Chiral Recognition Studies of α-(Nonafluoro- tert -butoxy)carboxylic Acids by NMR Spectroscopy

Article abstract of DOI:10.1021/acs.joc.5b00706

Three chiral α-(nonafluoro-tert-butoxy)carboxylic acids (R)-1, (RS)-2, (R)-3 were synthesized to examine their application as chiral solvating agents with amines. As a model compound, first (S)- and/or (RS)-α-phenylethylamine was used, and their diastereomeric salts were investigated by ¹H and ¹⁹F NMR and ECD spectroscopy. The NMR spectroscopic studies were carried out at room temperature using the slightly polar CDCl₃ and apolar C₆D₆ as solvents in 5 mM and 54 mM concentrations. The difference of the chemical shifts (Δδ) in the diastereomeric complexes is comparable with other, well-known chiral derivatizing and solvating agents (e.g., Mosher's acid, Pirkle's alcohol). Diastereomeric salts of racemic acids (RS)-1 and (RS)-2 with biologically active amines (1R,2S)-ephedrine and (S)-dapoxetine were also investigated by ¹⁹F NMR spectroscopy.

Full text of DOI:10.1021/acs.joc.5b00706

Subscriber access provided by NEW YORK UNIV
Article
Chiral recognition studies of alpha-(nonafluoro-
tert-butoxy)carboxylic acids by NMR spectroscopy
Aniko Nemes, Tamás Csóka, Szabolcs Béni, Viktor Farkas, Jozsef Rabai, and Dénes Szabó
J. Org. Chem., Just Accepted Manuscript • Publication Date (Web): 29 May 2015
Downloaded from http://pubs.acs.org on May 30, 2015
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
The Journal of Organic Chemistry is published by the American Chemical Society.
1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 23
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
Chiral recognition studies of αꢀ(nonafluoroꢀtertꢀbutoxy)carboxylic acids by NMR
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
spectroscopy
Anikó Nemes,∗† Tamás Csóka,† Szabolcs Béni,‡§ Viktor Farkas,║ József Rábai,† Dénes
Szabó†
∗ neagaft@caesar.elte.hu
†
Institute of Chemistry, Eötvös Loránd University, Pázmány Péter sétány 1/A, 1117
Budapest, Hungary
‡ Department of Pharmaceutical Chemistry, Semmelweis University, Hőgyes Endre utca
9, 1092 Budapest, Hungary
§ Department of Pharmacognosy, Semmelweis University, Üllői út 26, 1085, Budapest,
Hungary
║MTAꢀELTE Protein Modelling Research Group, Pázmány Péter sétány 1/A, 1117
Budapest, Hungary
1
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 2 of 23
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Abstract
Three chiral αꢀ(nonafluoroꢀtertꢀbutoxy)carboxylic acids (R)ꢀ1, (RS)ꢀ2, (R)ꢀ3 were
synthesized to examine their application as chiral solvating agents with amines. As model
compound first (S)ꢀ and/or (RS)ꢀαꢀphenylethylamine was used and their diastereomeric
salts were investigated by H and 19F NMR and ECD spectroscopy. The NMR
1
spectroscopic studies were carried out at room temperature using the slightly polar CDCl3
and apolar C6D6 as solvents in 5mM and 54 mM concentrations. The difference of the
chemical shifts (ꢁδ) in the diastereomeric complexes is comparable with other, wellꢀ
known chiral derivatizing and solvating agents (e.g. Mosher’s acid, Pirkle’s alcohol).
Diastereomeric salts of racemic acids (RS)ꢀ1 and (RS)ꢀ2 with biologically active amines
19
(1R,2S)ꢀephedrine and (S)ꢀdapoxetine were also investigated by F NMR spectroscopy.
2
ACS Paragon Plus Environment

