A new stereoselective synthesis of sterically constrained uncommon α,α'-dialkylated α-amino acids. Part 2

Article abstract of DOI:10.1016/S0957-4166(98)00360-7

The alkylation of the diastereomeric mixture of the chiral morpholinone derivatives 5 and 6 occurs with good yield and a prevalence of the cis isomer. Cleavage of the alkylated intermediates 7b,c yields enantiomerically pure sterically constrained uncommo

Full text of DOI:10.1016/S0957-4166(98)00360-7

TETRAHEDRON:
ASYMMETRY
Pergamon
Tetrahedron: Asymmetry 9 (1998) 3411–3420
A new stereoselective synthesis of sterically constrained
uncommon α,α0-dialkylated α-amino acids. Part 2 †
Gianni Porzi and Sergio Sandri ∗
Dipartimento di Chimica ‘G. Ciamician’, Via Selmi 2, Università degli Studi di Bologna, Bologna, Italy
Received 24 July 1998; accepted 7 September 1998
Abstract
The alkylation of the diastereomeric mixture of the chiral morpholinone derivatives 5 and 6 occurs with good
yield and a prevalence of the cis isomer. Cleavage of the alkylated intermediates 7b,c yields enantiomerically
pure sterically constrained uncommon α,α0-dialkylated α-amino acids. The absolute configuration of the new
stereocentres has been assigned on the basis of the 1H NMR spectra and NOE measurements. A model to explain
the observed diastereoselection is proposed. © 1998 Elsevier Science Ltd. All rights reserved.
1. Introduction
In a previous paper1 we described a new enantioselective approach to the synthesis of uncommon
α,α0-dialkylated α-amino acids, which have recently received considerable attention through the stereo-
selective alkylation of (3R,6S)-4-N[(S)-1-phenethyl]-3,6-dimethyl-1,4-morpholin-2,5-dione 1 reported in
Scheme 1.
Scheme 1.
∗
†
Corresponding author. E-mail: porzi@ciam.unibo.it
Ref. 1 is considered to be Part 1.
0957-4166/98/$ - see front matter © 1998 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(98)00360-7

3412
G. Porzi, S. Sandri / Tetrahedron: Asymmetry 9 (1998) 3411–3420
Therefore, as a continuation of our studies, we have directed our attention to investigating the influence
of the chiral auxiliary configuration (the phenethyl group) on the stereochemical outcome of this reaction.
2. Results and discussion
To this end, the alkylation was also performed on the morpholinone derivative 4 in which the (N-4)-
phenethyl group possesses the R instead of the S configuration. By comparing the stereochemical results
reported in Scheme 2 to those previously found for the alkylation of diastereomer 11 and summarized
in Scheme 1, it is evident that the change in the auxiliary configuration does not produce any effect on
the diastereomeric ratio of the reaction products: in fact, the alkylation at the C-6 occurs with the same
diastereoselectivity (1,4-trans induction with respect to the (C-3)–CH3).
Scheme 2.
The diastereomer 6 was presumably produced owing to the isomerization of the stereocentre C-3 of
the prevalent cis isomer 5, as previously1 observed for the alkylation of 1.
Thus, the alkylation appears prevalently stereocontrolled by the configuration of the C-3 stereocentre,
probably because the chiral auxiliary is remote enough from the C-6 centre to induce appreciable control
during the nucleophilic attack of the enolate ion on the alkylating reagent.
This result prompted us to also investigate the influence of the (R)-phenethyl group on the alkylation
of the diastereomeric mixture 5+6 (Scheme 3) in order to compare the stereochemical results with those
previously registered for the analogous diastereomeric mixture 2.1
Scheme 3.
The data reported in Scheme 3 show that in this case the cis isomer predominates, in contrast to the
previous1 observation for the alkylation of the diastereomeric mixture 2. A cis-induction by the (C-
6)–CH2Ph group is hardly conceivable, so we carried out the alkylation of compound 9, lacking the C-6
stereocentre, in order to clarify the influence of the chiral auxiliary configuration on the alkylation at C-3
(Scheme 4). The substrate 9 was obtained by alkylating 4 with methyl iodide.

