Ligand and base additive effects on the reversibility of nucleophilic addition in palladium-catalyzed allylic aminations monitored by nucleophile crossover

Article abstract of DOI:10.1016/j.tetlet.2016.12.049

A nucleophile crossover experiment was used to monitor the reversibility of nucleophilic addition of benzylamine to π-allylpalladium complexes. Dppe, dppp, dppb, and PHOX showed more crossover than PPh₃and dppm in both DMF and dichloromethane. Crossover was inhibited by the addition of DBU or Cs₂CO₃, but much less elimination to diene side products was observed with Cs₂CO₃. Analysis of percent crossover vs. percent reaction completion using the PHOX ligand revealed that with added DBU or Cs₂CO₃crossover only began occurring after 100% completion had been reached.

Full text of DOI:10.1016/j.tetlet.2016.12.049

Accepted Manuscript
Ligand and base additive effects on the reversibility of nucleophilic addition in
palladium-catalyzed allylic aminations monitored by nucleophile crossover
Bryan S. Holtzman, Eric T. Roberts, Nicholas S. Caminiti, Jacob A. Fox,
Madison B. Goodstein, Staci A. Hill, Zitong B. Jia, Isabelle N.-M. Leibler,
Michael L. Martini, Gina M. Mendolia, Sarah B. Nodder, Molly S. Costanza-
Robinson, Richard C. Bunt
PII:
S0040-4039(16)31697-5
DOI:
http://dx.doi.org/10.1016/j.tetlet.2016.12.049
TETL 48461
Reference:
Tetrahedron Letters
To appear in:
Received Date:
Accepted Date:
15 November 2016
19 December 2016
Please cite this article as: Holtzman, B.S., Roberts, E.T., Caminiti, N.S., Fox, J.A., Goodstein, M.B., Hill, S.A., Jia,
Z.B., Leibler, I.N., Martini, M.L., Mendolia, G.M., Nodder, S.B., Costanza-Robinson, M.S., Bunt, R.C., Ligand
and base additive effects on the reversibility of nucleophilic addition in palladium-catalyzed allylic aminations
monitored by nucleophile crossover, Tetrahedron Letters (2016), doi: http://dx.doi.org/10.1016/j.tetlet.2016.12.049
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Graphical Abstract
To create your abstract, type over the instructions in the template box below.
Fonts or abstract dimensions should not be changed or altered.
Ligand and base additive effects on the
reversibility of nucleophilic addition in
palladium-catalyzed allylic aminations
monitored by nucleophile crossover.
Leave this area blank for abstract info.
Bryan S. Holtzmana, Eric T. Robertsa, Nicholas S. Caminitia, Jacob A. Foxa, Madison B. Goodsteina, Staci A.
Hilla, Zitong B. Jiaa, Isabelle N. –M . Leiblera, Michael L. Martinia, Gina M. Mendoliaa, Sarah B. Noddera,
Molly S. Costanza-Robinsonb and Richard C. Bunta,

1
Tetrahedron Letters
Ligand and base additive effects on the reversibility of nucleophilic addition in
palladium-catalyzed allylic aminations monitored by nucleophile crossover.
Bryan S. Holtzmana, Eric T. Robertsa, Nicholas S. Caminitia, Jacob A. Foxa, Madison B. Goodsteina, Staci
A. Hilla, Zitong B. Jiaa, Isabelle N. –M. Leiblera, Michael L. Martinia, Gina M. Mendoliaa, Sarah B.
Noddera, Molly S. Costanza-Robinsonb and Richard C. Bunta,
a Department of Chemistry & Biochemistry, Middlebury College, Middlebury, Vermont 05753, United States.
b Department of Chemistry & Biochemistry and Program in Environmental Studies, Middlebury College, Middlebury, Vermont 05753, United States.
