Michal Kriegelstein1, David Profous1, Adam Přibylka1, Antonín Lyčka2, Petr Cankař1. 1. Department of Organic Chemistry, Faculty of Science, Palacký University, 17. Listopadu 12, 771 46 Olomouc, Czech Republic. 2. Faculty of Science, University of Hradec Králové, Rokitanského 62, CZ-500 03 Hradec Králové 3, Czech Republic.
Abstract
Axially chiral 2-(2-(trifluoromethyl)-1H-benzo[d]imidazole-1-yl)benzoic acid (TBBA) was used as a chiral derivatizing agent to evaluate the limits of absolute configuration assignment for β-chiral aminoalcohols. Seven Boc-aminoalcohols and eight variously N-substituted (S)-phenylglycinols were prepared, and their TBBA esters were analyzed by NMR spectroscopy. Diverse substitution at the β-position was employed to demonstrate the effect of structure on the general conformational model and reliability of the absolute configuration assignment. It was concluded that hydrogen bond formation and steric hindrance were the main factors affecting the correct assignment for Boc-aminoalcohols.
Axially chiral 2-(2-(trifluoromethyl)-1H-benzo[d]imidazole-1-yl)benzoic acid (TBBA) was used as a chiral derivatizing agent to evaluate the limits of absolute configuration assignment for β-chiral aminoalcohols. Seven Boc-aminoalcohols and eight variously N-substituted (S)-phenylglycinols were prepared, and their TBBA esters were analyzed by NMR spectroscopy. Diverse substitution at the β-position was employed to demonstrate the effect of structure on the general conformational model and reliability of the absolute configuration assignment. It was concluded that hydrogen bond formation and steric hindrance were the main factors affecting the correct assignment for Boc-aminoalcohols.
The assignment of the
absolute configuration of chiral compounds
is an essential part of structure elucidation in chemistry. For this
purpose, NMR spectroscopy is a valuable tool[1] among other available analytical techniques, such as X-ray crystallography,[2] circular dichroism (ECD, VCD),[3] or other chiroptical methods to determine the absolute
configuration. Most commonly, NMR methods designed for the assignment
of absolute configurations use chemical derivatization with various
chiral derivatization agents (CDAs) to convert the analyte into two
diastereomers, and NMR spectra (commonly 1H or 13C) are compared.[4−7] Then, the observed chemical shift differences are employed in a
proposed conformational model to determine the spatial arrangement
of substituents at the chiral center. Conformational models are generally
based on NMR analyses of known chiral compounds and in silico calculations.
The reliability of the most common CDAs has been assessed by multiple
investigations, where diverse structural motifs have been evaluated
to explore the scope and limits.[5,8−11]Recently, we reported a benzimidazole-based axially CDA, 2-(2-(trifluoromethyl)-1H-benzo[d]imidazole-1-yl)benzoic acid (TBBA),
and its application toward α-chiral[4] and β-chiral compounds.[12] While
the presented conformational model (Figure a–c) was reliable for various β-chiral
analytes, the analysis of (S)-Boc-phenylglycinol
assigned the opposite configuration (Figure d). We suspected that a hydrogen bond between
the carbamate NH group and CF3 group locked the compound
in a different conformation. This finding was unexpected since various
aminoalcohols, which were converted with TBBA into amides with free
hydroxyl groups, did not display any significant deviation from the
proposed model due to the hydrogen bond(s).[4] However, it was also reported that the presence of a polar group
at the chiral center could change the conformational equilibrium.[10] For these reasons, we decided to gain deeper
insight into the limitations of TBBA as a CDA for β-chiral aminoalcohols.
This study is focused on the relationship between the structure of
aminoalcohols and the shielding effect of the TBBA benzimidazole ring.
Figure 1
(a) Shielding
effect of the (P)-TBBA benzimidazole
group on the NHR[2] substituent, (b) shielding
effect of the (M)-TBBA benzimidazole group on the
R1 substituent, (c) conformational model for analysis of
primary β-chiral alcohols, and (d) inverted values of ΔδPM observed for (S)-Boc-phenylglycinol.
(a) Shielding
effect of the (P)-TBBA benzimidazole
group on the NHR[2] substituent, (b) shielding
effect of the (M)-TBBA benzimidazole group on the
R1 substituent, (c) conformational model for analysis of
primary β-chiral alcohols, and (d) inverted values of ΔδPM observed for (S)-Boc-phenylglycinol.
Results and Discussion
To explore
whether anomalous ΔδPM values
of (S)-Boc-phenylglycinol 1 are an exception
or trend from the previously proposed model (Figure ),[12] we prepared
more structurally diverse TBBA esters of Boc-aminoalcohols (Figure ).
Figure 2
Analyzed Boc-aminoalcohols 1–7 and their ΔδPM (ΔδPM = δR(P) –
δR(M)) values in CDCl3 as TBBA esters.
