Role of barium(II) in the determination of the absolute configuration of chiral amines by1H NMR spectroscopy

Article abstract of DOI:10.1021/jo0522207

The assignment of the absolute configuration of α-chiral primary amines by complexation of their MPA derivatives with Ba²⁺ and NMR analysis of the changes generated is presented. All that is required is (a) the derivatization of the amine of unknown configuration with one enantiomer of the auxiliary reagent (MPA), either (R) or (S)-α-methoxyphenylacetic acid, (b) the recording of the ¹H NMR spectrum of the resulting amide in MeCN-d₃, (c) the addition of Ba(ClO₄)₂ to the NMR tube, and (d) the recording of a second spectrum after a few minutes of shaking. The above steps take a few minutes and are followed by an analysis of the shifts (measured as Δδ^Ba) produced on the L ₁ and L₂ substituents of the amine by the addition of Ba²⁺ and their comparison with those expected from the conformational changes produced by the complexation. The conformational changes initiated by complexation have been subjected to NMR and CD studies, which showed that the formation of the complex shifts the equilibrium from an antiperiplanar (AP) to a synperiplanar (SP) form, leading to an increase of the shielding by the phenyl group of MPA of the substituent of the amine located on the same side. In addition, theoretical calculations [density functional theory (DFT)] provide further support for the formation, structure, and stability of the complexes. The general applicability of this method and the trustworthiness of the resulting configurational assignment were guaranteed with a series of amines of known absolute configuration and varied structures, used as test compounds. The method proposed is simple, fast, and inexpensive, and it requires a very small amount of sample, only one derivatization, and the recording of just two ¹H NMR spectra at room temperature. A graphical guide to simplify the application of this method is included.

Full text of DOI:10.1021/jo0522207

Role of Barium(II) in the Determination of the Absolute
1
Configuration of Chiral Amines by H NMR Spectroscopy
Rosa Garc´ıa,† Jose´ M. Seco,† Saulo A. Va´zquez,‡ Emilio Quin˜oa´,† and Ricardo Riguera*,†
Departamento de Qu´ımica Orga´nica and Departamento de Qu´ımica F´ısica, Facultad de Qu´ımica, and
Unidad de RMN de Biomole´culas Asociada al CSIC, UniVersidad de Santiago de Compostela,
E-15782 Santiago de Compostela, Spain
ricardo@usc.es
ReceiVed October 25, 2005
The assignment of the absolute configuration of R-chiral primary amines by complexation of their MPA
derivatives with Ba2+ and NMR analysis of the changes generated is presented. All that is required is (a)
the derivatization of the amine of unknown configuration with one enantiomer of the auxiliary reagent
(MPA), either (R) or (S)-R-methoxyphenylacetic acid, (b) the recording of the 1H NMR spectrum of the
resulting amide in MeCN-d3, (c) the addition of Ba(ClO4)2 to the NMR tube, and (d) the recording of a
second spectrum after a few minutes of shaking. The above steps take a few minutes and are followed
by an analysis of the shifts (measured as ∆δBa) produced on the L1 and L2 substituents of the amine by
the addition of Ba2+ and their comparison with those expected from the conformational changes produced
by the complexation. The conformational changes initiated by complexation have been subjected to NMR
and CD studies, which showed that the formation of the complex shifts the equilibrium from an
antiperiplanar (AP) to a synperiplanar (SP) form, leading to an increase of the shielding by the phenyl
group of MPA of the substituent of the amine located on the same side. In addition, theoretical calculations
[density functional theory (DFT)] provide further support for the formation, structure, and stability of
the complexes. The general applicability of this method and the trustworthiness of the resulting
configurational assignment were guaranteed with a series of amines of known absolute configuration
and varied structures, used as test compounds. The method proposed is simple, fast, and inexpensive,
and it requires a very small amount of sample, only one derivatization, and the recording of just two 1H
NMR spectra at room temperature. A graphical guide to simplify the application of this method is included.
Introduction
to secondary alcohols and primary amines. Currently, the
expansion experienced by this methodology has provided
researchers with new and more effective reagents that can be
applied to a variety of families of chiral substrates, such as
carboxylic acids, primary alcohols, diols, and, importantly, has
made available to those researchers an understanding of the
physical and chemical foundations of the method.2
The configurational assignment of chiral compounds in
solution based on H NMR spectroscopy has become a well-
established field of chemistry because of the impressive
development reached in the past decade. As originally developed
by Mosher,1 this approach resorted to just an auxiliary reagent
(R-methoxyphenyl acetic acid, MTPA, 1) that could be applied
1
† Departamento de Qu´ımica Orga´nica, Facultad de Qu´ımica and Unidad de
RMN de Biomole´culas Asociada al CSIC.
(1) (a) Dale, J. A.; Mosher, H. S. J. Am. Chem. Soc. 1973, 95, 512-
519. (b) Sullivan, G. R.; Dale, J. A.; Mosher, H. S. J. Org. Chem. 1973,
38, 2143-2147.
‡ Departamento de Qu´ımica F´ısica, Facultad de Qu´ımica.
10.1021/jo0522207 CCC: $33.50 © 2006 American Chemical Society
Published on Web 01/10/2006
J. Org. Chem. 2006, 71, 1119-1130
1119

