Effects of bulky ligands and water in Pd-catalyzed oxidative carbonylation of phenol

Article abstract of DOI:10.1016/j.molcata.2005.04.031

A diaryloxy Pd complex with a bulky 6,6′-dimethyl-2,2′-bipyridyl (6,6′-Me₂bpy) ligand reacted with pressurized CO (5 MPa) at 25 °C to produce a diaryl carbonate, whereas a diaryloxy Pd complex with an unsubstituted 2,2′-bipyridyl (bpy) ligand hardly reacted. ¹H and ¹³C NMR studies revealed that CO inserts into one of the Pd-O bonds in the latter complex to form a Pd aryloxycarbonyl complex, but that the subsequent reductive elimination of diaryl carbonate is slow. This is consistent with the much higher catalytic activity of the Pd-(6,6′-Me₂bpy) system for the oxidative carbonylation of phenol compared to the Pd-bpy system. To verify the steric effects of the ligands, the catalytic performance was also examined using 2,2′-bioxazolyl ligands with various substituents. Introducing bulky substituents at the 4,4′-position effectively promoted the catalytic reaction. The TONs of DPC increased in the following order: methyl < benzyl < iso-butyl < tert-butyl. The methylene-bridged bioxazolyl ligand with tert-butyl groups gave the highest TON (54 mol-DPC/mol-Pd in 3 h), which is higher than the TON for the 6,6′-Me₂bpy ligand. The addition of molecular sieve 3A to the reaction system further improved the TON and suppressed phenyl salicylate formation. The addition of the molecular sieve also prevented CO₂ formation, probably due to suppression of the reaction between CO and water, in addition to suppression of the hydrolysis of DPC.

Full text of DOI:10.1016/j.molcata.2005.04.031

Journal of Molecular Catalysis A: Chemical 236 (2005) 149–155
Effects of bulky ligands and water in Pd-catalyzed oxidative
carbonylation of phenol
Hiroyuki Yasuda, Keiji Watarai, Jun-Chul Choi, Toshiyasu Sakakura∗
National Institute of Advanced Industrial Science and Technology (AIST), Tsukuba Central 5, 1-1-1 Higashi,
Tsukuba, Ibaraki 305-8565, Japan
Received 10 March 2005; received in revised form 18 April 2005; accepted 18 April 2005
Abstract
A diaryloxy Pd complex with a bulky 6,6ꢀ-dimethyl-2,2ꢀ-bipyridyl (6,6ꢀ-Me2bpy) ligand reacted with pressurized CO (5 MPa) at 25 ◦C to
1
produce a diaryl carbonate, whereas a diaryloxy Pd complex with an unsubstituted 2,2ꢀ-bipyridyl (bpy) ligand hardly reacted. H and 13C
NMR studies revealed that CO inserts into one of the Pd–O bonds in the latter complex to form a Pd aryloxycarbonyl complex, but that the
subsequent reductive elimination of diaryl carbonate is slow. This is consistent with the much higher catalytic activity of the Pd-(6,6ꢀ-Me2bpy)
system for the oxidative carbonylation of phenol compared to the Pd–bpy system. To verify the steric effects of the ligands, the catalytic
performance was also examined using 2,2ꢀ-bioxazolyl ligands with various substituents. Introducing bulky substituents at the 4,4ꢀ-position
effectively promoted the catalytic reaction. The TONs of DPC increased in the following order: methyl < benzyl < iso-butyl < tert-butyl. The
methylene-bridged bioxazolyl ligand with tert-butyl groups gave the highest TON (54 mol-DPC/mol-Pd in 3 h), which is higher than the TON
for the 6,6ꢀ-Me2bpy ligand. The addition of molecular sieve 3A to the reaction system further improved the TON and suppressed phenyl
salicylate formation. The addition of the molecular sieve also prevented CO2 formation, probably due to suppression of the reaction between
CO and water, in addition to suppression of the hydrolysis of DPC.
© 2005 Elsevier B.V. All rights reserved.
