This article was downloaded by: [New York University]On: 17 October 2014, At: 06:35Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,37-41 Mortimer Street, London W1T 3JH, UK

Fuel Science and Technology InternationalPublication details, including instructions for authors and subscription information:http://www.tandfonline.com/loi/lpet19

METHANOL-TO-GASOLINE VS. DME-TO-GASOLNE II.PROCESS COMPARISON AND ANALYSISSunggyu Lee a , Makarand Gogate a & Conrad J. Kulik aa Process Research Center Department of Chemical Engineering , The University of Akron ,Akron, Ohio, 44325-3906Published online: 10 May 2007.

To cite this article: Sunggyu Lee , Makarand Gogate & Conrad J. Kulik (1995) METHANOL-TO-GASOLINE VS. DME-TO-GASOLNE II. PROCESS COMPARISON AND ANALYSIS, Fuel Science and Technology International, 13:8, 1039-1057, DOI:10.1080/08843759508947721

To link to this article: http://dx.doi.org/10.1080/08843759508947721

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in thepublications on our platform. However, Taylor & Francis, our agents, and our licensors make no representationsor warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Anyopinions and views expressed in this publication are the opinions and views of the authors, and are not theviews of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should beindependently verified with primary sources of information. Taylor and Francis shall not be liable for any losses,actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoevercaused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematicreproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in anyform to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

FUEL SCIENCE AND TECHNOLOGY INT'L., 13(8), 1039-1057 1995)

METHANOL-TO-GASOLINE YS. DME-TO-GASOLINEII. PROCESS COMPARISON AND ANALYSIS

Sunggyu Lee', Makarand Gogate, and Conrad 1. Kulik'

Process Research CenterDepartment of Chemical Engineering

The University of AkronAkron, Ohio 44325-3906

ABSTRACT

Methanol can be converted into gasoline boiling range hydrocarbons overzeolite ZSM-5 catalyst using the Mobil MTG process. Methanol feed in the MTGprocess can be derived from coal or natural gas based syngas. The Mobil MTGprocess involves the conversion steps of syngas-to-methanol and methanol-to­gasoline. Dimethyl Ether (DME), a product of methanol dehydrocondensation, isan intermediate species in the methanol-to-gasoline conversion.

Syngas can be directly converted to DME using the Liquid Phase DimethylEther Synthesis (LP-DME) process developed at the University of Akron inconjunction with Electric Power Research Institute. This direct one-stepconversion of syngas-to-DME can then be an ideal front end for further conversionto gasoline. This substitution (syngas-to-methanol by syngas-to-DME) is justifiedbecause DME results in an identical hydrocarbon distribution over the ZSM-5catalyst as methanol. The DME-to-Gasoline (DTG) process thus involves theconversion steps of syngas-to-DME and DME-to gasoline.

The UAIEPRI DTG process offers advantages over the Mobil MTG process inseveral areas. These include heat duty and heat of reaction, adiabatic temperaturerise, hydrocarbon product yield and selectivity, syngas conversion, and overallprocess efficiency. The conceptual benefits of the DTG process have been

*To whom all correspondence should be directedItCurrently with ANCON International, Newark, CA

1039

Copyright (C) 1995 by Marcel Dekker, Inc.

Dow

nloa

ded

by [

New

Yor

k U

nive

rsity

] at

06:

35 1

7 O

ctob

er 2

014

1040 LEE, GOGATE, AND KULIK

demonstrated experimentally in a fluidized bed reactor system operative at theUniversity of Akron. The salient features of the DTG process and processcomparison to the Mobil MTG process are discussed in this paper.

INTRODUCTION

The conversion of syngas to gasoline occurs typically in two stages. Syngas is

first converted to methanol over a copper-based hydrogenation catalyst. In the

second stage, methanol is converted to gasoline over a ZSM-S catalyst. These two

process steps form the basis for the Mobil Methanol-To-Gasoline (MTG) Process

(Chang and Silvestri, 1977; Chang, 1987; Chang and Silvestri, 1987; Meisel,

1988). Dimethyl Ether (DME) is a key intermediate chemical species in the

second stage. Mobil's MTG process in combination with the commercial syngas

to methanol technology thus provide a ready route to synthetic gasoline, i.e., with

feedstocks other than petroleum.

