CaCl2-catalyzed functionalization of saturated hydrocarbons with CO to carboxylic acids and esters

Article abstract of DOI:10.1006/jcat.2000.2972

The functionalization of saturated hydrocarbons, e.g., ethane, cyclopentane, propane, cyclohexane, cycloheptane, and cyclooctane, with CO by CaCl₂ catalyst in the presence of CF₃COOH and K₂S₂O₈ was studied. Carboxylic acids and alkyl trifluoroacetates were formed as dominant products and by-products, respectively. In the reaction of propane, isobutyric acid was the main product, and isopropyl trifluoroacetate and n-butyric acid were the by-products in 95% total yield based on propane when 1 bar of propane reacted with 30 bar of CO in the presence of CaCl₂, K₂S₂O₈, and CF₃COOH at 80°C for 24 hr. Ethane gave ~ 88% total yield, but 23% acetic acid was formed by the oxidation of ethane. Cyclopentane, cycloheptane, and cyclohexane underwent reaction, giving ~ 44, 12, and 33% yields, respectively. However, cyclooctane did not undergo this reaction.

Full text of DOI:10.1006/jcat.2000.2972

Journal of Catalysis 195, 180–186 (2000)
doi:10.1006/jcat.2000.2972, available online at http://www.idealibrary.com on
CaCl -Catalyzed Functionalization of Saturated Hydrocarbons
2
with CO to Carboxylic Acids and Esters
1
Mohammad Asadullah, Tsugio Kitamura, and Yuzo Fujiwara
Department of Chemistry and Biochemistry, Graduate School of Engineering, Kyushu University, 6-10-1 Hakozaki, Fukuoka 812-8581, Japan
Received April 6, 2000; revised June 19, 2000; accepted June 19, 2000
have conducted much theoretical and practical research
The functionalization of saturated hydrocarbons such as ethane, on propane transformations to useful products such as
propane, cyclopentane, cyclohexane, cycloheptane, and cyclooc- oxidative dehydrogenation of propane to propene (3–7),
tane with CO by CaCl2 catalyst in the presence of K2S2O8 and
CF3COOH was studied. In this reaction carboxylic acids were
formed as dominant products and alkyl trifluoroacetates were
formed as by-products. In the reaction of propane, isobutyric acid
was the main product and n-butyric acid and isopropyl trifluoroac-
etate were the by-products in 95% total yield based on propane when
oxidation of propane to alcohol (8, 9) and acrylic acid
(
10–12), carbonylation of propane with CO to butanal (13),
amoxidation of propane to acrylonitrile (14, 15), vapor-
phase epoxidation of propane (16), and aminomethylation
of propane with N,N-dialkylmethylamine N-oxides to
N,N-dialkylaminomethylated propane (17).
Carboxylation of propane with CO to isobutyric and
n-butyric acids is another important functinalization pro-
1
bar of propane reacted with 30 bar of CO in the presence of CaCl2
�
(0.5 mmol), K2S2O8 (5 mmol), and CF3COOH (5 ml) at 80 C for
2
4 h. Ethane gave about 88% total yield; however, in that case 23%
acetic acid was formed by the oxidation of ethane. Cyclopentane, cess; however, much fewer examples of this reaction are
cyclohexane, and cycloheptane underwent reaction, giving about reported (18). Recently, we found that Mg powder (19, 20)
4
4, 33, and 12% total yields, respectively. However, cyclooctane did and Co(OAc) (21) promote and catalyze the carboxylation
2
not undergo this reaction. The activation parameters for the reac-
tion of propane (5 bar) with CO (30 bar) have been determined to
of propane, respectively. The former is not catalytic and
gives a low yield but the latter gives moderate yield; how-
ever, the catalyst lifetime is short. Moreover, Co(OAc)2
catalyst is very selective to carboxylate only propane. In
this paper, we describe the high-yield functionalization of
propane to isobutyric acid (1), n-butyric acid (2), and iso-
propyl trifluoroacetate (3) by highly efficient and cheap
CaCl2 catalyst in the presence of K2S2O8 and CF3COOH
1
3
14
be Ea = 130.3, 138.0, and 153.8 kJ/mol; A = 7.14 � 10 , 5.83 � 10 ,
1
6
� 1
‡
‡
and 5.80 � 10
s
; 1H = 128.0, 134.7, and 150.5 kJ/mol; 1S =
‡
1
0.3, 28.6, and 66.8J/molK;and 1G = 124.4, 123.9, and 126.9kJ/
3
53
mol for the products isobutyric acid, n-butyric acid, and isopropyl
trifluoroacetate, respectively.
