Solid state properties and effective resolution procedure for guaifenesin, 3-(2-methoxyphenoxy)-1,2-propanediol

Article abstract of DOI:10.1016/j.tetasy.2006.10.027

Racemic expectorant guaifenesin, 3-(2-methoxyphenoxy)-1,2-propanediol 2 undergoes spontaneous resolution upon crystallization. This fact is confirmed by thermal analysis (single eutectic V-shape binary melting phase diagram, adequate entropy and free energy characteristics). Racemic 2 could be effectively resolved into (S)- and (R)-2 by a preferential crystallization procedure. Single enantiomer drugs levomoprolol and levotensin were obtained by starting from enantiomeric 2 through the sulfite route.

Full text of DOI:10.1016/j.tetasy.2006.10.027

Tetrahedron: Asymmetry 17 (2006) 3015–3020
Solid state properties and effective resolution procedure
for guaifenesin, 3-(2-methoxyphenoxy)-1,2-propanediol
Zemfira A. Bredikhina, Victorina G. Novikova,
Dmitry V. Zakharychev and Alexander A. Bredikhin*
A. E. Arbuzov Institute of Organic and Physical Chemistry, Russian Academy of Sciences, Arbuzov St., 8, Kazan 420088,
Russian Federation
Received 13 October 2006; accepted 20 October 2006
Available online 15 November 2006
Abstract—Racemic expectorant guaifenesin, 3-(2-methoxyphenoxy)-1,2-propanediol 2 undergoes spontaneous resolution upon crystal-
lization. This fact is confirmed by thermal analysis (single eutectic V-shape binary melting phase diagram, adequate entropy and free
energy characteristics). Racemic 2 could be effectively resolved into (S)- and (R)-2 by a preferential crystallization procedure. Single enan-
tiomer drugs levomoprolol and levotensin were obtained by starting from enantiomeric 2 through the sulfite route.
ꢀ 2006 Elsevier Ltd. All rights reserved.
1. Introduction
of epoxide 1, 3-(2-methoxyphenoxy)-1,2-propanediol 2,
expectorant guaifenesin,4a attracts our attention. We have
already reported preliminary data about conglomerate type
of crystallization for this diol based on the depression of
melting temperature of racemic 2 as compared with sca-
lemic, and on the coincidence of IR spectra for racemic
and scalemic crystalline samples of guaifenesin.5
The problem of an economically effective and environmen-
tally friendly production of single enantiomer compounds
is among the main goals of modern organic chemistry.
The phenomenon of the spontaneous resolution of race-
mates is one of the promising ways to reach a desirable
aim.1 However, even in the case of conglomerate forming
substances it is not always easy to use the benefits of a
spontaneous resolution. We have recently reported the
conglomerate nature of a valuable intermediate in chiral
drug synthesis of 1,2-epoxy-3-(2-methoxyphenoxy)-pro-
pane 1, but were unable to achieve an effective resolution
of this compound by an entrainment procedure.2 The
described situation is not uncommon. Coquerel et al.3 is
of opinion that almost half of conglomerate forming
compounds would demonstrate poor entrainment
characteristics.
O
O
O
*
OH
O
1
O
OH
*
OH
O
H
2
O
N
*
Guaifenesin
3
Moprolol
Conceivable tactics to overcome this problem are to deter-
mine the reasons for such a behavior and to take rational
steps to counteract them, if possible. Another tactic is
not to overcome, but get around the obstacles, that is, to
find another conglomerate forming compound structurally
similar with the initial one. Among the formal derivatives
The first aim of this report is to extend the scope of the evi-
dence of the guaifenesin conglomerate nature by thermo-
chemical investigations. The second aim is to develop a
practical method of direct resolution of (nearly) racemic
diol 2. The last part of this paper deals with the use of
diastereomeric cyclic sulfites, epoxide-like derivatives of
enantiomeric 2, in the short synthesis of single enantiomer
b-adrenoblocker levomoprolol 3.4b
*
Corresponding author. Tel.: +7 843 2727392; fax: +7 843 2731872;
e-mail: baa@iopc.knc.ru
0957-4166/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2006.10.027

3016
Z. A. Bredikhina et al. / Tetrahedron: Asymmetry 17 (2006) 3015–3020
20
2. Results and discussion
acterized by the value ½aꢂD ¼ þ9:5 ꢃ 0:1 (c 1.0, anhydrous
MeOH). The following reasons were taken into account to
establish the enantiomeric purity of such a sample. (1) The
value of the specific rotation was constant after several con-
secutive crystallizations; for conglomerate forming sub-
stances (only) it is a solid evidence for the enantiomeric
purity. (2) The dsc heating curve for this sample has virtu-
ally an ideal shape with a base line and a linear front part
2.1. Thermochemical investigations
Thermodynamic data on binary mixtures of enantiomers
are useful, for checking the purity (including enantiomeric
purity) of chiral compounds, and for obtaining informa-
tion concerning a particular technique to be used for
achieving enantiomeric resolution and/or enrichment. This
part of the work deals with binary mixtures of (R)- and (S)-
2 using differential scanning calorimetry (dsc) as a research
method. The temperature data were determined according
to the method of Ho¨hne et al.,6 and were treated as
described previously.1
of the curve that constitutes a nearly perfect angle. (3) This
25
value was close enough to the published value ½aꢂD ¼ ꢀ9:4
(c 1.0, MeOH) for (R)-2, 99.4% ee.8
From the dsc data the melting temperature against the
composition diagram was drawn as depicted in Figure 1.