9-Anthrylmethoxyacetic
acid (9-AMA) values are shown in green. The values of 9-AMA esters
for compounds 3 and 6 represent the inverse
absolute configuration of Boc-aminoalcohols.
Analyzed Boc-aminoalcohols 1–7 and their ΔδPM (ΔδPM = δR(P) –
δR(M)) values in CDCl3 as TBBA esters.
9-Anthrylmethoxyacetic
acid (9-AMA) values are shown in green. The values of 9-AMA esters
for compounds 3 and 6 represent the inverse
absolute configuration of Boc-aminoalcohols.The Boc group showed a very low ΔδPM value
(−0.002) in (S)-Boc-phenylglycinol 1 (Figure ). However,
based on the remaining ΔδPM differences at
the ortho-H of the phenyl ring (−0.1) and
NH group (+0.06), the opposite configuration would be deduced (Figure ). The ΔδPM value for the Boc group (−0.002) could be considered
anomalous and negligible in practical use to assign the absolute configuration
of unknown compounds. Partial epimerization (approx. 10%) was observed
during the preparation of 1; nevertheless, it did not
hamper the assignment of the NMR signals and absolute configuration.Other derivatives shown in Figure also followed the anomalous trend in sign distribution
in contrast to the previously proposed model. Boc-cyclohexylglycinol 2 showed a reliable distribution of ΔδPM: +0.53 for NH, +0.02 for Boc, and −0.1 ppm for the cyclohexyl
CH proton. The substitution of the cyclohexyl ring for the less sterically
demanding isopropyl group in valinol 3 achieved a similar
distribution of ΔδPM values: −0.38 for
NH; −0.01 for Boc; and +0.05, +0.06, and +0.15 for the isopropyl
group (reversed values due to the opposite configuration). Further
simplification of the structure by substituting the isopropyl group
with a methyl group in alaninol 4 led to a slight change
in the magnitudes of ΔδPM: the NH group now
displayed a value of only +0.13 ppm, which is significantly smaller
than that observed for compounds 2 and 3.The substitution of methyl for benzyl in phenylalaninol 5 significantly increased the ΔδPM value
at
the benzylic position (−0.39 and −0.09 vs −0.11
in 4) and NH group (+0.47 vs +0.13 in 4).
Further substitution of phenyl in 5 for isopropyl in
leucinol 6 caused a reduction in the magnitude of ΔδPM. The amino group displayed a value of −0.3 ppm, while
the methylene group showed a value of +0.19 ppm, which is significantly
less than that of the similar methylene group in 5. The
more remote isopropyl hydrogen atoms in 6 showed a difference
of less than 0.1 ppm. Finally, the more polar benzyl-protected hydroxymethyl
group of serine 7 caused a significant drop in ΔδPM values: −0.04 ppm for both methylene and benzyl protons
and only +0.08 ppm for the NH unit in the carbamate group. The methyl
groups in the Boc moiety showed no measurable difference among the
diastereomers.In addition, we also added reported values of
analogous esters
with 9-anthrylmethoxyacetic acid (9-AMA) for a comparison to Figure since 9-AMA is also
capable of projecting a strong shielding effect on remote positions.[13] Chemical shift differences of analogous esters
with Mosher’s acid were not reported in the literature.Since the experimental results summarized in Figure suggested a strong influence of the Boc
group, we decided to continue with variously N-substituted
(S)-phenylglycinols (Figure ). First, we prepared N-methylated
Boc-phenylglycinol 8 to remove any possible hydrogen
bonding between the NH group and the hydrogen bond acceptor. The eliminated
hydrogen bonding did not switch the sign distribution according to
the previous model (Figure ), but the observed ΔδPM value of ortho-Ar-H was smaller than that of nonmethylated derivative 1. A decreased ΔδPM value in 8 further indicates the presence of hydrogen bonds in conformational
equilibrium.
Figure 3
N-substituted (S)-phenylglycinols 8–15. ΔδPM values of minor rotamers
are underlined.