�c 2000 Academic Press
Key Words: saturated hydrocarbons; carbon monoxide; calcium
dichloride; carboxylic acid; catalysis.
(
TFA). The reactions of ethane and cyclopentane also give
high yields of their corresponding carboxylic acids and
esters.
INTRODUCTION
The lower alkanes such as ethane and propane are
abundantly available in the world, being utilized as clean
burning fuels for energy supply. For example, about 200
million tons of liquid petroleum gas (LPG) are supplied
in a year, most of which is being consumed as fuel.
Thus, a new chemical process for direct and selective
functionalization of lower alkanes to upgraded products
such as alcohols, aldehydes, ketones, acids, etc. under mild
conditions is one of the most promising methods for future
organic synthesis (1, 2). However, the process remains
challenging because of the low reactivity of lower hydro-
carbons and poor selectivity of their products. Scientists
METHODS
Carboxylation Reaction of Propane
The catalyst and K2S2O8 were placed in a 25-ml stain-
less steel autoclave equipped with a Teflon-coated mag-
netic stirring bar (12 mm, octagonal). Next, 5 ml of TFA
was introduced into the autoclave which was then closed
and flushed with propane three times to replace the air in-
side and finally pressurized with the desired pressures of
propane and CO. The autoclave was then heated with stir-
ring at a fixed temperature in an oil bath for the desired
length of time. After the reaction was finished, the auto-
clave was cooled in an ice bath for 15 min and then opened.
About 80 mg of an internal standard (n-valeric acid) was
1
To whom correspondence should be addressed. Fax: +81-92-642-3548.
E-mail: yfujitcf@mbox.nc.kyushu-u.ac.jp.
180
0021-9517/00 $35.00
Copyright �c 2000 by Academic Press
All rights of reproduction in any form reserved.

FUNCTIONALIZATION OF SATURATED HYDROCARBONS WITH CO
181
added to the reaction mixture and stirred for 5 min. The car-
TABLE 1
boxylic acids were analyzed by gas chromatography using a
Shimadzu GC-8A equipped with a flame ionization detec-
tor with a 3-m � 3.0-mm-i.d. stainless steel column packed
with Unisole10T + H3PO4 (5 + 0.5)% on 80/100 mesh Uni-
Effect of Alkaline Earth Compounds on the Carboxylation
a
Reaction of Propane with CO
Yield
�
port HP at an injection/detection temperature of 220 C
mmolb
�
% c
and a column temperature of 130 C. The ester product of
Catalyst
None
1
2
3
1 + 2 + 3
TONd
propane was characterized by NMR spectroscopy.
For the determination of the optimum temperature,
amount of catalyst, pressure of carbon monoxide, amount
of K2S2O8, pressure of propane, amount of TFA, and time
course, we varied the scale in each of the experiments.
The experiments on the carboxylation of ethane, cy-
clopentane, cyclohexane, cycloheptane, and cyclooctane
were carried out in the same way. However, the ester prod-
ucts of these hydrocarbons were analyzed by using the same
gas chromatograph with a 3-m � 3-mm-i.d. stainless steel
column packed with Silicone OV-17 under the conditions
0.02
0.62
0.54
0.46
0.31
0.10
0.01
Trace
0.35
Trace
Trace
0.01
0.31
0.19
0.16
0.11
0.04
0.01
—
0.12
0.20
0.19
0.12
0.10
0.05
0.04
0.15
0.02
Trace
Trace
3.3
25.0
20.3
16.3
11.5
4.2
1.3
3.3
10.6
—
—
—
CaCl2 � 2H2O
2.26
1.84
1.48
1.04
0.38
0.12
0.30
0.11
—
CaO
CaCO3
Ca(OAc)2 � 2H2O
Ca(acac)2 � 2H2O
Ca(OH)2
CaSO4 � 2H2O
e
Mg
0.16
—
Trace
MgCl2
BaCl2
—
�
of an injection/detection temperature of 220 C and a col-
umn temperature of 100 C.
a
Conditions:25-mlautoclave, propane (6bar, 4.54mmol), CO (30bar),
alkaline earth compound (0.5 mmol), K S O (4 mmol), and TFA (5 ml)
�
2
2
8
�
After the solution phase is separated from the reaction
mixture, the solvent may be separated from the products by
distillation because the boiling point difference is sufficient
for distillation. The solid spent catalyst and reagents were
gathered and treated with water to give a clear solution
which is now under investigation.
at 80 C for 15 h.
b
Determined by GC.
Based on propane.