G. Porzi, S. Sandri / Tetrahedron: Asymmetry 9 (1998) 3411–3420
3413
Scheme 4.
The C-3 stereocentre of the prevalent diastereomer 10 possesses the same configuration as 7c (see
Scheme 5) but the opposite to that of the more abundant isomer obtained by the alkylation of di-
astereomeric mixture 21 (Scheme 1). The absolute configurations of the new stereocentres have been
1
determined by H NMR data (see later the stereochemical assignments) using the approach previously
employed for similar molecules.1,3
Scheme 5.
However, in a model study, the compounds 7b,c and 10 were converted into the corresponding α,α0-
dialkyl α-amino acids following the same procedure already described1 (Scheme 5). The specific rotation
values, according to those reported in the literature, validated the approach used to assign the absolute
configuration of the stereocentre C-3 through the 1H NMR data.
3. Comments
The experimental data reported in Scheme 3 allowed us to deduce that the stereochemical outcome
of the second alkylation, i.e. at C-3, is largely stereocontrolled by the configuration of the vicinal chiral
auxiliary at N-4 and the C-6 stereocentre does not considerably influence the diastereoselectivity (see
Scheme 4). Thus, it is possible to prepare uncommon α-amino acids α,α0-dialkylated with R or S
configuration by using the chiral auxiliary of the S or R configuration, respectively.
The observed diastereoselection, i.e. the large proportion of the diastereomer with the 3S configuration
(7 or 10), could be explained by considering the differences in the steric hindrance of the lithium enolate
diastereotopic faces (induced by the chiral auxiliary) during the attack of electrophile R–X. For instance,
in the anti-periplanar conformation of the lithium enolate of 9, the phenyl ring gives rise to steric

3414
G. Porzi, S. Sandri / Tetrahedron: Asymmetry 9 (1998) 3411–3420
Fig. 1. Model proposed for the lithium enolate of 9.
hindrance on the re face when the chiral auxiliary is in the R configuration, thereby the electrophilic
attack on the si face appears preferred (see model A in Fig. 1), as results in Schemes 3 and 4. Conversely,
in the syn-periplanar conformation, the phenyl ring causes steric hindrance on the si face favouring the
electrophilic attack on the re face (see model B in Fig. 1). Following the same reasoning, the opposite
diastereoselection can be explained when the chiral auxiliary possesses the S configuration. In fact, as
reported in a previous paper1 and summarized in Scheme 1, the alkylations of the synthon 1 prevalently
gave the 3R diastereomer.
In order to substantiate the above proposed model, semiempirical quantomechanical calculations2
were performed on the lithium enolate of 9 and the results obtained were indeed consistent with the
model hypothesized in Fig. 1. In the preferred geometry of the enolate, the atoms O-1, C-2, C-3, and
N-4 are practically coplanar while C-5 and C-6 are about 20° and 26° out of the plane, respectively. It
is interesting to note that the molecular modelling studies give a geometry similar to that obtained by
Davies et al.4 for a morpholinone enolate. A complete conformational analysis, accomplished by full
rotation around the (N-4)–(C-10) bond, showed that the (R)-phenethyl side chain can exist in two low
energy geometries with the benzylic hydrogen quasi anti- or syn-periplanar (enolates A and B in Fig. 1,
respectively) with respect to the adjacent carbonyl group, the former being 1.5 kcal/mol more stable
than the latter. As already observed for similar tetrasubstituted morpholindiones,1 NOE experiments and
energy optimization performed, for instance, on the 6-allyl-3-benzyl-3,6-dimethyl-4-N-[(R)-phenethyl]-
1,4-morpholin-2,5-dione by the semiempirical quantomechanical AM1 method are consistent with a
preferred anti-periplanar conformation.
