Study of syntheses and specific rotations of (S)-3-phenylhexan-3-ol and its derivatives

Article abstract of DOI:10.1016/j.tetasy.2008.02.030

Tertiary alcohols, 3-phenylhexan-3-ol and 3-methylhexane-3-ol, and their derivatives were synthesized. The reaction conditions of the esterification of the tertiary alcohol with 2-NO₂PhCO₂Cl and 4-NO₂PhCO₂Cl were optimized. The absolute configuration of the derivative from (S)-3-phenylhexan-3-ol was identified by X-ray study and computational methods. Experimental results confirmed the computational specific rotation predictions by DFT-based and matrix methods.

Full text of DOI:10.1016/j.tetasy.2008.02.030

Available online at www.sciencedirect.com
Tetrahedron: Asymmetry 19 (2008) 808–815
Study of syntheses and specific rotations of
(S)-3-phenylhexan-3-ol and its derivatives
Tou-Gen Liao,a Jie Ren,a Hua-Fang Fan,a Ming-Jin Xieb and Hua-Jie Zhua,*
aOrganic Synthesis and Natural Product Laboratory, State Key Laboratory of Phytochemistry and Plant Resources in West China,
Kunming Institute of Botany, Chinese Academic of Sciences (CAS), Kunming, 650204 Yunnan, PR China
bDepartment of Chemistry, Yunnan University, Kunming, 650092 Yunnan, PR China
Received 10 January 2008; accepted 26 February 2008
Available online 15 April 2008
Abstract—Tertiary alcohols, 3-phenylhexan-3-ol and 3-methylhexane-3-ol, and their derivatives were synthesized. The reaction condi-
tions of the esterification of the tertiary alcohol with 2-NO2PhCO2Cl and 4-NO2PhCO2Cl were optimized. The absolute configuration
of the derivative from (S)-3-phenylhexan-3-ol was identified by X-ray study and computational methods. Experimental results confirmed
the computational specific rotation predictions by DFT-based and matrix methods.
Ó 2008 Elsevier Ltd. All rights reserved.
1. Introduction
ylhexan-3-ol, derivatives and specific rotation computa-
tions are reported herein.
It is well known that specific rotation calculations for rigid
chiral compounds can be obtained using quantum mechan-
ics.1–6 The computations of specific rotation for flexible,
especially, linear stereogenic compounds are also re-
ported.7,8 However, due to the difficulty in searching for
and computing most of the stable conformations in which
the molecule can exist during specific rotation calculations
of linear chiral compounds, successful examples are few
when compared to those in rigid compounds. Recently,
we reported a matrix method to predict the acyclic chiral
secondary alcohols, amines and others.9 This method can
use less computation time to predict the specific rotation
by use of the determinant value det(D) and k0 values in a
series of chiral compounds. Since the matrix method is still
in its infancy, it needs a more standard chiral molecule for
specific rotation computation tests. Chiral linear tertiary
alcohols have rarely been tested before; we synthesized
tertiary alcohols such as 3-phenylhexan-3-ol and its
derivatives. By enantiomer separations, (S)- and (R)-3-
phenylhexan-3-ols were obtained with almost 100% enan-
tiomeric purity. X-ray experiments confirmed the absolute
configuration of (S)-3-phenylhexan-3-ol by converting (S)-
3-phenylhexan-3-ol into an amide derivative, which had a
standard (S)-configuration. The syntheses of (S)-3-phen-
2. Results and discussion
2.1. Syntheses and identifications of (S)-3-phenylhexan-3-ol
and its derivatives
Synthetic routes to (S)-3-phenylhexan-3-ol and its deriva-
tives are illustrated below.10–13 The esterification of tertiary
alcohol is generally very difficult due to the big repulsive ef-
fect among the groups. For example, if the tertiary alcohol
was obtained after the addition of PhMgBr to the ketone 3,
and this tertiary alcohol was then used to esterify with 4-
NO2–PhCOCl, then the yield of ester 4 would be very
low or no ester product would be obtained. The reaction
products were also too many to be separated. After a series
of experiments using different solvents and temperatures,
continuing a reaction conditions were found to be efficient
for the conversion of 3 to 4. In this esterification, the chlo-
ride must be added at the low temperature, ꢀ76 °C, after
the addition of PhMgBr to ketone 3 and then kept for over
4 h. Resolution of the enantiomers, (R)- and (S)-4, was
tried using Chiralcel OD-H and Chiralcel OB columns.
Only Chiralcel OD-H was able to separate the enantio-
mers. The effect of the substituents on the benzoate on
the chiral separation is very big. For example, if the
–NO2 was not reduced to –NH2, no column could separate
*
Corresponding author. Tel.: +86 871 5223242; fax: +86 871 5216179;
e-mail: hjzhu@mail.kib.ac.cn
0957-4166/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2008.02.030

T.-G. Liao et al. / Tetrahedron: Asymmetry 19 (2008) 808–815
809
the enantiomers. The position of the –NO2 or –NH2 also
had the effect on chiral separation. Esterification of 2-
NO2–PhCOCl with the alcohol afforded the 2-NO2 ana-
logue ester. However, these two enantio-2-NO2 analogue
esters, or their reduced enantio-amines, could not be sepa-
rated by either Chiralcel OD-H or Chiralcel OB.
the carbon C3A that connected with an aromatic ester
could be directly configured. The X-ray experiment con-
firmed that carbon (C3A) had an (S)-configuration (Fig. 1).
