A very reliable method for determination of absolute configuration of chiral secondary alcohols by 1H NMR spectroscopy

Article abstract of DOI:10.1021/ol0479489

(Chemical equation presented) Surprisingly stable synperiplanar conformers of CFTA esters have led us to develop a new and very reliable method for assigning absolute configurations of even secondary alcohols having minimal structural differences, such as chiral benzhydrols and α-monodeuterated benzyl alcohols.

Full text of DOI:10.1021/ol0479489

ORGANIC
LETTERS
2004
Vol. 6, No. 24
4607-4610
A Very Reliable Method for
Determination of Absolute Configuration
of Chiral Secondary Alcohols by H
1
NMR Spectroscopy
Yoshio Takeuchi,* Hidehito Fujisawa, and Ryoji Noyori
Faculty of Pharmaceutical Sciences, Toyama Medical and Pharmaceutical UniVersity,
Sugitani Toyama, 930-0194 Japan, and Department of Chemistry and Research Center
for Materials Science, Nagoya UniVersity, Chikusa Nagoya, 464-8602 Japan
takeuchi@ms.toyama-mpu.ac.jp
Received October 6, 2004
ABSTRACT
Surprisingly stable synperiplanar conformers of CFTA esters have led us to develop a new and very reliable method for assigning absolute
configurations of even secondary alcohols having minimal structural differences, such as chiral benzhydrols and
alcohols.
r-monodeuterated benzyl
The determination of the absolute configuration of chiral
molecules is an essential tool for modern organic chemistry.
This is particularly true in the areas of asymmetric synthesis
and in structural studies of complex natural products. There
have been, however, no practical methods generally ap-
plicable to noncrystalline compounds and/or for samples
available only in very small quantities. In an important
approach to this problem using NMR spectroscopy,1 the
modified Mosher’s method2 uses R-methoxy-R-trifluoro-
methylphenylacetic acid (MTPA, 1)3 as a chiral derivatizing
agent (CDA). However, a substantial number of cases have
been reported where the MTPA procedure fails either because
of the low reactivity of MTPA chloride (MTPA-Cl, 2)4 or
because of the complex mixture of conformers observed in
the MTPA derivatives.5
To overcome these limitations, we have developed the
much more reactive R-cyano-R- fluorophenylacetic acid
(CFPA, 3)6 and related compounds as CDAs that possess a
unique structure with the fluorine atom located at the
stereogenic center.7 As a result of this work, we describe
here the development of the first very reliable and yet
(4) (a) Dutcher, J. S.; Macmillan, J. G.; Heathcock, C. H. J. Org. Chem.
1976, 41, 2663-2669. (b) Hietaniemi, L.; Pohjala, E.; Malkonen, P.;
Riekkola, M. L. Finn. Chem. Lett. 1989, 16, 67-73. (c) Svatos, A.;
Valterova, I.; Samon, D.; Vrkoc, J. Collect. Czech. Chem. Commun. 1990,
55, 485-490. (d) Takeuchi, Y.; Itoh, N.; Kawahara, S.; Koizumi, T.
Tetrahedron 1993, 49, 1861-1870.
(1) Seco, J. M.; Quinoa, E.; Riguera, R. Chem. ReV. 2004, 104, 17-
117.
(2) (a) Ohtani, I.; Kusumi, T.; Kashman, Y.; Kakisawa, H. J. Org. Chem.
1991, 56, 1296-1298. (b) Otani, I.; Kusumi, T.; Kashman, Y.; Kakisawa,
H. J. Am. Chem. Soc. 1991, 113, 4092-4096.
(3) (a) Dale, J. A.; Dull, D. L.; Mosher, H. S. J. Org. Chem. 1969, 34,
4, 2543-2549. (b) Dale, J. A.; Mosher, H. S. J. Am. Chem. Soc. 1973, 95,
512-519. (c) Sullivan, G. R.; Dale, J. A.; Mosher, H. S. J. Org. Chem.
1973, 38, 2143-2147.
(5) Latypov, Sh. K.; Seco, J. M.; Quinoa, E.; Riguera, R. J. Org. Chem.
1996, 61, 8569-8577.
(6) (a) Takeuchi, Y.; Itoh, N.; Note, H.; Koizumi, T.; Yamaguchi, K. J.
Am. Chem. Soc. 1991, 113, 6318-6320. (b) Takeuchi, Y.; Itoh, N.; Satoh,
T.; Koizumi, T.; Yamaguchi, K. J. Org. Chem. 1993, 58, 1812-1820.
(7) (a) Takeuchi, Y.; Asahina, M.; Murayama, A.; Hori, K.; Koizumi,
T. J. Org. Chem. 1986, 51, 955-956. (b) Takeuchi, Y.; Asahina, M.; Nagata,
K.; Koizumi, T. J. Chem. Soc., Perkin Trans. 1 1987, 2203-2207.
10.1021/ol0479489 CCC: $27.50
© 2004 American Chemical Society
Published on Web 11/04/2004

