On the chlorenium source in the asymmetric chlorolactonization reaction

Article abstract of DOI:10.1021/ol102850m

N-Acylated N-chlorohydantoins are shown to be competent chlorenium sources in the (DHQD)₂PHAL-mediated asymmetric chlorolactonization. The derivatives demonstrate the exact role of the N1 and N3 chlorine atoms in the parent dichlorohydantoins w

Full text of DOI:10.1021/ol102850m

ORGANIC
LETTERS
2011
Vol. 13, No. 4
608–611
On the Chlorenium Source in the
Asymmetric Chlorolactonization Reaction
Roozbeh Yousefi, Daniel C. Whitehead, Janet M. Mueller, Richard J. Staples, and
Babak Borhan*
Department of Chemistry, Michigan State University, East Lansing,
Michigan 48824, United States
babak@chemistry.msu.edu
Received November 24, 2010
ABSTRACT
N-Acylated N-chlorohydantoins are shown to be competent chlorenium sources in the (DHQD)2PHAL-mediated asymmetric chlorolactonization.
The derivatives demonstrate the exact role of the N1 and N3 chlorine atoms in the parent dichlorohydantoins with the N1 chlorine serving as an
inductive activator and the N3 chlorine being delivered to the substrate. The putative associated catalyst/chlorine source complex was
experimentally demonstrated through a series of matched/mismatched experiments employing chiral N-chlorinated hydantoins.
Recently, a number of new methods for the asymmetric
delivery of chiral halenium (Xþ) equivalents onto nonchiral
olefinic precursors have appeared in the literature.1 Begin-
ning with Taguchi’s carbocyclizations,2 Kang’s cobalt-
salen-mediated asymmetric iodoetherification reaction,3
and Ishihara’s iodonium induced polyene cascade cyclization,4
this traditionally stubborn5 class of asymmetric transfor-
mations has begun to return favorable results in terms of
enantioselectivity. Subsequent to these efforts, the Snyder
laboratory featured an asymmetric chlorination of an
isolated olefin in an advanced intermediate as a key step
in the total synthesis of (-)-napyradiomycin A1.6
Recently, we discovered a unique catalyst system com-
prising a catalytic amount of (DHQD)2PHAL and chlori-
nated hydantoins that allowed for good conversions and
enantioselectivities for the chlorolactonization of 4-substituted
4-pentenoic acids (see the transformation in Table 1 for an
example).7 Contemporary with our disclosure, the Tang
group reported a highly enantioselective bromolactoniza-
tion of conjugated (Z)-enynes governed by a cinchonidine-
based urea organocatalyst and NBS as a bromine source.8
Very recently, Veitch and Jacobsen disclosed an aminourea-
mediated asymmetric iodolactonization of alkenoic acids
that proceeds in high yield and excellent ee’s.9
(1) (a) For a recent review, see: Chen, G.; Ma, S. Angew. Chem., Int.
Ed. 2010, 49, 8306.
(2) (a) Inoue, T.; Kitagawa, O.; Kurumizawa, S.; Ochiai, O.; Taguchi, T.
Tetrahedron Lett. 1995, 36, 1479. (b) Inoue, T.; Kitagawa, O.; Ochiai, O.; Shiro,
M.; Taguchi, T. Tetrahedron Lett. 1995, 36, 9333. (c) Inoue, T.; Kitagawa, O.;
Saito, A.; Taguchi, T. J. Org. Chem. 1997, 62, 7384. (d) Kitagawa, O.; Taguchi,
T. Synlett 1999, 1191.
(3) (a) Kang, S. H.; Lee, S. B.; Park, C. M. J. Am. Chem. Soc. 2003,
125, 15748. (b) Kwon, H. Y.; Park, C. M.; Lee, S. B.; Youn, J.-H.; Kang, S. H.
Chem.;Eur. J. 2008, 14, 1023.
(4) Sakukara, A.; Ukai, A.; Ishihara, K. Nature 2007, 445, 900.