Based on Ca.
c
d
e
Conditions: Mg (5 mmol), TFA (3 ml), 30 h.
other metal compounds, MgCl2 and BaCl2, are also inactive
in this reaction. Although Mg powder gives a remarkable
result (19, 20), it is not catalytic. Therefore, it is obvious that
CaCl2 is a significant catalyst in the carboxylation reaction
of propane.
RESULTS AND DISCUSSION
Effect of Alkaline Earth Compounds on the Carboxylation
Reaction of Propane
First, we investigated the catalytic performance of al-
kaline earth compounds for the carboxylation reaction of
Effect of Temperature
Since CaCl2 is a catalyst for the carboxylation of propane,
the reaction was optimized. First the optimum temperature
for this reaction was determined. The results are shown in
Fig. 1.
This reaction proceeds at low temperature even at room
temperature to give low yields of 1 and 2 (0.08 and
propane with CO in the presence of K2S2O8 and TFA at
�
8
0 C for 15 h in a 25-ml stainless steel autoclave. The re-
action gives isobutyric acid (1) as a dominant product and
n-butyric acid (2) and isopropyl trifluoroacetate (3) as by-
products as shown in reaction [1]. The results of these ex-
periments are demonstrated in Table 1.
0
.04 mmol, respectively) and trace amounts of 3. The yields
of 1 and 2 increase with increasing temperature until 75–
�
8
0 C and then the yields gradually decrease with further
increases of temperature. The decreasing trend may be due
to the thermalor catalyticdecomposition ofacid productsat
This reaction proceeded in very low yields of 1, 2, and 3 high temperature to smaller molecular products which are
(
0.02, 0.01, and 0.12 mmol, respectively) when K2S2O8 was actually observed on the GC chart as several peaks. More-
used alone. The combination of Ca compounds with K2S2O8 over, the solubility of CO severely decreases with increas-
under the same reaction conditions gave remarkably high ing temperature. Thus, the carboxylation reaction becomes
yields of all the products.
slower due to the low concentration of CO in the solution.
Of the Ca compounds tested, CaCl2 gives the most re-
markable yields of 1, 2, and 3 (0.62, 0.31, and 0.20 mmol, increasing temperature beyond 75 C, maybe through a dif-
respectively; about 25% yield based on propane) where ferent mechanism. Thus, it is apparent that 75–80 C is the
On the other hand, the product 3 gradually increases with
�
�
the turnover number is 2.26 based on Ca atoms. CaSO4 did best temperature for the formation of acid products which
not bring about this reaction of propane significantly. The are the main target of this work. Therefore, the following

1
82
ASADULLAH, KITAMURA, AND FUJIWARA
when 1 mmol of K2S2O8 was used in the reaction mixture.
With increasing amounts of K2S2O8, the yields of 1 and 2 in-
creased up to 3 mmol of K2S2O8. Addition of K2S2O8 above
3
mmol causes an asymptotic increment of the yields to give
the maximum values (0.62 and 0.32 mmol of 1 and 2, re-
spectively) up to 5 mmol of K2S2O8. Further increasing the
amount of K2S2O8 causes a slight decrease in the yields of 1
and 2 while the yield of product 3 increases with increasing
amount of K2S2O8 up to 6 mmol. Excess of K2S2O8 beyond
6
mmol did not alter the yields of3. In the presence of a large
excess of K2S2O8, the overoxidation of the products 1 and
2
may occur which would lead to a decrease of the yields.
Moreover, increasing the amount of K2S2O8 resulted in
an increment of the density of the slurry inside the auto-
clave which decreased the rate of rotation of the magnetic
stirring bar. The yield of this reaction apparently depends
on the stirring rate which is also experimentally proved.
FIG. 1. Effect of temperature on the yields of CaCl2 (0.5 mmol)-
catalyzed carboxylation reaction of propane (6 bar, 4.54 mmol) with CO
(
30 bar) in the presence of K2S2O8 (4 mmol) in TFA (5 ml) for 15 h. Yield Thus, the yield gradually decreases with increasing amount
of isobutyric acid (1) (᭹), n-butyric acid (2) (᭿), and isopropyl trifluoro-
acetate (3) (᭡).
of K2S2O8.
Effect of CO Pressure
reactions were carried out at an optimum temperature of
The concentration of CO in the liquid phase is very im-
portant for the formation of the C–C bond in the carboxy-
lation reaction of propane with CO. Moreover, in this re-
action, CO acts not only as a carbonyl source but also as
a reducing agent. Both of these properties of CO increase
with increasing pressure. Therefore, the effect of various
pressures of CO in this reaction was investigated. The yields
are shown as a function of pressure in Fig. 3. This reaction
proceeded in low yields of 1 and 2 but high yield of 3 under
low pressure or absence of CO; under 1 bar of CO pressure,
the yields are 0.09, 0.13, and 0.63 mmol of 1, 2, and 3, re-
spectively. On the other hand, the reaction gives 0.65 mmol
of 3 only in the absence of CO.