Experimental points in Figure 1 form an obvious single eu-
tectic V-shape curve typical for a racemic conglomerate.1
Figure 1 also presents the theoretical solidus and liquidus
curves (dashed lines) deduced for conglomerate from
the reductive Schro¨der–Van Laar equation lnðnÞ ¼
ðDHfA=RÞð1=T fA ꢀ 1=Tf Þ. Both experimental and theoretical
sets correlate quite well.
The results obtained for the temperature and enthalpy of
fusion of the pure enantiomer and the pure racemate, as
well as calculated1,7 values of entropy of mixing for liquid
enantiomeric 2, DSml , and free energy of formation for race-
mic compound in the solid state, DG0, are presented in
Table 1.
The entropy of mixing for enantiomeric 2 in the liquid state
is equal to 5.15 J Kꢀ1 molꢀ1, which is slightly less but close
to the ideal value of 5.75 J Kꢀ1 molꢀ1 (Rln2) for conglo-
merates. The near zero value for DG0 also shows the same
peculiarity of chiral 2.7
2.2. Resolution of guaifenesin by preferential crystallization
Guaifenesin 2 is a popular object for developing and testing
of enantioselective preparative approaches. Thus nonrace-
mic 2 has been obtained by multistep synthesis from natu-
ral D-mannitol9 and through one step interaction of
guaiacol with scalemic glycidol8,10 or 3-chloropropane-
1,2-diol.11 Both enantiomers of 2 have been prepared by
Sharpless asymmetric dihydroxylation of prochiral 1-allyl-
oxy-2-methoxy-benzene.12,13 Racemic 2 was resolved via
diastereomeric glycosylated derivatives.14 Kinetic resolu-
tion approaches including lipase catalyzed esterification
of rac-215 and Jacobsen salen-catalyzed enantioselective
For the right evaluation of the binary phase diagram based
on the dsc experiment, the correct information concerning
an enantiomeric composition [mole fraction of each enan-
tiomer, nR and nS, (nR + nS) = 1] of the samples investi-
gated is of primary importance. We have used for this
purpose the value of optical purity [ai]/[amax] = op ꢁ ee =
j(nR ꢀ nS)j/(nR + nS). The closer is the value [amax] used
in the calculations of to the value of the specific rotation
for enantiopure sample, the better is equivalence between
the optical purity and the enantiomeric excess.
100
Using the specific rotation data for quantitative conclu-
sions one must take into account the pronounced solvent
dependence of this chiroptical property. In particular, for
enantiomeric 2 the value of the specific rotation strongly
depends upon the water content in alcohols commonly
used for the measurements. Thus for enantiopure (S)-2
90
20
20
½aꢂD ¼ þ11:7 (c 1.0) in anhydrous EtOH, whereas ½aꢂD
¼
80
þ15:0 (c 1.0) in absolute EtOH. The same is true for
MeOH as well. For the same sample the value of
20
20
½aꢂD ¼ þ9:5 (c 1.0, anhydrous MeOH) changes to ½aꢂD
¼
þ12:1 (c 1.0) in the MeOH/H2O 9:1 (v/v) mixture. Notice
20
that for (S)-2 ½aꢂD ¼ þ13:7 (c 1.1, H2O), hence the water
70
influence is not an additive one.
0.0
0.5
1.0
(
(
Mole fraction of
S
-enantiomer
With this fact in mind, we used methanol, freshly distilled
from sodium for the quantitative measurements. We be-
lieve that enantiopure samples (ee >99%) of (S)-2 are char-
Figure 1. Experimental (s) and calculated (---) melting point phase
diagram of 2.