(5) For asymmetric halolactonizations, see: (a) Kitagawa, O.; Hanano,
T.; Tanabe, K.; Shiro, M.; Taguchi, T. J. Chem. Soc., Chem. Commun 1992,
1005. (b) Grossman, R. B.; Trupp, R. J. Can. J. Chem. 1998, 76, 1233. (c) Cui,
X. L.; Brown, R. S. J. Org. Chem. 2000, 65, 5653. (d) Haas, J.; Piguel, S.; Wirth,
T. Org. Lett. 2002, 4, 297. (e) Wang, M.; Gao, L. X.; Yue, W.; Mai, W. P. Synth.
Commun. 2004, 34, 1023. (f) Wang, M.; Gao, L. X.; Mai, W. P.; Xia, A. X.;
Wang, F.; Zhang, S. B. J. Org. Chem. 2004, 69, 2874. (g) Haas, J.; Bissmire, S.;
Wirth, T. Chem.;Eur. J. 2005, 11, 5777. (h) Garnier, J. M.; Robin, S.;
Rousseau, G. Eur. J. Org. Chem. 2007, 3281. (i) Ahmad, S. M.; Braddock,
D. C.; Cansell, G.; Hermitage, S. A.; Redmond, J. M.; White, A. J. P. Tetra-
hedron Lett. 2007, 48, 5948. (j) Ning, Z.; Jin, R.; Ding, J.; Gao, L. Synlett 2009,
2291.
In our original disclosure, the use of dichlorohydantoins
as the chlorine source proved essential for obtaining the
highest possible enantioselectivities. We have subsequently
(6) Snyder, S. A.; Tang, Z.-Y.; Gupta, R. J. Am. Chem. Soc. 2009,
131, 5744.
r
10.1021/ol102850m
Published on Web 01/18/2011
2011 American Chemical Society

set out to explore the mechanism of the transformation
with the aim of understanding the drivers responsible for
enantioinduction, specifically as they relate to the role of
the hydantoin in the reaction. Our initial efforts have
focused on exploring the structural and electronic influ-
ence of the chlorine source (i.e., 3).
Table 1. Cyclization with Modified N-Chlorohydantoins
Inthispaper, wepresent a detailedstudyonthe influence
of structurally related N1 and N3 acylated N-chlorohy-
dantoins on the selectivity of the chlorolactonization reac-
tion. This study was facilitated by the development of a
convenient method for the selective synthesis of N1 and N3
acylated hydantoins. Additionally, we provide further evi-
dence for the associative complex between the (DHQD)2-
PHAL catalyst and the chlorine source that seems to be in-
fluential for the observed selectivity in the transformation.
During our initial optimization studies, it became ap-
parent that dichlorinated hydantoins (i.e., 3) were suited as
a perfectly tuned halenium source for our methodology.
The reagent appeared uniquely wedged in reactivity be-
tween N-bromosuccinimide, which resulted in a fast but
less selective bromolactonization, and N-chlorosuccinimide,
which was slow to react at low temperatures in the analogous
chlorolactonization (Table 1, 1, vide infra). The 35Cl nuclear
quadrupole resonance (NQR) frequency has been invoked as
a quantitative marker for the relative Clþ character of a given
N-Cl bond. In particular, a larger Clþ character corresponds
to a higher NQR resonance frequency.10 Several resonance
frequencies are collected in Figure 1.
Figure 1. Selected NQR resonance frequencies.
On the basis of the NQR resonance frequencies collected
in Figure 1, one would expect that the N3 chlorines (see 2
for numbering) of hydantoins 2 and 3 would be more
electrophilic than that of NCS (1), while the N1 chlorine
atoms should possess roughly the same level of electro-
philicity as 1. Additionally, one would expect TCCA (4)11
to possess even more electrophilic Cl atoms than the
N-chlorohydantoins and NCS.
a Yields are isolated yields after column. b As judged by chiral GC.
This general trend, as predicted by the NQR reso-
nance frequencies, is evident in the chlorolactonization of
p-fluorophenyl-substituted alkenoic acid 5 to give chloro-
lactone 6 (Table 1). Under our optimized conditions, the
cyclization of 5 governed by NCS (entry 1) proceeds in
a relatively poor (15%) yield with a reduced enantioselec-
tivity of just 52% ee. A series of dichlorinated hydantoins
(entries 3-6) produced the desired chlorolactone 6 in a
much higher yield (76-87%) and markedly improved ee’s
(81-89% ee) as compared to 1.12 Chlorolactonization
(7) Whitehead, D. C.; Yousefi, R.; Jaganathan, A.; Borhan, B. J. Am.
Chem. Soc. 2010, 132, 3298.
