Tartaric acid and its O-acyl derivatives. Part 14: Nucleophilic ring-opening reaction of nonsymmetrically substituted tartaric acid anhydride as a tool for the synthesis of totally differentiated tartaric acid derivatives

Article abstract of DOI:10.1016/j.tet.2015.04.063

The ring-opening reaction of nonsymmetrically substituted tartaric acid anhydride was used to synthesize monoamides and monoesters of O-benzoyltartaric acid, type I (a) and II (b) building blocks with all four functional groups differentiated. The correct

Full text of DOI:10.1016/j.tet.2015.04.063

Tetrahedron 71 (2015) 4047e4052
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Tartaric acid and its O-acyl derivatives. Part 14: Nucleophilic
ring-opening reaction of nonsymmetrically substituted tartaric acid
anhydride as a tool for the synthesis of totally differentiated tartaric
acid derivatives
*
ꢀ
Urszula Bernas, Halina Hajmowicz, Ludwik Synoradzki
Warsaw University of Technology, Faculty of Chemistry, Laboratory of Technological Processes, ul. Noakowskiego 3, 00-664 Warsaw, Poland
a r t i c l e i n f o
a b s t r a c t
Article history:
The ring-opening reaction of nonsymmetrically substituted tartaric acid anhydride was used to syn-
thesize monoamides and monoesters of O-benzoyltartaric acid, type I (a) and II (b) building blocks with
all four functional groups differentiated. The correct structure of the regioisomers, which had earlier
been misassigned, was established. The type Ietype II regioisomer rearrangement, which proceeded via
acyl migration, was examined. Moreover, it was shown that acyl migration induced by selective crys-
tallization may yield one of the regioisomers quantitatively.
Received 11 August 2014
Received in revised form 8 April 2015
Accepted 20 April 2015
Available online 25 April 2015
Keywords:
Acyl migration
Ó 2015 Elsevier Ltd. All rights reserved.
Desymmetrization
Ring-opening reaction
Tartaric anhydride
1. Introduction
independent modification of the carboxy and the hydroxy group, its
high toxicity limits its use.
Tartaric acid (1) plays an important role in all areas of
chemistry involving chirality.1,2 This four-carbon molecule pos-
sesses two chiral centers and two pairs of functional group-
sehydroxylic and carboxylicearranged in such a way that the
compound has C2 symmetry. This unique structure makes tar-
taric acid a valuable starting material but also complicates its
desymmetrization.3e5 Regioselective modifications of the hy-
droxy and carboxy groups, especially those leading to totally
differentiated tartaric acid derivatives with all four groups
functionalized, remains a challenge.
A literature search for evidence of type I and type II building
blocks revealed that tartaric acid derivatives bearing four distinct
substituents are very scarce.5e8 Bell obtained a few O-acyltartaric
monoamides via the aminolysis of the corresponding diacyltartaric
anhydride or amide followed by the deprotection of one hydroxy
group; however, the structure of the resulting compounds was
misassigned.7,8 Alternatively, one acyl function was cleaved by
rapid basic methanolysis.7 Recently, a strategy involving the use of
hexafluoroacetone (HFA) to enable the formation of O-protected
monotartrates was described.5 Although HFA allows for the
While working on the synthesis of compounds applicable as
biopolymers, we required a simple, scalable and ecologically viable
process for the preparation of totally differentiated tartaric acid
derivatives. Because the distinction between carboxylic groups can
be directly accomplished by the ring opening of diacyltartaric1,9a,b
or succinic anhydrides,10 a strategy involving the use of a five-
membered cyclic anhydride for both 2,3- and 1,4-
desymmetrization, similar to the application of O-acylmalic
anhydride,11aee was very promising. Thus, we chose an O-ben-
zoyltartaric acid anhydride (3) with differentiated 2,3-hydroxy
groups as a
starting material12,13 to achieve further 1,4-
desymmetrization of the tartaric unit, providing the desired tar-
taric acid derivatives with four distinct substituents (4e9) (Fig. 1).14
2. Results and discussion
We examined the regioselectivity of the ring-opening of anhy-
dride 3 by nitrogen and oxygen nucleophiles (Scheme 1, Table 1).
We found that O-protected tartaric anhydride 3 reacted in a non-
regioselective manner. The reaction proceeded with the formation
of the corresponding O-benzoyltartaric monoamides (4e6) and
monoesters (7e9) wherein two regioisomers, types I (a) and II (b),
were formed, depending on the carbonyl center attacked by the
nucleophile (Scheme 1). We also observed that, in contrast to the
* Corresponding author. E-mail address: synsa@ch.pw.edu.pl (L. Synoradzki).
http://dx.doi.org/10.1016/j.tet.2015.04.063
0040-4020/Ó 2015 Elsevier Ltd. All rights reserved.

