A photochemical approach to phenylalanines and related compounds by alkylation of glycine

Article abstract of DOI:10.1016/S0040-4020(01)00783-9

Phenylalanines can be prepared on UV photolysis of protected glycines in the presence of di-tert-butyl peroxide, substituted toluenes and the photosensitiser benzophenone. These reactions, which lead to highly selective mono-alkylation at the α-position of glycines, involve coupling of captodative α-glycine radicals with benzyl radicals. This method can be used to selectively alkylate a variety of glycine derivatives using a range of substituted toluenes under neutral reaction conditions.

Full text of DOI:10.1016/S0040-4020(01)00783-9

TETRAHEDRON
Pergamon
Tetrahedron 57 12001) 8115±8124
A photochemical approach to phenylalanines and related
compounds by alkylation of glycine
Haydn S. Knowles,a Keith Huntb and Andrew F. Parsonsa,p
aDepartment of Chemistry, University of York, Heslington, York YO10 5DD, UK
bA.H. Marks and Company Ltd, Wyke, Bradford, West Yorkshire BD12 9EJ, UK
Received 20 June 2001; accepted 30 July 2001
AbstractÐPhenylalanines can be prepared on UV photolysis of protected glycines in the presence of di-tert-butyl peroxide, substituted
toluenes and the photosensitiser benzophenone. These reactions, which lead to highly selective mono-alkylation at the a-position of glycines,
involve coupling of captodative a-glycine radicals with benzyl radicals. This method can be used to selectively alkylate a variety of glycine
derivatives using a range of substituted toluenes under neutral reaction conditions. q 2001 Elsevier Science Ltd. All rights reserved.
1. Introduction
because radical reactions not only proceed under mild
1neutral) conditions but radicals are compatible with a
wide range of common functional groups including amides
and esters. Thus, radical carbon±carbon bond forming
reactions offer a potential solution to the problems of
racemisation and polyalkylation, which can plague ionic
amino acid alkylations.
a-Amino acids are an important class of organic
compounds, which have been shown to exhibit an extra-
ordinary range of biological and physiological properties
as individual molecules or as constituent members of
proteins or peptides.1 One particularly interesting group of
a-amino acids are phenylalanines.2 A variety of phenyl-
alanine derivatives are of medicinal importance including
dopa, melphalan and levothyroxin, which are used in the
treatment of Parkinson's disease, cancer and myxoedema,
respectively. The nonpolar nature and steric bulkiness of the
aromatic side chain also results in phenylalanine derivatives
being one of the key pharmacophores in peptidomimetics.
For example, most peptidomimetics designed to inhibit
aspartyl proteases 1such as HIV protease, renin, and
cathepsins D and E) contain phenylalanine mimics or
other bulky hydrophobic groups.2c
The alkylation of glycine using an intermolecular radical
coupling reaction was ®rst reported by Elad and co-workers
1Scheme 1).7 This work showed that UV irradiation of ethyl
N-acetylglycinate 1 in the presence of acetone and excess
toluene produces phenylalanine 4, although the yield of 4
was extremely low 13±5%). The proposed mechanism for
this reaction involved coupling of captodative8 radical 2
with a benzyl radical 3, these radicals being derived from
abstraction of a hydrogen atom from glycine 1 and toluene,
respectively, by an excited acetone molecule. This method
was extended to selectively alkylate glycine residues within
peptides and proteins,9 and this could be accomplished
using visible light, when an a-diketone and di-tert-butyl
peroxide were used in place of acetone.10 For example,
irradiation of a mixture of biacetyl, di-tert-butyl peroxide,
toluene and the dipeptide Tfa-Gly-Val-OMe with visible
light, resulted in the selective formation of Tfa-Phe-Val-
OMe. The fact that glycine residues can be selectively
alkylated has been explained on the basis of radical
stability: the intermediate captodative radical derived from
glycine is more stable than related a-amino acid radicals
because this can adopt a favourable planar geometry,
which leads to maximum orbital overlapand radical
stabilisation.11
As these types of compounds have found widespread use in
pharmaceuticals, the synthesis of phenylalanines and other
a-amino acids has attracted the attention of organic
chemists for many years.3 This has led to the development
of several `classical' syntheses including the Strecker, Hell±
Volhardt±Zelinsky and Gabriel reactions. These ionic trans-
formations have long been used to prepare racemic amino
acids and asymmetric variations of these and many related
ionic processes are currently being developed.4 In com-
parison, the preparation of a-amino acids using radical
carbon±carbon bond forming reactions has received
considerably much less attention.5,6 This is surprising
Keywords: alkylation; amino acids and derivatives; photochemistry;
radicals and radical reactions.
Corresponding author. Tel.: 144-1904-432608; fax: 144-1904-432516;
e-mail: afp2@york.ac.uk
More recently, these radical coupling reactions have been
carried out in the absence of an alkylating agent, to form
amino acid dimers.12 For example, Obata and Niimura12a
p
0040±4020/01/$ - see front matter q 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020101)00783-9

8116
H. S. Knowles et al. / Tetrahedron 57 .2001) 8115±8124
Scheme 1.
Scheme 2.
have isolated dehydrodimerisation products on UV
irradiation of methyl pyroglutamate or methyl N-acetyl-
glycinate 5 in the presence of di-tert-butyl peroxide
1Scheme 2). In these reactions the tert-butoxyl radical
1tBuOz) can abstract an a-hydrogen atom from the amino
acids and, for example, for glycinate 5 this produces similar
amounts of the diastereoisomeric dimers 1^)-6 and meso-7.
Although the use of di-tert-butyl peroxide, in place of
acetone, leads to more ef®cient a-hydrogen-atom abstrac-
tion, related work by Easton and co-workers13 have reported
a competitive reaction 1to dimerisation) leading to the
formation of an alanine. This was attributed to coupling of
the intermediate captodative radical with a methyl radical,
which can be derived from b-scission of the tert-butoxyl
radical. With a view to developing an ef®cient photo-
chemical approach to phenylalanines, this paper describes
the novel alkylation of various glycine derivatives using
di-tert-butyl peroxide in the presence of substituted toluenes
and related compounds.14
toluene was used, phenylalanine 11a was formed in 27 or
59% yield based on recovered glycine 8a 1entry 1).
Increasing the number of equivalents of peroxide and/or
toluene did not improve the yield and/or selectivity for
11a 1entries 2±4). This reagent ratio was therefore
employed in subsequent reactions using alternative glycine
derivatives. Hence, although reaction using the correspond-
ing N-benzenesulfonyl protected glycine did not produce
any alkylated products,15 reaction of N-Boc and N-Cbz
glycines, 8b and 8c, formed phenylalanines 11b and 11c,
respectively, in similar yields to 11a 1entries 5 and 6). The
selective alkylation of 8c is of particular note because
competitive hydrogen-atom abstraction at the benzylic
position of the Cbz protecting group is also possible
although no products derived from this reaction were
isolated. Surprisingly, there was also no evidence of
benzylic radical formation from photolysis of related, but
simpler amines, including PhNHCbz and tBuNHCbz 1under
similar conditions).16
With a view to decreasing the reaction time, the alkylation
of glycine 8a was also carried out using a more powerful
400 W UV lamp1entry 7) and this dramatically increased
the rate of the reaction. After 6 h, 91% of the starting
material was consumed as compared to 70% after 64.5 h
when using the 125 W lamp. The use of the more powerful
lampalso gave rise to an additional product, namely a-tert-
butoxy-glycine 12, which was isolated in 13% yield. The
formation of 12 presumably re¯ects a higher concentration
2. Results and discussion
Initial investigations involved UV irradiation of methyl
N-benzoylglycinate 8a in the presence of toluene and
di-tert-butyl peroxide 1Table 1). The mixture was irradiated
using a 125 W lampin degassed benzene and a variety of
reactions were carried out using different concentrations of
the reagents 1entries 1±4). In all cases, 1,2-diphenylethane
1derived from dimerisation of benzyl radical 3) and small
amounts of butanedioate 9a 1as an approximately 1:1
mixture of racemic and meso isomers) and alanine 10a
were isolated, but the major product was the desired phenyl-
alanine 11a. When a 1:2:5 ratio of glycine 8a±peroxide±
t
of the BuOz, which could couple with the captodative
radical, when using the more powerful lamp. It should
also be noted that a-tert-butoxy-glycine 12 could be formed
from reaction of the captodative radical with di-tert-butyl
peroxide in an SH2 reaction.

