The first enantioenriched metalated nitrile possessing macroscopic configurationat stability

Article abstract of DOI:10.1021/ol070149g

Figure presented Magnesium - bromine exchange on enantiopure cyclopropyl bromonitrile 5 at -100°C for 1 min followed by a D₂O quench gives the deuterionitrile in 81% ee (retention); additional trapping experiments establish t_1/2(rac) = 11.4 h at -100°C. These experiments provide the first glimpse into the stereochemical aspects of Mg-Br exchange. The intermediate formed is the first metalated nitrile demonstrated to possess macroscopic configurational stability.

Full text of DOI:10.1021/ol070149g

ORGANIC
LETTERS
2007
Vol. 9, No. 7
1319-1322
The First Enantioenriched Metalated
Nitrile Possessing Macroscopic
Configurational Stability
Paul R. Carlier* and Yiqun Zhang
Department of Chemistry, Virginia Tech, Blacksburg, Virginia 24061
pcarlier@Vt.edu
Received January 19, 2007
ABSTRACT
Magnesium
−
bromine exchange on enantiopure cyclopropyl bromonitrile 5 at
11.4 h at
Br exchange. The intermediate formed is the first metalated nitrile demonstrated to possess macroscopic
−
100
°
C for 1 min followed by a D2O quench gives the deuterionitrile
in 81% ee (retention); additional trapping experiments establish t1 2(rac)
the stereochemical aspects of Mg
configurational stability.
)
−
100 C. These experiments provide the first glimpse into
°
/
−
By virtue of their excellent nucleophilicity, metalated nitriles
have found wide application in synthesis.1 Metalated nitriles
would prove even more useful if they could be generated as
enantiopure reactive intermediates (e.g., 1, 2, Figure 1) that
of highly retentive H/D exchange in 1.0 M NaOMe in
CH3OD at 50 °C,3 indicating microscopic configurational
stability of the sodium salt. This microscopic configurational
stability is a consequence of carbanion pyramidalization,
engendered by angle strain.4 However, deprotonation of (R)-
(-)-3 by LDA in Et2O gave an intermediate that racemized
within 10 min at -78 °C.5 This behavior clearly distinguishes
lithiated 3 from metalated cyclopropanes lacking electron
withdrawing groups, which are well-known to possess
macroscopic configurational stability at temperatures as high
as 35 °C.6
In this Letter we confirm the microscopic configurational
stability of the Na salt of 3, and determine an upper bound
for the racemization barrier of the rapidly inverting Li salt
Figure 1. Chiral metalated nitriles 1 and 2, and Walborsky’s
enantiopure cyclopropylnitrile 3
possess configurational stability on the macroscopic2 time
scale. Walborsky demonstrated that (R)-(-)-3 was capable
(3) Measured rate ratio kex/krac ) 9.060: Walborsky, H. M.; Motes, J.
M. J. Am. Chem. Soc. 1970, 92, 2445-2450.
(4) (a) X-ray crystallography of a related lithiated cyclopropylnitrile
shows C-lithiation (e.g. 1, Figure 1): Boche, G.; Harms, K.; Marsch, M. J.
Am. Chem. Soc. 1988, 110, 6925-6926. (b) Enantiopure acyclic nitriles
undergo racemizing H-D exchange (ref 3).
(1) (a) Arseniyadis, S.; Kyler, K. S.; Watt, D. S. Org. React. 1984, 31,
1-364. (b) Carlier, P. R.; Lo, K.-M.; Lo, M. M.-C.; Williams, I. D. J. Org.
Chem. 1995, 60, 7511-7517. (c) Fleming, F. F.; Zhang, Z. Tetrahedron
2005, 61, 747-789.
(5) Walborsky, H. M.; Hornyak, F. M. J. Am. Chem. Soc. 1955, 77,
6026-6029.
(2) (a) Kapeller, D.; Barth, R.; Mereiter, K.; Hammerschmidt, F. J. Am.
Chem. Soc. 2007, 129, 914-923. (b) Hoffmann, R. W.; Nell, P. G. Angew.
Chem., Int. Ed. 1999, 38, 338-340. (c) Schulze, V.; Hoffmann, R. W. Chem.
Eur. J. 1999, 5, 337-344.
(6) (a) Li Walborsky, H. M.; Impastato, F. J.; Young, A. E. J. Am. Chem.
Soc. 1964, 86, 3283-3288. (b) Na Pierce, J. B.; Walborsky, H. M. J. Org.
Chem. 1968, 33, 1962-1965. (c) Mg Walborsky, H. M.; Young, A. E. J.
Am. Chem. Soc. 1964, 86, 3288-3296.
10.1021/ol070149g CCC: $37.00
© 2007 American Chemical Society
Published on Web 03/10/2007