Keywords: Palladium; Oxidative carbonylation; Diphenyl carbonate; Bipyridyl; Bioxazolyl
1
2
1. Introduction
2PhOH + CO + O2 → PhOCO2Ph + H2O
(1)
Diphenyl carbonate (DPC) is an important precursor for
producing polycarbonates by a melt-polymerization process
[1,2]. DPC is currently synthesized by converting dimethyl
carbonate through transesterification or using highly toxic
and corrosive phosgene. Thus, a more direct, environmen-
tally benign synthesis of DPC is desirable. Oxidative car-
bonylation of phenol is a possible DPC synthetic method
(Eq. (1)). Numerous catalytic systems that include Pd have
been reported [3–13]. However, the catalytic activities must
be further improved for practical applications. In addition,
there are few studies on the reaction mechanism [14] and the
proposed mechanisms are speculative.
We recently isolated L2Pd(OPh)2 (L = ligand) and L2
Pd(CO2Ph)X (X = halogen) complexes, which are key inter-
mediates for the DPC forming reaction, and investigated their
reactivity [15,16]. We found that DPC is formed from both
complexes via L2Pd(OPh)(CO2Ph), although the reductive
elimination of DPC is slow. It is known that introducing bulky
substituents onto bidentate N–N ligands in Pd complexes ef-
fectively promotes the reaction [17–19]. It is assumed that the
reductive elimination of DPC from L2Pd(OPh)(CO2Ph) is ac-
celerated. However, the reductive elimination rates of DPC
using ligands with vastly different steric hindrances have not
been compared.
This study initially focused on bipyridyl ligands, which
exhibitmarkedsubstitutioneffectsonthePd-catalyzedoxida-
tive carbonylation of phenol [19]. We synthesized diaryloxy
∗
Corresponding author. Tel.: +81 29 861 4719; fax: +81 29 861 4719.
E-mail address: t-sakakura@aist.go.jp (T. Sakakura).
1381-1169/$ – see front matter © 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcata.2005.04.031

150
H. Yasuda et al. / Journal of Molecular Catalysis A: Chemical 236 (2005) 149–155
Pd complexes with a catalytically efficient 6,6ꢀ-dimethyl-
2,2ꢀ-bipyridyl (6,6ꢀ-Me2bpy) ligand 1a and an inefficient
2,2ꢀ-bipyridyl (bpy) ligand 1b. Then the relative rates of
reductive elimination of DPC were compared through the
reaction with pressurized CO at a complex level. The cat-
alytic performance using bioxazolyl ligands with various
substituents was also investigated to verify the steric ef-
fects of substituents. Bioxazolyl ligands are advantageous
since they are easily synthesized, stable in an oxidative
environment, and the steric effects are tunable. Further-
more, the effects of dehydrating agents and the CO2 for-
mation routes in the oxidative carbonylation of phenol were
examined.
Table 2
Reaction of diaryloxy Pd complexes with COa
Entry
Complex
Time (h)
Yield (%)b
1
2
3
4
a
2a
2a
2b
2b
0.5
3
0.5
3
57
61
0.3
8
Reaction conditions: Pd complex, 30 mg; THF, 5 ml; CO, 5 MPa; 25 ◦C.
Yield of di(p-tert-butyl)phenyl carbonate based on the Pd complex.
b
2.2. Reactivity of diaryloxy Pd complexes with bipyridyl
ligands toward CO
New diaryloxy Pd complexes 2a and 2b, which are pos-
sible model intermediates in this reaction, were prepared by
reacting Pd(OAc)2 with two equivalents of sodium p-tert-
butylphenoxide in the presence of 1a and 1b, respectively
2. Results and discussion
1
[20]. Complexes 2a and 2b were characterized by H and
2.1. Reaction catalyzed by Pd complexes with bipyridyl
ligands
13C NMR spectroscopies as well as elemental analysis. 2b
was characterized by X-ray crystallography, but the residual
values were unsatisfactory.
To ascertain the effectiveness of substituted 2,2ꢀ-
bipyridyl ligands for the Pd-catalyzed oxidative car-
bonylation of phenol, the catalytic activity of DPC
formation using 1a–1c was measured. The reaction was
conducted in the presence of Mn(TMHD)3 (TMHD =
2,2,6,6-tetramethyl-3,5-heptanedionate) and (Ph3P=)2NBr
(bis(triphenylphosphoranylidene)ammonium
bromide)
We examined the reactivity of 2a and 2b with pressurized
CO. Di(p-tert-butyl)phenyl carbonate was instantly formed
when 2a was reacted with CO (5 MPa) in THF at 25 ◦C. The
yield of carbonate formation based on the palladium complex
after 0.5 h was 57% (Table 2). In contrast, the reaction of 2b
with CO was very slow and the yields of the carbonate after
0.5 and 3 h were 0.3 and 8%, respectively.