Garc´ıa et al.
This knowledge about the chemical behavior of the various
auxiliary reagent/chiral substrate systems also allowed new ways
of simplifying the experimental procedure to be devised. Thus,
the usual approaches that required the use of two enantiomers
of the auxiliary reagent gave way to methods that required the
use of just a single enantiomer of the reagent and only one-half
the amount of substrate sample.3 The recent introduction of
solid-matrix-bound reagents has also contributed to these efforts
focused on making configuration determination an easier task.4
In the case of chiral secondary alcohols, there are two ways
to determine the absolute configuration using a single derivative.
They involve the controlled manipulation of the conformational
equilibrium of an R-methoxyphenylacetic acid (MPA, 2) ester
derivative. In one case,3a this is achieved simply by varying
the NMR probe temperature: the recording of two NMR spectra
at different temperatures reflects the variation of the populations
of the main conformers (synperiplanar, SP, and antiperiplanar,
AP) of the MPA ester, and this allows the absolute configuration
to be inferred by observing the selective shielding effects
produced on the substrate. In the other method,3c the in situ
complexation of the R-methoxyphenylacetic acid (MPA) ester
derivative with barium(II) induces conformational changes that
modify the chemical shifts of the alcohol part in a predictable
way and, therefore, allow the configuration to be deduced. In
both cases, theoretical studies and independent experimental
evidence such as circular dichroism results, give fully satisfac-
tory explanation to the effects observed.
FIGURE 1. Equilibrium between the two main conformers of (a) an
(R)-MPA amide and (b) an (S)-MPA amide [AP ) antiperiplanar
(major); SP ) synperiplanar (minor)]. Red curved arrows indicate the
anisotropic effect caused by the phenyl group on L1/L2 in each
conformer.
assignment of the absolute configuration of a chiral amine by
NMR spectroscopy of a single derivative are studied with a wide
range of structurally diverse amines. To facilitate the use of
this procedure, a chart for users has been incorporated at the
end of this article.
In the case of primary amines, the complexity of the
conformational equilibria with either MTPA or MPA amide
derivatives3 impedes the use of the approach based on temper-
ature variation. Therefore, double derivatization procedures are
still operative to elucidate the configuration of primary amines
with the stereogenic center in the R position. Although classical
auxiliary reagents (MTPA, MPA) are still employed for this
task,1b,5,6 more effective reagents such as Boc-phenylglycine
(BPG, 3) have been introduced in recent years.7
Results
1. NMR Studies: Addition of a Barium(II) Salt to MPA
Amides. Extensive theoretical calculations (MM, semiempirical,
and ab initio) and DNMR experiments have shown that, in
solution, arylmethoxyacetic acid amides (i.e., MPA amides) exist
mainly as two conformers in equilibrium.5 The major conformer,
denoted AP, has the CRsOMe and the CdO bonds antiperipla-
nar. In the minor conformer, denoted SP, these bonds are
synperiplanar. In both conformers, the aryl ring is virtually
coplanar with the CRsH bond (Figure 1). It is worth pointing
out that this conformational preference in arylmethoxyacetic acid
amides is opposite to that of arylmethoxyacetic acid esters. In
the latter, the SP conformers are the most populous.
We now present our results on the use of barium(II)
complexes generated in the NMR tube containing only one MPA
amide [either (R) or (S)]. A short communication describing
just the experimental technique and protocols has already been
published.8 In this article, we present theoretical (DFT calcula-
tions) and experimental (NMR and CD) data that provide a
detailed explanation and rationalization for the observed NMR
effects and the role of Ba2+ in the conformational population
dynamics, in how this affects the NMR spectra, and in the
differences observed between MPA amides and MPA esters.
The reliability, scope, and limitations of this method for the
According to these studies, in an (R)-MPA amide, L2 is
shielded by the phenyl group in the more stable AP conformer,
whereas L1 is shielded in the less stable SP conformer (Figure
1a). The opposite situation holds for the (S)-MPA amide (Figure
1b). As a result, the NMR signal corresponding to L2 is more
shielded by the phenyl group in the (R)- than in the (S)-MPA
amide. The difference between the chemical shifts of L2, when
the NMR spectra of both diasteromeric amides are compared,
is expressed as a negative ∆δRS value.9 In the other hand, L1 is
more heavily shielded in the (S)- than in the (R)-MPA amide,
and accordingly, its ∆δRS magnitude is expressed as a positive
value.
(2) (a) Seco, J. M.; Quin˜oa´, E.; Riguera, R. Chem. ReV. 2004, 104, 17-
117. (b) Seco, J. M.; Quin˜oa´, E.; Riguera, R. Tetrahedron: Asymmetry
2001, 12, 2915-2925.
(3) (a) Latypov, Sh. K.; Seco, J. M.; Quin˜oa´, E.; Riguera, R. J. Am.
Chem. Soc. 1998, 120, 877-882. (b) Seco, J. M.; Quin˜oa´, E.; Riguera, R.
Tetrahedron 1999, 55, 569-584. (c) Garc´ıa, R.; Seco, J. M.; Va´zquez, V.;
Quin˜oa´, E.; Riguera, R. J. Org. Chem. 2002, 67, 4579-4589.
(4) Porto, S.; Dura´n, J.; Seco, J. M.; Quin˜oa´, E.; Riguera, R. Org. Lett.
2003, 5, 2979-2982.
(5) (a) Latypov, Sh. K.; Seco, J. M.; Quin˜oa´, E.; Riguera, R. J. Org.
Chem. 1995, 60, 1538-1545. (b) Seco, J. M.; Latypov, Sh. K.; Quin˜oa´, E.;
Riguera, R. J. Org. Chem. 1997, 62, 7569-7574.
(6) Trost, B. M.; Bunt, R. C.; Pulley, S. R. J. Org. Chem. 1994, 59,
4202-4205.
The MPA amides of (-)-isopinocampheylamine10 (4) reflect
these facts. The phenyl group in the more stable AP conformer
(7) Seco, J. M.; Quin˜oa´, E.; Riguera, R. J. Org. Chem. 1999, 64, 4669-
4675.
(8) Lo´pez, B.; Quin˜oa´, E.; Riguera, R. J. Am. Chem. Soc. 1999, 121,
9724-9725.
(9) The ∆δRS value for a proton is the difference between its chemical
shift in the (R) derivative and its chemical shift in the (S) derivative.
1120 J. Org. Chem., Vol. 71, No. 3, 2006