Not every chiral compound could be separated in this way.
Another tertiary alcohol, 3-methylhexan-3-ol, was synthe-
sized using a similar route in Scheme 1, and this tertiary
alcohol was then converted into the corresponding esters
and others. The structures of compounds 8 to 13 are illus-
trated below. Enantiomers 8a and 8b could not be sepa-
rated by both chiral columns. The enantiomers of 9, 9a
and 9b could be separated using Chiralcel OD-H. How-
ever, enantiomers 9a and 9b decomposed to other complex
mixtures during the purification. To obtain more stable iso-
mers, compound 9 was converted into 10. Unfortunately,
10a and 10b could not be separated. Thus, 3-methylhex-
an-3-ol was then converted into 11–13. However, none of
these could be separated by Chiralcel OD-H or OB-H col-
umns. The trial was attempted to use L-tartaric acid to re-
act with one enantiomer of amines 9 or 12 to form the
corresponding crystal, respectively. Unfortunately, no crys-
tal was formed using different solvents, such as alcohols,
ethyl acetate, diether or their mixtures.
Enantiomer (S)-4 was hydrolyzed to the chiral tertiary
alcohol 5. The specific rotation of 5 was +11.6 deg/
[g/mL] (c 0.00805, >99% ee HPLC purity) determined in
chloroform at room temperature.
Control of the pH was required during the reduction of
(S)-4 to (S)-6. The hydrolysis of the ester was fast at
pH < 5.0 whereas the –NO2 was reduced to –NH2. No
amount of amine (S)-6 was formed after the reduction.
Once the pH value was over 7, the reduction reaction
was not finished and even the reaction time was increased
to two days. Thus, a suitable pH value for the reduction
was 6.
Amine (S)-6 was then reacted with (S)-2-(tert-butoxycar-
bonylamino)-4-methylpentanoic acid to afford amide 7
O
NO2
NH2
HN
NHBoc
O
O
O
O
O
O
3
1
6
2
8a: (3R)-
8b: (3S)-
9a: (3R)-
9b: (3S)-
10a: (3R)-
10b: (3S)-
O
O
O
NHBoc
H
N
NO2
NH2
O
O
O
O
11a: (3R)-
11b: (3S)-
13a: (3R)-
13b: (3S)-
12a: (3R)-
12b: (3S)-
with a dual-chiral-center. The condensation reagent,
CDMT or NMM, was normally required in the reaction,
and most of the time, either of them was enough to remove
the water formed in the condensation reaction. However,
in this condensation reaction, one mole of CDMT and
three moles of NMM were necessary for one mole of 6.
When either CDMT or NMM was used in the condensa-
tion, it afforded a very low yield of amide 7 (less than
10% yield).
Most chiral compounds have specific rotation values that
can be easily obtained. Therefore, it is useful to use a spe-
cific rotation value to determine the absolute configuration.
It is very important for researchers to use this method in
their practices, and this has led to the use of theoretical
methods.
2.2. QM model
The first example of using Hartree–Fork theory to calcu-
late the specific rotation was reported by Polavarapu in
1997.2a,b,e Stephens et al. studied the DFT-based specific
rotation computations;2c,d,3d,4b,c,5a–c Ruud,3a–c,4b,6c,7b,c
Crawford,1a,b,e Pederson3e,f and Koch1b,3e,f reported cou-
pled-cluster (CC) methods; while Wiberg et al.,4a,8a Coriani
et al.7b,c reported many valuable data for specific rotation
calculations. Over the past decade, Wipf et al.,1c,d,8b Helga-
ker,3a,c Grimme,7a Giorgio4d and other chemists6b,c
explored several computational approaches for use in spe-
cific rotation calculations. Studies related to conformation-
Solvents for crystal formation were optimized using meth-
anol, acetone, ethyl acetate, n-hexane and their mixtures
with different combinations and different ratios. The suit-
able solvents were a mixture of n-hexane/ethyl acetate
(2:1, v/v). The re-crystallization temperature was 3–5 °C
in a refrigerator. The crystals were then selected for X-
ray experiments, the X-ray structure is illustrated below.
Since the absolute configuration at C2B (in the chiral ami-
no acid section) was already known as (S), which meant
that once the whole molecule was pictured by X-ray, then,

810
T.-G. Liao et al. / Tetrahedron: Asymmetry 19 (2008) 808–815
Figure 1. The X-ray structure for (S)-3-phenylhexan-3-yl 4-((S)-2-(tert-butoxycarbonylamino)-4-methylpentanamido) benzoate, 7.
(1) PhMgBr / THF, -40 oC
(2) 4-NO2-PhCOCl, -76 oC
Et
O
OH
Et
O
EtMgBr
THF
PCC
CH2Cl2
n-Pr
H
n-Pr
n-Pr
1
2
3
O
O
Enantiomer separation
Chiralcel OD-H
Et
Ph
O
Et
O
n-Pr
NO2
Ph
NO2
n-Pr
(R)-4
4
+
O
Et
Ph
OH
n-Pr
Et
Ph
O
MeOH/NaOH / reflux, 2 h.
(S)-5
n-Pr
NO2
[α]D = +11.6 deg
(c 0.00805, CHCl3)
(S)-4
MeOH/Zn
powder HCl
pH = 6
O
CDMT/NMM,
THF, r.m.
