Hydroalkoxycarbonylation of olefins in the presence of palladium phosphine complexes: High activity and regioselectivity

Article abstract of DOI:10.1134/S0965544106020095

Several types of catalyst systems were examined in the olefin hydroalkoxycarbonylation reaction. The systems contained Pd(PPh ₃)₄, PdCl₂(PPh₃)₂, or some other palladium compounds as a principal component. The second component (promoter) was p-toluenesulfonic acid or diphenyl(m-sulfophenyl)phosphine, which combines both the ligand and promoter functions. An important feature of these systems is their high activity in the hydroalkoxycarbonylation of ethylene and a high regioselectivity (83-100%) in the hydroalkoxycarbonylation of α-olefins with respect to linear products. Thus, it was unnecessary to introduce additional stabilizing ligands to augment the catalyst and promoter. The esters obtained can find application in the pharmaceutical industry and perfumery, as well as in other industries. Nauka/Interperiodica 2006.

Full text of DOI:10.1134/S0965544106020095

ISSN 0965-5441, Petroleum Chemistry, 2006, Vol. 46, No. 2, pp. 117–121. © Pleiades Publishing, Inc., 2006.
Original Russian Text © E.G. Chepaikin, A.P. Bezruchenko, Kh.A. Suerbaev, K.M. Shalmagambetov, 2006, published in Neftekhimiya, 2006, Vol. 46, No. 2, pp. 134–138.
Hydroalkoxycarbonylation of Olefins in the Presence
of Palladium Phosphine Complexes: High Activity
and Regioselectivity
E. G. Chepaikina, 1, A. P. Bezruchenkoa, Kh. A. Suerbaevb, and K. M. Shalmagambetovb
aInstitute of Structural Macrokinetics and Materials Science, Chernogolovka, Moscow oblast, Russia
1e-mail: echep@ism.ac.ru
bAl-Faraby Kazakh National University, Almaty, Kazakhstan
Received July 1, 2005
Abstract—Several types of catalyst systems were examined in the olefin hydroalkoxycarbonylation reaction.
The systems contained Pd(PPh3)4, PdCl2(PPh3)2, or some other palladium compounds as a principal compo-
nent. The second component (promoter) was p-toluenesulfonic acid or diphenyl(m-sulfophenyl)phosphine,
which combines both the ligand and promoter functions. An important feature of these systems is their high
activity in the hydroalkoxycarbonylation of ethylene and a high regioselectivity (83–100%) in the hydroalkoxy-
carbonylation of α-olefins with respect to linear products. Thus, it was unnecessary to introduce additional sta-
bilizing ligands to augment the catalyst and promoter. The esters obtained can find application in the pharma-
ceutical industry and perfumery, as well as in other industries.
DOI: 10.1134/S0965544106020095
Catalytic carbonylation reactions of olefins lead to
An important characteristic of catalyst systems for
various oxygenated compounds of industrial impor- the carbonylation of α-olefins is regioselectivity, i.e.,
tance [1, 2]. The most important and widespread pro- the selectivity for an unbranched carbonylation product
cesses relate to the manufacture of acids and esters. [11–13]. Catalyst systems that contain an excess of free
Catalysts that provide the formation of alternating phosphine may be considered as ternary systems. Of
copolymers upon the carbonylation of olefins have also course, binary systems are simpler in use than their ter-
been found [3–5]. Olefin carbonylation catalysts are nary counterparts. However, such systems have not
cobalt and nickel carbonyls and palladium compounds. been studied in sufficient detail. Therefore, an impor-
The palladium compounds turned out to be so effective tant part of this work concerns a comparative study of
that they allowed the hydroxycarboxylation and the catalytic activity of two systems of this kind,
hydroalkoxycarbonylation reactions to be conducted Pd(PPh3)4–TsOH (I) and Pd(PPh3)2Cl2–TsOH (II).
under mild conditions [4, 6].
The aim of this study was to reveal whether an eth-
ylene methoxycarbonylation catalyst could be repeat-
edly used, as well as to compare some characteristics of
systems I and II and to determine the regioselectivity of
α-olefin alkoxycarbonylation reactions over system I.
Analyzing published data, we have concluded that
the stringent conditions required for the catalytic carbo-
nylation of olefins are due to the presence of halide
anionic ligands [7, 8]. The use of palladium acetate or
acetylacetonate and phosphine ligands leads to effec-
tive catalytic systems operating even at atmospheric
pressure [4–6].
