Competition between sp3-C-N vs sp3-C-F reductive elimination from PdIV complexes

Article abstract of DOI:10.1021/ja411433f

This communication describes the design of a model system that allows direct investigation of competing sp³-C-N and sp³-C-F bond-forming reductive elimination from a Pd^IV fluoro sulfonamide complex. The reductive elimination selectivity varies dramatically as a function of reaction additives. A mechanism is proposed that provides a rationale for these effects.

Full text of DOI:10.1021/ja411433f

Communication
pubs.acs.org/JACS
Competition between sp3‑C−N vs sp3‑C−F Reductive Elimination
from PdIV Complexes
Monica H. Perez-Temprano, Joy M. Racowski, Jeff W. Kampf, and Melanie S. Sanford*
́
́
Department of Chemistry, University of Michigan, 930 North University Avenue, Ann Arbor, Michigan 48109, United States
S
* Supporting Information
sp3-C−F reductive elimination upon heating to 80 °C (eq 2).17
We hypothesized that replacing the pyridine (pyr) ligand with
ABSTRACT: This communication describes the design
of a model system that allows direct investigation of
competing sp3-C−N and sp3-C−F bond-forming reductive
elimination from a PdIV fluoro sulfonamide complex. The
reductive elimination selectivity varies dramatically as a
function of reaction additives. A mechanism is proposed
that provides a rationale for these effects.
alladium-catalyzed reactions for constructing sp3-carbon−
P
heteroatom bonds have the potential to facilitate the
synthesis of valuable classes of organic molecules.1 In particular,
sp3-carbon−nitrogen bonds are ubiquitous in pharmaceuticals,
agrochemicals, and commodity chemicals.2 Despite the great
importance of alkyl amines, it remains challenging to form sp3-
C−N linkages via palladium catalysis.3 Indeed, there are only a
handful of examples of well-characterized sp3-C−N bond-
forming reductive elimination from any transition metal
complex.4−6
TsNH would produce PdIV complex B, which should enable
investigation of competing C−F and C−N coupling. Herein we
report the synthesis and reactivity of B. We demonstrate that
selective sp3-C−F and sp3-C−N bond formation can be
achieved, depending on the reaction conditions. These studies
represent the first example of sp3-C−N bond-forming reductive
elimination from an isolated PdIV complex.
Yu and co-workers recently pioneered a high valent Pd-
catalyzed strategy for C−N bond formation.7 This approach
involves the use of electrophilic fluorinating reagents8 as
bystanding oxidants in conjunction with nitrogen nucleophiles
to promote C−N coupling reactions.9 The earliest reports in
this area focused on arene C−H amination (i.e., sp2-C−N
coupling).10 However, subsequent efforts by Michael,11
As shown in Scheme 1, the treatment of PdIV complex 217
with 1 equiv of NH2SO2-p-tolyl (NH2Ts) and 1.5 equiv of
Scheme 1. Synthesis of Complexes 3a and 3b
Muniz,12 Liu,13 and others14 have adopted similar approaches
̃
to generate sp3-C−N bonds during alkene and alkane
amination reactions.
The latter transformations are proposed to proceed via PdIV
alkyl amino fluoride intermediates of general structure A (eq 1).
Cs2CO3 in CH3CN at room temperature results in rapid
substitution of triflate for NHTs to afford 3 in an 85% isolated
yield. This complex is formed as a 6:1 mixture of inseparable
isomers 3a and 3b. Each of these isomers was fully
These species can potentially undergo competing C−N and
C−F15,16 bond-forming reductive elimination to yield alkyl
amines or alkyl fluorides. During catalysis, good to excellent
selectivity is typically observed for sp3-C−N coupling.
However, a detailed understanding of the mechanism of
reductive elimination and the factors governing this selectivity
remains elusive.
We sought to design a model system that would allow direct
interrogation of competing sp3-C−N and sp3-C−F bond-
forming reductive elimination from a well-defined PdIV
complex. We recently reported that PdIV complex 1 undergoes
1
characterized by one- and two-dimensional H, 13C, and 19F
NMR spectroscopy (see Supporting Information (SI) for all
spectra and a full discussion of structural assignments). In
addition, an X-ray crystal structure of 3a was obtained, and the
ORTEP plot is shown in Figure 1.