H. S. Knowles et al. / Tetrahedron 57 .2001) 8115±8124
Table 1. Photoalkylation of glycines 8a±c leading to phenylalanines 11a±c
8117
Entry
8
Reaction
conditions
1W, h)
Conc.
of 8
Equiv. of
peroxide
Equiv. of
toluene
9a±c
1%)a
10a±c
1%)a
11a±c
1%)a
12
1%)a
13
1%)a
Totalb
yield
1%)
1mol dm23
)
1
2
3
4
5
6
7
8
a
a
a
a
b
c
a
a
a
a
a
a
a
125, 64.5
125, 48.0
125, 72.0
125, 50.5
125, 70
125, 64.5
400, 6
400, 12
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.13
0.13
0.26
0.13
0.13
0.13
2
4
4
2
2
2
2
2
5
5
5
5
5
5
5
10
10
5
5
5
5
5
5
5
7.5
10
113)
316)
316)
318)
9118)
±
112)
11121)
416)
418)
6115)
7112)
216)
27159)
25159)
26149)
18155)
27151)
21164)
19138)
29154)
31152)
19134)
30133)
26134)
37157)
±
±
±
±
±
±
±
±
±
±
±
±
±
±
70
80
67
69
69
76
91
89
77
77
82
61
84
±
4112)
14127)
6111)
315)
±
13124)
113)
9114)
417)
313)
±
9
400, 12
10
11
12
13
400, 12
7113)
18120)
415)
315)
±
400, 12c
400, 12c
400, 12c
21122)
13117)
11117)
314)
415)
416)
214)
±
a
Yields in brackets are based on recovered glycine 8a±c.
Total yield based on recovered glycine 8a±c.
In the presence of 5 equiv. of benzophenone.
b
c
In an attempt to minimise the formation of 12, the concen-
tration of glycine 8a was increased from 0.03 to
0.13 mol dm23. This had the desired effect and a-tert-
butoxy-glycine 12 was isolated in only 1% yield while the
yield of phenylalanine 11a increased from 19 to 29% 1entry
8). However, increasing the concentration of 8a also
resulted in more captodative radical dimerisation leading
to a greater yield of butanedioate 9a.17 When the number
of equivalents of peroxide was increased from 2 to 5 this
was found to lower the yield of 9a 1to 3%) and phenyl-
alanine 11a was isolated in an excellent 31% yield 1entry
9). Attempts to increase the yield of 11a even further by
increasing the concentration of 8a 1to 0.26 mol dm23) were
however, unsuccessful 1entry 10).
and benzophenone or 4,40-dimethoxybenzophenone, which
has been reported to give the corresponding tertiary
alcohols in only 7.7 and 6.4% yield, respectively.19 It should
also be noted that related intramolecular 1photochemical)
reactions of quinones with glycine residues have been
reported.20
Having optimised the formation of phenylalanine 11a, our
attention turned to the photolysis of toluene with alternative
glycine derivatives, namely chiral glycinates 14 and 16,
under similar reaction conditions. Reaction of 14 resulted
in the selective formation of phenylalaninate 15 in 28%
yield 1or 62% yield based on recovered 14) as a 1:1.1
mixture of inseparable diastereoisomers. Although a small
amount of the dimer was isolated 13%) no product derived
from alkylation at the amide side chain was isolated. The
menthyl ester 16 underwent a similar reaction to give
phenylalaninate 17 in 34% yield 1or 42% yield based on
recovered 16) as a 1:1.2 mixture of inseparable diastereo-
isomers. This demonstrates that a-alkylation can proceed
selectively for glycines with alternative secondary and
even tertiary C±H bonds. The piperazine-2,5-dione 18
could also be selectively alkylated in the presence of
10 equiv. of toluene, to give the mono-benzylated
compound 19 in 29% yield. Only a small amount of dialkyl-
ation 14%) to give 20 1as a 1:1 mixture of diastereoisomers)
was observed together with the a-tert-butoxy derivative 21
in 4% yield. Increasing the number of equivalents of toluene
to 20 also resulted in the predominant formation of 19 1in
20% yield); the yield of 20 only increased to 13%.
Further experiments investigated the addition of photo-
sensitisers and when benzophenone was used in combina-
tion with 10 equiv. of toluene an optimum yield of 37% 1or
57% based on recovered 8a) was isolated for 11a 1entries
11±13). In this case, the triplet state of benzophenone,
t
rather than BuOz, could abstract an a-hydrogen atom from
8a, although it was found that addition of both benzo-
phenone and di-tert-butyl peroxide are required for good
yields of 11a.18 The use of benzophenone also produced
minor amounts of the serine derivative 13, derived from
coupling of the captodative radical with Ph21HO)Cz.
Indeed, when the same photolysis was carried out in the
absence of toluene, tertiary alcohol 13 was isolated in
23% yield 1or 50% based on recovered 8a). This compares
favourably with photolysis of methyl N-acetylglycinate 5

8118
H. S. Knowles et al. / Tetrahedron 57 .2001) 8115±8124
were introduced at this site the yields of alkylation
decreased and, for example, reaction of glycinate 8a with
diphenylmethane or ¯uorene, produced 32 or 33 in only 10
and 8% yield, respectively. This could be a consequence of
t
the BuOz radical preferentially reacting with the weaker
benzylic C±H bonds in diphenylmethane and ¯uorene
1compared to toluene)23 resulting in a lower concentration
of the captodative radical derived from glycinate 8a.
The effect of introducing N-alkyl groups on the glycine
derivatives was also probed by reaction of the tertiary
amides 22 and 23. Photolysis of these derivatives under
the optimised conditions for the preparation of phenylalanine
11a produced at best, trace amounts of the phenylalanine
products. This surprising result may be explained by the
stability of the intermediate radicals of type 24; the introduc-
tion of N-alkyl groups may lead to `non-bonding' inter-
actions, which could prevent ef®cient orbital overlap.21
It is interesting to note that in all reactions, no products
derived from dialkylation of glycinate 8a 1or related
compounds) were observed. Indeed, attempts to deliberately
alkylate alanine 10a or phenylalanine 11a, by photolysis
with toluene/peroxide resulted in typically #1% yield of
dialkylated product 34. Similarly, the presence of a second
ester served to hinder the alkylation of malonate 35 and
photolysis in the presence of di-tert-butyl peroxide
12 equiv.) and toluene 15 equiv.) produced 36 in 13% yield
1or 58% yield based on recovered 29). This reaction also
produced a small amount 13%) of phenylalanine 11a,
presumably derived from photo-decarboxylation of 36.
The fact that highly selective monoalkylation occurs in
these reactions is presumably due to the exceptional
stability of the intermediate captodative radical derived
from glycinate 8a.11
Following the successful alkylation of numerous glycine
derivatives using toluene, our attention turned to the reac-
tion of substituted toluenes. Hence, a variety of toluenes
were reacted with 8a under the optimal conditions used
for the formation of phenylalanine 11a. This allowed the
formation of phenylalanines 25±31 in 12±36% isolated
yield or 35±87% yield based on recovered 8a. A variety
of functional groups can therefore be incorporated onto
the benzene ring without signi®cantly affecting the photo-
alkylation process. Thus, for example, although the intro-
duction of 4-bromo or 4-iodo groups on the toluene resulted
in preferential carbon±halogen bond cleavage, both chloro
and ¯uoro substituents can be successfully introduced. The
formation of ¯uoro-substituted phenylalanines is of particu-
lar interest because these derivatives have been used as
probes to elucidate biological pathways.22 All of these reac-
tions were carried out using a 400 W lampexcept for the
formation of 31, which employed a 1000 W lamp.