ARTICLE INFO
ABSTRACT
Article history:
Received
Received in revised form
Accepted
A nucleophile crossover experiment was used to monitor the reversibility of nucleophilic
addition of benzylamine to -allylpalladium complexes. Dppe, dppp, dppb, and PHOX showed
more crossover than PPh3 and dppm in both DMF and dichloromethane. Crossover was inhibited
by the addition of DBU or Cs2CO3, but much less elimination to diene side products was
observed with Cs2CO3. Analysis of percent crossover vs. percent reaction completion using the
PHOX ligand revealed that with added DBU or Cs2CO3 crossover only began occurring after
100% completion had been reached.
Available online
Keywords:
Palladium catalysis
Allylic amination
Reversibility
2009 Elsevier Ltd. All rights reserved.
Ligand effects
Elimination
Palladium-catalyzed allylic aminations have been widely
studied due to their synthetic utility1 and the increasing
availability of highly enantioselective versions.2 We recently
reported that reversible nucleophilic addition can lower the
observed enantioselectivity of allylic amination reactions with a
wide range of prototypical chiral ligands.3 Amatore, Jutand, et
al.4a have found that nucleophilic addition of amines was
reversible with the bidentate ligand dppb but irreversible with the
monodentate ligand PPh3. Based on our interest in chiral ligand
design and specific studies with PHOX ligands,5,6 we wanted to
investigate the key ligand, solvent, and base additive factors
governing the reversibility of nucleophilic addition of amines.
benchmark test reaction.7 Phenyl methyl substrate 2a was added
as a co-reactant along with 1a and 4 because our initial
investigations established the necessity of an actively functioning
catalyst to observe changes in the product ee. The reaction of 2a
also produced varying amounts of elimination8 product 5 that
depended on the reaction conditions and presence of added base9.
In some cases we also observed formation of 2c, which resulted
either from reaction of newly formed 3 with 2a or from
“secondary” crossover of 2b. Because 2c was formed in very low
amounts and only at longer reaction times, we defined the percent
crossover and elimination in terms of 1a/b, 2a/b, and 5 only (eq.
1-2).10 We quantified the amounts of all compounds by GC-MS.11
Percent Crossover = [1b / (1a + 1b)]  100
Percent Elimination = [5 / (2a + 2b + 5)]  100
(1)
(2)
Our previous study3 of chiral ligands used time-dependent
changes in the product ee to monitor the reversibility. Herein, we
employed a nucleophile crossover experiment that provided a
quantitative measure of product reversibility and allowed us to
study both achiral and chiral ligands. Specifically, we isolated the
back-reaction of 1a to reform the -allylpalladium intermediate
(i.e., reverse of nucleophilic addition) and trapped it with a
different amine nucleophile (Scheme 1). We focused on 1a due to
the widespread use of the 1,3-diphenylallyl system as a
Scheme 1. Crossover reaction design.
 Corresponding author. Tel.: +1-802-443-2559; fax: +1-802-443-2072; e-mail: rbunt@middlebury.edu

2
Tetrahedron Letters
.
Scheme 2. Nucleophile crossover mechanism.
The crossover mechanism is outlined in Scheme 2. As with
and prevents it from re-entering the catalytic cycle that leads to
crossover (Scheme 2). The impact of the added base on
elimination, however, was more complex. With DBU the amount
of elimination product (5) was increased in all cases, consistent
with general finding that amine bases promote elimination from
-allylpalladium intermediates.8 In contract, with Cs2CO3
increased elimination was only observed with PPh3 and dppm,
the two ligands that showed minimal crossover without added
base. We believe this correlation is mechanistically significant
and has a unifying explanation.
the metal-catalyzed isomerization of branched allylic amines to
linear isomers elucidated by Yudin,9,12 formation of 1b and free
benzylamine (3) are explained by 1a re-entering the catalytic
cycle to produce
a catalytically active -allylpalladium
intermediate (6). Thus, detection of crossover products serves as
a quantitative marker for the reversibility of an amination
reaction that would have initially formed 1a as a product. We
also observed crossover in the other direction—starting with 1b
and 3 to produce 1a and 4 as crossover products—to establish
that this effect was not just a result of the methoxy group label.13
Table 2.
