Salts of Mosher's thioacid: agents for determining the enantiomer excess of SN2 substrates

Article abstract of DOI:10.1016/j.tetlet.2010.03.041

The racemic and the (S)-enantiomer of Mosher's thioacid, 2-methoxy-2-trifluoromethylphenylacetic thioacid, form air-stable salts with Proton Sponge [1,8-bis(dimethylamino)naphthalene]. These salts are powerful nucleophiles that react cleanly (S_N2 inversion) in CDCl₃ with optically active alkyl halides ranging in reactivities from unactivated alkyl bromides and iodides to benzylic bromides. The diastereomeric excess (de) of the thioester products indicates the enantiomeric excess (ee) of the starting alkyl halides.

Full text of DOI:10.1016/j.tetlet.2010.03.041

Tetrahedron Letters 51 (2010) 2793–2796
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Salts of Mosher’s thioacid: agents for determining the enantiomer excess
of SN2 substrates
*
Jack E. Richman
Biochemistry, Molecular Biology and Biophysics, University of Minnesota, Gortner Laboratory, 1479 Gortner Ave., Saint Paul, MN 55108, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
The racemic and the (S)-enantiomer of Mosher’s thioacid, 2-methoxy-2-trifluoromethylphenylacetic
thioacid, form air-stable salts with Proton Sponge [1,8-bis(dimethylamino)naphthalene]. These salts
are powerful nucleophiles that react cleanly (SN2 inversion) in CDCl3 with optically active alkyl halides
ranging in reactivities from unactivated alkyl bromides and iodides to benzylic bromides. The diastereo-
meric excess (de) of the thioester products indicates the enantiomeric excess (ee) of the starting alkyl
halides.
Received 1 February 2010
Revised 11 March 2010
Accepted 12 March 2010
Available online 17 March 2010
Ó 2010 Elsevier Ltd. All rights reserved.
Occasionally in the accounts of science, modifications and
extensions of the art that would obviously be useful are over-
looked. As time passes, the very obviousness of such modifications
can deter further study—presumably the result of our tendency to
think that seemingly obvious things must have been tried and
must have encountered problems. This reasoning can be faulty.
This report examines such a case: the utility of Mosher’s thioacid.
Mosher’s acid,1 2-methoxy-2-trifluoromethylphenylacetic acid
(1), and the corresponding acid chloride (2), have found
extensive use as agents for determining the enantiomeric excess2a
(ee) of amines and alcohols and for predicting their absolute con-
figurations.2b,c
(Scheme 1) gives a practical laboratory synthesis of Mosher’s thio-
acid, which easily operates even on very small scale. Workup using
an excess of aqueous ammonia3a and then acid produces optically
active Mosher’s thioacid with retention of configuration and high
chemical and optical purities in high yield.7
The nucleophilicities of thioacids (pKa ꢀ3.5) are activated by
conversion to their salts, thiocarboxylate ions. We have found that
either the racemic or resolved salts, 5, of Mosher’s thioacid neutral-
ized with Proton Sponge [1,8-bis(dimethylamino)naphthalene,
Scheme 2] are readily recrystallized from minimal ethanol produc-
ing colorless shiny non-hygroscopic crystals that are remarkably
stable in air and soluble in chloroform-d and many other organic
solvents.8
H3CO
CF3
In general, even as a dilute solution in chloroform, the thiocarb-
oxylate anion of 5 is highly reactive in SN2 reactions. Because SN2
reactions normally occur with clean inversion at the reaction cen-
ter, resolved 5 can provide information about the configuration and
enantiomeric excess of chiral substrates.
X
1, X = OH
2, X = Cl
3, X = SH
C
O
C6H5
Considering this, and the well-known potent nucleophilicity of
thiocarboxylate ions,3 it is surprising that the thioacid analogue of
Mosher’s acid has not previously been reported.4 We have found
that thioacid 3, referred to here as Mosher’s thioacid, is readily pre-
pared and easily forms stable salts with organic bases. Unlike salts
of Mosher’s acid, the thioacid salts are powerful nucleophiles,
useful for accurately detecting the ee of small amounts of chiral
SN2-reactive compounds, possibly not thermally stable or readily
resolved by chiral chromatography.
