Novel and highly efficient synthesis of substituted dibenz[b,g]1,5- oxazocines. A direct comparison of the solution versus solid-phase approach

Article abstract of DOI:10.1016/S0040-4020(00)00120-4

We describe a novel and highly efficient synthesis of substituted dibenz[b,g] 1,5-oxazocines. The procedure is based on the S(N)Ar of fluoride with the phenolic hydroxide of the properly assembled acyclic intermediate. A direct comparison of the solution-

Full text of DOI:10.1016/S0040-4020(00)00120-4

TETRAHEDRON
Pergamon
Tetrahedron 56 (2000) 2369–2377
Novel and Highly Efficient Synthesis of Substituted
Dibenz[b,g]1,5-oxazocines. A Direct Comparison of the Solution
versus Solid-Phase Approach
*
Xiaohu Ouyang, Zhidong Chen, Longbin Liu, Celia Dominguez and Alexander S. Kiselyov
Small Molecule Drug Discovery, One Amgen Center Drive, Amgen, Inc., Thousand Oaks, CA 91320 USA
Received 14 December 1999; accepted 4 February 2000
Abstract—We describe a novel and highly efficient synthesis of substituted dibenz[b,g]1,5-oxazocines. The procedure is based on the SNAr
of fluoride with the phenolic hydroxide of the properly assembled acyclic intermediate. A direct comparison of the solution-phase vs the
solid-phase synthesis has been conducted. The speed of the synthesis, the purity of the desired eight-membered heterocycles (Ͼ95%), as well
as the diversity of the resultant library are definite advantages of the solid-phase procedure. Yields of dibenz[b,g]1,5-oxazocines synthesized
via the solid-phase approach were generally better, while the purity of the products synthesized by both methods were identical. ᭧ 2000
Elsevier Science Ltd. All rights reserved.
Application of combinatorial chemistry methods to the drug
discovery process considerably shortens the time required
for lead development and lead optimization. However,
along with the unequivocal advantages offered, especially
in the fast assembly of an array of compounds, solid-phase
synthesis poses several formidable challenges. For instance,
it is generally accepted that the purity, as well as the yields
of the products synthesized by a variety of a solid-phase
chemistry techniques, are inferior to the analogous
reactions conducted in solution phase. Additionally,
products synthesized on support contain a linker ‘trace’.
In many instances, the linker ‘trace’ is a polar, or meta-
bolically unstable, moiety (for example, carboxylic acid,
amide, or phenol), which may have an adverse effect on
the biological activity of the product. Solution chemistry,
on the other hand, is much more flexible with regard to the
experimental techniques and selection of substrates.
However, solution chemistry can be very tedious because
of the intermediate purification steps. In order to com-
pare the two synthetic methodologies, namely
‘traditional’ organic synthesis and solid-support synthesis,
we attempted the case study synthesis of dibenz[b, g][1,5]-
oxazocines. Related tricyclic aza-heterocyclic systems are
of interest due to their pronounced activity as strong central
nervous system supressants and anticancer agents. Some
selected examples of these compounds are presented in
Scheme 1.
Chloropromazine and imipramine are well-known agents
used currently for treatment of arteriosclerosis, Parkinson’s
disease, and various cancers.1 Heterocycles of the
dibenz[b,g]1,5-oxazocine series are CNS active, and are
used for the treatment of pain and/or inflammation.2 Two
main strategies have been devised to synthesize this hetero-
cyclic core. One approach involves the reaction of the
dibromide A with primary amines to afford the desired
oxazocines B.3 The alternative path proceeds through the
bromination of the biphenyl ether C followed by its treat-
ment with potassium phthalimide. Hydrazinolysis of the
resultant intermediate, and subsequent intramolecular cycli-
zation of the amine affords the targeted eight-membered
lactam D3 (Scheme 2).
Scheme 1.
Keywords: supported reagents/reactions; substitution.
* Corresponding author. Tel.: ϩ1-805-447-2522; fax: ϩ1-805-498-5714; e-mail: kiselyov@amgen.com
0040–4020/00/$ - see front matter ᭧ 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020(00)00120-4