We initially studied PPh3 and the series of bidentate analogs
with one to four carbons in the tether (dppm to dppb) in both
DMF and dichloromethane (DCM) (Table 1). The solvent choice
had little impact on the trends in crossover or elimination. This
result supports extending the kinetics findings of Amatore,
Jutand, et al. in DMF4a to less polar solvents commonly
employed in enantioselective reactions.7 The amount of
crossover, however, was very different among the ligands. Very
little crossover was observed for PPh3 and dppm in contrast to
dppe, dppp, and dppb, which showed much higher levels of
crossover. The dppm result was surprising based on the
hypothesis that amination is irreversible with bidentate ligands.4a
Crossover and elimination with achiral ligands in the presence
of added basesa
Ligand Solvent
Baseb
DBU
Crossover (%) Elimination (%)
PPh3
dppm
dppe
dppp
dppb
PPh3
dppm
dppe
dppp
dppb
PPh3
dppm
dppe
dppp
dppb
DMF
DMF
DMF
DMF
DMF
DCM
DCM
DCM
DCM
DCM
DCM
DCM
DCM
DCM
DCM
<1
<1
<1
<1
<1
<1
<1
<1
<1
<1
<1
<1
<1
<1
<1
24
21
50
27
50
30
29
31
44
86
38
27
5
DBU
DBU
DBU
DBU
DBU
Table 1.
DBU
Crossover and elimination with achiral ligandsa
DBU
Ligand
PPh3
dppm
dppe
dppp
dppb
PPh3
dppm
dppe
dppp
dppb
Solvent
DMF
DMF
DMF
DMF
DMF
DCM
DCM
DCM
DCM
DCM
Crossover (%)
Elimination (%)
DBU
3
2
2
DBU
2
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
65
63
83
4
1
<1
2
5
5
7
4
6
(a) Conditions: 40 C, 4 h, 10% Pd, average of 36 trials. (b) 3.2 equivalents.
43
43
79
<1
<1
<1
The amount of crossover observed with PPh3 (low) and dppb
(high) in the absence of added base are entirely consistent with
the kinetic findings of Amatore, Jutand, et al. on amination
reversibility.4a However their explanations—the higher
nucleophilicity of alkyldiarylphosphinepalladium(0) complexes
and the smaller bite angle of dppb favoring oxidative addition for
steric reasons—do not account for the low levels of crossover
observed with dppm, which is an alkyl diaryl phosphine and
should have the smallest bite angle of the bidentate ligands. The
(a) Conditions: 40 C, 4 h, 10% Pd; average of 36 trials.
When additional base (DBU or Cs2CO), beyond the excess
4-methoxybenzylamine (4), was added to the reaction, crossover
was almost completely suppressed (Table 2). Just as Yudin
showed for amine isomerizations9,14 and we proposed for
enantioselective reactions,3 added base limits protonation of 1a

3
low crossover (i.e., reversibility) observed with dppm could be
explained by the formation of off-cycle complexes that lower the
concentration and thus net reactivity of the active catalyst.
Dppm,15 in contrast to dppe, dppp, and dppb, readily forms
dinuclear palladium bridging complexes such as 7.16,17,18
effectively prevented crossover with the PHOX ligand (Table
3). Also as observed with the achiral ligands, use of Cs2CO3
results in much less elimination than DBU does. Elimination is
not an issue for the canonical 1,3-diphenylallyl test substrate (i.e.,
reactions leading to 1a as product), but for substrates that have -
hydrogens to eliminate, Cs2CO3 would be a much better base
additive.
Table 3.
