JOURNAL OF CATALYSIS 50, 61--66 (1979)

Kinetics and Mechanism of Methanol Carbonylation

over Rh-Y Zeolite

NOBUO TAKAHASHI~ YUICHI 0RIKASA, AND TATSUAKI YASIIIMA ~

Department of Chemistry, Tokyo Institute of Technology, Ookayama, Meguro-ku, Tokyo 152, Japan
Received July 27, 1978; revised January 23, 1979

Methanol carbonylation over Rh-Y zeolite was studied kinetically at 428 to 473 K under
atmospheric pressure. For low conversions of methanol, methyl acetate was the only carbonyla-
tion product observed. The reaction was first order in methyl iodide and zero order in both
methanol and carbon monoxide. The apparent activation energy was 5.65 X 104 J/tool in the
temperature range studied. Carbonylations in the ethanol-methyl iodide, methanol-ethyl
iodide, and deuterated methanol-methyl iodide systems were studied to clarify the reaction
mechanism. From these results, the reaction mechanism for this carbonylation is proposed as
follows: (i) the oxidative addition of methyl iodide to an active Rh site, (ii) the insertion of
carbon monoxide into the methyl-Rh bond, (iii) the formation of methyl acetate by methan-
olysis of the acetyl-Rh bond, (iv) the reduetive dissociation of hydrogen iodide to reform the
active site, and (v) the regeneration of methyl iodide by the reaction of hydrogen iodide with
methanol. This reaction mechanism is quite similar to that proposed for the homogeneous
rhodium complex catalyst.

INTRODUCTION over R h - X zeolite catalyst (4-6). These

results resemble that for the homogeneous
In our previous paper (1), we reported rhodium complex catalyst. So, it is expected
that 1%h-Y zeolite prepared by an ion ex- that the behavior of the R h - Y catalyst is
change procedure shows a high catalytic similar to that of the homogeneous rhodium
activity for the carbonylation of methanol complex catalyst.
in the presence of methyl iodide. In a con- A reaction mechanism which includes an
tinuous flow system , the R h - Y catalyst oxidative addition of methyl iodide to an
maintains constant activity for more than active monovalent rhodium species has been
10 hr under atmospheric pressure. For the proposed for the homogeneous rhodium
carbonylation, the specific activity of R h - Y complex catalysts (7-9). The methyl group,
exceeds that of other heterogeneous cata- from which the acetyl group is formed
lysts, such as charcoal-supported rhodium during carbonylation, is believed to be
chloride (2) and polymer-bound rhodium
supplied by the methyl iodide. This con-
complex (8). It has been found that methyl
clusion is based on kinetic results, i.e., the
acetate is an initial product of the carbo-
rate of carbonylation is proportional to the
nylation, but it is not observed without
concentration of methyl iodide and inde-
methyl iodide (1). The same phenomena
pendent of those of both carbon monoxide
have been observed in this carbonylation
and methanol. However, any further evi-
1To whom correspondence should be addressed. dence to support this mechanism has not

61

0021-9517/79/100061-06502.00/0
Copyright ~ 1979 by Academic Press, Inc.
AII rights of reproduction in any form reserved.
62 TAKAHASHI, ORIKASA, AND YASHIMA