2370
X. Ouyang et al. / Tetrahedron 56 (2000) 2369–2377
Scheme 2.
Scheme 3.
The double disadvantages of the reported procedures are: (i)
the fact that both are multistep syntheses; and (ii) the forma-
tion of side products. In our group, we were interested in the
application of nucleophilic aromatic substitution (SNAr)4
toward the assembly of medium- and large-ring hetero-
cycles. Some examples of these procedures in the literature
involve the syntheses of vancomycin and the related macro-
cyclic antibiotics,5 and parallel synthesis of dibenz[b,f]1,5-
oxazepines.6 The retrosynthetic analysis of the desired
heterocyclic core suggested that the eight-membered ring
could be accessed by the (SNAr) of fluorine with the phenolic
OH function in the properly assembled noncyclic precursor
(site c). This key intermediate, in its turn, could be prepared
by two sequential reductive amination steps involving
2-fluoro-5-nitrobenzaldehyde (1, site a), a primary amine
R1NH2 (2), and derivatives of salicyclaldehyde (3, site b)
(Scheme 3).
The application of a diverse set of primary amines and
salicylaldehydes introduces two elements of diversity into
the resulting set of dibenz[1,5-a]oxazocines. Subsequent
reduction of the NO2 group, followed by acylation of the
resultant aniline, adds yet another dimension into the result-
ing set of dibenz[b,g]1,5-oxazocines.
In the first set of experiments we developed a solution-phase
approach to dibenz[b,g]1,5-oxazocines. In a typical pro-
cedure, the reductive amination reaction of allylamine
(2B) with 2-fluoro-5-nitrobenzaldehyde (1) in toluene,
using an excess of CH(OMe)3 and acetic acid (AcOH),
Scheme 4. Solution-phase synthesis of dibenz[b,g]1,5-oxazocines. Reagents and conditions: (i) 2, NaBH(OAc)3, AcOH, CH(OMe)3, toluene, 6 h, RT; (ii) 3,
NaBH(OAc)3, AcOH, CH(OMe)3, CH2Cl2, 18 h, RT; (iii) DBU, DMF, 6 h, RT; (iv) H2 10% Pd/C; R3COCl, Hunig’s base, RT, 12 h.