Heating solutions of 3 for 30 min at 65 °C in CD3CN results
in competing sp3-C−N, sp3-C−F, and sp3-C−sp2-C bond-
Received: November 11, 2013
Published: February 28, 2014
© 2014 American Chemical Society
4097
dx.doi.org/10.1021/ja411433f | J. Am. Chem. Soc. 2014, 136, 4097−4100

Journal of the American Chemical Society
Communication
Table 1. Influence of Additives on Reductive Elimination
a
from 3
yield
4 (%)
yield
5 (%)
yield
6 (%)
Figure 1. ORTEP plot of complex 3a.
entry
additive
1
2
3
none
9
85
98
49
<1
<1
29
<1
<1
forming reductive elimination to form a mixture of products 4,
5, and 6, respectively (Scheme 2). Importantly, no products
1 equiv of NMe4NHTs
1 equiv of NMe4NHTs + 2 equiv
of bpy
a
Scheme 2. Reductive Elimination from 3
Yields of 4−6 determined by 1H NMR spectroscopic analysis relative
to 4-fluoroanisole as a standard.
bipyridine (bpy) is also formed in ∼10% yield. We propose that
the free bpy is derived from decomposition of product 4 via
ligand substitution of bpy with TsNH−. Consistent with this
proposal, subjecting an isolated sample of 4 to 3 equiv of
NMe4NHTs at room temperature for 1 h in CD3CN results in
the formation of a 61% yield of free bpy.
This undesired ligand exchange can be suppressed by adding
2 equiv of bpy along with 1 equiv of NMe4NHTs to reductive
elimination reactions of 3. Under these conditions, 4 is formed
in nearly quantitative yield (98%, Table 1, entry 3).
Furthermore, the rate of reductive elimination from 3 is
essentially identical in the presence and absence of 2 equiv of
bpy (see Figure S1). This supports the proposal that the bpy
additive does not influence the primary reductive elimination
process, but merely limits product decomposition.
With conditions in hand to achieve selective and high
yielding C−N bond-forming reductive elimination from 3, we
next conducted studies to probe the reaction mechanism. In the
presence of ≥1 equiv of NMe4NHTs, the reaction exhibits
clean kinetics with a first-order dependence on [3] and zero-
order dependence on [NMe4NHTs]. This rules out a pathway
involving a direct SN2 reaction between TsNH− and 3, since
this should show first-order kinetics in both 3 and the
nucleophile.
derived from sp2-carbon−heteroatom (C−X) bond-forming
reductive elimination were observed. Selectivity for sp3-C−X
coupling over sp2-C−X coupling is a hallmark of high valent Pd
and Pt chemistry (and stands in contrast to the selectivity
typically observed at other metal centers/oxidation states).18
Products 4, 5, and 6 were fully characterized using one- and
1
two-dimensional H, 13C, and 19F NMR spectroscopy (see SI
for all spectra and a full discussion of structural assignments).
Furthermore, X-ray crystal structures of 4 and 5 were obtained,
and ORTEP plots for these PdII complexes are shown in Figure
2.
We next probed the lability of the TsNH− ligand of 3 under
the reductive elimination reaction conditions. EXSY NMR
studies of a mixture of 1 equiv of 3 and 1 equiv of NBu4NHTs
show that exchange between free and bound NHTs is slow at
25 °C on the NMR time scale. However, the treatment of 3
with 1 equiv of NMe4NHMs (Ms = MeSO2) for 30 min at 65
°C in CD3CN results in rapid substitution of TsNH for MsNH
to form an equilibrium mixture of 3 and 3Ms (Scheme 3).19
These latter results show that sulfonamide exchange is
significantly faster than reductive elimination from 3.
Figure 2. ORTEP plots of (a) 4 and (b) 5.