of 3. Most importantly, we demonstrate macroscopic con-
figurational stability of the Mg salt of 3 in diethyl ether at
-100 °C (racemization t1/2 ) 11.4 h), indicating a racem-
ization barrier of 14 kcal/mol.
Scheme 2. Synthesis and Thermal Ellipsoid Plot (50%
Probability) of Enantiopure Bromonitrile (S)-(+)-5
To begin our investigations we resynthesized (R)-(-)-3
and treated it with NaOMe in CD3OD (Scheme 1).
Scheme 1. Enantiospecific H/D Exchange of 3
After 3 days the reaction was quenched to reveal >99%
deuteration and >99% ee, thus confirming the microscopic
configurational stability of the sodium salt of 3.7 We then
carried out numerous deprotonation/trapping experiments at
temperatures as low as -135 °C. Regardless of the identity
of base (LDA, LiHMDS, KHMDS), solvent (THF, Et2O,
Me2O, toluene), and electrophile (MeI, BnI, D2O), racemic
products were obtained, even under in situ trapping condi-
tions. An illustrative example: deprotonation of (R)-(-)-3
by LDA in THF at -100 °C for 1 min, followed by D2O
quench gave a 91:9 mixture of d1-3 and 3, in a combined
<2% ee. Given that total racemization would require at least
5 half-lives, the upper bound for the racemization t1/2
(t1/2(rac)) is 12 s, indicating an inversion barrier of e11.1
kcal/mol at -100 °C.8
Recent work by Fleming and Knochel has established that
halogen-metal exchange on R-halonitriles provides an
excellent non-deprotonative route to lithiated and magnesi-
ated nitriles.9 Magnesiated nitriles in particular appeared
attractive, as they might be expected to possess improved
configurational stability.10 We thus first prepared racemic
bromoacid (()-4 from 1,1-diphenylethene, according to the
literature procedure.11 Resolution with (-)-brucine12 and salt
break gave (+)-4. Conversion to the primary amide and
dehydration gave the desired bromonitrile (+)-5; X-ray
crystallography established (S)-configuration, thus providing
independent confirmation of the (S)-configuration of (+)-4
(Scheme 2).12
with 1.05 equiv of n-BuLi and 6 equiv of CH3I (Table 1,
entry 1). Unexpectedly, little of the desired methylation
product 6a was observed (32 mol %), and considerable
amounts of protionitrile 3 (36 mol %) and bromonitrile 5
(32 mol %) were seen. Furthermore, not only were the
products 6a and 3 racemic, but the recovered starting material
5 also had significantly reduced enantiopurity. Interestingly,
performance of the same reaction with CD3I and a D2O
quench gave no deuterionitrile d1-3, suggesting that the
protionitrile 3 results from E2 elimination of the n-butyl
bromide formed in the lithium-bromine exchange reaction
Table 1. Metal-Bromine Exchange/Trapping Reactions of 5a
mol %
mol % 5 mol % 6
(% ee) (% ee)
3:d1-3
(% ee)
entry
RM
EX
quench
Application of Fleming and Knochel’s exchange/in situ
trapping protocol was then carried out at -100 °C in Et2O
1
2
3
4
n-BuLi
n-BuLi
CH3I NH4Cl aq 32 (30 S)
CD3I D2O 23 (9 S)
32 (0) 36:0 (0)
52 (0) 25:0 (0)
6 (0) 46:0 (69 R)
7 (0) 25:31 (72 R)
(7) To the best of our knowledge this is the first preparation of
enantiopure d1-3 using Walborsky’s protocol; percent ee measured by chiral
stationary phase HPLC.
i-PrMgBr CD3I NH4Cl aq 48 (>95 S)
i-PrMgBr CD3I D2O 37 (>95 S)
(8) Eyring equation: Eyring, H. Chem. ReV. 1935, 17, 65-77.
(9) (a) Fleming, F. F.; Zhang, Z.; Knochel, P. Org. Lett. 2004, 6, 501-
503. (b) Fleming, F. F.; Zhang, Z.; Liu, W.; Knochel, P. J. Org. Chem.
2005, 70, 2200-2205.
a Product ratios measured by 1H NMR; % ee measured by chiral
stationary phase HPLC (5, 6: OD; 3/d1-3: AD). Note that the % ee of 3
and d1-3 cannot be deconvoluted by HPLC.
(10) For studies of the configurational stabilities of unstabilized primary
alkyl organolithiums and organomagnesiums see: Witanowski, M.; Roberts,
J. D J. Am. Chem. Soc. 1966, 88, 737-741.
(Table 1, entry 2). It should be noted that these reactions
are very clean and the product ratios are easily obtained from
(11) Baird, M. S.; Nizovtsev, A. V.; Bolesov, I. G. Tetrahedron 2002,
58, 1581-1593.
(12) Walborsky, H. M.; Barash, L.; Young, A. E.; Impastato, F. J. J.
Am. Chem. Soc. 1961, 83, 2517-2525. Correlation performed by the quasi-
racemate method; the absolute (S)-configuration of (+)-4 is correctly noted
in ref 6a.
1
the H NMR spectra of the crude products, due to the
excellent spectral dispersion of the cyclopropyl ring protons
in 3, d1-3, 5, and 6.13
1320
Org. Lett., Vol. 9, No. 7, 2007