Mosher’s thioacid, 3, can be prepared by treating the acid chlo-
ride, 2, with hydrogen sulfide.5 But this approach is not convenient
for a laboratory synthesis. Shin and Quinn6 describe the prepara-
tion of fatty thioacids from fatty acid chlorides using thioacetic
acid as a convenient carrier of hydrogen sulfide. This method
H3CO
C6H5
H3CO
C6H5
CF3
CF3
C
X
S
AcS−K+
THF
C
O
Ac
O
(S)-4
(S)-2, X = Cl
(R)-1, X = OH
1) NH4OH
2) H3O+
H3CO
C6H5
(S)-3
CF3
SH
C
O
* Tel.: +1 612 624 4278; fax: +1 612 625 5780.
E-mail address: richm005@umn.edu
Scheme 1. Synthesis of Mosher’s thioacid, 3.
0040-4039/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2010.03.041

2794
J. E. Richman / Tetrahedron Letters 51 (2010) 2793–2796
O
H3C
These benzylic bromides react with (S)-5 as dilute solutions in
CDCl3 at room temperature to cleanly produce SN2 products, the
benzylic thioesters 8 (Scheme 2 and Table 1 entries 1-1 and 1-4).
Elimination side reactions are insignificant and racemization is
minimal, especially in the early stages of these reactions.11 Moni-
toring by NMR12 or GC13, reagent (S)-5 readily determined the ee
of benzylic bromides 7 from the diastereomeric excess (de) of
the thioester products, 8.14
H
O
_
H3C
Ar
CF3
H
CF3
S
CDCl3
+
Ar
S
C6H5
Br
RT
C6H5
OCH3
H
OCH3
(R)-7a
(R)-7b
+
(S,S)-8a, de ~ 48%
(S,S)-8b, de = 98.8%
N
N
This analysis showed a surprising result. The de for each of the
two thioester products 8 was dramatically different. Both benzylic
bromides 7a and 7b were prepared from the corresponding (S)-
alcohols of high optical purity. Both preparations used conditions
essentially identical to those described by Stein.10 The results show
two very different ee values for the similar bromides 7a and 7b. At
least in our hands, the stereochemical integrity of Stein’s method
differed markedly even in these two closely related cases: clean
inversion for 7b (entry 1-4) and major racemization for 7a (entry
1-1). This difference was readily detected by reagent (S)-5.15
SN2 reactions of 5 with even more reactive benzylic methane-
sulfonate esters (entry 1-2) and also reaction (of 3) with benzylic
alcohols activated under Mitsunobu conditions16 (entry 1-3) are
compromised by competing reactions and racemization. This was
apparent from the time course of these reactions followed by GC
and NMR analyses: the de of thioester products 8 decreased signif-
icantly as these reactions proceeded.11b
Unactivated alkyl halides and methanesulfonate esters are
much less reactive SN2 substrates than benzylic bromides.9 Sur-
prisingly, the thiocarboxylate anion of 5 remains quite SN2-reactive
with even unactivated secondary alkyl iodides. Unactivated bro-
mides and methanesulfonate esters are considerably less reactive.
Chlorides are nearly inert (see Table 2).8c This order of reactivity
suggests an effect of hard and soft acid/base interactions.
a, Ar = 2-Naphthyl
b, Ar = 2-Bromophenyl
(S)-5
Scheme 2. Reactions of benzylic bromides with (S)-5.
As an indication of the range of reactivity of thiocarboxylate 5,
the following examples (collected in Table 1) show reactions with
different SN2 substrates. These range from more sluggish unacti-
vated primary and secondary alkyl halides to much more reactive
benzylic halides. On an approximate scale of SN2 reactivities,9
many classes of substrates normally exhibit intermediate reactivi-
ties and are also expected to show reactivities that are suitable.
These classes include, for example, primary and secondary halides
on sp3 carbons that are allylic, propargylic or alpha to carbonyl,
ether or amine groups.
Initially we chose to study the reactions of 5 with optically ac-
tive benzylic bromides. Fortuitously, this turned out to be a good
choice because these substrates appear to define the upper limit
of reactivity for clean SN2 reactions. Also, this demonstrated that
reagent 5 is excellent for detecting subtle differences in similar
substrates.