Table 1. Dsc measured melting point and enthalpy of fusion of racemic (low index R) and enantiopure (low index A) 3-(2-methoxyphenoxy)-1,2-
propanediol 2 and calculated thermodynamic characteristics for the substance
T0Af (ꢁC)
T0Rf (ꢁC)
DHfA (J molꢀ1
)
DHfR (J molꢀ1
)
DSml (J Kꢀ1 molꢀ1
)
DG0 (J molꢀ1
ꢀ118
)
97.2
80.2
41,800
37,600
5.15

Z. A. Bredikhina et al. / Tetrahedron: Asymmetry 17 (2006) 3015–3020
3017
hydrolysis of racemic glycidyl ether of guaiacol16 were also
used for scal-2 production.
tion (and the composition of deposited crystals as well)
then asymptotically tends to be racemic. It is apparent that
to achieve optimal output of the pure solid enantiomer, the
crystallization process must be interrupted not far from the
turning point of the curve. In the timescale of a real exper-
iment, this moment is difficult to fix for high starting con-
centrations (curves 1 and 2). On the other hand, overall
cycle times become too long for low concentrations (curves
4 and 5). As a result, the intermediate conditions were se-
lected for the preparative resolution of racemic guaifenesin.
The now uncovered conglomerate nature of guaifenesin
opens up possibilities for the direct resolution of this valu-
able substance. As a result we first examined entrainment
abilities of guaifenesin during seed induced crystallization
of its oversaturated slightly nonracemic solutions. We em-
ployed water as a ‘green’ solvent for the purpose. Mother
liquor optical rotation power was used as an indicator
quantity. The experimental time dependence for the mother
liquor optical rotation power is depicted in Figure 2.
An example of the successful resolution of near racemic 2 is
illustrated in Table 2. Over the course of the resolution, a
supersaturated solution of rac-2, including a small excess
of (R)-2, was prepared by heating, and then cooling to
22 ꢁC. A small amount of seed crystals of (R)-2 was added
and the stirred solution was allowed to crystallize for about
100 min. The weight of (R)-2 obtained after filtration was
more than common weight of the initial excess of the
(R)-enantiomer and seed added. To the mother liquor,
rac-2 was added in order that the overall quantity of 2 in
the solution could be recovered. The mixture was heated
until the solid was completely dissolved and then cooled
to 22 ꢁC. After the solution had been seeded with (S)-2
and stirred for about 100 min, precipitated (S)-2 was col-
lected in a similar manner. The cycle was repeated several
times.
As can be seen from Figure 2, investigated solutions of dif-
ferent concentrations but of equal starting enantiomeric
enrichment behave uniformly in the main features. During
the crystallization process primary enantiomeric prevalence
of solute decreases to zero, changes sign, and mounts to ini-
tial absolute value (even exceeds this value for high initial
concentrations). The mother liquor enantiomeric composi-
1
2
0.4
3
4
5
0.2
As cited in Table 2 values for yield of enantiomer [YE (g)]
and degree of resolution [DR (%)] were calculated by anal-
ogy with Shiraiwa et al.,17 according to the following
equations:
0.0
-0.2
-0.4
YE ðgÞ ¼ ½YieldðgÞꢄOP ð%Þꢂ=100ꢀ0:25
DR ð%Þ ¼ YE ðgÞꢄ100=½0:5ꢄðAmount of rac-2 ðgÞꢀ34:4Þꢂ
0
100
200
300
1200 1300
Resolution time (min)
where OP is the optical purity of the obtained (S)- or (R)-2,
0.25 (g) is the seed weight, and 34.4 (g) is the solubility
determined by us of rac-2 in 800 cm3 water at 22 ꢁC.
Figure 2. The dynamics of guaifenesin crystallization. Mother liquor
optical rotation power (T = 25 ꢁC; cell length 100 mm; k = 436 nm) time
dependence during seed induced crystallization of slightly enantiomeri-
cally enriched 2. Experimental conditions: T = 22 ꢁC; 25 mg of powered
seed crystals per run; 9 g of rac-2 + 1 g of (R)-2 in 60 (curve 1), 70 (curve
2), 80 (curve 3), 100 (curve 4), and 120 (curve 5) cm3 of H2O.