In the more populated conformer A of the enolate, the dihedral angle H–(C-10)–(N-4)–(C-5) is 160°,
while in the conformer B it is −14°. Thus, we can assert that the model proposed to explain the observed
diastereoselectivity is in good agreement with the results of the conformational analysis in spite of the
moderate difference in energy between the two more stable conformers.
In conclusion, in this report we have shown that the C-3 stereocentre completely controls the C-6
alkylation and subsequently is partially epimerized itself. In contrast, the C-6 stereocentre has no affect
on the stereocontrol of the alkylation at C-3. While the (N-4)-phenethyl group influences the adjacent
C-3 stereocentre, it has no effect on C-6.
4. Stereochemical assignments
The introduction of the PhCH2 group at the C-6 position in the derivatives 5 and 6 was shown through
NOE measurements. By irradiating the proton of the N-phenethyl group in both the diastereomers, a
significant NOE was registered on (C-3)–H (Fig. 2), but not on (C-6)–CH3 nor on CH2Ph. However,
we confirmed that the two reaction products are diastereomers with opposite configurations of the C-3,
because both 5 and 6, after metallation followed by water addition, give a mixture of 5 and 6.1
The absolute configuration of the diastereomers 5 and 6 was established on the basis of 1H NMR data

G. Porzi, S. Sandri / Tetrahedron: Asymmetry 9 (1998) 3411–3420
3415
Fig. 2.
using the approach often employed by us1,3,5 for similar molecules. The isomer 5 showed a significant
NOE between (C-3)–CH3 and (C-6)–CH3, as previously shown also in (3R,6S)-2,1 which is indicative
of a cis-diaxial arrangement. Thus, the C-3 configuration could be easily assigned by comparing the
chemical shifts of either the (C-3)–CH3 and (C-3)–H of 5 with those of the corresponding isomer
(3R,6S)-2. To this end, it is necessary to take into consideration: (a) the opposite configuration of the
chiral auxiliary phenethyl in 5 and in 2; (b) that the preferred geometry of the morpholindione derivatives
is a quasiboat (or sofa) conformation in which the (C-3)–CH3 preferentially lies in the pseudo-axial
position and the proton in the pseudo-equatorial position.3 As previously demonstrated,5 the shielding
effects can only take place in the rotamer in which the phenethyl hydrogen lies syn periplanar to the
carbonyl group. Therefore, the shielding of (C-3)–H and the deshielding of (C-3)–CH3 registered in 5,
with respect to (3R,6S)-2, are a result of the different shielding effects induced by the chiral phenethyl
group which possesses the R configuration in 5, but the S configuration in 2 (see Fig. 2). These findings
and the NOE between the (C-3)–CH3 and the (C-6)–CH3, reasonably allowed us to assign the 3R
configuration to the diastereomer 5. The 3S configuration of 6 was assigned owing to the strong upfield
shift suffered by the (C-3)–CH3 which can be exclusively induced by the shielding of –CH2Ph in the
cis-diaxial relationship, analogous to that already observed for (3S,6S)-2 (Fig. 2). In addition, because a
substituent in the pseudo-equatorial position can be exclusively shielded by the phenethyl moiety in the
S configuration, the chemical shift difference of the (C-3)–H between 6 and (3S,6S)-2 can be ascribed to
the opposite configuration of the auxiliary in the two isomers (see Fig. 2). Thus, we were able to infer
that in the isomers 5 and 6, diastereomers at C-3 as mentioned above, the C-6 stereocentre possesses the
S configuration also on the basis of the NOE between the methyls at C-3 and C-6 registered only on the
isomer 5.