O
O
Et
Ph
O
Et
Ph
O
n-Pr
i-Pr
NHBoc
N
O
H
n-Pr
NH2
OH
NHBoc
i-Pr
7
6
Scheme 1. The synthetic routes to 3-phenylhexan-3-ol and its derivatives.
ally flexible molecular specific rotation were also reporte-
d.4a,7a,b,8a,b Quantum mechanical theory relates specific
rotation to the following formula:
b(m) at the gas-phase equilibrium geometry, in which b(m)
is the frequency-dependent electric dipole–magnetic dipole
polarizability of molecule. Thus, the value of [a]v depends
upon the magnitude of the tensor of [b(m)]0.
28800p2NAm2
½aꢁv ¼
cs;v½bðmÞꢁ0
ð1Þ
c2M
For a chiral molecule with one or more stable conforma-
tions with close energy, all the conformational specific
rotation values should be calculated. Finally, these magni-
tudes can be summed up by using the Boltzmann formula.
The calculated values are the absolute specific rotation
value for this chiral molecule.
where NA is Avogadro’s number, M is the molecular
weight, c is the light speed in vacuum, c is the correction
from solvent which is either neglected (c = 1) or approxi-
mated by equation: c = (n2 + 2)/3. [b(m)]0 is the value of

T.-G. Liao et al. / Tetrahedron: Asymmetry 19 (2008) 808–815
811
2.3. Matrix model
minant of the matrix as det(D), which is proportional to
the specific rotation values, and this is the relative specific
rotation value. Also, the sodium D line is used to obtain
the specific rotation. As a result, one conformation has
In our recently reported matrix model, which described
four substituents surrounding a stereogenic center,9 four
substituents’ variables, comprehensive mass (m), radius
(r), electronegativity (v) and symmetry (s) were used in a
matrix F function. If an atom is directly connected to the
stereogenic center, its coefficient is b1. Coefficients of the
other atoms, which are further removed, are b2, b3, and
so on. Thus, the comprehensive mass of the substituent
group becomes
½aꢁDi ¼ k ꢃ a1 ꢃ a2 ꢃ a3 ꢃ a4 ꢃ detðDiÞ ¼ k0 ꢃ detðDiÞ ð4Þ
where
ꢀ
ꢀ
ꢀ
ꢀ
m1 r1 v1 s1
m2 r2 v2 s2
m3 r3 v3 s3
m4 r4 v4 s4
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
detðD Þ ¼
ð5Þ
i
ꢀ
ꢀ
ꢀ
ꢀ
m ¼ b1m1 þ Rb2m2 þ Rb3m3 þ Rb4m4 þ ꢂ ꢂ ꢂ
The different conformations of the chiral molecule have
different energies. Thus, the final observed specific rotation
must employ the Boltzmann distribution of all these
conformations.
Here m1 is the mass of atom 1, which is directly connected
to the stereogenic center, and m2 is the mass of atom 2,
which is directly bound to atom 1. Since more than one
atom can be bonded to atom 1, the summation term is used
to indicate that the contributions of all these atoms must be
included. If one substituent has the highest symmetry oper-
ation number N, and this symmetric axis passes through
the atom which is connected to the stereogenic atom, then
the symmetry factor for that substituent is s (s = [(N ꢀ 1)/
N]2). It was proposed that the determinant value (det(D)) is
proportional to the specific rotation magnitudes for this
chiral molecule. The matrix method gave a relative value
for chiral molecule which requires us to use both det(D)
and k0 values to characterize the prediction. This matrix
method was also extended to a dual-chirality molecular
specific rotation computation.10
detðDÞ ¼ RðdetðDiÞÞðQi=RQiÞ
½aꢁD ¼ k0 ꢃ detðDÞ
ð6Þ
As mentioned above, it is impossible to obtain a1, a2, a3,
and a4 and k values at this stage. Therefore, there is no
absolute value to [a]. However, it is useful to obtain the rel-
ative values by computing det(D) since det(D) is the initial
characteristic of that molecule and is proportional to the
specific rotation values when the outside factors are held
in a constant. It needs to use both det(D) and k0 values
to characterize the specific rotation prediction using Eq.
(6). In our previous report,9 the effect of the different con-
formations on the specific rotation in the QM method was
converted into the effect of different radius values on the
det(D) magnitude in a matrix model. In a series of chiral
compounds, the recorded specific rotation values were pro-
portional to their det(D) values.
The effects of temperature, solvent, and frequency of light
on a specific conformational specific rotation are the func-
tions f(t), f(s), and f(v). When these factors were held con-
stant in a specific rotation determination, the overall
contributions to the specific rotations for the ith conforma-
Our previously reported quantum method7a was used for
specific rotation calculations first via a GAUSSIAN 03 pack-
age.14 Alternatively, (S)-3-phenylhexan-3-ol was selected
for conformational searches, optimizations and specific
rotation computations before experiments. In total, 119
conformations of (S)-3-phenylhexan-3-ol within the energy
window of 0–5 kcal/mol were found from over 400 stable
conformations, which were obtained using an AM1 force
field via HYPERCHEM package, when the lowest conforma-
tional energy was used as the reference zero-point. The
selected 119 conformations were then selected for optimiza-
tions at the B3LYP/3-21G* level. There were 22 low energy
conformations (energies from 0 to 3 kcal/mol) found from
these 119 B3LYP/3-21G(d)-optimized geometries, and they
were selected for further optimization at the B3LYP/6-
31G(d) level. Finally, a total of 11 conformations, whose
energies were in the window of 0–2.0 kcal/mol, were used
in specific rotation calculations at the B3LYP/aug-cc-
pVDZ and B3PW91/aug-cc-pVDZ levels, respectively.
