Direct Reductive Amination of Carbonyl Compounds Catalyzed by a Moisture Tolerant Tin(IV) Lewis Acid

Article abstract of DOI:10.1002/adsc.201701418

Despite the ever-broadening applications of main-group ‘frustrated Lewis pair’ (FLP) chemistry to both new and established reactions, their typical intolerance of water, especially at elevated temperatures (>100 °C), represents a key barrier to their mainstream adoption. Herein we report that FLPs based on the Lewis acid ⁱPr₃SnOTf are moisture tolerant in the presence of moderately strong nitrogenous bases, even under high temperature regimes, allowing them to operate as simple and effective catalysts for the reductive amination of organic carbonyls, including for challenging bulky amine and carbonyl substrate partners. (Figure presented.).

Full text of DOI:10.1002/adsc.201701418

Title: Direct Reductive Amination of Carbonyl Compounds Catalyzed
by a Moisture Tolerant Tin(IV) Lewis Acid
Authors: Joshua Sapsford, Daniel Scott, Nathan Allcock, Matthew J.
Fuchter, Christopher Tighe, and Andrew Ashley
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Adv. Synth. Catal. 10.1002/adsc.201701418
Link to VoR: http://dx.doi.org/10.1002/adsc.201701418

10.1002/adsc.201701418
Advanced Synthesis & Catalysis
COMMUNICATION
DOI: 10.1002/adsc.201((will be filled in by the editorial staff))
Direct Reductive Amination of Carbonyl Compounds Catalyzed
by a Moisture Tolerant Tin(IV) Lewis Acid
Joshua S. Sapsford,a Daniel J. Scott,a Nathan J. Allcock,a Matthew J. Fuchter,a
Christopher J. Tigheb and Andrew E. Ashley*,a
a
Department of Chemistry, Imperial College London, London SW7 2AZ, UK.
Department of Chemical Engineering, Imperial College London, London SW7 2AZ, UK.
b
Phone: +44 (0)20 759 45810. E-mail: a.ashley@imperial.ac.uk.
Received: ((will be filled in by the editorial staff))
Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/adsc.201######.
Abstract. Despite the ever-broadening applications of
main-group ‘frustrated Lewis pair’ (FLP) chemistry to both
new and established reactions, their typical intolerance of
water, especially at elevated temperatures (> 100°C),
represents a key barrier to their mainstream adoption.
Herein we report that FLPs based on the Lewis acid
iPr3SnOTf are moisture tolerant in the presence of
moderately strong nitrogenous bases, even under high
temperature regimes, allowing them to operate as simple
and effective catalysts for the reductive amination of
organic carbonyls, including for challenging bulky amine
and carbonyl substrate partners.
poison, forming highly Brønsted acidic adducts with
the LA [e.g. H2O·B(C6F5)3: pKa = 8.4 (MeCN), < 1
(aq., est.), similar to HCl].[3] Such adducts can be
irreversibly deprotonated by even moderately strong
bases (e.g., alkyl imines/amines) to the corresponding
oxyborate anions, which are catalytically inactive
(Scheme 1). Furthermore, susceptibility to
decomposition via B–C protonolysis at relatively
modest temperatures (> 100°C)[4] means that
reversibility cannot be imparted through heating
(which also restricts the upper operating temperature,
narrowing the opportunity to optimise rates of
conversion).
Keywords: ‘frustrated Lewis pairs’, catalytic
hydrogenation, water tolerance, reductive amination, tin
Nevertheless, in recent years we,[5] and others,[6]
have separately reported the development of borane-
based protocols for the catalytic hydrogenation of
organic carbonyls, that are tolerant of H2O and
alcohol products. Notably, however, in none of these
cases was moisture tolerance reported in the presence
of basic functional groups (e.g. imines/amines),
which is consistent with the need to avoid
deprotonation of H2O·LA, as discussed above. This
of course presents a serious drawback in terms of
reaction scope. For example, the reductive amination
(RA) of organic carbonyls is a powerful and versatile
C–N bond forming methodology that is a key route to
secondary and tertiary amines in many industrially-
important compounds; it has been reported that 20%
of target drugs in leading pharmaceutical companies
incorporate a RA step.[7] While various stoichiometric
reductants have been incorporated into these
reactions, from an atom economy perspective direct
RA using H2 as the reductant is especially attractive.