Use of i-PrMgBr for Mg-Br exchange and CD3I as
electrophile gave dramatically different results (Table 1,
entries 3 and 4). First of all, little alkylation product 6b was
recovered in either case. Thus the magnesionitrile derived
from 5 is much less reactive than its lithium counterpart,
magnesiated cyclopropanes lacking directly attached electron-
withdrawing groups,14 and other magnesiated nitriles.9
Second, use of D2O as quench resulted in significant but
not exclusive formation of the deuterionitrile d1-3 (Table 1,
entry 4). Thus in the magnesium-bromine exchange reac-
tions some metalated nitrile remains before addition of the
quench. Third, the recovered bromonitrile 5 was found to
retain enantiopurity. Last, and most importantly, the protio-
and deuterionitriles 3 and d1-3 possess significant retention
of configuration, suggesting increased configurational sta-
bilility of the magnesiated cyclopropylnitrile relative to its
lithium analogue.
nance of deuterionitrile d1-3 relative to protionitrile 3. These
results are consistent with the idea that the excess Grignard
reagent preferentially induced elimination of the i-PrX
coproduct. As the delay (t) between the addition of the
Grignard and when the deuterium quench increased, the
enantiomeric excess of the combined protio- and deuterio-
nitriles at -100 °C decreased from 81% (1 min, Table 2,
entry 1) to 67% (180 min, Table 2, entry 4) suggesting slow
racemization of the magnesionitrile.
By plotting ln(% ee(d1-3/3)/100) vs time it is possible to
estimate the rate of racemization17 and similar studies were
then carried out at -78 °C (Figure 2). In this way t1/2(rac)
Our inability to deconvolute the enantiopurities of
protionitrile 3 and deuterionitrile d1-315 prompted us to carry
out magnesium-bromine exchange with 2.2 equiv of the
Grignard reagent, so as to minimize participation of the
magnesiated nitrile in elimination and thus improve the ratio
of d1-3 to 3. We found improved enantiopurities with
i-PrMgCl and carried out a series of sequential exchange/
deuteration experiments at -100 and -78 °C (Table 2).16
Table 2. Mg-Br Exchange/Deuteration Reactions of 5a with
USe of 2.2 equiv of i-PrMgCl
Figure 2. Plot of ln(% ee/100) of d1-3 and 3 vs delay time in
Mg-Br exchange/deuteration experiments on (S)-(+)-5 (Table 2).
of 11.4 ( 1 and 2.3 ( 0.1 h were determined at -100 and
-78 °C, respectively. Application of the Eyring equation
indicates a barrier to inversion of 14 kcal/mol at -100 °C.
Extrapolation of the data in Figure 2 to the Y-intercept (time
) 0 s) indicates (81 ( 1)% ee and (77 ( 1)% ee at -100
and -78 °C, respectively.18 Thus Mg-Br exchange of 5 is
not stereospecific, and enantioselectivity appears to dete-
riorate at higher temperature. Competition between stereo-
specific and racemizing exchange mechanisms may account
for these results; the latter process could involve an electron-
transfer chain mechanism.19
Attempts to alkylate the magnesiated nitrile at -100 and
-78 °C with a variety of carbon electrophiles were unsuc-
cessful. However, at -42 °C, a small amount (8 mol %) of
racemic methylated nitrile 6a was obtained by treatment of
the magnesiated nitriles with MeOTf (6 equiv, 30 min,
following a 2 min magnesium/bromine exchange) and a D2O
quench. Since d1-3 (72 mol %) and 3 (18 mol %) were
recovered from this reaction in a combined 41% ee, it appears
that methylation of the magnesiated cyclopropylnitrile is not
only slow, but racemizing. Thus it is possible that methyl-
temp
(°C)
time
(min)
mol % d1-3:3
(% ee, R)
entry
mol % 5
1
2
3
4
5
6
7
-100
-100
-100
-100
-78
1
5
30
180
1
90:6 (81)
88:7 (79)
89:7 (79)
85:14 (67)
90:7 (76)
90:7 (75)
88:8 (66)
4
5
4
1
3
3
4
-78
-78
5
30
a Product ratios measured by 1H NMR; % ee measured by chiral
stationary phase HPLC (3: AD; 5 OD).
As hoped, at -100 °C the use of 2.2 equiv of i-PrMgCl
gave excellent conversion (95-99 mol %) and a predomi-
(13) See Figure S2 in the Supporting Information for an overlay of the
1H NMR spectra of these compounds. No other products were observed in
the 1H NMR spectra of the crude products; mass balance following
chromatography in these reactions ranged from 60% to 75%.
(14) (a) Vu, V. A.; Marek, I.; Polborn, K.; Knochel, P. Angew. Chem.,
Int. Ed. 2002, 41, 351-352. (b) Kopp, F.; Sklute, G.; Polborn, K.; Marek,
I.; Knochel, P. Org. Lett. 2005, 7, 3789-3791.
(15) The isotopomers co-elute on HPLC; attempts to determine the
enantiopurity of protionitrile 3 apart from d1-3 by 1H NMR (chiral lanthanide
shift reagents, observation of the R-proton) were unsuccessful.
(16) Reactions with other Grignard reagents at -100 °C in Et2O were
also explored; EtMgBr and MeMgBr gave lower yields and enantioselec-
tivities, whereas t-BuMgCl failed to promote exchange.
(17) Carlier, P. R.; Lam, P. C.-H.; DeGuzman, J.; Zhao, H. Tetrahedron:
Asymmetry 2005, 16, 2998-3002.
(18) Standard error; at the 95% confidence limit, the % ee values are 81
( 3 (-100 °C) and 77 ( 4 (-78 °C), respectively (time ) 0 s).
(19) Hoffmann, R. W.; Bro¨nstrup, M.; Mu¨ller, M. Org. Lett. 2003, 5,
313-316.
Org. Lett., Vol. 9, No. 7, 2007
1321

