Lithiated carbamates: Chiral carbenoids for iterative homologation of boranes and boronic esters

Article abstract of DOI:10.1002/anie.200702146

(Chemical Equation Presented) Take your pick: Either enantiomer of either diastereomer of substrates bearing adjacent stereogenic centers is accessible through reaction of an (R)- or (S)-lithiated carbamate with an (R)- or (S)-boronic ester (see scheme; pin = pinacolate, OCb = substituted carbamate).

Full text of DOI:10.1002/anie.200702146

Angewandte
Chemie
DOI: 10.1002/anie.200702146
Asymmetric Synthesis
Lithiated Carbamates: Chiral Carbenoids for Iterative Homologation
of Boranes and Boronic Esters**
Jake L. Stymiest, Guillaume Dutheuil, Adeem Mahmood, and Varinder K. Aggarwal*
Dedicated to Professor Miguel Yus on the occasion of his 60th birthday
Table 1: One-pot lithiation/borylation of Hoppe-type carbamates.
The homologation of chiral boronic esters by Matteson and
co-workers represents a landmark contribution in the field of
asymmetric synthesis.[1] However, despite achieving excep-
tionally high enantioselectivities with simple, readily acces-
sible reagents, the methodology has not been widely adopted.
One possible factor may be that additional steps are required
to control stereochemistry during iterative homologation as a
consequence of employing substrate control.[1d,2] Since the
stereogenic center created during homologation is dictated by
the substrate diol of the boronic ester, if the opposite
stereoisomer is required, a three-step sequence is needed to
invert (by exchange) the diol stereochemistry.[2] A potentially
more powerful and efficient strategy is to employ reagent
control in the homologation process. This approach requires a
chiral carbenoid that shows high configurational and chemical
stability but sufficient reactivity to effect homologation of
boronic esters. We have shown that chiral sulfur ylides fulfil
some of these criteria and can be used effectively in the
homologation of a range of boranes with very high enantio-
selectivity.[3] However, these ylides do not react with boronic
esters and only give low enantioselectivity with borinic
esters.[3c] Very recently, Blakemore and co-workers described
the application of Hoffmannꢀs a-chloro Grignard reagents[4]
(although they found that the lithium derivatives worked
better) to effect iterative homologation of boronic esters.[5,6]
The chlorosulfoxide precursors were prepared in two steps,
but some degree of racemization occurred during the
homologation process.[5] Hoppe-type lithiated carbamates,
1a–e[7] (from lithiation of 2a–e with sBuLi in the presence of
(ꢀ)-sparteine; see Table 1, where the carbamate moiety is
abbreviated OCb) represent another class of chiral carbe-
noids, which is more readily obtained, and seemed to us to
fulfil the requirements for boronic ester homologation.
Indeed, Hoppe and co-workers have described the trapping
of lithiated carbamates with borates and the subsequent
(separate) treatment of the a-carbamoyl alkylboronate with a
Grignard reagent to effect 1,2-metalate rearrangement with
Entry
Carbenoid
precursor
R2
(R3)2
Lewis
acid
Yield [%]
(product)
e.r.[a]
1
2
3
4
5
6
7
8
2a
Et
Et
–
–
–
–
91 (3a)
90 (3b)
81 (3c)
85 (3d)
94 (3d)
90 (3a)
90 (3e)
71 (3 f)
75 (3e)
73 (3 f)
67 (3g)
65 (3h)
64 (3h)
68 (3i)
70 (3i)
70 (3j)
98:2
98:2
98:2
88:12
97:3
98:2
97:3[b]
95:5
97:3
98:2
95:5[c]
97:3
98:2
96:4
98:2
97:3
nHex
iPr
Ph
Ph
Et
Et
Ph
Et
9-BBN
9-BBN
9-BBN
9-BBN
pinacol
Et
9-BBN
pinacol
pinacol
Et
9-BBN
pinacol
9-BBN
pinacol
pinacol
MgBr2
MgBr2
–
MgBr2
MgBr2
MgBr2
–
MgBr2
MgBr2
MgBr2
MgBr2
MgBr2
2b
2c
9
10
11
12
13
14
15
16
Ph
Et
Ph
Ph
Ph
Ph
Ph
2d
2e
[a] Unless otherwise stated all enantiomeric ratios (e.r.) were calculated
using chiral HPLC(Chiracel OD column), [b] The e.r. was determined
using 1H NMR of (S)-(+)-a-methoxy-a-trifluoromethylphenyl acetic acid
ester,[16] [c] The e.r. was determined using chiral GCon a Supelco Alpha-
Dex column.