Homogeneous catalysts for RA typically use
precious TMs (e.g. Ru, Rh, Ir),[8] although a handful
of non-precious TM catalysts based on Fe or Cu have
been disclosed, all of which require high pressures,
anhydrous solvents and/or desiccants to perform
well.[9] In the quest for non-precious metal RA
Hydrogenations
catalyzed
by main-group
‘frustrated Lewis pairs’ (FLPs)[1] have attracted
enormous recent interest as potential alternatives to
the use of scarce, toxic, and expensive precious
transition metal (TM) catalysts. FLPs consist of
Lewis acid (LA) and Lewis base (LB) pairs which are
sterically precluded from irreversibly forming strong
classical adducts leading to unquenched reactivity
that can be utilised for bond activation processes. In
particular, heterolytic cleavage of H2 into protic [LB–
H]+ and hydridic [LA–H]– components can be
achieved and utilized for the polar hydrogenation of
various substrates. Early catalytic hydrogenation
protocols based on this reactivity[2] were established
for a variety of unsaturated organic functional groups
containing C=N and C=C bonds, almost exclusively
using organoboron-based LA catalysts, typified by
B(C6F5)3. Though they have provided a dramatic
proof-of-principle
for
TM-free
catalytic
hydrogenation, such systems suffer from a number of
common limitations. In particular, and in the vast
majority of cases, H2O (and other compounds
containing the hydroxyl group) is a potent catalyst
1
This article is protected by copyright. All rights reserved.

10.1002/adsc.201701418
Advanced Synthesis & Catalysis
catalyst candidates, main-group FLP systems seem
particularly appealing, given the status of imines as
the ‘archetypal’ FLP hydrogenation substrate.
However, successful RA necessarily requires H2O
tolerance in the presence of imine/amine bases.[10]
Very recently Soós et al. reported the first example of
FLP-catalysed RA (Scheme 2)[11] employing a
triarylborane as LA (I in Scheme 2), which is
impressive given the factors outlined above. The
authors noted, however, that “electronic tuning [in
BAr3 species] has reached its limit” for refining
moisture tolerance, and their success was based upon
very careful and specific design of the triarylborane
used, which focused on steric modification. This has
implications for reaction scope, which is known to be
highly dependent on LA structure.[2g] For example,
Soós’ borane design included the use of very high
steric bulk, even by FLP standards; consequently, the
reduction of bulky substrates was found to be
especially challenging. Thus, alternative and
complementary approaches to FLP-catalysed RA are
still desirable.
Scheme 2. Examples of previous moisture-tolerant FLP
hydrogenation systems relevant to this work.
Initially, we applied our protocol for carbonyl
hydrogenation with 1 under ‘wet’ conditions [10 bar
H2 (undried), reagent grade 1,2-dichlorobenzene
(DCB)] to archetypal imines PhC(H)=NPh (2a,
Scheme 3) and PhC(H)=NtBu (2b, Scheme 3). While
turnover can be successfully achieved at 120°C for
these substrates under anhydrous conditions,[12] when
moisture is present the temperature must be raised to
180°C to overcome its inhibitory effect, yet this is
made possible by the thermally robust nature of 1.