to p H 5.3 b y the addition of sodium hy-

l II A droxide. The rhodium content of the cata-
lyst, which was determined b y atomic
absorption spectrometry, was 4.2 X 10 - 5
g-atom/g-eat (4.4% of degree of ion ex-
"~2.0 change as Rh3+). No cation exchanged ex-
% cept Na and R h could be detected, b u t a
[0 o ~ ] o _ ~ _
small a m o u n t of proton would be present in
our catalyst. F r o m the measurement of
0 0.5 1.0 'o-3 PMeI(Pa)
L I I X - r a y diffraction it was estimated t h a t
0 2.5 5,0 Io-~Pco (Po) small portion of the R h - Y framework was
i I _ I
0 1.0 2.0 IO-~MeOH(Po) decomposed.
Methanol carbonylation was carried out
Fie-. 1. Effect of partial pressure of raw materials in a fixed-bed type apparatus with a con-
on methyl acetate formation rate. O, methyl tinuous flow system at atmospheric pres-
iodide (a) ; O, carbon monoxide (b) ; @, methanol
(c). Reaction conditions: catalyst, Rh-Y; tempera- sure. Details of the experimental procedure
ture, 443 K; W/F, 5.8 X 10~ g-see/mol; PM~I, (b) were described in the previous paper (1).
1.19 X l0 s Pa, (e) 1.03 X 10a Pa; Peo, (a) 3.07 The partial pressure of each reactant was
X 104 Pa, (e) 3.30 X 104 Pa; P~om (a) 1.19 X 104 changed while the total pressure was main-
Pa, (b) 1.19 X 104 Pa. tained at atmospheric pressure b y the addi-
tion of helium.
been found. During this investigation, the The earbonylation of deuterated metha-
carbonylation of methanol was kinetically nol was carried out in a glass reactor with
studied to obtain some information about a recirculating system (volume: 280 ml)
the catalytic behavior of R h - Y . Further- connected to a v a c u u m line. R h - Y (1.0 g)
more, to clarify the role of methyl iodide, was placed in the reactor and evacuated
the earbonylation of deuterated methanol at room temperature for 2 hr. After the
(CD~OD) was carried out and the dis- catalyst bed was heated to reaction tem-
tribution of deuterated methyl groups in perature b y an electric heater, deuterated
the product was determined. A mechanism methanol, m e t h y l iodide, and carbon mon-
for the carbonylation of methanol over oxide were introduced into the reaction
R h - Y zeolite is discussed based on these system to start the carbonylation. A small
results. portion of the reaction mixture, except
carbon monoxide, was collected in a trap
EXPERIMENTAL
cooled with liquid nitrogen.
Rhodium trichloride t r i h y d r a t e was ob- The reaction mixture was analyzed b y
tained from Nippon Engelhard Co. Metha- a gas chromatograph to determine the
nol and methyl iodide, obtained from Wako composition, and was also analyzed b y a
Pure Chemical Ind. Ltd., were used with- mass fragment chromatograph (Shimazu
out any further purification, and carbon L K B 9000 gas chromatograph-mass spee-
monoxide was obtained from Takachiho trometer) to determine the distribution of
Chemical Ind. Co. D e u t e r a t e d methanol deuterated methyl groups in m e t h y l ace-
(CD3OD) was obtained from Merck Japan tate and methyl iodide.
Co.
RESULTS AND DISCUSSION
Rhodium Y zeolite catalyst was pre-
pared b y a conventional cation exchange R h - Y is remarkably active for methanol
m e t h o d at 353 K using NaY (SK-40) and a carbonylation, and its specific activity is
rhodium chloride aqueous solution adjusted almost the same as that of the homo-
 METHANOL CARBONYLATION OVER Rh-Y 63

geneous rhodium complex catalyst. At less

-13..~
than 30% conversion of methanol, methyl
acetate is the only carbonylation product, g
with dimethyl ether formed as byproduct. ~' -14.(
g
It has been found that on R h - Y catalyst,
no methanol carbonylation occurs in the E -14.5
absence of methyl iodide, and almost all of <

the methyl iodide (over 95%) is recoverable

-15,0
from the reaction mixture.
The effect of the methyl iodide partial
pressure on the steady state formation rate -i5,5 - I
2.00 2.25 2.50
of methyl acetate (rMA) is shown in Fig. 1.
IO ~ I/ T ( K -I)
The rate is proportional to the partial
pressure of methyl iodide. The effects of FIG. 2. Effect of reaction temperature on methyl
the partial pressures of carbon monoxide acetate formation rate. Reaction conditions: cata-
and methanol on r~A are shown in Fig. 1. lyst, R h - Y ; W / F , 5.8 X l0 s g.sec/mol; PieI, 1.03
The formation rate is approximately zero X 10s P a ; Pco, 3.07 X 104 P a ; P~eog, 1.19 X 104
lma.
order with respect to the partial pressures
of both carbon monoxide and methanol.
The apparent activation energy for the proceed in the condensed liquid phase, be-
carbonylation in the reaction temperature cause it could be expected that a capillary
range from 428 to 473 K is 5.65 X 104 J/tool condensation of water and/or methanol in
from Fig. 2. the zeolite pore does not occur at such
From these results, the rate expression higher reaction temperature than their
and the apparent activation energy over boiling points, and sufficiently lower partial
Rh-Y are represented as Eq. (1). These pressures than their vapor pressures.
are quite similar to those for the homo- The following reaction mechanism has
been proposed for the homogeneous rho-
geneous rhodium complex catalyst [Eq.
dium complex catalyst (7-9).
(2)] (9).
d-CH~I I
Rh-Y : RhL~ ) CH3-RhLn