Role of Ba(II) in Absolute Configuration Determination
In fact, Ba2+ generated the largest shifts, when added either as
its perchlorate or as iodide salt, and it was selected for this study.
In addition, MeCN-d3 was selected as the solvent of choice after
proving to be the most adequate among several deuterated
solvent systems tested.
Thus, when NMR spectra of the (R)- and the (S)-MPA amides
of (-)-isopinocampheylamine (4) were recorded in MeCN-d3
saturated with Ba(ClO4)2, significant changes (measured as ∆δBa
values)11 with respect to the spectra recorded in the absence of
the salt (see above) were observed. These can be summarized
as follows: (a) In the (R)-MPA amide, the signals corresponding
to H(6′ax) and H(6′eq) shifted downfield upon addition of Ba2+
(∆δBa ) -0.07 and -0.11 ppm, respectively), whereas the
signal generated by CH3(10′) experienced an upfield shift (∆δBa
) +0.20 ppm), as shown in Figure 2a,b. (b) In the (S)-MPA
amide, the shifts upon addition of Ba2+ were in the opposite
directions to the above [∆δBa ) +0.27 and +0.05 ppm for
H(6′ax) and H(6′eq), respectively; ∆δBa ) -0.09 ppm for CH3-
(10′)], as shown in Figure 2c,d.
These changes can be ascribed to a redistribution of the
populations of the AP and SP conformers in equilibrium,
induced by the presence of the barium(II) salt. The process
would be similar to the one observed in MPA esters.3c,12 To be
precise, in the case of an MPA amide, as the concentration of
barium(II) increases, the less stable SP conformer becomes more
populated because of the formation of a complex among the
divalent metal and the two oxygens of the methoxy and carbonyl
groups. Correspondingly, the population of the more stable AP
conformer diminishes.
As a result, addition of the barium(II) salt to the (R)-MPA
amide of (-)-isopinocampheylamine (4) increases the number
of molecules in the SP conformation [CH3(10′) under the
shielding influence of the phenyl group], so in the average
spectrum, the signal of this methyl group appears more shielded.
On the other hand, the number of molecules in the AP
conformation decreases, and consequently, the signals of the
methylene group [H(6′ax) and H(6′eq) under the shielding
influence of the phenyl group] experience a deshielding effect
(Figure 2a,b).13 A parallel analysis applies to the (S)-MPA
amide, with opposite consequences for the shielding/deshielding
of those groups (Figure 2c,d).
2. Effect of Barium(II) on the Conformational Equilibria
of MPA Amides and MPA Esters as Shown by NMR
Spectroscopy: A Comparison. A way to confirm whether the
equilibrium present in MPA amides is modified by the metal
cation, as suggested by the NMR experiments described above,
consists of comparing the NMR behavior of an MPA amide
and an MPA ester containing the same carbon skeleton [i.e.,
(S)-butan-2-amine (5) versus (S)-butan-2-ol (6)] when treated
with barium(II). Because of the different conformer composi-
tions of the two substrates (SP, main conformer in esters; AP,
main conformer in amides; see Figure 3), the effect caused by
the metal (selective stabilization of the SP conformer) should
be the opposite for L1 (methyl group) and L2 (ethyl group) in
the two cases.
FIGURE 2. Partial 1H NMR spectrum of the (R)-MPA amide of (-)-
isopinocampheylamine (a) before and (b) after addition of barium(II).
1
Partial H NMR spectrum of the (S)-MPA amide of (-)-isopinocam-
pheylamine (c) before and (d) after addition of barium(II).
of the (R)-MPA amide shields the H(6′) protons, whereas the
same group shields CH3(10′) in the less stable SP conformer
(Figure 2a). In the (S)-MPA amide, the situation is the
opposite: CH3(10′) is shielded in the more stable AP conformer,
whereas the H(6′) protons are shielded in the less stable SP
form (Figure 2c). The overall result is that the H(6′) protons
are more shielded by the phenyl group in the (R)- than in the
(S)-MPA amide, giving rise to negative ∆δRS values [-0.075
and -0.053 ppm for H(6′ax) and H(6′eq), respectively], whereas
the opposite situation holds for CH3(10′) (a positive ∆δRS value,
+0.108 ppm).
To check whether a controlled modification of the equilib-
rium, useful for configurational assignment, could be possible
by selective formation of a complex with an adequate metal
ion, salts of Li+, Na+, K+, Cs+, Mg2+, Ca2+, and Ba2+ were
added to NMR tubes containing the (R)-MPA amide of (-)-
isopinocampheylamine in MeCN-d3 (progressive additions of
salt ranging from 1 equiv to saturation). The salts of the
(11) The ∆δBa value for a proton is the difference between its chemical
shift before the addition of barium(II) and its chemical shift after the addition
of barium(II).
(12) It is necessary to recall that, in MPA amides, the AP/SP equilibrium
in the absence of barium is shifted in favor of AP, whereas in MPA esters,
it is shifted toward SP.
1
monovalent cations produced no observable shifts in the H
NMR spectra, whereas those of Mg2+, Ca2+, and Ba2+ induced
visible changes in the chemical shifts of the substrate moiety.
(13) More detailed spectra of the (R)-MPA amide of (-)-isopinocam-
pheylamine and the assignment of its configuration can be found in Figure
1-SI in the Supporting Information.
(10) This amine was chosen as a model compound because it presents
NMR signals that are easily identified.
J. Org. Chem, Vol. 71, No. 3, 2006 1121

Garc´ıa et al.
FIGURE 3. (a) Main conformers present in the equilibria of an (R)-
MPA amide [X ) NH; i.e., (S)-butan-2-amine (5)] and an (R)-MPA
ester [X ) O; i.e., (S)-butan-2-ol (6)] in the absence and in the presence
of Ba2+, as suggested from the NMR data. (b) The same for the
corresponding (S)-MPA amide and ester.
FIGURE 4. 1H NMR spectra (MeCN-d3) of (a) the (R)-MPA amide
and (b) the (S)-MPA amide of (S)-butan-2-amine (5) without and with
barium(II). 1H NMR spectra (MeCN-d3) of (c) the (R)-MPA ester and
(d) the (S)-MPA ester of (S)-butan-2-ol (6) without barium(II).
In fact, this is the case when the NMR spectra are com-
pared: The shifts strongly suggest that the ethyl group in the
(R)-MPA amide is shielded in the AP conformer (more stable
and populated), whereas in the (S)-MPA amide, the same group
is shielded in the SP conformer (less stable and populated). In
the (R)-MPA ester, the ethyl group is shielded in the AP
conformer (less stable and populated), whereas in the (S)-MPA
ester, it is shielded in the SP conformer (more stable and
populated). The opposite reasoning holds for the methyl groups,
as shown in Figure 3 (L1) Me, L2) Et). As a consequence,
TABLE 1. Chemical Shifts (∆), ∆δRS and ∆δRS/Ba Values for the
(R)- and (S)-MPA Derivatives of (S)-Butan-2-amine (5) and
(S)-Butan-2-ol (6) (MeCN-d3, 250 MHz)
H(1′)
H(3′)
H(4′)
(R)-MPA
(S)-MPA
1.111
1.057
+0.054
1.005
1.100
-0.095
1.433
1.482
-0.049
1.468
1.352
+0.116
0.781
0.868
-0.087
0.813
0.624
+0.189
∆δRS
(R)-MPA + Ba2+
(S)-MPA + Ba2+
∆δRS/Ba
1
the H NMR spectrum of the (R)-MPA amide is similar to the
spectrum of the (S)-MPA ester, and the spectrum of the (S)-
MPA amide is similar to the spectrum of the (R)-MPA ester14
(Figure 4).
(R)-MPA
(S)-MPA
∆δRS
1.057
1.184
-0.127
1.544
1.445
+0.099
0.840
0.636
+0.204
1
Interestingly, the H NMR spectrum of the (R)-MPA amide
plus barium(II) is similar to the spectrum of the (R)-MPA ester
without barium(II), and the spectrum of the (S)-MPA amide
plus barium(II) looks like the spectrum of the (S)-MPA ester
without barium(II) (Figure 4). If we calculate the ∆δRS/Ba
values15 for the protons of the amide and compare them with
the ∆δRS values for those of the ester, we obtain very similar
values and identical signs (Table 1), strongly suggesting that
the most representative NMR conformer, in both cases, must
be the same SP conformer.
3. CD Studies. Additional support for this interpretation of
the NMR data was obtained from an independent technique such
as CD. The CD spectra of (R)- and (S)-MPA amides show a
Cotton effect at 226 nm associated to an n-π transition of the
carbonyl group. The Cotton effect is not affected by the chirality
of the amine moiety, but it depends on the configuration at the
asymmetric center of the auxiliary MPA: It is positive in the
(R)-MPA amides and negative in the (S)-MPA amides. The
octant rule16 predicts opposite signs and a similar intensity for
the AP and SP conformers [positive for the AP and negative
for the SP conformer of an (R)-MPA derivative and the opposite
behavior for an (S)-MPA derivative; Figure 5], and the intensity
and the sign of this band in the average CD spectrum constitutes
a measure of the AP/SP equilibrium. Thus, CD constitutes a
(14) The spectra should be alike but not identical because of the more
effective shielding caused by the phenyl group in SP than in AP conformers,
among other causes. The ∆δRS values should have opposite signs in esters
and amides, with smaller absolute values for the latter because of the
predominance of the AP conformer. See ref 5a and also: Latypov, Sh. K.;
Seco, J. M.; Quin˜oa´, E.; Riguera, R. J. Org. Chem. 1995, 60, 504-515.
(15) The ∆δRS/Ba value for a proton is the difference between its chemical
shift in the (R) derivative with barium(II) and its chemical shift in the (S)
derivative with barium(II).
(16) Eliel, E. L.; Wilen, S. H.; Mander, L. N. Stereochemistry of Organic
Compounds; Wiley-Interscience: New York, 1994.
1122 J. Org. Chem., Vol. 71, No. 3, 2006