�
8
0 C.
Effect of the Amount of K2S2O8
We then investigated the optimum amount of K2S2O8
for this reaction. The amount of K2S2O8 was varied from
1
to 12 mmol while the other conditions were kept con-
stant. The results of this experiments are shown in Fig. 2.
This figure reveals that the reaction of propane with CO
to corresponding acids did not take place in the absence
of K2S2O8. Very small amounts of 1, 2, and 3 were formed
FIG. 3. Effect of CO pressure on the yields of CaCl2 (0.5 mmol)-
FIG. 2. Effect of the amount of K2S2O8 on the yields of CaCl2
catalyzed carboxylation reaction of propane (6 bar, 4.54 mmol) in the
presence of K2S2O8 (5 mmol) and CF3COOH (5 ml) at 80 C for 15 h.
Yield of isobutyric acid (1) (᭹), n-butyric acid (2) (᭿), and isopropyl
trifluoroacetate (3) (᭡).
�
(
0.5 mmol)-catalyzed carboxylation reaction of propane (6 bar, 4.54 mmol)
�
with CO (30 bar) in TFA (5 ml) at 80 C for 15 h. Yield of isobutyric acid
(
1) (᭹), n-butyric acid (2) (᭿), and isopropyl trifluoroacetate (3) (᭡).

FUNCTIONALIZATION OF SATURATED HYDROCARBONS WITH CO
183
The yield of 1 increases sharply with increasing pressure
of CO until 30 bar where the maximum yield is 0.62 mmol
but the yield of 2 is the highest (0.34 mmol) at 70 atm of
CO. Further increasing the CO pressure beyond 30 atm
resulted in a lower yield of 1. In the case of 3, the yield de-
creases continuously with increasing CO pressure. Under
an atmosphere of CO, the alkyl radical formed in the reac-
tion mixture is supposed to react with CO to form the C–C
bond which gives the acid products in this reaction (22–25).
Under a low pressure of CO, it would be difficult to form the
C–C bond; thus, the yields of 1 and 2 in this reaction are low
under that condition. Furthermore, under a high pressure
of CO, the oxidation process may become difficult. There-
fore, the esterification reaction becomes slower at a high
pressure of CO. For maintaining the best concentration of
CO in the reaction mixture to facilitate the formation of the
C–C bond, the following reactions were carried out under
FIG. 5. Effect of the pressure of propane on its carboxylation reaction
with CO (30 bar) catalyzed by CaCl2 (0.5 mmol) in the presence of K2S2O8
�
(
(
5 mmol) and CF3COOH (5 ml) at 80 C for 15 h. Yield of isobutyric acid
1) (᭹), n-butyric acid (2) (᭿), and isopropyl trifluoroacetate (3) (᭡).
3
0 atm of CO pressure.
Effect of Pressure of Propane
Effect of the Amount of TFA
A high concentration of propane in the solution phase
facilitates the formation of a C–C bond rather than a C–O
bond. Thus, the yields of acid products 1 and 2 continue to
increase up to 6 bar of propane pressure (Fig. 5). Increas-
ing pressure of propane beyond 6 bar results in decreasing
yields. Here it is thought that the reaction proceeds through
the formation of propyl radicals which are generated by the
abstraction of a hydrogen atom from propane by calcium
Figure 4 demonstrates the effect of the amount of TFA
on the yield of the reaction of propane with CO in the pres-
ence of K2S2O8. In the reaction at least 2 ml of TFA was
used. It can be seen that the yields of all the products in-
crease with increasing the amount of TFA up to 5–6 ml,
where the maximum yields of 1, 2, and 3 are 0.65, 0.31,
and 0.40 mmol, respectively. The yields then decrease sud-
denly with increasing amount of TFA. The concentrations
of catalyst and oxidant play a very important role and are
controlled bythe amount ofTFA added. The higher amount
of TFA dilutes the reaction mixture and decreases the yield.
Considering the results of these experiments it can be con-
cluded that 5 ml of TFA is the optimum amount for this
reaction.
�
oxy radicals (Ca–O ). A high concentration of propane may
cause the formation of a large excess of propyl radical which
�
may be terminated by further reacting with Ca–O radicals.
However, the yield of product 3 gradually increases with
increasing propane pressure. It may be that the consumed
propoxide ion releases alcohol upon reaction with TFA.
The alcohol ultimately transforms to the ester after reac-
tion with another TFA molecule.