rMA = kPM~I1Pco °PMeoH °,

E -- 5.65 X 104 J/tool
(1)
I
I -HI

d-CH~OH I
+CO

H-RhLn ~- CH~CO-RhL~
Homogeneous rhodium complex: (A) --CH3COOCH3
HI ÷ CH3OH -~ CH3I + H20 (3)
r = k'[MeI][CO]°EMeOI-I] o,
(2)
E = 6.15 X 10 4 J / t o o l First, the oxidative addition of methyl
iodide to the active rhodium species occurs.
From a comparison of Eqs. (1) and (2), the After that, there is the successive insertion
active rhodium sites on the Rh-Y appear of carbon monoxide to C H r R h bond, the
to be quite similar to the active rhodium elimination of methyl acetate by methanol-
species in the homogeneous rhodium com- ysis, the reductive dissociation of HI from
plex catalyst, and the reaction mechanisms the rhodium species (A), and the regenera-
of both systems are the same. However, tion of methyl iodide by Eq. (3). The first
we believe that this reaction does not step, the oxidative addition of methyl
64 TAKAHASHI, ORIKASA, AND Y A S H I M A

TABLE 1
Carbonylations in the Ethanol-Methyliodide and Methanol-Ethyliodide Systems ~

Reactants W/F Conversion Percentage Formation rate

(104g-see/ of ROH of total (10.7 mol/sec.g-eat)
ROIt R'I tool) (mol%) R'I (mol%)
CHACO- C2H5CO-
CH~I C~HsI
rMA 5 rEAC TMpd rpA e

C~HsOH CHaI 0.5 6.0 94.8 5.2 Trace 3.5 N.D.S N.D.
C2HsOH CH3I 3.2 45.2 68.5 31.5 Trace 9.9 N.D. N.D.
CH~OH C2HsI 3.2 54.7 32.2 67.8 4.3 0.5 1.7 1.9

Reaction conditions: catalyst, Rh Y; reactant molar ratio, C O / R O H / R ' I = 8/2/1; temperature,

443 K.
b Methylacetate.
Ethylacetate.
Methylpropionate.
Propionic acid.
I N.D., not detected.

iodide, is believed to be the rate-determi- was also produced. Moreover, the forma-
ning step. The above mechanism is sup- tion rate of methyl acetate was the fastest
ported by the observation of a first order among the carbonylations. It is suggested
dependence of the earbonylation rate on that an exchange reaction between alkyl
the methyl iodide concentration. According groups occurs, and the methyl group reacts
to this reaction mechanism, the methyl more readily with carbon monoxide than
group in the aeetyl group must be derived the ethyl group. Therefore, the proposed
from methyl iodide. mechanism could not be confirmed by
In order to confirm this mechanism, these experiments only.
earbonylations of ethanol-methyl iodide In the ease of the deuterated methanol
and methanol-ethyl iodide systems were (CD3OD)-methyl iodide system, it is ex-
studied. If the rate of the alkyl group ex- pected that the difference of reactivity
change reaction between alcohol and iodine between methyl and deuterated methyl
compounds is slow, ethyl acetate or methyl groups is small, very much in contrast
propionate must be a major earbonylation with that between methyl and ethyl
product. The results are summarized in groups. So, it should be possible to distin-
Table 1. In the ease of the ethanol-methyl guish whether methyl iodide carbonylation
iodide system, the major earbonylation or methanol direct earbonylation occurs
product is ethyl acetate, and neither methyl over Rh-Y catalyst, when the distribution
propionate nor propionie acid is formed. of deuterated methyl groups is measured in
This result supports the proposed mecha- the methyl acetate produced from the
nism in which the alkyl group forming the CDaOD-CH3I system. But, this requires
acyl group is supplied by the iodine com- that the rate of the methyl group exchange
pound. However, in the ease of the metha- reaction between CD3OD and CHaI is not
nol-ethyl iodide system, not only propionie too fast as compared with the earbonyla-
acid and methyl propionate, which were tion. When methyl iodide is carbonylated,
formed from the earbonylation of the ethyl CHaCOOCDa must be produced, and
group were produced, but methyl acetate when methanol is directly carbonylated,
 METHANOL CARBONYLATION OVER Rh-Y 65

