Reductive elimination of ether from T-shaped, monomeric arylpalladium alkoxides

Article abstract of DOI:10.1002/anie.200702809

(Chemical Equation Presented) Isolated truth: The synthesis, structures, and reductive elimination chemistry of arylpalladium(II) phenoxide and alkoxide complexes with a single bulky phosphine ligand are reported. These complexes are true intermediates in palladium-catalyzed etherification of aryl halides. They undergo reductive elimination of the alkyl aryl ether directly from the isolated complex (see scheme), and the rates of these reductive eliminations are slower than those from related arylpalladium amido complexes. Ad = adamantyl.

Full text of DOI:10.1002/anie.200702809

Communications
DOI: 10.1002/anie.200702809
À
C O Bond Formation
Reductive Elimination of Ether from T-Shaped, Monomeric
Arylpalladium Alkoxides**
James P. Stambuli, Zhiqiang Weng, Christopher D. Incarvito, and John F. Hartwig*
Reductive elimination of ethers from organopalladium alk-
À
oxides is the C O bond-forming step of the palladium-
catalyzed etherification of aryl halides.[1] At this time,
examples of the reductive eliminations of ethers from isolated
organopalladium alkoxide complexes are limited. Only
reductive eliminations of alkyl aryl ethers from complexes
generated in situ, reductive elimination of an oxygen hetero-
cycle from a single oxapalladacycle,[2] reductive eliminations
from arylpalladium alkoxides containing activating groups on
the aryl ring,[3,4] and reductive eliminations of biaryl ethers in
low yield from isolated complexes[5,6] are documented. Most
importantly, the most active catalysts for the palladium-
catalyzed etherification of aryl halides contain sterically
hindered monophosphines,[1,6–15] and no reductive elimina-
tions from arylpalladium alkoxides containing hindered
monophosphines have been described.
We report reductive eliminations from isolated and
structurally characterized examples of such arylpalladium
alkoxides. We show that reductive eliminations occur from
the observed species containing one aryl, one alkoxide, and
one phosphine ligand. The structures of the reactive com-
plexes match those of proposed intermediates in the catalytic
etherifications,[1] and the rate of their eliminations of ethers
Scheme 1.
can be compared to those from related amido complexes.[16]
The synthesis of arylpalladium alkoxide complexes ligated
by PtBu3 or 1-AdPtBu2 (1-Ad = 1-adamantyl) is summarized
in Scheme 1. Reaction of the hindered sodium 2,4,6-tri-tert-
butylphenoxide with [(1-AdPtBu2)Pd(Ph)(Br)] (1a) formed
phenoxide complex 2 in 64% yield. This complex was
crystallized from pentane and was characterized by standard
spectroscopic methods and X-ray diffraction (see below).
Reaction of [(PtBu3)Pd(Ph)(Br)] (1b)[17,18] with NaOAr
(Ar= C6H4-4-OMe) formed the dimeric phenoxide complex
[{(PtBu3)Pd(Ph)(OAr)}2] (3) in 50% yield. The connectivity
of this complex in the solid state was determined by X-ray
diffraction, and this dinuclear structure appeared to be
preserved in solution, at least at low temperatures. Solutions
of [{(PtBu3)Pd(o-tol)(OAr)}2] and [(PtBu3)Pd(Ph)(OAr)]2
(6.25 mm) alone generated singlet 31P NMR resonances at
À608C. Amixture of the two compounds at À608C generated
four resonances, one for each homodimer and two that most
likely correspond to the inequivalent PtBu3 ligands in the
heterodimer [{(PtBu3)(Ph)Pd(OAr)2}{Pd(o-Tol)(PtBu3)}].[19]
Reaction of [(PtBu3)Pd(Ar)(Br)] (Ar= Ph 1b, A r= o-tol
1d)[17,18] with NaOtBu at À108C generated [(PtBu3)Pd(Ar)-
(OtBu)] (Ar= Ph 4a, A r= o-Tol 4b) in 92% and 89% yields,
as determined by NMR spectroscopy with an internal
standard. We could not isolate these complexes because
they reacted at 0–258C (see below). Thus, 4a and 4b were
characterized by low-temperature 1H and 13C NMR spectros-
copy (see Supporting Information).