To understand this difference, the reactions of 2a and
2b with CO were monitored by high-pressure NMR spec-
troscopy. A signal assigned to the carbon of PdCOO [15]
using (MeCN)2PdCl2 as a Pd source. Table 1 lists the
turnover number (TON) of DPC, TON of phenyl salicylate,
and the selectivity for DPC. The Pd complex with 1a
efficiently catalyzed the reaction to produce DPC. On
the other hand, 1b or 1c resulted in a low activity and
selectivity of DPC. These results are consistent with the
previous report [19]. Thus, a pronounced substitution
effect of the 2,2ꢀ-bipyridyl ligand at the 6,6ꢀ-position is
confirmed.
Table 1
Pd-catalyzed oxidative carbonylation of phenol using 2,2ꢀ-bipyridyl ligandsa
Entry
Ligand
TON (DPC)b
TON (PS)c
Selectivity of
DPC (%)d
1
2
3
1a
1b
1c
48
4
8
5
5
77
34
28
2
a
Reaction conditions: phenol, 32 mmol; (MeCN)2PdCl2, 0.012 mmol; ligand, 0.012 mmol; Mn(TMHD)3, 0.024 mmol; (Ph3P=)2NBr, 0.24 mmol; CO, 3 MPa;
air, 1.5 MPa; 100 ◦C; 3 h.
b
Turnover number of DPC (mol-DPC/mol-Pd).
Turnover number of phenyl salicylate (mol-PS/mol-Pd).
DPC percentage in the products evaluated from the peak areas on GC and the products identified by GC–MS.
c
d

H. Yasuda et al. / Journal of Molecular Catalysis A: Chemical 236 (2005) 149–155
151
that the methyl groups in the 6,6ꢀ-position of 1b acceler-
ate the reductive elimination of diaryl carbonate. Presum-
ably, the methyl groups sterically destabilize the aryloxycar-
bonyl intermediate. These results suggest that using biden-
tate ligands with sterically bulkier substituents give a bet-
ter catalyst for the Pd-catalyzed oxidative carbonylation of
phenol.
2.3. Reaction catalyzed by Pd complexes with
bioxazolyl ligands
Since we confirmed that the reductive elimination of DPC
from Pd(OPh)(CO2Ph)L2 is accelerated by increasing steric
hindrance around Pd, we chose bioxazolyl ligands and inves-
tigated the influence of the substituents at the 4,4ꢀ-position
of the 2,2ꢀ-bioxazolyl ligands (3) on the catalytic activity
and selectivity for the oxidative carbonylation. Table 3 sum-
marizes the TON and the selectivity of DPC formation at
100 ◦C for 3 h. Although DPC was produced without addi-
tional ligands, the TON was low and a large amount of phenyl
salicylate was formed (Table 3, entry 1). The addition of 2,2ꢀ-
bioxazolyl ligands greatly improved the catalytic activity and
selectivity, which strongly depended on the structure of the
ligands: Introducing methyl groups at the 4,4ꢀ-position of
3a was ineffective for producing DPC. On the other hand,
the TON of DPC formation significantly increased with in-
creasing bulkiness of the substituents (methyl (3b) < benzyl
(3c) < iso-butyl (3d) < tert-butyl (3e)), and reached 48 mol-
DPC/mol-Pd for the tert-butyl substitution (Table 3, entries
3–6). As expected, introducing bulky substituents is also
effective for promoting the reaction in the 2,2ꢀ-bioxazolyl
ligands.
Fig. 1. 13C NMR spectra of 2b in CD2Cl2: (a) without CO and (b) with CO
(5 MPa) at 25 ◦C.