As evident from Table 2, three cycles of entrainment reso-
lution using ordinary laboratory equipment and no resolv-
ing agents are sufficient enough to obtain (R)- and (S)-2
Table 2. Resolution by entrainment of rac-guaifenesin in water (800 cm3 H2O, 0.25 g crystal seeds on every run, crystallization temperature 22 0.5 ꢁC)
Run
Added amount of (R,S)-2 (g)
Operation
amount of (R)-
and (S)-2b (g)
Resolution time (min)
(R)- and (S)-2 obtained
(R)-2
(S)-2
Yield (g)
OPc (%)
YEd (g)
DRe (%)
1
2
3
4
5
6
90.46a
21.01
18.86
17.28
16.97
19.43
55.77
45.24
54.86
46.02
53.08
44.68
45.23
54.76
45.40
53.98
46.92
55.32
100
110
110
115
120
125
(R) 21.26
(S) 19.11
(R) 17.53
(S) 17.22
(R) 19.68
(S) 22.60
91
99
>99
84
87
89
19.10
18.67
17.11
14.21
16.87
19.84
68.1
66.6
60.7
49.3
56.8
72.2
a Additional amount of (R)-2 9.54 g.
b The operation amounts in runs 2–6 were calculated based on the results in runs 1–5 respectively.
20
c OP: Optical purity; the optical purities of (R)- and (S)-2 obtained, were calculated on the basis of the specific rotation of (S)-2: ½aꢂD ¼ þ9:5 (c 1.0,
MeOH).
d YE: Yield of enantiomer.
e DR: Degree of resolution.

3018
Z. A. Bredikhina et al. / Tetrahedron: Asymmetry 17 (2006) 3015–3020
samples of about 50 g each. The quality of each nonrace-
mic specimen is good enough, which single recrystallization
is sufficient to obtain (R)- and (S)-2 diols suitable for
further use.
4. Experimental
4.1. General
The NMR spectra were recorded on the Bruker WM-250
and the Avance-600 spectrometers in CDCl3 with TMS
or the signals of the solvent as the internal standard. Opti-
cal rotations were measured on a Perkin–Elmer model 341
polarimeter (concentration c is given as g/100 ml). Melting
points for general purposes were determined using a
Boe¨tius apparatus and are uncorrected.
2.3. Levomoprolol synthesis
Guaifenesin 2 has an obvious structural similarity with
the known b-adrenoblocker moprolol 3, which belongs
to an extensive class of 1-alkylamino-3-aryloxypropan-2-
ols with cardiovascular activity. Within this class, it has
been shown that (S)-enantiomers are eutomer components
of the racemic drug, whereas (R)-enantiomers (distomers)
usually display other (often undesirable) activities.18 The
availability of (S)-2 allowed us to obtain single enantio-
mer drug (S)-3, also known as levomoprolol and its
hydrochloride (S)-3ÆHCl known as levotenzine.4b We
used the well-known cyclic sulfite route19,20 outlined in
Scheme 1.
The melting curves of the samples of 3-(2-methoxyphen-
oxy)-1,2-propanediol were measured on a Setaram DSC-
111 differential scanning calorimeter in stainless steel cells
with the rate of heating of 1 ꢁC minꢀ1. Mass of the samples
amounted to approximately 2.5 mg. Temperature scale and
heat flux were calibrated against the data for a-corundum
(sapphire), phenol, and naphthalene. Experimental DSC
curves were treated according to Gallis et al.22
Due to the appearance of an additional chiral sulfur atom,
cyclic sulfites 4 were obtained as an epimer mixture with an
(R)-configuration at the carbon centre. The mixture was
used in the next step without the separation of individual
diastereomers. Overall nonoptimized isolated yields of
(S)-3 and (S)-3ÆHCl were better than 80%.
4.2. Synthesis
Racemic epichlorohydrin, 3-chloropropane-1,2-diol and
guaiacol are commercially available. (R)- and (S)-3-
chloro-1,2-propanediol were prepared by the enantio-
selective partial hydrolysis of racemic epichlorohydrin
according to the Jacobsen method without modifications.23
3. Conclusion
4.2.1. Racemic 3-(2-methoxyphenoxy)-propane-1,2-diol, rac-
guaifenesin, rac-2. Racemic 3-(2-methoxyphenoxy)-pro-
pane-1,2-diol, rac-guaifenesin, rac-2 was prepared follow-
ing the general procedure of Egri et al.24 To a solution of
guaiacol (83.4 g, 0.67 mol) in ethanol (400 ml), a solution
of NaOH (33.6 g, 0.84 mol) in water (135 ml) was added
and the resulting mixture stirred and heated at reflux for
30 min. A solution of racemic 3-chloropropane-1,2-diol
(111.5 g, 1.01 mol) in ethanol (84 ml) was then added with-
in 30 min and the mixture was further stirred and heated at
reflux for 8 h. After cooling, the volume of the resulting
mixture was reduced to about one third followed by the
addition of water (400 ml) and extraction with CH2Cl2
(3 · 200 ml). The combined organic layers were dried over
anhydrous Na2SO4 and the solvent was removed. Crude
diol 2 was purified by distillation (bp 110–120 ꢁC at
0.05 mmHg) and recrystallization from CCl4. Yield 81.2 g
(61%), mp 79–80 ꢁC; lit.25 mp 78.5–79.5 ꢁC. 1H NMR
(600 MHz, CDCl3): d = 2.74 (br s, 1H, OH), 3.48 (br s,
1H, OH), 3.76–3.81 (m, 2H, CH2), 3.85 (s, 3H, CH3),
4.04–4.06 (m, 2H, CH2), 4.13–4.14 (m, 1H, CH), 6.88–
6.92 (m, 3H, Ar), 6.95–6.98 (m, 1H, Ar).