(8) (a) Zhang, W.; Zheng, S.; Liu, N.; Werness, J. B.; Guzei, I. A.;
Tang, W. J. Am. Chem. Soc. 2010, 132, 3664. For two examples disclosed
during the preparation of this manuscript see: (b) Zhou, L.; Tan, C. K.; Jiang,
X.; Chen, F.; Yeung, Y.-Y. J. Am. Chem. Soc. 2010, 132, 15474. (c) Murai,
K.; Matsushita, T.; Nakamura, A.; Fukushima, S.; Shimura, M.; Fujioka, H.
Angew. Chem., Int. Ed. 2010, 49, 9174.
(9) Veitch, G. E.; Jacobsen, E. N. Angew. Chem., Int. Ed. 2010, 49, 7332.
(10) (a) Nagao, Y.; Katagiri, S. Sci. Rep. Hirosaki Univ. 1991, 38, 20.
(b) Nagao, Y.; Katagiri, S. Chem. Lett. 1993, 1169.
(12) For comparison with NCS (1), N-chlorophthalamide and 4-ni-
tro-N-chlorophthalamide were used in the standard reaction depicted in
Table 1, leading to 4% (13% ee) and 7% (32% ee) of 6 in 12 h,
respectively.
(11) Tilstam, U.; Weinmann, H. Org. Process. Res. Dev. 2002, 6, 384.
Org. Lett., Vol. 13, No. 4, 2011
609

with the more electrophilic TCCA (4, entry 2) returned
chlorolactone 6 in the highest yield (89%), but with a
reduced 70% ee as compared to the N-chlorohydantoin
series. The lower enantioselectivity with the more reactive
TCCA may be due to an inherently nonselective uncata-
lyzed background reaction (see the Supporting Information)
that is avoided with the less reactive chlorohydantoins.
This trend indicates that the N3 chlorine of hydantoin 3
is delivered to the substrate in the chlorolactonization
reaction, since the N1 chlorine atom is substantially less
electrophilic (i.e., on the order of 1, see Figure 1). Indeed,
we were able to observe the immediate formation of the
1-chloro-3-hydro hydantoin byproduct by 1H NMR dur-
ing the course of the reaction.7 The N1 chlorine of 3 must
then chiefly serve to inductively activate the N3 chlorine
atom and likely accounts for the molecule’s enhanced N3
NQR resonance frequency as compared to NCS. A sub-
stantial amount of evidence for this supposition was col-
lected by screening a series of N-chlorinated hydantoins.
The 1-chloro-N3-methyl-5,5-dimethylhydantoin (9, entry
7) was virtually inactive in the conversion of 5, returning just
7% yield of 6 in 72% ee. This outcome was not unexpected,
given that the remaining N1 chlorine of 9is flanked by just one
carbonyl and thus ought to be even less active than NCS (cf.
entry 1). More interestingly, when the inductively activating
N1 chlorine was replaced by a methyl (10, entry 8), the
remaining N3 chlorine was substantially less activated,
returning 6 in a reduced yield of 50% and with reduced
selectivity of 78% ee. Clearly, some modicum of inductive
activation at the N3 position via substitution at the N1
position is desirable in terms of both enhanced yield and
enantioselectivity.
We posited that one might be able to rescue both yield
and enantioselectivity in the transformation if the N1
chlorineofthechlorenium sourcewerereplacedbyanother
activating group. Additionally, by incorporating a series of
substitutions at the N1 site, we would be able to probe any
steric demands that might arise in the associative complex
between catalyst and hydantoin.7
After some experimentation, we arrived at a simple method
for the selective acylation of the N1 and N3 positions
of 5,5-dimethylhydantoin 21 (Scheme 1). The N3 position
(i.e., 22) could be selectively acylated under mild conditions
by treating 21 with the appropriate acyl chloride in pyridine
at room temperature. Alternatively, heating a pyridine
solution of 21 and the acyl chloride to 170 °C gave selective
access to the N1 acylated product 24. Suspecting the
intermediacy of 22 in the latter transformation, we also
demonstrated its quantitative conversion to 24 on heating
in pyridine at 170 °C. The N-acyl hydantoins were then
converted to their corresponding monochlorinated deriva-
tives 23 and 25 in good yield (80-87%) by employing our
TCCA mediated chlorination protocol.13 Thus, N1 and
N3 acyl and benzoyl hydantoins 11-20 were generated,
and the position of acylation and subsequent chlorination
Scheme 1. Preparation of N-Aacyl-N-chlorohydantions
was secured unequivocally by X-ray crystallography
(see the Supporting Information for details).