Benzylic bromides 7a and 7b were prepared from the (S)-ben-
zylic alcohols 6 by the diphos bromide method (Eq. 1).10
Ph2P-(C2H4)-PPh2
ArCH(OH)CH3
ArCH(Br)CH3
The methanesulfonate esters of the chiral alcohols (R)-2-buta-
nol and (S)-2-methyl-1-butanol were prepared (Scheme 3). Reac-
tion of (R)-2-butyl methanesulfonate with (S)-5 proceeded slowly
ð1Þ
CH2Cl2 + Br2
(S)-6a
(S)-6b
(R)-7a
(R)-7b
a, Ar = 2-Naphthyl
b, Ar = 2-Bromophenyl
Table 1
Reaction of thiocarboxylate (S)-5 with various chiral SN2 substrates in CDCl3 at rt
H
H3CO
N+
N
CF3
R2
CF3
R1
H3CO
CDCl3
+
S-
(S)-5
H
H
X
S
R1
O
R2
O
Entry
R1
R2
Substrate reacted with (S)-5
Primary SN2 product
X
Preparationa
Precursorb
½
a D
ꢁ
Stereochem
Stereochem
de (%)
1-1
1-2
1-3
1-4
1-5
1-6
1-7
1-8
CH3
2-Naphthyl
CH3
CH3
2-Bromophenyl
H
C2H5
C2H5/CH3
CH3
Br
1
2
3
1
4
5
4
6
(S)-Alc
(S)-Alc
(S)-Alc
(S)-Alc
(S)-Ms
(R)-Alc
(R)-Ms
(R)-Alc
ꢂ+23
(R)
(S)
(S)
(R)
(S)n
(R)
(RS)
(S)
(S,S)
(S,R)
(S,R)
(S,S)
(S,S)
(S,S)
(S,RS)
(S,R)
P47c, P48d
85–95c,e,f
94d,f
2-Naphthyl
2-Naphthyl
CH3
(S)-2-Methylbutyl
CH3
OMs
—
Mit.g
—
Br
I
OMs
I
I
ꢃ48
98.8d
+5.5, neat
90–100h,i
94i,j,k
—
—
CH3/C2H5
C2H5
ꢂ0i,j,l
+20, neat
50m
a
Substrate preparation conditions: (1) Diphos., 2Br2, 0 °C, CH2Cl2; (2) MsCl/Et3N, 0 °C, THF; (3) Mitsunobu, CDCl3; (4) NaI, acetone,
neat,
D; (5) MsCl/Et3N, 0 °C, CH2Cl2; (6) P/I2,
D
.
b
Substrate precursor stereochemistry: Alc = alcohol, Ms = mesylate.
c
d
e
f
By 19F NMR.
By GC/MS.
Extrapolated to t = 0 to minimize effects of competing racemization.
Significant unidentified reactions interfered.
Mitsunobu-activated complex.
g
h
i
Estimated by 1H NMR, which is poorly resolved.
SN2 reaction was run as a highly concentrated solution.
By FID-GC.
j
k
l
The starting (R)-2-butanol is about 95% optically pure.
The starting (R)-2-butyl methanesulfonate has ee 94%—see entry 1-6.
The ee for precursor (S)-2-iodobutane was 50% by chiral GC.
Stereochemistry is not altered.
m
n

J. E. Richman / Tetrahedron Letters 51 (2010) 2793–2796
2795
Table 2
early work done in the laboratory of the late Professor William E.
Truce at Purdue University.
SN2 chemistry of 5 in CDCl3 with achiral substrates
Substrate
Resulta
Supplementary data
n-C6H13Cl
Large excess ? slow pseudo first order reaction kinetics,
t1/2 ꢂ22 h at 37 °C
(CH3)2CHCl
C2H5OCH2Cl
CH2Cl2
CDCl3
CH3I
No reaction even in large excess
Supplementary data (experimental details, the IR spectra and
1H, 19F, and 13C NMR spectra for 3 and 5, and representative
NMR spectra for 8) associated with this article can be found, in
the online version, at doi:10.1016/j.tetlet.2010.03.041.
Rapid SN2 reaction. Can be used to quench 5
Slow sequential reactions (?diastereomers)
No reaction
Rapid SN2 reaction. Can be used to quench 5
a
Room temperature, except for first entry.