Perhaps unexpectedly, the use of either molecular
sieves (3 or 4 Å) or anhydrous MgSO4 as desiccants
proved to be deleterious to the reaction rate, which
we similarly ascribe to the competitive adsorption of
the Sn catalyst to the surface oxygen sites of these
materials.[6d] As when employing anhydrous
conditions, collidine (2,4,6-trimethylpyridine, Col;
pKa = 7.4 in H2O)[14] was required as an auxiliary
base only for 2a, which is too weakly basic to
activate H2 directly with 1 at a feasible rate;
conversely the higher basicity of 2b allows the imine
and product amine (pKa 10.5 in H2O)[15] to act as the
LBs for H2 cleavage. In this latter case, however, the
enhanced basicity also leads to a requirement for
longer reaction times, which we ascribe to increased
deprotonation of the aqua species [iPr3Sn·2H2O]+
(pKa = 6.37 in aqueous EtOH)[16] to off-cycle
iPr3SnOH/ (iPr3Sn)2O, thus reducing the concentration
of the active LA catalyst. Encouragingly, despite
observing partial hydrolysis of 2a/2b to PhCHO and
PhNH2/tBuNH2 immediately upon dissolving at RT
Scheme 1. Detrimental effects of hydroxylic species upon
catalytic activity of B(C6F5)3. R = alkyl, H; LB = Lewis
base.
[a] Brønsted acidification of H2O via coordination to
B(C6F5)3.
[b] Thermally induced protodeboronation. Quoted pKa
relates to aqueous conditions (est.).[3]
We have recently adopted a different approach to
achieving ROH tolerance by switching to LAs based
on ‘softer’ p-block elements than B, and reported that
i
inexpensive and readily-synthesised Pr3SnOTf (1; Tf
= SO2CF3) is a versatile catalyst for the FLP-type
hydrogenation of C=N, C=O and C=C bonds
(Scheme 2).[12] We also briefly noted that this LA
showed appreciable moisture tolerance for the
hydrogenation of acetone. Herein we extend our
i
initial study and demonstrate that Pr3SnOTf is an
effective RA catalyst for both aryl and alkyl amine
substrates with either aldehyde or ketone coupling
partners, using technical grade solvents and reagents
(i.e. ‘wet’ conditions), and without the need for
desiccants.[13]
2
This article is protected by copyright. All rights reserved.

10.1002/adsc.201701418
Advanced Synthesis & Catalysis
i
Table 1. Pr3SnOTf-catalysed hydrogenation of imines
under ‘wet’ conditions.[a]
i
Scheme 3. Pr3SnOTf-catalysed hydrogenation of imines
under ‘wet’ conditions.
[a] 10 mol% Col added. 10 bar refers to initial pressure at
RT. All reactions were prepared on the open bench and
degassed before pressurisation. Percentages are in situ
1
conversions determined by H NMR spectroscopy (see SI
for full details).
(by 1H NMR; see SI), only ca. 5% of the side-product
PhCH2OH was detected at the end of these reactions.
Based on these successful initial results, we
attempted the RA of PhCHO and PhNH2, as a model
reaction (Table 1, entry 3a). Upon mixing these
substrates with no catalyst, 31% conversion to imine
1
2a was observed by H NMR spectroscopy, over 24
hours. Subsequent addition of 1, however, resulted in
immediate further conversion to 2a (87%),
concomitant with a visible phase separation between
the DCB solvent and H2O generated from the
condensation reaction; evidently, 1 acts as an
efficient LA catalyst to promote imine formation
from carbonyls and amines. Gratifyingly, the
conditions used for ‘wet’ imine hydrogenation were
applicable to the RA, with an excellent conversion of
94% to the target amine; the exclusive side-product
was PhCH2OH. A longer reaction time for the RA
was required than that for the direct hydrogenation of
imine 2a under ‘wet’ conditions, which is to be
expected from the greater amount of H2O present,
formed from the initial condensation reaction. We
propose that the reduction mechanism is likely to be
the same as that proposed for imine hydrogenation
with 1, in which H2 activation by 1/Col precedes
protonation of the 2a by [Col–H]+[OTf]–, prior to
subsequent reduction of the [2a–H]+[OTf]– to 3a by
iPr3Sn–H (regenerating 1), all via
a
polar
mechanism.[12] Here, the effect of H2O is to bind to 1
and reversibly sequester it as off-cycle species (vide
supra), thereby retarding the rate of H2 activation.[17]
Attempts to lower the catalyst loading to 5 mol% led
to a dramatic drop in rate; given that no evidence of
appreciable decomposition was observed, this is
attributed simply to a doubling of the H2O/catalyst
ratio. Additionally, changing the solvent to toluene
detrimentally affected the reaction rate, primarily due
to the poor solubility of 1 in non-polar solvents.[12]
In addition to unsubstituted 3a, products bearing
functional groups on either of the aryl rings could
[a] 10 bar refers to initial pressure at RT. All reactions
were prepared on the open bench and degassed before
pressurisation. Percentages are in situ conversions
determined by 1H NMR spectroscopy (see SI for full
details). Cons. = consumption of carbonyl, Amine =
conversion to desired target pictured amine, Alcohol =
conversion of carbonyl to corresponding alcohol by direct
hydrogenation.