EXPERIMENTAL
The palladium complexes Pd(PPh3)4 [14],
PdCl2(PPh3)2 [15], and Pd(acac)2 [16] were prepared by
conventional procedures. p-Toluenesulfonic acid was
recrystallized from 96% ethanol and dried to have the
TsOH · H2O composition. Triphenylphosphine was
recrystallized from an ether–ethanol solvent blend until
a constant melting point. Diphenyl(m-sulfophe-
nyl)phosphine in both sodium and acid forms was pre-
pared as described in [4]. Reagent-grade alcohols and
An admixture of hydrogen to carbon monoxide lev-
els the inhibiting action of halide ligands [8, 9].Another
method that can produce the same result is the addition
of acids whose anions are weakly coordinating ligands,
for example, p-toluenesulfonic acid (TsOH) [4–6, 10–
13]. We have previously found that palladium com-
plexes containing a halide ligand need to be activated
by heating to a certain temperature [12, 13].
In some cases [4, 5], as well as in this work, we have liquid olefins were distilled prior to use. Ethylene
developed methods for the repeated use of catalysts. In (99.5%), isobutylene (99%), and carbon monoxide
this respect, good results were obtained in ethylene (99.8%) were used without additional purification. Cat-
methoxycarbonylation.
alytic experiments were conducted in 100-ml stainless
117

118
CHEPAIKIN et al.
Table 1. Stability of palladium–phosphine catalyst systems in ethylene hydromethoxycarbonylation: [Pd(PPh3)4] =
0.001 mol/l; [PPh3] = 0.006 mol/l; initial gas pressures: CO = 1.5 MPa, C2H4 = 1.5 MPa; T = 90°C; solvent, methanol (20 ml)
Number
of gas charges
Turnover frequency in successive runs,
in terms of moles of ester per mole of Pd per h
Ester,
mmol
Set no.
Promoter, mol/l
1
2
3
TsOH 0.02
5
5
7
11.5
10.6
12.0
10.6
9.5
10.5
7.6
10.7
4.9
10.5
3.6
115
TsOH 0.01
87.5
140.0
Ph2P–C6H4–(m)–SO3H 0.01
13.4
10.8
12.3
12.1
11.6
11.0
–1
Note: In set 3, Pd(acac) was used. TOF values of 11.0 and 10.8 min were found in runs 6 and 7, respectively.
2
steel reactors equipped with thermostatic jackets or was about 12 min–1. According to calculations, the
electric heating. A reactor was charged with compo- reaction rate was 0.7 mol/(l h) under the given condi-
nents of a catalyst system and an alcohol, hermetically tions. Note that the pressure dropped from 3.0 to
sealed, and purged twice with carbon monoxide to 1.5 MPa according to a linear law. Thus, the reaction
remove air. Then, an olefin was fed, carbon monoxide was zero order with respect to gases within this range.
was admitted to a required pressure, and heating and No side products of the reaction were found by GLC.
stirring were switched on. The progress of the reaction
was monitored by following the pressure drop in the
autoclave and by sampling. After completion of the
reaction, the reactor was cooled to room temperature,
and the excess gas was allowed to bleed. The reaction
mixture was subjected to fractional distillation to iso-
In the fourth set of experiments, we studied the
influence of solvent composition on the ethylene
hydromethoxycarbonylation rate in the presence of
the Pd(acac)2–Ph2P(m-C6H4SO3H) catalyst system
(Table 2). It turned out that the reaction rate had the
highest value in pure methanol and decreased when
water was added or methanol was diluted with methyl
propionate. The plot of the reaction rate on the metha-
nol concentration in methyl propionate has the form of
a curve with saturation. Over the range of 2.0–5.0 MPa,
the rate does not depend on the total pressure.
1
late the pure products. The isolated product fraction
was analyzed by GLC to determine the linear-to-
branched product ratio.
RESULTS AND DISCUSSION
Thus, the most effective system for ethylene
hydromethoxycarbonylation appears to be the
Pd(acac)2–Ph2P(m-C6H4SO3H) catalyst system (Table 1).
The reaction under the discussion is expressed by
the following general equation
R1(R2)C=CH2 + CO + R3OH
R1(R2)CHCH2C(O)OR3 + R1(R2)C(CH3)C(O)OR3
(1)
Isobutylene Hydroethoxycarbonylation
The experiments were carried out in the solvent-free
mode in the presence of Pd(PPh3)4–TsOH (I) and
PdCl2(PPh3)2–TsOH (II) systems. The dependence of
the ethyl isovalerate yield on the ratio between compo-
nents of the catalyst system is shown in Fig. 1. It is seen
that the molar ratio Pd/TsOH = 1/12 is optimal. As in
all other cases, system I turns out to be slightly more
R1 = ç, Me, Bu, Pent; R2 = H, Me; R3 = Me, Et, iso-Pr,
Bu.
