Organic Syntheses, Vol. 77, 64
Checked by Mark M. Gleason and William R. Roush.
1. Procedure
2. Notes
1. The glass components of the apparatus were dried overnight in a 150°C oven and allowed to cool in a desiccator over a drying agent before assembly.
2.
Acetyl chloride was purchased from the Acros Chimica and distilled before use. The checkers purchased
acetyl chloride from Aldrich Chemical Company, Inc., and used it without purification.
Caution! Acetyl chloride is a reactive substance that must be handled in a fume hood.
4.
L-Serine was purchased from the Acros Chimica or Aldrich Chemical Company, Inc., and used without purification.
5.
Di-tert-butyl dicarbonate was purchased from the Acros Chimica or Aldrich Chemical Company, Inc., and used without purification.
7.
Triethylamine was purchased from the Acros Chimica or Aldrich Chemical Company, Inc., and used without purification.
8. Submitters report an optical rotation value for crude N-Boc-
L-serine methyl ester
2 of
[α]D −19.1° (
MeOH,
c 4.07), very close to that reported by McKillop, et al.
2 (lit.
2 [α]D −18.9° (
MeOH,
c 5.0)). Checkers report the following data for
2:
[α]D23 17.0° (MeOH,
c 4.41);
1H NMR (300 MHz, CDCl
3) δ: 1.42 (s, 9 H), 3.03 (br s, 1 H), 3.75 (s, 3 H), 3.84 (dd, 1 H, J = 11, 3.3), 3.93 (br d, 1 H, J = 8.1), 4.33-4.36 (m, 1 H), 5.55 (br d, 1 H, J = 7.5);
13C NMR (75 MHz, CDCl
3) δ: 28.5, 52.7, 55.7, 63.3, 80.2, 155.6, 171.2; IR (neat) cm
−1: 3400, 1717. Anal. Calcd for C
9H
17NO
5: C, 49.31; H, 7.82; N, 6.39. Found C, 49.51; H, 7.86; N, 6.21.
11. TLC analysis on
silica gel 60F-254 plates eluting with (1:1)
cyclohexane-ethyl acetate showed the clean formation of ester
3 with R
f = 0.74 (visualized with
0.3% ninhydrin in (97:3)
butanol-acetic acid) at the expense of starting material with R
f = 0.4. The sample of the oxazolidine ester was neutralized with a little
triethylamine prior to TLC analysis.
12. Submitters report an optical rotation for the crude oxazolidine methyl ester
3 of
[α]D −54.4° (
CHCl3,
c 1.07), nearly identical to that found by McKillop, et al.
2 [lit.
2 [α]D −54.0° (
CHCl3,
c 1.3)]. Purification with flash chromatography on
silica gel eluting with (85:15)
cyclohexane-
ethyl acetate gave a product with a maximum rotation of
[α]D −58.3° (
CHCl3,
c 0.86), very close to that reported by Garner, et al.
3 4 (lit.
3,4 −57°). Checkers report the following data for
3:
[α]D23 −53.5° (
CHCl3,
c 1.05);
1H NMR (400 MHz, C
6D
6, 75°C) δ: 1.39 (s, 9 H), 1.54 (br s, 3 H), 1.82 (br s, 3 H), 3.34 (s, 3 H), 3.74 (m, 1 H), 3.80 (dd, 1 H, J = 8.8, 3.2), 4.26 (m, 1 H);
13C NMR (100 MHz, C
6D
6, 75°C) δ: 24.7, 25.3, 28.4, 51.6, 59.8, 66.4, 79.9, 95.5, 151.4, 171.3; IR (neat) cm
−1: 2980, 1759, 1708. Anal. Calcd for C
12H
21NO
5: C, 55.58; H, 8.16; N, 5.40. Found: C, 55.52; H, 8.29; N, 5.44.
13.
Lithium aluminum hydride was purchased from Acros Chimica or Aldrich Chemical Company, Inc., and used without further purification.
14. TLC analysis on
silica gel plates eluting with (7:3)
cyclohexane-
ethyl acetate showed the clean formation of the alcohol with R
f = 0.24 (visualized with
0.3% ninhydrin in (97:3)
butanol-
acetic acid) at the expense of the starting material with R
f = 0.55.
15. Checkers used an aqueous NaH
2PO
4/Na
2HPO
4 buffer with a phosphate concentration of ca. 0.5 M.
16. Submitters found the optical rotation of the crude alcohol to be
[α]D −23.9° (
CHCl3,
c 1.0). Purification with flash chromatography on
silica gel eluting with (7:3)
cyclohexane-
ethyl acetate gave a colorless syrup that solidified upon cold storage [mp
45-46°C;
[α]D −26.7° (
CHCl3,
c 1.0)], [lit.