Having established the C-6 configuration of 5 and 6, the stereochemistry of C-3 in the diastereomers
7 and 8 was ascertained on the basis of the shielding effects induced by the phenyl ring of (C-6)–CH2Ph,
according to that previously observed in similar substrates.1 In fact, in the isomers 8, the singlet of
(C-3)–CH3 (0.76, 0.82 and 0.88 ppm in 8a, 8c and 8d, respectively) is shifted upfield with respect to
distereomers 7 (1.56, 1.55 and 1.58 ppm in 7a, 7c and 7d, respectively) which is representative of a
cis-diaxial relationship between the methyl at C-3 and the (C-6)–CH2Ph (Fig. 3). Thus, we were able
to assign the 3S configuration to the isomers 7 and the 3R configuration to the isomers 8. The 3R
configuration of 8b was established considering that the (C-3)–CH3 and (C-6)–CH3 suffer the same

3416
G. Porzi, S. Sandri / Tetrahedron: Asymmetry 9 (1998) 3411–3420
Fig. 3.
shielding effect (1.03 and 1.07 ppm). On the contrary, in 7b, these methyl groups resonate at different
fields (1.54 and 1.67 ppm) and furthermore, are deshielded with respect to 8b. These findings suggested
a trans relationship between the (C-3)–CH3 and (C-6)–CH3 in 8b (both methyl groups being equally
shielded by the phenyl ring of both PhCH2 groups) and allowed us to assign the 3S configuration to the
diastereomer 7b.
5. Experimental
5.1. General information
1H and 13C NMR spectra were recorded on a Gemini spectrometer at 300 MHz using CDCl3 as
solvent, unless otherwise stated, and the chemical shifts are reported in ppm relative to the solvent.
Optical rotation values were measured on a Perkin–Elmer 343 polarimeter. Dry THF was distilled from
benzophenone ketyl. Chromatographic separations were performed with silica gel 60 (230–400 mesh).
Some products, after chromatographic separation were not isolated pure enough to measure the optical
rotation.
5.2. (3R,6S,10R)-3,6-Dimethyl-4-N-(10-phenethyl)-1,4-morpholin-2,5-dione 4
The title compound was synthesized by following the procedure already reported for 11 starting from
the methanesulfonate of (S)-ethyl lactate, (2S)-2-acetoxy-propanoylchloride and (R)-phenethyl amine.
1H NMR δ 1.57 (d, 3H, J=7.3 Hz), 1.63 (d, 3H, J=7.3 Hz), 1.68 (d, 3H, J=6.7 Hz), 3.95 (q, 1H, J=7.3
Hz), 5.0 (q, 1H, J=6.7 Hz), 5.88 (q, 1H, J=7.3 Hz), 7.32 (m, 5ArH); 13C NMR δ 16.0, 17.2, 18.4, 51.3,
51.7, 73.4, 126.7, 128.0, 128.8, 138.4, 166.0, 168.2. Anal. calcd for C14H17NO3: C, 68; H, 6.93; N, 5.66.
Found: C, 68.2; H, 6.95; N, 5.65.
5.3. (3R,6S,10R)-5 and (3S,6S,10R)-6 6-Benzyl-3,6-dimethyl-4-N-(10 -phenethyl)-1,4-morpholin-2,5-
diones
LHMDS (8 ml, 1M solution in THF) was added dropwise to a stirred solution of 4 (1.24 g, 5 mmol)
in dry THF (50 ml) cooled at −78°C under an inert atmosphere. After 1 h, benzyl bromide (0.85 g, 5
mmol) was added and the reaction mixture was stirred for 6 h. The reaction, slowly warmed to rt, was
quenched with water and extracted with ethyl acetate. The organic layer was dried, evaporated under
reduced presssure and the residue containing the diastereomeric mixture (90% yield) was separated by
silica gel chromatography (hexane:ethyl acetate as eluent).

G. Porzi, S. Sandri / Tetrahedron: Asymmetry 9 (1998) 3411–3420
3417
(3R,6S,10R)-5. 1H NMR δ 1.47 (d, 3H, J=7.2 Hz), 1.58 (d, 3H, J=7.2 Hz), 1.82 (s, 3H), 3.07 (d, 1H,
J=13.5 Hz), 3.46 (q, 1H, J=7.2 Hz), 3.50 (d, 1H, J=13.5 Hz), 5.51 (q, 1H, J=7.2 Hz), 6.55 (m, 2H), 7.25
(m, 8ArH); 13C NMR δ 17.4, 22.2, 26.8, 45.8, 50.8, 52.4, 85.3, 126.8, 127.0, 128.1, 128.3, 128.4, 130.9,
134.8, 137.1, 166.3, 167.0. Anal. calcd for C21H23NO3: C, 74.75; H, 6.87; N, 4.15. Found: C, 74.9; H,
6.9; N, 4.14.