This work has shown that a variety of glycine derivatives
can be alkylated to form phenylalanines on UV photolysis in
the presence of substituted toluenes, di-tert-butyl peroxide
and benzophenone 1as a photo-sensitiser). Whereas related
anionic alkylations are often complicated by poly-alkyl-
ation, this mild free-radical method allows the selective
formation of mono-alkylated products in reasonable to
good yields. The ef®ciency of these alkylation reactions
has been shown to be affected by the nature of the glycine
protecting group and particularly, by substitution of the
toluene alkylating agent.
Although a number of substituents could be introduced on
the benzene ring, the reaction proved to be very sensitive to
variation of the 1toluene) methyl group. When substituents
3. Experimental
1H and 13C NMR spectra were recorded using a Joel EX 270

H. S. Knowles et al. / Tetrahedron 57 .2001) 8115±8124
8119
Spectrometer; the carbon spectra were recorded at
3.3. 125 W Photolysis of methyl N-tert-butyloxy-
carbonylglycinate 8b
67.5 MHz and were assigned with the aid of DEPT experi-
ments. Coupling constants 1J) were recorded to the nearest
0.5 Hz. IR spectra were recorded on an ATI Mattson
Genesis Series FT IR spectrometer. Mass spectra were
recorded on a Fisons Instruments VG Analytical Autospec
Spectrometer system for both low and high-resolution EI
and CI spectra. Thin layer chromatography was performed
on Merck 5554 aluminium-backed silica gel plates and
compounds were visualised by a UV lamp or by staining
with alkaline potassium permanganate solution or iodine.
Column chromatography was carried out under gravity
using silica gel 1Matrex Silica 60 70±200 microns Fisons
Methyl N-tert-butyloxycarbonylglycinate 8b 11.00 g,
5.29 mmol) was dissolved in benzene 1135 ml) and di-tert-
butyl peroxide 11.55 g, 1.9 ml, 10.6 mmol) together with
toluene 12.43 g, 2.8 ml, 26.5 mmol) were added and the
mixture photolysed for 70 h. The crude orange oil 11.78 g)
was puri®ed by column chromatography 1silica; petroleum
ether±ethyl acetate, 2:1) to afford methyl N-tert-butyloxy-
carbonylphenylalaninate 8b [0.40 g, 27% 151% based on
recovered 8b)],28 1,2-diphenylethane 1.0.27 g), recovered
starting material 8b 10.47 g, 47%) and meso- and 1^)-
dimethyl 2,3-bis-tert-butyloxycarbonylaminosuccinate 9b
[0.19 g, 9%, 118% based)]29 as a 1:1 mixture of diastereo-
isomers.
Ê
or ICN Biomedicals GmbH ¯ash silica 32±63 A) and the
speci®ed eluent. Melting points were recorded on Ko¯er
hot-stage melting point apparatus. Petroleum ether refers
to the fraction of boiling range 40±608C. Photolyses were
carried out using the following lamps: a 125 W medium
pressure mercury arc UV lamp; a 400 W medium pressure
mercury arc UV lamp; an ILC 3102 300 W Xenon UV±
Visible lamp; or a 1000 W ESR spectrometer high pressure
mercury arc UV lamp. Quartz immersion-well reaction
vessels were used in all photolyses except those initiated
by the 1000 W lamp.
3.4. 125 W Photolysis of methyl N-benzyloxycarbonyl-
glycinate 8c
A mixture of Cbz-protected glycine 8c 11.00 g, 4.48 mmol),
di-tert-butyl peroxide 11.31 g, 1.64 ml, 8.96 mmol) and
toluene 12.06 g, 2.4 ml, 22.4 mmol) in benzene 1156 ml)
was irradiated for 64.5 h. This gave an orange oil 11.46 g)
which was puri®ed by column chromatography 1silica;
petroleum ether±diethyl ether, 1:3) to produce methyl
N-benzyloxycarbonylphenylalaninate 11c30 [0.30 g, 21%
164% based on recovered 8c)] and methyl N-benzyloxy-
carbonylalaninate 10c31 [0.042 g, 4% 112%)]. Recovered
starting material 8c 10.67 g, 67%) and 1,2-diphenylethane
1#0.82 g) were also eluted.
3.1. General procedure for photolysis of glycinate 8a
using a 125 W lamp
Methyl N-benzoylglycinate 8a24 10.83±1.00 g, 4.27±
5.18 mmol) was dissolved in benzene 1106±135 ml) and
di-tert-butyl peroxide 12.0±4.0 equiv.) and toluene 15.0±
10.0 equiv.) were added. The total volume of the solution
was such that the concentration was always 0.03 mol dm23
and photolysis using the 125 W lamp was continued for
48±72 h. The solutions were degassed 1by bubbling nitro-
gen gas through the solutions) prior to photolysis and
agitated by the same means through the reaction. The
water ¯ow was regulated to maintain the temperature of
the reaction mixture between 25±408C. At the end of the
irradiation, the lampwas cooled and evaporation of the
mixture gave an orange oil 11.32±2.00 g), which was
puri®ed by column chromatography 1silica; petroleum
ether±ethyl acetate, 1:1). The products were eluted in the
following order: 1,2-diphenylethane,25 methyl N-benzoyl-
phenylalaninate 11a,26 methyl N-benzoylalaninate 10a,27
recovered starting material 8a and a 1:1 mixture of 1^)-
and meso-dimethyl 2,3-dibenzamidobutanedioate 9a.13
3.5. Representative example for photolysis of glycinate
8a using a 400 W lamp in the absence of benzophenone
Methyl N-benzoylglycinate 8a 15.00 g, 25.9 mmol), di-tert-
butyl peroxide 15.0 equiv., 18.91 g, 23.8 ml, 0.13 mol) and
toluene 111.91 g, 13.8 ml, 0.13 mol) in benzene 1162 ml)
were photolysed 112 h) using a 400 W UV lamp. After
photolysis, the solvent was removed and a small sample
11.23 g, 12.5% of the crude mass) was puri®ed to identify
and quantify the products formed. Column chromatography
1silica, petroleum ether±ethyl acetate, 1:1) afforded phenyl-
alaninate 11a [0.286 g, 2.28 g scaled, 31%, 152% based)],
the tert-butoxy adduct 12 [0.074 g, 0.59 g, scaled, 9%,
114%)], alaninate 10a [0.019 g, 0.152 g scaled, 3%, 15%)],
recovered glycine 8a 10.248 g, 1.98 g scaled, 40%) and
dimers 9a [0.023 g, 0.184 g scaled, 4%, 16%)] as a 1:1
mixture of diastereoisomers.
3.2. Representative example: 125 W photolysis using a
1:4:5 ratio of glycinate 8a±peroxide±toluene
Photolysis 148 h) of glycinate 8a 11.00 g, 5.18 mmol), di-
tert-butyl peroxide 14.0 equiv., 3.03 g, 3.8 ml, 20.7 mmol)
and toluene 15.0 equiv., 2.38 g, 2.8 ml, 25.9 mmol) as a
solution in benzene 1133 ml) afforded an orange oil
11.81 g) after evaporation in vacuo. Column chromato-
graphy 1silica; petroleum ether±ethyl acetate, 1:1) afforded
1,2-diphenylethane 10.44 g), methyl N-benzoylphenyl-
alaninate 11a [0.36 g, 25% 159% based on recovered 8a)],
alaninate 10a [0.068 g, 6% 115%)], recovered glycinate 8a
10.58 g, 58%) and a 1:1 mixture of 1^): meso-butanedioate
9a [0.053 g, 3% 16%)].
3.5.1. Methyl N-benzoyl-2-tert-butoxyglycinate 12. RF
0.6 1petroleum ether±ethyl acetate, 3:2); nmax 1CH2Cl2)
3544 1s), 3439 1s), 3330 1s), 1656 1s), 1526 1w), 1481
1w), 1192 1w) cm21; dH 1270 MHz, CDCl3) 7.87±7.80
and 7.60±7.41 15H, m, aromatics), 7.14 11H, d, Jˆ9.5
Hz, NH), 6.03 11H, d, Jˆ9.5 Hz, NHCHCO), 3.78 13H,
s, CO2CH3), 1.31 19H, s, C1CH3)3); m/z 1CI, NH3) 283
1M1NH41, 80%), 266 1M1H1, 100), 227 120), 210
150), 194 110), 139 130), 122 160), 105 130). Found:
M1H1, 266.1390. C14H19NO4 requires for M1H1,
266.1392.