The first stage of the MTG process is the synthesis of methanol from syngas.

In a typical Liquid Phase Methanol Synthesis (LPMeoH™) process, the synthesis

catalyst (composed of CuD, ZnO, and AhO,) is slurried in an inert hydrocarbon

oil. Syngas (H2, CO, and CO2) reacts over the active catalyst to produce methanol

in-situ. The reaction chemistry for methanol synthesis in the liquid phase from

CO-rich syngas has been well established to be (Gogate, et al., I991a; Gogate,

1992, Lee, 1990; Lee and Parameswaran, 1991):

CO2 + 3H2 = CH,OH + H20CO + H20 = CO2 + H2

(1)(2)

Syngas-to-methanol conversion technology has recently been modified and

improved to synthesize Dimethyl Ether (DME) directly from syngas in a single

Dow

nloa

ded

by [

New

Yor

k U

nive

rsity

] at

06:

35 1

7 O

ctob

er 2

014

METHANOL VS. DME COVERSION TO GASOLINE. II 1041

reactor stage (Gogate, et aI., 1991a, 1991b; Gogate, et aI., 1992; Lee, et aI.,

1992). This process augments the per-pass syngas converison and volumetric

reactor productivity as a result of reduced chemical equilibrium limitation

governing the syngas-to-DME conversion. The reaction chemistry for typical

single-stage DME synthesis from CO-rich syngas can be written as (Gogate, 1992;

Lee, et aI., 1992a; Lee and Gogate, 1992; Lee, et aI., 1992b):

CO2 + 3H2 = CH30H + H20

CO + H20 = CO2 + H2

2CH30H = CH30CH3+ H20

(3)(4)(5)

In the Liquid Phase DME Synthesis (LP-DME) process, DME is thus directly

produced from syngas, in a single reactor stage. As seen from the reaction

sequence, the reaction scheme is a combination of an equilibrium limited reaction

(#3, methanol synthesis) and an equilibrium unlimited reaction (#5, DME

synthesis, or, methanol dehydration). Reactions #3 and #4 occur over the

coprecipitated CU/ZnO/Ah03 catalyst, while reaction #5 occurs over gamma-

alumina. The LP-DME synthesis is thus based on the application of dual catalysis

in the liquid phase (Lee, et aI., 1990; Gogate, et aI., 1992; Lee and Gogate, 1992).

This process can be an effective substitute for the syngas-to-methanol step in the

Mobil MTG Process. This substitution is justified based on the following facts (a)

DME results in virtually identical hydrocarbon product distribution as methanol

and (b) DME is a true intermediate in the Mobil MTG process. The single-stage

conversion of syngas to DME (and methanol) thus provides an ideal front-end for

further conversion to gasoline. A block diagram of the novel DTG process has

been shown in Figures Ia and lb.

Dow

nloa

ded

by [

New

Yor

k U

nive

rsity

] at

06:

35 1

7 O

ctob

er 2

014

a

1042

Steam

LEE, GaGATE, AND KULIK

WaJer Gasoline

b

Figure Ia. Syngas-to-Gasoline-Methanol Condensation-Low Pressure Gasoline Reactor

Water Gasoline

Figure lb. Syngas-to-Gasoline-DME Condensation-Low Pressure Gasoline Reactor

Dow

nloa

ded

by [

New

Yor

k U

nive

rsity

] at

06:

35 1

7 O

ctob

er 2

014

METHANOL VS. DME COVERS ION TO GASOLINE. II 1043

A novel syngas-to-gasoline process based on this unique integration scheme

(Syngas-to-DME and DME-to-Gasoline) has been conceptually demonstrated to

be potentially superior to the Mobil MTG process. This process has been termed

as the UA/EPRl Dimethyl Ether-To-Gasoline (DTG) process. The DTG process

merits over the Mobil MTG process are in the areas of heat duty, heat of reaction,

adiabatic temperature rise, space time and velocity, reactor size, hydrocarbon

product yield, and selectivity. These have been validated in the following sections.