[*] Dr. Z. Weng, Prof. Dr. J. F. Hartwig
Department of Chemistry
To generate more stable alkoxide analogues of 4a and 4b,
we treated 1 with fully and partially fluorinated versions of
NaOtBu. Reaction of [(1-AdPtBu2)Pd(Ar)(Br)] (Ar= Ph 1a,
Ar= o-tol 1c)[17,18] with the perfluoro tert-butoxide NaOC-
(CF3)3 or the trifluoro tert-butoxide NaOC(CH3)2CF3, and
subsequent AgOTf, generated [(1-AdPtBu2)Pd(Ph){OC-
(CF3)3}] (5) and [(1-AdPtBu2)Pd(Ar){OC(CH3)2CF3}] (6a,b)
cleanly, and these complexes were isolated in 28%, 52%, and
60% yields. Complexes 5 and 6a,b were characterized by
spectroscopic and microanalytical methods, and complexes
6a and 6b were characterized by X-ray crystallography.
University of Illinois Urbana-Champaign
Urbana, IL 61801 (USA)
Fax: (+1)217-244-8024
E-mail: jhartwig@uiuc.edu
Dr. J. P. Stambuli, Dr. C. D. Incarvito, Prof. Dr. J. F. Hartwig
Department of Chemistry, Yale University
NewHaven, CT 06520-8107 (USA)
[**] We are grateful to the NIH-NIGMS (GM-55382) for support of this
work and Johnson-Matthey for palladium.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
7674
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 7674 –7677

Angewandte
Chemie
ORTEP diagrams of aryloxide 2 and trifluoro-tert-but-
oxide 6a are shown in Figures 1 and 2. Both complexes are T-
shaped monomers. An ORTEP drawing of 6b is provided in
the Supporting Information. Each complex contains an
two complexes. One might envision that an o-tolylpalladium
alkoxide containing a single phosphine ligand would be
À
stabilized by an agostic interaction with a C H bond of the o-
tolyl group. However, the solid-state structure of 6b is nearly
identical to that of 6a. The same close Pd–H distance to the
adamantyl methylene hydrogen atom (2.156 Š) was
observed, and there was no evidence for an interaction of
À
palladium with the C H bond of the o-tolyl group.
The reductive eliminations of complexes 2–6 are sum-
marized in Equations (1) and (2) and in Scheme 2. Heating of
Figure 1. ORTEP diagram of [(1-AdPtBu2)Pd(OC6H2-1,3,5-tBu)(Ph)] (2)
with thermal ellipsoids set at 30% probability. Selected bond lengths
[Š]: Pd(1)–C(19) 1.980(4), Pd(1)–O(1) 2.057(2), Pd(1)–P(1)
2.2795(11), Pd(1)–H(2B) 2.14.
hindered phenoxide 2 in C6D6 with added 1-AdPtBu2 formed
the Pd0 complex [Pd(1-AdPtBu2)2], but no biaryl ether.
Biphenyl and 2,4,6-tri-tert-butyl phenol were formed instead
[Eq. (1)].[22] However, heating of the less hindered phenoxide
3 in [D8]toluene with added PtBu3 at 708C generated diphenyl
ether in 67% yield and [Pd(PtBu3)2] in 79% yield.
Several of the alkoxide complexes also underwent reduc-
tive elimination (Scheme 2). Warming C6D6 solutions of the
tert-butoxide complexes 4a and 4b generated in situ with
added PtBu3 led to the formation of ArOtBu (Ar= Ph, o-tol)
in 93–97% yield at room temperature. Perfluoro-tert-but-
oxide 5 was stable at room temperature and did not form any
perfluoroalkyl aryl ether upon decomposition at 808C in
C6D6. However, heating of arylpalladium trifluoro-tert-but-
oxide complexes 6a and 6b at 708C with added 1-AdPtBu2
generated ArOCMe2CF3 (Ar= Ph, o-Tol) in 66% and 73%
yield. In all cases the [L2Pd0] species was the only palladium
Figure 2. ORTEP diagram of [(1-AdPtBu2)Pd(OC(CH3)2CF3)(Ph)] 6a
with thermal ellipsoids set at 30% probability. Selected bond lengths
[Š]: Pd(1)–C(19) 1.976(2), Pd(1)–O(1) 2.0193(17), Pd(1)–P(1)
2.2639(8), Pd(1)–H(1) 2.18.
adamantyl methylene hydrogen atom that is close (2.14, and
2.18(3) Š) to palladium. This hydrogen atom in 2 and 6a was
refined in an idealized position and is unlikely to deviate from
this position significantly because of the rigidity of the
adamanyl unit; the hydrogen atom in 6b was located and
refined. Both Pd···H distances are consistent with an agostic
interaction[20] that occupies a coordination site of these
formally unsaturated species.[21] If this C H bond is included
À
as a fourth ligand, the palladium complexes are square-
planar; the sum of the angles about the palladium involving
the heavy atoms are 360 and 3578, respectively. The Pd–O and
the Pd–P distances in 6a are about 0.04 Š and 0.015 Š smaller
than those in 2. The Pd–C distances are nearly identical in the
Scheme 2.