Once again, an atmospheric pressure of propane gives
about 95% total yield based on propane. The yield% based
on propane continuously decreases with increasing pres-
sure of propane. Since the amount of unreacted propane
increases with increasing pressure of propane in the auto-
clave, the yield based on propane decreases. From these re-
sults it can be concluded that the reaction should be carried
out under 1–2 bar of propane for high yields of products.
Effect of the Amount of CaCl2
The variation of the yields as a function of the amount
of CaCl2 is depicted in Fig. 6. In the absence of catalyst,
propane reacted slightly, giving 0.02, 0.01, and 0.12 mmol of
1
, 2, and 3, respectively. Addition of the catalyst remarkably
FIG. 4. Effect of the amount of CF3COOH on the yields of CaCl2
0.5 mmol)-catalyzed carboxylation reaction of propane (6 bar, 0.98–
increased the yields. However, a low amount of catalyst
(
4
8
(
0.05 mmol) does not result in a good yield of the products.
As the amount of catalyst increased, the yield dramat-
.54 mmol) with CO (30 bar) in the presence of K2S2O8 (5 mmol) at
�
0 C for 15 h. Yield of isobutyric acid (1) (᭹), n-butyric acid (2) (᭿),
and isopropyl trifluoroacetate (3) (᭡).
ically increased up to 0.5 mmol where the highest yields

1
84
ASADULLAH, KITAMURA, AND FUJIWARA
to the rate of the thermal or catalytic decomposition of the
products being higher than the rate of the formation.
The variation in the total yield as a function of time of the
reaction of propane (6 atm) with CO (30 atm) in the pres-
ence of CaCl2 (0.5 mmol), K2S2O8 (5 mmol), and TFA (5 ml)
�
at 80 C is shown in plot c. As the reaction time increases,
the yield gradually increases up to 35 h to the maximum
value of 33.9% . On the other hand, the reaction gives as
high as about 95% total yield when 1 atm of propane is
used for 24 h (curve d) under the same reaction conditions
as those described for curve c. A reaction time longer than
2
4 h resulted in the lower yield, maybe for the same reasons
as those described above.
FIG. 6. Effect of the amount of CaCl2 on the carboxylation reaction
of propane (8 bar, 6.06 mmol) with CO (30 bar) in the presence of K2S2O8
Reaction of Various Saturated Hydrocarbons with CO
�
(
(
5 mmol) in CF3COOH (5 ml) at 80 C for 15 h. Yield of isobutyric acid
1) (᭹), n-butyric acid (2) (᭿), and isopropyl trifluoroacetate (3) (᭡).
The carboxylation reaction of various saturated hydro-
carbons such as ethane, cyclopentane, cyclohexane, cyclo-
heptane, and cyclooctane was carried out using the opti-
mum conditions for the reaction of propane: CO (30 bar),
CaCl2 (0.5 mmol), K2S2O8 (5 mmol), and TFA (5 ml)
are 0.65, 0.33, and 0.23 mmol of 1, 2, and 3, respectively.
After that the yield of 1 gradually decreases with increas-
ing amount of catalyst whereas under the same conditions
the yield of 2 gradually increases with increasing amount
of catalyst and finally, 2 became the major product when
�
at 80 C. Table 2 presents the results of this experiments.
Ethane also undergoes the carboxylation reaction with CO
efficiently, giving as high as 88% total yield. In this case
about 23% of acetic acid is formed in place of ester. Of the
cycloalkanes used in this reaction, cyclopentane and cyclo-
hexane undergo the reaction remarkably, giving about 44
and 33% total yields, respectively. In the case of cyclopen-
tane, cyclopentanecarboxylic acid is the main product but
in the case of cyclohexane the ester is the main one. On
the other hand, cycloheptane gives less remarkable yields
where ester is the dominant product. However, cyclooctane
did not react with CaCl2 catalyst in the presence of K2S2O8
and TFA. Thus, it is apparent that the CaCl2 catalyst is more
1
.5 mmol of catalyst was used; however, the reason is not
clear at present. On the other hand, the yield of 3 decreases
with increasing amount of catalyst beyond 0.5 mmol of
CaCl2.
This figure also indicates that the overall amount of
1
+ 2 + 3 decreases with increasing amount of CaCl2 be-
yond 0.5 mmol. The decreasing trend in the yield of the
total products is due to the decomposition of the products
which is observed experimentally in the presence of a large
excess of CaCl2.
Time Course
Figure 7 demonstrates the effect of reaction time on the
yields based on propane and turnover number based on
calcium atom under various conditions. The most important
aspect of the reaction conditions has been noticed in the
time course of the reaction. It is apparent from this figure
that no reaction product is observed within 2 h under any
reaction conditions. All the products were first observed
after 2 h induction period which can be correlated with the
�
formation of the catalytic CaO radical species.