ꢀ
4048
U. Bernas et al. / Tetrahedron 71 (2015) 4047e4052
Fig. 1. An approach to the totally differentiated tartaric acid derivatives, monoamides (4e6) and monoesters (7e9) of O-acyltartaric acid.
Scheme 1. Regioselectivity of the ring-opening of anhydride 3 by nitrogen and oxygen nucleophiles.
Table 1
Table 2
Regioselectivity in the ring opening of O-benzoyltartaric anhydride (3) by
nucleophiles
The ratio of regioisomers (type I/type II)a of monobenzoyl monoamides (4e6) and
monoesters (7e9) formed in the ring opening of 3 by nucleophiles according to
reaction conditions ib or iic
Compounda
Nucleophiles XeH
Regioisomer I, a
Regioisomer II, b
Compound
X
Procedure ib
Procedure iic
X
(%) HPLCb
Type I,
a (%)
Type II,
b (%)
Type I,
a (%)
Type II,
b (%)
4
5
6
7
8
9
NH2
73
60
39
60
90
70
27
40
61
40
10
30
NHCH2Ph
(S)-NHCHMePh
OEt
OCH2Ph
OCH2CH2OMe
4
5
6
7
8
9
NH2
73
60
39
60
90
70
27
40
61
40
10
30
97
100d
94
40
48
3
d
6
60
52
51
NHCH2Ph
(S)-NHCH(Me)Ph
OEt
OCH2Ph
OCH2CH2OCH3
a
Conditions: anhydride 3 was dissolved in MeCN at rt followed by the addition of
1 equiv of the nucleophile, reaction was stopped rapidly by the acidification with
MeCOOH.
49
a
The regioisomer ratio was determined by the dissolution of the reaction mixture
in a water/MeCN/MeCOOH system and HPLC analysis.
b
The regioisomer ratio was determined by the dissolution of the reaction mixture
b
in a water/MeCN/MeCOOH system and HPLC analysis.
Conditions: 1 equiv of amine XeH, rt, 5 s or 1 equiv of alcohol XeH, rt, 1 day.
c
Conditions: 3 equiv of amine XeH, rt, 7 days or 1 equiv of alcohol, addition of
10 mol % of benzylamine, 50 ꢀC, 2e4 days.
d
Crystallization occurred.
nucleophilic ring opening of malic anhydride,11aef the major
regioisomer formed was usually the product of the reaction at the
C-3 site of tartaric anhydride 3 (type I regioisomer, (a)).
Despite a lack of regioselectivity we found that the compound
crystallizing from the reaction mixture in each case was the
regioisomer of type I. Thus, we synthesized several monoamides
(4ae6a) and monoesters (7ae9a) of monobenzoyl-L-tartaric acid
(type I, yield 22e65%) utilizing 3 as a starting material in a very
simple procedure.
It was much more difficult to obtain type II building blocks with
both hydroxy and carboxy functionalities attached to the same C3
carbon. Despite considerable efforts to isolate high-purity type II
regioisomers, we only succeeded with derivatives of (S)-2-
phenylethylamine (6b) and 2-methoxyethanol (9b). For other
substituents, both multiple recrystallizations and column chro-
matography failed, most likely due to the higher polarity of the type
II building blocks compared to those of type I.
behavior, we decided to perform the nucleophilic ring opening of
benzoyltartaric anhydride (3) using two procedures (Table 2).
The first procedure was rapid (Table 2, Procedure i) and entailed
the use of stoichiometric amounts of the nucleophile and sub-
sequent acidification; therefore, the results revealed the initially
formed product ratio obtained just after the ring-opening reaction
reached completion. The second method (ii) (Table 2, Procedure ii)
involved the use of an excess of amine or the addition of catalytic
amine in the case of monoesters (7e9) and prolonged reaction
times or heating. We observed that the product ratio differed be-
tween that measured directly after ring opening and those obtained
after extended reaction, namely an excess of amine-mediator. We
assumed that it was a result of interconversion between type I and
type II regioisomers due to the intramolecular acyl migration,
which most likely occurs via a five-membered transition state
(Scheme 2). This acyl migration phenomenon is encountered in
carbohydrate chemistry and may significantly disturb the selected
synthetic route.15aed
During attempts to produce monoamide 5b, we noticed that the
ratio of regioisomers formed after the reaction of 3 with benzyl-
amine was dependent on the reaction conditions. To explain this

ꢀ
U. Bernas et al. / Tetrahedron 71 (2015) 4047e4052
4049
Scheme 2. Plausible mechanism of intramolecular acyl migration via a five-membered transition state.