expulsion of the carbamate moiety. Oxidation of the resultant
boronic esters affords the corresponding alcohols with high
enantiomeric excesses.[8] However, the direct reaction of
lithiated carbamates, such as 1a–e, with boranes or boronates
and further iterative homologations had not been described.[9]
We found that lithiated carbamates 1a–e reacted directly
with boranes or boronic esters, thus furnishing secondary
alcohols 3a–j in good yield and with high enantioselectivity
(Table 1). A number of points are worthy of note:
[*] Dr. J. L. Stymiest, Dr. G. Dutheuil, A. Mahmood, Prof. V. K. Aggarwal
School of Chemistry, University of Bristol
Cantock’s Close, Bristol, BS81TS (UK)
Fax: (+44)117-929-8611
E-mail: v.aggarwal@bristol.ac.uk
[**] We thank Interbioscreen for a generous supply of (ꢀ)-cytisine and
EPSRCfor support of this work. V.K.A. thanks the Royal Society for a
Wolfson Research Merit Award.
1) Reactions
with
9-BBN
(9-BBN = 9-borabicyclo-
[3.3.1]nonane) derivatives resulted in clean migration of
the boron substituent in 4 rather than the boracycle in all
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
Angew. Chem. Int. Ed. 2007, 46, 7491 –7494
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
7491

Communications
cases. This is highly unusual and has only previously been
observed with halide leaving groups.[10]
2) In the case of B-Ph-9-BBN, higher e.r. was obtained in the
presence of MgBr2 (Table 1, entry 4 versus entry 5)
whereas in other cases involving aliphatic groups MgBr2
was not found to be necessary.[11,12]
3) 1,2-Metalate rearrangement of ate complexes is much
slower in the case of boronic esters than for boranes[13] and
required MgBr2 in Et2O at reflux[14] in the former case,
whereas the ate complexes derived from boranes began to
rearrange at ꢀ408C without further additives.[15]
4) A broad range of alkyl carbamates, including those with
methyl and with primary and branched secondary alkyl
chains can be employed together with a broad range of
aryl and alkyl boranes and boronic esters. The method-
ology, therefore, shows considerable substrate scope.
5) Since the absolute stereochemistries of many of the
secondary alcohols are known, and the stereochemistry
of Hoppe lithiation is well-established, we are able to
conclude that reactions of lithiated carbamates with both
the boranes and boronic esters occurred with retention of
configuration (1!4). This result is consistent with pre-
vious findings.[8,9]
6) Secondary boronic esters can undergo a host of further
transformations, including 1-, 2-, 3-carbon chain exten-
sions, transmetallation to zinc and subsequent trapping
with electrophiles, and conversion to amines. Herein, we
illustrate the further utility of the borate esters by
transformation into amines (Scheme 1).[17] In all cases,
the amines 5–7 were formed in moderate to good yields
and high e.r.
Scheme 2. Proposed iterative homologation of boronic esters.
homologation reaction was not influenced by the stereo-
chemistry in A and B. This condition was easily tested by
reacting (ꢁ )-8 with the racemic lithiated carbamate (ꢁ )-1e,
which is available from lithiation of 2e with sBuLi in the
presence of TMEDA (Scheme 3). If (R)-8 reacted with equal
rates with both (R)- and (S)-1e then a 1:1 mixture of
diastereomers 9 and 10 would result, whereas if the rates were
different, one diastereomer would be formed predominantly,
thus making it much more difficult to form all four
stereoisomers. Experimentally, we found that a 1:1 mixture
of 9 and 10 was obtained, thus showing that the rates of
reaction k1 and k2 are not influenced by the stereochemistry of
either partner (Scheme 3).