8120
H. S. Knowles et al. / Tetrahedron 57 .2001) 8115±8124
3.6. Representative example for photolysis of glycinate
8a using a 400 W lamp in the presence of benzophenone
13H, d,Jˆ7.0 Hz, CH3CH); dC 167.5 MHz, CDCl3) 171.5
1CO2CH3), 169.7 1CON), 151.2, 127.5, 124.7 1Cv CH),
130.3, 127.9, 116.1 1Cv CH), 76.5 1OCHCH3), 52.5
1CO2CH3), 40.9 1NHCH2CO), 18.4 1CH3CH); m/z 1CI,
NH3) 323 135,35M1NH41, 100%), 308 135,37M1H1, 60),
306 135,35M1H1, 90) 270 130), 161 120), 144 120), 116
110). Found: 35,35M1H1, 306.0297. C12H1335,35Cl2NO4
requires for M1H1, 306.0300.
The mixture of glycinate 8a 15.00 g, 26.0 mmol), di-tert-
butyl peroxide 15.0 equiv., 18.98 g, 23.8 ml, 0.13 mol),
benzophenone 15.0 equiv., 23.57 g, 0.13 mol) and toluene
110.0 equiv., 23.83 g, 27.7 ml, 0.26 mol) in benzene
1149 ml) was photolysed for 12 h using a 400 W UV
lamp. After evaporation an orange oil 140.30 g) was isolated
and a fraction of this [1.70 g 14.2% of crude mass)] was
puri®ed by column chromatography 1silica; petroleum
ether±ethyl acetate, 1:1). This afforded the b,b-diphenyl-
serinate 13 [0.015 g, scaled to 0.36 g, 4%, 16%)], the
phenylalaninate 11a [0.115 g, scaled to 2.74 g, 37%,
157%)], alaninate 10a [0.025 g, scaled to 0.60 g, 11%,
117%)], recovered 8a 10.074 g, scaled to 1.76 g, 35%) and
a 1:1 ratio of meso-:1^)-9a [0.005 g, scaled to 0.119 g, 2%,
14%)].
3.7. Photolysis of methyl 2R-[N-22,4-dichlorophenoxy)-
propanoyl]glycinate 14
Methyl 2R-[N-12,4-dichlorophenoxy)propanoyl]glycinate
14 17.96 g, 25.9 mmol), di-tert-butyl peroxide 15.0 equiv.,
18.98 g, 0.13 mol), benzophenone 15.0 equiv., 23.57 g,
0.13 mol) and toluene 110 equiv., 23.92 g, 0.26 mol) in
benzene 1149 ml) were photolysed for 12 h. After evapora-
tion in vacuo, the crude mixture of products was isolated as
a yellow oil 140.07 g). A small sample 11.38 g, 3.4% of
crude mass) was puri®ed by column chromatography
1silica; petroleum ether±ethyl acetate, 2:1) to afford
3.6.1. Methyl N-benzoyl-b,b-diphenylserinate 13. RF 0.5
1petroleum ether±ethyl acetate, 1:1); White solid; mp 151±
1548C; 1found: C, 73.49; H, 5.60; N, 3.38. C23H21NO4
requires C, 73.58; H, 5.64; N, 3.73%); nmax 1CH2Cl2)
3443 1s), 3358 1s), 3321 1m), 1738 1m), 1650 1s), 1525
1w), 1487 1w), 1445 1w) cm21; dH 1270 MHz, CDCl3)
7.83±7.14 115H, m, aromatics), 7.04 11H, br d, Jˆ8.0 Hz,
NH), 5.85 11H, d, Jˆ8.0 Hz, NHCHCO), 3.53 13H, s,
CO2CH3), 1.64 11H, br s, OH); dC 167.5 MHz, CDCl3)
172.9 1CO2CH3), 167.2 1NCO), 143.5, 142.0, 133.4
1CvCH), 131.8, 130.0, 128.6, 128.4, 128.2, 127.6, 127.4,
127.0, 125.3, 125.1 1CvCH), 79.3 1Ph2COH), 57.4
1NHCHCO), 52.5 1CO2CH3); m/z 1CI, NH3) 376 1M1H1,
80%), 358 1MH12H2O, 100), 298 110), 183 120), 105 120).
Found: M1H1, 376.1544. C23H21NO4 requires for M1H1,
376.1549.
methyl
2R-[N-12,4-dichlorophenoxy)propanoyl]-1R,S)-
phenylalaninate 15 [0.14 g, scaled to 2.91 g, 28%,
162% based)] as an inseparable 1:1.1 mixture of dia-
stereoisomers 1from the 1H NMR sp ectrum) and
dimethyl
2,3-bis-[2R-12,4-dichlorophenoxy)propanoyl-
amino]succinate [0.008 g, scaled to 0.233 g, 3%, 16%)]
as a mixture with recovered glycinate 14 10.149 g,
scaled to 4.33 g, 54%).
3.7.1. Methyl 2R-[N-22,4-dichlorophenoxy)propanoyl]-
R,S-phenylalaninate 15. RF 0.3 1petroleum ether±ethyl
acetate, 2:1); nmax 1thin ®lm) 3412 1s), 2959 1m), 1746 1s),
1675 1s), 1521 1s), 1477 1s), 1445 1m), 1283 1s), 1259 1s),
1104 1s) cm21. Diastereoisomer 1: This was indicated by:
dH 1270 MHz, CDCl3) 7.38±7.08 18H, m, aromatics), 6.94
11H, d, Jˆ5.5 Hz, NH), 4.88 11H, m, NHCHCO), 4.62 11H,
apparent quintet, Jˆ7.0 Hz, OCH1CH3)CO), 3.75 13H, s,
CO2CH3), 3.22 11H, dd,Jˆ14.0 and 5.5 Hz, CHAHBPh),
1.60 13H, d, Jˆ6.5 Hz, CH3CH). Diastereoisomer 2: This
was indicated by: dH 1270 MHz, CDCl3)) 7.38±7.08 18H, m,
aromatics), 6.68 11H, d, Jˆ9.0 Hz, NH), 4.88 11H, m,
NHCHCO), 4.62 11H, apparent quintet, Jˆ7.0 Hz,
OCH1CH3)CO), 3.70 13H, s, CO2CH3), 3.02 11H,
dd,Jˆ14.0 and 7.5 Hz, CHAHBPh), 1.50 13H, d, Jˆ6.5 Hz,
CH3CH); dC 167.5 MHz, CDCl3) 171.4, 171.3, 170.9, 170.6
1CO2CH3 and CON of each diastereomer), 76.3, 76.2
1OCH1CH3)CO), 52.7, 52.6, 52.4, 52.3 1NHCHCO and
CO2CH3), 37.9, 37.7 1CH2Ph), 18.4, 18.2 1CH3CH); m/z
1CI, NH3) 398 137,35M1H1,60%), 396 135M1H1, 100),
362 1MH12Cl, 20), 234 170), 206 110), 120 110). Found:
35M1H1, 396.0767. C19H2035Cl2NO4 requires for M1H1,
396.0769.
3.6.2. Synthesis of methyl 2R-[N-22,4-dichlorophenoxy)-
propanoyl]glycinate 14. A solution of R-2-12,4-dichloro-
phenoxy)propionic acid 110.00 g, 42.0 mmol), thionyl
chloride 11.5 equiv., 7.50 g, 4.6 ml, 63.0 mmol) together
with 2±3 drops of N,N0-dimethylformamide 1Care: noxious
fumes!) was heated for 2 h. The excess thionyl chloride was
removed by rotary evaporation to afford the crude acid
chloride as an orange oil. This was added dropwise to a
solution of glycine methyl ester hydrochloride
11.05 equiv., 5.53 g, 44.0 mmol) and triethylamine
11.05 equiv., 4.44 g, 6.12 ml, 44.0 mmol) in dichloro-
methane 1100 ml) at 08C. The mixture was allowed to stir
for 2 h during which time the solution was allowed to warm
to room temperature. The solution was washed with water
1100 ml), separated, dried 1MgSO4) and evaporated to afford
crude product 113.06 g). Puri®cation by column chromato-
graphy 1silica; petroleum ether±ethyl acetate, 1:1) gave
methyl 2R-[N-12,4-dichlorophenoxy)propanoyl]glycinate
15 17.49 g, 58%) as an off-white solid; mp97±100 8C.