operationally facile procedure for the determination of
absolute configurations of chiral secondary alcohols through
the use of straightforward H NMR spectroscopy.
Table 1. ∆δ Values for CFTA and MTPA Ester Diastereomers
of Some (S)-Benzhydrols
1
We originally developed CFPA (3) for ee determinations
of optically enriched molecules through 19F NMR spectros-
copy of CFPA diastereomers. We found that CFPA chloride
(CFPA-Cl, 4), in contrast to MTPA-Cl (2), reacts even with
sterically hindered nucleophiles.8 The CFPA moiety in a
molecule allows the 19F NMR differentiation of diastereomers
having a stereogenic center up to several bonds removed
from the attachment point of the CFPA residue.6 To further
exploit CFPA-based methods as a general procedure for
chiral differentiation, including the method for absolute
stereochemical assignment, we prepared and examined
several CFPA analogues.9
This research led to our development of R-cyano-R-fluoro-
p-tolylacetic acid (CFTA, 5),10 a compound closely related
to CFPA, as the agent of choice for the development of our
procedure. This agent retains all of the structural merits of
CFPA (3), including high reactivity of the corresponding
chloride (CFTA-Cl, 6), but is much more readily prepared.11
1
We also found the H NMR data for the CFTA derivatives
to be more readily analyzed and characterized than were the
CFPA derivatives (Scheme 1).12
Scheme 1
a NA ) not assignable. b Data from 19F NMR. c Derivatives not obtained.
all-synperiplanar F-CR-CO-O-C-H array13 on the “CFTA
ester plane” as a predominant rotamer of the CFTA esters,14
even in CDCl3 on the NMR time scale.13 To validate the
rationale of our CFTA-based procedure for absolute stereo-
chemical assignment, we examined some chiral benzhydrols
of known absolute configurations.15 These secondary carbinols
have two very similar aromatic substituents. Each (S)-
benzhydrol 7-12 was condensed with both (S)- and (R)-
CFTA-Cl (6) to give (S)- and (R)-CFTA ester diastereomers,
respectively. All aromatic proton signals for each of the (S)-
and (R)-CFTA diastereomers were assigned by means of
COSY and other NMR techniques, and the chemical shift
Preliminary results showing a consistent relationship
between stereochemistry of chiral alcohols and chemical
shifts of the diastereomeric CFTA esters strongly suggested
the formation of an unusually stable conformer having an
(8) Takeuchi, Y.; Itoh, M.; Koizumi, T. J. Chem. Soc., Chem. Commun.
1992, 1514-1515 (see also Chem. Ind. 1992, 874).
(9) Takeuchi, Y.; Iwashita, H.; Yamada, K.; Gotaishi, M.; Kurose, N.;
Koizumi, T.; Kabuto, K.; Kometani, T. Chem. Pharm. Bull. 1995, 43, 1668-
1673. Some of the CFPA analogues prepared were examined with respect
to their abilities to differentiate enantiomers based on NMR analysis.
(10) Takeuchi, Y.; Konishi, M.; Hori, H.; Takahashi, T.; Kometani, T.;
Kirk, K. L. Chem. Commun. 1998, 365-366.
(11) (a) Takeuchi, Y.; Konishi, M.; Hori, H.; Takahashi, T.; Kirk, K. L.
Enantiomer 1999, 4, 339-344. (b) Fujiwara, T.; Sasaki, M.; Omata, K.;
Kabuto, C.; Kabuto, K.; Takeuchi, Y. Tetrahedron: Asymmetry 2004, 15,
555-563.
(13) Takahashi, T.; Fukushima, A.; Tanaka, Y.; Takeuchi, Y.; Kabuto,
K.; Kabuto, C. Chem. Commun. 2000, 787-788.
(14) Abraham, R. J.; Tormena, C. F.; Rittner, R. J. Chem. Soc., Perkin
Trans. 2 2001, 815-820.
(15) Ohkuma, T.; Koizumi, M.; Ikehira, H.; Yokozawa, T.; Noyori, R.
Org. Lett. 2000, 2, 659-662.
(12) It should be noted that assigning each proton in the 1H NMR
spectrum of the p-Tol group is much easier than that of the Ph group
(because of the simple AB quartet of p-Tol), especially for those chiral
alcohols having more than one aromatic residue.
4608
Org. Lett., Vol. 6, No. 24, 2004