With 11-20 in hand, they were next screened as chlor-
enium sources in the chlorolactonization of 5, providing
chlorolactone 6. Confirming our prediction, all eight of the
N1 acylated chlorine sources returned the desired lactone 6
in comparable yields and good enantioselectivities (on
average 72% yield, 86% ee). The N1 acetylated hydantoin
11 gave 6 in 71% yield and 83% ee (entry 9). Interestingly,
11 was nearly equally selective as the parent 1,3-dichloro-
5,5-dimethylhydantoin 2 and markedly more active and
selective than the N1 methyl derivative 10. The N1 benzoyl
derivative 12 surpasses the parent hydantoin 2 in terms of
selectivity, returning 6 in 88% ee with a slightly reduced
yield of 78% (entry 10). Conversely, the N3 benzoyl con-
gener (13) produced 6 in substantially lower yield and
reducedenantioselectivity (21% yield, 43% ee, entry 11). A
second exampleofanN3p-nitrobenzoyl-substitutedchlor-
ohydantoin alsoperformedpoorlyin the conversion of 5 to
6 (14% yield, 46% ee, entry 13). The latter two examples
clearly point to the need for an electron-withdrawing group
on N1 for the activation and reactivity of the N3 chlorine.
Next, we screened a series of N3-chlorinated hydantoins
containing a variety of substituted benzoyl groups at the
N1 position (14, 16-20). Interestingly, all of the compounds
returned 6 with good ee’s ranging from 84 to 88% ee
(entries 12, 14-18).
To more closely probe the magnitude of the electronic
effects of the N1 benzoyl substituent, we measured the rate
of the formation of lactone 6 by NMR studies (see the
Supporting Information for details). We employed hydan-
toins 12, 14, 16, 17, 19, and 20 in order to scan a range of
electron-donating and -withdrawing substituents. Interest-
ingly, we observed a clear rate acceleration when hydan-
toins harboring electron-withdrawing N-benzoyl substituents
Table 2. Calculated Rate Constants for the Formation of the
Lactone 6 by Using Different N1 Benzoylated Hydantoins
(1 mol % Catalyst)
hydantoin
k (s-1
)
hydantoin
k (s-1
)
17
12
19
1.6 ꢀ 10-4
2.1 ꢀ10-4
2.4 ꢀ 10-4
16
20
14
4.2 ꢀ 10-4
5.6 ꢀ 10-4
7.4 ꢀ 10-4
(13) Whitehead, D. C.; Staples, R. J.; Borhan, B. Tetrahedron Lett.
2009, 50, 656.
610
Org. Lett., Vol. 13, No. 4, 2011

were employed (Table 2). For example, by removing the
methyl group from the benzoyl group the rate constant
increases from 1.6 ꢀ 10-4 to 2.1 ꢀ 10-4 (cf. 17 vs 12). By
adding another fluoro group to the ortho postion of the
hydantoin 19 the rate constant was almost doubled (4.2 ꢀ
10-4 vs 2.4 ꢀ 10-4; 16 vs 19). Moreover, the rate acceleration
was even more pronounced when more strongly electron-
withdrawing substituents were employed (i.e., 5.6 ꢀ 10-4
and 7.4 ꢀ 10-4 for 20 and 14, respectively).
Table 3. Matched/Mismatched Behavior with Chiral N-Chlor-
ohydantoins
That the reaction maintains a comparable degree of selec-
tivity regardless of the substitution on the N1 benzoyl
substituent illustrates two interesting points. First, the
overall electronics of the benzoyl substituent does not seem
to influence the enantioselectivity of the transformation to
any appreciable extent aside from behaving as a surrogate
activating group in place of the N1 chlorine in DCDMH
(2) and DCDPH (3). Sadly, this frustrates any strategy
toward further optimization of the enantioselectivity of the
transformation by means of electronic fine-tuning of the
N1 benzoyl substituent, although the electronic effects of
the substituted benzoyl group do translate to the reactivity
of the distal N3 chlorine to some extent as shown by the
rate data in Table 2.