RESULTS AND DISCUSSION

Heat of Reaction:

The conversion of methanol to hydrocarbons is an exothermic reaction.

Depending on the nature and spectrum of the hydrocarbon product, the heat of

reaction will vary. In the simple case when water is the only byproduct, the

conversion can be represented stoichiometrically as:

(6)

The heat of reaction of methanol conversion to hydrocarbons is a function of

the product distribution, or alternately, that of space velocity. Thus, to calculate

the heat of reaction of methanol (or, DME) conversion to hydrocarbons, the

product distribution spectrum at a fixed space time is necessary. These have been

summarized elsewhere (Chang and Silvestri, 1977; Chang, 1983). Based on these

calculations (at one sample space velocity of 108), the heat of reaction for DME

conversion to hydrocarbons (-142 cal/g) was found to be 32% less than that for

Dow

nloa

ded

by [

New

Yor

k U

nive

rsity

] at

06:

35 1

7 O

ctob

er 2

014

1044 LEE, GOGATE, AND KULIK

methanol conversion to hydrocarbons (-211 cal/g). Overall, since methanol

dehydration to DME contributes about 25% to the total heat of reaction of the

methanol conversion to hydrocarbons, it is expected that the heat of reaction for

DME conversion to hydrocarbons at complete conversion would be about 25%

lower than that for methanol conversion.

At complete and stoichiometric conversion, the heat of reaction for methanol

conversion to gasoline hydrocarbons is -398 cal/g at 371 C. The heat of reaction

for DME conversion would then be only about -296 cal/g, Based on the value of

-398 cal/g, the adiabatic temperature rise in the methanol-to-gasoline reactor is

estimated to be about 650 C. Based on the heat of reaction, the adiabatic

temperature rise levels are expected to follow the same trend. The adiabatic

temperature rise levels in the DME-to-gasoline reactor are thus expected to be

about 150 C lower than those for methanol-to-gasoline reactor.

Selectivity and Product yield:

The conversion of methanol to water and hydrocarbons is essentially

stoichiometric and complete. The stoichiometric representation of the reaction is:

(7)

where [CH2] is the average composition of the hydrocarbon product comprising of

paraffins and aromatics. Water is the only co-product. The final stoichiometric

composition for methanol conversion then would be [12+2 = 14] gat oms of [CH2]

and [2+16 = 18] gatoms of(H20 ]. The stoichiometric product ratio then becomes

14 g to 18 g. This translates to a 44:56 weight ratio basis of the hydrocarbons to

Dow

nloa

ded

by [

New

Yor

k U

nive

rsity

] at

06:

35 1

7 O

ctob

er 2

014

METHANOL VS. DME COVERSION TO GASOLINE. II 1045

water. The selectivity towards hydrocarbons for methanol conversion is then 0.44

g of hydrocarbons! g methanol.

For the conversion of DME to hydrocarbons and water, the stoichiometric

representation of the reaction is:

(8)

where [CH2.CH2] is the average composition of the hydrocarbon product. The

final stoichiometric composition for DME conversion would then be [12+2+12+2

= 28 gatoms] of hydrocarbons and [2 +16 = 18 gatoms] of water. This

stoichiometric ratio then becomes 28 g to 18 g. This translates to a product ratio

of60.8 : 39.2 weight basis ratio of hydrocarbons to water. The selectivity towards

hydrocarbons is then 0.61 g hydrocarbons! g DME. The reactor productivity

would then be 0.61 g/unit volume of the reactor, when compared to 0.44 g/ unit

volume for the methanol case, this is 38% higher for the DME conversion to

hydrocarbons.

Heat Duties:

The heat of reaction comparison between the MTG and DTG process has

already been illustrated and quantified in Section 1. This section considers the heat

duties. In this comparison, a fixed number of moles of CO-rich syngas is taken as

the basis and the following individual heat components have been considered:

For the MTG Process

(a) Sensible heat for feed syngas from 25 C to 250 C

(b) Heat of reaction for methanol synthesis at 250 C

Dow

nloa

ded

by [

New

Yor

k U

nive

rsity

] at

06:

35 1

7 O

ctob

er 2

014

1046 LEE, GOGATE, AND KULIK

(c) Sensible heat for methanol from 250 C to 315 C

(d) Heat of methanol dehydration reaction at 315 C

(e) Sensible heat of the equilibrium product mixture from 315 C to 375 C

(I) Heat of reaction for gasoline synthesis at 375 C

For the DTG Process

(g) Same as (a)

(h) Heat of DME synthesis reaction at 250 C

(i) Sensible heat ofmethanol-DME mixture from 250 C to 375 C

G) Heat of reaction for gasoline synthesis at 375 C

The exact computation of the individual components have been given elsewhere

(Gogate, et aI., 1992). Overall, on the basis of 2.68 gmol of CO-rich syngas

(H2:CO:C02:CH. = 37.4:46.3:7.7:8.6), total heat input for the above process steps

is 4886 cal for the MTG process. The total heat removal duty for the MTG

process is 12394 cal. The corresponding values for the DTG process are 5062 cal

and 17891 cal. It is thus shown that the heat input and removal duties for the

DTG process are higher than those for the MTG process. However, the yield of

hydrocarbons for the MTG process (based on 50 g CO-rich syngas feed to the first

step) is only 4.8 g compared to the DTG process at 7.5 g. Thus the product yield

is substantially higher for the DTG process. Therefore, the heat input and removal

quantities per g hydrocarbon basis are substantially lower for the DTG process.

The novel DTG process has the dual benefit of improving the hydrocarbon yield

per unit feed of syngas basis and simultaneously lowering the heat duties per g

hydrocarbon basis.

Dow

nloa

ded

by [

New

Yor

k U

nive

rsity

] at

06:

35 1

7 O

ctob

er 2

014

METHANOL VS. DME COVERSION TO GASOLINE. II 1047

Space Velocity and Reactor Size:

The conversion of methanol and DME to hydrocarbons is a function of space

velocity (or, space time in the reactor). The reaction paths for methanol and DME

conversion (Chang, 1983) bring out this fact in greater detail. A substantially more

space time (or, contact time) is needed for complete and stoichiometric conversion

of methanol to hydrocarbons when compared to that needed for DME conversion

to hydrocarbons. The relative difference between the space times for these two

cases is estimated at 1 magnitude of space time (I/[g reactantlg cat. h]). Based on

the reaction paths given in the literature (Chang and Silvestri, 1977), it is estimated

that for identical space time, the DTG reactor size can be about 25% smaller than

the size of the MTG reactor, for complete and stoichiometric conversion.

Conversely, for identical MTG and DTG reactor size, the DTG reactor will be

25% more productive (i.e., reactor productivity per unit volume) than the MTG

reactor.

Overall Syngas Efficiency:

In computing the overall syngas efficiency, a feed of 2.68 gmol of syngas is

taken as the basis. The syngas composition is H2:CO:C02:CH4

37.4:36.3:7.7:8.6. In the MTG process, syngas is first converted to methanol.

The experimental conditions for this step are 250 C, 70 atm, 1500 RPM stirrer

speed, 150 g methanol synthesis catalyst in 500mL of Witco-40 oil. The synthesis

is carried out in a one-liter stirred autoclave. The experimental conditions for

single-step DME synthesis are the same, except 109 of DME catalyst (gamma­

alumina) are added to the slurry. For the MTG conversion, the conversion of

Dow

nloa

ded

by [

New

Yor

k U

nive

rsity

] at

06:

35 1

7 O

ctob

er 2

014

1048 LEE, GOGATE, AND KULIK

methanol feed from the first step is asumed to be complete and stoichiometric. For

the DTG conversion, the conversion of DME is assumed to be complete and

stoichiometric. The product hydrocarbon to water ratio for complete and

stoichiometric conversion is outlined earlier in the section on Selectivity and

Product Yield.