The 11 conformations were further optimized at the
MP2/6-31+G* level, and these 11 MP2/6-31+G*-optimized
conformations were used in specific rotation computations
at the B3LYP/aug-cc-pVDZ and B3PW91/aug-cc-pVDZ
levels, respectively, again. The energetics after zero-point
energy (ZPE) correction, free energy and totally elec-
tronic energy were used in the Boltzmann sums for whole
tion could be represented by function Fconi
.
2
32
3
m1 r1 v1 s1
a1
6
6
6
6
4
76
76
76
76
54
7
7
7
7
5
m
m
r2 v2 s2
r3 v3 s3
a
2
3
2
Fconi ¼ k
ð2Þ
a
3
m4 r4 v4 s4
a4
where k is the sum of all the constants (e.g., k1 + k2 + k3
ꢂ ꢂ ꢂ). k1 = f(m), k2 = f(s), k3 = f(t),ꢂ ꢂ ꢂ
The total contribution F from all the conformations can be
written in the form of the Boltzmann distribution
F ¼ RðFconiÞðQi=RQiÞ
ð3Þ
where Qi = kexp (ꢀDGi/RT), Qi is the amount of the ith
conformation, k and S are constants, DGi is the difference
between ith conformation’s free energy and the lowest con-
formational free energy.
The coefficients, a1, a2, a3, and a4 were unknown and the
specific functions of f(t), f(s), and f(v) were also unknown.
However, they are constant when light frequency and tem-
perature are fixed. Thus, we can study the middle matrix.
Since a matrix is not a scalar number, we defined the deter-

812
T.-G. Liao et al. / Tetrahedron: Asymmetry 19 (2008) 808–815
Table 1. The calculated specific rotation for (S)-3-phenylhexan-3-ol using
Table 2. The calculated specific rotation for (S)-3-phenylhexan-3-ol using
four methods
full optimized geometries in chloroform via the PCM model
a
b
c
a
0 b
c
[a]E
½aꢁE0
[a]G
½aꢁCHCl
[a]E
½aꢁE
[a]G
3
Method A
Method B
Method C
Method D
+2.8
ꢀ1.2
+2.7
ꢀ1.3
+3.2
ꢀ13.0
+3.1
ꢀ12.9
+4.4
ꢀ12.3
+1.6
ꢀ12.2
+2.9
ꢀ1.0
+2.8
ꢀ1.2
Method E
Method F
Method G
Method H
+2.2
ꢀ9.3
+2.0
ꢀ9.5
+3.1
ꢀ8.5
+2.7
ꢀ8.4
+2.3
ꢀ6.2
+0.7
ꢀ7.3
a The total electronic energy data at the B3LYP/aug-cc-pVDZ level were
used in the specific rotation computations in methods A and B, and at
the B3PW91/aug-cc-pVDZ level in methods C and D.
a The total electronic energy data at the B3LYP/aug-cc-pVDZ were used
in methods E and F in the specific rotation computations, and those at
the B3PW91/aug-cc-pVDZ level were used in methods G and H.
b The total electronic energy data at the MP2/6-31+G(d) level used.
c The total free energy in solution at the MP2/6-31+G(d) level were used.
b Energetics after the zero-point energy correction at the B3LYP/6-31G(d)
level in methods A and C, or at the MP2/6-31+G(d) level in methods B
and D, were used.
c The free energy data at the B3LYP/6-31G(d) level in methods A and C or
at the MP2/6-31+G(d) level in methods B and D were used.
respectively. Four methods E to H were applied to the spe-
cific rotation computations: method E, B3LYP/aug-cc-
pVDZ//B3LYP/6-31G(d)-CHCl3; method F, B3LYP/aug-
cc-pVDZ//MP2/6-31+G(d)-CHCl3; method G, B3LYP/
aug-cc-pVDZ//B3LYP/6-31G(d)-CHCl3; and method H,
B3LYP/aug-cc-pVDZ//B3LYP/6-31G(d)-CHCl3. The re-
sults are listed in Table 2. The calculated specific rotation
magnitudes obtained at the B3LYP/aug-cc-pVDZ level
were about 3 using the fully B3LYP/6-31G(d)-optimized
geometries in chloroform via PCM model (methods E
and G, Table 2). The specific rotation values became about
ꢀ5.7 to ꢀ9.0 when the full MP2/6-31+G(d)-optimized
structures in chloroform were used in specific rotation com-
putations (methods F and H, Table 2).
molecular specific rotation calculations. Here, method A
means that the B3LYP/6-31G(d)-optimized geometries in
the gas phase were used in specific rotation calculations
(B3LYP/aug-cc-pVDZ//B3LYP/6-31G(d)), method
B
means that the MP2/6-31+G(d)-optimized conformations
were used in specific rotation computations (B3LYP/
aug-cc-pVDZ//MP2/6-31+G(d)). In methods C and D,
B3PW91/aug-cc-pVDZ//B3LYP/6-31G(d) and B3PW91/
aug-cc-pVDZ//MP2/6-31+G(d) were used in the specific
rotation predictions. (S)-3-Phenylhexan-3-ol was predicted
to have +1.3 to 1.9 values via methods A and C (Table 1),
or ꢀ2.3 to ꢀ13.0 by methods B and D. Indeed, chiral
tertiary alcohol 5 has an (S)-configuration with a value of
+11.6. DFT-optimized geometries provided a more correct
prediction of (S)-5 specific rotation (methods A and C)
instead of MP2-optimized geometries (methods B and D).