[b] 10 mol% Col added.
3
This article is protected by copyright. All rights reserved.

10.1002/adsc.201701418
Advanced Synthesis & Catalysis
t
also be prepared with excellent conversions, with
both electron-withdrawing and electron-donating
groups being tolerated (Table 1, 3c-3e). Notable
exceptions are NO2-substituted arenes, which resulted
in very complicated mixtures and intractable
products; this is presumably due to radical-mediated
reduction of ArNO2 by the tin hydride, as has been
previously documented.[18] Reactions employing
alkylamines as reagents gave mixed results. Although
the least hindered primary amine substrates formed
the expected products in moderate yields (Table 1, 3f,
3g), the reactions suffered from over-alkylation as
evidenced by the formation of (PhCH2)2N–nBu (from
nBuNH2) or (PhCH2)2N–R (R = H, CH2Ph; from
PhCH2NH2) as side-products. Interestingly, when the
Gratifyingly, the bulky amine BuNH2 is coupled
very effectively[22] to aromatic aldehydes and even
the bulky aliphatic partner PrCHO (Table 1, 3b, 3j,
3k). This qualitative difference in applicability
i
between the two systems is consistent with the lower
steric bulk of Pr3Sn–H relative to [I–H]–, which
i
would allow for a closer approach to even very
hindered imines, thus facilitating H– transfer.[23] As
t
well as BuNH2, other very bulky amines could also
successfully be employed (Table 1, 3l-n). Notably the
very hindered secondary amine Pr2NH can even be
i
used (Table 1, 3l), albeit proceeding at a rather
sluggish rate; the side-product profile in this reaction
mirrors that from the synthesis of 3i, indicating a
general propensity for iPr-substituted amines to
undergo RA-transimination reactions under these
conditions. The relatively high production of
PhCH2OH is attributed to a slow initial condensation
reaction (observed in the H NMR), which leaves a
greater amount of PhCHO to compete as a
hydrogenation substrate.
Initial attempts to expand the carbonyl scope to
ketones led to a significant drop in chemoselectivity
for hydrogenation, with substrates PhNH2 and
CH3COCH3 or PhCOCH3 yielding ~1:1 ratios of the
target amine and the alcohol side-product.
Fortunately, and in contrast to the findings for
aldehydes, for ketone substrates this selectivity is
improved by reducing the reaction temperature to
150°C, albeit at the cost of reduced reaction rate.
Accordingly, under otherwise identical conditions,
both acetone and acetophenone could be successfully
coupled (Table 1, 3o-q).
i
slightly bulkier PrNH2 or CyNH2 (Cy = cyclohexyl;
Table 1, 3h, 3i) were reacted with PhCHO, the target
products were formed as the major species, alongside
traces of (PhCH2)2N–R (R = H, CH2Ph); additionally,
acetone and cyclohexanone were also observed in the
1
1
respective H NMR spectra, indicating some C–N
i
bond cleavage within the Pr–N and Cy–N moieties.