N-substituted (S)-phenylglycinols 8–15. ΔδPM values of minor rotamers
are underlined.Dimethylamino derivative 9 without a Boc group fully
followed the previously proposed conformational model (−0.1
for methyl groups and +0.08 and +0.03 for the phenyl ring). The substitution
of the Boc group for a smaller acetyl group in 10 showed
significant differences compared to 1: +0.13 and +0.02
for acetyl and NH, respectively, and −0.2 and −0.11
for aromatic protons with the opposite sign distribution to the conformational
model again.To rule out NH as the hydrogen bond donor, N-methylated
analogue 11 was synthesized and isolated as a mixture
of rotamers, which complicated the structural assignment. Nevertheless,
the sign distribution of the major isomer (+0.17, +0.09, and −0.01)
did not follow the conformational model, as was previously observed
for Boc analogue 8, and moreover, the minor rotamer complicated
the configuration assignment with the irregular sign distribution:
−0.17 for the acetyl group, +0.11 for the methyl group, and
+0.03 for the phenyl ring. Further reduction of sterically bulkier
acetyl to formyl did not offer an improvement. Compound 12 was isolated again as a mixture of two rotamers, showing an ambiguous
sign distribution of ΔδPM for both rotamers.Total N-deprotection of 1 with trifluoroacetic
acid (TFA) yielded aminoester 13, which fully followed
the conformational model, with ΔδPM values
of −0.29 ppm for the amino group and +0.05 and +0.01 for the
phenyl ring.Dibenzyl derivative 14 and phthalimide 15 were prepared to evaluate the influence of synthetically
interesting N-substituted groups that serve as ammonia
equivalents.[14,15]Derivative 14 displayed positive ΔδPM values at the phenyl
ring for the ortho and meta protons
(+0.18 and +0.02, respectively)
and an anomalous value of −0.01 for the most remote para position. The benzyl groups complicated the assignment,
with a zero difference at the aromatic hydrogens and opposing ΔδPM values at the benzylic methylene protons (+0.02 and −0.03).
Uncertain evidence of the absolute configuration by 1H
NMR spectra was arbitrated by the differences in 13C signals.
The quaternary carbon atom in the phenyl ring displayed a ΔδPM value of +0.06 ppm, while a value of −0.07 ppm was
observed for the benzylic methylene carbon atoms.Ester 15 displayed a +0.01 ppm difference at the phthalimide
hydrogen atoms, while the phenyl protons showed ΔδPM values of −0.04, +0.05, and +0.01 ppm for the ortho, meta, and para protons,
respectively. Then, we analyzed the 13C NMR spectra to
resolve the observed inconsistency in the sign distribution of the
ΔδPM values. A positive difference (+0.19 ppm)
was observed for the carbonyl carbon atoms, and a negative difference
(−0.07 ppm) was observed for the quaternary carbon atom in
the phenyl ring.The chemical shift differences of TBBA esters
in Figure clearly
revealed the significant
role of the N-carbonyl moiety present as a carbamate 1–8, amide 10–12, or imide 15 functionality to change the equilibrium of conformers.The supposed effect of the NHBoc group on conformational equilibrium
in a nonpolar solvent was studied with software Spartan 18 (B3LYP-D3/6-31G*)
to identify the theoretical lowest-energy conformers of 1 (Figure ). It was
revealed that the Boc group is always located out of the shielding
zone of the benzimidazole cycle, which is in accordance with the observed
small chemical shift differences of tert-butyl hydrogens.
The position of the phenyl depends on the presence of intramolecular
hydrogen bonds. If formed, the phenyl was positioned inside of the
shielding zone of TBBA in ( (Figure a). Oppositely, the
phenyl was located outside without an intramolecular hydrogen bond
in ( (Figure b). These calculations were in agreement
with the negative difference assigned for the ortho-positioned hydrogens of TBBA ester 1.
Figure 4
Theoretical lowest-energy
conformers of 1 in a nonpolar
solvent (software Spartan 18). The conformer distribution was calculated
with the MMFF model (≤100 kJ/mol), followed by the calculation
of energy at the ground state using DFT in a nonpolar solvent (B3LYP-D3/6-31G*)
to account for long-range nonbonded dispersion interactions. Hydrogens
were omitted for clarity. (a) Most stable conformer of ( (Boltzmann weight: 0.901) with the hydrogen
bond (light-blue dashed line) between the NHBoc and benzimidazole
nitrogen. (b) Most stable conformer of ( (Boltzmann weight: 0.627) without hydrogen bonds. Please see the Supporting Information for more details.
Theoretical lowest-energy
conformers of 1 in a nonpolar
solvent (software Spartan 18). The conformer distribution was calculated
with the MMFF model (≤100 kJ/mol), followed by the calculation
of energy at the ground state using DFT in a nonpolar solvent (B3LYP-D3/6-31G*)
to account for long-range nonbonded dispersion interactions. Hydrogens
were omitted for clarity. (a) Most stable conformer of ( (Boltzmann weight: 0.901) with the hydrogen
bond (light-blue dashed line) between the NHBoc and benzimidazole
nitrogen. (b) Most stable conformer of ( (Boltzmann weight: 0.627) without hydrogen bonds. Please see the Supporting Information for more details.The calculations showed the formation of intramolecular
hydrogen
bonds predominantly between the NHBoc and benzimidazole nitrogen at
position 3. Since the NMR spectra did not show interactions between
fluorine and hydrogen atoms (please see more details in the Supporting Information), we can exclude a strong
hydrogen bonding between these atoms. The positive difference (+0.06)
of NH hydrogen in phenylglycinol 1 (Figure ) can be attributed to the
higher ratio of conformers with intramolecular hydrogen bonds in diastereomer ( compared to ( (please see the Supporting Information for more details).To evaluate the presence of the hydrogen
bond, we conducted simple 1H NMR experiments in acetone-d6 as a possible hydrogen bond acceptor (Figure ).