Plots a and b of this figure demonstrate the total yield
(
based on propane) and turnover number (based on Ca
atom), respectively, as a function of time of the reaction of
propane (6 bar) with CO (30 bar) in the presence of CaCl2
(
0.02 mmol), K2S2O8 (5 mmol), and TFA (5 ml). It can
be seen that the yield% and turnover number of the cata-
lyst gradually increase with time up to 40 h. The highest
turnover number obtained is 26.5 (based on Ca) where the
yield is 11.7% (based on propane). A prolonged reaction
FIG. 7. Time course of the carboxylation reaction of propane with
CO using 5 ml of TFA at 80 C. Conditions: (a and b) Propane (6 bar,
.54mmol), CO (30bar), CaCl2 (0.02mmol), and K2S2O8 (5mmol);(c) pro-
pane (6 bar), CO (30 bar), CaCl2 (0.5 mmol), and K2S2O8 (5 mmol);
�
4
time beyond 40 h resulted in lower results; this may be due (d) propane (1 bar), CO (30 bar), CaCl2 (0.5 mmol), and K2S2O8 (5 mmol).

FUNCTIONALIZATION OF SATURATED HYDROCARBONS WITH CO
185
TABLE 2
the products of 1, 2, and 3. The energies of activation (Ea)
for the formation of 1, 2, and 3 have been determined from
the slopes of the plots to be 130.3, 138.0, and 153.8 kJ/mol,
respectively. In each case the energy of activation is suffi-
cient for breaking the bond in the transition state and quite
comparable with the activation energy for the formation of
methyl radical in the Li/MgO-catalyzed oxidative coupling
of methane (22). The frequency factors (A) have also been
Reaction of Various Saturated Hydrocarbons with CO
CaCl
2
R–H + CO ������→ R-COOH + R-O-COCF
3
�
K
2
S
2
O
8
, 80 C
acid
ester
Product (mmol)a
acid ester
0.51
iso, 0.34
n, 0.18
0.31
Yield (% )b
Entry
Substrate
acid + ester
1
2
C2H6c
C3H8
—d
0.20
88
95
13
determined from the intercept of the plots to be 7.14 � 10 ,
14
16 � 1
5
.83 � 10 , and 5.80 � 10 s , respectively.
3
4
5
6
c-C5H10
c-C6H12
c-C7H14
c-C8H16
0.13
0.18
0.07
—
44
33
12
—
‡
‡
The activation enthalpy (1H ), activation entropy (1S ),
0.15
0.05
—
‡
and activation free energy (1G353) for the formation of 1, 2,
and 3 from the reaction of propane (5 bar) with CO (30 bar)
using the same reaction conditions as those described above
have been determined by Eyring plots (Fig. 9). The activa-
tion enthalpies were calculated for the formation of the
products 1, 2, and 3 from the slopes of the plots h, i, and j to
be 128.0, 134.7, and 150.5 kJ/mol, respectively. In each case
the enthalpy of activation is sufficient for breaking the bond
in the transition state. The entropies of activation which
were derived from the intercepts of the plots are 10.3, 28.6,
Reaction conditions: 25-ml autoclave, ethane, propane (1 bar,
0
.76 mmol), other hydrocarbons (1 mmol), CO (30 bar), CaCl2 (0.5 mmol),
�
K2S2O8 (5 mmol), TFA (5 ml), 80 C, 24 h.
a
Determined by GC.
b
Based on hydrocarbon.
c
Reaction time 48 h.
d
Acetic acid (0.16 mmol) was also formed.
efficient for functionalization of lower alkanes rather than and 66.8 J/mol K, respectively. The activation free energies
cycloalkanes.
are also calculated to be 124.4, 123.9, and 126.9 kJ/mol for
The activation parameters for the reaction of propane the products 1, 2, and 3, respectively. The activation param-
5 bar) with CO (30 bar) catalyzed by CaCl2 (0.5 mmol) in eters for the formation of 1 are very close to those for 2 but
(
the presence of K2S2O8 (5 mmol) and TFA (5 ml) within differ from those for the formation of 3 which reveals that
the temperature range of 343 and 358 K for the formation two reactions proceed in the two different pathways.