Role of Ba(II) in Absolute Configuration Determination
FIGURE 5. Cotton effects for the main conformers of MPA amides and octant rules. The sign of the Cotton effect is defined by the quadrant
occupied by the phenyl group.
very useful technique for detecting the changes in the AP/SP
ratio suggested by the NMR studies.
Strong evidence in favor of both the modification of the
equilibrium by the divalent cation and the existence of Cotton
effects of opposite signs for the two MPA conformers (SP/AP)
was provided by the CD monitoring of the addition of barium-
(II) to the (R)-MPA ester of methanol (7) and to the (R)-MPA
amide of methylamine (8). Prior to the addition, the Cotton effect
of the ester is opposite to that of the amide, because although
both exhibit the same chirality, the conformational compositions
are different (SP is the major conformer in the MPA ester, and
AP is the major conformer in the MPA amide). After the
addition, the Cotton effect enhances its magnitude and keeps
its negative sign in the case of the ester, suggesting an increase
in the population of the major SP conformer. In the amide, the
Cotton effect diminishes in magnitude keeping its positive sign,
suggesting a decrease of the population of the AP conformer
(Figure 6).
Further verification was obtained when a series of CD spectra
of the (R)-MPA amide of (+)-bornylamine (9) were recorded
in the absence and in the presence of increasing amounts of
Ba(ClO4)2, and the following facts were observed: (a) In the
absence of barium(II), the positive Cotton effect suggests the
presence of a predominant AP conformer (Figure 7). (b) The
addition of increasing amounts of barium(II) causes a decrease
in the intensity of the positive Cotton band, indicating that the
equilibrium is being modified and that the population of the
SP conformer is growing, in full accordance with the NMR
studies (Figure 7).
4. Additional NMR Studies. Both the CD and NMR
experiments are coincident when pointing to the coordination
of barium(II) to the oxygens of the methoxy and carbonyl groups
as the cause of the increase of the SP populations and the
corresponding responses in the spectra.
FIGURE 6. CD spectra of the (R)-MPA ester of methanol (7) and the
(R)-MPA amide of methylamine (8) in the presence and in the absence
of Ba2+ [c ) 10-3 M, V ) 300 mL, MeCN, sequential additions of 10
mg of Ba(ClO4)2].
Additional experimental evidence that gives further support
to this hypothesis is the following:
spectra after the addition of barium(II), strongly suggesting that
these groups are not directly involved in the complexation
processes.
(a) In the 13C NMR spectrum of the (R)-MPA amide of (S)-
butan-2-amine (5), the carbonyl signal at 169.1 ppm shifts
downfield to 171.4 ppm after the addition of the barium(II) salt
(a 2.3 ppm shift). In the (S)-MPA amide, the shift is even larger,
from 169.2 to 174.2 ppm (a 5.0 ppm shift). The increase of the
electropositive character experienced by the sp2 carbon atoms
after the coordination of the carbonyl groups to the cation
justifies these downfield shifts.
(c) The signal corresponding to CRH of an MPA amide shifts
downfield after the addition of barium(II), and the same
phenomenon was observed for the MPA esters. The explanation
can be found in the following fact: In the AP conformer, this
proton lies in the shielding cone of the neighboring carbonyl
group, whereas in the SP conformer, it lies in the deshielding
zone (Figure 8). Therefore, a redistribution of the AP/SP
populations in favor of the SP conformer can generate the
(b) In all of the compounds studied, no significant shifts were
1
observed for the signals of the NH groups in the H NMR
J. Org. Chem, Vol. 71, No. 3, 2006 1123