3 mp
38-39°C;
[α]D −24.0° (
CHCl3,
c 1.61)]. Occasionally the compound crystallized at room temperature to give colorless prisms with a maximum mp of
49-51°C. Checkers report the following data for
4:
[α]D23 −26.2° (
CHCl3,
c 0.79);
1H NMR (400 MHz, C
6D
6, 70°C) δ: 1.37 (s, 9 H), 1.43 (br s, 3 H), 1.55 (br s, 3 H), 3.20 (br s, 1 H), 3.48 (m, 1 H), 3.63-3.68 (m, 3 H), 3.87 (m, 1 H);
13C NMR (100 MHz, C
6D
6, 70°C) δ: 24.3, 27.3, 28.4, 59.7, 64.0, 65.5, 80.1, 94.1, 153.4; IR (neat) cm
−1: 3430, 1699, 1380, 1260, 1174, 1050, 848. Anal. Calcd for C
11H
21NO
4: C, 57.12; H, 9.15; N, 6.00. Found: C, 56.84; H, 9.10; N, 6.06.
18.
Dimethyl sulfoxide was purchased from Acros Chimica or Aldrich Chemical Company, Inc., and distilled under reduced pressure before use.
19.
Oxalyl chloride was purchased from Acros Chimica or Aldrich Chemical Company, Inc., and distilled under an atmosphere of
nitrogen before use.
Caution! Oxalyl chloride is a reactive substance that must be handled in a fume hood.
21. The optical rotation of the crude aldehyde was
[α]D −89.7° (
CHCl3,
c 1.65). Purification by flash chromatography on
silica gel eluting with (4:1)
cyclohexane-
ethyl acetate gave a product with a rotation of
[α]D −95.5° (
CHCl3,
c 0.78), (lit.
3,4 [α]D −105°). Checkers report the following data for
5:
[α]D23 −93.3° (
CHCl3,
c 1.10);
1H NMR (400 MHz, C
6D
6, 70°C) δ: 1.34 (s, 9 H), 1.40 (br s, 3 H), 1.58 (br s, 3 H), 3.57 (d, 1 H, J = 7.6), 3.67 (d, 1 H, J = 7.6), 3.91 (m, 1 H), 9.33 (br s, 1 H);
13C NMR (100 MHz, C
6D
6, 70°C) δ: 24.1, 26.0, 28.3, 63.7, 65.1, 80.6, 95.0, 151.8, 198.0; IR (neat) cm
−1: 1739, 1710. Anal. Calcd for C
11H
19NO
4: C, 57.63; H, 8.35; N, 6.11. Found: C, 57.59; H, 8.27; N, 5.91.
22. Enantiomeric purity was determined to be 96-98% by
1H NMR analysis of the Mosher esters
5 of the alcohols
4 and ent-
4 obtained by reduction of the aldehydes
5 and ent-
5. To an ice-cold solution of
aldehyde 5 (0.10 g, 0.44 mmol) in
5 mL of methanol was added solid
sodium borohydride (33 mg, 0.88 mmol). After the mixture was stirred for 30 min at this temperature, the TLC in (7:3)
cyclohexane-
ethyl acetate showed the clean formation of the alcohol
4. The mixture was treated with
0.05 mL of acetone and concentrated to dryness under reduced pressure. The residue was partitioned between water (10 mL) and
ethyl acetate (10 mL) and the phases were separated. The aqueous phase was extracted with three
10-mL portions of ethyl acetate. The combined organic phases were dried with anhydrous
sodium sulfate and concentrated under reduced pressure. The residue was purified by flash chromatography on
silica gel eluting with (7:3)
cyclohexane-
ethyl acetate to give
71 mg (
70% yield) of pure alcohol
4 as a colorless oil:
[α]D −26.4° (
CHCl3,
c 0.58). To a solution of the alcohol (50 mg, 0.22 mmol),
N,N'-dicyclohexylcarbodiimide (54 mg, 0.26 mmol) and
4-dimethylaminopyridine (3.0 mg, 0.025 mmol) in dry methylene chloride (0.5 mL) was added a solution of
(R)-(+)-α-methoxy-α-(trifluoromethyl)phenylacetic acid (61 mg, 0.26 mmol, Aldrich Chemical Company, Inc.) in 0.26 mL of dry methylene chloride. The mixture was stirred overnight (14 hr) at room temperature, filtered to remove
N,N'-dicyclohexylurea, and partitioned between
ethyl acetate (3 × 5 mL) and water (5 mL). The combined organic phases were washed with
5-mL each of 1 M hydrochloric acid, water, and saturated aqueous
sodium bicarbonate solution, then dried with anhydrous
sodium sulfate and concentrated under reduced pressure. The residue was purified by flash chromatography over
silica gel, eluting with (9:1)
cyclohexane-
ethyl acetate, to give
83 mg (
84% yield) of product with the following properties:
[α]D +11.1° (
CHCl3,
c 0.7);
1H NMR (300 MHz, DMSO-d