Interestingly, further substituting H-atoms at the 4,4ꢀ-
position of 3b with methyl groups (3f) also enhanced the
TON and the selectivity (Table 3, entry 7), possibly due to
larger steric effects compared to mono-substitution. We also
examined the carbon-spaced bioxazolyl ligands, in which the
oxazoline rings are bridged with a methylene unit (4) or an
isopropylene unit (5). The methylene-bridged bioxazolyl lig-
and with tert-butyl groups at the 4,4ꢀ-position (4a) gave the
highest TON of DPC (54 mol-DPC/mol-Pd) (Table 3, en-
try 8). In contrast, the bioxazolyl ligand with phenyl groups
(4b) was inefficient. This suggests that introducing bulky, but
electron-withdrawing substituents are ineffective. A similar
tendency was obtained for the isopropylene-bridged bioxa-
zolyl ligands (5a versus 5b). The bioxazolyl ligands (3e, 3f,
and 4a) gave similar TONs and slightly higher selectivities,
compared to the previously reported bulky diimine ligand
6a [18] and the bulky bipyridyl ligand 1a (Table 1, entry
1). It has been assumed that the diimine ligands are effec-
tive for oxidative carbonylation of phenol because conjuga-
tion with aromatic substituents decreases the electron den-
sity of the Pd center, which facilitates the coupling between
CO and phenol on the Pd [18]. However, the present results
show that even nonresonant bioxazolyl ligands are highly
efficient.
appeared at 184.78 ppm in the 13C NMR spectrum for the re-
action of 2b with CO (5 MPa) at 25 ◦C in CD2Cl2 (Fig. 1(b)).
In addition, a signal due to C(CH3) in the tert-butyl groups
(33.24 ppm) splits into two peaks with almost a 1:1 ratio
(34.04 and 33.32 ppm) and the C(CH3) signal (31.13 ppm)
also splits into two close to 1 to 1 (31.21 and 30.85 ppm)
1
(Fig. 1(a) and (b)). In H NMR, the splitting of the sin-
glet due to the tert-butyl groups (1.23 ppm) was observed.
On the other hand, signals due to di(p-tert-butyl)phenyl
1
carbonate were not detected in H and 13C NMR spectra.
These results presumably demonstrate that at 25 ◦C CO read-
ily inserts into one of the two Pd–O bonds in 2b to form
(bpy)Pd(CO2Ar)(OAr), while subsequent reductive elimina-
tion of diaryl carbonate is slow. This is consistent with our
previous findings on (tmeda)Pd(OAr)2 (tmeda = N,N,Nꢀ,Nꢀ-
tetramethylethylenediamine)[15]. Itisnoteworthythattmeda
is an inefficient ligand for the Pd-catalyzed oxidative car-
bonylation of phenol [18].
In contrast, the 1H and 13C NMR spectra for the reaction
of 2a with CO at 25 ◦C revealed the formation of di(p-tert-
butyl)phenyl carbonate. In this case, a signal due to PdCOO
was unobservable in 13C NMR. This clearly demonstrates

152
H. Yasuda et al. / Journal of Molecular Catalysis A: Chemical 236 (2005) 149–155
Table 3
Pd-catalyzed oxidative carbonylation of phenol using 2,2ꢀ-bioxazolyl ligandsa
Entry
1
Ligand
None
TON (DPC)b
TON (PS)c
13
Selectivity of DPC (%)d
53
8
2
3a
20
18
9
68
3
3b
6
75
4
5
3c
26
33
5
6
76
79
3d
6
7
3e
48
44
6
5
85
87
3f
8
9
4a
54
8
8
8
84
66
4b
10
11
12
5a
42
5
6
4
83
54
5b
6a
6b
60
52
9
81
77
13
10
a
Reaction conditions: phenol, 32 mmol; (MeCN)2PdCl2, 0.012 mmol; ligand, 0.012 mmol; Mn(TMHD)3, 0.024 mmol; (Ph3P=)2NBr, 0.24 mmol; CO, 3 MPa;
air, 1.5 MPa; 100 ◦C; 3 h.
b
Turnover number of DPC (mol-DPC/mol-Pd).
Turnover number of phenyl salicylate (mol-PS/mol-Pd).
DPC percentage in the products evaluated from the peak areas on GC and the products identified by GC–MS.
c
d

H. Yasuda et al. / Journal of Molecular Catalysis A: Chemical 236 (2005) 149–155
153
Fig. 2. Effects of the amount of molecular sieve 3A: (a) on TON (᭹) and selectivity of DPC (ꢀ); and (b) on TON of CO2. Reaction conditions: phenol,
32 mmol; (MeCN)2PdCl2, 0.012 mmol; 3f, 0.012 mmol; Mn(TMHD)3, 0.024 mmol; (Ph3P=)2NBr, 0.24 mmol; CO, 3 MPa; air, 1.5 MPa; 100 ◦C; 3 h.