Ethylene Hydromethoxycarbonylation
To determine the activity and stability of the catalyst
systems for ethylene hydromethoxycarbonylation, we
performed three sets of experiments (Table 1). The
experiments were run in the presence of Pd(PPh3)4 and
excess PPh3 (molar ratio Pd(PPh3)4/PPh3 = 6) in meth-
anol using two TsOH concentrations. Each set included
at least five 100-min runs. At an TsOH concentration of
0.02 mol/l, the catalytic activity (turnover frequency)
of the system was about 11 moles of ester/(mole of
Pd min). At [TsOH] = 0.01 mol/l, the turnover fre-
quency (TOF) dropped approximately by a factor of
three after the fifth cycle. Interesting results were
obtained using the Pd(acac)2–Ph2P(m-C6H4SO3H) sys-
tem. When the catalyst was used seven times, the TOF
Table 2. Influence of the solution composition on the rate
of ethylene hydromethoxycarbonylation in the presence of the
Pd(acac)2–Ph2P(m-C6H4SO3H) catalyst system: [Pd(acac)2] =
0.002 mol/l, [Ph2P(m-C6H4SO3H)] = 0.04 mol/l, T = 90°C.
Initial gases pressures: CO = 2.5 MPa, C2H4 = 2.5 MPa
CH3OH,
ml
C2H5COOCH3,
ml
r,
Set no.
H2O, ml
mol/(l h)
1
2
3
4
5
25
24
15
10
5
–
1
–
–
–
–
1.7
1.5
1.6
1.4
0.9
–
10
15
20
1
Because of the close boiling points, we failed to separate the lin-
ear and branched products of 1-hexene and 1-heptene hydro-
ethoxycarbonylation by fractional distillation.
PETROLEUM CHEMISTRY Vol. 46 No. 2 2006

HYDROALKOXYCARBONYLATION OF OLEFINS
119
Yield, mol %
Yield, mol %
100
100
80
60
40
20
80
II
60
II
I
40
I
20
0
2
4
6
8
10
12
14
0
50
60
70
80
90 100 110 120
[n-TsOH]/[Pd], mol
T, °ë
Fig. 1.Yield of ethyl isovalerate (mol %) as a function of the
component ratio of catalyst systems at p = 2.0 MPa, T =
CO
Fig. 2. Temperature dependence of the ethyl isovalerate yield
100°C, and
τ
=
4
h: (I) Pd(PPh ) –TsOH and
at p = 2.0 MPa, τ = 4 h, and [Pd] : [TsOH] = 1 : 12 (mol)
3 4
CO
(II) PdCl (PPh ) –TsOH.
on (I) Pd(PPh ) –TsOH and (II) PdCl (PPh ) –TsOH.
2
3 2
3 4 2 3 2
Yield, mol %
100
Yield, mol %
100
I
80
60
40
20
80
60
40
20
I
II
II
0
1
2
3
4
5
6
7
8
0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
On-stream time, h
P, MPa
Fig. 4.Yield of ethyl isovalerate as a function of the reaction time
Fig. 3. Yield of ethyl isovalerate as a function of pressure at
T = 100°C, τ = 4 h and [Pd] : [TsOH] = 1 : 12 (mol)
at T = 100°C, p = 2.0 MPa, and [Pd] : [TsOH] = 1 : 12 (mol)
CO
on (I) Pd(PPh ) –TsOH and (II) PdCl (PPh ) –TsOH.
on (I) Pd(PPh ) –TsOH and (II) PdCl (PPh ) –TsOH.
3 4
2
3 2
3 4 2 3 2
active, presumably because it contains a larger excess
of triphenylphosphine with respect to palladium. The
temperature dependence for the formation of ethyl
isovalerate (Fig. 2) flattens out at 100°C for system I
and has a maximum at the same temperature for system
II. It may be assumed that a further increase in temper-
ature leads to the decomposition of catalytically active
complexes.