Crossover and elimination with PHOXa
Ligand Solvent
Baseb
–
Crossover (%) Elimination (%)
PHOX
PHOX
PHOX
PHOX
PHOX
PHOX
DMF
DMF
DMF
DCM
DCM
DCM
14
<1
<1
29
<1
2
1
48
7
We conducted a crossover reaction with dppm in CD2Cl2 and
monitored it by 31P NMR to investigate possible off-cycle
complexes. We observed a strong 31P chemical shift at -3.0 ppm,
along with a smaller peaks at 13.5 ppm and 25 ppm. The -3.0
ppm shift corresponds quite closely to the reported value19 of -2.5
ppm for compound 7. The 13.5 ppm shift (initially larger than the
25 ppm peak) was tentatively assigned to alkene complex 8, the
most likely resting state of the catalytic cycle.4a,20 Other palladium
complexes of the general form Pd(0)Ln (9) are possibly
responsible for the 25 ppm shift, but we did not investigate them
further. It is unlikely that these shifts arise from the -
allylpalladium complex (10) as the reported chemical shifts of the
analogous dppe complex (phosphines not equivalent) are 46.3
and 49.0 ppm.21 Formation of complex 7 (or other off-cycle
palladium complexes) with dppm would result in a less active /
slower ligand/catalyst system and explain the lower amount of
crossover observed. This explanation is similar to what Amatore,
Jutand, et al. proposed for PPh3 due to steric factors.4a
DBU
Cs2CO3
–
1
DBU
Cs2CO3
64
4
(a) Conditions: 40 C, 4 h, 10% Pd.
In addition to providing insight into elimination issues, the
conversion of 2a to amine product 2b and elimination product 5
(eq. 3) also provided a way to analyze the crossover data that
offered additional insight into the catalyst behavior.
Percent Conversion = [(2b + 5) / (2a + 2b + 5)]  100
(3)
We ran crossover reactions with the PHOX ligand for varying
amounts of time (5 min to 4 h in DMF, 5 min to 36 h in DCM) to
obtain a range of percent conversion values without and with
added DBU or Cs2CO3. The graphs of percent crossover vs.
percent conversion look generally similar in DMF (Figure 1) and
DCM (Figure 2). Crossover products started appearing (> 1%)
around 50% conversion and increased steadily thereafter. In the
presence of either DBU or Cs2CO3 little to no crossover was
observed prior to 100% conversion.
The lower catalyst activity with PPh3 and dppm also accounts
for the unusual elimination results with Cs2CO3. Elimination to 5
can occur directly from 2a or from -allylpalladium
intermediate8 11 prior to formation of amination product 2b
(Scheme 3). In the absence of additional base, the background
elimination pathway is slower than the palladium-catalyzed
amination pathway (data in Table 1). Soluble organic bases such
as Et3N8a and DBU8b are known to accelerate elimination from
-allylpalladium intermediates, which is consistent with our
observations (data in Table 2). Added base also accelerates the
amination of 11 by providing more unprotonated 4 in solution to
serve as nucleophile.22 With DBU these two pathways for 11 are
more evenly competitive. With the heterogeneous Cs2CO3 base
the amination pathway is accelerated more when dppe, dppe, or
dppb are used as the ligand. Only with PPh3 and dppm as ligands
are higher levels of elimination observed. If these catalyst
systems are slower, as hypothesized, then the background
elimination reaction, also likely faster with added base, can
become more significant compared to the net flux through 11.
Figure 1. Correlation of of crossover with conversion with PHOX ligand
in DMF with and without added bases. Reaction times 5 min to 4 h.
Scheme 3. Elimination pathways for 2a.
The success of our crossover experiment at elucidating these
subtle ligand and base effects prompted us to study crossover
with the PHOX ligand as a prototypical, privileged23 chiral
ligand. As with the achiral ligands both DBU and Cs2CO3

4
Tetrahedron
1.
1 (a) Trost, B. M.; Zhang, T.; Sieber, J. D. Chem. Sci., 2010, 1,
427–440; (b) Faller, J. W.; Wilt, J. C.; Organometallics 2005, 24,
5076–5083; (c) Trost, B. M.; Crawley, M. L.; Chem. Rev. 2003,
103, 2921–2943; (d) Johannsen, M.; Jørgensen, K. A.; Chem. Rev.
1998, 98, 1689–1708; (e) Togni, A.; Burckhardt, U.; Gramlich,
V.; Pregosin, P. S.; Salzmann, R. J. Am. Chem. Soc. 1996, 118,
1031–1037.
2.
3.
4.