(3S,6S,10R)-6. 1H NMR δ 0.30 (d, 3H, J=7.1 Hz), 1.65 (d, 3H, J=6.6 Hz), 1.69 (s, 3H), 3.01 (d, 1H,
J=13.8 Hz), 3.45 (d, 1H, J=13.8 Hz), 4.17 (q, 3H, J=7.1 Hz), 5.3 (q, 3H, J=6.6 Hz), 7.25 (m, 10ArH);
13C NMR δ 16.2, 19.9, 27.8, 45.9, 53.3, 53.8, 85.8, 127.1, 127.6, 127.9, 128.4, 128.6, 131.1, 134.9,
140.2, 166.6, 167.2.
5.4. Alkylation of diastereomeric mixture 5+6: general procedure
The diastereomeric mixture of 5 and 6 (0.67 g, 2 mmol) in dry THF (20 ml) was treated with LHMDS
(2 ml, 1M solution in THF) then submitted to the alkylation with the appropriate halide (3 mmol). The
reaction was performed by following the same procedure employed for the alkylation of 4. From the crude
reaction product the diastereomers were separated by silica gel chromatography (hexane:ethyl acetate as
eluent).
5.5. (3S,6S,10R)-6-Benzyl-3-ethyl-3,6-dimethyl-4-N-(10 -phenethyl)-1,4-morpholin-2,5-dione 7a
Ethyl iodide was used as the alkylating reagent. The reaction was completed in 8 h and the pure
product was isolated in 75% yield. 1H NMR δ 0.17 (t, 3H, J=7.4 Hz), 1.45 (m, 1H), 1.56 (s, 3H), 1.69
(s, 3H), 1.76 (d, 3H, J=6.9 Hz), 1.80 (m, 1H), 2.86 (d, 1H, J=13.5 Hz), 3.50 (d, 1H, J=13.5 Hz), 4.31 (q,
1H, J=6.9 Hz), 7.27 (m, 10ArH); 13C NMR δ 7.6, 19.4, 26.8, 28.6, 32.3, 44.8, 54.5, 66.5, 84.1, 126.1,
126.7, 127.1, 128.1, 128.2, 131.3, 135, 141.4, 166.1, 168.9. [α]D −81.2 (c 0.9, CHCl3). Anal. calcd for
C23H27NO3: C, 75.59; H, 7.45; N, 3.83. Found: C, 75.75; H, 7.44; N, 3.82.
5.6. (3R,6S,10R)-6-Benzyl-3-ethyl-3,6-dimethyl-4-N-(10 -phenethyl)-1,4-morpholin-2,5-dione 8a
The pure product was isolated in 15% yield. 1H NMR δ 0.67 (t, 3H, J=7.2 Hz), 0.76 (s, 3H), 1.72 (d,
3H, J=7 Hz), 1.73 (s, 3H), 1.79 (m, 2H), 2.92 (d, 1H, J=13.5 Hz), 3.44 (d, 1H, J=13.5 Hz), 4.14 (q, 1H,
J=7 Hz), 7.34 (m, 10ArH); 13C NMR δ 9.1, 20.1, 25.1, 28.2, 32.7, 46.7, 56.1, 66.9, 84.3, 127.1, 127.5,
127.9, 128.1, 128.5, 131.5, 134.5, 140.9, 166.6, 168.9.