X. Ouyang et al. / Tetrahedron 56 (2000) 2369–2377
2371
Scheme 5. Solid-phase synthesis of dibenz[1,5-a]oxazocines. Reagents and conditions: (i) 20% piperidine, DMF, 30 min, RT; (ii) 1, NaBH(OAc)3, AcOH,
DMF, 8 h, RT; (iii) 3, NaBH(OAc)3, AcOH, DMF, 8 h, RT; (iv) 5% DBU in DMF, 12 h, RT; (v) SnCl2 H2O (1 M in DMF), 12 h, RT; (vi) R3COCl, Hunig’s
base, CH2Cl2, 4 h, RT; (vii) R1CH2Br, DMF, 8 h, RT; Hunig’s base, CH2Cl2, 2 h, RT.
afforded the expected benzylic amine 4 in a 63% yield
(Scheme 4). The second reductive amination step was
successfully accomplished in CH2Cl2 under similar experi-
mental conditions to yield the tertiary amine 5. The intra-
molecular cyclization proceeded smoothly upon treatment
of 5 with an excess of 1,8-diazabicyclo-[5.4.0]undec-7-ene
(DBU) in dimethylformamide (DMF) to afford the expected
heterocycle 6b in a 66% yield from 4 (42% overall yield).
Similarly, a set of dibenz[b,g]1,5-oxazocines 6a, 6c, and 6d
were prepared in 12–77% overall yields starting from 1. We
attempted to use several other bases to facilitate this cycli-
zation. For example, K2CO3,7 Cs2CO3 (suspension in DMF,
5–15 fold excess relative to substrate), KOt-Bu or Bu4NF8
(0.4 M solutions in THF, 20 fold excess relative to
substrate), and CsF9 were found to afford the desired
eight-membered heterocycles, albeit in much lower yields.
Reduction of the nitro group in 6 with H2 over 10% Pd on
charcoal, followed by the acylation of the resultant tricyclic
anilines with MeCOCl or p-Cl–C6H4COCl, proceeded
smoothly to afford the trisubstituted dibenz[b,g]1,5-oxazo-
cines 7a,b in 4–7% overall yield from 1 (for the yields of the
individual steps, see Table 2 vide infra).
12–75% yield). Alternatively, cyclization of 10 with DBU
was followed by the reduction of the nitro group with 1 M
SnCl2·H2O in DMF. Subsequent treatment of the resin with
CH3COCl or p-Cl–C6H4COCl in DMF, alkylation with
R1CH2Br, and cleavage with Hunig’s base afforded the
expected eight-membered heterocycles in 20–78% yield
(Table 1, 7a–h).
Neither the substitution pattern of the salicyclic aldehyde 3,
nor the nature of the substituents affected the yields of the
targeted heterocycles. The notable exception, however, was
2-hydroxy-1-naphthoic aldehyde (Table 1, 12% isolated
yield of the corresponding dibenz[b,g]1,5-oxazocine 6h).
Our attempts to further optimize reaction conditions for
this entry were unsuccessful, primarily due to the poor
yield of the reductive amination step (Scheme 5 step iii).
The direct comparison of the isolated yields for the selected
9-nitro-, and 9-N-acylated-dibenz[b,g]1,5-oxazocines syn-
thesized via the two synthetic strategies is summarized in
Table 2.
In general, the yields of the targeted dibenz[b,g]1,5-oxazo-
cines synthesized on solid support are considerably higher
(except for the entry 6d). The solid-support strategy
appears to be most beneficial for the sequences involving
more than 2–3 steps, given that the chemistry for each step
of the sequence is well validated. The syntheses of
dibenz[b,g]1,5-oxazocines 7a and 7b are illustrative.
Whereas the solid support synthesis resulted in 78 and
45% yield of these heterocycles, respectively, a similar
sequence conducted via the ‘standard’ solution-phase
approach afforded only 7 and 4% overall yields. Impor-
tantly, the purities of the final products 6 and 7 synthesized
by both strategies are virtually identical by LC MS. Both 1H
NMR and combustion analyses data indicate that the puri-
ties of the materials cleaved off the solid support were
greater than 95%.
In the solid-support approach, b-alanine immobilized on
Wang resin was selected as a starting point of the synthesis
(Scheme 5, 8). The rationale for the choice of resin was that
the final products can be conveniently cleaved off the
support using the alkylation-retro-Michael reaction
sequence
to
afford
the
targeted
N-alkylated
dibenz[b,g]1,5-oxazocines.10 In the optimized procedure,11
reductive amination of b-alanine resin with 2-fluoro-5-
nitrobenzaldehyde and NaBH(OAc)3 in DMF, in the
presence of 1% TFA, took place smoothly to afford the
desired secondary amine (9). The second reductive amina-
tion with salicylic aldehyde (3a) afforded the expected
tertiary amine in 83% yield (10, as determined by TFA
cleavage). The intermediate was treated with 5% DBU in
DMF to afford the targeted immobilized 9-nitro-
dibenz[b,f ]1,5-oxazocines. The resulting resin was treated
with the alkylating agent (R1CH2Br) for 8 h at room
temperature followed by Hunig’s base to result in the
expected 9-nitro-dibenzo[b,f ]oxazocines (Table 1, 6a–i,
In summary, we have described an efficient synthesis of
dibenz[b,g]1,5-oxazocines using either a solution-phase or
solid support approach. The procedure is based upon the