Table 4
2.4. Effects of water
Consideration of the CO2 formation routesa
Entry
Catalyst
Air
Amount of CO2 formation (mmol)
The addition of dehydrating agents is effective for the ox-
idative carbonylation of phenol [3,6,10]. We examined the
effects of adding molecular sieve 3A (MS-3A) to the sys-
tem using 3f. Fig. 2(a) shows the dependences of TON and
selectivity of DPC on the amount of MS-3A added. Upon
the addition of 0.5 g of MS-3A, the TON reached 150 mol-
DPC/mol-Pd (44 mol-DPC/mol-Pd without MS-3A). The se-
lectivity of DPC also improved from 87 to 96%. The reason
the activity is enhanced by adding MS-3A is unclear, but
may be that the hydrolysis of DPC is prevented (Eq. (2)). It
has been mentioned that water reacts with DPC to produce
CO2 and phenol [10]. Another possible reason the activity is
enhanced is that Pd does not have to catalyze the water–gas-
shift reaction between CO and formed water (Eq. (3)), as
discussed below. Further additions of MS-3A decreased the
amounts of DPC formation. Yin et al. hypothesized that
excessive MS-3A increases the viscosity of molten phenol
and decreases the probability of reoxidation of Pd(0) by
cocatalysts [10].
1
2
3
a
ꢁ
–
ꢁ
–
–
ꢁ
8.1 × 10−2
2.9 × 10−3
4.3 × 10−1
Reaction conditions: DMF, 3 ml; water, 0.1 ml; (MeCN)2PdCl2,
0.012 mmol; 3f, 0.012 mmol; Mn(TMHD)3, 0.024 mmol; (Ph3P=)2NBr,
0.24 mmol; CO, 3 MPa; air, 1.5 MPa; 100 ◦C; 3 h.
we investigated the possibility of CO2 formation via Eq.
(3).
When the reaction between CO and water was conducted
under conditions similar to Table 3, the water–gas-shift re-
action proceeded to form CO2 (Table 4, entry 1), while CO2
formation was negligible without metal catalysts (Table 4,
entry 2). These observations suggest that CO2 is formed via
Eq. (3). The presence of air in the above reaction created a
larger amount of CO2 formation, indicating a simultaneous
occurrence of CO oxidation (Table 4, entry 3). Therefore,
we believe that adding MS-3A decreased the amount of CO2
formed not only by suppressing Eq. (2) but also by suppress-
ing Eq. (3).
DPC + H2O → CO2 + 2PhOH
CO + H2O → CO2 + H2
(2)
(3)
Notably, adding MS-3A remarkably prevented the for-
mation of CO2 (Fig. 2(b)). The TON of CO2 was 56 mol-
CO2/mol-Pd without MS-3A, whereas the TONs decreased
to around 7 mol-CO2/mol-Pd with MS-3A. Therefore, it is
considered that water is involved in CO2 formation. Pos-
sible CO2 formation routes in which water takes part are
Eqs. (2) and (3). The oxidative carbonylation of phenol
was conducted in the presence of H218O. Indeed, we con-
firmed the formation of CO18O; the gas phase analyzed by
GC–MS had CO2 peaks from two parent peaks with mass
numbers of 44 and 46, and an intensity ratio of 2:1. This
demonstrates that Eq. (2) or (3) is involved. Since the oc-
currence of Eq. (2) has already been suggested [10,13],
3. Conclusion
We demonstrated that the reductive elimination rate of
diaryl carbonate is much faster using 2a than 2b in the
presence of highly pressurized CO. We also revealed that
2,2ꢀ-bioxazolyl with a bulky substituent at the 4,4ꢀ-position
is an efficient ligand for Pd-catalyzed oxidative carbony-
lation of phenol. Increasing the steric hindrance around
Pd effectively promoted the reaction. In the oxidative car-
bonylation of phenol, CO2 is probably formed via the re-
action between CO and the formed water (water–gas-shift
reaction), in addition to the hydrolysis of DPC and CO
oxidation.