1found: C, 47.14; H, 4.18; N, 4.34. C12H13Cl2NO4 requires
C, 47.08; H, 4.28; N, 4.58%); RF 0.3 1petroleum ether±ethyl
acetate, 1:1); nmax 1CHCl3) 3430 1m), 3010 1m), 1752 1s),
1680 1s), 1526 1s), 1477 1s), 1439 1m), 1284 1m), 1243 1s),
1105 1m) cm21; dH 1270 MHz, CDCl3) 7.42±7.41 11H, m,
aromatic), 7.22±7.18 11H, m, aromatic), 6.90±6.86 12H, m,
aromatic and NH), 4.74 11H, q, Jˆ7.0 Hz, OCHCH3), 4.10
12H, d, Jˆ5.5 Hz, NHCH2CO), 3.77 13H, s, CO2CH3), 1.64
3.7.2. Dimethyl 2,3-bis-[2R-22,4-dichlorophenoxy)pro-
panoylamino]succinate. The presence of this was
indicated by: dH 1270 MHz, CDCl3) 5.52±5.16 11H, m,
NHCHACO), 5.13±5.09 11H, m, NHCHBCO); m/z 1CI,
NH3) 628 135M1NH41, 10%), 611 135M1H1, 100), 575
110), 447 140), 378 120), 285 120), 214 130). Found:
35M1H1, 609.0363. C24H2435Cl4N2O8 requires for M1H1,
609.0365.

H. S. Knowles et al. / Tetrahedron 57 .2001) 8115±8124
8121
3.8. Photolysis of 21S,2R,5S)-menthyl N-benzoylglycinate
16
25±408C. Upon completion the lamp was extinguished and
allowed to cool for 0.5 h. The lampwas then removed and
the resultant orange solution transferred to a round-
bottomed ¯ask and the solvent removed in vacuo to leave
an orange sticky liquid 120.39±71.69 g) from which a small
sample was puri®ed by column chromatography on silica.
11S,2R,5S)-Menthyl N-benzoylglycinate 1632 18.24 g, 25.9
mmol), di-tert-butyl peroxide 15.0 equiv., 18.98 g, 23.8 ml,
0.13 mol), benzophenone 15.0 equiv., 23.57 g, 0.13 mol)
and toluene 110.0 equiv., 23.92 g, 27.7 ml, 0.26 mol) in
benzene 1149 ml) were photolysed using a 400 W lamp.
After 12 h irradiation, the solution was evaporated in
vacuo to afford crude product as an orange oil 134.87 g).
A small sample 10.94 g, 2.7% of crude mass) was puri®ed
by column chromatography 1silica; petroleum ether±ethyl
acetate, 15:1) to afford 11S,2R,5S)-menthyl N-benzoyl-
phenylalaninate 1733 10.097 g, scaled to 0.36 g, 34%,
[42% based on recovered 16]) as an inseparable 1:1.2
3.10.1. Methyl N-benzoyl-2-¯uorophenylalaninate 25.
White solid; mp119±120 8C 1lit.,22b 111±1128C); RF 0.5
1petroleum ether±ethyl acetate, 1:1); nmax 1CHCl3) 3327
1w), 1744 1s), 1648 1s), 1581 1w), 1531 1s), 1490 1s), 1445
1m), 1281 1s), 1235 1s) cm21; dH 1270 MHz, CDCl3) 7.73±
7.70 12H, m, ortho-H aromatics), 7.56±6.99 17H, m,
aromatics), 6.78 11H, br d, Jˆ7.5 Hz, NH), 5.06 11H, dt,
Jˆ7.5 and 6.0 Hz, NHCHCO), 3.75 13H, s, CO2CH3), 3.33
11H, dd, Jˆ15.0 and 6.0 Hz, NHCHAHBAr), 3.26 11H,
dd,Jˆ15.0 and 6.0 Hz, NHCHAHBAr); dF 1254 MHz,
CDCl3) 2117.6 11F, s, CHvCF); dC 167.5 MHz, CDCl3)
171.9 1CO2CH3), 166.9 1CON), 161.2 1d, 1Jˆ243.0 Hz, CF),
1
mixture of diastereoisomers 1from the H NMR spectrum)
and recovered glycinate 16 10.044 g, scaled to 1.62 g, 26%).
3.9. Photolysis of 1,4-diacetylpiperazine-2,5-dione 18
3
133.7 1C±CvO), 131.7 1d, Jˆ4.0 Hz, C±CHvCH±CF),
3
1,4-Diacetylpiperazine-2,5-dione 1834 15.15 g, 25.9 mmol),
di-tert-butyl peroxide 15.0 equiv., 18.98 g, 23.8 ml,
0.13 mol), benzophenone 15.0 equiv., 23.57 g, 0.13 mol)
and toluene 110.0 equiv., 23.92 g, 27.7 ml, 0.26 mol) in
benzene 1149 ml) were photolysed for 12 h. Evaporation
of the volatile components afforded an orange oil
139.80 g). A sample 11.51 g, 3.8% by mass) was puri®ed
by column chromatography 1silica; petroleum ether±diethyl
ether, 1:1) to elute 1,4-diacetyl-3-benzylpiperazine-2,5-
dione 1935 10.083 g, scaled to 2.19 g, 29%) as a mixture
with 1^)-cis-1,4-diacetyl-3,6-dibenzylpiperazine-2,5-dione
2035,36 10.016 g, scaled to 0.42 g, 4%) and a mixture of
trans-1,4-diacetyl-3,6-dibenzylpiperazine-2,5-dione 2035,36
10.014 g, scaled to 0.37 g, 4%) with 1,4-diacetyl-3-tert-
butoxypiperazine-2,5-dione 21 10.016 g, scaled to 0.42 g,
6%).
131.7 1CvCH±CHvC±CO), 129.0 1d, Jˆ8.0 Hz, CH±
CHvCF), 128.5, 1CHvC±CO), 127.1 1CH±CHvC±
4
CO), 124.1 1d, Jˆ4.0 Hz, FC±CHvCH±CH), 122.9 1d,
2Jˆ16.0 Hz, FCvC), 115.3 1d, 2Jˆ22.0 Hz, FCvCH),
52.8 1NHCHCO), 52.4 1CO2CH3), 31.3 1CHCH2Ar); m/z
1CI, NH3) 302 1M1H1, 100%), 180 120), 122 110), 105
150), 77 110). Found: M1H1, 302.1187. C17H16FNO3
requires for M1H1, 302.1192.
3.10.2. Methyl N-benzoyl-4-¯uorophenylalaninate 26.
White solid; mp79±80 8C 1lit.,22b 96±978C); RF 0.15
1petroleum ether±diethyl ether, 1:1); nmax 1CHCl3) 3433
1w), 3022 1w), 3009 1w), 1740 1m), 1662 1s), 1511 1s),
1486 1w), 1219 1s) cm21; dH 1270 MHz, CDCl3) 7.74±
7.71 12H, apparent d, Jˆ7.0 Hz, ortho-H aromatics),
7.56±7.40 13H, m, aromatics), 7.12±6.93 14H, m,
aromatics), 6.64 11H, br d, Jˆ7.0 Hz, NH), 5.07 11H, dt,
Jˆ7.0 and 5.5 Hz, NHCHCO), 3.77 13H, s, CO2CH3), 3.29
11H, dd, Jˆ14.0 and 5.5 Hz, NHCHAHBAr), 3.19 11H,
dd,Jˆ14.0 and 5.5 Hz, NHCHAHBAr); dF 1254 MHz,
CDCl3) 2115.4 11F, s, CHvCF); dC 167.5 MHz, CDCl3)
171.9 1CO2CH3), 166.7 1CON), 162.0 1d, 1Jˆ245.0 Hz, CF),
3.9.1. 3-tert-Butoxy-1,4-diacetylpiperazine-2,5-dione 21.