Scheme 3. Effects of stereochemistry on formation of 9 and 10.
Scheme 1. Amination of intermediate boronate complexes.
The iterative homologation process began with lithiation
of 2a (in the presence of (ꢀ)-sparteine) with subsequent
trapping with EtB(pinacol), which gave the desired boronic
ester intermediate 8 in 78% yield and 98:2 e.r. (Scheme 4).
Reaction of boronic ester 8 with lithiated carbamate 1e gave,
after oxidative workup, alcohol 9 as a 96:4 mixture of
diastereomers with e.r. > 98:2 (major diastereomer 9;
Scheme 4).[19] The other diastereomer (10) is potentially
accessible by using the opposite enantiomer of the lithiated
carbamate, ent-1e. In practice, this goal was easily achieved
using OꢀBrienꢀs enantiomeric sparteine surrogate (+)-11,
derived from (ꢀ)-cytisine.[20] Indeed, OꢀBrien and co-workers
have shown that diamine (+)-11 can be effectively used in
Hoppe-type lithiation of carbamates to deliver the opposite
Having shown that lithiated carbamates are effective for
reagent-controlled homologation of boranes and boronates,
we then considered the iterative process with boronic esters.
Indeed, this methodology should allow us to set up scaffolds
bearing multiple adjacent stereocenters with high diastereo-
and enantioselectivities. Realization of this goal would
require the isolation of pinacol boronic ester intermediate
A and its subsequent use in a second homologation reaction
with lithiated carbamate B to yield the double homologation
product C (Scheme 2).
To effectively prepare all four isomers of the alcohol,
corresponding to the oxidation of C, it was important that the
7492 www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 7491 –7494

Angewandte
Chemie
Scheme 4. Iterative homologation of boronic esters 8 and ent-8.
[3] a) V. K. Aggarwal, G. Y. Fang, A. T. Schmidt, J. Am. Chem. Soc.
enantiomer to (ꢀ)-sparteine with high enantiomeric ratios.
Thus, using the opposite enantiomeric carbamate ent-1e in
the same process gave the alternative diastereomer 10 with
similarly high diastereo- (94:6 d.r.) and enantioselectivities
(e.r. > 98:2; Scheme 4). The enantiomeric pair to 9 and 10 was
readily obtained with similarily high d.r. and e.r. by using the
same protocol but starting from ent-8, which was itself derived
from the first homologation using OꢀBrienꢀs enantiomeric
sparteine surrogate (+)-11 (Scheme 4).
These levels of enantio- and diastereoselectivity clearly
show that the reaction is dominated by reagent control with
no “matching” issues affecting the outcome of the second
homologation, even though adjacent stereocenters are cre-
ated.
In conclusion, we have developed a process for the
homologation of boranes and boronic esters using Hoppe-
type lithiated carbamates which shows very broad substrate
scope. The Hoppe-type lithiated carbamates are effectively
chiral carbenoids and are derived from the simplest of
reagents: primary alcohols. The power of the methodology
lies in its iterative use. Thus, through two cycles of the
homologation process, either enantiomer of either diastereo-
mer can be easily accessed through appropriate choice of the
chiral diamine employed ((ꢀ)-sparteine 12 or (+)-11).
Application of this versatile methodology in natural product
synthesis is currently underway.
2005, 127, 1642; b) G. Y. Fang, V. K. Aggarwal, Angew. Chem.
2007, 119, 363; Angew. Chem. Int. Ed. 2007, 46, 359; c) G. Y.
Fang, O. Wallner, N. di Blasio, V. K. Aggarwal, unpublished
results.
[4] a) R. W. Hoffmann, P. G. Nell, R. Leo, K. Harms, Chem. Eur. J.
2000, 6, 3359; b) R. W. Hoffmann, Chem. Soc. Rev. 2003, 32, 225;
c) T. Satoh, K. Tahano, Tetrahedron 1996, 52, 2349.
[5] P. R. Blakemore, M. S. Burge, J. Am. Chem. Soc. 2007, 129, 3068.