As it was revealed that solvent may play a significant role in the
acyl migration process, we analyzed equilibrium mixtures of type I
(5a) and type II (5b) N-benzyltartramides in two different solvents
(Table 3).
presence of a catalytic amount of acid rearrangement did not
occur (Table 4, column E). To the best of our knowledge this is the
first evidence that acidic conditions may prevent unwanted acyl
migration.
Table 3
Influence of the crystallization of 5a on the equilibrium between 5a and 5b
Entriesa
Temperature (ꢀC)
Time
MeCN
DMF
Type I, 5a %HPLC
Type II, 5b %HPLC
Type I, 5a %HPLC
Type II, 5b %HPLC
1
2
3
rt
25
80
5 s
7 days
20 min
61
100b
100b
39
0
0
69
82
70
31
18
30
a
Ring-opening reaction of anhydride 3 with 3 equiv of benzylamine.
Crystallization occurred.
b
We noticed relatively small changes in product ratios obtained
Table 4
Influence of the conditions on the transformation of monobenzoyl monoamides (5a
under ‘kinetic’ conditions in acetonitrile versus dimethylforma-
mide (61:39 vs 69:31, Table 3, entry 1). In contrast, we observed the
large changes in regioisomer ratios after extended reactions (pro-
longed reaction time or heating). In DMF solution, where no crys-
tallization occurred, ratio of 5a to 5b slowly equilibrated to 82:18
(entries 2 and 3). In contrast, in the case of MeCN preferential
crystallization from the reaction mixture of 5a entailed further
benzoyl migration and the quantitative transformation of the
mixture of type I and II regioisomers into the type I regioisomer
(5a), (100:0). We assumed it was a consequence of the influence of
the crystallization of 5a on the equilibrium between 5a and 5b. We
found it very promising that acyl migration induced by selective
crystallization may yield one of the regioisomers quantitatively.
Further crystallization tests of the other compounds are in progress.
An opposite approach was to prevent the uncontrolled migra-
tion of the benzoyl group, so we investigated the effect of the re-
action conditions on the transformation of regioisomers (Scheme 3,
Table 4, Fig. 2). The studies were conducted in DMSO-d6, which
similarly to DMF provided regioisomers dissolution avoiding un-
wanted crystallization.
to 5b) and monoesters (7a to 7b) observed by 1H NMR spectroscopy
Additive
A.
B.
Benzylamine No
7a/7b 7a/7b
C.
D.
E.
No
Benzylamine AcOH
Entrya Timeb Tempc 5a/5b 5a/5b
(%)d
(%)
5a/5b
(%)
(%) (%)
1
2
3
4
5
6
7
0
rt
rt
rt
100/0 100/0
100/0 97/3
100/0 100/0
100/0 95/5
100/0
100/0
100/0
100/0
98/2
1d
4d
7dc
2w
3w
2m
99/1
95/5
91/9
79/21
98/2
88/12
50 ꢀC
89/11 51/49
86/14 50/50
84/16 50/50
60/40 50/50
50 ꢀ
50 ꢀ
50 ꢀ
C
C
C
88/12 78/22
85/15 78/22
84/16 78/22
98/2
98/2
a
The starting regioisomer 5a or 7a was dissolved in DMSO-d6 and 10 mol % of
PhCH2NH2 (B or D) or AcOH (E) was added if selected.
b
d¼days, w¼weeks, m¼months.
c
After 4 days at rt temperature was increased to 50 ꢀC.
d
The regioisomers ratio was determined based on 2a, 2b, 3a, 3b protons in-
tegration, see Scheme 3.
Scheme 3. The effect of the reaction conditions on the transformation of regioisomers.
We observed that samples of 5a and 7a dissolved in DMSO-d6
slowly interconverted to 5b and 7b regioisomers. The process
was slightly accelerated by the temperature increase and signif-
icantly, by the addition of base (benzylamine), (Table 4, Fig. 2, B
and D).16 The most important was that we found that in the
The important issue was the elucidation of the correct structures
of the synthesized monoacyl monotartramides (4e6) and mono-
tartrates (7e9), thus standard spectroscopic techniques (1H NMR,13
C
NMR, IR) were useless for identifying type I or type II regioisomers.