Garc´ıa et al.
shielded in the (S) than in the (R) derivative, whereas those of
CH3(4′) and CH2(3′) are more shielded in the (R) than in the
(S). The sequential addition of barium(II) changes the confor-
mational preference, and correspondingly, the signals that
previously were shielded in the (R) amide are now shielded in
the (S) and vice versa.
(e) The determinations of stoichiometry and association
constant (Ka) were undertaken from solution NMR data. The
stoichiometry was calculated by means of the method of
continuous variation (Job’s method).17 It involved preparing a
series of solutions containing both the (R)-MPA amide of (+)-
bornylamine (9) and barium perchlorate in varying proportions
(0 < [Ba2+]0/{[Ba2+]0 + [(R)-MPA amide]0} < 1) and where
the total initial concentration of Ba2+ + (R)-MPA amide was
constant for each solution. The chemical shift corresponding to
H(6′ax) of the amide, which is sensitive to complex formation,
was selected as the experimentally observed parameter. The data
were plotted in the form X(R)-MPA amide∆δBa versus XBa , and
2+
FIGURE 7. CD spectrum of the (R)-MPA amide of of (+)-
the position of the maximum indicated a 1:1 stoichiometry for
2+
bornylamine (9) in the presence of increasing amounts of Ba [c )
2+
10-3 M, V ) 300 µL, MeCN, sequential additions of 10 mg of Ba-
(ClO4)2].
the complex (XBa ) 0.5) (Figure 10a).
The association constant (Ka) was calculated according to the
Benesi-Hildebrand (Hanna-Ashbaugh) treatment.17a,18 The plot
of 1/∆δBa [δH(6ax)] against 1/[Ba2+]0 gave a linear representa-
tion with a slope 1/Ka∆δBamax and intercept 1/∆δBamax, affording
a value of Ka ) 5.39 M-1 (Figure 10b).
5. Theoretical Calculations. Next, we decided to perform
theoretical calculations to determine whether the conclusions
about the formation of the complexes obtained from the
experimental NMR and CD studies could be reproduced and
confirmed by computational methods. The (R)-MPA amide of
propan-2-amine (10) [(R)-N-isopropyl-2-methoxy-2-phenyl-
acetamide] was chosen as a model compound, and the stability
and structure of its complexes with Ba2+ were investigated.
First, a series of geometry optimizations was carried out by
an appropriate selection of the initial values for the
OCRsCdO, HNsCH, and MeOsCRC(dO) dihedral angles.
observed downfield shift. For instance, -0.287 and -0.289 ppm
downfield shifts are experienced by the (R)- and (S)-MPA
amides of (-)-isopinocampheylamine (4). The shielding ob-
served for the signal of the methoxy group is also consistent
with the proposed AP/SP redistribution. This group is deshielded
in the AP form and shielded in the SP form; thus, it moves
upfield because of the increase in the SP conformer after
complexation with barium(II) (Figure 8).
(d) The changes produced in the NMR spectra of the amides
by addition of barium(II) were quantitatively monitored. Figure
9 shows the spectral evolution of a mixture of the (R)- and (S)-
MPA amides of (S)-(+)-butan-2-amine (0.7 equiv and 0.3 equiv,
respectively) in MeCN-d3 during the addition of barium
perchlorate. Before the addition, the CH3(1′) signals are more
FIGURE 8. Effect of the shielding cone of the carbonyl group on the CRH and MeO groups of the (R)- and (S)-MPA amides of
(-)-isopinocampheylamine (8) (a) before and (b) after the addition of Ba.2+
1124 J. Org. Chem., Vol. 71, No. 3, 2006

Role of Ba(II) in Absolute Configuration Determination
FIGURE 9. Evolution of the 1H NMR spectra (250 MHz) of a mixture of (R)- and (S)-MPA amides of (S)-(+)-butan-2-amine (5) (0.7/0.3 equiv;
MeCN-d3) in the presence of increasing amounts of Ba(ClO4)2.
torsion angles calculated for the optimized structures are detailed
in Figure 11. The acronyms used to name the conformers include
three pairs of letters, which refer to the classification of torsion
angles proposed by Klyne and Prelog.19 The first two (capital)
letters refer to the OCsCO torsion angle, the two letters in the
middle refer to the HNsCH torsion angle, and the last two
letters refer to the MeOsCC(dO) dihedral angle.
At the B3LYP/LanL2DZ level of theory, all of the stable
conformations in the gas phase present an antiperiplanar (AP)
OCCO arrangement, and in all of these structures, a
MesO‚‚‚HsN intramolecular hydrogen bond is exhibited. The
present results contrast with those reported for the (R)-MPA
ester of propan-2-ol,3c as the stabilities of the two different
OCsCdO conformations of the ester were calculated to be
rather similar. To a certain degree, the different behavior found
here for the (R)-MPA amide of propan-2-amine arises from the
hydrogen-bond interaction, which stabilizes the AP conforma-
tions. In addition, the effect of electron conjugation of the N
atom on the carbonyl group, expected to be more pronounced
than that of the O(sCdO) atom, can result in a significant
negative partial charge on the carbonyl oxygen, which, in turn,
destabilizes the SP OCCO arrangement.
The antiperiplanar (ap) orientation of the MeOsCC(dO)
fragment is energetically favored over the (-)-synclinal (-sc)
by about 0.7 kcal/mol. To a certain extent, this might be a
consequence of the different strength of the hydrogen-bonding
interactions in the ap and -sc MeOsCC(dO) arrangements,
given that the O‚‚‚H distance in the former conformation (∼2.0
Å) is about 0.2 Å shorter than that in the latter. As far as the
orientation of the propan-2-amine moiety is concerned, the
calculations predict that, in the gas phase, the synperiplanar (sp)
arrangement is energetically disfavored by approximately 1.7
kcal/mol with respect to the ap and anticlinal (ac) arrangements.
For the gas phase, all attempts to optimize a conformation
with a synperiplanar OCsCO arrangement were unsuccessful,
FIGURE 10. (a) Job plot for the determination of stoichiometry. (b)
Benesi-Hildebrand data treatment for the determination of the as-
sociation constant.
and the optimizations led to AP conformations. However, SP
conformations could be optimized when solvent effects, as
calculated with the Onsager model, were taken into consider-
ation. The reason for this is that the interaction with the medium
is stronger for SP conformations (showing a dipole moment of
The phenyl group was placed in its preferred orientation as
predicted in our previous work3c for the (R)-MPA ester of
propan-2-ol. The relative energies and the values of the above
J. Org. Chem, Vol. 71, No. 3, 2006 1125

Garc´ıa et al.
FIGURE 11. Geometries, relative energies, relative populations, and relevant torsional angles for the antiperiplanar conformations of the (R)-MPA
amide of propan-2-amine (10). Torsion angles (in degrees) in the (a) gas phase and (b) Onsager model; relative energies in kcal/mol (B3LYP/
landL2DZ); and relative polpulations (Boltzmann, 298.15 K) in the (c,d) gas phase and (e,f) Onsager model, respectively.
TABLE 2. (a) Relative Energiesa (in kcal/mol), (b) Relative
Populations,b and Relevant Torsion Angles (in degrees) for the
about 5.4 D) than for AP conformations (µ ≈ 4.2 D). The
Conformations of the (R)-MPA Amide of Propan-2-amine (10)
stabilization of the SP conformations by the interaction with
the solvent is, however, not sufficient for these conformations
to contribute substantially to the total population at 298.15 K.
As shown in Table 2, the relative energies of the SP conforma-
tions are larger than 5.0 kcal/mol, and consequently, the
populations at 298.15 K, calculated by the Boltzmann distribu-
tion, are negligible. It is worth noticing that the relative energies
for the AP conformations do not vary significantly when going
from the gas phase to solution. This is because the dipole
torsion angles (deg)
(Me)Os
CRsCdO
HsNs
C(1)sH
MesOs
CRsC(dO)
conformer
∆E
population
(SP,ac,sc)
(SC,-ac,sc)
(SC,sp,sc)
(SC,ac,ap)
(SP,-ac,ap)
(SC,sp,ap)
5.7
5.8
7.7
6.2
6.1
8.1
28.3
32.1
30.5
63.1
66.9
66.0
149.5
-144.1
2.6
143.9
-149.6
-3.0
61.3
-60.9
61.6
172.4
172
negligible
negligible
negligible
negligible
negligible
negligible
172.3
a All calculations were performed with the B3LYP/LanL2DZ basis set.
b Values calculated from the Boltzmann distribution at 298.15 K.
(17) (a) Fielding, L. Tetrahedron 2000, 56, 6151-6170. (b) Gil, V. M.
S.; Oliveira, N. C. J. Chem. Educ. 1990, 473-478.
(18) Hanna, M. W.; Ashbaugh, A. L. J. Phys. Chem. 1964, 68, 811-
816.
moment of the molecule has almost the same value in the six
AP conformations.
(19) Klyne, W.; Relog, V. P. Experientia 1960, 16, 521-523.
1126 J. Org. Chem., Vol. 71, No. 3, 2006