To the best of our knowledge glycerol ether 2, guaifenesin
is the first registered chiral drug resolved by the direct crys-
tallization as itself, not in the form of a derivative of any
kind. Guaifenesin belongs to the family of chiral 3-(2-R-
phenoxy)-1,2-propanediols within the range of which the
conglomerate forming representatives are common ones.
Thus, based on thermochemical properties we disclosed
the conglomerate nature of 2-I-, 2-Br-, and 2-Cl-phenoxy-
propanediols, whereas 2-F- and unsubstituted ether forms
unstable and rather stable racemic compounds, respec-
tively.21 It could be suggested with a great deal of certainty
that 2-CN- and 2-EtO-derivatives are conglomerate
forming substances. Among 3-(2-alkylphenoxy)-1,2-prop-
anediols at least 2-Me-substituted compound, another reg-
istered drug, muscle relaxant mephenesin4c also forms
conglomerate upon crystallization.5 We believe that a more
detailed search for conglomerates within this highly potent
class of chemicals would lead to more deep understanding
of the nature of spontaneous resolution and more intent
practical use of this unappreciated phenomenon.
OMe
OH
OH
H
O
O
O
H
PriNH
2
SOCl
H
OH
NHPri
2
S
O
O
O
OMe
OMe
(S
)-
2
(R
)
-
4
(S)-
3
Scheme 1.

Z. A. Bredikhina et al. / Tetrahedron: Asymmetry 17 (2006) 3015–3020
3019
4.2.2. (R)-3-(2-Methoxyphenoxy)-propane-1,2-diol, (R)-
2. (R)-3-(2-Methoxyphenoxy)-propane-1,2-diol, (R)-2
(0 ꢁC) solution of (S)-3-(2-methoxyphenoxy)-propane-1,2-
diol (S)-2 (1.0 g, 5 mmol) in CH2Cl2 (50 ml), a solution
of SOCl2 (0.62 g, 5.2 mmol) in CH2Cl2 (10 ml) was added
dropwise. The reaction mixture was stirred for an extra
1 h, and the volatile material removed under reduced pres-
which was used as the seed was obtained from the (R)-3-
20
chloropropane-1,2-diol {13.3 g, 0.12 mol; ½aꢂD ¼ ꢀ6:4 (c
5, H2O)} and guaiacol (12.4 g, 0.1 mol) as described for
racemic compound. Crude (R)-2 (16.38 g, 83%) was crys-
tallized from a mixture of CCl4 (200 ml) and EtOH
(12 ml) to give colourless needles {12.23 g (62%); mp 97–
sure to afford 1.2 g {solid, mixture of cis and trans isomers
20
(43:57), mp 63–69 ꢁC, ½aꢂD ¼ þ42:3 (c 1.0, CH2Cl2)} of
(2RS,4R)-4, which was used in the next step without fur-
20
20
1
98 ꢁC; ½aꢂD ¼ ꢀ9:5 (c 1.0, MeOH), ½aꢂD ¼ ꢀ11:7 (c 1.0,
ther purification. H NMR (250 MHz, CDCl3): d = 3.85
25
EtOH), >99% op}; {lit.8 mp 96.8–99.1 ꢁC, ½aꢂD ¼ ꢀ9:4 (c
(s, 3H, OCH3); 3.99–4.19 (m, 1.14H, CH2OS (trans));
4.25–4.46 (m, 0.86H, CH2OS (cis)); 4.53–4.59, 4.68–4.71,
4.79–4.85 (all m, totally 2H, CH2OAr (cis, trans)); 4.90–
4.95 (m, 0.43H, CHOS (cis)); 5.23–5.32 (m, 0.57H, CHOS
(trans)); 6.89–7.05 (4H, Ar). 13C NMR (62.9 MHz,
CDCl3): d = 56.17 (CH3); 68.84 (CH2OAr (trans)); 69.00
(CH2OAr (cis)); 70.19 (CH2OS (trans)); 70.85 (CH2OS
(cis)); 78.13 (CHOS (trans)); 80.17 (CHOS (cis)); 112.76,
112.81 ðC3ArÞ; 116.37, 116.60 ðCA6 rÞ; 121.15, 121.17 ðCA4 rÞ;
123.19, 123.36 ðCA5 rÞ; 147.71 ðCA1 rÞ; 150.37, 150.50 ðCA2 rÞ.