For N-benzyltartramide 5a we used X-ray analysis of the crystal

ꢀ
U. Bernas et al. / Tetrahedron 71 (2015) 4047e4052
4050
Fig. 2. The effect of the reaction conditions on the transformation of regioisomers in DMSO-d6 solution (based on 2a, 2b, 3a, 3b protons integration by 1H NMR, see Scheme 3,
Table 4).
structure.17 For the benzoyltartaric N-phenylethylamide (6a) and
other compounds, the structure was assigned by the interpretation
of the HMBC correlations from the protons attached to the chiral
carbon atoms in the tartaric backbone (H-2 and H-3) to the carbonyl
carbons of carboxylic (C-1), benzoyl (C-5) and amide (C-4) or ester
functional groups (Fig. 3) (see also Supplementary data).
benzoyl and amide functional groups attached to different carbons
(see Supplementary data).
Our structural assignments were inconsistent with earlier data,
which proposed type II structures,7,8 with both hydroxyl and car-
boxyl groups attached to the same carbon, for compounds 5a and
6a characterized using the same mp., 1H NMR, and [
a]D. As the type
First, we assigned 13C NMR chemical shifts to the corresponding
carbonyl carbons based on the HMBC correlations from H-7 and H-
8 to C-5 and H-16 to C-4. Next, the correlations from H-2 to C-5, H-2
to C-1 and H-3 to C-4 indicated the type I regioisomer (6a) with the
II regioisomer structure were not confirmed by any extended
analyses (e.g., X-ray or correlation spectra), we assumed that
structures of 5a and 6a were originally misassigned, as our results
indicated that they are type I building blocks.

ꢀ
U. Bernas et al. / Tetrahedron 71 (2015) 4047e4052
4051
Fig. 3. Key HMBC signals for the structural assignment of compound 6a.
3. Conclusion
After the reaction completion, the mixture was acidified with 10%
HCl and filtered. The resulting white solid was rinsed with water
and dried under reduced pressure. For compounds 4a and 6a, the
crude products were recrystallized from MeOHewater.
We reported the ring opening of unsymmetrical tartaric anhy-
dride 3 as a method for obtaining totally differentiated tartaric
derivatives monoamides (4e6) and monoesters (7e9) of mono-
benzoyltartaric acid (type I, yield 22e65%). The regioselectivity of
the reaction was examined, and the phenomenon of regioisomer
interconversion was reported and analyzed. Moreover, the solvent
influence on the equilibrium mixtures was revealed and benzoyl-
tartaric N-benzylamide (5a) was used to show that acyl migration
induced by selective crystallization may yield one of the
regioisomers quantitatively. Some general assumptions have been
made, namely that acidic conditions may prevent uncontrolled acyl
migration. The structures of the originally misassigned monoacyl
monotartates and monotartramides were verified.
4.3.1. Amide of O-benzoyl- -tartaric acid, type I (4a). Compound 4a
L
was obtained following general procedure A. Aqueous NH3 was
used (2.2 mL/1 mmol of 3), and the crude product was obtained
after concentration under reduced pressure. 2.7 g, 50% yield; white
25
crystals; mp 154e156 ꢀC [
(400 MHz, DMSO-d6)
a]
þ65.8 (c 1.0, EtOH); 1H NMR
D
d 13.33 (br, 1H), 7.50 (s, 2H), 8.02e7.53 (m,
5H), 6.20 (br, 2H), 5.47 (d, J¼2.4 Hz, 1H), 4.48 (s, 1H); 13C NMR
(100 MHz, DMSO)
d 172.6 (CONH2), 169.0 (COOH), 165.1 (PhCOO),
133.8, 129.6, 129.1, 128.8 (ArH), 74.0, 71.2 (CH); Anal. Calcd for
11H11NO6: C 52.18, H 4.38, N 5.53; Found: C 52.19, H 4.52, N 5.55;
FTIR (ATR) (
, cmꢂ1) 3643, 3252, 1706, 1659, 1253, 1222, 717, 681;
FTIR (EtOH) (
, cmꢂ1) 1732, 1717, 1684.