Syngas-to-Methanol

Feed 2.68 gmol (=50.1 g) CO-rich Syngas

Product 0.34 gmol (=10.9 g) methanol + 0002 gmol (=0.036 g) water

Syngas (H2 + CO) conversion: 53.3 % (on gmol basis)

Methanol-to-Gasoline

Feed

Product

Syngas-to-DME

Feed

Product

10.9 g methanol

4.8 g [CH2) + 6.1 g [H20 )

2.68 gmol (=50.1 g) CO-rich Syngas

0.04 gmol (=1.28 g) methanol + 0.25 gmol (= 11.5 g) DME +

trace water

Syngas (H2 + CO) conversion: 73.3 % (on gmol basis)

DME-to-Gasoline

Feed

Product

1.21 g methanol + 11.5 g DME

7.54 g [CH2) + 5.17 g [H20 )

Percent Increase in [CH2) Yield for Syngas-to-DME-to-Gasoline

= [(7.54-4.8)/4.8) x 100 = 56 % (w/w)

Dow

nloa

ded

by [

New

Yor

k U

nive

rsity

] at

06:

35 1

7 O

ctob

er 2

014

METHANOL VS. DME COVERSION TO GASOLINE. II 1049

Overall Comparison and Analysis:

Earlier sections have elucidated the rationale and the philosophy behind

operating the novel Syngas-to-DME-to-Gasoline process, as opposed to the

conventional Syngas-to-Methanol-to-Gasoline process. This section summarizes

advantages of the UNEPRI DTG process over the MTG process. These are

summarized in Table 1. Comparative mass and energy balances for DTG process

from coal and natural gas are shown in Figures 2 and 3.

[1] One-step conversion of syngas to DME improves the per-pass conversion

and reactor productivity over syngas to methanol. The comparison of our

experimental results are as follows. At 250 C, 70 atm, I-liter slurry reactor, per­

pass syngas conversion to methanol is 53%. At nominally identical operating

conditions, per-pass syngas conversion to DME is 53%. Starting with I gmol of

syngas, this translates to about 0.47 gmol of recycle / gmol of syngas for the

syngas to methanol case. The recycle load is reduced to only 0.26 gmol / gmol of

syngas for the syngas to DME case. The recycle load is thus reduced by about

46%.

[2] The conversion of syngas to methanol uses copper-based Cu/ZnO/AhO)

catalyst. This catalyst is susceptible to deactivation by crystal growth in the liquid

phase rich in methanol and water. However, the conversion of syngas to DME

uses a dual catalyst system based on a combination of CU/ZnO/Al20 , catalyst and

gamma-alumina catalyst. This conversion offers a favorable scenario for crystal

growth of copper catalyst because of two reasons. First, the liquid phase is lean in

methanol because of in-situ conversion to DME over gamma-alumina. Second,

Dow

nloa

ded

by [

New

Yor

k U

nive

rsity

] at

06:

35 1

7 O

ctob

er 2

014

1050 LEE, GOGATE, AND KULIK

TABLE I

Syngas-to-Methanol-to-GasolineVs.

Syngas-to-DME-to-Gasoline

Syngas-to-methanolvs

Syngas-to-DME

MTG DTG

1.2.3.4.5.6.7.

productivity, g product'hydrogen conversion, %CO conversion, %syngas conversion, %recycle, gmol / gmol feed syngasheat of reaction, cal (exothermic)catalyst life

10.972.438.253.00.478008good

11.887.762.373.40.2713745better

'On basis of experimental data, 2.68 mol (=50 g) syngas feed, 250 C, 70 atm.

methanol-to-gasolinevs

DME-to-gasoline

8.9.10.II.12.13.14.15.16.17.18.19.

productivity, g [CH 2) '

conversion, %heat of reaction, cal / g (exothermic)adiabatic temperature rise, Creactor size at identical WHSVcatalyst lifemethanol dehydration reactorwater concentration in reactoroverall water production, g [H20)

[CH2) selectivity, g /g converted[H20) selectivity, g / g convertedoverall hydrogen consumption, g [CH 2] / g

4.791003986001.25poornecessaryhigh6.160.440.563.42

7.46100300475Ibetternot neclow5.150.610.394.31

'On basis of complete and stoichiometric conversion, 375 C, I atm, feed item (I).