Figure 5
Comparison of observed
ΔδPM values in CDCl3 (black) and
acetone-d6 (red)
for compounds 1–8. Esters 1 and 4 were also measured in acetonitrile-d3 (blue).
Comparison of observed
ΔδPM values in CDCl3 (black) and
acetone-d6 (red)
for compounds 1–8. Esters 1 and 4 were also measured in acetonitrile-d3 (blue).The solvent change had
a significant effect on the TBBA esters.
Complete inversion of the ΔδPM sign was observed
in two cases (Figure ; compounds 1 and 4), and this effect was
further confirmed in acetonitrile-d3.
A partial inversion of ΔδPM was revealed in
the case of derivatives 3, 5, and 7. Esters 2 and 6 did not show partial
inversion; however, the magnitude of the difference approached the
inverted value. Further evidence of intermolecular hydrogen bonds
between the NH hydrogen and acetone carbonyl illustrates a drop of
all NH differences in esters 1–7 since both diastereomers
(P)- and (M)-TBBA participate more
evenly in hydrogen bonding with acetone. A less significant decrease
of ΔδPM for the NH hydrogens in 7 can originate from competitive formation of an intramolecular hydrogen
bond between the NH and ether oxygen. As expected, no significant
change was observed for ΔδPM in the case of
N-methylated derivative 8. The change in chemical shift
differences in Figure supports the formation of hydrogen bonds, but the influence of steric
hindrance (ester 2 vs 4) is also evident.
Conclusions
In summary, a small library of β-chiral N-Boc aminoalcohol TBBA esters was prepared. Their NMR spectra confirmed
incongruity with the previously reported conformation model for β-chiral
primary alcohols.[12] Further expansion of
the library with variously modified N-substituted
phenylglycinols revealed similar nonconformity. The increased acidity
of the hydrogen atom in the carbamate or amide functionality has a
strong influence on the formation of hydrogen bonds capable of changing
the conformational equilibrium. Moreover, the repulsion between the N-carbonyl moiety and trifluorobenzimidazole ring significantly
impacts the conformer ratio. Both effects cause incorrect assignment
since the resulting predominant conformers differ from the general
conformational model.To conclude, the presence of a carbonyl
group at the N-substituent, where the nitrogen atom
is not a constituent of a ring,
is a limitation of the TBBA method to assign absolute configuration
of β-chiral primary aminoalcohols. Analysis of such compounds
with TBBA should be carried out with caution, and if possible, an
alternative method including N-deprotection is much
better for analyzing aminoalcohols such as TBBA amides with the chiral
carbon at the α-position.[4]
Experimental
Section
All reactions were carried out under normal conditions
without
any specific precautions to exclude moisture or air from the reaction
unless otherwise stated. Reaction workup and column chromatography
(CC) were performed with commercial-grade solvents without further
purification. 1H NMR, 13C NMR, and 19F NMR spectra were measured on a Jeol ECA400II (400 MHz) or Jeol
ECX-500SS (500 MHz) instrument in CDCl3, DMSO-d6, acetone-d6, or acetonitrile-d3 as a solvent. 1H and 13C spectra were calibrated using residual a nondeuterated solvent
as an internal reference (7.26 and 77.16 ppm for CDCl3,
2.50 and 39.52 ppm for DMSO-d6, 2.05 and
29.84 ppm for acetone-d6, and 1.94 and
1.32 ppm for acetonitrile-d3). 19F spectra were calibrated by the addition of CFCl3 as
an internal reference (δ = 0.0 ppm). All 13C NMR
spectra were measured with broad-band 1H decoupling. 1H NMR data are reported as follows: δ, chemical shift;
coupling constants (J are given in hertz, Hz), and integration. Abbreviations
to denote the multiplicity of a particular signal were s (singlet),
d (doublet), t (triplet), q (quartet), m (multiplet), app (appears
as), and br (broad).Analytical thin-layer chromatography (TLC)
was performed using
Kieselgel 60 F254 plates (Merck). Compounds were detected
by UV light (255 nm) and then by basic KMnO4 solution.
Flash chromatography was performed using silica gel (35–70
μm particle size). HRMS analyses were carried out using an Exactive
Plus Orbitrap high-resolution mass spectrometer with electrospray
ionization (Thermo Fisher Scientific, MA, USA). Chromatographic pre-separation
was performed using a HPLC system Dionex UltiMate 3000 (Thermo Fisher
Scientific, MA, USA) equipped with a Phenomenex Gemini column (C18,
50 × 2 mm, 3.0 μm). The samples were dissolved in MeOH
or acetonitrile and injected by an autosampler. Mobile phase compositions:
isocratic elution of MeOH/water 95:5 + 0.1% (v/v) HCOOH with a flow
rate of 0.3 mL/min.The calculation of theoretical lowest-energy
conformers was done
using Spartan 18 (Wavefunction, USA). The conformer distribution was
calculated with molecular mechanics applying the MMFF force field
(≤100 kJ/mol), followed by the calculation of energy at the
ground state using DFT in a nonpolar solvent with the B3LYP-D3 functional
(6-31G* basis set) to account for long-range nonbonded (dispersion)
interactions.