2 Transition Metal Catalyzed Enantioselective Allylic Substitution
in Organic Synthesis, Kazmaier, U., Ed.; Topics in
Organometallic Chemistry 38; Springer-Verlag: Berlin, 2012.
3 Caminiti, N. S.; Goodstein, M. B.; Leibler, I. N.-M.; Holtzman,
B. S.; Jia, A. B.; Martini, M. L.; Nelson, N. C.; Bunt, R. C.
Tetrahedron Lett. 2015, 56, 5445–5448.
4 (a) Amatore, C.; Genin, E.; Jutand, A.; Menash, L.
Organometallics 2007, 26, 1875–1880. (b) Amatore, C.; Jutland,
A.; Mensah, L.; Mayer, G.; Fiaud, J.-C.; Legros, J.-Y. Eur. J. Org.
Chem. 2006, 11851192. Amatore, C.; Bahsoun, A. A.; Jutland,
A.; Mensah, L.; Meyer, G.; Ricard, L. Organometallics 2005, 24,
1569–1577. (d) Amatore, C.; Gamez, S.; Jutland, A.; Meyer, G.;
Moreno-Manas, M.; Morral, L.; Pleixats, R. Chem. Eur. J. 2000,
6, 33723376.
Figure 2. Correlation of of crossover with conversion with PHOX ligand
in DCM with and without added bases. Reaction times 5 min to 36 h.
5.
5 (a) Armstrong, P. B.; Dembicer, E. A.; Desbois, A. J.; Fitzgerald,
J. T.; Gehrmann, J. K.; Nelson, N. C.; Noble, A. L.; Bunt, R. C.
Organometallics 2012, 31, 6933–6946; (b) Armstrong, P.B.;
Bennett, L.M.; Constantine, R.N.; Fields, J.L.; Jasinski, J.P.;
Staples, R.J.; Bunt, R.C. Tetrahedron Lett. 2005, 46, 1441–1445;
(c) Constantine, R. N.; Kim, N.; Bunt, R. C. Org. Lett. 2003, 5,
2279–2282.
At 100% conversion, the amount of crossover was quite
variable—the graph has no measure of how far or long past
completion the reaction had proceeded. In DCM in particular we
allowed several reactions to go long past (24-36 h) the time
necessary for complete reaction of 2a. In these cases much larger
amounts of crossover were obtained, even in the presence of
added base (although less crossover than without base). Based on
the 3:1 net ratio of 4/3 in the reaction system, 70% crossover is
likely close to the thermodynamic mixture of 1b/1a.24
6.
7.
6 For reviews of PHOX ligands see: (a) Helmchen, G.; Pfaltz, A.
Acc. Chem. Res. 2000, 33, 336–345; (b) Helmchen, G. J.
Organomet. Chem. 1999, 576, 203–214; (c) Williams, J. M. J.
Synlett 1996, 705–710.
7 (a) Zagidullin, A.; Miluykov, V.; Sinyashin, O.; Hey-Hawkins,
E. Catal. Today 2017, 279, 142146; (b) Clavero, P.; Grabulosa,
A.; Rocamora, M.; Muller, G.; Font-Bardia, M. Eur. J. Inorg.
Chem. 2016, 4054-4065; (c) Théveau, L.; Bellini, R.; Dydio, P.;
Szabo, A.; van der Weft, A.; Sander, R. A.; Reek, J. N. H.;
Moberg, C. Organometallics 2016, 35, 19561963; (d) Liu, Q. –
L.; Chen W.; Jiang, Q. –Y.; Bai, X. –F.; Li, Z.; Xu, Z. Xu, L.-W.
ChemCatChem 2016, 8, 14951499; (e) Rovira, L.; Fernández-
Pérez, H.; Vidal-Ferran, A. Organometallics 2016, 35, 528533;
(f) Zhang, R.; Xie, B.; Che, G. –S.; Qiu, L.; Chen, Y. –X.