Angew. Chem. Int. Ed. 2007, 46, 7674 –7677
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
7675

Communications
product observed. These results are consistent with the
In summary, we have described the direct observation of
reductive elimination from a series of monomeric arylpalla-
dium(II) aryloxide and alkoxide complexes, including the first
reductive elimination from an isolated T-shaped alkoxide
complex. These reactions allow a direct comparison of the
rates of these eliminations to related processes and constitute
a rare case in which unsaturated intermediates in a catalytic
process are observed, isolated, and studied directly.
intermediacy of 3 and 4 in the palladium-catalyzed formation
of biaryl and tert-butyl aryl ethers from aryl halides and
NaOAr[5,6,9,14] and NaOtBu.[10,23,24] Complexes 6a and 6b
constitute rare isolated and fully characterized alkoxide
complexes that reductively eliminate ether.
Most importantly, the stability of trifluoro tert-butoxides
6a,b at room temperature allowed kinetic data to be obtained
without potential complications from impurities.[2] Kinetic
data on the reactions of 6b were consistent with direct
reductive elimination from the isolated complex. Reaction of
6b (12 mm in C6D6) at 738C in the presence of 1, 5, and
10 equivalents of added 1-AdPtBu2 occurred with rate con-
stants of 4.6 Æ 0.4, 4.2 Æ 0.3, and 5.8 Æ 0.8 ” 10À4 sÀ1, respec-
tively. Thus, the reaction is zero-order in added ligand, and
reversible dissociation or association of ligand does not occur
Experimental Section
Representative procedure for generation and characterization of 4a
at low temperature: A600
mL CD2Cl2 solution of
[(PtBu3)Pd(Ph)(Br)] (1b) (12.0 mg, 0.0260 mmol) was placed into a
screw-capped NMR tube and cooled to 08C. To this tube was added
by syringe a solution of NaOtBu (216 mL, 0.12m, 0.026 mmol) in
CD2Cl2. The reaction mixture was agitated and immediately turned
from yellow to deep orange. The tube was placed into a NMR probe
that was precooled to À108C. The reaction was monitored by
31P NMR spectroscopy for approximately 1 h or until
[(PtBu3)Pd(Ph)(Br)] was completely consumed. 1H NMR (CD2Cl2,
400 MHz, À108C): d = 1.06 (s, 9H), 1.41 (d, J = 12.0 Hz, 27H), 6.82–
6.85 (m, 3H), 7.34–7.36 ppm (m, 2H); 13C{1H} NMR (CD2Cl2,
100 MHz, À108C): d = 31.1 (d, J = 4.0 Hz), 34.5, 38.4 (d, J = 9.6 Hz),
71.0, 122.2, 125.7, 136.0 (d, J = 1.9 Hz), 136.1 ppm (d, J = 5.7 Hz);
31P NMR (CD2Cl2, 202 MHz, À108C): d = 65.9 ppm.
À
prior to C O bond formation.
These rate data were similar to those for reductive
eliminations from tert-butoxides 4a,b. Reductive elimination
from 4a was also independent of the concentration of added
ligand. Complex 4a (24 mm in CH2Cl2) reacted at 308C with
rate constants of 2.6 Æ 0.2, 2.4 Æ 0.2, and 2.5 Æ 0.2 ” 10À4 sÀ1,
and 4b (12 mm in [D8]toluene) reacted at 158C with rate
constants of 4.1 Æ 0.4, 4.4 Æ 0.2, and 5.3 Æ 0.5 ” 10À4 sÀ1 with 1,
5, and 10 equivalents of added ligand. The faster elimination
from these complexes than from trifluoro tert-butoxide 6a is
consistent with faster carbon–heteroatom bond-forming
reductive elimination from complexes with more electron-
rich amido and alkoxo groups.[25,26]
Preparation of 6b: In a glovebox, [(1-AdPtBu2)Pd(o-tol)(Br)]
(1c, 31 mg, 0.055 mmol) and NaOC(CH3)2(CF3) (8.5 mg, 0.057 mmol)
were dissolved in toluene (2 mL), and the solution was stirred. In a
separate vial, silver triflate (15 mg, 0.058 mmol) was dissolved in
toluene (1 mL). The triflate solution was added dropwise to the
stirred solution of palladium and alkoxide. Aprecipitate formed
immediately. The reaction was stirred for an additional 5 min. After
this time, the reaction mixture was filtered through celite, and all of
the toluene was evaporated under reduced pressure. The resulting
yellow-orange residue was dissolved in a minimum amount of ether
(ca. 2 mL), layered with pentane (ca. 6 mL), and cooled at À358C.