C
n
n
4. Experimental section
4.1. General
4.3.2. N-Benzylamide of O-benzoyl-
L
-tartaric acid, type
I
(5a).17 Compound 5a was obtained following general procedure A
in 4.7 g, 65% yield. The addition of benzylamine (73.5 mmol) and
All commercial reagents and solvents were used as received
unless otherwise indicated. TLC was performed with TLC aluminum
sheets coated with silica gel 60 RP-18 F254S. Spots were observed
with UV. HPLC: HP 1100 Agilent Technologies LiChrospher 100 RP-
a prolonged reaction time (7 days) resulted in 7.2 g, 100% yield;
25
white crystals; mp 192e194 ꢀC [
(400 MHz, DMSO-d6)
a
]
þ40.9 (c 1.0, EtOH); 1H NMR
D
d 13.98 (br, 1H), 8.69 (t, 1H), 7.99e7.42 (m,
5H), 6.91e6.81 (m, 5H), 6.42 (br, 1H), 5.53 (d, 1H), 4.62 (d, 1H),
18 (250 mmꢁ4 mmꢁ5
MeCN/(AcOH/H2O, 2 mL:1000 mL, pH¼3.2) (1:1), 1 mL/min.
GCeMS: Hewlett Packard HP 6890 HP-5MS
(30 mꢁ0.25 mmꢁ0.25
m), 100 ꢀCe320 ꢀC, 10 ꢀC/min, 1.2 mL/
mm) with UVevis detection (230 nm);
4.43e4.15 (m, 2H); 13C NMR (100 MHz, DMSO)
d
170.3 (CONH),
168.9 (COOH), 165.0 (PhCOO), 139.3, 133.7, 129.5, 129.1, 128.8, 127.9,
126.4 (ArH), 74.0, 71.3 (CH), 41.9 (CH2); Anal. Calcd for C18H17NO6: C
m
min, samples were derivatized with BSA. 1H NMR spectra were
acquired at 400.125 MHz, 13C at 100.625 MHz with TMS as internal
reference. IR spectra were performed using BRUKER FTIR ALPHA
with ATR module. The following abbreviations are used to indicate
multiplicity: s, singlet; d, doublet, dd, double doublet; t, triplet; dt,
62.97; H 4.99; N 4.08; Found: C 62.92, H 4.99, N 4.11; FTIR (ATR) (
n,
cmꢂ1) 3314, 1729, 1706, 1264, 1239, 708, 693.
4.3.3. N-(2-(S)-Phenylethylamide) of O-benzoyl- -tartaric acid, type I
L
(6a). Compound 6a was obtained following general procedure A.
1.7 g, 22% yield; white crystals; mp 204e206 ꢀC [
a
]
25 þ29.3 (c 1.0,
D
double triplet; m, multiplet, br, broad signal. Chemical shifts (d) are
expressed in parts per million downfield from tetramethylsilyl
chloride.
EtOH). 1H NMR (400 MHz, DMSO-d6)
d
13.38 (br, 1H), 8.40 (t,
J¼8.8 Hz, 1H), 7.95e7.49 (m, 5H), 7.22e6.86 (m, 5H), 6.45 (br, 1H),
5.46 (d, J¼2.4 Hz, 1H), 4.87e4.93 (m, 1H), 4.58 (d, 1H), 1.37 (d,
J¼6.8 Hz, 3H); 13C NMR (100 MHz, DMSO)
d 169.3 (CONH), 168.9
(COOH), 165.1 (PhCOO), 144.1, 133.7, 129.5, 129.0, 128.7, 127.8, 126.4,
4.2. General procedure for studying the regioselectivity of the
ring opening of O-benzoyl-L-tartaric anhydride (3)
125.8 (ArH), 74.0, 71.3 (CH), 47.9 (CH), 22.2 (CH3); Anal. Calcd for
C
19H19NO6: C 63.86, H 5.36, N 3.92; Found: C 63.78, H 5.38, N 3.93;
A sample of anhydride 312 (120 mg, 0.5 mmol) was dissolved in
MeCN (3 mL). Next, the corresponding nucleophile (0.5 mmol) was
added. The reaction was stopped by the addition of conc. MeCOOH
after 5 s when using amine as the nucleophile or 24 h when using
alcohol instead. The mixture was quenched by the water/MeCOOH
solution (200 mL/40 mL). The regioisomer ratio was determined by
HPLC using the sample prepared as described above.
FTIR (ATR) (
n
, cmꢂ1) 3351, 1726, 1681, 1228, 1107, 712, 693.