differences (∆δ values)16 for each protons were obtained.
When the benzhydrol esters are depicted in a manner such
that the two aromatic substituents are in the plane of Table
1 and the carbinyl protons are coming out of the plane, the
protons having positive ∆δ values (written in red) are
invariably on the left side of the CFTA ester plane and those
having negative ∆δ values (written in blue) are on the right
side of the plane. These results are shown in the middle
column of Table 1. From these results, we conclude that our
procedure is consistent even for chiral secondary alcohols
such as these, which have quite subtle structural differences.
To date we have found no exception to this pattern, including
the other chiral alcohols shown as the preliminary data.12
In contrast, the MTPA method of stereochemical assign-
ment gives very inconsistent results for this set of chiral
secondary alcohols, as shown in the right-hand column of
Table 1. For 7 and 8, the ∆δ values for the protons on both
the left and right sides of the MTPA ester plane2,17 are
positive.18 For 9 and 10, some of the proton shifts cannot be
assigned.12 Furthermore, in the case of 11 and 12, the
reliability of the procedure could not be investigated because
the corresponding MTPA derivatives could not be prepared,
presumably as a result of the poor reactivity of MTPA-Cl
(2).4 These results reveal that the modified Mosher’s
procedure1 is not applicable to this series of chiral benzhy-
drols.
structural hypothesis is also supported by the X-ray crystal-
lographic structure of neomenthyl CFTA ester and by the
ab initio calculations on menthyl CFPA ester.13
Steric arguments can be invoked to explain the proposed
sp conformation that we use to rationalize the observed ∆δ
values (Figure 1). To view the diastereomeric CFTA esters
of chiral alcohol (RSRRCHOH)20 from the left side via an
“extended Newman projection”, the intervening ester link-
ages are omitted for convenience. Conformational arguments
are used to explain the algebraic signs of the ∆δ values. In
the (S)-CFTA diastereomer, the shifts of the protons of the
RS substituent that eclipses the p-Tol group should always
be upfield, relative to the shifts in the nonderivatized alcohol,
as a result of the anisotropic shielding of the aromatic ring.
In contrast, for the (R)-diastereomer, the protons of the RR
group are shielded and therefore should appear upfield.
Reinforcing these differences, the CN group induces the
anisotropic deshielding effect21 on the proximate RR and RS
groups in the (S)- and (R)-CFTA diastereomers, respectively.
1
To account for the observed patterns in the H NMR of
CFTA esters, we propose a synperiplanar (sp) F-CR-CO-
O-C-H conformation, as shown by Figure 1. Indeed, there
Figure 2. ∆δH and ∆δD values for CFTA esters of (S)-dR-benzyl
alcohols.
Accordingly, the ∆δ () δS - δR) values for the protons in
the RR group, namely, those for all of the protons on the left
side of the CFTA ester plane, should be positive, and those
(16) The ∆δ values are defined as δS - δR, where δS and δR are the
chemical shifts of the corresponding protons in the (S)- and (R)-CFTA esters,
respectively.
(17) The term “MTPA ester plane” is used in a same manner as the
term “CFTA ester plane”.
(18) (a) Seco, J. M.; Quinoa, E.; Riguera, R. Tetrahedron: Asymmetry
2000, 11, 2781-2791. (b) Kusumi, T.; Fujita, Y.; Ohtani, I.; Kakisawa, H.
Tetrahedron Lett. 1991, 32, 2923-2926. (c) Rieser, M. J.; Hui, Y.;
Rupprecht, J. K.; Kozlowski, J. F.; Wood, K. V.; McLaughlin, J. L.; Hanson,
P. R.; Zhuang, Z.; Hoye, T. R. J. Am. Chem. Soc. 1992, 114, 10203-
10213. (d) Duret, P.; Waechter, A.; Figadere, B.; Hocquemiller, R.; Cave,
A. J. Org. Chem. 1998, 63, 4717-4720.
(19) (a) Nieminen, J.; Pettersson, M.; Rasanen, M. J. Phys. Chem. 1993,
97, 10925-10936. (b) Durig, J. R.; Phan, H. V.; Hardin, J. A.; Little, T. S.
J. Chem. Phys. 1989, 90, 6840-6851. (c) Amin, M. R.; Harper, D. B.;
Moloney, J. M.; Murphy, C. D.; Howard, J. A. K.; O’Hagan, D. Chem.
Commun. 1997, 1471-1472.
Figure 1. Conformations of (S)- and (R)-CFTA esters of chiral
secondary alcohols RSRRCHOH.
are spectral data and calculations reported that support a sp
F-CR-CO-X conformation as the most stable rotamer, for
a highly conformational preference for various R-fluoro
carbonyl compounds (X ) OMe, OH, Cl, R).14,19 This
(20) In a formula, the “pro-S group R” and the “pro-R group R” are
written as RS and RR, respectively.
(21) Noe, C. R.; Knollmuller, M.; Oberhauser, B.; Steinbauer, G.;
Wagner, E. Chem. Ber. 1986, 119, 729-743.
Org. Lett., Vol. 6, No. 24, 2004
4609