General Procedure for Esters 1–12, 14, and 15
TBBA (15 mg, 0.05 mmol, 1
equiv) was dissolved in dry DCM (1.5 mL). Then, alcohol (1 equiv,
0.05 mmol), DMAP (6 mg, 0.05 mmol, 1 equiv), and DCC (11 mg, 0.05
mmol, 1 equiv) were added. The mixture was stirred at room temperature
for 16 h. After that, the precipitate was filtered off. The resulting
filtrate was washed twice with 10% aq. HCl (2 mL) and 10% K2CO3 (2 mL) and once with brine and dried with MgSO4. After evaporation of DCM, the residue was purified by CC.
The washing steps can be skipped if the analyte contains labile functional
groups.
Following the literature procedure[16](S)-Phenylglycine methylester hydrochloride
(603 mg, 3 mmol, 1 equiv) was dissolved in DI water (10 mL), and aq.
K2CO3 solution was added (10 mL, 10 wt %). The
solution was extracted with diethylether (3 × 20 mL), dried with
MgSO4, and evaporated to yield freebase (S)-phenylglycine methylester [360 mg of a clear oil (70%)]. This oil
was dissolved in formic acid (30 mL) and cooled in an ice bath. Acetic
anhydride (8.3 mL) was added dropwise while cooling. After the addition
was complete, the reaction mixture was stirred for 16 h. After 16
h, DI water was added (20 mL) and the solution was stirred for 20
min and evaporated. The oily residue was dissolved in EtOAc (50 mL)
and extracted with 10% aq. HCl (3 × 50 mL) and 10% aq. K2CO3 (3 × 50 mL), dried with MgSO4, and evaporated to yield a clear oil, which solidified upon standing
on room temperature or under high vacuum. Yield: 371 mg of a white
solid (75%). The reaction was reproduced on a 10 mmol scale, yielding
1.2 g (65%) of a white solid. 1H NMR (400 MHz, CDCl3): δ 8.25 (s, 1H), 7.38–7.33 (m, 5H), 6.60 (s,
1H), 5.67 (d, J = 7.4 Hz, 1H), 3.75 (s, 3H). 13C NMR {1H} (101 MHz, CDCl3): δ
171.1, 160.2, 136.2, 129.2, 128.9, 127.3, 55.2, 53.1. HRMS (ESI) m/z: [M + H]+ calcd for C10H12N3O1, 194.0812; found,
194.0813. +87.62 (c 0.42, CHCl3).
(S)-2-(Methylamino)-2-phenylethan-1-ol
Modified literature procedure[16]Methyl (S)-2-formamido-2-phenylacetate (400
mg, 2 mmol, 1 equiv) was added portionwise to a suspension of LiAlH4 (380 mg, 10 mmol, 5 equiv) in dry THF (15 mL) at 5 °C
(ice/water bath). After addition was completed, the mixture was refluxed
for 16 h. After reaction completion (TLC, EtOAc/MeOH 2:1), the reaction
mixture was cooled to room temperature and further cooled in an ice
bath, and aq. NaOH solution (15% by weight, 0.75 mL/mmol LiALH4) was added dropwise. The resulting suspension was filtered
through Celite and washed thoroughly with EtOAc, dried with MgSO4, and evaporated. The residual oil was purified by CC (EtOAc/MeOH
2:1), yielding 242 mg of a white solid (80%). The reaction was reproduced
on a 6.2 mmol scale, yielding a white solid, which was suspended in
chloroform and filtered, and after evaporation, 800 mg (85%) of a
white solid was obtained. 1H NMR (400 MHz, CDCl3): δ 7.39–7.34 (m, 2H), 7.32–7.27 (m, 3H), 3.75
(dd, J = 10.1, 4.1 Hz, 1H), 3.71–3.66 (m,
1H), 3.61 (dd, J = 10.0, 8.0 Hz, 1H), 2.69 (s, 2H),
2.36 (s, 3H). 13C NMR {1H} (101 MHz, CDCl3): δ 179.1, 129.4, 129.2, 128.2, 66.2, 64.5, 31.5, 23.8.
HRMS (ESI) m/z: [M + H]+ calcd for C9H14NO, 152.1070; found, 152.1070. +39.89 (c 0.88, CHCl3).