Page 3 of 23
The Journal of Organic Chemistry
1
2
3
4
5
6
Introduction
7
8
9
Chiral solvating agents (CSAs) are powerful reagents to determine enantiomeric excess
(ee) of chiral organic compounds.1,2 With the application of CSAs, there is no need to
modify covalently the analyte before the examination because there are only secondary
interactions between the molecules in the solution (such as Coulombic and πꢀπ
interaction, hydrogen bonding). The formed diastereomeric complexes are
distinguishable by spectroscopic methods (NMR, ECD), therefore the enantiomeric
composition of the sample can be determined.3,4
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
To date, numerous chiral discriminating agents are available. Among them, several
contain fluorine atom or trifluoromethyl group.5,6,7,8 The determination of ee with these
19
19
fluorineꢀcontaining reagents is advantageous by F NMR spectroscopy, since F nuclei
has favorable nuclear properties, and there are no or minimal overlapping peaks in the
recorded 19F NMR spectra of the diastereomeric complexes.
Synthesis and utilization of nonafluoroꢀtertꢀbutoxy groupꢀcontaining chiral molecules are
still limited.9 , 10 , 11 Previously we reported the synthesis of three αꢀ(nonafluoroꢀtertꢀ
butoxy)carboxylic acids, two of them in optically active form [(R)ꢀ1, (RS)ꢀ2, (R)ꢀ3]
(Figure 1.).12
Figure 1. αꢀ(Nonafluoroꢀtertꢀbutoxy)carboxylic acids
3
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 4 of 23
1
2
3
4
5
6
7
8
9
The structural properties of these carboxylic acids are highly beneficial for their
application as chiral solvating agents. They contain (i) an acidic functional group which
responsible for the interaction between the chiral reagent and the analyte, (ii) a bulky
hydrophobic OC(CF3)3 moiety, which possesses a significant steric hindrance on one side
of the stereogenic center and decreases the rotational movements around the chiral center,
(iii) and in the case of (RS)ꢀ2 and (R)ꢀ3 an aromatic ring, which also helps to fix a stable
conformation by πꢀπ interaction. The simple structures of these compounds are coupled
with simple spectral properties, consequently there are only a few signals belonging to
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
19
the acids in their H NMR spectra. Additionally F NMR spectra have only one strong
singlet related to the nine chemically equivalent fluorine atoms. In this paper, we present
an extended experimental study on the enantiomeric discrimination ability of the above
αꢀ(nonafluoroꢀtertꢀbutoxy)carboxylic acids.
Results and Discussion
Synthesis
The synthesis of αꢀ(nonafluoroꢀtertꢀbutoxy)carboxylic acids 1ꢀ3 was described in our
preliminary work in detail.12 It is worthy to note that, as expected, the stereospecific
synthesis of caboxylic acids 1 and 3 resulted in the corresponding enantiomers, but
similar methods yielded racemic product in the case of 2 (Scheme 1).
4
ACS Paragon Plus Environment

Page 5 of 23
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 1. Syntheses of αꢀ(nonafluoroꢀtertꢀbutoxy)carboxylic acids 1ꢀ3
Because the application of 2 as chiral agent is possible in its enantiomerically pure form,
we tried to carry out its optical resolution. Testing several methods we obtained the best
result by using half equivalent of organic base (S)ꢀphenylethylamine [(S)ꢀPEA] as
resolving agent in warm aqueous media (see Scheme 2). To one equivalent of racemic 2
one equivalent of NaOH and half equivalent of (S)ꢀPEA × HCl were added. One of the
diastereomeric salt separated in crystalline form, from which (ꢀ)ꢀ2 was regenerated by
aqueous Na2CO3 solution. This method provided enantiomerically enriched carboxylic
acid (ꢀ)ꢀ2 (ee: 33 %, [α]578= ꢀ24°, c = 0.5, methanol). The other enantiomer was prepared
from the mother liquor by acidification in 30% ee ([α]578= +14°, c = 0.5, methanol).
Unfortunately our further efforts to obtain optically pure 2 from these enantiomer
mixtures failed. The efficiency of resolution was analyzed by 1H NMR spectroscopy.
5
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 6 of 23
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 2. Resolution experiments of (RS)ꢀ2 with (S)ꢀPEA
NMR measurements
There are two main approaches to determine ee by NMR spectroscopy.1,2 First, chiral
derivatizing agents (CDAs) are covalently bonded to the analyte, therefore the structure
of the formed diastereomers are strictly fixed. Here, the typical difference of the chemical
shifts in the two diastereomers is approximately 0.1 – 0.2 ppm in 1H NMR and 0.3 – 0.7
ppm in 19F NMR, respectively.3,4 The second approach is using chiral solvating agents for
this purpose. These reagents form diastereomeric complexes with the dissolved
enantiomers via rapid reversible equilibrium in a competition with the bulk solvent. This
equilibrium is faster than the time scale of the NMR measurement, thus there is one set of
peaks for each diastereomeric complex.13,14 The observed chemical shift anisochrony (~
1
0.05 ppm in H NMR)13 is the result of: (i) the relative position of magnetically
anisotropic groups (e.g. aryl, carbonyl) of the low energy conformers in the solution (ii)
the relative value of diastereomeric complex formation constants KRS and KSS, i.e., the
strength of interactions between the ion pairs, which depends primarily on the nature of
the functional groups for a given molecule. Using this method have an advantage of
6
ACS Paragon Plus Environment