References and notes
1. Dale, J. A.; Dull, D. L.; Mosher, H. S. J. Org. Chem. 1969, 34, 2543–2549.
2. (a) Jacques, J.; Collet, A.; Wilen, S. H.. In Enantiomers, Racemates, and
Resolutions; Wiley: New York, 1981 p. 335; for advanced Mosher method
see: (b) Ohtani, I.; Kusumi, T.; Kashman, Y.; Kakisawa, H. J. Am. Chem. Soc. 1991,
113, 4092–4096; (c) Porto, S.; Seco, J. M.; Ortiz, A.; Quinoa, E.; Riguera, R. Org.
Lett. 2007, 9, 5015–5018.
H3CO
CF3
O
(S)-5
X
S
C6H5
(CH2)
n
(S), X= MsO or I
3. (a) Strijtveen, B.; Kellogg, R. M. J. Org. Chem. 1986, 51, 3664–3671; (b)
McFadden, J. M.; Frehywot, G. L.; Townsend, C. A. Org. Lett. 2002, 4, 3859–3862.
4. For resolved camphanic thioacid, see: de March, P.; Figueredo, M.; Font, J.;
Gonzalez, L.; Salgado, A. Tetrahedron: Asymmetry 1996, 7, 2603–2606.
5. For examples of thioacids prepared from hydrogen sulfide and acid chlorides
see: (a) Ellingboe, E. K. Org. Synth. 1963, CV 4, 928–931. and references therein;
(b) Loeliger, P.; Flükiger, E. Org. Synth. 1988, CV 6, 776–780. see note 14 on page
779.
(S,S)-11, n = 1
(S,S)-10, n = 0
OMs
(R)
(S)-5
Finkelstein
(RS)-2-Iodobutane
(S)-5
(S)-5
(S,RS)-thioester
(S,R)-thioester
P/I2
(R)-2-Butanol
(S)-2-Iodobutane
6. Shin, H.-C.; Quinn, D. M. Lipids 1993, 28, 73–74.
7. Details for preparation and properties of (S)-3 [90% yield from (R)-1] and
comments about (RS)-3, (RS)-5 and (R)-5 are presented in the Supplementary
data.
Scheme 3. Reactions of (S)-5 with unactivated substrates.
8. (a) Elemental analyses (C, H, N, S) were acceptable ( 0.3%) after a single
recrystallization; (b) salts 5 are moderately photolabile; (c) dichloromethane
reacts slowly.
9. For an approximate scale of reactivities of SN2 substrates, see: Smith, M. B.;
March, J. March’s Advanced Organic Chemistry, 5th ed.; Wiley: New York, 2001, p
439.
but smoothly in CDCl3 and indicated that the optical purity of the
starting alcohol (ca. 95%) was preserved (entry 1-6). Both methane-
sulfonate esters were then converted to the corresponding alkyl io-
dides using Finkelstein conditions. As is expected, the reaction of
(S)-5 with (S)-2-methyl-1-iodobutane, where the SN2 reaction cen-
ter is adjacent to the chiral center, showed that the integrity of the
(S)-center was preserved (entry 1-5).
10. Stein, A. R. Can. J. Chem. 1994, 72, 1789–1796.
11. (a) A comparison with the much less reactive Na+ salt of Mosher’s acid is given
in the Supplementary data; (b) methodologies described in the Supplementary
data and note 13 below minimize effects of racemization and compensate for
kinetic sorting and detector response factors.
12. Short approximate NMR analysis of 7b: 6.7 mg (14 lmol) of (S)-5 and 3.5 mg
In marked contrast, the reaction of thiocarboxylate (S)-5 with 2-
iodobutane prepared from the (R)-methanesulfonate ester (entry
1-7) clearly indicated that this iodide was racemic—racemized by
the standard Finkelstein reaction conditions.
That reagent (S)-5 successfully detected the optical purity of 2-
iodobutane was demonstrated by reaction with (S)-2-iodobutane
prepared otherwise17 (entry 1-8). In this case polarimetry failed
to determine the ee of 2-iodobutane.18 However, chiral GC did suc-
cessfully separate the enantiomers19 and confirm the results ob-
tained with (S)-5.
Further observations on the reactivities of 5 are shown in Table
2. 1-Chlorohexane, in high excess, reacts only slowly following
pseudo first order kinetics. Chloroform and 2-chloropropane are
essentially inert. Dichloromethane, however, is slowly reactive
and proceeds to mono and disubstituted products at long reaction
times. At the opposite extreme, methyl iodide and chloromethyl
ethyl ether react rapidly and can be used to quench the reactions
of 5.
