Direct observation of C-O reductive elimination from palladium aryl alkoxide complexes to form aryl ethers

Article abstract of DOI:10.1021/ja963324p

Reaction of KOCH₂CMe₃ with [(R)-Tol-BINAP]Pd(p-C₆H₄CN)Br formed [(R)-Tol-BINAP]Pd(p-C₆H₄-CN)(OCH₂CMe₃) (5) in quantitative yield (¹H NMR spectroscopy). Thermolysis of 5 in THF-d₈ at 47 °C led to C-O reductive elimination with formation of p-neopentoxybenzonitrile (85 ± 2%). A secondary P-C bond-cleavage process led to formation of 4,4'-dimethylbiphenyl (16 ± 2%). Kinetic analysis of the decomposition of 5 at 47 °C in the presence of excess potassium neopentoxide established the two-term rate law, rate = k[5] + k'[5][KOCH₂-CMe₃], where k = 1.50 ± 0.07 x 10^-3 s^-1 and k' = 6.2 ± 0.4 x 10^-3 s^-1 M^-1, consistent with reductive elimination via competing alkoxide-dependent and alkoxide-independent pathways. In addition, excess KOCH₂CMe₃ exchanged rapidly with the palladium-bound alkoxide ligand of 5 at 47 °C according to the rate law: rate exchange = k(ex)[5][KOCH₂CMe₃], where k(ex) = 1.0 ± 0.1 x 10² s^-1 M^-1. Thermolysis of the related palladium p-cyanophenyl alkoxide complexes (P- P)Pd(p-C₆H₄CN)(OR) [P-P = (S)-BINAP, R = CH₂CMe₃; P-P = (R)-Tol-BINAP, R = CHMe₂, CMe₃; P-P = dppf, R = CH₂CMe₃, CMe₃] and (dppf)Pd[o- C₆H₄(CH₂)₂C(Me)₂O] led to aryl ether formation in 46-91% yield.

Full text of DOI:10.1021/ja963324p

J. Am. Chem. Soc. 1997, 119, 6787-6795
6787
Direct Observation of C-O Reductive Elimination from
Palladium Aryl Alkoxide Complexes To Form Aryl Ethers
Ross A. Widenhoefer, H. Annita Zhong, and Stephen L. Buchwald*
Contribution from the Department of Chemistry, Massachusetts Institute of Technology,
Cambridge, Massachusetts 02139
ReceiVed September 23, 1996X
Abstract: Reaction of KOCH2CMe3 with [(R)-Tol-BINAP]Pd(p-C6H4CN)Br formed [(R)-Tol-BINAP]Pd(p-C6H4-
CN)(OCH2CMe3) (5) in quantitative yield (1H NMR spectroscopy). Thermolysis of 5 in THF-d8 at 47 °C led to
C-O reductive elimination with formation of p-neopentoxybenzonitrile (85 ( 2%). A secondary P-C bond-cleavage
process led to formation of 4,4′-dimethylbiphenyl (16 ( 2%). Kinetic analysis of the decomposition of 5 at 47 °C
in the presence of excess potassium neopentoxide established the two-term rate law, rate ) k[5] + k′[5][KOCH2-
CMe3], where k ) 1.50 ( 0.07 × 10-3 s-1 and k′ ) 6.2 ( 0.4 × 10-3 s-1 M-1, consistent with reductive elimination
via competing alkoxide-dependent and alkoxide-independent pathways. In addition, excess KOCH2CMe3 exchanged
rapidly with the palladium-bound alkoxide ligand of 5 at 47 °C according to the rate law: rate exchange )
kex[5][KOCH2CMe3], where kex ) 1.0 ( 0.1 × 102 s-1 M-1. Thermolysis of the related palladium p-cyanophenyl
alkoxide complexes (P-P)Pd(p-C6H4CN)(OR) [P-P ) (S)-BINAP, R ) CH2CMe3; P-P ) (R)-Tol-BINAP, R )
CHMe2, CMe3; P-P ) dppf, R ) CH2CMe3, CMe3] and (dppf)Pd[o-C6H4(CH2)2C(Me)2O] led to aryl ether formation
in 46-91% yield.
Introduction
Scheme 1
Reductive elimination from a low-valent group 10 metal
center to form an (aryl)C-C bond represents the key bond-
forming step in a variety of synthetically relevant catalytic cross-
coupling protocols.1 As a result, the mechanisms of C-C
reductive elimination from well-defined group 10 metal com-
plexes have been intensely investigated.2 Similarly, C-N and
C-S reductive elimination presumably serve as the key bond-
forming steps in the corresponding palladium-catalyzed cross-
coupling protocols,3,4 and reductive elimination from well-
characterized group 10 metal aryl amido5 and aryl sulfido6
complexes to form arylamines and aryl sulfides, respectively,
X Abstract published in AdVance ACS Abstracts, July 1, 1997.
(1) (a) Stille, J. K. Angew. Chem., Int. Ed. Engl. 1986, 25, 508. (b)
Miyaura, N.; Suzuki, A. Chem. ReV. 1995, 95, 2457. (c) de Meijere, A.;
Meyer, F. E. Angew. Chem., Int. Ed. Engl. 1994, 33, 2379.
(2) (a) Collman, J. P.; Hegedus, L. S.; Norton, J. R.; Finke, R. G. In
Principles and Applications of Organotransition Metal Chemistry; Univer-
sity Science: Mill Valley, CA, 1987; p 322. (b) Stille, J. K. In The Chemistry
of the Metal-Carbon Bond, Vol. II. The Nature and CleaVage of the Metal-
Carbon Bond; Hartley, F. R., Patai, S., Eds.; Wiley: New York, 1985; pp
625-787.
(3) (a) Kosugi, M.; Ogata, T.; Terada. M.; Sano, H.; Migita, T. Bull
Chem. Soc. Jpn. 1985, 58, 3657. (b) Takagi, K. Chem. Lett. 1987, 2221.
(c) Dickens, M. J.; Gilday, J. P.; Mowlem T. J.; Widdowson, D. A.
Tetrahedron 1991, 47, 8621. (d) Carpita, A.; Rossi, R.; Scamuzzi, B.;
Tetrahedon Lett. 1989, 30, 2699. (e) Cristau, H. J.; Chabaud, B.; Chene,
A.; Christol, H. Synthesis 1981, 892.
(4) (a) Kosugi, M.; Kameyama, M.; Migita, T. Chem. Lett. 1983, 927.
(b) Guram, A. S.; Buchwald, S. L. J. Am. Chem. Soc. 1994, 116, 7901. (c)
Guram, A. S.; Rennels, R. A.; Buchwald, S. L. Angew. Chem., Int. Ed.
Engl. 1995, 34, 1348. (d) Wolfe, J. P.; Buchwald, S. L. J. Org. Chem.
1996, 61, 1133. (e) Wolfe, J. P.; Rennels, R. A.; Buchwald, S. L.
Tetrahedron 1996, 52, 7525. (f) Wolfe, J. P.; Wagaw, S.; Buchwald, S. L.
J. Am. Chem. Soc. 1996, 118, 7215. (g) Driver, M. S.; Hartwig, J. F. J.
Am. Chem. Soc. 1996, 118, 7217. (h) Wagaw, S.; Buchwald, S. L. J. Org.
Chem. 1996, 61, 7240. (i) Paul, F.; Patt, J.; Hartwig, J. F. J. Am. Chem.
Soc. 1994, 116, 5969. (j) Louie, J.; Hartwig, J. F. Tetrahedron Lett. 1995,
36, 3609.
has been observed. However, despite numerous examples of
group 10 metal aryl alkoxide complexes, direct thermal reductive
elimination to form an (aryl)C-O bond has not been observed.7,8
We have recently developed a palladium-catalyzed procedure
for the formation of aryl ethers from aryl bromides and alcohols
that employed bulky bidentate phosphine ligands such as 1,1′-
bis(diphenylphosphino)ferrocene (dppf), 2,2′-bis(di-p-tolylphos-
phino)-1,1′-binaphthyl [Tol-BINAP], or 2,2′-bis(diphenylphos-
phino)-1,1′-binaphthyl [BINAP] in the presence of a base
(Scheme 1). For example, treatment of (2-bromophenyl)-2-
methyl-2-butanol with a catalytic amount of Pd(OAc)2 and dppf
and a stoichiometric quantity of sodium tert-butoxide led to ring
closure and formation of 2,2-dimethylchroman in 69% yield
(Scheme 1).9 Likewise, cross-coupling of 4-bromobenzonitrile
and methanol in the presence of excess sodium hydride was
efficiently catalyzed by a mixture of Pd2(DBA)3 and Tol-
BINAP, forming p-methoxybenzonitrile in 81% yield (Scheme
1).10 Significantly, these systems appeared to provide an
opportunity to observe (aryl)C-O reductive elimination from
a group 10 metal center. Here we report the generation of
(7) Bryndza, H. E.; Tam, W. Chem. ReV. 1988, 88, 1163.
(8) After this paper was submitted, an example of aryl(C)-O reductive
elimination from a palladium (aryl)tert-butoxide complex was reported by
Hartwig: Mann, G.; Hartwig, J. F. J. Am. Chem. Soc. 1996, 118, 13109.
(9) Palucki, M.; Wolfe, J. P.; Buchwald, S. L. J. Am. Chem. Soc. 1996,
118, 10333.
(5) (a) Driver, M. S.; Hartwig, J. F. J. Am. Chem. Soc. 1996, 118, 4206.