The presence of this was con®rmed by the following: RF
0.4 1petroleum ether±diethyl ether, 1:1); dH 1270 MHz,
CDCl3) 6.28 11H, s, NHCHOC1CH3)3), 5.09 11H, d,
Jˆ17.5 Hz, NCHAHBCO), 4.31 11H, d, Jˆ17.5 Hz,
4
NCHAHBCO), 2.54 13H, s, CH3CO), 2.52 13H, s, CH3CO),
133.7 1C±CvO), 131.8 1CvCH), 131.6 1d, Jˆ11.0 Hz,
1
1.28 19H, s, C1CH3)3); m/z 1CI, NH3) 288 1M1NH4
,
C±CHvCH±CF), 130.8 1d, 3Jˆ8.0 Hz, CH±CHvCF),
128.6, 126.9 1CHvC), 115.6 1d, 2Jˆ22.0 Hz, FCvC),
53.5 1NHCHCO), 52.5 1CO2CH3), 37.1 1CHCH2Ar); m/z
1CI, NH3) 302 1M1H1, 100%), 180 120), 122 110), 105
130). Found: M1H1, 302.1187. C17H16FNO3 requires for
M1H1, 302.1192.
100%), 271 1M1H1, 30). Found: M1H1, 271.1292.
C12H18N2O5 requires for M1H1, 271.1294.
3.10. General procedure for photolysis of 8a in the
presence of substituted toluenes and related compounds
3.10.3. Methyl N-benzoyl-2-chlorophenylalaninate 27.
Colourless oil; RF 0.2 1petroleum ether±ethyl acetate,
3:1); nmax 1CHCl3) 3466 1br, m), 3066 1w), 2976 1w),
2955 1w), 1749 1s), 1664 1s), 1524 1s), 1484 1s) cm21; dH
1270 MHz, CDCl3) 7.73 12H, d, Jˆ7.0 Hz, aromatics),
7.71±7.17 17H, m, aromatics), 6.78 11H, d, Jˆ7.5 Hz,
NH), 5.10 11H, dt,Jˆ7.5 and 6.0 Hz, NHCHCO), 3.75
13H, s, CO2CH3), 3.43 11H, dd, Jˆ14.0 and 6.0 Hz,
CHAHBAr), 3.33 11H, dd, Jˆ14.0 and 6.0 Hz, CHAHBAr);
dC 167.5 MHz, CDCl3) 172.0 1CO2CH3), 166.9 1CON),
134.3, 134.1, 133.6 1CvCH), 132.0, 131.7, 131.3, 129.4,
128.6, 128.4, 127.6 1CvCH), 52.9 1NHCHCO), 52.6
1CO2CH3), 35.3 1CH2Ar); m/z 1CI, NH3) 320 137M1H1,
To the pyrex reaction vessel of the 400 W lamp was added
methyl N-benzoylglycinate 8a 15.00 g, 25.9 mmol) and
benzene 1133±176 ml). The solution was degassed and
agitated to aid the dissolving process by use of a steady
stream of dry nitrogen passing through the gas inlet
10.5 h). After this period, di-tert-butyl peroxide
15.0 equiv., 18.98 g, 23.8 ml, 0.13 mol), benzophenone
15.0 equiv., 23.57 g, 0.13 mol) and the toluene, diphenyl-
methane or ¯uorene 110.0 equiv., 0.26 mol) were added to
give in all cases, homogenous solutions. The quartz immer-
sion well was inserted along with the lampand photolysis
proceeded for 12 h. The water ¯ow was regulated to
maintain the temperature of the reaction mixture between

8122
H. S. Knowles et al. / Tetrahedron 57 .2001) 8115±8124
30%), 318 135M1H1, 100), 282 110), 194 110), 161 110),
122 110), 105 120). Found: 35M1H1, 318.0893.
C17H1635ClNO3 requires for M1H1, 318.0897.
1424 1m), 1264 1s) cm21; dH 1270 MHz, CDCl3) 7.76±7.73
12H, m, ortho-H aromatics), 7.59±7.19 112H, m, aromatics),
6.65 11H, d, Jˆ7.5 Hz, NH), 5.14 11H, dt,Jˆ7.5 and 5.5 Hz,
NHCHCO), 3.79 13H, s, CO2CH3), 3.34 11H, dd, Jˆ14.0
and 5.5 Hz, CHAHBAr), 3.27 11H, dd, Jˆ14.0 and 5.5 Hz,
CHAHBAr); dC 167.5 MHz, CDCl3) 172.0 1CO2CH3), 166.8
1NCO), 140.6, 140.0, 134.8, 133.8 1CvCH), 131.8, 129.7,
128.7, 128.6, 127.3, 126.9, 126.8, 126.7 1CvCH), 53.5
1NHCHCO), 52.5 1CO2CH3), 37.5 1CH2Ar); m/z 1CI, NH3)
360 1M1H1, 100%), 238 110), 208 110), 167 110), 122 110),
105 110). Found: M1H1, 360.1589. C23H21NO3 requires for
M1H1, 360.1600.
3.10.4. Methyl N-benzoyl-3-chlorophenylalaninate 28.
White solid; mp91±94 8C; RF 0.7 1petroleum ether±ethyl
acetate, 1:1); nmax 1thin ®lm) 3419 1w), 3324 1m), 3061 1w),
2953 1m), 2929 1w), 1745 1vs), 1660 1vs), 1646 1vs), 1602
1m), 1578 1m), 1525 1s), 1484 1s), 1440 1m), 1360 1w), 1266
1m), 1216 1m) cm21; dH 1270 MHz, CDCl3) 7.93±7.72 12H,
m, aromatics), 7.60±7.00 16H, m, aromatics), 6.69 11H, br d,
Jˆ7.5 Hz, NH), 5.08 11H, dt, Jˆ7.5 and 5.5 Hz,
NHCHCO), 3.77 13H, s, CO2CH3), 3.28 11H, dd, Jˆ14.0
and 6.0 Hz, CHAHBAr), 3.19 11H, dd, Jˆ14.0 and 6.0 Hz,
CHAHBAr); dC 167.5 MHz, CDCl3) 171.8 1CO2CH3), 167.0
1PhCON), 138.0, 134.4, 133.6 1CvCH), 131.9, 129.8,
129.5, 128.7, 127.5, 127.4, 127.2, 127.0 1CvCH), 53.4
1NHCHCO), 52.6 1CO2CH3), 37.6 1CHCH2Ar); m/z 1CI,
NH3) 320 137M1H1, 30%), 318 135M1H1, 100), 284 120),
122 110), 105 110). Found: 35M1H1, 318.0897.
C17H1635ClNO3 requires for 35M1H1, 318.0897.
3.10.8. Methyl N-benzoyl-b-phenylphenylalaninate 32.
White solid; mp134±136 8C; RF 0.1 1dichloromethane±
toluene, 10:1); nmax 1CHCl3) 3437 1m), 3065 1w), 3031
1w), 2955 1w), 2247 1m), 1739 1s), 1666 1s), 1514 1s),
1484 1s), 1451 1m), 1360 1m), 1227 1m) cm21; dH
1270 MHz, CDCl3) 7.61±7.54 12H, m, ortho-H aromatics),
7.50±7.18 113H, m, aromatics), 6.40 11H, br d, Jˆ8.5 Hz,
NH), 5.60 11H, t, Jˆ8.5 Hz, NHCHCO), 4.60 11H, d,
Jˆ8.5 Hz, Ph2CH), 3.54 13H, s, CO2CH3); dC 167.5 MHz,
CDCl3) 171.3 1CO2CH3), 166.2 1CON), 140.0, 139.4
1CvCH), 128.5, 127.2, 127.1, 24.1 1CvCH), 52.8
1CO2CH3), 39.3 1NCH2CO); m/z 1CI, NH3) 360 1M1H1,
100%), 194 1100), 183 160), 167 140), 122 140), 105 170).
Found: M1H1, 360.1594. C23H21NO3 requires for M1H1,
360.1600.