After 24 h, yellow crystals were collected, washed quickly with
pentane, and dried under vacuum to yield 60% of 6b. 1H NMR (C6D6,
500 MHz): d = 1.13 (d, J = 12.5 Hz, 9H), 1.15 (d, J = 13 Hz, 9H), 1.43
(s, 3H), 1.44–1.52 (m, 6H), 1.59 (s, 3H), 1.71 (brs, 3H), 2.01–2.14 (m,
6H), 3.26 (s, 3H), 6.77–6.79 (m, 1H), 6.86–6.90 (m, 2H), 7.45 ppm (d,
J = 8 Hz, 1H); 13C{1H} NMR (C6D6, 126 MHz): d = 27.0, 27.4, 28.9 (d,
J = 7.8 Hz), 29.0, 31.9, 32.0, 36.3, 39.0 (d, J = 9.8 Hz), 39.2 (d, J =
9.8 Hz), 41.2, 45.8 (d, J = 9.8 Hz), 75.55 (q, J = 25 Hz), 123.7, 124.2,
128.9, 130.90 (q, J = 290 Hz), 136.3 (d, J = 4.9 Hz), 137.4 (d, J =
1.9 Hz), 142.5 ppm; 31P NMR (C6D6, 202 MHz): d = 64.5 ppm.
19F NMR (C6D6, 376 MHz) d = À82.5 ppm. Elemental analysis calcd
(%) for C29H46F3OPPd: C 57.57, H 7.66; found: C 57.53, H 7.84.
Crystal data for 2: Mr = 725.33, monoclinic, space group C2/c, a =
36.023(7), b = 9.6881(19), c = 22.328(5) Š, a = 90, b = 92.06(3), g =
908, V= 7787(3) Š3, Z = 8, 1 = 1.237 MgmÀ3, m = 0.547 mmÀ1, T=
183(2) K, crystal dimensions: 0.20 ” 0.20 ” 0.15 mm. Of 8881 reflec-
tions measured, 5397 unique reflections were used in refinement.
Final R = 0.0544, (Rw = 0.1119). Crystal data for 6a: Mr = 591.00,
monoclinic, space group P2(1)/c, a = 13.887(3), b = 11.847(2), c =
17.853(4) Š, a = 90, b = 112.79(3), g = 908, V= 2707.9(9) Š3, Z = 4,
1 = 1.450 MgmÀ3, m = 0.784 mmÀ1, T= 183(2) K, crystal dimensions:
0.30 ” 0.25 ” 0.25 mm. Of 6687 reflections measured, 5250 unique
reflections were used in refinement. Final R = 0.0363, (Rw = 0.0850).
Crystal data for 6b: Mr = 605.03, monoclinic, space group P2(1)/c, a =
10.6142(3), b = 18.5875(6), c = 14.9558(5) Š, a = 90, b = 103.338(2),
These data allow comparison of the rate of reductive
elimination of ether from a discrete three-coordinate and an
isolated four-coordinate palladium alkoxo complex. Reduc-
tive eliminations from binap-ligated alkoxides (binap = 2,2’-
bis(diphenylphosphanyl)-1,1’-binaphthyl) containing elec-
tron-neutral aryl groups occur in low yields. These reactions
occur in higher yields from binap-ligated alkoxides containing
electron-poor aryl groups. However, the reductive elimina-
tions from alkoxides 4 and 6 containing a single phosphine
and an electron-neutral aryl group are comparable or faster
than those from binap-ligated arylpalladium complexes
generated in situ containing electron-poor aryl groups.[2,4]
Although there are differences between the electron-donat-
ing properties of arylphosphines and alkylphosphines and
between bisphosphines and monophosphines, this compar-
ison fits the trend of faster reductive elimination from three-
coordinate species than from four-coordinate species[25]
observed for alkylpalladium and amidopalladium com-
plexes.[26]
These data also allow comparison of the rates of reductive
elimination from three-coordinate alkoxo and three-coordi-
nate amido complexes. Reductive elimination of tert-butyl
aryl ether from alkoxide 4a is slower than reductive
elimination of triarylamine from the related diarylamido
complex [(PtBu3)Pd(Ar)(NTol2)]. Alkoxide 4a reacted at
308C at the same timescale as [(PtBu3)Pd(Ar)(NTol2)]
reacted at À108C.[16] Further studies are needed to determine
if this difference results from steric effects, basicity of
alkoxide versus amide ligands, polarizability of the hetero-
atom, or thermodynamic driving force.