Page 7 of 23
The Journal of Organic Chemistry
1
2
3
4
5
6
“easy” performance, by simply adding the reagent to a chiral compound in a small
amount of deuterated solvent.
7
8
9
In our experiments the diastereomeric salt formation was investigated between fluorous
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
carboxylic acids 1ꢀ3 and αꢀphenylethylamine (PEA) and characterized by H and 19F
NMR spectroscopy. PEA has been widely used by others as a model substrate for
enantioselective recognition studies.15,16
In these cases the recorded chemical shifts of the diastereomeric complexes are weighted
averages of chemical shifts of the distinct protonation forms of the free and the
1
protonated amine, according to their relative amounts. The integrals of the signals in H
and 19F NMR spectra are proportional with the amount of the enantiomers therefore
enantiomeric composition of the sample can be simply determined.
In our study the first attempt was to observe the enantiorecognition phenomena on the 1H
nuclei, followed by the 19F NMR measurements. In several cases when the
1
enantiodifferentiation is inhibited by overlapping peaks or if the H NMR spectra are
19
rather complex to interpret without spectral simulations, F NMR experiments posses a
great advantage. To enhance the ion pair formation between the CSA and the analyte, i.e.
the differences in the chemical shifts between the diastereomers, the slightly polar CDCl3
and the apolar C6D6 were used as solvents. Our preliminary experiments on the effect of
acid : amine molar ratio revealed, that 1:1 stoichiometry provides the highest ꢀδ, thus this
1:1 ratio was applied. NMR measurements were carried out on samples in two different
1
concentrations of the diastereomeric salts (54 and 5 mM) in both solvents. H and 19F
NMR spectra were recorded at room temperature immediately after mixing. The results
7
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 8 of 23
1
2
3
4
are summarized in Figure 2ꢀ4, where the differences in the chemical shifts (ꢀδ) and
5
6
7
partial 1H NMR spectra are given.
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(RS)ꢀ1
× (S)ꢀPEA and (R)ꢀ1 × (RS)ꢀPEA
The racemic (RS)ꢀ1 was analyzed with enantiopure (S)ꢀPEA, while optically active lactic
acid derivative (R)ꢀ1 was tested with racemic amine (RS)ꢀPEA. In the first set of
1
experiments we recorded the H NMR spectra in 54 mM concentration, using CDCl3 as
solvent. The diastereomeric salts formed between (RS)ꢀ1 and (S)ꢀPEA have chemical
shift differences (ꢀδ) 0.038 ppm for the methine protons (quartets, 4.3ꢀ4.5 ppm) and
0.030 ppm for the methyl protons (doublets, 1.0ꢀ1.2 ppm) of (RS)ꢀ1 (Figure 2). These
values are close to those given in the literature, e.g. 0.05 ppm, using Pirkle’s alcohol as
CSA.6, 17 The signals of (R)ꢀ1 × (RS)ꢀPEA salt show broad signals for the methine
(quartets, 4.1ꢀ4.2 ppm) and 0.034 ppm for the methyl hydrogen atoms (doublets, 1.4ꢀ1.6
ppm) of (RS)ꢀPEA. Our results indicate that chiral discrimination by NMR between 1 and
PEA is possible to detect if one of the two compounds is enantiomerically pure. The
measured ꢀδs are significant in both cases, but not baseline resolution was observed due
to the multiplet resonances.
8
ACS Paragon Plus Environment