This is unexpected. The reason might be that the solvent
effect on the specific rotation values was not investigated
in the specific rotation calculations.
The DFT theory (methods A and E) predicted the specific
rotation sign direction correctly although the absolute spe-
cific rotation values were a little lower than the recorded
magnitudes. MP2 theory (methods B and F) predicted
the wrong specific rotation direction. However, the abso-
lute values approached the experimental one. Indeed, it is
necessary to use the sign of the specific rotation magnitudes
to judge the absolute configuration. Thus, in this example,
although the computational times in methods B and F were
much longer than those in methods A and E, the correct
prediction of the configuration is obtained using methods
A and E. This is an unexpected result. However, this does
not mean that the high computational method is less valu-
able. Indeed, the selection of the computational method
and the basis set in calculation could have a great influence
on the results of the specific rotations.7b This is similar to
what Crawford reported that ‘in spite of these advances,
it is not yet understood what level of theory is necessary
to obtain ‘the right answer for the right reason’ for specific
rotation, and many successes rely implicitly on fortuitous
cancellation of errors (e.g., limited basis sets, lack of expli-
cit solvation, and vibrational averaging).1a It is possible to
find a suitable QM method to give good predictions for the
linear chiral molecular specific rotation values, including
compound 5 in this research, although this is time-
consuming.
The effect of solvent could have a big effect on the specific
rotation values or even change the sign of specific rotation.
Thus, the effect of chloroform on specific rotation (½aꢁCHCl
)
3
was investigated. Single point energy (SPE) calculations for
both B3LYP/6-31G(d)- and MP2/6-31+G(d)-optimized
geometries were first performed at the B3LYP/6-
311+G(d,p) level in chloroform using PCM model (Table
1). The SPE was used to correct the specific rotation values
in solvent chloroform (d,p) level in chloroform using PCM
model (Table 1). However, it did not change the specific
rotation direction (the sign of specific rotation) in this way.
The effect of the geometries had a big effect on the specific
rotation values, and the effect of solvent on the geometries
was also large as reported in Coriani’s study.7c The geom-
etries used in methods A and B were obtained in the gas
phase, the structures of the different conformations could
be different from those in solution. This could be the reason
that the computed specific rotations using methods A to D
did not give the expected predictions. Thus, optimizations
of the conformations in chloroform were performed at
the B3LYP/6-31G(d) and MP2/6-31+G(d) levels, respec-
tively, using a PCM model. The geometries obtained were
then used for specific rotation computations again at the
B3LYP/aug-cc-pVDZ and B3PW91/aug-cc-pVDZ levels,
It would be time-consuming to compute the specific rota-
tions for linear chiral molecules if to use the above methods
A to H, since the numbers of stable conformations with
low energy were huge. Once the stable conformation num-
bers were over a hundred obtained at the B3LYP/6-
31G(d), the specific rotation prediction for a linear chiral

T.-G. Liao et al. / Tetrahedron: Asymmetry 19 (2008) 808–815
813
molecule would be expensive. For example, compounds 4
or 7 in Scheme 1 were very flexible. The number of stable
conformations, which had an energy of 0–5 kcal/mol,
and needed to be re-optimized at the B3LYP/3-21G(d)
level were over 2000 and 8000, respectively, after the con-
formational search using AM1 force field via the HYPER-
CHEM package. The stable conformations needed to be
optimized at the B3LYP/6-31G* level would be over hun-
dreds and thousands, respectively, for 4 and 7. Thus, to
use any one of methods A to H to predict the specific rota-
tions of 4 and 7 would be very time-consuming. If a com-
puter’s computational rate is fast enough, this is still a good
choice to calculate specific rotations for linear chiral mole-
cules as 4 and 7. However, if there is another method to do
the predictions of specific rotation for linear chiral mole-
cules, such as 4 and 7, it would prove to be helpful. The
matrix method provides another choice to do the predic-
tions of specific rotation for 4 and 7.9 Matrix model is de-
duced from mathematical theory, and is different from the
quantum theory. The parameters used at the present time
were calculated from the standard molecular structures.
For example, the bond length of C–C was regarded as
cific rotation is ꢀ33.3 (c 0.01485, CHCl3). Its experimental
k0,7 is ꢀ15.1.
3. Conclusion
The quantum methods to predict the specific rotation val-
ues for acyclic chiral compounds proved to be difficult since
the number of stable conformations is very large. The long-
er the carbon chain in the acyclic chiral compound, is the
more the numbers of the stable conformations would be.
The specific rotation computations for these large acyclic
compounds would be time-consuming although the predic-
tions match the experimental results. Relatively, matrix
method could provide an additional method in the specific
rotation predictions for these acyclic chiral compounds.
This is another mathematic method to predict specific rota-
tion values for tertiary alcohol derivatives; these derivatives
require one of the four substituents which are connected to
the stereogenic center (one of the s values in det(D) is not
equal to zero).
3
˚
1.54 A, the angle sp hybridized carbon was 108°, and so
on. These parameters were used in the specific rotation pre-
dictions for the reported 94 chiral acyclic compounds, and
it worked well via the combination with k0.9 The same ser-
ies of chiral compounds had similar constant k0 values as
mentioned in the matrix model section.
4. Experimental
A Brucker-AV-400 instrument was used for NMR determi-
nations using CDCl3 as the solvent unless another solvent
is specifically stated. The chemicals were used as received.