(b) Villanueva, L. A.; Abboud, K. A.; Boncella, J. M. Organometallics
1994, 13, 3921. (c) Koo, K.; Hillhouse, G. L. Organometallics 1995, 14,
4421. (d) Koo, K.; Hillhouse, G. L. Organometallics 1996, 15, 2669. (e)
Berryhill, S. R.; Price, T.; Rosenblum, M. J. Org. Chem. 1983, 48, 158.
(6) Ba˜ranano, D.; Hartwig, J. F. J. Am. Chem. Soc. 1995, 117, 2937.
(10) Palucki, M.; Wolfe, J. P.; Buchwald, S. L. J. Am. Chem. Soc. 1997,
119, 3395.
S0002-7863(96)03324-0 CCC: $14.00 © 1997 American Chemical Society

6788 J. Am. Chem. Soc., Vol. 119, No. 29, 1997
Widenhoefer et al.
Scheme 2
Scheme 4
Scheme 5
Scheme 3
by H and 31P NMR spectroscopy without isolation. The H
NMR spectrum of 5 in THF-d8 displayed a 1:1:1:1 ratio of
p-tolyl peaks at δ 2.38, 2.19, 1.98, and 1.93 and a single tert-
butyl resonance at δ 0.17; the ratio of these resonances
established the 1:1 ratio of neopentoxide ligands to [Tol-BINAP]-
PdAr groups. A pair of doublets at δ 2.76 and 2.62 (J ≈ 9 Hz)
assigned to the diastereotopic methylene protons of the alkoxide
ligand confirmed binding of the alkoxide to the chiral metal
fragment. The 31P NMR spectrum of 5 displayed two doublets
at δ 29.3 and 14.1 (JPP ) ∼37 Hz), which established bidentate
coordination of the phosphine ligand to the palladium alkoxide
fragment.12
1
1
thermally unstable palladium aryl alkoxide complexes that
undergo reductive elimination to form aryl ethers.
Results
Formation of Palladium Aryl Alkoxide Complexes. Our
approach to generate palladium aryl alkoxide complexes with
bidentate phosphine ligands involved metathesis of a palladium
p-cyanophenyl halide complex with potassium neopentoxide.
Neopentoxide was initially employed due to its diagnostic
signals in the 1H NMR spectrum and because neopentanol
coupled efficiently with 4-bromobenzonitrile under catalytic
conditions.11 The requisite palladium p-cyanophenyl halide
complexes [(R)-Tol-BINAP]Pd(p-C6H4CN)(Br) (1), [(S)-Tol-
BINAP]Pd(p-C6H4CN)(I) (2), [(S)-BINAP]Pd(p-C6H4CN)(Br)
(3), and (dppf)Pd(p-C6H4CN)(Br) (4) were prepared in good
yield (>75%) from reaction of the appropriate combination of
ligand and palladium tri-o-tolylphosphine dimer {Pd[P(o-tolyl)3]-
(p-C6H4CN)(µ-X)}2 (X ) Br, I) (Scheme 2). Complexes 1-4
were characterized by standard spectroscopic techniques and
elemental analysis.
Treatment of a pale yellow solution of 1 in THF-d8 with a
slight excess (∼1.1 equiv) of potassium neopentoxide rapidly
formed an orange solution of the palladium neopentoxide
complex [(R)-Tol-BINAP]Pd(p-C6H4CN)(OCH2CMe3) (5) in
quantitative yield by 1H NMR spectroscopy (Scheme 3).
Likewise, reaction of KOCH2CMe3 with iodide precursor 2 also
generated 5 as the exclusive palladium species. Solutions of 5
darkened within minutes at room temperature, and attempts to
isolate 5 from the corresponding preparative-scale reaction were
unsuccessful. As a result, alkoxide complex 5 was characterized
The related palladium p-cyanophenyl alkoxide complexes (P-
P)Pd(p-C6H4CN)(OR) [P-P ) (S)-BINAP, R ) CH2CMe3 (6);
P-P ) (R)-Tol-BINAP, R ) CHMe2 (7), CMe3 (8), p-C6H4-
Me (9); P-P ) dppf, R ) CH2CMe3 (10), CMe3 (11)] were
formed by analogous procedures. In each case, the desired
palladium alkoxide complex was formed as the exclusive
1
palladium species by H and 31P NMR spectroscopy and was
characterized without isolation. In addition, the oxapalladacycle
(dppf)Pd[o-C6H4(CH2)2C(Me)2O] (12) was isolated free from
KBr and excess alkoxide in 77% yield from treatment of the
aryl bromide complex (dppf)Pd[o-C6H4(CH2)2C(Me)2OH]Br
with KH in THF at room temperature (Scheme 4). Palladacycle
12 was characterized by spectroscopy and elemental analysis.
1
For example, the H NMR spectrum of 12 displayed a pair of
singlets at δ 0.55 and 0.07 assigned to the diastereotopic methyl
resonances, which established alkoxide coordination to pal-
ladium.
(12) Formation of a stable five-coordinate anionic palladium complex
possessing both an alkoxide ligand and a halide ligand appears highly
unlikely. For example, treatment of a 1:1 mixture of 1 and 2 with 1 equiv
of KOCH2CMe3 generated a single species (5) as determined by 1H and
31P NMR spectroscopy at -52 °C.
(11) Palucki, M.; Buchwald, S. L. Unpublished results.

Direct ObserVation of C-O ReductiVe Elimination
J. Am. Chem. Soc., Vol. 119, No. 29, 1997 6789
Figure 2. Eyring plot for the thermolysis of 5 in THF-d8 over the
temperature range 23-57 °C.
Figure 1. Representative first-order plots for disappearance of 5 in
THF-d8 at 23 (3), 37 (4), 47 (O), and 55 °C (×).
exchange,15 or P-C bond hydrogenolysis were observed.
Table 1. First-Order Rate Constants and Aryl Ether Yield for
Thermal Decomposition of Complexes 5-12 ([Pd]0 ≈ 1.5 × 10-2
M) in THF-d8
Likewise, no significant resonances corresponding to free or
1
ligated Tol-BINAP were observed by H or 31P NMR spec-
troscopy of the final reaction mixture.16 Observed rate constants
for disappearance of 5 were also obtained at temperatures
ranging from 23 to 57 °C in THF-d8 (Figure 1, Table 1).13 An
Eyring plot of the data provided the activation parameters: ∆Hq
) 19.8 ( 0.8 kcal mol-1; ∆Sq ) -9.3 ( 0.3 eu (Figure 2).
Thermolysis of 5 in the presence of PPh3 (0.15 M) as a
trapping agent led to no significant change in rate of decom-
position or yield of p-neopentoxybenzonitrile (Table 1) but
eliminated formation of 4,4′-dimethylbiphenyl. The failure of
PPh3 to enhance the yield of aryl ether formation is in contrast
to C-S reductive elimination from related (dppe)Pd(aryl)SCMe3
complexes [dppe ) (diphenylphosphino)ethane], which required
the presence of a trapping agent to generate high yields of
thioether.6 In contrast, thermolysis of 5 in the presence of
KOCH2CMe3 increased both the rate of decomposition and the
yield of p-neopentoxybenzonitrile (Table 1). In order to
determine the rate dependence on neopentoxide concentration,
observed rate constants for decomposition of 5 were measured
as a function of KOCH2CMe3 concentration from 0.0017 to 0.30
M at 47 °C in THF-d8 (Table 1). A plot of kobs versus alkoxide
concentration was linear with a significant positive intercept of
the ordinate, which established the two-term rate law shown in
eq 1, where k ) 1.50 ( 0.07 × 10-3 s-1 (∆Gq ) 22.9 ( 0.1
kcal mol-1) and k′ ) 6.2 ( 0.4 × 10-3 s-1 M-1 (∆Gq ) 22.0
( 0.1 kcal mol-1) (Figure 3). It is noteworthy that the rate of
decomposition of 5 in THF-d8 was not significantly accelerated
(e10%) in the presence of potassium salts such KOTf (98-
180 µmol/mL) (Table 1).
compd
T (°C)
[KOR] (mM)
104kobs (s-1
)
ArOR (%)
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
6
7
8
9
10
11
12
23
23
35
37
47
47
47
47
47
47
47
47
47
47
47
52
55
55
57
47
23
55
60
55
55
60
<2
<2
<2
<2
<2
<2
<2
<2
43
110
120
170
200
260
280
<2
<2
<2
<2
<2
<2
<2
<2
<2
<2
<2
1.43 ( 0.01
1.45 ( 0.02a
5.2 ( 0.2
6.3 ( 0.2
15.3 ( 0.4
15.6 ( 0.2b
16.7 ( 0.3c
16.1 ( 0.3d
18.1 ( 0.6
20.9 ( 0.6
23.0 ( 0.9
23.2 ( 0.3
25.5 ( 0.9
33 ( 1
76
72
85
71
75
72
97
97
94
33 ( 1
30 ( 1
43 ( 2
85
87
86
87
66
61
<5
46
47
91
45 ( 3a
58 ( 2
15.1 ( 0.2
14.1 ( 0.7
0.13 ( 0.01
14.3 ( 0.6
10.3 ( 0.6
11.1 ( 0.6
a Contained PPh3 (0.15 M). b Contained KOTf (98 µmol/mL).
c Contained KOTf (180 µmol/mL). d Contained excess KBr (270 µmol/
mL).
Thermolysis of Palladium Aryl Alkoxide Complexes.