1.0, MeOH), 99.4% ee}. 1H NMR (600 MHz, CDCl3):
d = 2.71 (br s, 1H, OH), 3.47 (br s, 1H, OH), 3.77–3.83
(m, 2H, CH2), 3.86 (s, 3H, CH3), 4.06–4.08 (m, 2H,
CH2), 4.13–4.15 (m, 1H, CH), 6.90–6.93 (m, 3H, Ar),
6.96–6.98 (m, 1H, Ar). 13C NMR (150.864 MHz, CDCl3):
d = 55.86 (CH3), 63.89 (CH2OH), 70.08 (CH2O),
72.29 (CH), 111.98 ðC3ArÞ, 114.95 ðCA6 rÞ, 121.13 ðCA4 rÞ,
122.22 ðC5ArÞ, 148.07 ðC1ArÞ, 149.79 ðCA2 rÞ.
4.2.3. (S)-3-(2-Methoxyphenoxy)-propane-1, 2-diol, (S)-
2. (S)-3-(2-Methoxyphenoxy)-propane-1,2-diol,
(S)-2
which was used as the seed was synthesized analogously
4.3.2.
(S)-1-Isopropylamino-3-(2-methoxyphenoxy)-pro-
20
from the (S)-3-chloropropane-1,2-diol {½aꢂD ¼ þ6:4
pane-2-ol, (S)-moprolol, levomoprolol, (S)-3. A solution
of dioxathiolane (2RS,4R)-4 (1.08 g, 4.4 mmol) and
PriNH2 (4 g, 68 mmol) in DMF (10 ml) was heated at
60–70 ꢁC for 45 h. After this period, excess amine and
DMF were removed in vacuo. A solution of NaOH
(35 ml, 1 M) was added, the mixture extracted with AcOEt
(3 · 40 ml), and extract was dried with Na2SO4. After re-
moval of the solvent in vacuo, the residue (1.1 g, 87%)
was crystallized from EtOAc to give (S)-3, mp 78–80 ꢁC;
(c 4.7, H2O)}. The colourless needles; yield 63%; mp 97–
20
25
98 ꢁC; ½aꢂD ¼ þ9:4 (c 1.0, MeOH). {lit.15 ½aꢂD ¼ þ11:2 (c
20
1, EtOH), 88% ee; lit.5 ½aꢂD ¼ þ9:4 (c 1.0, MeOH)}.
4.3. Resolution of racemic 3-(2-methoxyphenoxy)-1,2-pro-
panediol (guaifenesin, rac-2) by preferential crystallization
(entrainment)
Racemic guaifenesin rac-2 (90.46 g) and (R)-2 (9.54 g) was
dissolved in 800 ml of water at 42–45 ꢁC. The solution was
cooled to 23 ꢁC and seeded with finely pulverized (R)-2
(0.25 g). After stirring the mixture for 100 min at
20
20
½aꢂD ¼ ꢀ2:8 (c 2.25, EtOH); lit.26 mp 80–81 ꢁC, ½aꢂD
¼
ꢀ5:6 (c 4.5, EtOH). 1H NMR (400 MHz, CDCl3):
d = 1.06 (d, 3H, CH3), 1.07 (d, 3H, CH3), 2.20–2.86 (m,
5H, OH, NH, CH2O, NCH), 3.84 (s, 3H, OCH3), 3.96–
4.06 (m, 3H, CH2O, CH), 6.87–6.96 (m, 4H, ArH) [cf.