Page 9 of 23
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
0.006 (0)
CF3
CF3
F3C
H3C
CH3
O
COO
0.030 (0.031)
H3N
H
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
0.038 (0.057)
(RS)-1 x (S)-PEA in CDCl3
CF3
0.034 (0)
F3C
H3C
CF3
CH3
O
H3N
H
COO
broad signals (0.017)
(R)-1 x (RS)-PEA in CDCl3
0 (0.006)
CF3
F3C
H3C
CF3
CH3
O
0.089 (0.033)
COO
H3N
H
0.082 (0.016)
(RS)-1 x (S)-PEA in C6D6
0.035 (0.063)
CF3
0.012 (0)
F3C
CF3
CH3
O
H3N
H
H3C
COO
broad signals (0.055)
(R)-1 x (RS)-PEA in C6D6
9
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 10 of 23
1
2
3
4
Figure 2. Chemical shift differences (blue: methyl protons, red: methine proton,
5
6
1
green: fluorine atoms in ppm) and partial H NMR spectra (400 MHz) for the
7
8
9
diastereomeric salts of (RS)ꢀ1 × (S)ꢀPEA and (R)ꢀ1 × (RS)ꢀPEA in CDCl3 and C6D6
when the concentration is 54 mM, data for 5 mM are given in brackets
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In our second set of experiments C6D6 was used as solvent. In this case the formation of
the tight ion pairs is more favorable (larger K values), thus the average chemical shifts
are further away from those recorded for the pure carboxylic acid and the free amine. Our
results show that the differences between the signals in the diastereomeric salts (ꢀδ) are
more explicit in this apolar solvent. These data are shown also in Figure 2. The measured
ꢀδ of (R)ꢀ1 × (RS)ꢀPEA is 0.035 ppm for methyl protons, however the frequencies of
methine group are not resolved in 54 mM concentration due to solubility problems. In
this case applying lower concentration the signal separation is detectable; in 5mM
solution the chemical shift difference is 0.055 ppm for the methine and 0.063 ppm for
methyl protons. Even better chiral discrimination can be achieved using racemic fluorous
acids. (RS)ꢀ1 × (S)ꢀPEA shows 0.082 ppm difference for the methine and 0.089 ppm for
methyl hydrogens.
(RS)ꢀ2
× (S)ꢀPEA
1
Chemical shift differences and partial H NMR spectra for the salt of racemic mandelic
acid derivative (RS)ꢀ2 and optically active (S)ꢀPEA are shown in Figure 3.
10
ACS Paragon Plus Environment

Page 11 of 23
The Journal of Organic Chemistry
1
2
3
4
5
6
0.006 (0.007)
CF3
CF3
F3C
CH3
7
O
8
9
COO
H3N
H
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
0.130 (0.011)
(RS)-2 x (S)-PEA in CDCl3
0.028 (0.023)
CF3
F3C
CF3
CH3
O
H3N
COO
H
0.190 (0.178)
(RS)-2 x (S)-PEA in C6D6
Figure 3. Chemical shift differences (red: methine proton, green: fluorine atoms in
ppm) and partial 1H NMR spectra (400 MHz) for the diastereomeric salt of (RS)ꢀ2 ×
(S)ꢀPEA in CDCl3 and C6D6 when the concentration is 54 mM, data for 5 mM are
given in brackets
1
The signals in the H NMR spectrum of (RS)ꢀ2 × (S)ꢀPEA, which belong to the methine
protons of carboxylic acid 2, appear as two sharp fully baseline separated singlets which
do not overlap with other peaks (Figure 3). It is remarkable that this salt shows the
highest ꢀδ=0.130 ppm (at 5ꢀ5.3 ppm) in CDCl3 solvent which is comparable to those
reported for chiral derivatizing agents. The shifts of the resonances of this latter salt are
much larger than those in the spectrum of fluorous lactic acid salt [(RS)ꢀ1 × (S)ꢀPEA],
denoting a strong association between the ion pairs in the phenyl groupꢀcontaining
11
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 12 of 23
1
2
3
4
5
6
7
8
9
compound [(RS)ꢀ2]. This salt probably has a relatively rigid structure due to the directly
attached phenyl group to the chiral center resulting in a favorable πꢀπ interaction between
the aromatic groups. As expected, the integral ratio of the two separated methine peaks of
(RS)ꢀ2 is 1 : 1 while 1 : 2 for the enantiomerically enriched (33 % ee) compound (+)ꢀ2/(ꢀ
)ꢀ2.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Similarly to CDCl3 the highest ꢀδ=0.190 ppm was detected in C6D6 for methine protons
of (RS)ꢀ2 × (S)ꢀPEA. The signals of racemic fluorous carboxylic acids (RS)ꢀ1 and (RS)ꢀ2
show large chemical shift nonꢀequivalence, which are comparable to the literature
reported values for CDAs.
(R)ꢀ3
× (RS)ꢀPEA
In the case of (R)ꢀ3 × (RS)ꢀPEA ꢀδ of the methine protons of PEA is 0.019 ppm and
0.026 ppm for methyl protons (Figure 4) when CDCl3 was used as solvent. It is worthy to
note that the spectra of this salt show broad, strongly overlapping signals, which can be
attributed to the conformational flexibility of the formed diastereomeric complex. It
should also be noted that when PEA is in racemic form in the salts with (R)ꢀ1 and (R)ꢀ3,
the methyl protons show larger ꢀδ, compared to the methine protons, although they are
further from the chiral center. Unfortunately the (R)ꢀ3 × (RS)ꢀPEA salt has low solubility
in C6D6 resulting in broad signals therefore the diastereomers are not distinguishable.
12
ACS Paragon Plus Environment