[6] P. R. Blakemore, S. P. Marsden, H. D. Vater, Org. Lett. 2006, 8,
773.
[7] a) D. Hoppe, F. Hintze, P. Tebben, Angew. Chem. 1990, 102,
1457; Angew. Chem. Int. Ed. Engl. 1990, 29, 1422; For reviews on
a-lithiation of carbamates, see: b) D. Hoppe, F. Hintze, P.
Tebben, M. Paetow, H. Ahrens, J. Schwerdtfeger, P. Sommerfeld,
J. Haller, W. Guarnieri, S. Kolczewksi, T. Hense, D. Hoppe, Pure
Appl. Chem. 1994, 66, 1479; c) D. Hoppe, F. Hintze, Angew.
Chem. 1997, 109, 2376; Angew. Chem. Int. Ed. Engl. 1997, 36,
2282; d) A. Basu, S. Thayumanavan, Angew. Chem. 2002, 114,
740; Angew. Chem. Int. Ed. 2002, 41, 716; e) D. Hoppe, F. Marr,
M. Brüggemann in Organolithiums in Enantioselective Synthesis
(Ed.: D. M. Hodgson), Springer, Heidelberg, 2003, pp. 61 – 137,
and references therein; f) D. Hoppe, G. Christoph in The
Chemistry ofOrganolithium Compounds (Eds.: Z. Rappoport,
I. Marek), Wiley, Chichester, 2004, p. 1055.
[8] a) E. Beckmann, V. Desai, D. Hoppe, Synlett 2004, 2275; b) E.
Beckmann, D. Hoppe, Synthesis 2005, 217; c) M. J. McGrath, P.
OꢀBrien, Synthesis 2006, 2233.
[9] One isolated example of a reaction of a lithiated carbamate with
an arylboronate has been reported during the total synthesis of
N-acetylcolchinol: G. Besong, K. Jarowicki, P. J. Kocienski, E.
Sliwinski, F. T. Boyle, Org. Biomol. Chem. 2006, 4, 2193.
[10] In contrast, reactions of the same boranes with sulfur ylides only
show high selectivity in the cases of Ph, alkenyl, and iPr
substituents.[3c] In fact, the only other examples where the boron
substituent migrates in preference to the boracycle are cases
involving a (small) halide leaving group or carbonylation. In all
other cases, the boracycle migrates preferentially. Thus the large
carbamate moiety is behaving as a small leaving group. For a
discussion of the factors influencing the migration of groups on
nonsymmetrical ate complexes of organoboranes, see: V. K.
Aggarwal, G. Y. Fang, X. Ginesta, D. Howells, M. Zaja, Pure
Appl. Chem. 2006, 78, 215.
Received: May 16, 2007
Published online: July 20, 2007
Keywords: asymmetric synthesis · boranes · boronic esters ·
.
chiral carbenoids · homologation
[1] a) D. S. Matteson, D. Majumder, J. Am. Chem. Soc. 1980, 102,
7588; b) D. S. Matteson, D. Ray, J. Am. Chem. Soc. 1980, 102,
7590; For reviews on the use of boronic esters in asymmetric
syntheses, see: c) D. S. Matteson, Tetrahedron 1989, 45, 1859;
d) D. S. Matteson, Tetrahedron 1998, 54, 1055.
[2] D. S. Matteson, H.-W. Man, J. Org. Chem. 1996, 61, 6047.
Angew. Chem. Int. Ed. 2007, 46, 7491 –7494
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
7493

Communications
[11] We believe that this additive sequesters either (ꢀ)-sparteine 12
We have established that it is in fact the diamine ((ꢀ)-sparteine,
12) and not the leaving group (LiOC(O)NiPr2, 13) that is
responsible for the erosion in e.r. Reaction of 16 with nBuLi gave
the diamine-free lithiated carbamate, which after trapping with
PhEt2B and oxidative workup again furnished the alcohols 3a
and 3d, but this time with identical e.r. values.
or the carbamate leaving group 13, thus preventing it from
binding to the borane 14, which otherwise results in a degree of
homolysis of the benzylic carbon–boron bond, thus lowering the
e.r. as seen in the absence of MgBr2 (entry 4, Table 1).