lit.26]. Solid (S)-3 was dissolved in 30 ml of Et2O and gas-
eous HCl was passed through the resulting solution to give
0.95 g (82%) of (2S)-1-isopropylamino-3-(2-methoxyphen-
22 0.5 ꢁC, precipitated (R)-2 was collected by filtration
20
{21.26 g after drying; ½aꢂD ¼ ꢀ8:6 (c 1.1, MeOH), 91% op}
(Table 2, run 1). The extra portion of rac-2 (21.01 g)
was then dissolved in the mother liquor at 42 ꢁC; the result-
ing solution was cooled to 23 ꢁC. After the addition of (S)-
2 (0.25 g) as seed crystals to the solution, and stirring the
oxy)-propane-2-ol hydrochloride (levotensin) (S)-3ÆHCl,
20
mp 124–125 ꢁC (EtOAc/EtOH); ½aꢂD ¼ ꢀ16:5 (c 5.0,
20
mixture for 110 min at 22 0.5 ꢁC, (S)-2 {19.11 g after
EtOH). Lit.4b mp 121–123 ꢁC, ½aꢂD ¼ ꢀ16:3 (c 5.0, EtOH).
20
drying; ½aꢂD ¼ þ9:4 (c 1.0, MeOH), 99% op} was collected
1H NMR (600 MHz, CDCl3): d = 1.54 (d, J = 6.6, 3H,
3
by filtration (run 2). Further resolution was carried out at
22 0.5 ꢁC by adding amended amounts of rac-2 to the fil-
trate in a manner similar to that described above. The
CH3), 1.53 (d, 3J = 6.6, 3H, CH3), 3.26–3.30 (m, 1H,
1N+CH2), 3.44–3.46 (m, 1H, 1N+CH2), 3.51–3.53 (m,
1H, N+CH), 3.88 (s, 3H, OCH3), 4.10 (dd, 2J = 9.8,
3J = 5.9, 1H, 1OCH2), 4.22 (dd, 2J = 9.8, 3J = 4.2, 1H,
1OCH2), 4.63–4.67 (m, 1H, OCH), 5.40 (br s, OH), 6.91–
7.01 (m, 4H, Ar), 8.69 (br s, N+H), 9.50 (br s,
1H, N+H). 13C NMR (150.864 MHz, CDCl3): d =
19.11 (CH3), 18.98 (CH3), 48.45 (CH2N), 51.43 (CHN),
detailed conditions are given in Table 2. After second cycle,
20
17.53 g of (R)-2 {½aꢂD ¼ ꢀ9:6 (c 1.0, MeOH), >99% op}
20
and 17.22 g of (S)-2 {½aꢂD ¼ þ7:9 (c 1.0, MeOH), 84%
op} were collected. After third cycle, 19.68 g of (R)-2
20
{½aꢂD ¼ ꢀ8:2 (c 1.0, MeOH), 87% op} and 22.60 g of (S)-
20
2 {½aꢂD ¼ þ8:4 (c 1.1, MeOH), 89% op} were collected.
55.89 (OCH3), 65.52 (CH), 71.93 (OCH2), 112.22 ðC3 Þ,
Ar
A high degree of enantiomeric purity of collected diols
115.12 ðC6ArÞ, 121.17 ðCA4 rÞ, 122.40 ðC5ArÞ, 147.83 ðCA1 rÞ,
149.76 ðC2ArÞ.
can be achieved by simple recrystallization. For example:
20
a portion of (R)-2 (21.26 g, ½aꢂD ¼ ꢀ8:6 (c 1.1, MeOH),
91% op) was dissolved in a boiling mixture of CCl4
(250 ml) and EtOH (15 ml). After cooling the solution to
5–10 ꢁC, the crystallized (R)-2 was collected by filtration
20
(yield 19.75 g; ½aꢂD ¼ ꢀ9:5 (c 1.0, MeOH), >99% op).
Acknowledgement
4.3.1. (2RS,4R)-4-(2-Methoxyphenoxy)methyl-1,3,2-dioxa-
thiolane-2-one (2RS,4R)-4. To a stirred and cooled
The authors thank the Russian Fund of Basic Research for
financial support (Grant No. 06-03-32508).

3020
Z. A. Bredikhina et al. / Tetrahedron: Asymmetry 17 (2006) 3015–3020
14. Mereyala, H. B.; Mamidyala, S. K.; Chigurupati, K. P.;
References
Srinivasa, S. R. Synthesis 2003, 2378–2384.
15. Theil, F.; Weider, J.; Ballschuh, S.; Kunath, A.; Schick, H. J.
Org. Chem. 1994, 59, 388–393.
16. Bredikhin, A. A.; Strunskaya, E. I.; Novikova, V. G.;
Azancheev, N. M.; Sharafutdinova, D. R.; Bredikhina, Z.
A. Russ. Chem. Bull., Int. Ed. 2004, 53, 213–218, [Izv. Akad.
Nauk, Ser. Khim. 2004, 203–208 (in Russian)].