Page 13 of 23
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
CF3
CF3
0.026 (0.046)
CH3
F3C
O
H3N
9
COO
H
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
0.019 (0)
(R)-3 x (RS)-PEA in CDCl3
(0.010)
F3C
CF3
CF3
CH3
O
H3N
COO
H
Broad signals
(R)-3 x (RS)-PEA in C6D6
Figure 4. Chemical shift differences (blue: methyl protons, red: methine proton,
1
green: fluorine atoms in ppm) and partial H NMR spectra (400 MHz) for the
diastereomeric salt of (R)ꢀ3 × (RS)ꢀPEA in CDCl3 and C6D6 when the concentration
is 54 mM, data for 5 mM are given in brackets
19F NMR measurements
1H is the most commonly used nuclei for ee determination by NMR spectroscopy, since
protons are present in almost all organic molecules and it has the highest sensitivity.
1
However, analytical protocols for ee determination relying on H NMR spectroscopy
suffer from several disadvantages, namely the narrow spectral window (low chemical
shift dispersion in the range of 0 to 12 ppm) and the signal multiplicities due to Hꢀ1H
1
coupling, which can inflict severe overlapping peaks. The range of fluorine chemical
13
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 14 of 23
1
2
19
3
4
shifts and the sensitivity of F NMR to emphasize the details of the local environment
5
6
7
8
are higher than 1H NMR. At the same time the singlet nature of the signals in standard 1H
decoupled experiments makes 19F NMR a powerful alternative or complement method to
1H NMR, especially when studying pure compounds with crowded proton spectra or
mixtures of compounds. Observing 19F nuclei the spectra overlap is eliminated by the
wider chemical shift range of 19F signals and to the absence of signal multiplicities in
standard proton decoupled spectra.18
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Although in our cases fluorine atoms are further from the stereogenic center, 19F NMR
spectra also indicate the difference between the above diastereomeric salts. These ꢀδ
values are also demonstrated in Figure 2ꢀ4.
1
The detected 19F chemical shift differences show strong correlation with the H NMR
results. Accordingly, considerable ꢀδs were observed also for both (RS)ꢀ1 × (S)ꢀPEA
(0.006, 5 mM) and (R)ꢀ1 × (RS)ꢀPEA salts (0.012 ppm, 54 mM; Figure 2 and 5), while
the best resolution of the resonances was observed in the case of the carboxylic acid (RS)ꢀ
2 × (S)ꢀPEA in C6D6 in 54 mM concentration (0.028 ppm, Figure 3 and 6).
Figure 5. Partial 19F NMR spectra (376.40 MHz) for the (RS)ꢀ1 × (S)ꢀPEA salt in
CDCl3 in 54 mM (left) and in C6D6 in 5 mM (right)
14
ACS Paragon Plus Environment

Page 15 of 23
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 6. Partial 19F NMR spectra (376.40 MHz) for the (RS)ꢀ2 × (S)ꢀPEA salt in
CDCl3 in 54 mM (left) and in C6D6 in 54 mM (right)
The complexes of the benzylic derivative (R)ꢀ3 × (RS)ꢀPEA are not distinguishable by 19F
NMR spectroscopy using routine conditions (Figure 4 and 7). Although, measuring in
cold solution conformational movements may be reduced and thus broad signals can be
eliminated, but unfortunately the low solubility of the above salts at lower temperature
hinders the application of this technique.
Figure 7. Partial 19F NMR spectra (376.40 MHz) for the (R)ꢀ3 × (RS)ꢀPEA salt in
CDCl3 in 54 mM (left) and in C6D6 in 54 mM (right)
For further exploration of the potential use of (RS)ꢀ1 and (RS)ꢀ2 as chiral discriminating
agents we tested their applicability also with other known organic bases (1R,2S)ꢀ
ephedrine (EPH) and (S)ꢀdapoxetine [(S)ꢀDpx] (Figure 8). Ephedrine is a biologically
active amine used as a drug of low toxicity for treating asthma.19 It is available from
15
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 16 of 23
1
2
3
4
natural source in chiral form. The (S)ꢀenantiomer of dapoxetine is also a drug with
5
6
serotonine transporter inhibitory effect.20,21
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 8. Structure of (1R,2S)ꢀEphedrine and (S)ꢀDapoxetine
Both amines have complex structure, thus their NMR spectra are more difficult to
1
analyze than that of PEA. First, we recorded the H NMR spectra of the diastereomeric
complexes of ephedrine (EPH) as a potential CSA with the racemic acids (RS)ꢀ1 and
(RS)ꢀ2, but no chemical shift differences between the enantiomers were observed, thus
1
19
we could not differentiate them by H NMR. In contrary, in their F NMR spectra in 5
mM C6D6 solutions we observed two singlets, thus the diastereomeric salts formed
become distinguishable under this condition. The ꢁδ in (RS)ꢀ1 × EPH and (RS)ꢀ2 × EPH
are 0.008 ppm and 0.010 ppm, respectively (Figure 9). In order to prove the effective
chiral recognition ability of (RS)ꢀ2, we tested it also with optically active dapoxetine
using C6D6 as solvent. The 19F NMR spectrum shows two singlet signals for the
enantiomers with 0.014 ppm chemical shift difference (Figure 10).
16
ACS Paragon Plus Environment