Second, the steric demand of the substituents on the N1
benzoyl moiety seems to have a negligible effect on the
selectivity of the transformation. It seems logical that since
the active N3 chlorine is distal to the benzoyl group at N1,
one would not expect the steric bias at the N1 position to
influence the approach of the alkenoic acid to the pre-
organized catalyst/chlorine source complex. More inter-
esting is the implications of the N1 substitutent on the
catalyst/hydantoin complex itself. Apparently, the N1
nitrogen and its substituent, regardless of size, must be
splayed in a manner such that it does not clash with the
chiral catalyst, thus allowing for high enantioselectivity.
This fact, taken together with the clear dependence of ee on
the size of the C-5 substituents (see entries 3-6), may
suggest that the complex between the catalyst and the
hydantoin is governed by a hydrogen bond to the C-4
carbonyl in lieu of that housed in the C-2 position.
N-chlorohydantoin
entry
(DHQD)2PHAL
(DHQ)2PHAL
% yield % ee of
% yield
% ee
of 6
of 6
of ent-6
ent-6
1
2
3
4
(S)-26
(R)-26
(S)-27
(R)-27
78
44
78
71
83
69
79
75
30
65
71
71
-55
-62
-71
-73
observed (Table 2). In the presence of (S)-26, lactone 6
was returned in 78% yield and 83% ee, while the R
antipode gave 6 in a substantially reduced yield of 44%
with an eroded enantioselectivity of 69% ee. This result
again clearly demonstrates the formation of an associative
complex between the chiral catalyst and the chlorine
sources. The reduced yield for the mismatched case (i.e.,
with (R)-26) might indicate that the C-5 substituent is
displayed in the transition state in such a manner so as to
prevent facile approach of the substrate to the catalyst/
hydantoin complex. Importantly, the opposite effect was
observed when the pseudoenantiomer (DHQ)2PHAL was
employed as the chiral catalyst, albeit to a lesser extent.
In that case, the R enantiomer of 26 returned lactone
ent-6 in higher yield and ee (65% yield, 62% ee) as com-
pared to the S antipode (30% yield, 55% ee). The overall
lower selectivity when (DHQ)2PHAL is employed is likely
due to its diastereomeric relationship with its more selective
pseudoenantiomer (DHQD)2PHAL. A similar but less
pronounced effect was observed with (S)- and (R)-27
(Table 3, entries 3 and 4), possibly as a result of benzyl’s
smaller steric demand as compared iso-propyl.
In addition to screening the series of N-acyl hydantoins,
we also wished to probe the associative hydantoin/catalyst
complex observed previously by 1H NMR spectroscopy.7
We surmised that an appropriate means of experimentally
demonstrating the formation of the hydantoin/catalyst
complex might be to employ a series of enantiomerically
pure N-chlorinated hydantoins in the presence of the chiral
catalyst in the conversion of 5 to 6. If the enantiomerically
pure hydantoins engaged in an associative interaction with the
(DHQD)2PHAL catalyst, then the stereochemical anti-
podes of the chlorine source ought to reflect matched/
mismatched behavior in terms of enantioselectivity in the
preparation of 6. For this study, we prepared both enantio-
mers of 5-isopropyl (26) and 5-benzyl-1,3-dichlorohydantoin
(27). We have already demonstrated the ability to chlorinate
chiral hydantoins without eroding their enantiopurity.13
When the R and S enantiomers of 26 were employed in
the cyclization of 5 with (DHQD)2PHAL under optimized
conditions, clear matched/mismatched behavior was
In summary, we have probed the nature of the chlor-
enium source in our asymmetric chlorolactonization protocol
by preparing a series of previously unknown N-acylated N-
chlorohydantoins. We have conclusively secured the role of
the N1 chlorine as an inductive activator of the N3 chlorine,
which in turn is delivered to the substrate during the course
of the chlorolactonization event. Finally, we experimentally
demonstrated the formation of the previously observed
associative complex between the chiral catalyst and chlor-
ohydantoin by performing the chlorolactonization in the
presence of chiral N-chlorohydantoins.
Acknowledgment. We acknowledge the ACS PRF
(No. 47272-AC) and the NSF (CHE-0957462) for funding.
Supporting Information Available. Experimental de-
tails for synthesis of hydantoins along with X-ray crystal
structure of select compounds. This material is available
free of charge via the Internet at http://pubs.acs.org.
Org. Lett., Vol. 13, No. 4, 2011
611