6, 80°C) δ: 1.42 (s, 6 H), 1.44 (s, 9 H), 3.48 (s, 3 H), 3.69 (dd, 1 H, J = 2.5, 9.4), 3.96 (dd, 1 H, J = 6.0, 9.4), 4.01-4.11 (m, 1 H), 4.28 (dd, 1 H, J = 7.5, 10.5), 4.48 (dd, 1 H, J = 3.3, 10.5), 7.49 (s, 5 H). The same procedure was performed with ent-
5. The resulting ester showed the following properties:
[α]D +53.6° (
CHCl3,
c 0.68);
1H NMR (300 MHz, DMSO-d
6, 80°C) δ: 1.37 (s, 3 H), 1.41 (s, 3 H), 1.43 (s, 9 H), 3.50 (s, 3 H), 3.73 (dd, 1 H, J = 2.0, 9.4), 3.98 (dd, 1 H, J = 6.5, 9.4), 4.04-4.14 (m, 1 H), 4.28 (dd, 1 H, J = 7.3, 10.5), 4.46 (dd, 1 H, J = 3.1, 10.5), 7.49 (s, 5 H).
All toxic materials were disposed of in accordance with "Prudent Practices in the Laboratory"; National Academy Press; Washington, DC, 1995.
3. Discussion
Since the first appearance in the literature,
6 7 the ingeniously protected serine-derived aldehyde
5 (Garner aldehyde) has attracted considerable attention as a model chiral α-amino carbonyl compound for stereochemical studies,
8 9 10 11 12 13 14 and a precursor to interesting biologically active compounds such as amino sugars,
15 16 aza sugars,
17 sphingosines,
18 19 20 21 22 23 24 25 and unusual amino acids.
26 27 28 29 30 31 32 Compound
5 was designed
6,7 to meet some essential requirements for wide application in synthesis. These include: (1) easy and large scale preparation, (2) configurational and chemical stability, (3) high stereoselectivity during addition reactions, (4) easy and selective removal of O- and N-protective groups. Almost all these features occur and greatly increased the numerous synthetic applications of aldehyde
5 over the years. Nevertheless the issue regarding feature (1) is worth reconsideration since it is crucial to the exploitation of the advantages associated with features (2)-(4). The aim of this work is to provide an improved preparation of
5, partly along the lines of the previous procedures reported by Garner
3,4 and McKillop
2 and their co-workers, partly by a new reaction sequence described by Roush and Hunt,
16 and by the submitters.
33
The original Garner preparation
3,4 of
5 involves the conversion of
serine into the protected methyl ester
3 and controlled reduction of the latter by
DIBAL. The reaction sequence employed for the preparation of
3 involves the protection of the amino acid as N-Boc derivative using di-tert-butyl dicarbonate, esterification with
methyl iodide or
diazomethane, and acetonization with
2,2-dimethoxypropane under acid catalysis. The N-Boc methyl serinate and the ester
3 require purification by vacuum distillation or chromatography. In a modification to this procedure reported by McKillop,
2 the esterification reaction of
serine is carried out first by
methanol/
acetyl chloride. The resulting ester is then converted into the N-Boc derivative
2 with di-tert-butyl dicarbonate and the latter transformed into
3 by acetonization. This procedure avoids the use of
methyl iodide or
diazomethane and the toxic solvent
benzene and gives ester
3 pure enough for the reduction by DIBAL according to the Garner procedure above. Roush
16 and the submitters
33 have observed that the DIBAL reduction of
3 leads to a mixture of the aldehyde
5, primary alcohol
4, and unreacted methyl ester
3 that were difficult to separate. Therefore it proved more convenient to obtain aldehyde
5 by a two-stage reduction-oxidation sequence. Thus, Roush
16 reported the reduction of
3 to the protected serinol
4 by the use of
lithium aluminum hydride and Swern oxidation of the latter to
5 with DMSO/(COCl)
2 in the presence of
triethylamine. While the chemical yield of
5 was quite good (
85%) the enantiomeric purity was determined to be 86-87%, much lower than that reported by the Garner method (
93-95%).
Copyright © 1921-2002, Organic Syntheses, Inc. All Rights Reserved