The catalytic activities of the complexes Pd(PPh3)4
and PdCl2(PPh3)2 and the systems Pd(PPh3)4–TsOH,
PdCl2(PPh3)2–TsOH, Pd(PPh3)4–PPh3–TsOH, and
PdCl2(PPh3)2–PPh3–TsOH on their basis in the isobuty-
lene hydroethoxycarbonylation reaction are collated in
Table 3. In the presence of the PdCl2(PPh3)2 complex,
the product yield is small, 10 mol %. Under the same
conditions in the presence of p-toluenesulfonic acid as
a promoter at a [PdCl2(PPh3)2]/[TsOH] molar ratio of
1 : 12, the product yield sharply increases to 85.0 mol %
(Table 3, entries 1, 2). In the presence of Pd(PPh3)4, the
reaction does not occur. From the data presented in
Table 3, it also follows that the catalytic activities of the
systems I and II are comparable with those of the
Pd(PPh3)4–PPh3–TsOH and PdCl2(PPh3)2–PPh3–TsOH
The dependence of the ethyl isovalerate yield on the
carbon monoxide pressure follows the first-order rate
law up to a pressure of 2.0 MPa; then, the yield drops
(Fig. 3). This fall can be explained by the fact that,
beginning from 2.0 MPa, carbon monoxide inhibits the
coordination (activation) of the olefin. As follows from
Fig. 4, the maximal yield of ethyl isovalerate is attained
over 2 or 4 h with system I or II, respectively.
PETROLEUM CHEMISTRY Vol. 46 No. 2 2006

120
CHEPAIKIN et al.
Table 3. Influence of the triphenylphosphine to p-toluenesulfonic acid ratio on isobutylene hydroethoxycarbonylation: pCO
2.0 MPa, T = 100°C, τ = 4 h, [Pd] = 0.000115 mol/l, [Pd] : [C2H5OH] : [iso-C4H8] = 1 : 435 : 550. Solvent-free conditions
=
Component ratio of catalyst system
Ethyl isovalerate yield*,
%
Set no.
Catalyst system
[PPh3]
[TsOH]
1
2
3
4
5
6
PdCl2(PPh3)2
–
–
–
–
4
7
–
10.0
85.0
–
PdCl2(PPh3)2–TsOH
Pd(PPh3)4
12
–
Pd(PPh3)4–TsOH
12
12
12
95.4
92.5
96.0
PdCl2(PPh3)2–PPh3–TsOH
Pd(PPh3)4–PPh3–TsOH
* The yields are given in terms of the product isolated by fractional distillation.
ternary systems (Table 3, entries 2, 4–6) examined ear- anism of ethylene hydrocarboxylation on binuclear Pd
lier [6, 12, 13].
complex I in acetic or propionic acid solution [5].
The “alkoxide” [20] and “hydride” [21, 22] mecha-
nisms are more broadly known. The alkoxide mecha-
nism of ethylene carbonylation is confirmed by indirect
data obtained in the alternating copolymerization of
carbon monoxide with ethylene [3, 5, 20]. On the basis
of reactions of isolated intermediates, Cavinato and
Toniolo [22] give preference to the hydride mechanism.
Petrov and Noskov surveyed in [23] both their
detailed kinetic studies and in situ spectral investigation
of several catalyst systems. They believe that the multi-
route hydride mechanism is realized in this case. Con-
siderable attention was given to the problem of regiose-
lectivity in the hydrocarboxylation of α-olefins. Regi-
oselectivity depends on many factors, such as the CO
pressure, the water concentration, and the nature of the
anionic ligand. Note that the regioselectivity was not
more than 90% when ClO4– and CF3SO3– counterions
were used in cationic palladium complexes for styrene
hydrocarboxylation.
Therefore, a practically quantitative yield of ethyl
isovalerate is attained under the optimal conditions (a
temperature of 100°C, a CO pressure of 2.0 MPa, and a
molar ratio [Pd] : [TsOH] : [i-C4H8] : [C2H5OH] =
1 : 12 : 550 : 435).
High regioselectivity (100%) in isobutylene
hydroalkoxycarbonylation in the presence of ethanol,
isopropanol, and butanol is not surprising and seems to
be characteristic of isobutylene. The regioselectivities
of 1-hexene and 1-heptene hydroethoxycarbonylation
in the presence of system I were 88.2 and 83.0%,
respectively, with the ester yields being practically
quantitative (Table 4). Under similar conditions, the
regioselectivity of 1-nonene hydroethoxycarbonylation
in the presence of palladium chloride–phosphine sys-
tems was 68% [17]. A regioselectivity on the order of
64–69% was found in isopropanol and 88%, in butanol.
However, reaction rates in butanol were rather low, and
the total ester yield was only 48% for 6 h [18].