As shown in Scheme 2, reductive elimination from 3 affords
alkyl fluoride 5 as the major product in 49% yield. However, we
hypothesized that the selectivity could be shifted to favor C−N
reductive elimination product 4 by adding TsNH− to the
reaction mixture. This hypothesis was predicated on a study of
a related sp3-C−N bond-forming reductive elimination from
the PtIV complex fac-(dppbz)Pt(CH3)3(NHR). In this system,
the addition of exogeneous RNH− led to a dramatic increase in
the yield of C−N coupled product CH3NHR relative to that of
C−C coupled product H3C−CH3.4 Gratifyingly, an analogous
effect is observed with complex 3. Heating 3 in MeCN at 65 °C
in the presence of 1 equiv of NMe4NHTs leads to 4 as the sole
detectable reductive elimination product (Table 1, entry 2).
The yield of 4 is reasonably high (85%); however, free
Scheme 4 shows a mechanism for reductive elimination from
3a that is consistent with all of the data presented above.20 Step
i of this process involves pre-equilibrium dissociation of RNH−
Scheme 3. Substitution of the Sulfonamide Ligand
4098
dx.doi.org/10.1021/ja411433f | J. Am. Chem. Soc. 2014, 136, 4097−4100

Journal of the American Chemical Society
Communication
Scheme 4. Proposed Mechanism for sp3-C−N and sp3-C−F
Reductive Elimination from 3-a
of 3 (yields of 4−6 were 11%, 40%, and 29%, respectively,
under these conditions).
In summary, this communication describes the design of a
model complex for studying competing sp3-C−F and sp3-C−
NHTs bond-forming reductive elimination from PdIV. These
studies provide preliminary insights into the role of reaction
additives on these processes. Ongoing studies are focused on
obtaining additional mechanistic data on these transformations
as well as applying the insights learned from these studies to
catalysis.
ASSOCIATED CONTENT
* Supporting Information
■
S
Experimental procedures and complete characterization data
for all new compounds. This material is available free of charge
via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
mssanfor@umich.edu
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
to afford five-coordinate cationic intermediate 7. Importantly,
analogous five-coordinate intermediates are involved in the vast
majority of reductive elimination reactions from octahedral
PdIV and PtIV complexes.1c,21 Furthermore, the results in
Scheme 3 demonstrate the feasibility of step i under the
reductive elimination reaction conditions.
This work was supported by the National Science Foundation
Grant CHE-1111563. We also acknowledge funding from NSF
Grant CHE-0840456 for X-ray instrumentation. M.H.P.
gratefully acknowledges support from a Foundation Ramon
Areces Fellowship.
́
Proceeding in the forward direction, intermediate 7 is
proposed to participate in three competing reductive
elimination reactions: (1) C−C bond-forming reductive
elimination to release 6 (step iia); (2) C−F coupling to
generate 8 (step iib);22 and (3) C−N bond formation (likely by
an SN2 pathway1,4−6) to generate 9 (step iic). To complete the
reaction sequence, coordination of RNH− to 8 (step iii) would
yield 5, while loss of HF from 9 (step iv) would produce 4.
The observed effect of exogenous NMe4NHTs on the
product distribution is consistent with the mechanism proposed
in Scheme 4. The rates of C−C and C−F reductive elimination
are expected to be inverse order in [TsNH−], while that of C−
N coupling is zero order in [TsNH−]. As a result, the addition
of NMe4NHTs should lead to an increase in selectivity for 4
over 5 and 6, as observed.
The proposed mechanism suggests that the addition of acids
should also impact reductive elimination selectivity. In
particular, acids that can protonate TsNH− (pKa of TsNH2 =
16 in DMSO)23 should remove this nucleophile from solution,
thereby leading to less of product 4 and more of 5/6. Indeed,
the addition of 1 equiv of HF (pKa = 16 in DMSO, added as 1
equiv of NEt3·3HF)23 to reductive elimination reactions of 3
resulted in the formation of 5 and 6 as the major products
(yields of 4−6 were <1%, 32%, and 68%, respectively, under
these conditions). Similarly, <1% of the C−N reductive
elimination product 4 was observed upon the addition of 1
equiv of HOTf (pKa = 0.3 in DMSO; yields of 4−6 were <1%,
<1%, and 95%, respectively, under these conditions).24 This
mechanism also leads us to predict that H2O should not
appreciably impact the product ratios, since its pKa is too high
to protonate TsNH− (pKa of H2O in DMSO = 32).23 Indeed, a
nearly identical product distribution was observed upon the
addition of 10 equiv of H2O to reductive elimination reactions
REFERENCES
■
(1) For recent reviews on carbon−heteroatom bond-forming
reductive elimination, see: (a) Vigalok, A. Chem.Eur. J. 2008, 14,
5102. (b) Hartwig, J. F. Nature 200.8, 455, 314. (c) Racowski, J. M.;
Sanford, M. S. Top. Organomet. Chem. 2011, 35, 61.