Tetrahedron Lett. 2016, 57, 845848; (g) Magre, M.; Biosca, M.;
Norrby, P. –O. Pàmies, O.; Diéguez, M. ChemCatChem 2015, 7,
4091-4107; (h) Bayardon, J.; Maronnat, M.; Langlois, A.;
Rousselin, Y.; Harvery, P. D.; Jugé, S. Organometallics 2015, 34,
43404358; (i) Schmitz, C.; Leitner, W.; Franciò, G. Eur. J. Org.
Chem. 2015, 62056230; (j) Bravo, M. J.; Ceder, R. M.
Grabulosa, A.; Muller, G.; Rocamora, M.; Bayón, J. C.; Peral, D.
Organometallics 2015, 34, 37993808; (k) Wu, H.; Xie, F.;
Wang, Y.; Zhao, X.; Liu, D.; Zhang, W. Org. Biomol. Chem.,
2015, 13, 4248–4254; (l) Coll, M.; Pàmies, O.; Diéguez. M. Org.
Lett. 2014, 16, 1892–1895; (m) Grabulosa, A.; Doran, S.;
Brandariz, G.; Muller, G.; Benet-Buchholz, J.; Riera, A.;
Verdaguer, X. ; Organometallics, 2014, 33, 692–701; (n) Bravo,
M. J.; Favier, I, Saffon, N.; Ceder, R. M.; Muller, G.; Gómez, M.;
Rocamora, M. Organometallics, 2014, 33, 771–779; (o) Khiri-
Meribout, N.; Bertrand, E.; Bayardon, J.; Eymin, M. –J.;
Rousselin, Y.; Cattery, H.; Fortin, D.; Harvey, P. D.; Jugé, S.
Organometallics, 2013, 32, 2827–2839; (p) Wang, Y.; Vaismaa,
M. J. P.; Rissanen, K.; Franzen, R. Eur. J. Org. Chem. 2012,
1569–1576; (q) Harada, T.; Nemoto, T.; Jin, L.; Hamada, Y.
Chem. Pharm. Bull. 2011, 59, 412–415; (r) Sun, Y. –W.; Jiang, J.
–J.; Zhoa, M. –X.; Wang, F. –J.; Shi, M. J. Organomet. Chem.
2011, 696, 2850–2856; (s) Kultyshev, R. G.; Miyazawa, A.
Tetrahedron 2011, 67, 2139–2148; (t) Shen, C.; Xia, H.; Zheng,
H.; Zhang, P.; Chen, X. Tetrahedron: Asymmetry 2010, 21, 1936–
1941; (u) Panossian, A.; Fernandez-Perez, H.; Popa, D.; vidal-
Ferran, A. Tetrahedron: Asymmetry 2010, 21, 2281–2288; (v)
Caldentey, X.; Pericas, M. A. J. Org. Chem. 2010, 75, 2628–2644;
(w) Mazuela, J.; Paptchikhine, A.; Tolstoy, P.; Pamies, O.;
Dieguez, M; Andersson, P. G. Chem.—Eur. J. 2010, 16, 620–638.
8 (a) Ogawa, R.; Shigemori, Y.; Uehara, K.; Sano, J.; Nakajima,
T.; Shimizu, I. Chem. Lett. 2007, 36, 1338–1339; (b) Takacs, J.
M.; Lawson, E. C.; Clement, F. J. Am. Chem. Soc. 1997, 119,
Overall, these conversion vs. crossover graphs provide a
consistent picture of the PHOX catalyst behavior that extends
what was learned from the achiral ligand experiments. The
catalysts all prefer allyl carbonate 2a to allyl amine 1a as a
substrate. In the presence of additional base (DBU or Cs2CO3)
this preference is nearly absolute. In the absence of additional
base or at very long reaction times, the catalysts will more slowly
start converting 1a to 1b via -allylpalladium intermediate 6.
These findings substantiate our explanation that added base can
increase the observed enantioselectivity in allylic amination
reactions by preventing product equilibration through reversible
nucleophilic addition.3 This effect is now clearly understood in
terms of the crossover reaction mechanism which provides direct
evidence for the reversibility of nucleophilic addition under
similar reaction conditions. Finally, Cs2CO3 is a better base
additive than DBU if substrate elimination to form unwanted
diene side products is possible. We are currently using the
crossover reaction to investigate the generality of these findings
with other chiral ligands.