Role of Ba(II) in Absolute Configuration Determination
FIGURE 12. (a) Torsion angles in the gas phase (in degrees), (b) calculated relative energies (B3LYP/landL2DZ) (in kcal/mol), and (c) relative
populations (Boltzmann, 298.15 K) in the gas phase for the conformations of the complexes of the (R)-MPA amide of propan-2-amine with Ba2+
.
The results obtained for the complexes of the (R)-MPA amide
To summarize, the calculations suggest that the conforma-
of propan-2-amine with Ba2+ are given in Figure 12. As for
tional equilibrium for the (R)-MPA amide of propan-2-amine
favors the AP (OCCO) conformations, but in the presence of
Ba2+, the formation of complexes with SP conformations is
strongly preferred. At 298.15 K, virtually all of the complexes
with Ba2+ exhibit an SP (OCCO) conformation.
the complexes of the MPA ester with Ba2+ and Mg2+ 20
in
,
general, the calculations including solvent effects showed
convergence problems, and so we present only the gas-phase
results. In all of these optimized structures, the MeOsCC(dO)
fragment presents only one orientation. Therefore, only the OCs
CO and the HNsCH torsion angles were taken into account to
name the conformers. As for the complexes of the MPA ester
6. Application to the Assignment of Absolute Configura-
tions of Chiral Primary Amines. To check the general
applicability of this chelation process for the assignment of the
absolute configurations of amines, its scope, and its limitations,
the NMR spectra of a series of (R)- and (S)-MPA amides derived
from primary amines of known absolute configuration, including
all kind of structures (acyclic, cyclic, bicyclic, with aliphatic
and aromatic substituents, with the presence of other functional
groups, etc.), were studied in the absence and in the presence
with Ba2+ and Mg2+ 3c
, two different conformations were found
for the OCsCO arrangement (SP and SC). The energy
difference between these two conformations is very large (∼30
kcal/mol), almost 3 times larger than that predicted for the
complexes of the MPA ester with Ba2+ 3c
. This is an expected
result that can be rationalized from basic concepts of valence
bond theory. The hybridization of both the amidic N atom and
the O(sCdO) atom in the SC Ba2+ complex of the MPA amide
and in the AC Ba2+ (ap Ba2+ in ref 3c) complex of the MPA
ester, respectively, have a significant sp2 character. For the latter,
the interaction of the O(sCdO) atom with the cation takes place
via the sp2 lone pair, whereas for the former, the interaction
occurs basically through the electron pair conjugated with the
carbonyl group (weaker interaction). The present results,
therefore, point out that, in effect, at 298.15 K, all of the
calculated complexes of the MPA amide with Ba2+ present an
SP conformation. Concerning the HNsCH arrangement in the
SP conformations, the calculations predict the ap orientation to
be more stable than the sp by about 2.5 kcal/mol.
of Ba2+ 21
.
Figures 13 and 1422 show the structures of the amides along
with the displacement of the signals produced by the addition
of the barium salt, measured as ∆δBa.11 The signs of these shifts
are fully consistent with the selective formation of the Ba2+
chelate with the SP conformer and the corresponding increase
in its population. In all cases, substituents at one side of the
chiral center move to higher field upon addition of Ba2+
,
whereas those on the other side shift to lower field, as shown
in Figures 2 and 3. Thus, a regular distribution of the signs of
(21) The spectra were recorded in MeCN-d3, and the Ba2+ was added to
the NMR tube in the form of dry, finely powdered Ba(ClO4)2.
(22) A 2:1 molar ratio of Ba(ClO4)2 to amide is enough to obtain
satisfactory ∆δBa values, as shown in Figure 13. Nevertheless, larger ∆δBa
values can be obtained if the salt is added until saturation (see Figure 14
and Figure 2-SI in the Supporting Information).
(20) Onsager, L. J. Am. Chem. Soc. 1936, 58, 1486-1493.
J. Org. Chem, Vol. 71, No. 3, 2006 1127