of isobutyric and n-butyric acids and isopropyl trifluoro-
Although the mechanism of the functionalization of
acetate have been determined by Arrhenius and Eyring saturated hydrocarbons to carboxylic acids and alkyl
equations. Figure 8 shows the Arrhenius plots e, f, and g for
FIG. 9. Eyring plots for the products of 1, 2, and 3. The thermo-
‡
‡
‡
‡
FIG. 8. Arrhenius plots for the products 1, 2, and 3. Energies of activa-
tion for the products of 1, 2, and 3 are 130.3, 138.0, and 153.8 kJ/mol, respec-
dynamic terms 1H , 1S , and 1G
are determined to be 1H =
3
53
‡
128.0, 134.7, and 150.5 kJ/mol; 1S = 10.3, 28.6, and 66.8 J/mol K; and
1
3
14
16 � 1
‡
tively, and frequency factors are 7.14 � 10 , 5.83 � 10 , and 5.80 � 10
s
,
1G = 124.4, 123.9, and 126.9 kJ/mol, respectively. Conditions: propane
3
53
respectively. Conditions: propane (5 bar), CO (30 bar), CaCl2 (0.5 mmol),
(5 bar), CO (30 bar), CaCl2 (0.5 mmol), K2S2O8 (5 mmol), and TFA (5 ml)
�
�
K2S2O8 (5 mmol), and TFA (5 ml) at 70–85 C.
at 70–85 C.

1
86
ASADULLAH, KITAMURA, AND FUJIWARA
trifluoroacetates catalyzed by CaCl2 in the presence of temperature range of 343 and 358 K for the formation of
�
K2S2O8 in TFA at 80 C is unclear, we believe that the reac- 1, 2, and 3 have been determined by Arrhenius and Eyring
tion involves hydrogen atom abstraction from the alkane by equations to be Ea = 130.3, 138.0, and 153.8 kJ/mol; A =
�
13
14
16 � 1
‡
radicals such as calcium oxy radicals (Ca–O ) or radical-like 7.14 � 10 , 5.83 � 10 , and 5.80 � 10 s ; 1H = 128.0,
‡
species to generate the corresponding alkyl radicals. The 134.7, and 150.5 kJ/mol; 1S = 10.3, 28.6, and 66.8 J/mol K;
‡
generated alkyl radicals react with CO to give acyl radicals and 1G
= 124.4, 123.9, and 126.9 kJ/mol for the prod-
353
�
�
(
R–CO ). The acyl radical then reacts with Ca–O to form ucts 1, 2, and 3, respectively. Since the reaction of lower
a complex from which the acyl cation transfers to a TFA alkanes gives very high yields and uses a cheap catalyst, it
molecule to form a mixed anhydride. It has been reported has importance from the industrial point of view.
that this type of mixed anhydride reacts further with TFA to
form carboxylic acid and trifluoroacetic anhydride (23, 24).
ACKNOWLEDGMENTS
The role of K2S2O8 is to reoxidize the –CaOH catalyst to
�
This work was supported in part by the Grants-in-Aid for Scientific
Research on Priority Area No. 283 “Innovative Synthetic Reactions” and
Scientific Research (A) No. 09355031 from the Ministry of Education,
Science, Sports and Culture of Japan.
–
CaO . Finally, K2S2O8 is transformed to KHSO4 which is
confirmed by IR spectra of the solid phase after completion
�
of the reaction. Alkyl radicals (R ) actually react very fast
with CO to form the acyl radicals (25–30). The ester prod-
ucts 3 formed in the reaction mixture may be formed in a
different way.
REFERENCES
The IR studies of the solid sample which was prepared
by the reaction of CaCl2 with TFA under a nitrogen atmos-
phere reveal that the CaCl2 first reacts with TFA to form
Ca(CF3COO)2. It is assumed that in the final reaction mix-
ture, Ca(CF3COO)2 successively reacts with K2S2O8 and
CO to form calcium peroxide which then converts to the ac-
tive speciesCa–O radical. The sufficient activation energies
and enthalpies for the formation of 1, 2 and 3 indicate that
the reaction is endothermic and controlled by external heat.
1
. Crabtree, R. H., Chem. Rev. 85, 245 (1985).
2. “Activation and Functionalization of Alkanes” (C. L. Hill, Ed.). Wiley,
New York, 1989.
3. Beretta, A., Forzatti, P., and Ranzi, E., J. Catal. 184, 469 (1999).
4. Beretta, A., Piovesan, L., and Forzatti, P., J. Catal. 184, 455 (1999).
5. Arena, F., Frusteri, F., and Parmaliana, A., Catal. Lett. 60, 59 (1999).
6. Bardin, B. B., and Davis, R. J., Appl. Catal. A 185, 283 (1999).
�
7. Creaser, D., Anderson, B., Hudgins, R. R., and Silveston, P. L., Chem.
Eng. Sci. 54, 4437 (1999).