Following
the literature procedure[17](S)-2-(Methylamino)-2-phenylethan-1-ol (40
mg, 0.25 mmol, 1 equiv) was dissolved in EtOAc (10 mL). Boc2O was added at once, and the mixture was refluxed for 16 h. After
16 h, the reaction mixture was cooled to room temperature, washed
twice with water and once with brine, dried with MgSO4,
and evaporated to provide 53 mg of an oily product (85%). 1H NMR (400 MHz, CDCl3): δ 7.36–7.31 (m, 2H),
7.30–7.26 (m, 1H), 7.24–7.21 (m, 2H), 5.32–5.24
(m, 1H), 4.11–4.01 (m, 2H), 2.69 (s, 2H). 13C NMR
{1H} (101 MHz, CDCl3): δ 146.9, 128.8,
127.8, 127.5, 85.3, 80.4, 60.6, 28.6, 27.6. HRMS (ESI) m/z: [M + H]+ calcd for C14H22NO3, 252.1594; found, 252.1595. +55.17 (c 0.6, CHCl3).
(S)-2-(Dibenzylamino)-2-phenylethan-1-ol
Following the literature procedure[18](S)-Phenylglycinol (137 mg, 1 mmol, 1 equiv)
was dissolved in acetonitrile (7 mL). K2CO3 (280
mg, 2 mmol, 2 equiv) was added, followed by benzyl bromide (250 μL,
2.1 mmol, 2.1 equiv). The reaction mixture was stirred at 60 °C
for 24 h. After the reaction was complete (TLC, hexane/EtOAc 4:1),
the reaction mixture was filtered and the filtrate was evaporated
and purified by CC (hexane/EtOAc, gradient from 10:1 to 8:1). Isolated
as a colorless oil (199 mg, 62%). 1H NMR (400 MHz, CDCl3): δ 7.48–7.34 (m, 3H), 7.34 (d, J = 4.4 Hz, 8H), 7.27 (q, J = 3.9, 3.2 Hz, 4H), 4.14
(t, J = 10.6 Hz, 1H), 3.96–3.93 (m, 1H), 3.62
(dd, J = 10.8, 5.2 Hz, 1H), 3.16 (d, J = 13.4 Hz, 1H). 13C NMR {1H} (101 MHz, CDCl3): δ 139.3, 135.3, 129.4, 129.1, 128.7, 128.5, 128.2,
127.4, 63.2, 60.6, 53.7. HRMS (ESI) m/z: [M + H]+ calcd for C22H24NO, 318.1852;
found, 318.1853. +122.33 (c 0.6, CHCl3).
Following the literature procedure[19](S)-Phenylglycinol (420 mg, 3 mmol, 1 equiv)
was suspended in toluene (10 mL). Phthalic anhydride (450 mg, 3 mmol,
1 equiv) was added, followed by triethylamine (50 μL, 0.3 mmol,
0.1 equiv). The reaction mixture was refluxed for 16 h and then cooled
to room temperature and evaporated, and the residue was dissolved
in EtOAc (25 mL) and extracted with 10% aq. HCl (3 × 25 mL) and
10% aq. K2CO3 (3 × 25 mL). The combined
organic layers were washed with brine, dried with MgSO4, and evaporated. The residue was purified by CC (hexane/EtOAc 2:1).
Yield 390 mg (50%). 1H NMR (400 MHz, CDCl3):
δ 7.83 (dd, J = 5.6, 3.2 Hz, 2H), 7.71 (td, J = 5.3, 2.1 Hz, 2H), 7.46 (dd, J = 6.9,
1.5 Hz, 2H), 7.39–7.23 (m, 3H), 5.47 (dd, J = 9.0, 5.0 Hz, 1H), 4.65 (dd, J = 11.7, 8.9 Hz,
1H), 4.24 (dd, J = 11.7, 5.0 Hz, 1H). 13C NMR {1H} (101 MHz, CDCl3): δ 169.0,
137.0, 134.3, 132.0, 128.9, 128.3, 128.0, 123.6, 62.5, 57.7. HRMS
(ESI) m/z: [M + H]+ calcd
for C16H14NO3, 268.0968; found, 268.
0967. −45.17 (c 0.29,
CHCl3).
(S)-2-Acetamido-2-phenylethyl
Acetate
(S)-Phenylglycinol (670
mg, 5 mmol, 1 equiv)
and DMAP (70 mg, 0.5 mmol, 0.1 equiv) were dissolved in Ac2O (7 mL) and stirred at room temperature for 2.5 h. After 2.5 h,
the solution was added dropwise into an aq. solution of K2CO3 (10%, 15 mL). The solution was further neutralized
with solid K2CO3 until pH = 7 and then extracted
into DCM (3 × 30 mL). Organic layers were combined and dried
with MgSO4 and evaporated to yield a white solid (573 mg,
50%). 1H NMR (500 MHz, CDCl3): δ 7.37–7.33
(m, 2H), 7.31–7.27 (m, 3H), 6.09 (d, J = 6.4
Hz, 1H), 5.29 (td, J = 7.6, 4.7 Hz, 1H), 4.43 (dd, J = 11.5, 7.2 Hz, 1H), 4.26 (dd, J = 11.5,
4.7 Hz, 1H), 2.05 (s, 3H), 2.02 (s, 3H). 13C NMR {1H} (126 MHz, CDCl3): δ 171.4, 169.7, 138.5,
129.0, 128.1, 126.8, 66.2, 52.7, 23.5, 21.0. HRMS (ESI) m/z: [M + H]+ calcd for: C12H16NO3, 222.1130; found, 222.1125. +80.77 (c 0.13, CHCl3).