(2) (a) Henkel, T.; Brunne, R. M.; Muller, H.; Reichel, F. Angew.
̈
Chem., Int. Ed. 1999, 38, 643. (b) Hili, R.; Yudin, A. K. Nat. Chem. Biol.
2006, 6, 284. (c) Collet, F.; Dodd, R. H.; Dauban, P. Chem. Commun.
2009, 34, 5061.
(3) (a) Streuff, J.; Hoevelmann, C. H.; Nieger, M.; Muniz, K. J. Am.
Chem. Soc. 2005, 127, 14586. (b) Minatti, A.; Muniz, K. Chem. Soc.
̃
̃
Rev. 2007, 36, 1142. (c) Muniz, K. J. Am. Chem. Soc. 2007, 129, 14542.
(d) Fraunhoffer, K. J.; White, M. C. J. Am. Chem. Soc. 2007, 129, 7274.
̃
(e) Muniz, K.; Hoevelmann, C. H.; Streuff, J. J. Am. Chem. Soc. 2008,
̃
130, 763. (f) Iglesias, A.; Perez, E. G.; Muniz, K. Angew. Chem., Int. Ed.
̃
2010, 49, 8109. (g) Zalatan, D. N.; Du Bois, J. Top. Curr. Chem. 2010,
292, 347. (h) He, G.; Zhao, Y.; Zhang, S.; Lu, C.; Chen, G. J. Am.
Chem. Soc. 2012, 134, 3. (i) Nadres, E. T.; Daugulis, O. J. Am. Chem.
Soc. 2012, 134, 7. (j) Martinez, C.; Muniz, K. Angew. Chem., Int. Ed.
2012, 51, 7031.
(4) Pawlikowski, A. V.; Getty, A. D.; Goldberg, K. I. J. Am. Chem. Soc.
2007, 129, 10382.
(5) (a) Marquard, S. L.; Rosenfeld, D. C.; Hartwig, J. F. Angew.
Chem., Int. Ed. 2010, 49, 793. (b) Hanley, P. S.; Marquard, S. L.;
Cundari, T. R.; Hartwig, J. F. J. Am. Chem. Soc. 2012, 134, 15281.
̃
(6) Iglesias, A.; Muniz, K. Helv. Chim. Acta 2012, 95, 2007.
(7) Reviews on high valent Pd in catalysis: (a) Muniz, K. Angew.
̃
̃
Chem., Int. Ed. 2009, 48, 9412. (b) Canty, A. J. Dalton. Trans. 2009, 47,
10409. (c) Sehnal, P.; Taylor, R. J. K.; Fairlamb, I. J. S. Chem. Rev.
2010, 110, 824. (c) Xu, L.-M.; Li, B.-J.; Yang, Z.; Shi, Z.-J. Chem. Soc.
Rev. 2010, 39, 712. (d) Hickman, A. J.; Sanford, M. S. Nature 2012,
484, 177.
(8) For examples of Pd-catalyzed C−H fluorination using F+
reagents, see: (a) Hull, K. L.; Anani, W. Q.; Sanford, M. S. J. Am.
Chem. Soc. 2006, 128, 7134. (b) Wang, X.; Mei, T.-S.; Yu, J.-Q. J. Am.
4099
dx.doi.org/10.1021/ja411433f | J. Am. Chem. Soc. 2014, 136, 4097−4100

Journal of the American Chemical Society
Communication
Chem. Soc. 2009, 131, 7520. (c) Lou, S. J.; Xu, D. Q.; Xia, A. B.; Wang,
Y. F.; Liu, Y. K.; Du, X. H.; Xu, Z. Y. Chem. Commun. 2013, 49, 6218.
(9) Engle, K. M.; Mei, T.-S.; Wang, X. S.; Yu, J.-Q. Angew. Chem., Int.
