Systematic search for conglomerates among glycerol aromatic monoethers: Guaifenesin and mephenesin are the cases

Article abstract of DOI:10.1070/MC2003v013n03ABEH001752

Within the family of biologically active 3-aryloxy-1,2-propanediols, three new conglomerate-forming compounds have been found and resolved into enantiomers using the entrainment procedure.

Full text of DOI:10.1070/MC2003v013n03ABEH001752

, 2003, 13(3), 104–105
Systematic search for conglomerates among glycerol aromatic monoethers:
guaifenesin and mephenesin are the cases
Alexander A. Bredikhin,* Zemfira A. Bredikhina, Sergey N. Lazarev and Dmitry V. Savel’ev
A. E. Arbuzov Institute of Organic and Physical Chemistry, Kazan Scientific Centre of the Russian Academy of Sciences,
420088 Kazan, Russian Federation. Fax: +7 8432 73 2253; e-mail: baa@iopc.kcn.ru
10.1070/MC2003v013n03ABEH001752
Within the family of biologically active 3-aryloxy-1,2-propanediols, three new conglomerate-forming compounds have been found
and resolved into enantiomers using the entrainment procedure.
Louis Pasteur’s discovery of spontaneous resolution of Mitscherlich’s
salt1,2 has had diverse and far going consequences. The practical
use of spontaneous resolution for the efficient production of single
enantiomer compounds is an example. Industrial applications of
the phenomenon can be illustrated by the Merck process for
antihypertensive methyldopa, the Haarmann and Reimer process
for l-menthol, the Roussel–Uclaf process for chloroamphenicol,
and industrial processes for artificial α-amino acids.3
However, the potential of spontaneous resolution procedures
is restrained by the necessity for a chiral compound subjected
to resolution being a conglomerate. It should crystallise as a
mechanical mixture of single crystals formed by homochiral
molecules. Until now the list of conglomerate-forming chiral
substances may number only in hundreds,4 which is a tiny part
of millions of known chemical compounds. The pettiness of
this list is a real obstacle for not only fruitful practical use but
also theoretical explanations of spontaneous homochiral crystal-
lization. Thus, a search for conglomerates among relatively simple
molecules is of practical importance.
Table 1 Fusion temperatures and IR spectra of scalemic and racemic
3-aryloxypropane-1,2-diols ArOCH2CH(OH)CH2OH 1–11.
IR spectrab
of scal- vs.
rac-diol
No.
Tfscal /°Ca
Tfrac /°C
∆Ts-r/°C
1
2
3
4
5
6
7
8
9
91–92
99
69–70
79
22
20
20
12
11
10
5
Identical
Identical
Similar
91–92
111–112
62–64
74–76
47–49
83
79–80
60–62
64–65
71–72
99–100
51–53
64–66
43–44
80
80–81
67–69
73–74
Different
Different
Different
Different
Different
Different
Different
Different
3
–1
–7
–9
10
11
aThe enantiomeric purities of scalemic samples are characterised in
footnote.† bPolycrystalline samples in KBr pellets.
During our investigations of synthetic approaches to nonracemic
β-adrenoblockers of 3-aryloxy-1-alkylaminopropan-2-ol family,5
we have obtained a representative set of 3-aryloxypropane-1,2-
diols 1–11 through the interaction of phenols and glycidol
(Scheme 1).
of equal amounts of opposite enantiomers lies between higher
fusion temperature regions restricted by mirror-like liquidus curves
started from pure enantiomers.4 Therefore, a necessary (yet not
only) attribute for this type of crystallization is the higher melting
point for scalemic samples than for racemic ones. Table 1 presents
a comparison of fusion temperatures for the pairs of scalemic
and racemic polycrystalline samples of diols 1–11 and the dif-
ference ∆Tfs-r = Tfscal – Tfrac for each pair. For 8 from the 11 test
compounds, ∆Tfs-r was positive, and for entries 1–6 this value