in the RS group, namely, those for all of the protons on the
right side of the plane, should be negative.
positions of the two equilibria will be essentially the same,
because the proton and deuterium are essentially the same,
both chemically and sterically. Only the designated deuterium
and proton shown in the structures A and C, respectively,
will be subjected to the shielding effect of the p-Tol group,
and therefore the ∆δD and ∆δH values for 13-17 should be
always negative and positive, respectively. All the data shown
in Figure 2 are in good agreement with these considerations
based on steric factors.
The theoretical basis for the surprisingly stable sp con-
formation will be thoroughly investigated and clarified in
the future. It seems reasonable at this time to propose that
structures having both F and CN groups on the R-position
of aryl acetate lend themselves to creating the specific
stereoelectronic environment that produces this sp conforma-
tion.
In summary, because of the robust reactivity of CFTA
chloride (6) and the consistency in the signs of the ∆δ values
for the derivatized chiral secondary alcohols so far examined,
the CFTA procedure appears to be much more reliable than
any of the other existing methods.24 We stress that this
method is especially suitable for examining sterically hin-
dered secondary alcohols. The method can also be routinely
applied to very small amounts of material. Further clarifica-
tion of the theoretical basis for the above results and
acquisition of additional data to study the generality of the
CFTA procedure are objectives of our ongoing investigations.
To look further at the scope and limitation of our
technique, we next investigated the CFTA esters of several
chiral R-monodeuterated benzyl alcohols 13-17.22 The
stereogenic carbon in these structures carries two substituents
that chemically and sterically are almost identical. The ∆δH
values for the benzylic protons are always positive, and the
∆δD values for the benzylic deuteriums are always negative
(Figure 2).23 Because the stereochemistry of the proton and
the deuterium in each compound is reversed, the ∆δD values
obtained by 2D NMR should be reversed from the ∆δH values
obtained by 1H NMR, yet as observed, their absolute values
should be the same within the instrumental errors in
measurement.
In the conformational equilibria shown in Figure 3, there
should be two sp conformers, A and B, and C and D, for
the (S)- and (R)-CFTA ester diastereomers, respectively. The
Acknowledgment. We thank Drs. K. Kabuto of Tohoku
University and T. Yamazaki of TUAT for helpful discussions
2
and Dr. K. Omata of Tohoku University for measuring D
NMR spectra.
Supporting Information Available: Experimental details
of the synthesis and characterization of 7-17. This material
is available free of charge via the Internet at http://pubs.acs.org.
OL0479489
(22) Yamada, I.; Noyori, R. Org. Lett. 2000, 2, 3425-3427.
1
(23) The H NMR data of the corresponding MTPA esters showed no
consistent relationship between the signs of the ∆δH values and stereo-
Figure 3. Conformational equilibria for (S)- and (R)-CFTA esters
of (S)-dR-benzyl alcohols.
chemistry.
(24) Both enantiomers of the CFTA agent are now being commercially
developed.
4610
Org. Lett., Vol. 6, No. 24, 2004