In conclusion, salts of Mosher’s thioacid, (S)-3, and in particular
the salt (S)-5 formed with Proton Sponge, are effective nucleo-
philes that cleanly react with SN2 substrates ranging in reactivities
from unactivated alkyl bromides, iodides, and mesylates to ben-
zylic halides. We expect that Mosher’s thioacid and corresponding
salts will find a wide range of uses.20
(13 lmol) of (R)-7b in 0.75 mL of CDCl3 showed no remaining 7b after 6 days
(rt). 19F NMR integration of singlets for (S,S)- and (S,R)-8b at ꢃ69.49 and
ꢃ69.30 ppm show the de ꢂ96%. More accurate GC integrations, uncorrected for
differences in detector response, show de P 93.4. Racemization, especially
near completion, sets the lower limit to de.
13. Accurate GC analysis of 7b: A mix of 1.8 mg (7
lmol) of (R)-7b and 7.5 mg
(16.1 mol) of (S)-5 in 0.1 mL CDCl3 was quenched (CH3I) 2 min after mixing
l
(rt). GC shows de = 98.5%, but this result is not corrected for kinetic sorting and
detector response factors. GC analysis of an analogous experiment using (RS)-
7b and (S)-5 (or vice versa) shows the diastereomeric ratio [(S,S)/(S,R)] is 0.80
indicating that the corrected de above is 98.8% early in the reaction when
racemization is insignificant.
14. (a) Assigning the ee value for the substrate (7) as being equivalent to the de
values of substitution products (8) rests on the assertion that the ee of the
nucleophile is high (>>99%) and that the stereointegrity of the SN reactions and
the stereo stabilities of the reactants and products are high; (b) the
configuration of the
a-carbon of Mosher’s acyl group is listed first followed
by the configuration of the thiol chiral carbon.
15. Confirmation of the low optical purity of benzylic bromide 7a came from two
additional observations. First, recrystallization of 7a of optical rotation +23
(first preparation) or +37 (second preparation) increased the rotation to a
maximum value of +64.3. Analysis of this purified 7a by reaction with (S)-5
indicated that the ee still had not reached 100%, only 94%. Second, another
derivative, benzyl 1-(2-naphthyl)ethyl sulfide (9, reported by Givens, et al. J.
Am. Chem. Soc. 1984, 106, 1779–1789), was prepared from 7a of optical
rotation +23. On recrystallization, the optical rotation of 9 increased 2.36
times. Both these polarimetry results indicate that the initial ee of 7a was low,
approximately 0.4 but are lower limits set by the unknown purities for both
crude 7a and 9. Note also that observations reported in the Supplementary
data indicate that both (RS)-7a and 9 are examples of racemic conglomerates.
See Ref. 2a, p 7.
16. See, for a closely related example: Volante, R. P. Tetrahedron Lett. 1981, 22,
3119–3122.
Acknowledgments
17. (a) Kornblum, N.; Patton, J. T.; Nordmann, J. B. J. Am. Chem. Soc. 1948, 70, 746–
749; (b) Clarke, L. J. Am. Chem. Soc. 1908, 30, 1144–1151.
18. Results suggesting that ½
a 2D7
ꢁ +40 for optically pure (S)-2-iodobutane are
The author thanks Professor Larry P. Wackett for his support
during the course of this work. Inspiration came from Professor
D. Y. Curtin (University of Illinois) and (for sulfur chemistry) from
presented in the Supplementary data.
19. Chiral GC on 2-iodobutane: Miranda, E.; Sanchez, F.; Sanz, J.; Jimenez, M. I.;
Martinez-Castro, I. J. High Resolut. Chromatogr. 1998, 21, 225–233. Note also

2796
J. E. Richman / Tetrahedron Letters 51 (2010) 2793–2796
that GC separation of the (S,R)- and (S,S)-thioesters of 10 [prepared from the
(RS)-thiol] was previously reported: Kukula, P.; Dutly, A.; Sivasankar, N.; Prins,
R. J. Catal. 2005, 236, 14–20.
151: 173065 (2009)]. This patent position is not intended to restrict
noncommercial research use, but will be used to preclude gifts, purchases,
sales or commercial use of the compounds by anyone other than authorized
licensees.
20. Claims to Mosher’s thioacids, their salts, and related compounds are pending
in US Patent Application 2009/0181462, by Jack E. Richman [Chem. Abstr.