154
H. Yasuda et al. / Journal of Molecular Catalysis A: Chemical 236 (2005) 149–155
4. Experimental
4.1. General
Diaryloxy Pd complex 2b was analogously prepared by
reacting Pd(OAc)2 with sodium p-tert-butylphenoxide in the
presence of 2a to give a yellowpowder in58% yield. 1HNMR
(400 MHz, CD2Cl2, 25 ◦C, ppm): δ 8.76 (d, J = 6.0 Hz, 2H,
H6), 8.08–7.97 (m, 4H, H3, H4), 7.57 (t, J = 6.8 Hz, 2H, H5),
7.14–7.00 (m, 8H, Hm, Ho), 1.23 (s, 18H, C4H9). 13C{1H}
NMR (100.4 MHz, CD2Cl2, 25 ◦C, ppm): δ 164.89, 155.10,
149.29, 139.39, 136.92, 126.34, 124.97, 121.52, 118.44,
33.24 (C(CH3)), 31.13 (C(CH3)). Analytically calculated for
C30H34N2O2Pd: C, 64.22; H, 6.11; N, 4.99. Found: C, 64.35;
H, 6.22; N, 4.73.
Phenol was purified by recrystallization from a petroleum
ether solution. Solvents were dried in a typical manner and
were distilled prior to use. Molecular sieve 3A (MS-3A) was
purchased from Aldrich and activated prior to use by heat-
ing to 300 ◦C in a vacuum for 3 h. All other reagents were
used without further purification. 2,2ꢀ-Bipyridyl ligands 1a,
1b, and 1c, and 2,2ꢀ-bioxazolyl ligands 3c, 4a, 4b, 5a, and
5b were used as received from Aldrich. Carbon monoxide
(Takachiho Chemical Industrial Co., >99.95% purity) and air
(Nippon Sanso Co., G1 grade) were also used without further
purification. 2,2ꢀ-Bioxazolyl ligands 3a [21], 3d [22], 3e [23],
and 3f [24], and diimine ligands 6a [25] and 6b [26] were
prepared according to the literature. (MeCN)2PdCl2 was
prepared by reacting PdCl2 with acetonitrile. (Ph3P=)2NBr
(bis(triphenylphosphoranylidene)ammonium bromide) was
prepared by anion-exchange of (Ph3P=)2NCl [27]. Di(p-
tert-butyl)phenyl carbonate was prepared by reacting p-tert-
butylphenol with 1,1ꢀ-carbonyldiimidazole [28]. The 1H and
13C{1H} NMR spectra were recorded on a JEOL LA400WB
4.3. Preparation of 3b
(S,S)-4,4ꢀ-Dimethyl-4,4ꢀ,5,5ꢀ-tetrahydro-2,2ꢀ-bioxazolyl
ligand 3b was prepared by modifying the procedure reported
for 3a [21]. IR (KBr, cm−1): 1619 (νC N), 1145 (νC–O). 1H
NMR (400 MHz, CDCl3, 25 ◦C, ppm): δ 4.50 (t, J = 8.5 Hz,
2H, OCHHCH(CH3)), 4.41–4.32 (m, 2H, OCH2CH(CH3)),
3.95 (t, J = 8.5 Hz, 2H, OCHHCH(CH3)), 1.32 (d, J = 6.8 Hz,
6H, OCH2CH(CH3)). 13C{1H} NMR (100.4 MHz, CDCl3,
25 ◦C, ppm): δ 154.59 (OC N), 74.77 (OCH2CHCH3),
62.51 (OCH2CHCH3), 20.87 (OCH2CHCH3). Analytically
calculated for C8H12N2O2: C, 57.12; H, 7.19; N, 16.66.
Found: C, 56.48; H, 7.04; N, 16.22.
1
spectrometer (400 MHz for H). IR spectra were measured
on a JASCO FT/IR-610 spectrometer. The elemental analyses
were conducted using a CE Instruments EA1110 elemental
analyzer. The reaction products were analyzed by a Shimadzu
GC-14A gas chromatograph (GC) equipped with a flame ion-
ization detector (FID) using a capillary column (J&W Sci-
entific, DB-1, 30 m) and by a Shimadzu GC-14B GC with a
thermal conductivity detector (TCD) using a packed column
(GL Sciences Inc., Unibeads C, 3 m). The products were fur-
ther identified using a Shimadzu GC-17A GC connected to a
QP-5000 mass spectrometer (GC–MS) (70 eV EI).