5.7. (3S,6S,10R)-3,6-Dibenzyl-3,6-dimethyl-4-N-(10 -phenethyl)-1,4-morpholin-2,5-dione 7b
Benzyl bromide was used as the alkylating reagent. The reaction was completed in 5 h and the pure
1
product was isolated in 70% yield. H NMR δ 1.54 (s, 3H), 1.67 (s, 3H), 1.90 (d, 3H, J=6.9 Hz), 2.33
(d, 1H, J=14.3 Hz), 2.97 (d, 1H, J=13.5 Hz), 3.0 (qAB, 2H, J=13.6 Hz), 4.54 (q, 1H, J=6.9 Hz), 7.2 (m,
15ArH); 13C NMR δ 21.2, 26.9, 27.5, 44.8, 44.9, 55.8, 66.4, 83.8, 125.9, 126.5, 127.0, 128.0, 128.1,
130.0, 130.8, 134.5, 134.7, 141.8, 167.1, 167.5. [α]D −30 (c 2.1, CHCl3). Anal. calcd for C28H29NO3:
C, 78.66; H, 6.84; N, 3.28. Found: C, 78.45; H, 6.87; N, 3.27.

3418
G. Porzi, S. Sandri / Tetrahedron: Asymmetry 9 (1998) 3411–3420
5.8. (3R,6S,10R)-3,6-Dibenzyl-3,6-dimethyl-4-N-(10 -phenethyl)-1,4-morpholin-2,5-dione 8b
The pure product was isolated in 20% yield. 1H NMR δ 1.03 (s, 3H), 1.07 (s, 3H), 1.70 (d, 3H, J=7.1
Hz), 2.82 (d, 1H, J=13.5 Hz), 3.03 (qAB, 2H, J=14 Hz), 3.34 (d, 1H, J=13.5 Hz), 4.36 (q, 1H, J=7.1 Hz),
6.52 (m, 2H), 7.31 (m, 13ArH); 13C NMR δ 21.1, 25.7, 26.8, 44.9, 46.9, 57.9, 67.8, 84.2, 127.2, 127.4,
127.5, 128.0, 128.3, 129.3, 130.5, 131.5, 134.2, 134.6, 141.4, 166.7, 168.1. [α]D −102.5 (c 0.9, CHCl3).
5.9. (3S,6S,10R)-3-Allyl-6-benzyl-3,6-dimethyl-4-N-(10 -phenethyl)-1,4-morpholin-2,5-dione 7c
Allyl iodide was used as the alkylating reagent. The reaction was completed in 4 h and the pure
1
product was isolated in 67% yield. H NMR δ 1.55 (s, 3H), 1.72 (s, 3H), 1.77 (d, 3H, J=6.9 Hz), 2.15
(m, 1H), 2.45 (m, 1H), 2.91 (d, 1H, J=13.6 Hz), 3.52 (d, 1H, J=13.6 Hz), 4.35 (q, 1H, J=6.9 Hz), 4.79
(m, 2H), 4.92 (m, 1H), 7.27 (m, 10ArH); 13C NMR δ 20.0, 26.3, 28.4, 43.3, 45.2, 55.1, 65.7, 84.2, 119.6,
126.1, 126.7, 127.0, 128.2, 131.3, 134.9, 141.5, 166.2, 168.4. [α]D −57.9 (c 0.6, CHCl3). Anal. calcd for
C24H27NO3: C, 76.36; H, 7.21; N, 3.71. Found: C, 76.65; H, 7.19; N, 3.7.
5.10. (3R,6S,10R)-3-Allyl-6-benzyl-3,6-dimethyl-4-N-(10 -phenethyl)-1,4-morpholin-2,5-dione 8c
1
The pure product was isolated in 22% yield. H NMR δ 0.82 (s, 3H), 1.70 (d, 3H, J=7 Hz), 1.72 (s,
3H), 2.47 (m, 2H), 2.95 (d, 1H, J=13.6 Hz), 3.43 (d, 1H, J=13.6 Hz), 4.22 (q, 1H, J=7 Hz), 4.9 (m, 2H),
5.31 (m, 1H), 7.3 (m, 10ArH); 13C NMR δ 20.5, 25.0, 28.3, 43.4, 46.6, 56.6, 66.7, 84.4, 121.2, 127.2,
127.4, 127.9, 128.1, 128.4, 130.9, 131.5, 134.6, 141.0, 166.5, 168.2. [α]D −56 (c 0.3, CHCl3).