There are several schemes for the mechanism of
hydrocarboxylation and hydroalkoxycarbonylation of
olefins. A four-membered palladacyclic scheme was
proposed in [19]. A variant of this scheme is the mech-
Thus, elucidation of the high regioselecivity of these
catalyst systems as compared with published data [17,
18] calls for further, more detailed investigation of the
kinetics and mechanism of the process.
REFERENCES
Table 4. Hydroalkoxycarbonylation of olefins in the pres-
ence of the Pd(PPh3)4–TsOH system: [Pd] = 0.000115 mol/l,
[Pd] : [TsOH] : [hexane (heptene)] : [ethanol] = 1 : 12 : 550 :
435, pCO = 2 MPa, T = 100°C, and τ = 4
1. L. Falbe, Carbon Monoxide in Organic Synthesis
(Springer, New York, 1970).
2. Carbonylation of Unsaturated Hydrocarbons, Ed. by
D. N. Rudkovskii (Khimiya, Leningrad, 1968) [in Rus-
sian].
Product yields, mol %
No.
Olefin
Alcohol
linear
product
branched
product
3. A. Sen, Adv. Polym. Sci. 73/74, 125 (1986).
4. E. G. Chepaikin, A. P. Bezruchenko, G. N. Boiko, and
A. A. Leshcheva, Izv. Akad. Nauk, Ser. Khim., No. 3,
401 (1994).
1
2
3
4
5
Isobutylene Ethanol
Isobutylene Isopropanol
Isobutylene Butanol
95.4
90.1
89.0
88.2
83.0
–
–
5. E. G. Chepaikin, A. P. Bezruchenko, and A. A. Le-
–
shcheva, Kinet. Katal. 40, 348 (1999).
1-Hexene
1-Hexene
Ethanol
Ethanol
11.8
17.0
6. E. G. Chepaikin, A. P. Bezruchenko, A. Ben’ei, and F. Io,
Izv. Akad. Nauk SSSR, Ser. Khim., No. 3, 743 (1989).
PETROLEUM CHEMISTRY Vol. 46 No. 2 2006

HYDROALKOXYCARBONYLATION OF OLEFINS
121
7. K. Bittler, N. Kutepov, D. Neubauer, and H. Reis, 15. J. Chatt and L. M. Venanzi, J. Chem. Soc., 2351 (1957).
Angew. Chem., Int. Ed. Engl. 7, 329 (1968).
8. D. M. Fenton, J. Org. Chem. 38, 3192 (1973).
16. M. C. Datta and C. R. Saha, J. Appl. Chem. Biotechnol.
28, 709 (1978).
17. A. D. Karpyuk, M. I. Terekhova, N. D. Kolosova, et al.,
Izv. Akad. Nauk SSSR, Ser. Khim., No. 3, 639 (1985).
9. A. R. El’man, V. M. Matveev, E. V. Slivinskii, and
S. M. Loktev, Khim.-Farm. Zh., No. 3, 47 (1990).
18. A. D. Karpyuk, M. I. Terekhova, N. D. Kolosova, et al.,
Izv. Akad. Nauk SSSR, Ser. Khim., No. 1, 233 (1985).
10. Z. Hongying, F. Hongxiang, L. Shiyie, et al., Fenzi
Cuihua 5, 268 (1991).
19. E. S. Petrov, Zh. Fiz. Khim. 62, 2858 (1988).
11. Kh. A. Suerbaev, G. M. Abyzbekova, K. M. Shalmagam-
betov, and K. A. Zhubanov, Zh. Obshch. Khim. 70, 553
(2000).
20. A. D. Karpyuk, N. D. Kolosova, M. I. Terekhova, et al.,
Dokl. Akad. Nauk SSSR 277, 1402 (1984).
12. Kh. A. Suerbaev, K. M. Shalmagambetov, and
21. E. Drent and P. Budzelaar, Chem. Rev., No. 2, 663
(1996).
K. A. Zhubanov, Zh. Obshch. Khim. 70, 2043 (2000).
13. Kh. A. Suerbaev, K. M. Shalmagambetov, and
22. G. Cavinato and L. Toniolo, J. Organomet. Chem. 398,
K. A. Zhubanov, Zh. Obshch. Khim. 70, 2046 (2000).
187 (1990).
14. Y. Tatsuno, T. Yoshida, and Seiotsuka, Inorg. Synth. 19, 23. E. S. Petrov and Yu. G. Noskov, Ross. Khim. Zh. 42 (4),
220 (1979).
149 (1998).
PETROLEUM CHEMISTRY Vol. 46 No. 2 2006