4.4. Oxidative carbonylation of phenol
The oxidative carbonylation of phenol was conducted
with a high-pressure-resistant glass vessel (15 ml). Phenol
(3.0 g, 32 mmol), (MeCN)2PdCl2 (3.1 mg, 0.012 mmol), a
ligand (0.012 mmol), Mn(TMHD)3 (14.5 mg, 0.024 mmol),
(Ph3P=)2NBr 148.3 mg (0.24 mmol), and o-terphenyl
(50 mg, internal standard for GC analysis) were charged to
the vessel. The vessel was then heated to 100 ◦C using an
oil bath and charged with CO (3 MPa) and air (1.5 MPa).
After stirring the mixture at 100 ◦C for 3 h, the vessel was
cooled to room temperature and the gaseous contents were
vented. The resulting liquid products were diluted with ace-
tone and analyzed using FID–GC and GC–MS. The same
procedure was used for reactions with MS-3A except a stain-
lesssteelautoclave(30 ml)equippedwithamechanicalstirrer
wasemployedandphenol, (MeCN)2PdCl2, 3f, Mn(TMHD)3,
(Ph3P=)2NBr, o-terphenyl, and the activated MS-3A were
charged to the autoclave in a glove box under a nitrogen at-
mosphere.
4.2. Preparations of 2a and 2b
Diaryloxy Pd complex 2a was prepared according to the
literature [20]. CH2Cl2 (100 ml), 1a (0.82 g, 4.45 mmol), and
sodium p-tert-butylphenoxide (9.35 mmol, 0.94 M in THF
solution) were added to a Schlenk tube containing Pd(OAc)2
(1.00 g, 4.45 mmol). After stirring for 12 h at room tem-
perature, the reaction mixture was fully evaporated to dry-
ness. The crude product was extracted with CH2Cl2, filtered
through Celite, and fully evaporated to dryness. The resulting
product was repeatedly washed with pentane and dried under
a vacuum. Recrystallization from a CH2Cl2–ether solution
gave an ocher crystalline solid of 2a (1.72 g, 66% yield).
1H NMR (400 MHz, CD2Cl2, 25 ◦C, ppm): δ 7.89–7.81 (m,
4H, H3, H4), 7.25 (d, 2H, J = 7.5 Hz, H5), 6.97–6.93 (m, 8H,
Hm, Ho), 2.87 (s, 6H, CH3), 1.22 (s, 18H, C4H9). 13C{1H}
NMR (100.4 MHz, CD2Cl2, 25 ◦C, ppm): δ 164.55, 164.30,
156.24, 138.70, 127.16, 125.86, 124.70, 118.77, 118.54,
33.21 (C(CH3)3), 31.13 (C(CH3)3), 23.72 (CH3). Analyti-
cally calculated for C32H38N2O2Pd: C, 65.24; H, 6.50; N,
4.76. Found: C, 64.94; H, 6.47; N, 4.66.
4.5. Reactions of 2a and 2b with CO
In a glove box, 2a (30 mg, 0.051 mmol) and THF (5 ml)
were charged to a high-pressure-resistant glass vessel (15 ml)
under a nitrogen atmosphere. After the vessel was sealed and
removed from the glove box, the vessel was pressurized with
CO (5 MPa). After stirring the mixture at 25 ◦C for 0.5 h,
CO was released and o-terphenyl (50 mg) was added to the
reaction mixture as an internal standard for GC analysis. The

H. Yasuda et al. / Journal of Molecular Catalysis A: Chemical 236 (2005) 149–155
155
resulting products were analyzed by FID–GC and GC–MS.
In a separate experiment, the reaction time was prolonged to
3 h. The reaction of 2b with CO was conducted in a similar
manner.
High-pressure NMR spectra for the reaction of 2a with
CO were recorded using a sapphire NMR tube (10 mm in
diameter). In the NMR tube, 2a (100 mg, 0.17 mmol) and
CD2Cl2 (2 ml) were charged under a nitrogen atmosphere.
The NMR tube was then pressurized with CO (5 MPa) and
the 1H and 13C NMR spectra were recorded at 25 ◦C. A sim-
ilar measurement was also performed for the reaction of 2b
with CO.
[5] M. Takagi, H. Miyagi, T. Yoneyama, Y. Ohgomori, J. Mol. Catal. A
129 (1998) L1.
[6] M. Goyal, R. Nagahata, J. Sugiyama, M. Asai, M. Ueda, K.
Takeuchi, J. Mol. Catal. A 137 (1999) 147.
[7] H. Ishii, M. Goyal, M. Ueda, K. Takeuchi, M. Asai, J. Mol. Catal.
A 148 (1999) 289.