Page 17 of 23
The Journal of Organic Chemistry
1
2
3
4
Figure 9. Partial 19F NMR spectra (376.40 MHz) for the ephedrine salts in C6D6 in 5
5
6
7
mM; EPH × (RS)ꢀ1 (left) and EPH × (RS)ꢀ2 (right)
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 10. Partial 19F NMR spectra (376.40 MHz) for the (S)ꢀdapoxetine salt (S)ꢀ
Dpx × (RS)ꢀ2 in C6D6 in 5 mM
ECD measurements
As a part of our above studies we investigated the complex formation of (R)ꢀ1, (RS)ꢀ2
and (R)ꢀ3 by farꢀUV ECD spectroscopy (190ꢀ250 nm). First, we recorded the ECD
spectra of (R)ꢀ1 and (R)ꢀ3, and their complexes with (R)ꢀ and (S)ꢀPEA. The spectrum of
the lactic acid derivative (R)ꢀ1 has a broad negative band in both acetonitrile and
isooctane (Figure 11) and its 1:1 complex with one of the enantiomers of PEA show only
a very weak effect in this region (data not shown).
In the case of the aromatic (R)ꢀ3 the spectrum is more intensive. The explanation of the
ECD spectra of benzene chromophore is well known and has been discussed in detail.22
Thus we can conclude that in this region the bands near 197 and 220 nm correspond to
the 1Bb and 1La transitions of the benzene chromophore, respectively (Figure 11).
17
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 18 of 23
1
2
3
4
5
3
2
6
7
1
8
0
9
-1
-2
-3
-4
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
190
200
210
220
230
240
250
λ
/ nm
Figure 11. FarꢀUV ECD spectra of (R)ꢀ1 (dash) and (R)ꢀ3 (solid) in acetonitrile
(black) and isoꢀoctane (red)
The complex formation of the phenyllactic acid derivative (R)ꢀ3 with the enantiomers of
PEA is more effective in isooctane than in acetonitrile, so we discuss these results in the
latter solvent (Figure 12).
a)
b)
4
3
4
3
2
2
1
1
0
0
-1
-2
-3
-4
-5
-6
-7
-8
-9
-10
-1
-2
-3
-4
-5
-6
-7
-8
-9
-10
190
200
210
220
230
240
250
190
200
210
220
230
240
250
λ
/ nm
λ
/ nm
Figure 12. a) FarꢀUV ECD spectra of heterochiral (green solid) and homochiral (red
solid) 1:1 complexes of (R)ꢀ3 (black) with (S)ꢀPEA (green dash) and (R)ꢀPEA (red
dash) in isooctane. b) Difference spectra of the complexes, ꢁ(ꢁε)=ꢁεcomp ꢀ (ꢁε(R)ꢀ3
+
ꢁεPEA); (R)ꢀ3 and (R)ꢀPEA (red curve), (R)ꢀ3 and (S)ꢀPEA (green curve).
18
ACS Paragon Plus Environment