CDaCOOCDa must be produced. The re- The dotted line in Fig. 3 shows the
sults of carbonylation in the CD3OD-CH3I calculated value of CH3CO-/(CH3CO-
system are shown in Fig. 3. The fraction of 4- CD3CO-) which is based on the fraction
CH3I in the total methyl iodide decreased of CH3I and following assumptions, i.e.,
with increasing methanol conversion. This there is no isotopic effect, the aeetyl group
result shows that the alkyl group exchange is made only from methyl iodide, and there
reaction does not reach the equilibrium is no exchange of methyl groups between
state during this run. Therefore, an analysis the acetyl group and other compounds. As
of the deuterated methyl group distribu- shown in the Fig. 3, the experimental
tion in the methyl acetate produced could values agree very well with the calculated
provide some useful information with re- values. Moreover, the change in the fraction
gard to the reaction path. Mass numbers of CI-I3CO- with methanol conversion con-
43 and 46 correspond to CH3CO- and sistently agrees with that in the fraction
CDaCO-, and mass numbers 74, 77, and 80 of mass number 77 (MN77), (MN77)/
correspond to CHaCOOCH3, CDaCOOCH~ [(MN74) -t- (MN77) 4- (MN80)]. This
or CH3COOCDa, and CDaCOOCDa, respec- result shows that almost all of the mass
tively. If methanol were directly carbon- number 77 product is CH3COOCD3. It is
ylated, the fraction of CH3CO-, CHACO-/ clear that neither inter- nor intramoleeular
(CH3CO- + CD~CO-) would increase with exchange reaction of methyl groups of
methanol conversion, because the fraction methyl acetate occurs.
of CKaOH in the total methanol should It is concluded that on Rh-Y catalyst,
increase with reaction time by the ex- methanol is not directly earbonylated, but
change reaction of methyl groups between methyl iodide is carbonylated. Namely,
methyl iodide and methanol. However, the methyl acetate formation over Rh-Y pro-
fraction of CHACO- in the acetyl groups eeeds through the following paths which
decreased with methanol conversion. are the same as those proposed for the
homogeneous rhodium complex catalyst:
(i) the oxidative addition of methyl iodide
1.0 to the active rhodium site, (ii) the insertion
of carbon monoxide into the methyl-Rh
bond, (iii) the formation of methyl acetate
by methanolysis of the acetyl-Rh bond,
0", o (iv) the reductive dissociation of HI from
.o 0.5 (i',,..o the rhodium species (A) to reform the
u_
active rhodium site, and (v) the regenera-
tion of methyl iodide by the reaction of HI
with CHsOH.
r
25 50 ACKNOWLEDGMENT
Conversion of Methanol (%)
We wishto thank ShimazuSeisakushoLtd. for the
FIG. 3. Changes in the fractions: CH3I/ cooperation concerned with measurements by a
(CH3I 4- CD~I), CH3CO-/(CH~CO-+ CD~CO-), gas chromatograph-massspectrometer.
and MN(77)/(MN(74) q- MN(77) + MN(80)),
with methanolconversion. O, CHsI/(CH3I+CD3I) ; REFERENCES
e, CH3CO-/(CHaCO- + CD3CO-); (}, MN (77)/
(MN (74) 4- MN(77) + MN (80)). Reaction condi- 1. Yashima, T., Orikasa, Y., Takahashi, N., and
tions: catalyst, Rh-Y; temperature, 443 K; initial Hara, N., J. Catal. 5g, 53 (1979).
pressure of CD~OD, 1.1 X 105 Pa; CO~ 2.6 X 104 2. Robinson,K. K., Hershman, A., Craddock,J. H.,
Pa; CH3I, 1.1 X 103 Pa. and Roth, J. F., J, Catal. 27, 389 (1972).
66 TAKAHASHI, ORIKASA, AND YASHIMA

3. Jarrel, M. S., and Gates, B. C., J. Catal. 40, 255 6. Nefedov, B. K., Sergeeva, N. S., and Krasnova,
(1975). L. L., Izv. Akad. Nauk, SSSR, Ser. Khim. 614
4[. Nefedov, B. K., Sergeeva, N. S., Zueva, T. V., (1977).
Schutkina, E. M., and Eidus, Ya. T., Izv. 7. Paulik, F. E., and Roth, J. F., Chem. Commun.
Akad. Nauk, SSSR, Ser. Khim. 582 (1976). 1578 (1968).
5. Nefedov, B. K., Sergeeva, N. S., and Eidus, Ya. 8. Forster, D., J. Amer. Chem. Soc. 98, 846 (1976).
T., Izv. Akad. Nauk, SSSR, Ser. Khim. 2271 9. Hjortkjaer, J., and Jensen, V. W., Ind. Eng.
(1976). Chem. Prod. Res. Dev. 15, 46 (1976).