The formation of these carbonyl compounds likely
results from a transimination reaction, which could
proceed via 1-mediated β-N H– abstraction from the
iPr–N and Cy–N groups in 3h and 3i respectively; the
resultant iminium ions would rapidly hydrolyse to
acetone or cyclohexanone,[19] and the liberated
PhCH2NH2 would undergo subsequent RA reactions
with PhCHO to produce (PhCH2)2N–R (R = H,
CH2Ph), directly analogous to the aforementioned
synthesis of 3g (see SI, Fig. S20 for further details). It
is noteworthy that parallel H– abstraction reactivity
has been previously documented for combinations of
the ubiquitous LA in FLP chemistry, B(C6F5)3, and
iPr2NH /iPr2NEt.[20]
Both of these side reactions are attributed to the
high temperatures required to achieve productive
catalysis with 1 when moisture is present, which
reduce selectivity.[21] Attempts to lower the
temperature to 150°C resulted in similar product
distributions accompanied by a substantial decrease
in reaction rate (e.g. for 3h, reaction at 150°C
achieved 60% conversion to the target amine in 49 h).
Nevertheless it should be emphasised that, despite
these competing reactions, the desired singly-
alkylated amine was the major product for all of the
above reactions. While the reaction times using 1 are
mostly shorter than using Soós’ catalyst I for
identical coupling partners (vide supra; Scheme 2),
these side reactions were not observed by the latter,
which is highly likely a result of the lower operating
temperature. Since our attempts to reduce the reaction
temperature with 1 detrimentally affected the rate of
turnover, we considered substrates which were more
problematic for I, namely those exhibiting a larger
steric profile.[11]
Scheme 4. Scaled-up reductive amination of benzaldehyde
with aniline catalysed by 1. 50 bar refers to final pressure
at 150°C (35 bar at RT). The reaction was prepared on the
open bench and sparged with N2 before pressurisation with
H2 (see SI). The in situ conversion was determined by H
NMR spectroscopic analysis (see SI), while the yield was
calculated from the mass of isolated pure product.
1
Finally, in order to demonstrate the ability of 1-
catalysed RA to produce larger quantities of material,
the hydrogenative coupling of model substrates
PhNH2 and PhCHO was conducted on an increased
scale. Conducting the reaction at 150°C and using
slightly modified conditions (DCB was replaced with
1,2-difluorobenzene to facilitate solvent removal
during workup; an increased pressure of 50 bar was
4
This article is protected by copyright. All rights reserved.

10.1002/adsc.201701418
Advanced Synthesis & Catalysis
To a solution of amine (0.2 mmol), carbonyl (0.2 mmol)
and, when aniline or its derivatives are used (e.g. 3a, 3g),
2,4,6-collidine (2.6 µL, 0.02 mmol, 10 mol%) in 1,2-
dichlorobenzene (0.7 mL) was added to 1 (7.9 mg, 0.02
mmol, 10 mol%) in a Wilmad high pressure NMR tube
fitted with a PV-ANV PTFE valve. The solution was
freeze-pump-thaw degassed once. After complete thawing,
H2 was admitted up to a pressure of 10 bar at RT. The
reaction mixture was heated in an Al bead bath, and the
results are presented in Table 1.
used to compensate for the lower reaction
temperature), the reaction furnished target 3a with an
isolated yield of 75% (343 mg; Scheme 4), following
a simple work-up.
In conclusion, we have developed a simple and
practical FLP-type protocol for the RA of various
amines and organic carbonyls, catalysed by
iPr3SnOTf (1). This simple ‘R3Sn+’-based Lewis acid,
which can be readily prepared from inexpensive
starting materials, displays a remarkable tolerance to
H2O, elevated temperatures and strong amine bases.
Notably, this protocol shows a qualitative substrate
scope that is complementary to that of the only other
reported FLP RA catalytic system, which was
recently reported by Soós et al. and employed
sterically-tuned triarylboranes. Since our approach to
develop a H2O-tolerant LA (required for RA)
manipulated electronic factors (i.e. incorporating a
softer p-block element), rather than augmenting the
sterics of an existing LA series, the lower steric
profile of 1 enables the successful reduction of more
hindered substrates while still being competent for H2
activation, in the presence of moisture. In addition to
the inherent appeal of developing new methods for
precious metal-free RA, we would also suggest that
these results further emphasise the value of pursuing
‘alternative’ non-boron-based LAs as targets for FLP
chemistry.