Neopentoxide complex 5 decomposed rapidly at or above room
temperature to form mixtures of aryl ether and biaryl side
products (Scheme 5). For example, thermolysis of a freshly
prepared solution of 5 in THF-d8 at 47 °C led to first-order
d[5]
dt
rate ) −
) k[5] + k′[5][KOCH2CMe3]
(1)
In addition to promoting reductive elimination, excess
KOCH2CMe3 underwent rapid, associative exchange with the
decay to >3 half-lives with an observed rate constant of kobs
)
1.53 × 10-3 s-1 (t1/2 ) 7.5 min) (Figure 1, Table 1).13 1H NMR
and GCMS analysis of the resulting black solution revealed the
formation of p-neopentoxybenzonitrile (85 ( 2%) and 4,4′-
dimethylbiphenyl (16 ( 2%). No significant amounts (<3%)
of products resulting from â-hydrogen elimination,14 P/Pd aryl
(14) (a) Bryndza, H. E.; Calabrese, J. C.; Marsi, M.; Roe, D. C.; Tam,
W.; Bercaw, J. E. J. Am. Chem. Soc. 1986, 108, 4805. (b) Goldman, A. S.;
Halpern, J. J. Am. Chem. Soc. 1987, 109, 7537. (c) Hoffman, D. M.; Lappas,
D.; Wierda, D. A. J. Am. Chem. Soc. 1993, 115, 10538. (d) Blum, O.;
Mielstein, D. Angew. Chem., Int. Ed. Engl. 1995, 34, 229. (e) Bernard, K.
A.; Rees, W. M.; Atwood, J. D. Organometallics 1986, 5, 390.
(15) (a) Morita, D. K.; Stille, J. K.; Norton, J. R. J. Am. Chem. Soc.
1995, 117, 8576. (b) Kong, K.-C.; Cheng, C.-H. J. Am. Chem. Soc. 1991,
113, 6313. (c) Segelstein, B. E.; Butler, T. W.; Chenard, B. L. J. Org. Chem.
1995, 60, 12. (d) Gillie, A.; Stille, J. K. J. Am. Chem. Soc. 1980, 102,
4933.
(13) (a) Rate measurements employing different batches of 5 and KOCH2-
CMe3 provided values for kobs that differed by <7.5%. (b) These solutions
contained small amounts (<2 mM) of free KOCH2CMe3 and 1 equiv of
KBr. Despite the small concentration of free alkoxide present under these
conditions, the reaction rate is dominated by the first-order pathway k >>
k′[KOCH2CMe3] and kobs ≈ k; the rate of decomposition of 5 generated
with <1 equiv of alkoxide was not significantly different. Thermolysis of
5 in the presence of excess KBr (270 µmol/mL) or thermolysis of 5
generated from iodide precursor 2 led to no significant change in the rate
or efficiency of reductive elimination. However, the effect of 1 equiv KBr
on C-O reductive elimination from 5 is not known.
(16) The 1H NMR spectrum of the reaction mixture revealed a broad
resonance at δ 2.1 possibly corresponding to (R)-Tol-BINAP palladium
complexes, while the 31P NMR spectrum displayed a broad resonance at δ
∼2. No resonances in the region expected for free (R)-Tol-BINAP (δ -16.1)
or Pd[(R)-Tol-BINAP]2 were observed. The 31P NMR resonance for Pd-
[(R)-BINAP]2 is δ ∼25 [Ozawa, F.; Kubo, A.; Hayashi, T. Chem. Lett.
1992, 2177].

6790 J. Am. Chem. Soc., Vol. 119, No. 29, 1997
Widenhoefer et al.
Scheme 6
reaction mixtures generated by thermolysis of 5 or 6. Although
the ultimate fate of palladium in these thermolysis reactions is
unclear, thermal decomposition of group 10 transition metal
phosphine complexes has been shown to generate µ-phosphide
clusters.18
Figure 3. Potassium neopentoxide concentration dependence of the
rate of reductive elimination of 5 in THF-d8 at 47 °C.
In an effort to distinguish between intra- and intermolecular
pathways for the formation of biaryl side products in the
decomposition of complexes 5 and 6, an equimolar mixture of
5 and 6 was thermolyzed and analyzed for cross-over products
(Scheme 6).19 Intramolecular biaryl formation should form
biphenyl and 4,4′-dimethylbiphenyl only, while an intermo-
lecular process is expected to form a statistical 1.0:2.6:1.7
mixture of 4,4′-dimethylbiphenyl, 4-phenyltoluene, and biphe-
nyl. Significantly, thermolysis of a 1.0 ( 0.1:1 mixture of 5
and 6 at 55 °C produced a 1.0:2.5:1.8 mixture of 4,4′-
dimethylbiphenyl, 4-phenyltoluene, and biphenyl in 21% total
yield, consistent with biaryl formation via an intermolecular
process.
Thermal decomposition of isopropoxide complex 7 was
considerably faster than decomposition of 5. Although kinetics
were not performed, the half-life for decomposition of 7 at 23
°C was e5 min, which is g15 times faster than decomposition
of 5 under comparable conditions. Analysis of the resulting
solution revealed formation of p-isopropoxybenzonitrile (66%),
benzonitrile (22%), and 4,4′-dimethylbiphenyl (18%) (Scheme
7, Table 1). The yield of p-isopropoxybenzonitrile formed in
the thermolysis of 7 increased to 81% in the presence of excess
KOCHMe2. Reductive elimination from the tert-butoxide
complex 8 in THF-d8 at 55 °C was approximately three times
slower than reductive elimination from neopentoxide complex
5 with formation of p-tert-butoxybenzonitrile (61%) and 4,4′-
dimethylbiphenyl (26%) (Scheme 7, Table 1). Thermal de-
composition of the p-cresolate complex 9 was >450 times
slower than decomposition of neopentoxide complex 5 and led
to no detectable formation of diaryl ether.
Figure 4. Potassium neopentoxide concentration dependence of the
rate of alkoxide exchange with 5 in THF-d8 at 47 °C.
palladium-bound neopentoxide group of 5. For example, in the
1H NMR spectrum of 5 at 55 °C, the resonances for the
neopentoxide ligand remained sharp with no loss of coupling
between the diastereotopic methylene protons. However, ad-
dition of KOCH2CMe3 led to considerable broadening of these
resonances in the 1H NMR spectrum. Observed rate constants
1
for alkoxide exchange were determined from excess H NMR
line broadening (∆ω1/2 ) k/π)17 of the tert-butyl resonance of
5 as a function of alkoxide concentration from 0.0017 to 0.3 M
KOCH2CMe3 at 47 °C. A plot of kobs versus [KOCH2CMe3]
established the first-order dependence of the rate of exchange
on alkoxide concentration and the second-order rate law shown
in eq 2, where kex ) 1.0 ( 0.1 × 102 s-1 M-1 (∆Gq ) 15.8 (
0.1 kcal mol-1) at 47 °C (Figure 4). The second-order rate
constant for alkoxide exchange (kex) is >5 orders of magnitude
greater than the second-order rate constant for reductive
elimination (k′).
Thermal decomposition of the dppf-ligated palladium aryl
alkoxide complexes was also investigated. For example, thermal
decomposition of 10 at 55 °C in THF-d8 led to first-order decay
at a rate approximately three times slower than decomposition
of 5 under comparable conditions (Table 1). Analysis of the
resulting black solution revealed the formation of p-neopen-
toxybenzonitrile (46%), pivaldehyde (36%), benzonitrile (43%),
and biphenyl (12%) (Table 1, Scheme 8). Thermolysis of tert-
butoxide complex 11 at 55 °C formed p-tert-butoxybenzonitrile
(47%) and biphenyl (12%) (Scheme 7), while thermolysis of
tertiary oxapalladacycle complex 12 at 60 °C in THF-d8
produced 2,2-dimethylchroman in high yield (91%) (Scheme
4).
rate of alkoxide exchange ) kex[5][KOCH2CMe3] (2)
Thermolysis of BINAP-ligated complex 6 at 47 °C in THF-
d8 also led to first-order decomposition at a rate not significantly
different from that of 5, with formation of p-neopentoxyben-
zonitrile (87 ( 2%) and biphenyl (27 ( 2%) (Table 1, Scheme
5). The formation of biphenyl from 6 and 4,4′-dimethylbiphenyl
from 5 strongly implicates the phosphorus-bound aryl groups
as the source of biaryl side products.18 The presence of a
phosphine degradation pathway is also consistent with the
absence of significant resonances corresponding to either free
1
or bound phosphine ligand in the H or 31P NMR of the final
(17) Bovey, F. A. Nuclear Magnetic Resonance Spectroscopy, 2nd ed.;
Academic Press: San Diego, CA, 1988.
Discussion
(18) (a) Garrou, P. E. Chem. ReV. 1985, 85, 171. (b) Coulson, D. R.
Chem. Commun. 1968, 1530. (c) Fahey, D. R.; Mahan, J. E. J. Am. Chem.
Soc. 1976, 98, 4499. (d) Taylor, N. J.; Chieh, P. C.; Carty, A. J. J. Chem.
Soc., Chem. Commun. 1975, 448. (e) Blake, D. M.; Nyman, C. J. J. Am.
Chem. Soc. 1970, 92, 5359. (f) Bellon, P. L.; Ceriotti, A.; Demartin, F.;
Longoni, G.; Heaton, B. T. J. Chem. Soc., Dalton Trans. 1982, 1671.
C-O Bond Formation. Thermolysis of palladium aryl
alkoxide complexes led to C-O reductive elimination and
(19) The S enantiomer [(S)-Tol-BINAP]Pd(p-C6H4CN)OCH2CMe3 gen-
erated from [(S)-Tol-BINAP]Pd(p-C6H4CN)Br was employed in these cross-
over experiments.