Page 19 of 23
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
By using difference spectra [ꢁ(ꢁε)=ꢁεcompꢀ(ꢁε(R)ꢀ3 + ꢁεPEA)] we are able to compare the
discriminating efficiency. Figure 12b clearly shows that the spectra of the complexes are
not only a simple sum of the enantiomers of 3 and PEA but also there is an interaction
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
between them. This effect is located in the region of the La transition (205ꢀ222 nm)
1
rather than Bb. These results are in correlations with the data received from NMR
measurements (cf. Figure 6 and 9).
Figure 13 shows the recorded ECD spectra of the racemic mandelic acid derivative (RS)ꢀ
2 with (S)ꢀPEA [ie. the sum of (R)ꢀ2 × (S)ꢀPEA and (S)ꢀ2 × (S)ꢀPEA] and (R)ꢀPEA [ie.
the sum of (R)ꢀ2 × (R)ꢀPEA and (S)ꢀ2 × (R)ꢀPEA], (S)ꢀ and (R)ꢀPEA are also indicated.
As can be seen the 1:1 complexes with the enantiomers of PEA show mirror image ECD
spectra. The change of intensity and the shape of curves indicates interactions between
the racemic fluorous acid and the enantiomers of PEA.
5
4
3
2
1
0
-1
-2
-3
-4
190
200
210
220
230
240
250
λ
/ nm
Figure 13. FarꢀUV ECD spectra of the complexes of (RS)ꢀ2 × (S)ꢀPEA (green solid)
and (RS)ꢀ2 × (R)ꢀPEA (red solid) and (S)ꢀPEA (green dash) and (R)ꢀPEA (red dash)
in isooctane
Conclusion
19
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 20 of 23
1
2
3
4
5
6
7
8
9
αꢀ(Nonafluoroꢀtertꢀbutoxy)carboxylic acids (1ꢀ3) show chiral recognition ability towards
chiral amines in apolar solvents. Thus using optically active 1ꢀ3 as CSA a simple
determination of enantiomeric excess (ee) of chiral amines can be achieved by NMR
spectroscopy. This method is fast, accurate and requires only small amounts of sample
and reagent. The analysis of the enantiomeric ratio can be executed within a short time
with the sample used for the identification of the product, just adding the optically active
carboxylic acid as CSA. Observing the chemical shift difference of the protons close to
the chiral centers in the diastereomeric salts by 1H NMR spectroscopy was feasible. The
measured ꢀδs for the methine protons are 0.017–0.190 ppm which values are comparable
to those of the known chiral derivatizing and solvating agents.3 Concurrently, these
fluorous carboxylic acids contain nine chemically equivalent fluorine atoms which have
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
19
only one strong singlet in F NMR. Since these fluorine atoms are extremely sensitive
towards their chemical environment, 19F NMR spectroscopy offers great potential to
1
determine ee of samples bearing complex H NMR spectra. The measured 19F NMR
chemical shift differences were 0.006ꢀ0.028 ppm which values are also comparable to
those of known chiral solvating agents.6 In addition 1ꢀ3 were tested using (1R,2S)ꢀ
ephedrine and (S)ꢀdapoxetine as amines. Employing (RS)ꢀ2 as acidic component both
diastereomeric salts show significant 19F NMR chemical shift differences 0.010 and
0.014 ppm in C6D6. Therefore, optically active αꢀ(nonafluoroꢀtertꢀbutoxy)carboxylic
acids 1ꢀ3 are promising CSAs both in 1H and 19F NMR spectroscopy.
Experimental section
Resolution experiments
20
ACS Paragon Plus Environment