Acknowledgments
We thank the Donors of The Petroleum Research Fund,
administered by the American Chemical Society, and the
National Science Foundation (CHE-071451) for support of this
research. N. S. C., M. B. G., and I. N.-M. L. thank Middlebury
College for summer financial support.
Supplementary Material
Supplementary data associated with this article can be found,
in the online version, at doi.
8.
References and notes

5
22. 22 The reaction goes to completion (as judged by TLC and GC-
MS) in 2-3 h without added base and 20-30 min with DBU or
Cs2CO3 see reference 3 for data.
5956–5957; (c) Andersson, P. G.; Schab, S. Organometallics,
1995, 14, 1-2; (d) Tsuji, J.; Yamakawa, T.; Kaito, M.; Mandai, T.
Tetrahedron Lett. 1978, 2075–2078.
23. 23 Privileged Chiral Ligands and Catalysts; Zhou, Q. –L., Ed.;
Wiley-VCH: Weinheim, Germany, 2011.
9.
9 (a) Dubovyk, I.; Watson, I. D. G.; Yudin, A. K. J. Org. Chem.
2013, 78, 1559–1575; (b) Dubovyk, I.; Pichugin, D.; Yudin, A. K.
Angew. Chem. Int. Ed. 2011, 50, 5924–5926; (f) De luliis, M. Z.;
Watson, I. D. G. ; Yudin, A. K.; Morris, R. H. Can. J. Chem.
2009, 87, 54–62. (g) Dubovyk, I.; Watson, I. D. G.; Yudin, A. K.
J. Am. Chem. Soc. 2007, 129, 14172–114173. (h) Watson, I. D.
G.; Yudin, A. J. Am. Chem. Soc. 2005, 127, 17516–17529.
24. 24 75% crossover would be the purely statistical mixture ignoring
the differences in basicity of 4 and 3 and free energy differences
between 1b and 1a.
10. 10 Inclusion of 2c in the calculation, when observed, changes the
data by <1%.
11. 11 For examples of mass spectrometric catalyst screening see: (a)
Ebner, C.; Müller, C. A.; Markert, C.; Pfaltz, A.; J. Am. Chem.
Soc. 2011, 133, 47104713. (b) Müller, C. A.; Markert, C.;
Teichert, A. M.; Pfaltz, A.; Chem. Commun. 2009, 16071618. (c)
Martha, C. T.; Elders, N.; Krabbe, J. G.; Kool, J.; Niessen W. M.
A.; Orru; R. V. A.; Irthi, H. Anal. Chem. 2008, 80, 71217127. (d)
Müller, C. A.; Pfaltz, A.; Angew. Chem. 2008, 120, 34113414.
(e) Chen, P. Angew. Chem., Int. Ed., 2003, 42, 28322847.
12. 12 For other examples of allyl amine isomerization see: (a) Yang,
M. –H.; Orsi, D. L.; Altman, R. A. Angew. Chem. Int. Ed. 2015,
54, 23612365; (b) Ferraccioli, R.; Pignataro, L. Curr. Org.
Chem. 2015, 19, 106120; (c) Pieri, C.; Combes, S.; Brunel, J. M.
Tetrahedron 2014, 70, 97189725; (d) Kawatsura, M.; Terasaki,
S.; Minakawa, M.; Hirakawa, T.; Ikeda, K.; Itoh, T. Org. Lett.
2014, 16, 24422445; (e) Hirakawa, T.; Kawatsura, M.; Itoh, T. J.
Fluorine Chem. 2013, 152, 62 69; (f) Wang, X.; Meng, F.;
Wang, Y.; Han, Z.; Chen, Y. –J.; Liu, L.; Wang, Z.; Ding, K.