5.11. (3S,6S,10R)-6-Benzyl-3,6-dimethyl-4-N-(10 -phenethyl)-3-[(E)-3-phenyl-2-propenyl]-1,4-
morpholin-2,5-dione 7d
Cinnamyl bromide was used as the alkylating reagent. The reaction was completed in 4 h and the
1
product was isolated in 65% yield. H NMR δ 1.58 (s, 3H), 1.71 (s, 3H), 1.83 (d, 3H, J=6.9 Hz), 1.95
(m, 1H), 2.45 (m, 1H), 2.98 (d, 1H, J=13.7 Hz), 3.43 (d, 1H, J=13.7 Hz), 4.46 (q, 1H, J=6.9 Hz), 5.75
(m, 1H), 6.33 (d, 1H, J=15.7 Hz), 7.03–7.52 (m, 15ArH); 13C NMR δ 20.4, 25.8, 27.8, 42.8, 45.5, 55.2,
65.8, 84.0, 122.8, 126.1, 126.4, 126.7, 127.6, 128.1, 128.3, 131.1, 134.5, 134.9, 136.6, 141.4, 166.8,
168.2. Anal. calcd for C30H31NO3: C, 79.44; H, 6.89; N, 3.09. Found: C, 79.65; H, 6.85; N, 3.1.
5.12. (3R,6S,10R)-6-Benzyl-3,6-dimethyl-4-N-(10 -phenethyl)-3-[(E)-3-phenyl-2-propenyl]-1,4-
morpholin-2,5-dione 8d
The pure product was isolated in 24% yield. 1H NMR δ 0.88 (s, 3H), 1.59 (s, 3H), 1.76 (d, 3H, J=6.8
Hz), 2.62 (m, 2H), 2.90 (d, 1H, J=13.4 Hz), 3.42 (d, 1H, J=13.4 Hz), 4.30 (q, 1H, J=6.8 Hz), 5.70 (m,
1H), 5.97 (d, 1H, J=15.5 Hz), 7.25 (m, 15ArH); 13C NMR δ 20.9, 25.0, 27.9, 42.4, 46.5, 56.9, 66.9, 84.4,
121.9, 126.0, 127.2, 127.4, 127.5, 127.9, 128.2, 128.4, 131.4, 134.8, 135.5, 136.3, 141.4, 166.6, 168.4.
5.13. (3R,10R)-3,6,6-Trimethyl-4-N-(10-phenethyl)-1,4-morpholin-2,5-dione 9
A solution of 4 (0.5 g, 2 mmol) in dry THF (20 ml) was treated with LHMDS (2 ml, 1M solution
in THF) then alkylated with methyl iodide (0.42 g, 3 mmol). The reaction was performed by following
the same procedure employed to obtain the products 5 and 6 and the reaction product (90% yield) was

G. Porzi, S. Sandri / Tetrahedron: Asymmetry 9 (1998) 3411–3420
3419
1
purified by silica gel chromatography (hexane:ethyl acetate as eluent). H NMR δ 1.6 (d, 3H, J=7 Hz),
1.65 (d, 3H, J=7.1 Hz), 1.7 (s, 3H), 1.72 (s, 3H), 3.90 (q, 1H, J=7 Hz), 5.75 (q, 1H, J=7.1 Hz), 7.37 (m,
5ArH). Anal. calcd for C15H19NO3: C, 68.94; H, 7.33; N, 5.36. Found: C, 68.7; H, 7.35; N, 5.37.
5.14. (3S,10R)-10 and (3R,10R)-11 3-Allyl-3,6,6-trimethyl-4-N-(10 -phenethyl)-1,4-morpholin-2,5-dione
The allylation of 9 was performed with allyl iodide by following the same procedure described for the
alkylation of the mixture 5+6. The diastereomers, obtained in about 90% yield, were separated by silica
gel chromatography (hexane:ethyl acetate as eluent).