Direct ObserVation of C-O ReductiVe Elimination
J. Am. Chem. Soc., Vol. 119, No. 29, 1997 6791
Scheme 7
Scheme 8
palladium concentration and the activation parameters (∆Hq )
19.8 ( 0.8 kcal mol-1; ∆Sq ) -9.3 ( 0.3 eu) for the alkoxide-
independent pathway are consistent with unimolecular reductive
elimination directly from 5 to form p-neopentoxybenzontrile
and Pd[Tol-BINAP].13 The latter species is expected to be
highly reactive28 and presumably decomposes via P-C bond
cleavage to form 4,4′-dimethylbiphenyl (see below).
We considered two plausible mechanisms for the unimolecu-
lar reductive elimination pathway. One pathway is a concerted
process analogous to that proposed for H-H, C-H, and C-C
reductive elimination (Scheme 9, path a).2 Although our kinetics
do not distinguish this pathway from a mechanism initiated by
rapid and reversible dissociation of a single phosphorus center,
C-S,6 C-C,29,30 and C-H31 reductive elimination have been
shown to occur directly from a four-coordinate platinum group
bis(phosphine) complex without prior ligand dissociation. We
also considered a mechanism initiated by inner-sphere nucleo-
philic attack of the alkoxide ligand at the ipso carbon atom of
the palladium-bound aryl group via a Meisenheimer-type species
such as I (Scheme 9, path b).32 This C-O reductive elimination
pathway is analogous to the mechanisms proposed for R-migra-
formation of aryl ethers. Reductive elimination to form an
(aryl)C-O bond is unprecedented, and related C-O reductive
elimination processes have been observed in only two cases.8
For example, thermolysis of the palladium bis(phosphine) acyl
aryloxide complex (PPh3)2Pd(COMe)(OAr) led to C-O reduc-
tive elimination and formation of aryl esters.20 The analogous
nickel complexes underwent C-O reductive elimination upon
addition of π-acids such as acrylonitrile or CO.20 Likewise,
treatment of the nickel aryloxide complex (phen)Ni(O-o-C6H4-
CMe2CH2) (phen ) 1,10-phenanthralene) with I2 formed 4,4-
dimethyl-3,4-dihydrocoumarin in 55% yield.21 C-O reductive
elimination is also implicated in a variety of stoichiometric
transformations,22 as well as nickel-,23 palladium-,24 and copper-
catalyzed25 reactions. Several related transformation such as
C-O bond cleavage26 and O-H reductive elimination27 have
also been observed.
(25) (a) Keegstra, M. A.; Peters, T. H. A.; Brandsma, L.. Tetrahedron
1992, 48, 3633. (b) Aalten, H. L.; van Koten, G.; Grove, D. M.; Kuilman,
T.; Piekstra, O. G.; Hulshof, L. A.; Sheldon, R. A. Tetrahedron 1989, 45,
5565. (c) Capdevielle, P.; Maumy, M. Tetrahedron Lett. 1993, 34, 1007.
(d) Nobel, D. J. Chem. Soc., Chem. Commun. 1993, 419. (e) Pert, D. J.;
Ridley, D. D. J. Chem. 1989, 42, 421. (f) Whitesides, G. M.; Sadowski, J.
S.; Lilburn, J. J. Am. Chem. Soc. 1974, 96, 2829. (g) Lindley, J. Tetrahedron
1984, 40, 1433. (h) Bacon, R. G. R.; Rennison, S. C. J. Chem. Soc. C
1969, 308. (i) Bacon, R. G. R.; Rennison, S. C. J. Chem. Soc. C 1969, 312.
(26) Yamamoto, A. AdV. Organomet. Chem. 1992, 34, 111.
(27) Glueck, D. S.; Winslow, L. J. N.; Bergman, R. G. Organometallics
1991, 10, 1462.
(28) Otsuka, S. J. Organomet. Chem. 1980, 200, 191.
(29) Stang, P. J.; Kowalski, M. H. J. Am. Chem. Soc. 1989, 111, 3356.
(30) Braterman, P. S.; Cross, R. J.; Young, G. B. J. Chem. Soc., Dalton
Trans. 1977, 1892.
(31) (a) Michelin, R. A.; Faglia, S.; Uguagliati, P. Inorg. Chem. 1983,
22, 1831. (b) Abis, L.; Sen, A.; Halpern, J. J. Am. Chem. Soc. 1978, 100,
2915.
The experimental rate law for decomposition of 5 (eq 1) in
the presence of KOCH2CMe3 is consistent with C-O reductive
elimination via competing alkoxide-dependent and alkoxide-
independent pathways. The first-order rate dependence on
(32) A mechanism initiated by intermolecular attack of the Pd-O bond
of 5 at the ipso carbon atom of a second molecule of 5 is not in accord
with the first-order dependence of the reaction rate on [5]. Likewise, a
mechanism initiated by rate-limiting alkoxide dissociation is not in line
with the experimental entropy of activation. However, we cannot distinguish
the pathway shown in Scheme 9, path b, from a mechanism initiated by
rapid and reversible alkoxide dissociation to form [Tol-BINAP]Pd(p-C6H4-
CN)]+ [OCH2CMe3]-, followed by attack of the alkoxide at the ipso carbon
atom.
(33) (a) Dockter, D. W.; Fanwick, P. E.; Kubiak, C. P. J. Am. Chem.
Soc. 1996, 118, 4846. (b) Bryndza, H. E.; Calabrese, J. C.; Wreford, S. S.
Organometallics 1984, 3, 1603. (c) Bryndza, H. E.; Kretchmar, S. A.; Tulip,
T. H. J. Chem. Soc., Chem. Commun. 1985, 977. (d) Michelin, R. A.; Napoli,
M.; Ros, R. J. Organomet. Chem. 1979, 175, 239. (e) Bennett, M. A.;
Yoshida, T. J. Am. Chem. Soc. 1978, 100, 1750. (f) Arnold, D. P.; Bennett,
M. A.; Crisp, G. T.; Jeffery, J. C. AdV. Chem. Ser. 1982, 196, 195. (g)
Bennett, M. A. J. Organomet. Chem. 1986, 300, 7. (h) Bennett, M. A.;
Rokicki, A. Aust. J. Chem. 1985, 38, 1307. (i) Appleton, T. G; Bennett, M.
A. J. Organomet. Chem. 1973, 55, C88. (j) Bennett, M. A. J. Mol. Catal.
1987, 41, 1. (k) Bennett, M. A.; Rokicki, A. J. Organomet. Chem. 1983,
244, C31. (l) Bennett, M. A.; Rokicki, A. J. Organometallics 1985, 4, 180.
(m) Bryndza, H. E. Organometallics 1985, 4, 1686.
(20) Komiya, S.; Akai, Y.; Tanaka, K.; Yamamoto, T.; Yamamoto, A.
Organometallics 1985, 4, 1130.
(21) (a) Koo, K.; Hillhouse, G. L.; Rheingold, A. L. Organometallics
1995, 14, 456. (b) Matsunaga, P. T.; Mavropoulos, J. C.; Hillhouse, G. L.
Polyhedron 1995, 14, 175.
(22) (a) Bernard, K. A.; Churchill, M. R.; Janik, T. S.; Atwood, J. D.
Organometallics 1990, 9, 12. (b) Komiya, S.; Tane-ichi, S.; Yamamoto,
A.; Yamamoto, T. Bull. Chem. Soc. Jpn. 1980, 53, 673. (c) Bryndza, H.
E.; Calabrese, J. C.; Wreford, S. S. Organometallics 1984, 3, 1603.
(23) (a) Cristau, H.-J.; Cesmurs, J. -R. Ind. Chem. Libr. 1995, 7, 240.
(b) Cramer, R.; Coulson, D. R. J. Org. Chem. 1975, 40, 2267.
(24) (a) Kundu, N. G.; Pal, M.; Mahanty, J. S.; Dasgupta, S. K. J. Chem.
Soc., Chem. Commun. 1992, 41. (b) Stanton, S. A.; Felman, S. W.; Parkhurst,
C. S.; Godleski, S. A. J. Am. Chem. Soc. 1983, 105, 1964. (c) Larock, R.
C.; Gou. L. Synlett 1995, 465. (d) Larock, R. C.; Berrios-Pen˜a, N. G.; Fried,
C. A. J. Org. Chem. 1991, 56, 2615. (e) Larock, R. C.; Berrios-Pen˜a, N.
G.; Fried, C. A. Yum, E. K.; Tu, C.; Leong, W. J. Org. Chem. 1993, 58,
4509. (f) Larock, R. C.; Harrison, L. W.; Hsu, M. H. J. Org. Chem. 1984,
49, 3664. (g) Larock, R. C.; Berrios-Pen˜a, N. G.; Narayanan, K.; J. Org.
Chem. 1990, 55, 3447.

6792 J. Am. Chem. Soc., Vol. 119, No. 29, 1997
Widenhoefer et al.
Scheme 9
Scheme 10
tory insertion of both CO33 and isocyanides34 into group 10
metal-alkoxide bonds.7 The high nucleophilicity of late-
transition-metal alkoxide ligands is evidenced by the strong
tendency of the oxygen atom of these complexes to serve as a
hydrogen-bond acceptor in the presence of free alcohols.35
Several experimental observations point to an insertion
mechanism (path b) in preference to a symmetric concerted
pathway (path a). For example, the decreasing rate of C-O
reductive elimination in the order OCHMe2 (7) > OCH2CMe3
(5) > OCMe3 (8) . OC6H4Me (9) appears unusual for concerted
reductive elimination36 but roughly parallels the nucleophilicity
of the respective palladium hydrocarboxide. In addition,
preliminary studies probing the electronic effects indicate that
the rate of C-O reductive elimination increases dramatically
with the decreasing electron density of the palladium-bound aryl
goup.37 The extent of rate acceleration appears greater than
that expected through strictly inductive effects and strongly
suggests the delocalization of negative charge within the aryl
ligand in the transition state for C-O reductive elimination.