ation is occurring via an SET pathway,20 rather than the SE2-
ret pathway that appears likely for protonation and deuter-
ation. The extremely low nucleophilicity of the magnesiated
cyclopropylnitrile relative to other magnesiated nitriles9 is
puzzling; we attribute the lack of reactivity to the steric bulk
posed by the â,â-diphenyl substitution.
Scheme 3. Proposed Racemization Mechanism for (S)-(+)-5
In contrast to the behavior seen with i-PrMgCl/Br, when
lithium-bromine exchange of (S)-(+)-5 was carried out with
2.2 equiv of n-BuLi for 1 min at -100 °C, followed by a
D2O quench, d1-3 (68 mol %) and 3 (31 mol %) were
obtained as racemates. Use of sec-BuLi under these condi-
tions again gave high conversion and a racemic outcome,
affording d1-3 (94 mol %) and 3 (3 mol %). On the basis of
our deprotonation/trapping experiments on (R)-(-)-3, we
attribute the unfavorable outcome to rapid racemization of
lithiated nitrile, rather than racemizing Li/Br exchange. If
this is true we can again set an upper bound of 12 s for
t1/2(rac) of lithiated 3, indicating a racemization barrier of
less than 11.1 kcal/mol at -100 °C.
Finally, to account for the significant racemization of
starting bromonitrile 5 seen in reactions employing only 1.05
equiv of n-BuLi (Table 1, entries 1 and 2), we considered
the reaction of bromonitrile 5 and lithiated nitrile rac-7 to
transiently form ate complex 8 as a mixture of diastereomers
(Scheme 3). Collapse of the ate complex would lead to 5 in
racemic form or as the starting (S)-enantiomer; several cycles
through the ate complex (coupled with rapid racemization
of 7) would lead to complete racemization of 5.
racemization of bromonitrile 5. At this point we cannot
determine whether racemization of 5 is due to ate complex
formation (Scheme 3) or an SET pathway. However, to date
we have not detected any radical coupling products.
In conclusion, we have demonstrated that Mg-Br ex-
change on (S)-(+)-5 produces a magnesiated nitrile that
racemizes slowly at -100 °C (t1/2(rac) ) 11.4 h). Although
this species has highly attenuated reactivity toward electro-
philes, it represents the first example of a metalated nitrile
possessing configurational stability on the macroscopic time
scale. Finally, these experiments allow the stereoselectivity
of Mg-Br exchange to be observed for the first time; further
studies are in progress.
Acknowledgment. We gratefully acknowledge financial
support from the NSF (CHE-0213525) and the Virginia Tech
Department of Chemistry.
To test this proposed mechanism we prepared rac-7 by
deprotonation of rac-3 with LDA (1.1 equiv), and added it
to 1 equiv of (S)-(+)-5 at -100 °C in THF. After 30 min
the reaction was subjected to protic quench, and bromonitrile
5 was recovered in 77% yield and 4% ee; the protionitrile 3
was recovered in 75% yield. Thus the lithiated nitrile induces
Supporting Information Available: Synthetic procedures
and analytical data. This material is available free of charge
via the Internet at http://pubs.acs.org.
(20) Hoffmann, R. W. Chem. Soc. ReV. 2003, 32, 225-230.
OL070149G
1322
Org. Lett., Vol. 9, No. 7, 2007