Procedure and for the scaled-up reductive
amination catalysed of PhCHO and PhNH2
catalysed by 1
A
solution of 1 (99.3 mg, 0.25 mmol) in 1,2-
difluorobenzene (35 mL) was prepared in a 100 mL Parr
5500 high pressure compact laboratory reactor. The reactor
was sealed and sparged with N2 for 5 minutes, then
pressurised with nitrogen (10 bar) and stirred for a further
5 minutes. The reactor was depressurised, and aniline
(0.228 mL, 2.50 mmol), benzaldehyde (0.254 mL, 2.50
mmol) and 2,4,6-collidine (33.0 µL, 0.25 mmol) were
injected. The reactor was pressurised with hydrogen (35.0
bar, which equates to 50 bar at 150 °C) and heated to
150 °C whilst stirring at 200 rpm. Upon completion of the
reaction, the stirrer was stopped, whereupon the reactor
was cooled to room temperature and depressurised. The
solvent was removed under reduced pressure, resulting in a
dark brown oil. The product was extracted into pentane (10
mL), where it was recrystallised by cooling to -20 °C to
obtain 3a as an off-white crystalline solid (343 mg, 1.87
mmol, 75%).
Experimental Section
Acknowledgements
All reactions were prepared on the open bench unless
i
stated otherwise. Pr3SnOTf (1) was synthesised according
We thank the EPSRC, GreenCatEng, Eli Lilly and
to literature.[12] All substrates, 2,4,6-collidine and solvents
(1,2-dichlorobenzene (DCB), 1,2-difluorobenzene (DFB))
were purchased from commercial suppliers (Sigma Aldrich,
Fluorochem, Acros Organics). Solid imines were dried
under vacuum and stored under N2, while liquid imines
and aldehydes were degassed, dried over 4Å molecular
sieves and stored under N2. All other compounds and
solvents were used as supplied. H2 was purchased from
BOC (research grade) and used without further drying or
purification.
GlaxoSmithKline (Pharmacat consortium) for funding PhD
studentships (JSS, NJA and DJS), and the Royal Society for a
University Research Fellowship (AEA; UF/110061).
References
[1] a) G. C. Welch, R. R. S. Juan, J. D. Masuda, D. W.
Stephan, Science, 2006, 314, 1124–1126; b) G. C.
Welch, D. W. Stephan, J. Am. Chem. Soc., 2007, 129,
1880–1881; c) P. A. Chase; G. C. Welch, T. Jurca, D.
W. Stephan, Angew. Chem. Int. Ed., 2007, 46, 8050–
8053; d) P. A. Chase, T. Jurca, D. W. Stephan, Chem.
Commun., 2008, 1701–1703; e) D. Chen, J.
Klankermayer, Chem. Commun., 2008, 2130–2131; f)
A. E. Ashley, A. L. Thompson, D. O’Hare, Angew.
Chem. Int. Ed., 2009, 48, 9839–9843; g) K.
Chernichenko, A. Madarász, I. Pápai, M. Nieger, M.
Leskelä, T. Repo, Nat. Chem., 2013, 5, 718–723; h)
M.-A. Légaré, M.-A. Courtemanche, E. Rochette, F.-
G. Fontaine, Science, 2015, 349, 513–516.
Typical procedure for the ‘open bench’
hydrogenation of imines catalysed by 1
To a solution of imine (0.2 mmol) (and, for imine 2a only,
2,4,6-collidine (2.6 µL, 0.02 mmol, 10 mol%)) in 1,2-
dichlorobenzene (0.7 mL) was added to 1 (7.9 mg, 0.02
mmol, 10 mol%) in a Wilmad high pressure NMR tube
fitted with a PV-ANV PTFE valve. The solution was
freeze-pump-thaw degassed once. After complete thawing,
H2 was admitted up to a pressure of 10 bar at RT. The
reaction mixture was heated in an Al bead bath; the results
are presented in Scheme 3.
[2] a) D. W. Stephan, S. Greenberg, T. W. Graham, P.