3.10.5. Methyl N-benzoyl-4-chlorophenylalaninate 29.
White solid; mp100±102 8C 1lit.,2e 98±998C); RF 0.62
1petroleum ether±ethyl acetate, 1:1); nmax 1CHCl3) 3424
1m), 3361 1m), 3303 1m), 2951 1w), 1741 1s), 1653 1s),
1529 1m), 1444 1m), 1356 1w), 1298 1w) cm21; dH
1270 MHz, CDCl3) 7.74±7.71 12H, m, aromatics), 7.58±
7.19 15H, m, aromatics), 7.05 12H, d, Jˆ8.5 Hz,
ClCCHvCH or ClCCHvCH±CH), 6.66 11H, d,
Jˆ7.5 Hz, NH), 5.07 11H, dt, Jˆ7.5 and 5.5 Hz,
NHCHCO), 3.76 13H, s, CO2CH3), 3.29 11H, dd, Jˆ14.0
and 6.0 Hz, CHAHBAr), 3.18 11H, dd, Jˆ14.0 and 6.0 Hz,
CHAHBAr); dC 167.5 MHz, CDCl3) 171.8 1CO2CH3), 166.8
1PhCON), 133.6, 132.0, 131.8 1CvCH), 130.6, 128.6,
128.6, 127.1, 126.9 1CvCH), 53.4 1NHCHCO), 52.4
1CO2CH3), 37.1 1CH2Ar); m/z1CI, NH3) 320 137M1H1,
30%), 318 135M1H1, 100), 196 120), 122 110), 105 110),
49 110). Found: 35M1H1, 318.0896. C17H1635ClNO3
requires for 35M1H1, 318.0897.
3.10.9. Methyl N-benzoyl-29H-¯uoren-9-yl)glycinate 33.
Orange solid; mp154±156 8C; RF 0.5 1petroleum ether±
ethyl acetate, 1:1); nmax 1CHCl3) 3434 1w), 3023 1w),
1742 1m), 1665 1s), 1513 1s), 1483 1m), 1448 1w) cm21
;
dH 1270 MHz, CDCl3) 7.82±7.64 13H, m, aromatics),
7.48±7.23 110H, m, aromatics), 5.99 11H, d, Jˆ9.0 Hz,
NH), 5.70 11H, dd, Jˆ9.0 and 3.5 Hz, NHCHCO), 4.72
11H, d, Jˆ3.5 Hz, NHCHCHAr), 3.79 13H, s, CO2CH3);
dC 167.5 MHz, CDCl3) 171.5 1CO2CH3), 167.3 1CON),
142.8, 142.2, 141.1, 141.0, 133.8 1CvCH), 131.6, 128.4,
128.0, 127.6, 127.4, 126.8, 126.7, 124.8, 124.7, 120.2, 119.8
1CvCH), 54.0 1NHCHCO), 52.6 1CO2CH3), 49.2
1NHCHCHAr2); m/z 1CI, NH3) 358 1M1H1, 100%), 326
110), 236 110), 194 120), 162 110), 122 110), 105 120).
Found: M1H1, 358.1429. C23H19NO3 requires for
M1H1, 358.1443.
3.10.6. Methyl N-benzoyl-4-methylphenylalaninate 30.
RF 0.2 1petroleum ether±ethyl acetate, 3:111% toluene);
nmax 1thin ®lm) 3422 1br, w), 3337 1br, w), 2952 1w),
1744 1s), 1646 1s), 1579 1w), 1533 1s), 1520 1s), 1488 1m),
1442 1m), 1216 1m) cm21; dH 1270 MHz, CDCl3) 7.75±7.72
12H, m, aromatics), 7.60±7.26 13H, m, aromatics), 7.10 12H,
d, Jˆ8.0 Hz, 2£CH3C±CHvCH), 7.02 12H, d,Jˆ8.0 Hz,
2£CH3±CHvCH), 5.07 11H, dt,Jˆ7.5 and 5.5 Hz,
NHCHCO), 3.77 13H, s, CO2CH3), 3.26 11H, dd, Jˆ14.0
and 5.5 Hz, CHAHBAr), 3.19 11H, dd, Jˆ14.0 and 5.5 Hz,
CHAHBAr), 2.32 13H, s, CH3CvCH); dC 167.5 MHz,
CDCl3) 172.3 1CO2CH3), 167.0 1CON), 137.0, 134.1,
132.8 1CvCH), 132.0, 129.2, 128.9, 128.5, 127.2
1CvCH), 53.6 1NHCHCO), 52.7 1CO2CH3), 37.6
1CH2Ar); m/z 1CI, NH3) 298 1M1H1, 100%), 194 110),
176 120), 105 120). Found: M1H1, 298.1439. C18H19NO3
requires for M1H1, 298.1443.
3.11. Photolysis of dimethyl 2-2N-benzoylamino)-
malonate 35
A mixture of malonate 3537 14.55 g, 18.1 mmol), di-tert-
butyl peroxide 15.29 g, 6.6 ml, 36.2 mmol) and toluene
18.33 g, 9.6 ml, 90.5 mmol) in benzene 1184 ml) was
photolysed for 12 h using a 400 W lamp. After evaporation
in vacuo, the crude product was isolated as an orange oil
18.13 g). A sample 10.98 g, 12.0% of the crude mass) was
puri®ed by column chromatography 1silica; petroleum
ether±ethyl acetate, 2:1) to afford 1,2-diphenylethane
10.211 g), methyl 2-1N-benzoylamino)-2-benzylmalonate
36 [0.098 g, 0.817 g upon scale up, 13%, 158% based on
recovered malonate)], methyl N-benzoylphenylalaninate
11a [0.017 g, 0.142 g scaled, 3%, 112%)] and starting
malonate 35 10.42 g, scaled upto 3.53 g, 78%).
3.10.7. Methyl N-benzoyl-4-phenylphenylalaninate 31.
White solid; mp130±132 8C; RF 0.5 1petroleum ether±
ethyl acetate, 3:1); nmax 1CHCl3) 3423 1w), 3053 1s), 2985
1m), 1741 1m), 1662 1m), 1542 1m), 1485 1m), 1440 1m),

H. S. Knowles et al. / Tetrahedron 57 .2001) 8115±8124
8123
J.; Elad, D. J. Am. Chem. Soc. 1973, 95, 6418. 1c) Sperling, J.;
Elad, D. J. Am. Chem. Soc. 1971, 93, 3839.
3.11.1. Dimethyl 2-2N-benzoylamino)-2-benzylmalonate
36. Colourless oil; RF 0.4 1petroleum ether±ethyl acetate,
2:1); nmax 1CHCl3) 3419 1m), 3030 1w), 2955 1m), 1744 1s),
1665 1s), 1511 1s), 1480 1s), 1441 1m), 1316 1m), 1284 1s),
1219 1s) cm21; dH 1270 MHz, CDCl3) 7.77±7.74 12H, m,
aromatics), 7.54±7.41 13H, m, aromatics), 7.25±7.20 14H,
m, aromatics), 7.02±6.98 12H, m, aromatic and NH), 3.84
16H, s, 2£CO2CH3), 3.78 12H, s, CH2Ph); dC 167.5 MHz,
CDCl3) 168.0 1CO2CH3), 166.2 1PhCON), 135.0, 133.3
1CvCH), 132.1, 129.8, 128.6, 128.4, 127.3, 127.1
1CvCH), 67.4 1COCCO), 53.6 1CO2CH3), 38.1 1CH2Ph);
m/z 1CI, NH3) 342 1M1H1, 100%), 220 110), 105 120).
Found: M1H1, 342.1342. C19H19NO5 requires for
M1H1, 342.1341.
11. Easton, C. J.; Hay, M. P. J. Chem. Soc., Chem. Commun. 1986,
55.
12. 1a) Obata, N.; Niimura, K. J. Chem. Soc., Chem. Commun.
Â
1977, 238. 1b) Viehe, H. G.; Merenyi, R.; Stella, L.; Janousek,
Z. Angew. Chem., Int. Ed. Engl. 1979, 18, 917. 1c) Benson, O.;
Demirdji, S. H.; Haltiwanger, R. C.; Koch, T. H. J. Am. Chem.
Soc. 1991, 113, 8879.
13. Burgess, V. A.; Easton, C. J.; Hay, M. P.; Steel, P. J. Aust. J.
Chem. 1988, 41, 701.
14. For a preliminary account of this work see: Knowles, H. S.;
Hunt, K.; Parsons, A. F. Tetrahedron Lett. 2000, 41, 7121.