was sufficiently high to suspect conglomerate formation.
The identity of IR spectra for the pairs of racemic and scalemic
polycrystal samples of chiral substances is another diagnostic for
conglomerate. We have also done this test, and the results are
presented in Table 1. For the upper three members of the column,
the IR spectra are identical or closely similar within each pair.
The X-ray studies of single crystals picked up from racemic
samples of o-substituted phenyloxy compounds 1–3 have revealed
that they all belong to ‘chiral’ space groups P21, P212121 and
P212121, respectively. This can be considered as evidence for
the conglomerate nature of solid diols 1–3.
Aromatic monoethers of glycerol (as well as their carbamates)
form a family of highly biologically active compounds. Among
their number o-tolyloxy- and o-methoxyphenyloxy-derivatives
1 and 2 are the registered drugs known under the nonpropri-
etary names mephenesin and guaifenesin. In this family, one
enantiomer is more active in vivo than the other or a racemate.6
Because of their importance for medicinal chemistry, 3-aryloxy-
propane-1,2-diols have received much attention from the synthetic
standpoint. Among many approaches to single enantiomer glycerol
ethers, one can find the ‘Chiral Pool’ based synthesis,7 catalytic
enantioselective synthesis, especially Sharpless asymmetric di-
hydroxylation of aryl allyl ethers,8 biotransformations through
the aid of living micro-organisms9 and the kinetic resolution of
racemic derivatives (esters) with the use of natural enzymes.10
The crystallization of a chiral substance as a racemic con-
glomerate makes it possible to resolve it into enantiomers with-
out resort to enantiopure chiral reagents and/or auxiliaries. This
problem can be solved, for example, using resolution by the
entrainment technique,4,11 which takes its roots in an early work
O
Et3N
ArOH +
ArO–CH –CH(OH)–CH OH
2 2
HO
1–11
1 Ar = 2-MeC6H4
2 Ar = 2-MeOC6H4
3 Ar = 2-ClC6H4
4 Ar = 1-Naphthyl
5 Ar = Ph
7 Ar = 2-Allyl-C6H4
8 Ar = 4-ClC6H4
9 Ar = 4-MeOC6H4
10 Ar = 3-MeC6H4
11 Ar = 4-MeC6H4
6 Ar = 3-ClC6H4
Scheme 1
Dealing with scalemic glycidol of modest enantiomeric purity,
we usually have received scalemic aryloxydiols of the same
quality. However, in all cases, simple recrystallization was suf-
ficient for producing samples of high optical purity.† This be-
haviour implies that a zone of racemate formation on the ternary
solubility phase diagram (and on the binary melting phase dia-
gram) is relatively narrow for the whole family, and it may be
altered to zero width for conglomerate-forming representatives.
On the binary melting phase diagram of a conglomerate-
forming chiral substance, the only low-melting eutectic consisting
(R)-1: [a]D20 +19.3 [c 1.15, hexane–PriOH (4:1)], lit.,10 [a]D20 +19.8 [c 0.9,
†
hexane–PriOH (4:1), ee > 99%]; (S)-2: [a]D20 +9.4 (c 1.0, MeOH), lit.,13
for (R)-2: [a]D20 –9.4 (c 1.0, MeOH, ee 99.4%); (S)-3: [a]D20 –13.3 [c 1.0,
hexane–EtOH (4:1)], lit.,10 [a]2D0 –13.4 [c 0.9, hexane–EtOH (4:1), ee
99%]; (S)-4: [a]D20 +7.5 (c 1.0, MeOH), lit.,14 [a]D20 +7.6 (c 1.0, MeOH);
(S)-5: [a]D20 +9.1 (c 1.7, EtOH), lit.,10 [a]D20 +10.2 (c 1.0, EtOH, ee
91%); (S)-6: [a]D20 +11.2 (c 1.0, EtOH), lit.,10 [a]2D0 +13.7 (c 1.0, EtOH,
ee 98%); (S)-7: [a]D20 –2.1 (c 2.8, EtOH); (S)-8: [a]2D0 +9.9 (c 1.1,
MeOH), lit.,13 for (R)-8: [a]D20 –10.2 (c 1.0, MeOH, ee 99.4%); (S)-9:
[a]D20 +7.3 (c 1.3, EtOH), lit.,10 [a]D20 +7.9 (c 1.0, EtOH, ee 96%);
(S)-10: [a]D20 +8.1 (c 0.84, EtOH), lit.,10 [a]D20 +9.5 (c 1.0, EtOH, ee
97%); (R)-11: [a]D20 –8.4 (c 1.0, EtOH), lit.,10 [a]D20 –9.2 (c 1.0, EtOH,
ee 97%).
– 104 –