Chase, J. J. Hastie, S. J. Geier, J. M. Farrell, C. C.
Brown, Z. M. Heiden, G. C. Welch, M. Ullrich, Inorg.
Chem., 2011, 50, 12338–12348; b) G. Erker, D. W.
Stephan, Frustrated Lewis Pairs, Vols. I and II,
Typical procedure for the ‘open bench’
hydrogenation aminations catalysed by 1
5
This article is protected by copyright. All rights reserved.

10.1002/adsc.201701418
Advanced Synthesis & Catalysis
Springer, Berlin, 2013; c) D. W. Stephan, Acc. Chem.
[15] H. K. J. Hall, J. Am. Chem. Soc., 1957, 79, 5441–
5444.
[16] M. J. Janssen, J. G. A. Luijten, Rec. Trav. Chim.,
1963, 82, 1008–1014.
Res., 2015, 48, 306–316; d) D. W. Stephan, G. Erker,
Angew. Chem. Int. Ed., 2015, 54, 6400–6441; e) D.
W. Stephan, J. Am. Chem. Soc., 2015, 137, 10018–
10032; f) D. W. Stephan, Science, 2016, 354, 1248–
1256; g) D. J. Scott, M. J. Fuchter,; A. E. Ashley,
Chem. Soc. Rev., 2017, 46, 5689–5700.
[17] Direct spectroscopic evidence for binding of H2O to
1
[iPr3Sn]+ in the RA experiments comes from their H
NMR spectra, where broad signals of varying
lineshape are observed at ca. δ = 4.90-5.30 ppm, and
occasionally 119Sn{1H} NMR spectra showed a single
broad resonance at δ = 30-40 ppm. An independent
experiment mixing 1, Col and H2O (1:1:10) in DCB
showed that free H2O displays a resonance at δ = 5.16
ppm, whereas H2O bound to the Sn centre appears at
δ = 4.90 ppm (correlated to a broad 119Sn{1H} NMR
[3] C. Bergquist, B. M. Bridgewater, C. J. Harlan, J. R.
Norton, R. A. Friesner, G. Parkin, J. Am. Chem. Soc.,
2000, 122, 10581–10590.
[4] a) M. Lindqvist, N. Sarnela, V. Sumerin, K.
Chernichenko, M. Leskelä, T. Repo, Dalton Trans.,
2012, 41, 4310–4312; b) L. E. Longobardi, C. Tang,
D. W. Stephan, Dalton Trans., 2014, 43, 15723–
15726.
1
resonance at δ = 35 ppm). A H NOESY experiment
[5] a) D. J. Scott, M. J. Fuchter, A. E. Ashley, J. Am.
Chem. Soc., 2014, 136, 15813–15816; b) D. J. Scott,
T. R. Simmons, E. J. Lawrence, G. G. Wildgoose, M.
J. Fuchter, A. E. Ashley, ACS Catal., 2015, 5, 5540–
5544.
[6] a) M. Bakos, A. Gyömöre, A. Domján, T. Soós,
Angew. Chem. Int. Ed., 2017, 56, 5217–5221; b) A.
Gyömöre, M. Bakos, T. Földes, I. Pápai, A. Domján,
T. Soós, ACS Catal., 2015, 5, 5366–5372; c) T.
Mahdi, D. W. Stephan, J. Am. Chem. Soc. 2014, 136,
15809–12; d) T. Mahdi, D. W. Stephan, Angew.
Chem. Int. Ed., 2015, 54, 8511–8514.
[7] J. S. Carey, D. Laffan, C. Thomson, M. T. Williams,
Org. Biomol. Chem., 2006, 4, 2337. More recently, it
was found that 15.6% of all syntheses published in
the Journal of Medicinal Chemistry utilised a RA
step: D. G. Brown, J. Boström, J. Med. Chem., 2016,
59, 4443–4458.
reveals the latter signal to display though–space
coupling between H2O and the Sn(CH(CH3)2)3
protons of the LA, and to the α-CH3 protons in the LB.