The rate of reductive elimination from 5 was accelerated in
the presence of potassium neopentoxide. The absence of
significant rate acceleration in the presence of potassium salts
such as KOTf argues against a medium effect and points to a
mechanism initiated by direct attack of potassium neopentoxide
on 5. One possible mechanism involves rapid and reversible
attack of neopentoxide at palladium to generate the anionic five-
coordinate bis(alkoxide) intermediate {[Tol-BINAP]Pd(p-C6H4-
CN)(OCH2CMe3)2}- (II) (Scheme 10).38 Rate-limiting C-O
reductive elimination from II would then generate p-neopen-
toxybenzonitrile and the three-coordinate palladium alkoxide
fragment {[Tol-BINAP]Pd(OCH2CMe3)}-. Alternately, inner-
sphere (II f III) or outer sphere (5 f III) attack of alkoxide
at the ipso-carbon atom of the palladium-bound aryl group
would form the palladium alkoxide Meisenheimer complex III
(Scheme 10). Collapse of intermediate III via Pd-C bond
cleavage would then generate p-neopentoxybenzonitrile and
{[Tol-BINAP]Pd(OCH2CMe3)}-.
(34) Michelin, R. A.; Ros, R. J. Organomet. Chem. 1979, 169, C42.
(35) (a) Kegley, S. E.; Schaverien, C. J.; Freudenberger, J. H.; Bergman,
R. G.; Nolan, S. P.; Hoff, C. D. J. Am. Chem. Soc. 1987, 109, 6563. (b)
Osakada, K.; Kim, Y.-J.; Yamamoto, A. J. Organomet. Chem. 1990, 382,
303. (c) Kim, Y. J.; Osakada, K.; Takenaka, A.; Yamamoto, A. J. Am.
Chem. Soc. 1990, 112, 1096. (d) Osakada, K.; Kim, Y. J.; Tanaka, M.;
Ishiguro, S.; Yamamoto, A. Inorg. Chem. 1991, 30, 197. (e) Shubina, E.
S.; Belkova, N. V.; Krylov, A. N.; Vorontsov, E. V.; Epstein, L. M.; Gusev,
D. G.; Niedermann, M.; Berke, H. J. Am. Chem. Soc. 1996, 118, 1105.
(36) Although data are limited, existing studies indicate that the rate of
reductive elimination should increase with increasing steric interaction within
the complex (a) Jones, W. D.; Kuykendall, V. L. Inorg. Chem. 1991, 30,
2615. (b) Brown, J. M.; Guiry, P. J. Inorg. Chim. Acta 1994, 220, 249.
(37) Thermolysis of [(R)-Tol-BINAP]Pd(p-C6H4NO2)(OCH2CMe3) [σpara
NO2 ) 0.78] at 25 °C was ˜100 times faster (∆∆Gq ≈ 3.5 kcal mol-1) than
reductive elimination from 5 [σpara CN ) 0.66], forming p-neopentoxyni-
trobenzene in >90% yield. Likewise, thermolysis of [dppf]Pd(p-C6H4-
NO2)(OCH2CMe3) was considerably more facile than decomposition of 7,
The intermediacy of II in the alkoxide-dependent C-O
reductive elimination pathway is supported by the presence of
a facile, associative alkoxide-exchange pathway for 5.37 Like-
wise, both ligand-promoted C-C reductive elimination and
ligand-promoted CO insertion have been observed. For ex-
ample, reductive elimination from the nickel aryl methyl
complexes (dmpe)Ni(Ar)Me [dmpe ) (dimethylphosphino)et-
(38) Presumably, the Tol-BINAP ligand of I will span equatorial and
axial sites in the trigonal bipyramid with a P-M-P bond angle ≈ 90°,
close to the value of (∼92°) observed in square planar palladium-BINAP
complexes [Hayashi, T.; Konishi, M.; Kobori, Y.; Kumada, M.; Higuchi,
T.; Hirotsu, K. J. Am. Chem. Soc. 1984, 106, 158]. Likewise, the palladium-
bound aryl ligand is expected to occupy the second axial position in the
trigonal bipyramid [Rossi, A. R.; Hoffmann, R. Inorg. Chem. 1975, 14,
365].
occurring readily at 25 °C (kobs ) 1.62 ( 0.2 s-1, ∆∆Gq ≈ 2.3 kcal mol-1
)
with formation of p-neopentoxynitrobenzene in 81% yield. In contrast,
extended thermolysis of either [(S)-BINAP]Pd(p-C6H4Cl)(OCH2CMe3) or
(dppf)Pd(p-C6H4Cl)(OCH2CMe3) [σpara Cl ) 0.23] led to slow decomposi-
tion with no detectable formation of p-neopentoxybenzonitrile: Widen-
hoefer, R. A.; Buchwald, S. L. Unpublished observations.

Direct ObserVation of C-O ReductiVe Elimination
J. Am. Chem. Soc., Vol. 119, No. 29, 1997 6793
Scheme 11
hane] was promoted by addition of tertiary phosphines such as
PEt3,39,40 while reductive elimination from the platinum bis-
(phosphine)diaryl complex (PPh3)2Pt(p-C6H4Me)2 was acceler-
ated by addition of excess PPh3.30 In addition, migratory CO
insertion into the methyl-molybdenum bond of the cyclopen-
tadienyl tri(carbonyl) complex (η5-C5H5)Mo(CO)3CH3 to form
the acetyl molybdenum complex (η5-C5H5)Mo(CO)2(PMePh2)-
COCH3 in the presence of PMePh2 in substituted tetrahydrofuran
solvents occurred by competing first- and second-order path-
ways.41
Biaryl Formation. Thermolysis of palladium aryl alkoxide
complexes, which forms aryl ethers, also led to the formation
of varying amounts (12-27%) of biaryl side products. Exclu-
sive formation of 4,4′-dimethylbiphenyl from thermolysis of 5
and exclusive formation of biphenyl from thermolysis of 6
implicates the phosphorus-bound aryl groups as the source of
biaryl, while the formation of a statistical amount of 4-phenyl-
toluene from thermolysis of a mixture of 5 and 6 points to an
intermolecular process for biaryl formation.19 In addition,
several observations indicate that P-C bond cleavage occurs
subsequent to C-O reductive elimination. For example, no
significant quantities of benzonitrile were incorporated into the
biaryl side products. In addition, thermolysis of 5 in the
presence of a trapping agent eliminated formation of 4,4′-
dimethylbiphenyl but affected neither the rate of decomposition
nor yield of aryl ether.
The above observations are in accord with biaryl formation
via P-C oxidative addition18 to the reactive28 14-electron
fragment Pd[P-P] [P-P ) BINAP, Tol-BINAP] generated by
C-O reductive elimination from a palladium (aryl)alkoxide
complex (Scheme 10). In one possible mechanism, intramo-
lecular P-C oxidative addition to Pd[P-P] followed by
dimerization of the resulting mononuclear phosphido fragment
IV would form the bridging phosphide dimer V (path a).
Dinuclear reductive elimination42 directly from V or ligand
rearrangement followed by unimolecular reductive elimination
from VI would then form biaryl. Alternatively, µ-phosphide
dimer V could also be generated via two consecutive intermo-
lecular P-C oxidative additions to Pd[P-P] (path b). However,
it appears somewhat unlikely that Pd[P-P] could attack a second
molecule of Pd[P-P] without also attacking the parent palladium
aryl alkoxide complex prior to C-O reductive elimination,
which was not observed. Phosphine degradation via P-C
oxidative addition has also been observed in the thermal
decomposition of the platinum diaryl bis(phosphine) complex
(PPh3)2Pt(p-C6H4Me)2, which formed 4,4′-dimethylbiphenyl and
biphenyl.43
Conclusions
Thermal decomposition of palladium(aryl)alkoxide complexes
led to C-O reductive elimination with formation of aryl ethers.
Significantly, these transformations represent the first examples
of C-O reductive elimination from group 10 metal (aryl)-
alkoxide complexes.8 Thermolysis of these palladium aryl
alkoxide complexes also formed biaryl side products derived
from the palladium-bound aryl groups formed via intermolecular
decomposition of the 14-electron Pd(0) fragment generated via
initial reductive elimination. Kinetic analysis of the decomposi-
tion of [(R)-Tol-BINAP]Pd(p-C6H4CN)(OCH2CMe3) (5) in the
presence of excess alkoxide established a two-term rate law
consistent with the presence of both an alkoxide-independent
and alkoxide-dependent pathway for C-O reductive elimination.
Significantly, unimolecular C-O reductive elimination appears
to be initiated by inner-sphere nucleophilic attack of the alkoxide
lone pair at the ipso carbon atom of the palladium-bound aryl
group. Likewise, an alkoxide-dependent pathway initiated by
direct attack of KOCH2CMe3 at palladium was supported by
the presence of an associative alkoxide-exchange pathway for
5.
Experimental Section
General Methods. All manipulations and reactions were performed
under an atmosphere of nitrogen or argon in a glovebox or by standard
Schlenk techniques. Preparative-scale reactions were performed in
flame- or oven-dried Schlenk tubes equipped with a stir bar, side arm
joint, and a septum. NMR spectra were obtained in oven-dried 5 mm
thin-walled NMR tubes capped with a rubber septum on a Varian XL-
300 spectrometer at 23 °C unless otherwise noted. Gas chromatography
was performed on a Hewlett-Packard Model 5890 gas chromatograph
using a 25 m polymethylsiloxane capillary column. Elemental analyses
were performed by E+R Microanalytical Laboratories (Corona, NY).