, 2003, 13(3), 104–105
of Pasteur’s student Gernez.12 Using the procedure, one should
seed the oversaturated solution of a racemic or slightly scalemic
substrate with the appropriate single enantiomer seed. After a
time, the crop of a scalemic product having the configuration of
the seed substance is filtered off. Next, the racemic substrate is
dissolved in the mother liquor to compensate the removed crys-
tals, and the oversaturated solution of a slightly scalemic substrate
is seeded with the single enantiomer seed having an opposite
configuration to obtain the new crop of nonracemic crystals of
another enantiomer. Theoretically, one can obtain any desirable
quantity of both enantiomers with minimal losses and waste pro-
ducts using the above cyclic procedure (Scheme 2).
We hope that our finding will be useful for the large-scale
production of this pharmaceutically active single enantiomer sub-
stances.
This work was supported by the Russian Foundation for
Basic Research (grant no. 03-03-33084). We are grateful to
I .A. Litvinov and A. T. Gubaidullin for performing X-ray experi-
ments and to I. I. Vandyukova for her assistance in the measure-
ment of IR spectra.
References
1
(a) L. Pasteur, Ann. Chim. Phys., 1848, 24, 442; (b) L. Pasteur, Compt.
Rend. Acad. Sci., 1848, 26, 535.
H
OH
2
3
4
G. B. Kauffman and R. D. Myers, J. Chem. Educ., 1975, 52, 777.
J. Crosby, Tetrahedron, 1991, 47, 4789.
O
OH
R
J. Jaques, A. Collet and S. H. Wilen, Enantiomers, Racemates, and
Resolutions, Krieger Publishing Co., Malabar, FL, 1994.
Z. A. Bredikhina, D. V. Savel’ev and A. A. Bredikhin, Zh. Org. Khim.,
2002, 38, 233 (Russ. J. Org. Chem., 2002, 38, 213).
E. Souri, M. Sharifzadeh, H. Farsam and N. Gharavi, J. Pharm.
Pharmacol., 1999, 51, 853.
W. L. Nelson, J. E. Wennerstrom and S. R. Sankar, J. Org. Chem.,
1977, 42, 1006.
(a) Z.-M. Wang, X.-L. Zhang and K. B. Sharpless, Tetrahedron Lett.,
1993, 34, 2267; (b) H. C. Kolb, M. S. VanNieuwenhze and K. B.
Sharpless, Chem. Rev., 1994, 94, 2483.
(–)-isomer
(–)-seed
OH
5
6
7
8
(S)-1–3
O
OH
+
+
/–
/–
–/+
R
(+/–)-1–3
–
+
/+
–
(+)-seed
HO
H
rac-1–3
O
OH
(+)-isomer
R
1 R = Me
2 R = OMe
3 R = Cl
9
G. Egri, A. Kolbert, J. Balint, E. Fogassy, L. Novak and L. Poppe,
Tetrahedron: Asymmetry, 1998, 9, 271.
10 (a) F. Theil, J. Weidner, S. Ballschuh, A. Kunath and H. Schick, J. Org.
Chem., 1994, 59, 388; (b) L. Salazar, J. L. Bermudes, C. Ramirez, E. F.
Llama and J. V. Sinisterra, Tetrahedron: Asymmetry, 1999, 10, 3507.
11 A. Collet, J. Jacques and M. J. Brienne, Chem. Rev., 1980, 80, 215.
12 D. Gernez, Compt. Rend. Acad. Sci., 1866, 63, 843.
13 K. Kitaori, Y. Furukawa, H. Yoshimoto and J. Otera, Tetrahedron, 1999,
55, 14381.
14 P. H. J. Carlsen and K. Aase, Acta Chem. Scand., 1993, 47, 737.
(R)-1–3
Scheme 2
We have found that mephenesin, guaifenesin and o-chloro-
phenyl derivative 3 are really capable of spontaneous resolution
upon crystallization. Table 2 gives an example of successful
resolution by the entrainment of mephenesin 1. Immediately in
the course of spontaneous resolution, only a moderate degree of
optical purity can be reached. But the sample of high optical
purity can be obtained after single recrystallization of the one-
cycle or overall crop of crystals formed by molecules of the same
configurations. Similar results have been obtained for guaifenesin
2 and chloro derivative 3.
Table 2 Resolution by the entrainment of rac-mephenesin 1 (165 ml of
H2O; crystallization temperature of 33 °C).
Resolu-
tion
time/
min
(R)-1
(S)-1
Addi- Addi-
Entry tional tional Seed/g
rac-1/g (R)-1/g
Yield/g op (%) Yield/g op (%)
1
2
3
4
5
6
4.50
0.86
0.82
0.90
0.92
1.10
0.50
(R),
0.025
(S),
0.025
(R),
0.025
(S),
0.025
(R),
0.025
(S),
0.025
(R),
60
40
60
40
50
30
0.86
0.90
1.10
2.86
87.7
87.3
80.9
84.9
0.82
0.92
86.1
86.3
0.88
2.62
86.1
86.2
Overall 9.10
0.50
0.075
(S),
0.075
Received: 9th April 2003; Com. 03/2078
– 105 –