Dow

nloa

ded

by [

New

Yor

k U

nive

rsity

] at

06:

35 1

7 O

ctob

er 2

014

Syngas-to-DME-to-Gasoline-Coal ~

~>ZoI'""'

<en

e~rn

~rnGl<3zdo>enoI'""'ZtTl

::::

[llzOlj 0.21!M; gmol_ I<> SbJJ\

AM' (Prodoct CooIlDaIo~"'::)

= -14.860 cal

. LAM" (Jleaolor C4oIln«J= -3914 cal

+ 0.2111 lIllol HtAHO(26"c-25o"C)

= :lf13 oaI

Cp'l,... ) 0 11.8 GaI/'J<AH' (Cool!nl I<> eoodalse 1leOR) • .al oaI 1C.'l!nI lI)AR' lCool!nll<> Coo4e>oo DlIi) • 2lmi oaI c.ol!nl Dj

rCO2 0.5f3l

L~o O.I47J-rCoollDa I<> a.m... co. 0 2412 eo!C001lD& I<> RemoT. Hz!! • 1726 eo!

112 0.264 smoIt'n d ~44 smol_

B2 0.717 lIllolco o.m lIllolCO2 e.es smoI

[B2 42.n~CO 55.0%CO2 2..~

-lip

~.

B2 O.n. lIllolell 0.:13. lIllolco.. 0.100 lIllol

Figure 2. Overall Mass and Energy Balance for Syngas-to-Gasoline, Coal-Based Syngas~

Dow

nloa

ded

by [

New

Yor

k U

nive

rsity

] at

06:

35 1

7 O

ctob

er 2

014

Syngas-to-DME-to- Gasoline-NG ~...,

HZ 0.883 gmoI~)CO 0.124 gmoI IU)CO2 0.122 gmoI 10.8")

~1+6BO • -wi cal

Figure 3. Overall Mass and Energy Balance for Syngas-to-Gasoline, NG-Based Syngas

rmmooo~m

>zo;>;c:r;;;

------.l .:iG -:. I 5.2OZ S ('llU5210 s D!20]

!mO~ 000lIDil= -lZ06O cal

6BOCH30B 0..201 smoI ~ (M+n) 14 aoo°C) I n.l+tll IH2O 0.006 smoI • 2788 c.aI

lIlIJ! 0.082 smol j1 d (lleodor Coobil

Q -3'78ll calCpO (....) • 12.8 001/"1:

6 B" ICooll"I 14 Con4cw 1de01I) = Z204 oeI (Coolbli IjJ6 SO Coollnc '" e-Ic:sloo Il!IlE) • 113 oaI (Coolbli nl

I Pn41>o\~ Ba 0.883 smoI

CO 0.1.24 &molCO2 0.1Zl smol

6 II' (Z5"1:- 260" C)• 382'7 oaI

tor IL>otlnc i'lled SyDps

H! 0.883 gmoICO 0.274 rmol

COz 0.060 gmoI 1~_I j

.E:!ll!t.- nlllE,H2 I.m smoJ·co 0.411 &molCO 2 UII smoJ

[~ ~~lCO! 6~

11oB"~ CooIlD&l

• -65U oaI

Ilecydo

hlaIc....l1Jl

H2 0.882 &mol Imlco G.l!1"1 &mol l1X)COz 0.061 &mol n)

Dow

nloa

ded

by [

New

Yor

k U

nive

rsity

] at

06:

35 1

7 O

ctob

er 2

014

METHANOL VS. DME COVERSION TO GASOLINE. II 1053

water produced by both methanol synthesis (C02 hydrogenation) and DME

synthesis (methanol dehydration) constantly is shifted by the forward shift reaction.

This is perhaps the most significant merit of the one-step conversion of syngas to

DME, a fact well-proven by the research at the University of Akron. The reaction

chemistry of syngas to DME conversion is as follows, starting with CO-rich or the

so-called unbalanced syngas:

CO2 + 3 H2

CO + H20

2 CHJOH

= CH,OH+H20

CO2 + H2

= CHJOCHJ + H20

(9)(10)(II)

[3] The one-step conversion of syngas to DME improves the volumetric

reactor productivity by as much as 100%, over that of syngas to methanol

conversion. This is because of the fact that conversion of syngas to DME is not

nearly limited by chemical equilibrium as syngas to methanol. This fact has been

well-proven by the research at the University of Akron. For example, at 250 C

and 70 atm, starting with I gmol of unbalanced syngas, methanol yield is about

0.14 gmol, or, 0.14 gmol (CHJ) equivalent. At identical conditions, the yield of

DME is about 0.12 gmol, or, 0.24 gmol of (CH J) equivalent. This corresponds to

an increase of about 72% in the reactor productivity, at identical conditions.