4.4. Procedure B. N-(2-(S)-phenylethylamide) of O-benzoyl-L-
tartaric acid, type II (6b)
Compound 6b was obtained following general procedure A with
(S)-phenylethylamine from the concentrated filtrates after isolation
4.3. General procedure A. Monoamides of O-benzoyl-
L
-tarta-
of 6a. 1.1 g, 14% yield; white crystals; mp 167e172 ꢀC [
a
]
25 ꢂ22.27 (c
D
ric acid, type I (4ae6a)
1.0, EtOH); 1H NMR (400 MHz, DMSO-d6)
d
8.52 (d, J¼8.0 Hz, 1H),
8.04e7.65 (m, 5H), 7.36e7.20 (m, 5H), 5.47 (d, J¼2.8 Hz, 1H),
A mixture of anhydride 3 (5.0 g, 21.0 mmol) and amine
(42.0 mmol) in MeCN (80 mL) was stirred at room temperature.
4.93e4.99 (m, 1H), 4.67 (d, J¼2.8 Hz, 1H), 1.34 (d, J¼6.8 Hz, 3H); 13
C
NMR (100 MHz, DMSO) d 172.9 (COOH), 165.9 (CONH), 165.6

ꢀ
U. Bernas et al. / Tetrahedron 71 (2015) 4047e4052
4052
(PhCOO), 144.6, 133.9, 130.2, 129.0, 128.6, 127.0, 126.5 (ArH), 75.7,
C14H16O8: C 53.85, H 5.16; Found. C 53.39, H 5.30. FTIR (ATR) (
cmꢂ1) 3459, 2944, 1767, 1739, 1710, 1200, 1112, 711, 689.
All structures were confirmed by HMBC spectroscopy.
n,
71.1 (CH), 48.3 (CH), 22.6 (CH3); Anal. Calcd for C19H19NO6: C 63.86,
H 5.36, N 3.92; Found: C 63.61, H 5.32, N 3.92; FTIR (ATR) (
3304, 1737, 1721, 1668, 1242, 1118, 725, 701.
n )
, cmꢂ1
Acknowledgements
4.5. General procedure C. Monoesters of O-benzoyl-L-tartaric
acid, type I (7a, 8a)
The above research project was financially supported by War-
saw University of Technology. The authors kindly thank Mr. Pawe1
Zuk for the valuable advice and fruitful discussions and Mrs. Renata
_
A mixture of anhydride 3 (8.35 g, 35.0 mmol) with ethanol
(30 mL) for 7a or benzyl alcohol (20 mL) in MeCN (15 mL) for 8a was
stirred at room temperature. After the reaction completion (min.
24 h), the precipitated product was filtered off, rinsed with water
and dried under reduced pressure (7a) or crystallized after the
concentration of the reaction mixture under reduced pressure (8a,
9a).
Przedpe1ska for the editorial assistance.
Supplementary data
Supplementary data (Copies of the 1H NMR, 13C NMR, HMBC and
IR spectra of compounds 4e9) related to this article can be found at
http://dx.doi.org/10.1016/j.tet.2015.04.063.
4.5.1. Ethyl O-benzoyltartrate, type I (7a). Compound 7a was ob-
References and notes
tained following general procedure C. 3.2 g, 32% yield; white
25
crystals; mp 175e178 ꢀC [
a
]
þ11.1 (c 1.0, EtOH); 1H NMR
D
ꢀ
ꢀ
1. Gawronski, J.; Gawronska, K. Tartaric and Malic Acids in Synthesis; Wiley: New
(400 MHz, DMSO-d6) 8.03e7.53 (m, 5H), 6.24 (br, 1H), 5.47 (d,
York, NY, 1999.
J¼2.4 Hz, 1H), 4.76 (s, 1H), 4.08 (m, 2H), 1.04 (t, J¼6.8 Hz, 3H); 13C
ꢀ
ꢀ
2. Synoradzki, L.; Bernas, U.; Ruskowski, P. Org. Prep. Proced. Int. 2008, 40, 163.
3. Desymmetrization of tartaric acid 1,4-carboxy groups: (a) Uray, G.; Lindner, W.
Tetrahedron 1988, 44, 4357; (b) Bestmann, H. J.; Roth, D. Synlett 1990, 751; (c)
Guarna, A.; Bucelli, B.; Machetti, F.; Menchi, G.; Occhiato, E. G.; Scarpi, D.;
Trabocchi, A. Tetrahedron 2002, 58, 9865; (d) McNulty, J.; Grunner, V.; Mao, J.
Tetrahedron Lett. 2001, 42, 5609; (e) Barros, M. T.; Burke, A. J.; Lou, J. D.; May-
cock, C. D.; Wahnon, J. R. J. Org. Chem. 2004, 69, 7847.