This is consistent with H-bonding between Col and
coordinated
H2O,
possibly
formulated
as
or
[iPr3Sn·(H2O)(HO···H-Col)]+
[iPr3Sn·(H2O)(HOH···Col)]+, and indicates that the
Brønsted acidity of [Col-H]+ and [iPr3Sn·2H2O]+ are
very similar. The observed H and 119Sn{1H} NMR
1
chemical shifts were found to vary depending on the
concentration of H2O and the identity of the LB
present, which explains the small chemical shift range
seen in the RA experiments. See SI for full details.
n
[18] Bu3SnH undergoes various radical reactions with
aromatic NO2 groups, even without a radical initiator.
For an example study, see: W. Bowman, D. Crosby, P.
Westlake, J. Chem. Soc., Perkin Trans. 2, 1991, 73-
80.
[8] R. Tripathi, S. Verma, J. Pandey, V. Tiwari, Curr.
Org. Chem., 2008, 12, 1093–1115.
[19] The failure of these ketones to undergo discernible
reduction to the respective alcohols in these reactions
reflects the greater difficulty of hydrogenating these
C=O bonds vs imines using 1, which has previously
required longer reaction times (> 30 h for Me2CO;
see ref. 27), and more weakly basic conditions (e.g.
Col), than those used for the synthesis of 3i.
[20] V. Sumerin, F. Schulz, M. Nieger, M. Leskelä, T.
Repo, B. Rieger, Angew. Chem. Int. Ed., 2008, 47,
6001-6003.
[9] a) M. D. Bhor, M. J. Bhanushali, N. S. Nandurkar, B.
M. Bhanage, Tetrahedron Lett., 2008, 49, 965–969;
b) S. Fleischer, S. Zhou, K. Junge, M. Beller, Chem.
Asian J., 2011, 6, 2240–2245; c) S. Werkmeister, K.
Junge, M. Beller, Green Chem., 2012, 14, 2371–2374.
[10] a) V. Fasano, M. J. Ingleson, Chem. Eur. J., 2017, 23,
2217–2224; b) V. Fasano, J. E. Radcliffe, M. J.
Ingleson, ACS Catal., 2016, 6, 1793–1798.
[11] E. Dorkó, M. Szabó, B. Kótai, I. Pápai, A. Domján, T. [21] S. Gomez, J. A. Peters, T. Maschmeyer, Adv. Synth.
Soós, Angew. Chem. Int. Ed., 2017, 56, 9512–9516. Catal., 2002, 344, 1037–1057.
[12] D. J. Scott, N. A. Phillips, J. S. Sapsford, A. C. Deacy, [22] For comparison, the synthesis of 3b using the
M. J. Fuchter, A. E. Ashley, Angew. Chem. Int. Ed.,
2016, 55, 14738–14742.
protocol of Soós et al. (ref. 11) achieved 85%
conversion in 72 h.
[13] RAs have previously been performed using
nBu2SnCl2, however hydrosilanes were used as
sacrificial reductants: see R. Apodaca, W. Xiao, Org.
Lett., 2001, 3, 1745–1748 and references therein.
[14] K. Clarke, K. Rothwell, J. Chem. Soc., 1960, 1885–
1895.
[23] The poorer hydride donor ability of very bulky
borohydrides has been previously ascribed to steric
effects e.g. [(C6Cl5)3B–H]–; see D. J. Scott, M. J.
Fuchter, A. E. Ashley, Angew. Chem., Int. Ed., 2014,
53, 10218−10222.
6
This article is protected by copyright. All rights reserved.

10.1002/adsc.201701418
Advanced Synthesis & Catalysis
COMMUNICATION
Direct Reductive Amination of Carbonyl
Compounds Catalyzed by a Moisture Tolerant
Tin(IV) Lewis Acid
Adv. Synth. Catal. Year, Volume, Page – Page
Joshua S. Sapsford, Daniel J. Scott, Nathan J.
Allcock, Matthew J. Fuchter, Christopher J. Tighe
and Andrew E. Ashley*
7
This article is protected by copyright. All rights reserved.