[4] The conversion of methanol to gasoline is highly exothermic with a heat of

reaction of 398 callg of methanol converted. For this reason, the conversion is

usually split in two parts, for exothermic heat management. In the first, methanol

is first converted to an equilibrium mixture of DME, water (and unconverted

methanol). In the second, this mixture of methanol, DME, and water is converted

Dow

nloa

ded

by [

New

Yor

k U

nive

rsity

] at

06:

35 1

7 O

ctob

er 2

014

1054 LEE, GOGATE, AND KULIK

to gasoline. The first step contributes to about 25% of the total exothermic heat of

reaction. The conversion of DME to gasoline is thus about 25% less exothermic,

with a heat of reaction of about 300 call g ofDME converted.

[5] As elucidated in point [4], the conversion of methanol to gasoline requires

the conversion of methanol to methanol, DME, and water first, to manage the

exothermic heat of reaction. The one-step conversion of syngas-to-DME followed

by DME conversion to gasoline thus obviates the need for a separate methanol

dehydration reactor. This contributes to significant savings in capital and

operating cost.

[6] The theoretical adiabatic temperature rise in the conversion of methanol to

gasoline is estimated at about 650 C, while that for the conversion of DME to

gasoline is only at about 475 C.

[7] In methanol to gasoline conversion, equilibrium mixture of DME, water,

and methanol is usually fed to the gasoline synthesis reactor. The ZSM-5 based

zeolite catalyst is very sensitive to water in the feed (water is a catalyst poison).

For this reason, water is removed from the feed. In the conversion of DME to

gasoline, very little or no water is fed to the gasoline synthesis reactor, because the

single-step syngas-to-DME conversion produces DME with extremely high

selectivity (>99%). Catalyst life cycle in the gasoline synthesis reactor is favorable

in the direct conversion of DME-to-gasoline because of (a) lower concentrations

of water in the reactor and (b) about 25% reduced exothermicity of the reaction.

Dow

nloa

ded

by [

New

Yor

k U

nive

rsity

] at

06:

35 1

7 O

ctob

er 2

014

METHANOL VS. DME COVERSION TO GASOLINE. II 1055

[8] The stoichiometric conversion of methanol to gasoline produces ,'4 %

(w/w) to 56 % (w/w) mixture of hydrocarbons to water. For example, 1 :; of

methanol yields 0.44 g hydrocarbons [CH2] and 0.56 g water, at complete and

stoichiometric conversion. On the contrary, 1 g DME yields 0.61 g hydrocarbons

[CH2] and only 0.39 g water. The yield of gasoline hydrocarbons for DME

conversion is thus 38 % higher than that for methanol conversion. The yield of

water is concomitantly 30 % lower for the DME conversion case.

[9] Hydrocarbon product distribution is directly related to the space time for

both methanol and DME conversion. For methanol conversion, the reactor space

time is estimated to be 20% higher than that for DME conversion, to achieve a

stable hydrocarbon product distribution. Intuitively also, this can be immediately

validated considering that reaction path of methanol conversion to hydrocarbons

involves DME as the intermediate. At identical space time, the DTG reactor size

can be 25% lower than the conventional MTG reactor.

[10] Finally, there are important differences between two cases of syngas-to­

DME-to-gasoline process, where syngas is derived from coal or from natural gas.

Syngas derived from coal is normally CO-rich (or, unbalanced), while that derived

from natural gas is Hj-rich (or, stoichiometric). These differences are in the areas

of:

(a) recycle load (about 0.34 gmol / gmol for coal vs 0.52 for NG case).

(b) CO2 rejection may be necessary in coal-based syngas. CO addition is

necessary for NG.

Dow

nloa

ded

by [

New

Yor

k U

nive

rsity

] at

06:

35 1

7 O

ctob

er 2

014

1056 LEE, GOGATE, AND KULIK

(c) overall [CH21productivity per pass is about 30% higher for coal-based

syngas.