Garc´ıa et al.
FIGURE 14. Selected ∆δBa values (ppm) obtained from the 1H NMR
spectra (MeCN-d3) of the (R)- (in bold) and (S)- (in italics) MPA amides
of structurally representative chiral amines upon the addition of
Ba(ClO4)2 into the NMR tube until saturation.
second spectrum is recorded after solid anhydrous Ba(ClO4)2
is added to the NMR tube until saturation,21 and the signals of
L1 and L2 are assigned once more. Then, the ∆δBa values and
signs are calculated, and the graphical models shown in Figure
15 are used to place L1 and L2 in space in accordance with the
signs (positive or negative) of ∆δBa.
7. Scope and Limitations. As the shifts employed for the
assignments are based on the selective formation of the barium-
(II) complex, the presence in the substrate of functional groups
(i.e., carbonyl and NH groups)23 that could compete in the
formation of the required complex might pose a limitation on
the application of the method. In this sense, it is important to
realize that Figures 13 and 14 show a number of amines with
additional carbonyl (11, 15, 16, 18, 21) and NH (23, 24) groups
and that, upon addition of Ba2+, they behave according to the
model of Figure 3. However, caution should be taken in the
case of amines bearing a 1,4-ketoester moiety. Inadequate shifts
have been reported in those cases,24 most likely because of the
competing complexation brought about by the extra carbonyls.
FIGURE 13. Selected ∆δBa values (ppm) obtained from the 1H NMR
spectra (MeCN-d3) of the (R)- (in bold) and (S)- (in italics) MPA amides
of structurally representative chiral amines upon the addition of 2 equiv
of Ba(ClO4)2 into the NMR tube.
∆δBa, which are positive for one side of the chiral center and
negative for the other, is obtained.
These results demonstrate that the signs of ∆δBa, obtained
with either (R)-MPA or (S)-MPA, can be used to correlate the
position of the phenyl group of the auxiliary agent (MPA of
known configuration) with the spatial position of the L1 and L2
substituents of the amine chiral center and therefore of its
absolute configuration. This is so because the amine substituent
that faces the phenyl group (the one under its shielding cone)
in the AP conformer (i.e., L2 in Figure 3a or L1 in Figure 3b)
undergoes a downfield shift (negative ∆δBa value) upon the
addition of the barium salt because of the increase of the
population of the SP conformer. The other substituent (i.e., L1
in Figure 3a or L2 in Figure 3b) undergoes the opposite
phenomenon, experiencing an upfield shift (positive ∆δBa value).
It is important to point out that this correlation operates with
the (R)- and (S)-MPA derivatives, and therefore, the assignment
can be carried out using only one of those derivatives, either
the (R)- or the (S)-MPA amide, which constitutes a clear
advantage when compared to the usual procedures that require
the preparation of both MPA derivatives.
8. Conclusion. We have shown that the absolute configuration
of an R-primary amine can be assigned by NMR monitoring of
the changes produced upon addition of Ba(ClO4)2 to either its
(R)- or (S)-MPA amide in MeCN-d3. A representative number
of amines of known configuration and diverse structural features
have been used to demonstrate the general applicability of this
method and to assess its scope and limitations.
As a guide to researchers interested in the use of Ba2+
chelation shifts for the assignment of the absolute configuration
of R-chiral primary amines, we present a stepwise summary of
the procedure in Figure 15: The amine (unknown configuration)
(23) (a) Shibasaki, M.; Yamada, Y. M. A. Tetrahedron Lett. 1998, 39,
5561-5564. (b) Rees, W. S.; Caballero, C. R.; Hesse, W. Angew. Chem.,
Int. Ed. Engl. 1992, 31, 735-737. (c) Norman, J. A. T.; Pez, G. P. J. Chem.
Soc., Chem. Commun. 1991, 971-972. (d) Rees, W. S.; Carris, M. W.;
Hesse, W. Inorg. Chem. 1991, 30, 4479-4481. (e) Rabhofer, W.; Mu¨ller,
W. M.; Vo¨gtle, F. Chem. Ber. 1979, 112, 2095-2119.
1
is derivatized either with (R)- or with (S)-MPA. The H NMR
spectrum of the resulting MPA amide is recorded in MeCN-d3,
and the signals of L1 and L2 are unambiguously assigned. A
(24) Earle, M. A.; Hultin, P. G. Tetrahedron Lett. 2000, 41, 7855-7858.
1128 J. Org. Chem., Vol. 71, No. 3, 2006

Role of Ba(II) in Absolute Configuration Determination
FIGURE 15. Diagram to deduce the absolute configuration of an R-chiral primary amine from the ∆δBa experimental signs of either the (R)- or
the (S)-MPA amide.
Huzinaga full double-ú (D95)29 functions for first row and the Los
Alamos effective core potentials plus double-ú functions on Na-
Bi.30-32
High-level theoretical calculations and experimental data (CD,
NMR) support the preferential chelation of one of the conform-
ers (SP) of the MPA amides with the Ba2+ cation. This change
is transmitted via an aromatic shielding effect to the NMR
spectra, where the modifications in the chemical shifts of the
L1/L2 substituents of the amine are measured as ∆δBa values.
For the conformers of the (R)-MPA amide of propan-2-amine,
solvent effects were considered using the Onsager model,20,33 which
considers a polarizable reference molecule as being in vacuum in
a spherical cavity. The surroundings of the molecule are treated as
a continuum with the macroscopic dielectric constant of the solvent
(ꢀ ) 36.64 for acetonitrile at room temperature). The reference
molecule is treated as a point dipole placed in the center of the
cavity. A dipole in the molecule will induce a dipole in the medium,
and the electric field applied by the solvent dipole will, in turn,
interact with the molecular dipole, leading to net stabilization. With
this model, full geometry optimizations were performed by using
the self-consistent reaction field method.33 Attempts were made to
calculate the conformational energies for the complexes of Ba2+
using the Onsager solvation model, but in general, success was
not achieved. All calculations were performed with the Gaussian
98 series of programs.34
General Procedures. The MPA amides were prepared by
treatment of the amine (1.0 equiv) with the corresponding (R)- and
(S)-MPA (1.0 equiv) in the presence of 1-(3-dimethylaminopropyl)-
3-ethylcarbodiimide hydrochloride (EDC, 1.2 equiv) and DMAP
(1.0 equiv) in dry CH2Cl2 and under a nitrogen atmosphere. The
reaction was stirred at room temperature for 2 h. The organic layer
was then washed sequentially with water, HCl (1 M), water,
This procedure is fast, reliable, and inexpensive. It requires
a very small amount of sample, only one derivatization with
one enantiomer of the auxiliary agent, and the recording of just
two 1H NMR spectra at room temperature. It represents a great
advantage in relation to the previous procedures available to
date that require two derivatizations, twice the amount of sample,
and the use of two enantiomers of the auxiliary agent.
Unpublished results indicate that the “mix-and-shake” method25
can also be adapted to the Ba2+ complexation procedure.
A graphical model is presented to facilitate the assignment
of the absolute configurations of amines from the NMR spectra.
Experimental Section
Computational Methods. Density functional theory (DFT)
calculations were performed to study the conformational preferences
of the (R)-MPA amide of propan-2-amine (used as a model
compound) as well as its complexes with Ba2+. In particular, we
employed the B3LYP method, which combines Becke’s three-
parameter nonlocal hybrid exchange potential26 with the nonlocal
correlation functional of Lee, Yang, and Parr.27 This formalism has
been found to be very reliable in the description of ion-molecule
complexes.28 The basis set employed in the calculations was the
standard LanL2DZ basis set, which consists of the Dunning/
(28) (a) Luna, A.; Amekraz, B.; Tortajada, J. Chem. Phys. Lett. 1997,
266, 31-37. (b) Hoyeau, S.; Ohanessian, G. Chem. Phys. Lett. 1997, 280,
266-272. (c) Luna, A.; Amekraz, B.; Morizur, J.-P.; Tortajada, J.; Mo´,
O.; Ya´n˜ez, M. J. Phys. Chem. 1997, 101, 5931-5941. (d) Luna, A.;
Amekraz, B.; Tortajada, J.; Morizur, J.-P.; Alcam´ı, M.; Mo´, O.; Ya´n˜ez, M.
J. Am. Chem. Soc. 1998, 120, 5411-5426. (e) Kemper, P. R.; Weis, P.;
Bowers, M. T.; Maitre, P. J. Am. Chem. Soc. 1998, 120, 13494-13500.
(29) Dunning, T. H., Jr.; Hay, P. J. In Modern Theoretical Chemistry;
Schaefer, H. F., Ed.; Plenum Press: New York, 1976; Vol. 3.
(30) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 270-283.
(31) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 284-298.
(32) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 299-310.
(33) (a) Wong, M. W.; Wiberg, K. B.; Frisch, M. J. J. Chem. Phys. 1991,
95, 8991-8998. (b) Wong, M. W.; Frisch, M. J.; Wiberg, K. B. J. Am.
Chem. Soc. 1991, 113, 4776-4782.
(25) These results will be published elsewhere. The mix-and-shake
method is described in ref 4.
(26) (a) Becke, A. D. J. Chem. Phys. 1993, 98, 5648. (b) Becke, A. D.
J. Chem. Phys. 1992, 96, 2155-2160.
(27) Lee, C.; Yang, W.; Parr, R. G. Phys. ReV. 1988, B37, 785-789.
J. Org. Chem, Vol. 71, No. 3, 2006 1129