A Waters model 2695 high pressure liquid chromatography
instrument was used for HPLC analyses and chiral separa-
tions. Thermometers used in the reaction temperature
determinations were not corrected. Silica gel (200–400
mesh) was used. Chiralcel-OD-H column was used in the
analysis and separation for compound 4. The procedures
for the formations of compounds 4 to 7 are described
below.
The matrix method predicted the specific rotation by using
the det(D) and k0 values. In a similar series of chiral com-
pounds, the k0 values should be constant. By computing
the det(D) for the target chiral molecule and to use the k0
value of target molecular analogue, the specific rotation
could be predicted for the target chiral molecule. Thus, ma-
trix method was used to calculate the k0 for (S)-3-phenyl-
hexan-3-ol, 5. The magnitudes of m, r, v, and s for
substituents in 5 were selected directly from our recent
9
report. The determinant det(D) was ꢀ0.42. Thus, the k0
4.1. 3-Phenylhexan-3-yl-4-nitrobenzoate, 4
value was ꢀ27.6 (k0,5 = 11.6/(ꢀ0.42)). The specific rotation
value could be measured in different solvents, various k0 for
different solvents could be obtained.
A 250 mL three-necked round-bottomed flask equipped
with a magnetic stirbar and reflux condenser was flushed
with nitrogen, and charged with 50 mL anhydrous THF
and 800 mg (8 mmol) of 3. A solution of phenylmagnesium
bromide (2.7 mL, 3 M, 8.1 mmol) was added dropwise over
a period of 0.5 h at ꢀ30 °C, the solution was then warmed
to 50 °C for 10 min. The reaction mixture was cooled down
to ꢀ78 °C, and a solution of 4-NO2–PhCOCl (8 mmol in
5 mL THF) was added dropwise into the flask over a per-
iod of 10 min and kept for 20 min. After being stirred over-
night at room temperature, the solution was treated with
2 mL water and extracted with EtOAc (6 ꢃ 15 mL). The
organic layer was dried over anhydrous Na2SO4, concen-
trated, and purified by column chromatography (1:20
EtOAc/hexanes) to afford 4 as a yellow liquid (1.3 g, 50%
yield), which solidified slowly to give a pale yellow product.
Chiral separation of 4 was carried out by HPLC using Chi-
ralcel-OD-H column. 1H NMR (400 MHz, CDCl3) d 0.77–
0.80 (m, 3H), 0.87–0.91(m, 3H), 1.17–1.21 (m, 2H), 2.24–
2.34 (m, 2H), 2.48–2.60 (m, 2H), 7.28–7.41 (m, 5H), 8.27
(dd, J = 6.8 Hz, 2.0 Hz, 2H), 8.34 (dd, J = 6.8 Hz,
2.0 Hz, 2H); 13C NMR (100 MHz, CDCl3), d 7.7, 14.1,
As mentioned above, it is possible to use this k0,5 to predict
the specific rotation values of chiral 4. The calculated deter-
minant value for chiral compound 4 is ꢀ0.07. This value of
ꢀ0.07 predicted that its specific rotation value should be
+1.9 in chloroform using k0,5 (ꢀ27.6) value via Eq. (5)
since 4 is the tertiary ester derived from 5. The determined
specific rotation for chiral compound 4 is +1.6 (c 0.026,
CHCl3). As can be seen, this value is close to the prediction
of 1.9. Its experimental k0 is ꢀ22.9 (k0,4).
The specific rotation for 7 was predicted using this method
too. Compound 7 has two kinds of stereogenic centers, one
center has a proton, whose det(D) value was +0.05; an-
other has not proton and its det(D) magnitude was
+2.16. The sum of the two det(D) values was 2.21. The size
of 7 is more close to that of 4 instead of 5. Thus, the k0,4
(ꢀ22.9) was used for the prediction of a specific rotation
of 7. The calculated specific rotation for 7 was ꢀ50.6
(2.21 ꢃ (ꢀ22.9) = ꢀ50.6) in chloroform. The recorded spe-

814
T.-G. Liao et al. / Tetrahedron: Asymmetry 19 (2008) 808–815
16.6, 30.5, 39.6, 89.8, 123.5, 125.0, 127.0, 128.2, 130.5,
136.8, 143.0, 150.4, 162.9.
(100 MHz, CDCl3) d 7.6, 14.1, 16.5, 22.9, 24.6, 28.8, 30.6,
40.0, 40.7, 53.8, 80.5, 88.0, 118.8, 125.0, 126.5, 126.6,
127.9, 130.5, 142.0, 143.7, 156.6, 164.3, 171.8. CCDC
No. 680024.
4.2. (R)-3-Pheny-3-hexanol, 5
To an aqueous solution of KOH (220 mg, 4 mmol, 7 mL)
was added the solution of methanol (3 mL) and lithium
chloride (168 mg, 4 mmol), and then 4a was added
(53 mg, 0.16 mmol). The mixture was refluxed for 2 h and
diluted with water. The solution was extracted with diethyl
ether (6 ꢃ 25 mL). The combined organic phase was
washed with brine (3 ꢃ 5 mL), dried over Na2SO4 and fil-
tered. The solvents were removed under reduced pressure.