8
9
. Raja, R., Jacob, C. R., and Ratnasamy, P., Catal. Today 49, 171 (1999).
. Hogan, T., and Sen, A., J. Am. Chem. Soc. 119, 2642 (1997).
1
1
1
0. Li., W., Oshihara, K., and Ueda, W., Appl. Catal. A 182, 357 (1999).
1. Grasselli, R. K., Catal. Today 49, 141 (1999).
2. Han, Y. F., Wang, H. M., Cheng, H., and Deng, J. F., Chem. Commun.
521 (1999).
SUMMARY AND CONCLUSIONS
Calcium dichloride has been found to be an efficient
catalyst for the reaction of lower alkanes such as ethane,
propane, cyclopentane, and cyclohexane with CO to
afford the corresponding carboxylic acids. The reaction
conditions were optimized for the reaction of propane
1
3. Sakakura, T., Ishiguro, K., Okano, M., and Sako, T., Chem. Lett. 1089
(
1997).
1
1
4. Hinz, A., and Andersson, A., Chem. Eng. Sci. 54, 4407 (1999).
5. Nilsson, J., Landa-Canovas, A. R., Hansen, S. A., and Andersson, A.,
J. Catal. 186, 442 (1999).
with CO. Propane gave about 95% total yield (based on 16. Uphade, B. S., Okumura, M., Tsubota, S., and Haruta, M., Appl. Catal.
A 190, 43 (2000).
propane) of 1, 2, and 3 when 1 bar of propane reacted with
1
1
7. Taniguchi, Y., Kitamura, T., Fujiwara, Y., Horie, S., and Takaki, K.,
Catal. Today 36, 85 (1997).
8. Miyata, T., Nakata, K., Yamaoka, Y., Taniguchi, Y., Takaki, K., and
Fujiwara, Y., Chem. Lett. 1005 (1993).
3
0 bar of CO in the presence of CaCl2 (0.5 mmol), K2S2O8
�
(
5 mmol), and TFA (5 ml) at 80 C for 24 h. In this reaction
1
is formed as the main product and 2 and 3 are formed
as by-products. Using the optimum reaction conditions for 19. Asadullah, M., Kitamura, T., and Fujiwara, Y., Chem. Lett. 449 (1999).
2
2
2
0. Asadullah, M., Kitamura, T., and Fujiwara, Y., Appl. Organomet.
Chem. 13, 539 (1999).
1. Asadullah, M., Taniguchi, Y., Kitamura, T., and Fujiwara, Y., Tetrahe-
dron Lett. 40, 8867 (1999).
propane, we attempted the reaction of several saturated
hydrocarbons with CO. Ethane gave about 88% total yield;
however, in this case 23% acetic acid was formed by the
oxidation of ethane. The ester product of the reaction of
2. Lunsford, J. H., Angew. Chem., Int. Ed. Engl. 34, 970 (1995).
ethane was not detected by GC. Of the cycloalkanes used 23. Taniguchi, Y., Yamaoka, Y., Nakata, K., and Fujiwara, Y., Chem. Lett.
3
45 (1995).
in this reaction, cyclopentane and cyclohexane undergo
the reaction remarkably, giving about 44 and 33% total
yields, respectively. In the case of cyclopentane, acid is the
main product but in the case of cyclohexane the ester is
2
2
4. Nakata, K., Yamaoka, Y., Miyata, T., Taniguchi, Y., Takaki, K., and
Fujiwara, Y., J. Organomet. Chem. 473, 329 (1994).
5. Nagahara, K., Ryu, I., Kambe, N., Komatsu, M., and Sonoda, N.,
J. Org. Chem. 60, 7384 (1995).
the main one. On the other hand, cycloheptane gives less 26. York, A. P. E., Claridge, J. B., Brungs, A. J., Tsang, S. C., and Green,
M. L. H., Chem. Commun. 39 (1997).
remarkable yields where ester is the dominant product.
Cyclooctane did not react.
The activation parameters for the reaction of propane
2
2
2
7. Bitter, J. H., Seshan, K., and Lercher, J. A., J. Catal. 171, 279 (1997).
8. Lu, Y., Xue, J., Liu, Y., and Shen, S., Chem. Lett. 515 (1997).
9. Olsbye, U., Wurzel, T., and Mleczko, L., Ind. Eng. Chem. Res. 36, 5180
(
5 bar) with CO (30 bar) catalyzed by CaCl2 (0.5 mmol) in
(
1997).
the presence of K2S2O8 (5 mmol) and TFA (5 ml) within the 30. Bradford, M. C. J., and Vannice, M. A., J. Catal. 173, 157 (1998).