate

Cry

sta

llit

e s

ize (

nm

)