Angew. Chem. 2012, 124, 9410–9416; (g) Wang, Y.; Liu, L.;
Wang, D.; Chen, Y. –J. Org. Biomol. Chem. 2012, 10, 6908–6913;
(h) Hirakawa, T.; Ikeda, K.; Ikeda, D.; Tanaka, T.; Ogasa, H.;
Kawatsura, M.; Itoh, T. Tetrahedron 2011, 67, 8238–8247.
13. 13 In the “reverse” direction larger total amounts of crossover were
observed (data not shown), likely due to the increased nitrogen
basicity of 1b vs. 1a.
14. 14 For an example of bases effects on ruthenium-catalyzed allylic
amination see: Kawatsura, M.; Uchida, K.; Terasaki, S.; Tsuji, H.;
Minakawa, Itoh, T. Org. Lett. 2014, 16, 14701473.
15. 15 (a) Murahashi, T.; Kurosawa, H. Coord. Chem. Rev. 2002, 231,
207228; (b) Chaudret, B.; Delavaux, B.; Poilblanc, R. Coord.
Chem. Rev. 1988, 86, 191243; (c) Puddephatt, R. J. Chem. Soc.
Rev. 1983, 12, 99127.
16. 16 For the initial isolation of 7 see: (a) Holloway, R. G.; Penfold,
B. R.; Colton, R.; McCormick, M. J. J. Chem. Soc., Chem.
Commun. 1976, 485486; (b) Colton, R. Farthing, R. H.;
McCormick, M. J. Aust. J. Chem. 1973, 26, 26072614.
17. 17 For examples of 7 and related complexes see: (a) Burrows, A.
D.; Mahon, M. F.; Nolan, S. P.; Varrone, M. Inorg. Chem. 2003,
42, 72277238; (b) Besenyei, G.; Párkányi, L.; Gács-baitz, Z.;
James, B. R. Inorg. Chim. Acta 2002, 327, 179187; (c) Lee, C. –
L.; Yang, Y. –P.; Rettig, S. J.; James, B. R.; Nelson, D. A.; Lilga,
M. A. Organometallics 1986, 5, 22202228; (d) Maekawa, M.;
Munakata, M.; Kuroda-Sowa, T.; Suenaga, Y. Polyhedron 1998,
17, 36573663.
18. 18 For examples of di- and polynuclear palladium allyl complexes
see: (a) Hazari, N.; Hruszkewycz, D. P. Chem. Soc. Rev. 2016, 45,
28712899; (b) Markert, C.; Neuburger, M.; Kulicke, K.;
Meuwly, M.; Pfaltz, A. Angew. Chem. Int. Ed. 2007, 46,
58925895; (c) Tromp, M.; Sietsma, J. R. A.; van Bokhoven, A.
A.; van Strijdonck, G. P. F.; van Haaren, R. J.; van der Eerden, A.
M. J.; van Leeuwen, P. W. N. M.; Koningsgerger, D. C. Chem.
Commun. 2003, 128129l; (d) Hayashi, Y.; Matsumoto, K.;
Nakamura, Y. J. Chem. Soc. Dalton Trans. 1989, 15191525.
19. 19 (a) Lindsay, C. H.; Benner, L. S.; Balch, A. L. Inorg. Chem.
1980, 19, 35033508; (b) Benner, L. S.; Balch, A. L. J. Am.
Chem. Soc. 1978, 100, 60996106.
20. 20 Evans, L. A.; Fey, N.; Harvey, J. N.; Hose, D.; Lloyd-Jones, G.
C.; Murray, P.; Orpen, A. G.; Osbourne, R.; Owen-Smith, G. J. J.;
Purdie, M. J. Am. Chem. Soc. 2008, 130, 1447114471.
21. 21 Schott, D.; Pregosin, P. P.; Veiros, L. F.; Calhorda, M. J.
Organometallics 2005, 24, 57105717.

Highlights
 Nucleophile crossover quantitates reversibility of benzylamine addition.
 DBU and Cs2CO3 inhibit crossover / reversibility with all ligands tested.
 Cs2CO3 causes much less elimination to dienes than DBU.
 Percent conversion analysis shows timing of crossover / reversibility.
 Results support reversibility as mechanism for base effects on enantioselectivity.