[13] For examples, see: a) J. M. Stoddard, K. J. Shea, Chem.
Commun. 2004, 830; b) A. Bottoni, M. Lombardo, A. Neri, C.
Trombini, J. Org. Chem. 2003, 68, 3397.
[14] Rearrangement did not occur at 258C (11B NMR). Heating of
ate complex 4 at refluxin Et 2O in the presence of MgBr2 for 12 h
was required to achieve 1,2-metalate rearrangement.
[15] Observed using varible-temperature 11B NMR spectroscopy; the
reaction was monitored beginning at ꢀ808C and warmed up by
108C increments every 5 min over 35 min. Rearrangement of ate
complex 4 began at ꢀ408C.
[12] Although it is possible for B-Ph-9-BBN to react with the
lithiated carbamate with different enantioselectivity to other
boranes, we were able to show that the difference in e.r. occurs
after ate complexformation. Treatment of lithiated 1a with the
mixed borane PhEt2B gave, after oxidative workup, a mixture of
the two alcohols (3a and 3d) with enantiomeric ratios of 96:4
and 78:22, respectively (see below). Since the two alcohols are
derived from the common intermediate 15, this species must be
formed with at least 96:4 e.r. Intermediate 15 undergoes
stereospecific migration of Et and Ph groups leading to boranes
C and D, respectively. Oxidation of borane D led to substantial
loss of e.r. whilst C did not. However, boranes related to D had
been generated in our sulfur ylide reactions with no loss of e.r.[3a]
The difference in this case is that D is generated in the presence
of agents capable of binding strongly to boron (LiOC(O)NiPr2,
13) or good electron donors ((ꢀ)-sparteine, 12), which, after
association, could cause some degree of homolytic cleavage of
[16] B. D. Johnston, A. C. Oehlschlager, J. Org. Chem. 1986, 51, 760.
[17] a) H. C. Brown, M. M. Midland, A. B. Levy, J. Am. Chem. Soc.
1973, 95, 2394; b) E. Hupe, I. Marek, P. Knochel, Org. Lett. 2002,
4, 2861.
[18] The e.r. was determined via conversion of amine 5 to the
acetamide by hydrogenolysis and subsequent treatment with
Ac2O in pyridine.
[19] The enhancement of the enantiomeric ratio always occurs when
two enantioselective operations are mapped onto the same
substrate. In theory, if the first homologation gives an x:y. ratio
of enantiomers (97–98:3–2 ratio observed for both (ꢀ)-sparteine
12 and (+)-11), then, assuming there is no kinetic resolution
(both enantiomers of the boronic ester react at the same rate) or
double stereo differentiation (the second homologation is not
affected by the asymmetric center formed in the first step of the
reaction), the ratio of enantiomers for the second homologation
will be x2:y2. (99.8:0.2 expected; e.r. > 98:2 observed). The
enantiomeric excess of the doubly homologated product is
therefore considerably increased. One can view this as the x2:y2
rule. This process also generates the diastereomeric compound
(amount = 2xy; 94:6 expected; 90–96:10–4 observed). See the
following for previous discussions of enantiomeric amplification:
a) E. Negishi, Dalton Trans. 2005, 827; b) V. K. Aggarwal, B. N.
Esquivel-Zamora, G. R. Evans, E. Jones, J. Org. Chem. 1998, 63,
7306; c) S. E. Baba, K. Sartor, J.-C. Poulin, H. B. Kagan, Bull.
Chim. Soc. Fr. 1994, 131, 525.
ꢀ
the C B bond. This process would, in turn, cause the erosion in
e.r. observed.
[20] a) M. J. Dearden, C. R. Firkin, J-P. R. Hermet, P. OꢀBrien, J. Am.
Chem. Soc. 2002, 124, 11870; b) A. J. Dixon, M. J. McGrath, P.
OꢀBrien, Org. Synth. 2006, 83, 141.
7494 www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 7491 –7494