NMR (400 MHz, DMSO-d6)
d 170.4 (COOEt), 168.0 (COOH), 164.9
(PhCOO), 133.9, 129.6, 128.8,128.7 (ArH), 74.0, 70.3 (CH), 60.8 (CH2),
14.0 (CH3); Anal. Calcd for C13H14O7: C 55.32, H 5.00; Found: C
55.35, H 5.24; FTIR (ATR) (
1276, 1257, 1111, 712, 693; FTIR (EtOH) (
n
, cmꢂ1) 3353, 3035, 1782, 1750, 1717,
n
, cmꢂ1) 1733, 1715.
4. Desymmetrization of tartaric acid 2,3-hydroxy groups: (a) Albizati, K. F.; Babu,
S.; Birchler, A.; Busse, J. K.; Fugett, M.; Grubbs, A.; Haddach, A.; Pagan, M.; Potts,
B.; Remarchuk, T.; Rieger, D.; Rodriguez, R.; Shanley, J.; Szendroi, R.; Tibbetts, T.;
Whitten, K.; Borer, B. C. Tetrahedron Lett. 2001, 42, 6481; (b) Furuta, K.; Miwa, Y.;
Iwanaga, K.; Yamamoto, H. J. Am. Chem. Soc. 1988, 110, 6254; (c) Iwasaki, F.;
Maki, T.; Onomura, O.; Nakashima, W.; Matsumura, Y. J. Org. Chem. 2000, 65,
996; (d) Clarke, P.; Holton, R. A.; Kayaleh, N. E. Tetrahedron Lett. 2000, 41, 2687.
4.5.2. Benzyl O-benzoyltartrate, type I (8a). Compound 8a was ob-
tained following the general procedure C. Crude product was
crystallized from MeOHewater. 5.1 g, 42% yield; white crystals; mp
25
160e162 ꢀC [
d6)
a
]
þ61.11 (c 1.0, EtOH); 1H NMR (400 MHz, DMSO-
D
ꢀ
5. Spengler, J.; Fernandez-Llamazares, A. I.; Ruiz-Rodiguez, J.; Burger, K.; Albericio,
d
13.48 (br, 1H), 7.97e7.52 (m, 5H), 7.10e7.09 (m, 5H), 6.27 (br,
F. J. Org. Chem. 2010, 75, 5746 and references cited herein.
6. Baderschneider, B.; Winterhalter, P. J. Agric. Food Chem. 2001, 49, 2788.
7. Bell, K. H. Aust. J. Chem. 1987, 40, 399.
1H), 5.48 (d, J¼2.8 Hz, 1H), 5.13 (m, 2H), 4.85 (d, 1H); 13C NMR
(400 MHz, DMSO-d6)
d 170.4 (COOBn), 167.9 (COOH), 164.9
8. Bell, K. H. Aust. J. Chem. 1987, 40, 1723.
(PhCOO), 135.6, 133.8, 129.6, 128.7, 128.7, 128.2, 127.9, 127.5 (ArH),
73.9, 70.3 (CH), 66.1 (CH2); Anal. Calcd for C18H16O7: C 62.79, H 4.68;
9. The use of diacyltartaric anhydride as a building block was reported: (a) Ghosh,
A. K.; Koltun, E. S.; Bilcer, G. Synthesis 2001, 1281; (b) Zheng, Y. S.; Hu, Y. J. J. Org.
Chem. 2009, 74, 5660.
10. The use of succinic anhydride as a building block was reported: Serrano, P.;
Casas, J.; Llebaria, A.; Zucco, M.; Emeric, G.; Delgado, A. J. Comb. Chem. 2007, 9,
635.
11. The ring opening of O-acylmalic anhydride: (a) Mhaske, S. B.; Argade, N. P. J. Org.
Chem. 2001, 66, 9038; (b) Shiuey, S. J.; Partrid, J. J.; Uskokovic, M. R. J. Org. Chem.