Diethyl ether, hexane, and THF-d8 were distilled from solutions of
sodium/benzophenone ketyl under argon or nitrogen. Pd2(DBA)3, P(o-
tol)3, (R)-Tol-BINAP, (S)-Tol-BINAP, dppf (Strem), (S)-BINAP (Pfiz-
er), 4-bromobenzonitrile, 4-iodobenzonitrile, pivaldehyde, biphenyl,
4,4′-dimethylbiphenyl, p-phenylbenzonitrile, 4-phenyltoluene, and ben-
zonitrile (Aldrich) were used as received. KOTf and KBr (Aldrich)
(39) Komiya, S.; Abe, Y.; Yamamoto, A.; Yamamoto, T. Organome-
tallics 1983, 2, 1466.
(40) Tatsumi, K.; Nakamura, A.; Komiya, S.; Yamamoto, A.; Yamamoto,
T. J. Am. Chem. Soc. 1984, 106, 8181.
(41) Wax, M. J.; Bergman, R. G. J. Am. Chem. Soc. 1981, 103, 7028.
(42) (a) Hersh, W. H.; Bergman, R. G. J. Am. Chem. Soc. 1983, 105,
5846. (b) Hembre, R. T.; Scott, C. P.; Norton, J. R. J. Org. Chem. 1987,
52, 3650. (c) Chetcuti, M. J.; Chisholm M. H.; Folting, K.; Haitko, D. A.;
Huffman, J. C. J. Am. Chem. Soc. 1982, 104, 2138. (d) Halpern, J. Inorg.
Chim. Acta 1982, 62, 31.
(43) Braterman, P. S.; Cross, R. J.; Young, G. B. J. Chem. Soc., Chem.
Commun. 1976, 1310.

6794 J. Am. Chem. Soc., Vol. 119, No. 29, 1997
Widenhoefer et al.
were dried under vacuum at 220 °C prior to use. Potassium alkoxides
were synthesized from reaction of anhydrous alcohol with 1 equiv of
KH in THF.
via syringe to a colorless solution of 1 (12 mg, 1.25 × 10-2 mmol)
and PhSiMe3 (1.75 mg, 1.16 × 10-2 mmol) in THF-d8 (0.70 mL). The
tube was shaken briefly at room temperature to form an orange solution
of 5 in quantitative (98 ( 5%) yield by 1H NMR spectroscopy versus
PhSiMe3 internal standard, along with a small resonance corresponding
to free alkoxide (δ 0.85). 5 was thermally unstable and was analyzed
{Pd[P(o-tolyl)3](p-C4H4CN)(µ-Br)}2. A purple solution of Pd2-
(DBA)3 (1.0 g, 1.1 mmol), P(o-tol)3 (1.3 g, 4.3 mmol), and p-
bromobenzonitrile (2.0 g, 11 mmol) in benzene (60 mL) was stirred at
room temperature for 1 h. The resulting green/brown solution was
filtered through Celite, and benzene was evaporated under vacuum.
The oily residue was dissolved in Et2O (25 mL) and allowed to stand
at room temperature overnight. The resulting yellow precipitate was
filtered, washed with Et2O, and dried under vacuum to give {Pd[P(o-
tolyl)3](p-C4H4CN)(µ-Br)}2 (0.95 g, 75%) as a yellow powder. 1H
NMR (CHCl3, 55 °C): δ 7.33, 7.13, 6.90, 6.73, 2.10. 31P {1H} NMR
(CDCl3, 55 °C): δ ∼22.5 (br s). IR (THF): ν[CtN] 2222 cm-1. Anal.
Calcd for C56H50Br2N2P2Pd2 (found): C, 56.73 (56.57); H, 4.25 (4.51).
{Pd[P(o-tolyl)3](p-C6H4CN)(µ-I)}2. Reaction of Pd2(DBA)3 (1.0
g, 1.1 mmol), P(o-tol)3 (1.3 g, 4.3 mmol), and p-iodobenzonitrile (1.75
g, 8.6 mmol) in benzene (60 mL) employing a procedure analogous to
that used to synthesize {Pd[P(o-tolyl)3](p-C4H4CN)(µ-Br)}2 led to the
isolation of {Pd[P(o-tolyl)3](p-C4H4CN)(µ-I)}2 (1.17 g, 84%) as a
yellow powder. Anal. Calcd for C56H50I2N2P2Pd2 (found): C, 52.57
(52.80); H, 3.94 (4.12).
1
without isolation by H and 31P NMR spectroscopy. 1H NMR (THF-
d8): δ 7.83 (t, J ) 8.4 Hz), 7.78-7.59 (m), 7.51-7.25 (m), 7.12 (t, J
) 7.4 Hz), 7.04-6.93 (m), 6.81 (d, J ) 7.0 Hz), 6.44 (d, J ) 7.4 Hz),
6.29 (d, J ) 7.4 Hz) 2.76 (d, J ) 9.0 Hz, 1 H), 2.62 (d, J ) 8.8 Hz,
1 H), 2.38 (s, 3 H), 2.19 (s, 3 H), 1.98 (s, 3 H), 1.93 (s, 3 H), 0.17 (s,
9 H). 31P{1H}NMR (THF-d8): δ 29.3 (d, J ) 36.6 Hz), 14.1 (d, J )
36.7 Hz). IR (THF): ν[CtN] 2218 cm-1
.
[(S)-BINAP]Pd(p-C6H4CN)(OCH2CMe3) (6). Reaction of KOCH2-
Me3 (1.6 mg, 1.25 × 10-2 mmol) and 3 (11 mg, 1.21 × 10-2 mmol)
using a procedure analogous to that used to form 5 generated a bright
yellow solution of 6. 1H NMR (THF-d8): δ 7.90 (m), 7.69 (m), 7.61
(t, J ) 9.0 Hz), 7.44 (m), 7.14 (d, J ) 8.5 Hz), 6.99 (t, J ) 9.3 Hz),
6.69 (t, J ) 7.3 Hz), 6.52 (br s), 2.79 (d, J ) 8.5 Hz, 1 H), 2.62 (d, J
) 8.5 Hz, 1 H), 0.16 (s, 9 H). 31P {1H} NMR (THF-d8): δ 31.7 (d, J
) 37.2 Hz), 15.5 (d, J ) 37.1 Hz).
[(R)-Tol-BINAP]Pd(p-C6H4CN)(OCHMe2) (7). Reaction of
KOCHMe2 and 1 using a procedure analogous to that used to prepare
5 gave 7 in ∼90% yield along with minor decomposition products and
a small resonance corresponding to free alkoxide. 7 was thermally
unstable and was analyzed without isolation by 1H and 31P NMR
spectroscopy at -15 °C. 1H NMR (THF-d8, -15 °C): δ 8.16 (t, J )
8.81 Hz, 2 H), 7.63 (m, 4 H), 7.55 (d, J ) 7.26 Hz, 2 H), 7.44 (d, J )
10.1 Hz, 2 H), 7.36 (m, 4 H), 7.20 (d, J ) 7.31 Hz, 2 H), 7.10 (m, 4
H), 6.93 (d, J ) 7.46 Hz, 2 H), 6.87 (d, J ) 7.20, 2 H), 6.77 (d, J )
8.32 Hz, 2 H), 6.66 (d, J ) 8.39 Hz, 2 H), 6.37 (br s, 4 H), 3.21 (m,
1 H), 2.36 (s, 3 H), 2.25 (s, 3 H), 1.95 (s, 3 H), 1.94 (s, 3 H), 0.65 (d,
J ) 5.8 Hz, 3 H), 0.63 (d, J ) 5.8 Hz, 3 H). 31P{1H}NMR (THF-d8,
-15 °C): δ 30.2 (d, J ) 36.7 Hz), 14.2 (d, J ) 37.4 Hz).
[(R)-Tol-BINAP]Pd(p-C4H4CN)(Br) (1). A solution of {Pd[P(o-
tolyl)3](p-C4H4CN)(µ-Br)}2 (200 mg, 0.17 mmol) and (R)-Tol-BINAP
(240 mg, 0.35 mmol) in CH2Cl2 (10 mL) was stirred at room
temperature for 5 h and then evaporated under vacuum. The oily
residue was dissolved in Et2O (10 mL) and allowed to stand at room
temperature for 4 h. The resulting precipitate was filtered, washed with
Et2O, and dried under vacuum to give 1 (308 mg, 94%) as a cream-
colored solid that contained traces of ether (<5%) by 1H NMR analysis.
1H NMR (THF-d8): δ 8.26 (dd, J ) 8.7, 10.5 Hz), 8.04 (t, J ) 8.2
Hz), 7.98 (t, J ) 9.3 Hz), 7.91 (q, J ) 7.7 Hz), 7.77 - 7.60 (m, 7 H),
7.38 (m, 4 H), 7.28 - 7.14 (m, 5 H), 6.90 (d, J ) 8.4 Hz, 1 H), 6.77 (d,
J ) 6.8 Hz, 2 H), 6.67 (d, J ) 7.0 Hz, 2 H), 2.76 (s, 3 H), 2.56, (s, 3
H), 2.31, (s, 3 H), 2.29 (s, 3 H). 31P{1H}NMR: δ 26.7 (d, J ) 38.1
Hz), 11.4 (d, J ) 37.9 Hz). IR (THF): ν[CtN] 2219 cm-1. Anal. Calcd
for C55H44BrNP2Pd (found): C, 68.30 (68.36); H, 4.59 (4.87).