Garc´ıa et al.
NaHCO3 (saturated), and water. Next, it was dried with Na2SO4
(anhydrous) and concentrated under reduced pressure to yield the
MPA amide. Final purification was achieved by HPLC (µ-Porasil,
3 mm × 250 mm or Spherisorb S5W 5 µm, hexanes/ethyl acetate).
All compounds were characterized by optical rotation, NMR
spectroscopy (1D, 2D), and MS (EI).
1D 13C NMR Spectra. Size, 64 K; pulse length, 3.5 µs (30°);
1024 acquisitions.
2D COSY Spectra. Sequence: D1-90-t1-90-t2; relaxation
delay D1 ) 0.5 s; 90° pulse, 8.5 µs.
2D NOESY Spectra. Sequence: D1-90-t1-90-τmix-90-t2;
relaxation delay D1 ) 0.5 s; mixing time (τmix) ) 0.5s, 90° pulse,
8.5 µs; TPPI mode, NS ) 64.
For experimental details on compounds 9, 11, 15, and 17, see
ref 5a.
(-)-Isopinocampheyl-(R)-2-methoxy-2-phenylacetamide [(R)-
1
4]. [R]D) -96.20 (c ) 0.005, CHCl3); H NMR (250.13 MHz,
General Procedure for Configuration Assignment. The amine
is reacted with either the (R)- or the (S)-MPA reagent, the spectrum
of the resulting MPA amide is recorded in MeCN-d3 (i.e., 5 mg of
amide in 0.5 mL of MeCN-d3), and the signals of L1 and L2 are
assigned. Solid anhydrous Ba(ClO4)2 is added to the NMR tube
until saturation is attained (about 200 mg per tube), and a new
spectrum is recorded. The signals are assigned, and the ∆δBa values
and signs are calculated. The models shown in Figure 15 are used
to place L1 and L2 in space in accordance with the signs of ∆δBa.
CDCl3) δ (ppm) 0.91 (d, J ) 9.74 Hz, 1H), 1.02 (s, 3H), 1.15 (d,
J ) 7.22 Hz, 3H), 1.48 (ddd, J ) 2.19, 6.28, 13.97 Hz, 1H), 1.78-
1.90 (m, 2H), 1.93 (m, 1H), 2.41 (m, 1H), 3.37 (s, 3H), 4.26 (m,
1H), 4.61 (s, 1H), 6.62 (br, J ) 9.11 Hz, 1H), 7.27-7.42 (m, 5H);
13C NMR (62.9 MHz, CDCl3) δ (ppm) 21.7, 23.7, 28.4, 37.4, 35.6,
38.8, 41.4, 46.6, 47.6, 48.1, 57.6, 84.1, 127.3, 128.7, 128.9, 137.7,
170.0; MS (EI) m/z % 301 (M+).
Acknowledgment. We thank the Ministerio de Educacio´n
y Ciencia and the Xunta de Galicia for financial support (BQU-
2002-01195, SAF2003-08765-C03-01, PGIDT02BTF20902PR,
PGIDT03PXIC20908PN, PGIDIT04PXIC20903PN) and the
Centro de Supercomputacio´n de Galicia (CESGA) for their
assistance with the computational work. We are also grateful
to Yamakawa Chemical Industry Co. Ltd. (Japan) for their gift
of MPA.
1
NMR Spectroscopy. H and 13C NMR spectra of samples in
CDCl3 were recorded at 500 and 250 MHz. Chemical shifts (ppm)
are internally referenced to the TMS signal (0 ppm) in all cases. J
values are recorded in hertz.
1D 1H NMR Spectra. Size, 32 K; pulse length, 2.8 µs (30°); 16
acquisitions.
(34) (a) Cheeseman, J. R.; Trucks, G. W.; Keith, T. A.; Frisch, M. J. J.
Chem. Phys. 1996, 104, 5497-5509. (b) Frisch, M. J.; Trucks, G. W.;
Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Zakrzewski,
V. G.; Montgomery, J. A., Jr.; Stratmann, R. E.; Burant, J. C.; Dapprich,
S.; Millam, J. M.; Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.;
Tomasi, J.; Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.;
Adamo, C.; Clifford, S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui,
Q.; Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J. B.; Cioslowski, J.; Ortiz, J. V.; Stefanov, B. B.; Liu, G.;
Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R. L.;
Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.;
Gonzalez, C.; Challacombe, M.; Gill, P. M. W.; Johnson, B. G.; Chen, W.;
Wong, M. W.; Andres, J. L.; Head-Gordon, M.; Replogle, E. S.; Pople, J.
A. Gaussian 98, revision A.7; Gaussian, Inc.: Pittsburgh, PA, 1998.
Supporting Information Available: Experimental section
[computational methods (level of theory, specific program, basis
set, Cartesian coordinates, and computed total energies of optimized
structures), general procedures, general procedure for configuration
assignment, NMR spectroscopy, spectroscopic data for compounds
(S)-4, (R) and (S)-5, (R)- and (S)-12, (R)- and (S)-13, (R)- and (S)-
14, (R)- and (S)-18, (R)- and (S)-19, (R)- and (S)-20, (R)- and (S)-
21, (R)- and (S)-22, (R)- and (S)-24)]. Figures 1-SI and 2-SI. This
material is available free of charge via the Internet at http://
pubs.acs.org.
JO0522207
1130 J. Org. Chem., Vol. 71, No. 3, 2006