The residue was purified by column chromatography
(1:20 EtOAc/hexanes) to give 5 as a colorless oil
Acknowledgments
H.J.Z. thanks the financial support from NSFC
(30770235), CAS (YZ-06–01) and the Science and Technol-
ogy Committee of Yunnan Province (2005B0048M). The
Super-Computer Centers of CAS in Beijing and of Yunnan
University are thanked for computation supports. H.J.Z.
thanks the comments from the referees and C. U. Pittman,
Jr. in the department of Chemistry of Mississippi State
University for partial English editing.
1
(21.5 mg, 75% yield). H NMR (CDCl3), d 0.73–0.76 (m,
3H), 0.83–0.86 (m, 3H), 1.01–1.11 (m, 1H), 1.28–1.30 (m,
1H), 1.75–1.89 (m, 4H), 7.19–7.37 (m, 5H); 13C NMR
(CDCl3), d 7.7, 14.4, 16.7, 35.2, 44.8, 77.2, 125.2, 126.1,
127.9, 146.0.
References
1. (a) Crawford, T. D. Theor. Chem. Acc. 2006, 115, 227–245;
(b) Crawford, T. D.; Owens, L. S.; Tam, M. C.; Schreiner, P.
R.; Koch, H. J. Am. Chem. Soc. 2005, 127, 1368–1369; (c)
Kondru, R. K.; Wipf, P.; Beratan, D. N. J. Am. Chem. Soc.
1998, 120, 2204–2205; (d) Goldsmith, M. R.; Jayasuriya, N.;
Beratan, D. N.; Wipf, P. J. Am. Chem. Soc. 2003, 125, 15696–
15697; (e) Tam, M. C.; Russ, N. J.; Crawford, T. D. J. Chem.
Phys. 2004, 121, 3550–3557.
4.3. 3-Phenylhexan-3-yl-4-aminobenzoate, 6
To a suspension of Zn powder (800 mg) in a mixture of
methanol (20 mL) and water (2 mL) was added 4a
(200 mg, 0.6 mmol), then 3 M HCl was added dropwise
into the mixture and the pH value was carefully controlled
at 6.0. The suspension was stirred at ambient temperature
for about 50 h until TLC analysis indicated the complete
conversion of 4a to 6. The mixture was filtered and the
solution was washed with 10 mL brine and dried over
Na2SO4 (anhydrous). Evaporation of the solvent afforded
6 (160 mg, 90% yield) as a yellow oil without any further
2. (a) Polavarapu, P. L. Mol. Phys. 1997, 91, 551–554; (b)
Polavarapu, P. L. Chirality 2006, 18, 348–356; (c) Cheeseman,
J. R.; Frisch, M. J.; Delvin, F. J.; Stephens, P. J. J. Phys. Chem.
A 2000, 104, 1039–1046; (d) McCann, D. M.; Stephens, P. J.;
Cheeseman, J. R. J. Org. Chem. 2004, 69, 8709–8717; (e)
Polavarapu, P. L. Tetrahedron: Asymmetry 1997, 8, 3397–3410.
3. (a) Pecul, M.; Ruud, K.; Rizzo, A.; Helgaker, T. J. Phys.
Chem. A 2004, 108, 4269–4276; (b) Ruud, K.; Zanasi, R.
Angew. Chem., Int. Ed. 2005, 44, 3594–3596; (c) Helgaker, K.;
Ruud, T. Chem. Phys. Lett. 2002, 352, 533–539; (d) Stephens,
P. J.; Devlin, F. J.; Cheeseman, J. R.; Frisch, M. J.; Rosini, C.
Org. Lett. 2002, 4, 4595–4598; (e) Pedersen, T. B.; Koch, H.;
Boman, L.; Sanchez de Meras, A. M. J. Chem. Phys. Lett.
2004, 393, 319–326; (f) Pedersen, T. B.; Sanchez de Meras, A.
M. J.; Koch, H. J. Chem. Phys. 2004, 120, 8887–8897.
4. (a) Wiberg, K. B.; Wang, Y. G.; Vaccaro, P. H.; Cheeseman,
J. R.; Luderer, M. R. J. Phys. Chem. A 2005, 109, 3405–3410;
(b) Ruud, K.; Stephens, P. J.; Devlin, F. J.; Taylor, P. R.;
Cheeseman, J. R.; Frisch, M. J. Chem. Phys. Lett. 2003, 373,
606–614; (c) Stephens, P. J.; Devlin, F. J.; Cheeseman, J. R.;
Frisch, M. J.; Mennucci, B.; Tomasi, J. Tetrahedron: Asym-
metry 2000, 11, 2443–2448; (d) Giorgio, E.; Minichino, C.;
Viglione, R. G.; Zanasi, R.; Rosini, C. J. Org. Chem. 2003,
68, 5186–5192.
5. (a) Mennucci, B.; Tomasi, J.; Cammi, R.; Cheeseman, J. R.;
Frisch, M. J.; Devlin, F. J.; Gabriel, S.; Stephens, P. J. J.
Phys. Chem. A 2002, 106, 6102–6113; (b) Stephens, P. J.;
Devlin, F. J.; Cheeseman, J. R.; Frisch, M. J. J. Phys. Chem.
A 2001, 105, 5356–5371; (c) McCann, D. M.; Stephens, P. J.