[(S)-Tol-BINAP]Pd(p-C6H4CN)(I) (2). Reaction of {Pd[P(o-tolyl)3]-
(p-C4H4CN)(µ-I)}2 (230 mg, 0.36 mmol) and (S)-Tol-BINAP (250 mg,
0.37 mmol) employing a procedure analogous to that used to synthesize
1 led to the isolation of 2 (273 mg, 75%) as a yellow powder. 1H
NMR (THF-d8): δ 7.99 (dd, J ) 10.2, 10.5 Hz, 1 H), 7.83 (dd, J )
8.7, 1.5 Hz, 1 H), 7.70 - 7.55 (m, 6 H), 7.50 - 6.80 (m, 19 H), 6.53
(d, J ) 8.7 Hz, 1 H), 6.44 (d, J ) 6.9 Hz, 2 H), 6.32 (d, J ) 6.9 Hz,
2 H), 2.42 (s, 3 H), 2.20 (s, 3 H), 1.96 (s, 3 H), 1.95 (s, 3 H). 31P-
{1H}NMR: δ 20.4 (d, J ) 38.6 Hz), 8.9 (d, J ) 38.6 Hz). Anal.
Calcd for C55H44INP2Pd (found): C, 65.13 (65.17); H, 4.37 (4.64).
[(S)-BINAP]Pd(p-C6H4CN)(Br) (3). Reaction of {Pd[P(o-tolyl)3]-
(p-C4H4CN)(µ-Br)}2 (200 mg, 0.17 mmol) and (S)-BINAP (220 mg,
0.35 mmol) using a procedure analogous to that used to form 1 gave
3 (246 mg, 80%) as a white solid that contained traces of ether (<5%)
[(R)-Tol-BINAP]Pd(p-C6H4CN)(OCMe3) (8). Reaction of 1 (12
mg, 1.2 × 10-2 mmol) and KOCMe3 (0.95 mg, 0.013 mmol) employing
a procedure analogous to that used to generate 5 gave 8 in 82 ( 5%
yield by 1H NMR analysis. 8 was thermally unstable and was analyzed
without isolation by 1H and 31P NMR spectroscopy. 1H NMR (22 °C,
THF-d8): in addition to a small resonance corresponding to free
alkoxide (δ 1.15), resonances were observed at δ 7.94 (t, J ) 8.1 Hz),
7.68 (q, J ) 9.3 Hz), 7.60 (d, J ) 8.6 Hz), 7.46 (t, J ) 7.1 Hz, 2 H),
7.35 (q, J ) 8.3 Hz), 7.23 (d, J ) 7.9 Hz), 7.13 (t, J ) 8.1 Hz), 7.00
(t, J ) 7.9 Hz), 6.86 (dd, J ) 2.0, 8.2 Hz), 6.68 (t, J ) 9.4 Hz, 2 H),
6.38 (d, 2 H, J ) 7.3 Hz), 6.31 (d, J ) 7.0 Hz, 2 H), 2.38 (s, 3 H),
2.28 (s, 3 H), 1.97 (s, 3 H), 1.90 (s, 3 H), 0.54 (s, 9 H). 31P{1H}NMR
(22 °C, THF-d8): δ 25.2 (d, J ) 40.0 Hz), 12.1 (d, J ) 39.8 Hz). IR
(THF): ν[CtN] 2218 cm-1
.
[(R)-Tol-BINAP]Pd(p-C6H4CN)(OC6H4Me) (9). A solution of
potassium p-cresolate (26 mg, 0.18 mmol) and 2 (180 mg, 0.18 mmol)
in THF (15 mL) was stirred at room temperature for 15 min, and the
resulting orange solution was filtered through Celite. Concentration
of solvent under vacuum (∼3 mL) and dropwise addition of hexane
(20 mL) formed a precipitate that was washed with ether and dried
under vacuum to give 9 (164 mg, 89%) as orange microcrystals. 1H
NMR (C6D6): δ 7.96 (dd, J ) 8.4, 10.6 Hz), 7.84 (t, J ) 8.9 Hz), 7.67
(t, J ) 8.6 Hz), 7.61 (m), 7.31 (dd, J ) 8.13, 11.3 Hz), 7.28 (t, J )
10.0 Hz), 7.18 (t, J ) 8.73 Hz), 6.96 (m), 6.85 (q, J ) 8.9 Hz), 6.70
(dd, J ) 2.1, 8.14 Hz), 6.57 (m), 6.24, (d, J ) 7.05 Hz), 6.14 (d, J )
7.05 Hz), 2.19 (s, 3 H), 1.98 (s, 3 H), 1.87 (s, 3 H), 1.71 (s, 6 H).
31P{1H}NMR (C6D6): δ 31.1 (d, J ) 38.3 Hz), 14.4 (d, J ) 38.3 Hz).
1
by H NMR analysis. 1H NMR (THF-d8): δ 8.01 (t, J ) 8.8 Hz, 1
H), 7.81 (d, J ) 9.3 Hz, 3 H), 7.68-7.54 (m, 6 H), 7.46 (s, 6 H), 7.39
(t, J ) 7.8 Hz, 1 H), 7.33 (t, J ) 7.3 Hz, 1 H), 7.20-7.06 (m, 5 H),
7.04-6.94 (m, 6 H), 6.80-6.50 (m, 7 H). 31P{1H}NMR: δ 27.9 (d,
J ) 38.4 Hz), 12.9 (d, J ) 38.4 Hz). Anal. Calcd for C51H36BrNP2Pd
(found): C, 67.23 (67.02); H, 3.98 (4.20).
(dppf)Pd(p-C4H4CN)(Br) (4). A solution of {Pd[P(o-tolyl)3](p-
C4H4CN)(µ-Br)}2 (200 mg, 0.17 mmol) and dppf (258 mg, 0.47 mmol)
in CH2Cl2 (10 mL) was stirred at room temperature for 12 h. The
resulting solution was concentrated under vacuum. Addition of Et2O
(10 mL) formed a precipitate that was filtered, washed with Et2O, and
dried under vacuum to give 4 (280 mg, 79%) as a bright yellow solid
1
Although solutions of 9 were homogeneous and >95% pure by H
and 31P NMR analysis, C analysis was consistently low and H analysis
was consistently high. Anal. Calcd for C62H51NOP2Pd (found): C,
74.88 (73.13); H, 5.17 (5.74). Reaction of KOC6H4Me with bromide
precursor 1 provided 9 which was spectroscopically and analytically
identical to that generated from 2.
1
that contained traces of ether (<5%) by H NMR analysis. 1H NMR
(CDCl3): δ 8.01 (dt, J ) 2.7, 9.7 Hz), 7.47 (m. 6 H), 7.33 (t, J ) 11.2
Hz, 6 H), 7.12 (dt, J ) 1.7, 15.4 Hz, 6 H), 6.76 (d, J ) 7.54 Hz, 2 H),
4.68 (d, J ) 1.95 Hz, 2 H), 4.51 (s, 2 H), 4.14 (d, J ) 2.23 Hz, 2 H),
3.59 (d, J ) 1.74 Hz, 2 H). 31P{1H}NMR (CDCl3): δ 30.0 (d, J )
29.2 Hz), 10.8 (d, J ) 31.6 Hz). IR (CH2Cl2): ν[CtN] 2220 cm-1. Anal.
Calcd for C41H32BrFeNP2Pd (found): C, 58.43 (58.41); H, 3.83 (3.98).
[(R)-Tol-BINAP]Pd(p-C6H4CN)(OCH2CMe3) (5). A 0.54 M solu-
tion of KOCH2Me3 in THF-d8 (25 µL, 1.35 × 10-2 mmol) was added
(dppf)Pd(p-C6H4CN)(OCH2CMe3) (10). A yellow suspension of
4 (11 mg, 1.3 × 10-2 mmol) in THF-d8 (0.70 mL) was treated with
aliquots of KOCH2Me3 in THF-d8. Addition of 1.1 equiv of alkoxide
formed an orange solution of 10 and a small amount of free KOCH2-
Me3 as the exclusive products by 1H NMR spectroscopy. Despite the

Direct ObserVation of C-O ReductiVe Elimination
J. Am. Chem. Soc., Vol. 119, No. 29, 1997 6795
relatively slow decomposition of 10 at room temperature (t1/2 ≈ 4 h),
attempts to isolate 10 from the corresponding preparative-scale reaction
gave only impure brown solids. 1H NMR (22 °C, THF-d8): δ 8.21
(m, 4 H), 7.46 (m), 7.40 (d J ) 11.5 Hz), 7.37 (dd, J ) 1.2, 11.6 Hz),
7.31 (t, J ) 7.3 Hz), 7.09 (dt, J ) 2.1, 8.0 Hz), 6.75 (dd, J ) 2.2, 8.2
Hz, 2 H), 4.82 (q, J ) 2.0 Hz, 2 H), 4.58 (br s, 2 H), 4.20 (t, J ) 1.6
Hz, 2 H), 3.55 (q, J ) 1.8 Hz, 2 H), 2.68 (s, 2 H), 0.23 (s, 9 H).
31P{1H}NMR (22 °C, THF-d8): δ 30.9 (d, J ) 32 Hz), 11.9 (d, J )
the respective peaks in the GC spectrum versus the integration of
PhSiMe3. The first-order rate constant for disappearance of 5 was
determined from a plot of ln [[5]t/[5]0] versus time (Table 1). The
corresponding plot of reciprocal concentration versus time deviated
considerably from linearity. Kinetics and product analysis for ther-
molysis of complexes 6-9 were performed analogously.