(S)-N-(2-Hydroxy-1-phenylethyl)acetamide
(S)-2-Acetamido-2-phenylethyl acetate (300
mg, 1.35 mmol, 1 equiv) was dissolved in MeOH (15 mL). Then, a solution
of NaOH (270 mg, 6.75 mmol, 5 equiv dissolved in 5 mL of DI water)
was added and stirred at room temperature for 12 h. After 12 h, the
mixture was filtered through a pad of Celite and washed with 30 mL
of EtOAc/MeOH (1:1), and the filtrate was dried with MgSO4 and evaporated to yield a white solid (228 mg, 93%). 1H NMR (500 MHz, CDCl3): δ 7.37–7.33 (m, 2H),
7.31–7.27 (m, 3H), 6.38 (d, J = 3.5 Hz, 1H),
5.04 (dt, J = 7.1, 5.1 Hz, 1H), 3.85 (d, J = 5.1 Hz, 2H), 3.12 (s, 1H), 2.02 (s, 3H). 13C NMR {1H} (126 MHz, CDCl3): δ 171.0,
139.1, 129.0, 128.0, 126.9, 66.6, 56.1, 23.4 HRMS (ESI) m/z: [M + H]+ calcd for: C10H14NO2, 180.1019; found, 180.1019. +45.26 (c 0.19, CHCl3).
(S)-N-(2-Hydroxy-1-phenylethyl)-N-methylacetamide
Following the literature procedure[20](S)-N-Methyl-phenylglycinol
(50 mg, 0.33 mmol, 1 equiv) was dissolved in DCM (1.5 mL), and acetylchloride
(30 μL, 0.4 mmol, 1.2 equiv) was added, followed by a dropwise
addition of 0.5 M NaOH (840 μL, 0.4 mmol, 1.2 equiv). The biphasic
system was stirred vigorously for 1 h. Then, the mixture was diluted
with water (10 mL) and extracted with DCM (3 × 10 mL). Organic
layers were combined, dried with MgSO4, and purified by
CC (EtOAc/MeOH 20:1) to yield 50 mg of a white solid (78%) as a mixture
of rotamers in a 10:4 ratio. Peaks belonging to the major rotamer
are designated as M, and peaks belonging to a minor rotamer are designated
as m. 1H NMR (400 MHz, CDCl3): δ 7.41–7.19
(m, 10H, both rotamers), 5.83 (dd, J = 9.3, 4.9 Hz,
1H, M), 5.09 (dd, J = 9.2, 4.9 Hz, 1H, m), 4.23–4.02
(m, 4H, both rotamers), 2.78 (s, 4H, both rotamers), 2.42 (dd, J = 7.2, 4.7 Hz, 1H, M), 2.28 (s, 3H, m), 2.19 (s, 3H, M),
2.15–2.05 (m, 1H, m). 13C NMR {1H} (126
MHz, CDCl3): δ 172.8 m, 172.4 m, 137.2 m, 137.0 m,
129.1 m, 128.8 m, 128.2 m, 127.93 m, 127.89 m, 127.0 m, 62.6 m, 61.9
m, 61.5 m, 58.4 m, 32.0 m, 28.1 m, 22.5 m, 22.2 m. HRMS (ESI) m/z: [M + H]+ calcd for: C11H16NO2, 194.1176; found, 194.1176. −440.0 (c 0.13,
CHCl3).