Ed. 2011, 50, 1478.
(10) Mei, T.-S.; Wang, X.; Yu, J.-Q. J. Am. Chem. Soc. 2009, 131,
10806.
(11) (a) Sibbald, P. A.; Michael, F. E. Org. Lett. 2009, 11, 1147.
(b) Rosewall, C. F.; Sibbald, P. A.; Liskin, D. V.; Michael, F. E. J. Am.
Chem. Soc. 2009, 131, 9488. (c) Sibbald, P. A.; Rosewall, C. F.; Swartz,
R. D.; Michael, F. E. J. Am. Chem. Soc. 2009, 131, 15945. (d) Liskin, D.
V.; Sibbald, P. A.; Rosewall, C. F.; Michael, F. E. J. Org. Chem. 2010,
75, 6294.
(12) Iglesias, A.; Alvarez, A.; de Lera, A. R.; Muniz, K. Angew. Chem.,
Int. Ed. 2012, 51, 2225.
(13) Xu, T.; Qiu, S.; Liu, G. J. Organomet. Chem. 2011, 696, 46.
(14) Xiong, T.; Li, Y.; Mao, L.; Zhang, Q.; Zhang, Q. Chem. Commun.
2012, 48, 2246.
(15) Rare examples of C−F reductive elimination from PdIV: (a) Ball,
N. D.; Sanford, M. S. J. Am. Chem. Soc. 2009, 131, 3796. (b) Furuya,
T.; Benitez, D.; Tkatchouk, E.; Strom, A. E.; Tang, P.; Goddard, W. A.,
III; Ritter, T. J. Am. Chem. Soc. 2010, 132, 3793.
(16) Examples where C−F reductive elimination from PdIV is slow
relative to other processes: (a) Racowski, J. M.; Ball, N. D.; Sanford,
M. S. J. Am. Chem. Soc. 2011, 133, 18022. (b) Ball, N. D.; Gary, J. B.;
Ye, Y.; Kampf, J.; Sanford, M. S. J. Am. Chem. Soc. 2011, 133, 7577.
(17) Racowski, J. M.; Kampf, J. W.; Sanford, M. S. Angew. Chem., Int.
Ed. 2012, 51, 3414.
(18) Canty, A. J. Acc. Chem. Res. 1992, 25, 83.
(19) At room temperature, only 3a underwent fast exchange with
MsNH to form 3Ms-a. See SI for full details.
(20) An alternative pathway that is consistent with all of the data
would involve pre-equilibrium isomerization of 3a to 3b followed by
direct C−N bond-forming reductive elimination from 3b. This
pathway is discussed in detail in the SI (p S15). We preliminarily
believe that this mechanism is less likely because the vast majority of
reductive elimination reactions from PdIV proceed via five-coordinate
intermediates (see ref 21).
(21) Bartlett, K. L.; Goldberg, K. I.; Borden, W. T. J. Am. Chem. Soc.
2000, 122, 1456.
(22) See ref 17 for a detailed discussion of the mechanism of C−F
reductive elimination from 1, which is believed to proceed through
intermediate 7.
(23) http://evans.harvard.edu/pdf/evans_pka_table.pdf. The pKa for
TsNH2 is approximated based on that of PhSO2NH2.
(24) The impact of reaction additives on the relative ratio of C−F
and C−C coupling products 5 and 6 is the subject of ongoing
investigations. Preliminary results suggest that H-bonding between the
additive and the fluoride ligand on PdIV may play a role in the
selectivity.
NOTE ADDED AFTER ASAP PUBLICATION
■
The amounts of reagents used in Scheme 1, and a typographical
error in the sentence describing the rates of reductive
elimination in the third-from-last paragraph, were corrected
after the ASAP version was published February 28, 2014. The
revised version was re-posted on March 5, 2014.
4100
dx.doi.org/10.1021/ja411433f | J. Am. Chem. Soc. 2014, 136, 4097−4100