[8] A. Vavasori, L. Toniolo, J. Mol. Catal. A 149 (1999) 321.
[9] H.Y. Song, E.D. Park, J.S. Lee, J. Mol. Catal. A 154 (2000)
243.
[10] G. Yin, C. Jia, T. Kitamura, T. Yamaji, Y. Fujiwara, J. Organomet.
Chem. 630 (2001) 11.
[11] K.J.L. Linsen, J. Libens, P.A. Jacobs, Chem. Commun. (2002)
2728.
[12] M. Higashijima, K. Nagayama, Stud. Surf. Sci. Catal. 145 (2003)
551.
[13] K. Okuyama, J. Sugiyama, R. Nagahata, M. Asai, M. Ueda, K.
Takeuchi, J. Mol. Catal. A 203 (2003) 21.
4.6. Reaction between CO and water
[14] I. Hatanaka, N. Mitsuyasu, G. Yin, Y. Fujiwara, T. Kitamura, K.
Kusakabe, T. Yamaji, J. Organomet. Chem. 674 (2003) 96.
[15] H. Yasuda, J.-C. Choi, S.-C. Lee, T. Sakakura, Organometallics 21
(2002) 1216.
[16] H. Yasuda, N. Maki, J.-C. Choi, T. Sakakura, J. Organomet. Chem.
682 (2003) 66.
[17] A. Vavasori, L. Toniolo, J. Mol. Catal. A 151 (2000) 37.
[18] H. Ishii, M. Goyal, M. Ueda, K. Takeuchi, M. Asai, Catal. Lett. 65
(2000) 57.
[19] H. Ishii, M. Goyal, M. Ueda, K. Takeuchi, M. Asai, Appl. Catal. A
201 (2000) 101.
A high-pressure-resistant glass vessel (15 ml) contain-
ing DMF (3 ml), water (0.1 ml), (MeCN)2PdCl2 (3.1 mg,
0.012 mmol), 3f (0.012 mmol), Mn(TMHD)3 (14.5 mg,
0.024 mmol), and (Ph3P=)2NBr 148.3 mg (0.24 mmol) was
heated to 100 ◦C and then pressurized with CO (3 MPa) and
air (1.5 MPa). After stirring the mixture at 100 ◦C for 3 h. The
vessel was cooled to room temperature and the gaseous con-
tents were collected in a gas-sampling bag and analyzed by
TCD–GC. Similar reactions were conducted without adding
air and without adding (MeCN)2PdCl2, 3f, Mn(TMHD)3,
(Ph3P=)2NBr, and air.
[20] G.M. Kapteijn, D.M. Grove, H. Kooijman, W.J.J. Smeets, A.L. Spek,
G. van Koten, Inorg. Chem. 35 (1996) 526.
[21] M.G. Burnett, V. McKee, S.M. Nelson, J. Chem. Soc., Dalton Trans.
(1981) 1492.
[22] S.E. Denmark, R.A. Stavenger, A.-M. Faucher, J.P. Edwards, J. Org.
Chem. 62 (1997) 3375.
[23] D. Mu¨ller, G. Umbricht, B. Weber, A. Pfaltz, Helv. Chim. Acta 74
(1991) 232.
References
[1] Y. Ono, Pure Appl. Chem. 68 (1996) 367.
[2] W.B. Kim, U.A. Joshi, J.S. Lee, Ind. Eng. Chem. Res. 43 (2004)
1897.
[24] I. Butula, G. Karlovic´, Liebig Ann. Chem. (1976) 1455.
[25] T.V. Laine, M. Klinga, A. Maaninen, E. Aitola, M. Leskela¨, Acta
Chem. Scand. 53 (1999) 968.
[3] J.E. Hallgren, G.M. Lucas, R.O. Matthews, J. Organomet. Chem.
204 (1981) 135.
[26] H. tom Dieck, M. Svoboda, T. Greiser, Z. Naturforsch. 36b (1981)
823.
[4] I.I. Moiseev, M.N. Vargaftik, T.V. Chernysheva, T.A. Stromnova,
A.E. Gekhman, G.A. Tsirkov, A.M. Makhlina, J. Mol. Catal. A 108
(1996) 77.
[27] A. Martinsen, J. Songstad, Acta Chem. Scand.
645.
[28] H.A. Staab, Liebigs Ann. Chem. 609 (1957) 75.
A 31 (1977)