(S)-2-(N-Methylformamido)-2-phenylethyl
Formate
Following the literature procedure[21]HCOOH (150 μL, 4 mmol, 4 equiv)
dissolved in CHCl3 (2 mL) was added dropwise under cooling
into a solution of
DCC (412 mg, 2 mmol, 2 equiv) in CHCl3 (3 mL). After 5
min, a white suspension was added dropwise into the solution of (S)-N-methyl-phenylglycinol (151 mg, 1 mmol,
1 equiv) in a mixture of CHCl3 (3 mL) and pyridine (1.5
mL) and stirred in an ice bath for 16 h. After 16 h, the reaction
mixture was evaporated, suspended in diethylether (10 mL), and filtered,
and the filtrate was evaporated. The residue was then dissolved in
ethylacetate and extracted twice with 10% HCl, 10% K2CO3, and brine; dried with MgSO4; and purified by
CC (hexane/EtOAc 1:1) to yield 100 mg of an oily product (50%) as
a mixture of two rotamers in a 10:6 ratio. Peaks belonging to a major
rotamer are designated as M, and peaks belonging to a minor rotamer
are designated as m. 1H NMR (500 MHz, CDCl3):
δ 8.31 (s, 1H, M), 8.18 (s, 1H, m), 8.10 (s, 1H, M), 8.08 (s,
1H, m), 7.43–7.31 (m, 5H, both rotamers), 7.29–7.22
(m, 5H, both rotamers), 5.91 (dd, J = 9.5, 5.3 Hz,
1H, m), 4.91 (dd, J = 10.0, 4.6 Hz, 1H, M), 4.79–4.72
(m, 1H, both rotamers), 4.67–4.62 (m, 1H, both rotamers), 2.76
(s, 3H, m), 2.69 (s, 3H, M). 13C NMR {1H} (126
MHz, CDCl3): δ 163.5 m, 163.1 m, 160.6 m, 160.4 m,
135.2 m, 134.8 m, 129.3, 129.1, 128.9, 128.6, 127.9, 127.2, 60.8 m,
59.6 m, 52.7 m, 49.3 m, 34.1, 30.6 HRMS (ESI) m/z: [M + H]+ calcd for: C11H14NO3, 208.0968; found, 208.0967. +89.47 (c 0.19, CHCl3).
(S)-N-(2-Hydroxy-1-phenylethyl)-N-methylformamide
(S)-2-(N-Methylformamido)-2-phenylethyl
formate (80 mg, 0.38 mmol, 1 equiv) was dissolved in MeOH (8 mL),
and NH3 was added (25% aq. solution, 90 μL, 1.15
mmol, 3 equiv). The reaction mixture was stirred at room temperature
for 2 h. Then, the solvent was evaporated. The resulting residue was
dissolved in EtOAc and extracted with brine three times. The organic
layer was separated, dried with MgSO4, and evaporated.
The residue was purified by CC (EtOAc) to yield 21 mg (30%) of a colorless
oil as a mixture of two rotamers in an aprox. 10:6 ratio. Peaks belonging
to the major rotamer are designated as M, and peaks belonging to the
minor rotamer are designated as m. 1H NMR (500 MHz, CDCl3): δ 8.33 (s, 1H, M), 8.21 (s, 1H, m), 7.40–7.22
(m, 10H, both rotamers), 5.41 (dd, J = 8.4, 5.4 Hz,
1H, m), 4.68 (dd, J = 8.7, 5.3 Hz, 1H, M), 4.17–4.08
(m, 4H, both rotamers), 2.80 (s, 3H, m), 2.70 (s, 3H, M). 13C NMR {1H} (126 MHz, CDCl3): δ 164.3,
163.9, 136.2, 136.1, 129.1, 129.0, 128.5, 128.3, 127.9, 127.4, 63.5,
61.6, 60.7, 58.7, 32.1, 26.6. HRMS (ESI) m/z: [M + H]+ calcd for: C10H14O2N1, 180.1019; found, 180.1019. +41.51 (c 0.21, CHCl3).
(S)-2-(Dimethylamino)-2-phenylethan-1-ol
Following the literature procedure[22](S)-Phenylglycinol (550 mg, 4 mmol, 1 equiv)
was dissolved in HCOOH (0.6 mL), and formaldehyde was added (38% aq.
solution, 0.6 mL). The reaction mixture was heated at 90 °C for
16 h. After 16 h, the solution was cooled to room temperature, neutralized
with the ammonia solution (25% aq. solution, 0.5 mL), and extracted
three times with DCM. The organic phases were combined and dried with
MgSO4, evaporated, and purified by CC (EtOAc/MeOH 20:1)
to yield 400 mg (60%) of a brown oil, which solidified after standing
at room temperature. 1H NMR (400 MHz, CDCl3):
δ 7.41–7.28 (m, 3H), 7.25–7.16 (m, 2H), 3.93 (dd, J = 10.7, 9.0 Hz, 1H), 3.68 (dd, J = 10.6,
5.3 Hz, 1H), 3.57 (dd, J = 9.0, 5.3 Hz, 1H), 2.21
(s, 3H). 13C NMR {1H} (101 MHz, CDCl3): δ 135.9, 129.1, 128.3, 128.0, 70.3, 61.4, 41.5. HRMS (ESI) m/z: [M + H]+ calcd for: C10H16NO, 166.1266; found, 166.1266. +32.5 (c 0.36, CHCl3).
Authors: José M. Andrés; Roberto Barrio; María A. Martínez; Rafael Pedrosa; Alfonso Pérez-Encabo Journal: J Org Chem Date: 1996-06-26 Impact factor: 4.354
Authors: Michal Kriegelstein; David Profous; Antonín Lyčka; Zdeněk Trávníček; Adam Přibylka; Tereza Volná; Sandra Benická; Petr Cankař Journal: J Org Chem Date: 2019-08-30 Impact factor: 4.354