Page 21 of 23
The Journal of Organic Chemistry
1
2
3
4
5
6
Resolution of (RS)ꢀ2ꢀPhenylꢀ2ꢀ[1,1ꢀbis(trifluoromethyl)ꢀ2,2,2ꢀtrifluoroethoxy]acetic acid
[(±)ꢀ2]
7
8
9
To the solution of (±)ꢀ2 (1.00 g, 2.7 mmol) in 0.1 M NaOH (27 ml) was added the
solution of (S)ꢀPEA (0.164 g, 1.35 mmol) in 1 M HCl (1.35 ml) at 90 ºC. The mixture
was allowed to stand for a night at rt and filtered to give the diastereomeric salt with (ꢀ)
rotation {0.75 g, mp 110ꢀ135 °C, [α]578 = ꢀ17° (c = 0.5, DMF)}. To liberate the acid (ꢀ)ꢀ2
130 mg of the latter salt was suspended in 1 M HCl (pH ~ 2), the resulting white
precipitate was filtered and dried over P2O5 to yield acid (ꢀ)ꢀ2 {50 mg; [α]578 = ꢀ24° (c =
0.5, methanol), mp 104 – 106 °C}.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The clear aqueous filtrate of the (+) salt was acidified with 1 M HCl to pH ~ 2, the
precipitated acid (+)ꢀ1 was filtered off and dried to give the partially resolved acid (+)ꢀ1
{180 mg; [α]578 = +14° (c = 0.5, methanol), mp 104.5ꢀ108 °C}.
NMR experiments
NMR sample solutions were made as follows: 0.027 mmol carboxylic acid was dissolved
in 600 ꢁl of deuterated solvent. After 1H and 19F NMR measurements 0.014 mmol, 0.027
mmol, 0.040 mmol, 0.054 mmol amine was added and measured the NMR spectra. As
deuterated solvent CDCl3 and C6D6 were used. All measurements were carried out at 298
K.
CD experiments
21
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 22 of 23
1
2
3
4
5
6
7
8
9
ECD spectra were recorded at room temperature using 0.1 cm cell for measurements
between 190 and 250 nm. Acetonitrile and 2,2,4ꢀtrimethylpentane (isoꢀoctane) were used
as solvents and the concentration was 25 ꢂM.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Acknowledgments
The authors would like to thank Hungarian Scientific Research Foundation (OTKA PD
109373) for the financial support of the NMR measurements and András Darcsi for (S)ꢀ
Dapoxetine.
Supporting Information. 2D NMR spectra for 1, 2 and 3 and characterization data for
diastereomeric salts. This material is available free of charge via the Internet at
http://pubs.acs.org.
References
1 Wenzel, T. J.; Chisholm, C. D. Chirality, 2011, 23, 190ꢀ214.
2 Parker, D. Chem. Rev. 1991, 91, 1441ꢀ1457.
3 Seco, J. M.; Quiñoá E.; Riguera R. Chem. Rev. 2012, 112, 4603ꢀ4641.
4 Seco, J. M.; Quiñoá E.; Riguera R. Chem. Rev. 2004, 104, 17ꢀ117.
5 Dale, J. A.; Mosher, H. S. J. Am. Chem. Soc. 1968, 90, 3732ꢀ3738.
6 Pirkle, W. H.; Sikkenga, D. L.; Pavlin, M. S. J. Org. Chem. 1977, 42, 384ꢀ387.
7 Apparu, M.; Tiba, Y. B.; Léo, PꢀM.; Hamman, S.; Coulombeau, C. Tetrahedron:
Asymmetry, 2000, 11, 2885ꢀ2898.
22
ACS Paragon Plus Environment

Page 23 of 23
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8 Pomares, M.; SánchezꢀFerrando, F.; AlvarezꢀLarena, A.; Piniella, J. F. J. Org. Chem.
2002, 67, 753ꢀ758.
8
9 Szabó, D.; Nemes, A.; Kövesdi, I.; Farkas, V.; Hollósi, M.; Rábai, J. J. Fluorine Chem.
2006, 127, 1405ꢀ1414.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
10 Chu, Q.; Henry, C.; Curran, D. P. Org. Lett. 2008, 10, 2453ꢀ2456.
11 Yu, Y. B. Nanomed. Nanobiotech. 2013, 5, 646ꢀ661.
12 Csóka, T.; Nemes, A.; Szabó, D. Tetrahedron Lett. 2013, 54, 1730ꢀ1733.
13 Analytical Methods in Supramolecular Chemistry Ed.: Schalley, C. 2007, WileyꢀVCH
Verlag GmbH&Co. KGaA, Weinheim, ISBN: 978ꢀ3ꢀ527ꢀ31505ꢀ5.
14 Quinn, T. P.; Atwood, P. D.; Tanski, J. M.; Moore, T. F.; FolmerꢀAndersen, J. F. J.
Org. Chem. 2011, 76, 10020ꢀ10030.
15 Sachin Rama Chaudhari; Suryaprakash, N. J. Org. Chem. 2012, 77, 648ꢀ651.
16 Schrader, T. J. Org. Chem. 1998, 63, 264ꢀ272.
17 Pedras, M. S. C.; Hossain, M.; Sarwar, M. G.; Montaut, S. Bioorg. Med. Chem. Lett.
2004, 14, 5469ꢀ5471.
18 Guide to Fluorine NMR for Organic Chemists Dolbier, W. R. 2009, John Wiley &
Sons Inc. Hoboken, New Jersey, ISBN: 978ꢀ0ꢀ470ꢀ19341ꢀ9.
19 Drew, C. D. M.; Knight, G. T.; Hughes, D. T. D.; Bush, M. Br. J. Clin. Pharmacol.
1978, 6, 221ꢀ115.
20 Khatik, G. L.; Sharma, R.; Kumar, V.; Chouhan, M.; Nair, V. A. Tetrahedron Lett.
2013, 54, 5991ꢀ5993.
21 Darcsi, A; Tóth, G.; Kökösi, J.; Béni, S. J. Pharm. Biomed. Anal. 2014, 96, 272ꢀ277.
22 Smith, H. E. Chem. Rev. 1998, 98, 1709ꢀ1740.
23
ACS Paragon Plus Environment