Communications
binding and acidity to selectivity exists within the series of
greater solubility of LiNBu2 over LiN(H)Ph likely contributes
to the complete chemoselectivity reversal observed in diox-
ane. In DME, where both lithium amides are completely
soluble, steric and nucleophilic differences between the
lithium amides likely influence the observed selectivity the
most. Thus, in cases where the acidity of the amine complex
determines the overall chemoselectivity, use of the corre-
sponding pre-deprotonated lithium amide permits a reversal
of chemoselectivity.
In conclusion, we have isolated and crystallographically
characterized the first neutral aryl palladium amine complex
(2) that is an intermediate within a catalytic cycle. Using
complex 1, we have shown that between isosteric aliphatic
amines, amine acidity is the primary influence on the
selectivity (C–H control), while between isosteric anilines,
amine binding is more influential (no C–H control). When
amine acidity has more influence on selectivity, the intrinsic
selectivity can be circumvented by employing the correspond-
ing lithium amides and removing the deprotonation step from
the catalytic cycle. The trends established by correlating
binding and nucleophilicity to selectivity demonstrate that the
origin of amine selectivity cannot be explained simply by
steric effects, and that electronic properties of the amines
must also be considered. The information gathered from these
anilines (Table 1, entries 1–4) and within the series of
aliphatic amines (Table 1, entries 5–11). Within the series of
aliphatic amines, the relative acidity of the amine complex has
more influence on the observed selectivity than the relative
binding affinity of the amine. This trend can be observed by
comparing groups of isosteric aliphatic amines (Table 1,
entries 5 and 6 or 7–9). When the butyl groups of dibutyl-
amine are replaced with methoxyethyl groups (Table 1,
entries 5 and 6), selectivity is increased more than two-fold
owing to enhanced acidity, despite a seven-fold decrease in
binding ability. When electron-withdrawing groups are
inserted into the backbone of piperidine (Table 1, entries 7–
9), greater selectivity is likewise observed for the more
electron-deficient (lower pKa)[11,12] amine, despite a necessary
decrease in binding ability. Such electronic control is also
observed when an aliphatic amine competes against an aniline
derivative, as the coupling is invariably more selective for the
more acidic, less nucleophilic aniline substrates. These results
are all consistent with C–H control.
In contrast, within the aniline series, product determina-
tion appears to be governed by relative binding ability.[13] The
S value is highest for p-Me2NC6H4NH2 (Table 1, entry 5),
which binds the strongest. For anilines that are more electron-
deficient than p-Me2NC6H4NH2, it appears that the effect of
enhanced acidity gained from binding to palladium no longer
outweighs the decreased binding affinity, and poorer selec-
tivity results. Thus, selectivity within the aniline series does
not seem to be under C–H control.
À
studies provides insight both into the mechanism of C N
couplings using biaryl phosphines and into the feasibility of
À
achieving chemoselective C N cross-coupling reactions.
Received: May 15, 2007
Published online: August 6, 2007
Finally, when amine binding differs on account of steric
differences rather than pKa differences, as with sBuNH2 and
tBuNH (Table 1, entries 10 and 11) or Bu2NH and piperidine
(Table 1, entries 6 and 9), arylation is more selective for the
less sterically encumbered amine as a result of its greater
binding affinity.[14–16]
To illustrate the important role that the deprotonation
step plays in product determination, three competition
reactions analogous to those of Table 1 were performed in
Keywords: amination · chemoselectivity ·
.
homogeneous catalysis · palladium · phosphanes
[1] X. Huang, K. W. Anderson, D. Zim, L. Jiang, A. Klapars, S. L.
Buchwald, J. Am. Chem. Soc. 2003, 125, 6653.
À
[2] For reviews on C N cross-coupling reactions, see: a) J. F.
Hartwig in Handbook on Organopalladium Chemistry for
Organic Synthesis (Ed.: E. Negishi), Wiley-Interscience, New
York, 2002, p. 1051; b) L. Jiang, S. L. Buchwald in Metal-
Catalyzed Cross-Coupling Reactions, 2nd ed. (Eds.: A. de Mei-
jere, F. Diederich), Wiley-VCH, Weinheim, 2004, p. 699.
[3] T. E. Barder, M. R. Biscoe, S. L. Buchwald, Organometallics
2007, 26, 2183.
Table 2: Competition reactions of lithium amides with PhCl using 1.[a]
[4] For the use of [Me2Pd(tmeda)] as a Pd0 source, see: a) W.
de Graaf, J. Boermsa, W. J. J. Smeets, A. L. Spek, G. van Koten,
Organometallics 1989, 8, 2907; b) S. M. Reid, R. C. Boyle, J. T.
Mague, M. J. Fink, J. Am. Chem. Soc. 2003, 125, 7816; c) Crystal
PhNBu2
:
5
:
5
:
5
R2NH/PhNH2[b]
R2NLi/PhNHLi[c]
R2NLi/PhNHLi[d]
<1
>99
72
:
:
:
>99
<1
28
9
86
95
:
:
:
91
14
5
6
96
88
:
:
:
94
4
12
data for 1: C32H40ClO2PPd, crystals from CH2Cl2/hexane, Mr =
3
¯
629.46, 0.26 0.23 0.22 mm , tetragonal, space group P4c2, a =
b = 16.7915(2), c = 21.7992(4) , V= 6146.38(15) 3,Z = 8,
[a] Ratios determined by GC. [b] NaOtAm, toluene. [c] Dioxane. [d] Di-
methoxyethane (DME).
1calcd = 1.360 MgmÀ3
,
T= 100(2) K, F(000) = 2608, 2qmax =
56.588, monochromated MoKa radiation, l = 0.71073 , m =
0.769 mmÀ1, Siemens Platform three-circle diffractometer equip-
ped with a CCD detector, 123200 measured and 7638 independ-
ent reflections, Rint = 0.0350. Data processed using the program
SAINT supplied by Siemens Industrial Automation, Inc.,
structure determination by direct methods (SHELXTL V6.10,
G. M. Sheldrick, University of Göttingen, and Siemens Indus-
trial Automation, Inc.), structure refined on F2 by full matrix
least-squares methods, absorption correction applied with
SADABS. All non-hydrogen atoms were refined anisotropically.
which lithium amides were employed instead of neutral
amines (Table 2). In each reaction, a dramatic reversal of
selectivity from that of Table 1 was observed. For reactions
involving lithium amides, the binding/deprotonation step is
eliminated from the catalytic cycle, and selectivity is deter-
mined by the relative nucleophilicity and steric bulk of the
lithium amides as well as by the solvent employed. The
7234
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 7232 –7235