(d) overall, coal-based syngas process is more attractive. However, the

availabilityand price of coal in comparison to NG remains a primary

concern.

CONCLUSIONS

Process comparison and analysis of Syngas-to-Methanol-to-Gasoline and

Syngas-to-OME-to-Gasoline proves that synthesis of gasoline via direct OME

route has definitive process advantages over the synthesis via methanol route.

These process merits are in the areas of higher gasoline yield, higher syngas

conversion, good adaptability to coal-based syngas, and integrated energy

efficiency. Further experimental investigation to establish these process merits is

currently underway at the University of Akron.

ACKNOWLEDGEMENTS

This work was supported by the Electric Power Research Institure, via the

research contract RP 317-6.

REFERENCES

Chang, CO., and Silvestri, A.I., "The Conversion of Methanol and other 0­Compounds to Hydrocarbons over Zeolite Catalysts", I. Catal., 47, 249-259,1977.

Chang, CO., "Hydrocarbons from Methanol", Cat. Rev.-Sci. Eng., 2S (I), 1-118,1983.

Chang, CD., and Silvestri, A.I., "MTG Origin, Evolution, Operation,CHEMTECH, 10, 624-631, 1987.

Dow

nloa

ded

by [

New

Yor

k U

nive

rsity

] at

06:

35 1

7 O

ctob

er 2

014

METHANOL VS. DME COVERS[ON TO GASOLINE. II 1057

Gogate, M.R., Lee,S., and Kulik, C.1., "A Single-Stage, Liquid-Phase DimethylEther (DME) Synthesis Process from Syngas I. Dual Catalytic Activity andProcess Feasibility", Fuel Sci. Tech. [n1'l, 9(6), 653-679, [99Ia.

Gogate, M.R., Lee, S., and Kulik, C.1., "A Single-Stage, Liquid-Phase DimethylEther (DME) Synthesis Process from Syngas II. Comparison of Per-pass SyngasConversion, Reactor Productivity, and Hydrogenation Extent", Fuel Sci. Tech.Int'l, 9(7), 889-912, 1991b.

Gogate, M.R., Lee, S., and Kulik, C.J., "Synthesis of Hydrocarbons Via Single­Stage DME Route", Paper Presented at the AIChE Spring National Meeting, NewOrleans, Louisiana, March 29 - April 2, 1992.

Gogate, M.R., "A Novel Single-Step Dimethyl Ether (DME) Synthesis fromSyngas", Ph.D. Dissertation, The University of Akron, Akron, Ohio, 1992.

Lee, S., Methanol Synthesis Technology, CRC Press, Boca Raton, Florida, 1990.

Lee, S., Gogate, M.R., Lee, B.G., and Kulik, C.J., "Kinetics of Liquid PhaseCatalytic Dehydration of Methanol to Dimethyl Ether", Fuel Sci. Tech. Int'I, 8(6),637-671, 1990.

Lee, S., and Parameswaran, V.R., "Reaction Mechanisms in Liquid PhaseMethanol Synthesis Process", EPRI Electric Power Research Institute, Palo Alto,California, 1991.

Lee, S., Gogate, M.R., and Kulik, C.1., "A Novel Single-Stage Dimethyl Ether(DME) Synthesis Process over Dual Catalyst System from CO-rich Syngas",Chern. Eng. Sci., 47, 3769-3776, 1992a.

Lee, S., and Gogate, M.R., "Dimethyl Ether (DME) Synthesis Process", FinalReport, Electric Power Research Institute, Palo Alto, California, 1992.

Lee, S., Gogate, M.R., and Vijayaraghavan, P., "Modeling of Methanol and DMEProduction from Syngas", Proceedings: Sixteenth Annual EPRI Conference onFuel Science, Electric Power Research Institute, Palo Alto, California, In Press,1992b.

Meisel, S.L., "Catalytic Research Bears Fruit", CHEMTECH, 1,32-37, 1988.

RECEIVED: November 10, 1994

ACCEPTED: December 1, 1994

Dow

nloa

ded

by [

New

Yor

k U

nive

rsity

] at

06:

35 1

7 O

ctob

er 2

014