First-order rate constants for disappearance of 5 in the absence of
added KOCH2CMe3 (<2 mM) were also obtained at 23, 35, 37, 52,
and 57 °C (Table 1); activation parameters were obtained from a plot
ln [k/T] versus 1/T (Figure 2). First-order rate constants for disap-
pearance of 5 were also measured as a function of [KOCH2CMe3] from
0.043 to 0.30 M at 47 °C in THF-d8 (Table 1). Solutions of 5 with
KOCH2CMe3 concentrations ranging from 0.43 to 0.12 M were obtained
by a procedure analogous to that described above. Solutions of 5 with
KOCH2CMe3 concentrations >0.12 M were prepared by adding a THF-
d8 solution of 5 (17 mM) via syringe to an NMR tube containing solid
KOCH2CMe3. The first-order rate constant k was obtained as the
intercept of a plot of kobs versus [KOCH2CMe3] (Figure 3). The second-
order rate constant k′ was obtained from the slope of this plot.
Alkoxide Exchange with 5. An NMR tube containing a freshly
prepared solution of 5 (12 mg, 1.2 × 10-2 mmol, 19 mM), PhSiMe3
(1.75 mg, 1.16 × 10-2 mmol), and KOCH2CMe3 (3.4 mg, 0.0271 mmol,
0.043 mM) in THF-d8 (0.63 mL) was placed in the probe of an NMR
spectrometer heated at 47 °C. The excess line broadening (∆ω1/2) of
the tert-butyl resonance of 5 was determined by measuring the peak
width at half-height (ω1/2) for the tert-butyl resonance of 5 (δ 0.17)
relative to ω1/2 for the trimethylsilyl resonance of PhSiMe3 (δ 0.25) in
the 1H NMR spectrum [∆ω1/2 (5) ) ω1/2(5) - ω1/2(PhTMS)]. Because
the separation of the tert-butyl peaks for PdOCH2CMe3 and KOCH2-
CMe3 (∆ν > 200 Hz) was much larger than the excess broadening of
31.7 Hz). IR (THF): ν[CtN] 2218 cm-1
.
(dppf)Pd(p-C6H4CN)(OCMe3) (11). Reaction of 4 (11 mg, 1.3 ×
10-2 mmol) and KOCMe3 (1.05 mg, 1.4 × 10-2 mmol) employing a
procedure analogous to that used to generate 10 gave 11 as the exclusive
product by 1H NMR analysis. 11 was thermally unstable and was
analyzed without isolation by 1H and 31P NMR spectroscopy. 1H NMR
(22 °C, THF-d8): δ 8.25 (br t, J ) 8.4 Hz, 4 H), 7.46 (m), 7.39 (d J
) 7.5 Hz), 7.35 (d, J ) 8.0 Hz), 7.29 (d, J ) 8.1 Hz), 7.25 (s), 7.08
(br t, J ) 7.5 Hz, 4 H), 6.67 (br d, J ) 7.8 Hz, 2 H), 4.84 (br d, J )
1.86 Hz, 2 H), 4.59 (br s, 2 H), 4.15 (br s, 2 H), 3.48 (br d, J ) 1.68
Hz, 2 H), 0.46 (s, 9 H). 31P{1H}NMR (22 °C, THF-d8): δ 33.1 (d, J
) 36.8 Hz), 11.7 (d, J ) 36.7 Hz). IR (THF): ν[CtN] 2219 cm-1
.
(dppf)Pd[o-C6H4CH2CH2C(Me)2O] (12). A suspension of Pd(dp-
pf)[o-C6H4CH2CH2C(Me)2OH](Br) (50 mg, 0.55 mmol) and dry KH
(15 mg, 0.38 mmol) in THF (3 mL) was stirred at room temperature
for 3 min. The resulting yellow solution was filtered through Celite
into hexane (10 mL), and the filtrate was concentrated under vacuum
to 2.5 mL. The resulting precipitate was collected, washed with
pentane, and dried under vacuum to give 12 (35 mg, 77% yield) as a
yellow powder. 1H NMR: 8.39 (t, J ) 8.1 Hz, 2 H), 8.18 (t, J ) 8.4
Hz, 2 H), 7.61 (dd, J ) 7.7, 11 Hz, 2 H), 7.45 (m, 2 H), 7.30 (m, 8 H),
7.12 (q, J ) 8.0 Hz, 2 H), 6.96 (t, J ) 6.38 Hz, 2 H), 6.58 (dt, J ) 3.4,
12.0 Hz, 1 H), 6.50 (br d, J ) 4.8 Hz), 6.42, (t, J ) 7.2 Hz, 1 H), 6.17
(t, J ) 7.1 Hz), 4.83 (br s, 2 H), 4.55, (s, 1 H), 4.52 (s, 1 H), 4.17, (br
s, 2 H), 4.08 (dt, J ) 6.3, 12.7 Hz, 1 H), 3.71 (s, 1 H), 3.43 (s, 1 H),
2.58 (dd, 3.8, 11.5, 1 H), 1.75 (m, 1 H, partially obscured by THF),
1.52 (dt, 5.7, 12.8, 1 H), 0.55 (s, 3 H), 0.07 (s, 3 H). 31P{1H}NMR
(THF-d8): δ 35.0 (d, J ) 38.4 Hz), 11.6 (d, J ) 38.4 Hz). Anal.
Calcd for C45H42FeOP2Pd (found): C, 65.67 (65.51); H, 5.14 (5.33).
Kinetic Measurements. Samples for kinetic analysis were prepared
from stock solutions of the appropriate palladium aryl halide complex
and were performed in oven-dried 5 mm thin-walled NMR tubes capped
with rubber septa. Solvent volume in the NMR tubes was calculated
from the solvent height measured at 25 °C according to the relationship
V (mL) ) H (mm) × 0.01384 - 0.006754 and from the temperature
dependence of the density of benzene.44 Kinetic data were obtained
by 1H NMR spectroscopy in the heated probe of a Varian XL-300
spectrometer. Probe temperatures were measured with an ethylene
glycol thermometer and were maintained at ( 0.5 °C throughout data
acquisition. Syringes employed in measuring liquids for kinetic
measurements were calibrated by mercury displacement and were
accurate to >95%. Error limits for rate constants refer to the standard
deviation of the corresponding least-squares-fit line or to the standard
deviation of two or more separate experiments.
the tert-butyl resonance of 5 (ω1/2 ) 5.5 Hz), the slow-exchange
17
approximation (∆ω1/2 ) kobs/π) was employed to convert ∆ω1/2 to kobs
.
Observed rate constants for alkoxide exchange were also determined
at KOCH2CMe3 concentrations ranging from 0.0017 to 0.30 M. The
second-order rate constant kex for exchange of alkoxide with 5 was
determined from the slope of a plot of kobs versus [KOCH2CMe3] (Figure
4).
Thermolysis of 10. An NMR tube containing a freshly prepared
solution of 10 (11 mg, 1.2 × 10-2 mmol, 18 mM) and mesitylene (1.72
mg, 1.14 × 10-2 mmol) in THF-d8 (0.70 mL) was placed in the probe
of an NMR spectrometer preheated to 55 °C. The concentrations of
10, p-neopentoxybenzonitrile, and pivaldehyde were determined by
integrating the tert-butyl resonances for 7 (δ 0.23), p-neopentoxyben-
zonitrile (δ 1.06), and pivaldehyde (δ 1.04) versus the methyl resonance
1
of mesitylene (δ 2.12) in the H NMR spectrum. The concentrations
of biphenyl and benzonitrile were determined from the integration of
the respective peaks in the GC spectrum versus the mesitylene peak.
The first-order rate constant for disappearance of 10 was determined
from a plot of ln{[10]t/[10]0} versus time (Table 1). Kinetics and
product analysis for thermolysis of complexes 11 and 12 were
performed analogously.
Acknowledgment. We thank the National Science Founda-
tion, Dow Chemical, and Pfizer for their support of this work.
R.W. is an NCI Postdoctoral Trainee supported by NIH Cancer
Training Grant No. CI T32CA09112. We thank Dr. M. Palucki
for providing samples of p-neopentoxybenzonitrile, p-isopro-
poxybenzonitrile, and p-tert-butoxybenzonitrile, and we thank
Mr. J. P. Wolfe for providing a sample of (dppf)Pd[o-C6H4-
(CH2)2C(Me)2OH]Br. We would like to thank a reviewer for
several insightful suggestions including the possibility of an
alkoxide-promoted reductive elimination pathway.
Thermolysis of 5. An NMR tube containing a freshly prepared
solution of 5 (12 mg, 1.2 × 10-2 mmol) and PhSiMe3 (1.75 mg, 1.16
× 10-2 mmol) in THF-d8 (0.70 mL) ([KOCH2CMe3] ≈ 1.7 mM) was
placed in the probe of an NMR spectrometer heated at 47 °C. The
concentrations of 5 and p-neopentoxybenzonitrile were determined by
integrating the tert-butyl resonances for 5 (δ 0.17) and p-neopentoxy-
benzonitrile (δ 1.06) versus the trimethylsilyl resonance of PhSiMe3
(δ 0.25) in the 1H NMR spectrum. The concentrations of 4,4′-
dimethylbiphenyl and benzonitrile were determined by integration of
(44) International Critical Tables of Numerical Data, Physics, Chemistry,
and Technology; Washburn, E. W., Ed.; McGraw-Hill: London, 1928; Vol.
III, pp 29, 39, 221.
JA963324P