1988, 53, 1040; (c) Krell, C. M.; Seebach, D. Eur. J. Org. Chem. 2000, 1207; (d)
Found: C 62.56, H 4.76; FTIR (ATR) (
1709, 1269, 1132, 715, 697.
n
, cmꢂ1) 3361, 3064, 1775, 1751,
4.5.3. 2-Methoxyethyl O-benzoyltartrate, type I (9a). Compound 9a
was obtained following general procedure C. 3.5 g, 32% yield; white
25
crystals; mp. 121e124 ꢀC [
(400 MHz, DMSO-d6)
a]
þ16.9 (c 1.0, MeOH). 1H NMR
D
ꢀ
Arencibia-Mohar, A. A.; Ariosa-Alvarez, A.; Madrazo-Alonso, O.; Gonzalez
Abreu, E.; Garcia-Imia, L.; Sierra-Gonzalez, G.; Verez-Benzomo, V. Carbohydr.
d 8.01e7.54 (m, 5H), 6.28 (br, 1H), 5.44 (d,
ꢀ
Res. 1998, 306, 163; (e) Mitsos, C. A.; Zografos, A. L.; Igglessi-Markopoulou, O. J.
Org. Chem. 2000, 65, 5852; (f) Liesen, G. P.; Sukenik, C. N. J. Org. Chem. 1987, 52,
455.
J¼2.4 Hz, 1H), 4.77 (br, 1H), 4.22e4.28 (1H), 4.06e4.11 (m, 1H),
3.32e3.43 (m, 2H), 2.99 (s, 3H); Anal. Calcd for C14H16O8: C 53.85, H
5.16; Found: C 53.48, H 5.28; FTIR (ATR) (
1755, 1709, 1234, 1110, 712, 685.
n
, cmꢂ1) 3311, 2947, 1777
12. Monobenzoyl tartaric anhydride (3) was obtained via cyclization of O-ben-
ꢀ
zoyltartaric acid (2): Bernas, U.; Hajmowicz, H.; Madura, I. D.; Majcher, M.;
Synoradzki, L.; Zawada, K. Arkivoc 2010, xi, 1.
13. Madura, I. D.; Zachara, J.; Bernas, U.; Hajmowicz, H.; Klis, T.; Serwatowski, J.;
ꢀ
ꢀ
Synoradzki, L. J. Mol. Struct. 2010, 984, 75.
14. Synoradzki, L.; Hajmowicz, H.; Bernas, U.; Jerzak, A. Process of preparing de-
4.6. Procedure D. 2-Methoxyethyl O-benzoyltartrate, type II
(9b)
ꢀ
rivatives of O-acyltartaric acid. PL Pat. 210126, 2010; Chem. Abstr. 2012, 156,
257286.
ꢀ
ꢀ
15. Acyl migration: (a) Tomic, S.; Petrovic, V.; Matanovic, M. Carbohydr. Res. 2003,
338, 491; (b) Berces, A.; Whitfield, D. M.; Nukada, T.; do Santos, Z. I.; Ob-
uchowska, A.; Krepinsky, J. J. Can. J. Chem. 2004, 82, 1157; (c) Pugh, C. Org. Lett.
2000, 2, 1329; (d) Li, W.; Du, W.; Li, Q.; Sun, V.; Liu, D. J. Mol. Catal. B: Enzym.
2010, 63, 17.
A mixture of 3 (8.35 g, 35.0 mmol) and 2-methoxyethanol (6.9 g,
91.0 mmol) was stirred at room temperature for 24 h. Next, the
reaction mixture was concentrated and the crude product was
crystallized from MeCN (100 mL). 1.8 g, 16% yield; white crystals;
25
16. These assumption were consistent with observation of Sprenger at al.5 who
worked on the tartaric acid derivatives and noticed base-promoted migration
of the acetyl group in the methyl acetyltartrate in the aq NaHCO3 as well as Li
et al.15d who researched influence of water and temperature on acyl migration
phenomenon in glycerides
mp. 154e158 ꢀC [
a
]
D
þ0.0 (c 1.0, MeOH); 1H NMR (400 MHz,
DMSO-d6)
d
8.05e7.54 (m, 5H), 5.54 (d, J¼2.6 Hz, 1H), 4.67 (d,
J¼2.6 Hz, 1H), 4.18e4.30 (m, 2H), 3.53 (t, J¼4.6 Hz, 2H), 3.23 (s, 3H);
13C NMR (400 MHz, DMSO-d6)
d
171.6 (COOH), 166.8 (COO-
17. The structural characterization of (2R,3R)-3-O-Benzoyl-N-benzyltartramide (5a)
was disclosed in Madura, I. D.; Zachara, J.; Bernas, U.; Hajmowicz, H.; Synor-
adzki, L. Acta Crystallogr. 2012, E68, o1891.
C2H4OMe), 165.0 (PhCOO), 134.0, 129.6, 128.9, 128.6 (ArH), 74.1
(CH), 70.1 (CH), 69.5 (CH2), 64.6 (CH2), 58.2 (CH3); Anal. Calcd for