(3S,10R)-10. 1H NMR δ 1.6 (s, 3H), 1.65 (s, 3H), 1.70 (s, 3H), 1.95 (d, 3H, J=7 Hz), 2.8 (m, 2H), 4.55
(q, 1H, J=7 Hz), 5.30 (m, 2H), 5.80 (m, 1H), 7.3 (m, 5ArH). Anal. calcd for C18H23NO3: C, 71.73; H,
7.69; N, 4.65. Found: C, 71.85; H, 7.67; N, 4.66.
(3R,10R)-11. 1H NMR δ 1.65 (s, 6H), 1.8 (s, 3H), 1.9 (d, 3H, J=7 Hz), 2.7 (m, 2H), 4.55 (q, 1H, J=7
Hz), 4.98 (m, 2H), 5.4 (m, 1H), 7.37 (m, 5ArH).
5.15. (2S,4R)-3-Aza-2-benzyl-2-methyl-4-phenyl-pentanoic acid hydrochloride 12
The product was recovered in practically quantitative yield by hydrolysis of 7b with LiOH for 40 h at
reflux and following the same procedure previously described.1 The NMR data are consistent with those
reported.1 [α]D 80 (c 1, 1N HCl). Anal. calcd for C18H21NO2: C, 76.3; H, 7.47; N, 4.94. Found: C, 76.6;
H, 7.5; N, 4.92.
5.16. (2S)-2-Benzylalanine hydrochloride 13
The product, obtained in about 90% yield by hydrogenolysis of 12 using the same conditions
previously employed,1 was then converted into the hydrochloride. [α]D −9.4 (c 0.2, H2O) [lit.6 −9.6
(c 0.2, H2O)].
5.17. (2S,4R)-3-Aza-2-allyl-2-methyl-4-phenyl-pentanoic acid hydrochloride 14
The product was obtained in practically quantitative yield by refluxing 7c or 10 for 60 h with LiOH
and following the same procedure previously used.1 The NMR data are consistent with those reported1
[α]D 19 (c 1, 1N HCl). Anal. calcd for C14H19NO2: C, 72.07; H, 8.21; N, 6. Found: C, 72.3; H, 8.2; N,
6.02.
5.18. (2S)-2-Propylalanine 15
The product was obtained by hydrogenolysis of 14, as previously reported1 and it was recovered pure
in about 90% yield after treatment with the ion exchange resin Amberlist H-15. [α]D 1.5 (c 1, 1N HCl).
Acknowledgements
Investigation supported by the University of Bologna (Funds for selected research topics and Fondi
Ricerca Istituzionale, ex quota 60%).

3420
G. Porzi, S. Sandri / Tetrahedron: Asymmetry 9 (1998) 3411–3420
References
1. Carloni, A.; Porzi, G.; Sandri, S. Tetrahedron: Asymmetry 1998, 9, 2987.
2. Energy calculations have been performed by the AM1 method (HyperChem program, 1994) by using the ‘Polak-Ribiere
(conjugate gradient) algorithm’ with an RMS gradient 0.01 kcal.
3. (a) Porzi, G.; Sandri, S. Tetrahedron: Asymmetry 1996, 7, 189; (b) Favero, V.; Porzi, G.; Sandri, S. Tetrahedron: Asymmetry
1997, 8, 599; (c) Porzi, G.; Sandri, S.; Verrocchio, P. Tetrahedron: Asymmetry 1998, 9, 119.
4. Bull, S. D.; Davies, S. G.; Fox, D. J.; Sellers, T. G. R. Tetrahedron: Asymmetry 1998, 9, 1483.
5. Orena, M.; Porzi, G.; Sandri, S. J. Org. Chem. 1992, 57, 6532.
6. Williams, R.; Im, M.-N. J. Am. Chem. Soc. 1991, 113, 9276.