17. Shiraiwa, T.; Okhubo, M.; Miyazaki, H.; Kubo, M.; Nishi-
gawa, H.; Tsujimoto, T.; Kurokawa, H. Bull. Chem. Soc. Jpn.
1998, 71, 735–739.
1. Jacques, J.; Collet, A.; Wilen, S. H. Enantiomers, Racemates
and Resolutions; Krieger Publishing Company: Malabar, FL,
1994, 447pp.
2. Bredikhin, A. A.; Strunskaya, E. I.; Zakharychev, D. V.;
Krivolapov, D. B.; Litvinov, I. A.; Bredikhina, Z. A.
Tetrahedron: Asymmetry 2005, 16, 3361–3366.
3. Dufour, F.; Perez, G.; Coquerel, G. Bull. Chem. Soc. Jpn
2004, 77, 79–86.
4. (a) The Merck Index, 12th ed.; Budavari, S., Ed.; Merck:
Whitehouse Station, 1996; p 4582; (b) The Merck Index, 12th
ed.; Budavari, S., Ed.; Merck: Whitehouse Station, 1996; p
6345; (c) The Merck Index, 12th ed.; Budavari, S., Ed.;
Merck: Whitehouse Station, 1996; p 5895.
18. Mehvar, R.; Brocks, D. R. J. Pharm. Pharm. Sci. 2001, 4,
185–200.
19. Carlsen, P. H. J.; Aase, K. Acta Chem. Scand. 1993, 47, 737–
738.
20. Bredikhina, Z. A.; Savel’ev, D. V.; Bredikhin, A. A. Russ. J.
Org. Chem. 2002, 38, 213–219, [Zh. Organ. Khim. 2002, 38,
233–239 (in Russian)].
21. Zakharychev, D. V.; Lazarev, S. N.; Bredikhina, Z. A.;
Bredikhin, A. A. Russ. Chem. Bull., Int. Ed. 2006, 55, 230–
237, [Izv. Akad. Nauk, Ser. Khim. 2006, 225–232 (in
Russian)].
22. Gallis, H. E.; Bougrioua, F.; Oonk, H. A. J.; van Ekeren, P.
J.; van Miltenburg, J. C. Thermochim. Acta 1996, 274, 231–
242.
5. Bredikhin, A. A.; Bredikhina, Z. A.; Lazarev, S. N.; Savel’ev,
D. V. Mendeleev Commun. 2003, 13, 104–105.
6. Ho¨hne, G. W. H.; Cammenga, H. K.; Eysel, W.; Gmelin, E.;
Hemminger, W. Thermochim. Acta 1990, 160, 1–12.
7. Li, Z. J.; Zell, M. T.; Munson, E. J.; Grant, D. J. W. J.
Pharm. Sci. 1999, 88, 337–346.
8. Kitaori, K.; Furukawa, Y.; Yoshimoto, H.; Otera, J. Tetra-
hedron 1999, 55, 14381–14390.
9. Nelson, W. L.; Wennerstrom, J. E.; Sankar, S. R. J. Org.
Chem. 1977, 42, 1006–1012.
23. Schaus, S. E.; Brandes, B. D.; Larrow, J. F.; Tokunaga, M.;
Hansen, K. B.; Gould, A. E.; Furrow, M. E.; Jacobsen, E. N.
J. Am. Chem. Soc. 2002, 124, 1307–1315.
24. Egri, G.; Kolbert, A.; Balint, J.; Fogassy, E.; Novak, L.;
Poppe, L. Tetrahedron: Asymmetry 1998, 9, 271–283.
25. Yale, H. L.; Pribyl, E. J.; Braker, W.; Bergeim, F. H.; Lott,
W. A. J. Am. Chem. Soc. 1950, 72, 3710–3716.
26. Hou, X. L.; Li, B. F.; Dai, L. X. Tetrahedron: Asymmetry
1999, 10, 2319–2326.
10. Chen, J.; Shum, W. Tetrahedron Lett. 1995, 36, 2379–
2380.
11. Iriuchijima, S.; Kojima, N. Agric. Biol. Chem. 1982, 46, 1153–
1157.
12. Wang, Z.-M.; Zhang, X.-L.; Sharpless, K. B. Tetrahedron
Lett. 1993, 34, 2267–2270.
13. Sayyed, I. A.; Thakur, V. V.; Nikalje, M. D.; Dewkar, G. K.;
Kotkar, S. P.; Sudalai, A. Tetrahedron 2005, 61, 2831–
2838.