g = 908, V= 2871.06(16) Š3, Z = 4, 1 = 1.400 MgmÀ3, m = 0.741 mmÀ1
,
T= 193(2) K, crystal dimensions: 0.59 ” 0.48 ” 0.06 mm. Of 43885
reflections measured, 8368 unique reflections were used in refine-
ment. Final R = 0.0316, (Rw = 0.0765). CCDC-651861 (2), 651862
7676 www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 7674 –7677

Angewandte
Chemie
(6a), and 651863 (6b) contain the supplementary crystallographic
data for this paper. These data can be obtained free of charge from
The Cambridge Crystallographic Data Centre via www.ccdc.cam.
ac.uk/data_request/cif.
[11] S. Kuwabe, K. E. Torraca, S. L. Buchwald, J. Am. Chem. Soc.
2001, 123, 12202.
[12] K. Torraca, S. Kuwabe, S. Buchwald, J. Am. Chem. Soc. 2000,
122, 12907.
[13] A. V. Vorogushin, X. H. Huang, S. L. Buchwald, J. Am. Chem.
Soc. 2005, 127, 8146.
Received: June 25, 2007
[14] C. H. Burgos, T. E. Barder, X. H. Huang, S. L. Buchwald,
Angew. Chem. 2006, 118, 4427; Angew. Chem. Int. Ed. 2006,
45, 4321.
[15] S. Harkal, K. Kumar, D. Michalik, A. Zapf, R. Jackstell, F.
Rataboul, T. Riermeier, A. Monsees, M. Beller, Tetrahedron
Lett. 2005, 46, 3237.
[16] M. Yamashita, J. F. Hartwig, J. Am. Chem. Soc. 2004, 126, 5344.
[17] J. P. Stambuli, C. D. Incarvito, M. Buhl, J. F. Hartwig, J. Am.
Chem. Soc. 2004, 126, 1184.
[18] J. P. Stambuli, M. Bꢀhl, J. F. Hartwig, J. Am. Chem. Soc. 2002,
124, 9346.
[19] These resonances broadened at higher temperatures, possibly
because of equilibria between monomers and dimers.
[20] M. Brookhart, M. L. H. Green, L.-L. Wong, Prog. Inorg. Chem.
1988, 36, 1.
[21] The IR spectra do not contain low frequency bands, and the
1H NMR spectra do not contain upfield signals. The same
absence of spectroscopic indicators of agostic interations were
discussed extensively and supported by theory in reference [17].
[22] The origin of the proton to form the phenol is not known.
[23] G. Mann, J. F. Hartwig, J. Org. Chem. 1997, 62, 5413.
[24] M. Watanabe, M. Nishiyama, Y. Koie, Tetrahedron Lett. 1999, 40,
8837.
Published online: September 4, 2007
Keywords: cross-coupling · ether · palladium ·
.
reductive elimination
[1] J. F. Hartwig in Handbook of Organopalladium Chemistry for
Organic Synthesis, Vol. 1 (Ed.: E. I. Negishi), Wiley-Interscience,
New York, 2002, p. 1097.
[2] R. A. Widenhoefer, H. A. Zhong, S. L. Buchwald, J. Am. Chem.
Soc. 1997, 119, 6787.
[3] G. Mann, J. F. Hartwig, J. Am. Chem. Soc. 1996, 118, 13109.
[4] R. A. Widenhoefer, S. L. Buchwald, J. Am. Chem. Soc. 1998, 120,
6504.
[5] G. Mann, C. Incarvito, A. L. Rheingold, J. F. Hartwig, J. Am.
Chem. Soc. 1999, 121, 3224.
[6] Q. Shelby, N. Kataoka, G. Mann, J. F. Hartwig, J. Am. Chem. Soc.
2000, 122, 10718.
[7] N. Kataoka, Q. Shelby, J. P. Stambuli, J. F. Hartwig, J. Org. Chem.
2002, 67, 5553.
[8] K. W. Anderson, T. Ikawa, R. E. Tundel, S. L. Buchwald, J. Am.
Chem. Soc. 2006, 128, 10694.
[9] K. E. Torraca, X. H. Huang, C. A. Parrish, S. L. Buchwald, J.
Am. Chem. Soc. 2001, 123, 10770.
[10] C. A. Parrish, S. L. Buchwald, J. Org. Chem. 2001, 66, 2498.
[25] J. F. Hartwig, Acc. Chem. Res. 1998, 31, 852.
[26] J. F. Hartwig, Inorg. Chem. 2007, 46, 1936.
Angew. Chem. Int. Ed. 2007, 46, 7674 –7677
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
7677