2372
X. Ouyang et al. / Tetrahedron 56 (2000) 2369–2377
Table 1. Yields and purities of dibenz[b,g]-oxazocines 6, and 7 synthesized via the solid support strategy
HPLC
purity,a
(%)
Retention
time
(min)
Isolated
yield
(%)
R1
H
R2
R3
6a
6b
H
H
NO2
NO2
100
100
1.01
1.12
59
50
6c
H
H
NO2
NO2
100
100
1.62
1.85
65
65
6d
6e
6f
6-OMe
5-OMe
4-OMe
NO2
NO2
NO2
100
98
1.25
1.28
1.33
75
55
65
6g
100
6h
NO2
100
1.57
12
6i
5-Br
H
NO2
98
1.48
0.68
23
78
7a
H
H
NHCOMe
100
7b
7c
H
100
100
1.65
1.65
45
69
6-OMe
7d
7e
5-OMe
4-OMe
100
100
1.68
1.73
65
20
7f
100
100
1.28
1.50
45
42
7g
5-Br
6-F
NHCOMe
NHCOMe
7h
100
1.30
43
a The column employed was YMC Pak ProC18 column, 50×4.6 mm2. The solvent system was MeCN/H2O (start: 20:80; finish: 100:0; 3 min runs; 0.1% TFA
added), with a flow rate of 3 mL/min.

X. Ouyang et al. / Tetrahedron 56 (2000) 2369–2377
2373
Table 2. Selected yields for 9-nitro- and 9-N-substituted-dibenz[b,g]1,5-oxazocines synthesized in solution vs on solid support (yields are reported for isolated,
analytically pure compounds)
Compound
Solution synthesis
Solid-phase
synthesis overall
yield (%)
Yield of
step 1
(%)
Yield of
steps 2ϩ3
Yield of
steps 4ϩ5
(%)
Overall
yield
(%)
(%)
6a
6b
6c
6d
7a
7b
54
63
76
96
54
54
22
66
78
80
22
22
12
42
59
77
7
59
50
65
65
78
45
58
36
4
intramolecular nucleophilic aromatic substitution of fluorine
in the derivatives of 2-fluoro-5-nitrobenzaldehyde with the
OH function of the immobilized phenols. Direct comparison
of solution vs solid-phase synthesis has been conducted. The
relative speed of the synthesis, purity of the desired eight-
membered heterocycles (Ͼ95%), as well as the diversity of
the resultant library are definite advantages of the solid-
phase procedure. Yields of dibenz[b,g]1,5-oxazocines
synthesized via the solid-phase approach were generally
better, while the purity of the products synthesized by
both methods were identical.
[(2-Fluoro-5-nitrophenyl)methyl]prop-2-enylamine (4B).
To a 1000-mL, round-bottomed flask was added 2-fluoro-5-
nitrobenzaldehyde (1, 15 g, 88.8 mmol) and allylamine (2B,
10.0 mL, 133 mmol) in 300 mL toluene, followed by the
addition of acetic acid (25 mL, 88.8 mmol), trimethyl
orthoformate (10 mL, 88.0 mmol), and NaBH(OAc)3
(20 g, 94 mmol). The solution was stirred at room
temperature for 6 h. 30 mL of MeOH were added, and the
solution was stirred for an additional 0.5 h. The reaction
solvent was evaporated in vacuo, and the resulting
crude material was purified by flash chromatography on
silica gel with MeOH/CH2Cl2/NH4OH (5:94:1) as
1
eluant to afford 11.75 g (63%) of 4B as a brown oil. H
NMR (CDCl3): d 3.30 (d, Jˆ6.0 Hz, 2H), 3.92 (s, 2H),
5.10–5.30 (m, 2H), 5.80–6.00 (m, 1H), 8.10–7.40 (m,
2H). MS m/z: 211 (Mϩ1). Anal. calcd for C10H11FN2O2.:
C, 57.14; H, 5.27; N, 13.33. Found: C, 57.03; H, 5.11; N,
13.26.
Experimental Procedure
Materials and methods
All solid-phase reactions were carried out in peptide
synthesis vessels and agitated on an orbit shaker at room
temperature. Reagents were purchased from Aldrich and
used without further purification. The Fmoc-b-Ala-Wang
resin (100–200 mesh) purchased from Novabiochem with
a loading of 0.80 mmol/g was deprotected with 40% piper-
idine in DMF, followed by washing with DMF, MeOH, and
DCM before use. Concentration of the solutions after