J. Org. Chem. 2006, 71, 6074–6098.
6. (a) Liu, D.-Z.; Wang, F.; Liao, T.-G.; Tang, J.-G.; Steglich,
W.; Zhu, H.-J.; Liu, J.-K. Org. Lett. 2006, 8, 5749–5752; (b)
De Meijere, A.; Khlebnikov, A. F.; Kostikov, R. R.;
Kozhushkov, S. I.; Schreiner, P. R.; Wittkopp, A.; Yufit,
D. S. Angew. Chem., Int. Ed. 1999, 38, 3474–3477; (c) Pecul,
M.; Ruud, K. Adv. Quantum Chem. 2005, 50, 185–212; (d)
Rinderspacher, B. C.; Schreiner, P. R. J. Phys. Chem. A 2004,
108, 2867–2870.
1
purification. H NMR (400 MHz, CDCl3) d 0.82–0.86 (m,
3H), 0.92–0.96 (m, 3H), 1.18–1.31 (m, 1H), 1.32–1.36 (m,
1H), 2.22–2.33 (m, 2H), 2.56–2.65 (m, 2H), 4.15 (br, 2H),
6.77 (dd, J = 6.8 Hz, 1.6 Hz, 2H), 7.33–7.52 (m, 5H), 8.02
(dd, J = 6.8 Hz, 1.6 Hz, 2H); 13C NMR (100 MHz,
CDCl3), d 7.7, 14.2, 16.6, 31.0, 40.2, 87.4, 113.7, 121.1,
125.1, 126.5, 127.9, 131.5, 144.2, 150.6, 165.0.
4.4. Compound 7
To a solution of Boc-L-leucine (64 mg, 0.3 mmol) in THF
(8 mL), at room temperature, were added 55 mg (0.3
mmol) of 2-chloro-4,6-dimethoxy-1,3,5-trizaine (CDMT)
and 90 mg (0.9 mmol) of N-methylmorpholine (NMM).
After a white precipitate formed, a solution of amine 6
(80 mg, 0.3 mmol) was added into the flask. The mixture
was stirred overnight and then quenched with 0.5 mL
water. The solution was then extracted with EtOAc
(5 ꢃ 15 mL). The combined organic phases were washed
successively with saturated sodium carbonate, water, 1 M
HCl, water, and brine, respectively. The organic layer
was dried and subjected to silica gel chromatography (1:6
EtOAc/hexanes) to give 7 as a colorless solid (65 mg,
42.5% yield). 1H NMR (400 MHz, CDCl3) d 0.70–0.73
(m, 3H), 0.81–0.84 (m, 3H), 0.93–0.96 (m, 6H), 1.11–1.28
(m, 2H), 1.43 (s, 9H), 1.71–1.75 (m, 3H), 2.17–2.22 (m,
2H), 2.50–2.51 (m, 2H), 4.41 (m, 1H), 5.46 (d,
J = 5.2 Hz, 1H), 7.22–7.39 (m, 5H), 7.61 (d, J = 6.8 Hz,
2H), 7.98 (d, J = 6.8 Hz, 2H), 9.31(s, 1H); 13C NMR

T.-G. Liao et al. / Tetrahedron: Asymmetry 19 (2008) 808–815
815
7. (a) Grimme, S.; Bahlmann, A.; Haufe, G. Chirality 2002, 14,
793–797; (b) Marchesan, D.; Coriani, S.; Forzato, C.; Nitti,
P.; Pitacco, G.; Ruud, K. J. Phys. Chem. A 2005, 109, 1449–
1453; (c) Coriani, S.; Baranowska, A.; Ferrighi, L.; Forzato,
C.; Marchesan, D.; Nitti, P.; Pitacco, G.; Rizzo, A.; Ruud, K.
Chirality 2006, 18, 357–369.
8. (a) Wiberg, K. B.; Vaccaro, P. H.; Cheeseman, J. R. J. Am.
Chem. Soc. 2003, 125, 1888–1896; (b) Ribe, S.; Kondru, R.
K.; Beratan, D. N.; Wipf, P. J. Am. Chem. Soc. 2000, 122,
4608–4617.
13. (a) Basel, Y.; Hassner, A. J. Org. Chem. 2000, 65, 6368–6380;
(b) Byung, H. L.; Marvin, J. M. J. Org. Chem. 1983, 48, 24–31.
14. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.;
Vreven, T.; Kudin, K. N.; Burant, J. C.; Millam, J. M.;
Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi,
M.; Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.;
Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.;
Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.;
Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.;
Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratemann, R. E.;
Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski,
J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador,
P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.;
Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.;
Rabuk, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J.
V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.;
Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komar-
omi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M.
A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P.
M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.;
Pople, J. A. GAUSSIAN 03 User’s Reference, Gaussian,
Carnegie, PA 15106, USA, 2003.
9. Zhu, H.-J.; Ren, J.; Pittman, C. U., Jr. Tetrahedron 2007, 63,
2292–2314.
10. (a) Kasmai, H. S.; Mischke, S. G.; Blake, T. J. J. Org. Chem.
1995, 60, 2267–2270; (b) The Merck Index; Martha, W., Ed.,
11th ed.; Merk: Rahway (New Jersey), 1988, 6511.
11. (a) De Luca, L.; Giacomelli, G.; Taddei, M. J. Org. Chem.
2001, 66, 2534–2537; (b) Kunishima, M.; Kawachi, C.;
Monta, J.; Terao, K.; Iwasaki, F.; Tani, S. Tetrahedron
1999, 55, 13159–13170.
12. (a) Taylor, E. C.; Dowling, J. E. J. Org. Chem. 1997, 62,
1599–1603; (b) Wessel, H. P.; Mitchell, C. M.; Lobato, C. M.;
Schmid, G. Angew. Chem., Int. Ed. Engl. 1995, 34, 2712–
2713.