15. This could be explained by b-scission of the captodative
z
radical to expel the sulfonyl radical PhSO2 . 1a) Parsons,
A. F.; Pettifer, R. M. Tetrahedron Lett. 1996, 37, 1667.
1b) Knowles, H.; Parsons, A. F.; Pettifer, R. M. Synlett
1997, 271.
Acknowledgements
16. For photo-deprotection of Cbz-protected glycine see: Pallai,
V. N. R. Synthesis 1980, 1.
17. When this 10.13 mol dm23) reaction was carried out in the
absence of toluene, butanedioate 9a was formed in 48%
yield 155% based on recovered 8a) together with alaninate
10a in 15% yield 117% based on recovered 8a).
We thank A. H. Marks and Co. Ltd and the EPSRC for
funding 1research grant; GR/L58538).
References
18. Reaction of 8a with benzophenone 15 equiv.) and toluene
15 equiv.), in the absence of di-tert-butyl peroxide, gave 11a
in lower yield [12% 125% based on recovered 8a)]. The use of
12 equiv. of) 4,40-dimethoxybenzophenone proved to be
ineffective as a photosensitiser in these reactions.
19. Deseke, E.; Nakatani, Y.; Ourisson, G. Eur. J. Org. Chem.
1998, 243.
1. 1a) Barrett, G. E.; Elmore, D. T. Amino Acids and Peptides;
Cambridge University: Cambridge, 1998. 1b) Calmes, M.;
Daunis, J. Amino Acids 1999, 16, 215.
2. 1a) Malan, C.; Morin, C. J. Org. Chem. 1998, 63, 8019.
1b) Liu, S.; Dockendorff, C.; Taylor, S. D. Org. Lett. 2001,
3, 1571. 1c) Lee, Y.; Silverman, R. B. J. Am. Chem. Soc. 1999,
121, 8407. 1d) Kanoh, K.; Kohno, S.; Katada, J.; Takahashi, J.;
Uno, I.; Hayashi, Y. Bioorg. Med. Chem. 1999, 7, 1451.
1e) Taudien, S.; Schinkowski, K.; Krause, H.-W. Tetrahedron:
Asymmetry 1993, 4, 73.
20. 1a) Maruyama, K.; Hashimoto, M.; Tamiaki, H. J. Org. Chem.
1992, 57, 6143. 1b) Tamiaki, H.; Hashimoto, M.; Maruyama,
K. Bull. Chem. Soc. Jpn 1994, 67, 1987.
21. Methyl N-benzoylsarcosinate 22 can undergo free-radical
halogenation using NBS or sulfuryl chloride in boiling carbon
tetrachloride: Easton, C. J.; Hay, M. P.; Love, S. G. J. Chem.
Soc., Perkin Trans. 1 1988, 265.
3. Tetrahedron Symposia-in-print No. 33; Ed. O'Donnell, M.J.
1988, 44, 5253.
4. See for example: 1a) Boesten, W. H. J.; Seerden, J.-P. G.; de
Lange, B.; Elsenberg, H. J. A.; Kaptein, B.; Moody, H. M.;
Kellogg, R. M.; Broxterman, Q. B. Org. Lett. 2001, 3, 1121.
1b) Myers, A. G.; Gleason, J. L.; Yoon, T.; Kung, D. W. J. Am.
Chem. Soc. 1997, 119, 656. 1c) Ooi, T.; Kameda, M.;
Maruoka, K. J. Am. Chem. Soc. 1999, 121, 6519. 1d) Lygo,
B.; Wainwright, P. G. Tetrahedron Lett. 1997, 38, 8595.
1e) Corey, E. J.; Noe, M. C.; Xu, F. Tetrahedron Lett. 1998,
39, 5349. 1f) Sigman, M. S.; Jacobsen, E. N. J. Am. Chem. Soc.
1998, 120, 4901. 1g) Thierry, B.; Plaquevent, J.-C.; Cahard, D.
Tetrahedron: Asymmetry 2001, 12, 983.
22. 1a) Sutherland, A.; Willis, C. L. Nat. Prod. Rep. 2000, 17, 621.
È
1b) Krause, H.-W.; Kreuzfeld, H.-J.; Dobler, C. Tetrahedron:
Asymmetry 1992, 3, 555.
23. The bond dissociation energies of the benzylic C±H bonds are
as follows: dephenylmethane 1341 kJ mol21); ¯uorene
1,340 kJ mol21); and toulene 1376^2 kJ mol21) CRC Hand-
book of Chemistry andPhysics , Lide, D. R., Ed.; CRC:
Florida, 1999 Table 9±64.
24. Skorna, G.; Ugi, I. Chem. Ber. 1979, 112, 776.
25. Clive, D. L. J.; Anderson, P. C.; Moss, N.; Singh, A. J. Org.
Chem. 1982, 47, 1641.
5. 1a) Easton, C. J. Chem. Rev. 1997, 97, 53. 1b) Easton, C. J.;
Hutton, C. A. Synlett 1998, 457.
26. Iselin, B. M.; Huang, H. T.; MacAllister, H. V.; Niemann, C.
J. Am. Chem. Soc. 1950, 72, 1729.
6. For a recent asymmetric synthesis of a-amino acids using free
radicals see: Miyabe, H.; Ushiro, C.; Ueda, M.; Yamakawa,
K.; Naito, T. J. Org. Chem. 2000, 65, 176.
27. Hamon, D. P. G.; Massy-Westropp, R. A.; Razzino, P.
Tetrahedron 1992, 24, 5163.
28. 1a) Iwama, S.; Katsumura, S. Bull. Chem. Soc. Jpn 1994, 67,
7. Elad, D.; Sinnreich, J. J. Chem. Soc., Chem. Commun. 1965,
471.
È
3363. 1b) Kreuzfeld, H.-J.; Dobler, C.; Kause, H. W.; Facklam,
8. Viehe, H. G.; Janousek, Z.; Merenyi, R. Acc. Chem. Res. 1985,
18, 148.
C. Tetrahedron: Asymmetry 1993, 4, 2047.
29. Fernandez-Megia, E.; Paz, M. M.; Sardina, F. J. J. Org. Chem.
1994, 59, 7643.
9. 1a) Elad, D.; Sperling, J. J. Chem. Soc., Chem. Commun. 1968,
655. 1b) Elad, D.; Sperling, J. J. Chem. Soc. .C) 1969, 1579.
1c) Elad, D.; Sperling, J. J. Chem. Soc., Chem. Commun. 1969,
234. 1d) Sperling, J.; Elad, D. J. Am. Chem. Soc. 1971, 93, 967.
10. 1a) Elad, D.; Schwarzberg, M.; Sperling, J. J. Chem. Soc.,
Chem. Commun. 1970, 617. 1b) Schwarzberg, M.; Sperling,
30. Janes, L. E.; Kazlauskas, R. J. Tetrahedron: Asymmetry 1997,
8, 3719.
31. Shono, T.; Ishige, O.; Uyama, H.; Kashmura, S. J. Org. Chem.
1986, 51, 546.

8124
H. S. Knowles et al. / Tetrahedron 57 .2001) 8115±8124
32. Kober, R.; Papadopoulos, K.; Miltz, W.; Enders, D.; Steglich,
W.; Reuter, H.; Puff, H. Tetrahedron 1985, 41, 1701.
33. Berens, U.; Fischer, C.; Selke, R. Tetrahedron: Asymmetry
1995, 6, 1105.
35. 1a) Kanoh, K.; Kohno, S.; Katada, J.; Takahashi, J.; Uno, I.;
Hayashi, Y. Bioorg. Med. Chem. 1999, 7, 1451. 1b) Machin,
P. J.; Sammes, P. G. J. Chem. Soc., Perkin Trans. 1 1976, 624.
36. Marcuccio, S. M.; Elix, J. A. Aust. J. Chem. 1984, 37, 2397.
37. Dewar, M. J. S.; Turchi, I. J. J. Am. Chem. Soc. 1994, 96, 6148.
34. Peters, D. A.; Beddoes, R. L.; Joule, J. A. J. Chem. Soc.,
Perkin Trans. 1 1993, 1217.