workup was performed by reduced pressure rotary evapora-
2-({[(2-Fluoro-5-nitrophenyl)methyl]prop-2-enylamino}-
methyl)phenol (5Ba). To a 25-mL, round-bottomed flask
was added 4B (163 mg, 0.776 mmol) and salicylaldehyde
(3a, 63 mg, 0.517 mmol) in 8 mL CH2Cl2, followed by the
addition of acetic acid (310 mg, 5.17 mmol), trimethyl
orthoformate (55 mg, 0.517 mmol), and NaBH(OAc)3
(0.55 g, 2.58 mmol). The solution was stirred at room
temperature for 18 h. 5 mL of MeOH were added, and the
solution was stirred for an additional 0.5 h. The reaction
solvent was evaporated in vacuo, and the resulting crude
material was purified by flash chromatography on silica
gel with EtOAc/Hexane (20:80) as eluant to afford
130 mg (80%) of 5Ba as a yellow oil. 1H NMR
(CDCl3): d 3.20 (d, Jˆ6.72 Hz, 2H), 3.80 (s, 4H), 5.24–
5.33 (m, 2H), 5.90–6.00 (m, 1H), 6.70–8.20 (m, 7H), 10.10
(s, 1H). MS m/z: 317 (Mϩ1). Anal. calcd for C17H17FN2O3:
C, 64.55; H, 5.42; N, 8.86. Found: C, 64.32; H, 5.33; N, 8.71.
1
tion on a Bu¨chi 535 apparatus. H NMR and 13C NMR
spectra were obtained on a Bruker 400 instrument. MS
analyses (ES and CI modes) were performed on a Perkin
Elmer API 165 instrument. Melting points were measured
with a Bu¨chi 535 melting point apparatus. Elemental
analyses were performed by Atlantic Microlab Inc,
Norcrosss, GA. HPLC analyses were performed on a
Beckman Gold Analytic 126 apparatus with a diode array
detector model 168 at the wavelengths of 220 nm and
254 nm. The column employed was a YMC Pak ProC18
column, 50×4.6 mm2. The solvent system was MeCN/H2O
(start: 20:80; finish: 100:0; 3 min runs; 0.1% TFA added),
with a flow rate of 3 mL/min.
3-Nitro-6-prop-2-enyl-5H,
7H-dibenzo[b,g]1,5-oxazo-
cine (6b). To a 10 mL, round-bottomed flask was added
2-fluoro-5-nitro-(N-allyl-N-2-hydroxybenzyl) benzylamine
(5Ba, 88 mg, 0.28 mmol) and 1,8-diazabicyclo[5.4.0]-
undec-7-ene (50 mg, 0.33 mmol) in 3 mL of DMF. The
solution was stirred at room temperature for 6 h. The reac-
tion solvent was evaporated in vacuo, and the resulting
crude material was purified by flash chromatography on
silica gel with EtOAc/Hexane (20:80) as eluant to afford
71 mg (86.5%) of 6b as a light brown oil.
Solution phase synthesis of 3-nitro-6-prop-2-enyl-5H,
7H-dibenzo[b,g]1,5-oxazocine (6b)
The procedure reported below for the synthesis of 6b is the
typical protocol for the solution-phase synthesis of
dibenz[b,g]1,5-oxazocines 6a–6d.

2374
X. Ouyang et al. / Tetrahedron 56 (2000) 2369–2377
Example of solution phase synthesis of N-acylated
dibenzo[b,g]1,5-oxazocines (7)
(25 mL) in DMF for 8 h. The resin containing the quater-
nized amine on solid support was filtered, washed with
MeOH, CH2Cl2, and DMF, and treated with a 5% Hunig’s
base in CH2Cl2 (50 mL) for 2 h. The filtrate was collected
and washed twice with a saturated NaHCO3 solution. The
extract was dried over MgSO4, filtered and co-evaporated
twice with MeOH to afford 266 mg of the desired product as
a yellow oil (HPLC purity: 100%). The product was further
purified by flash chromatography on silica gel with EtOAc/
hexane (20:80) as eluent to afford 237 mg (50% yield) of
analytically pure 6b as a light brown oil.
N-[6-Methyl-5H, 7H-benzo[b]benzo[3,4-g]1,5-oxazocin-
3-yl)acetamide (7a). To a 150-mL flask was added 6a
(65 mg, 0.24 mmol), 10% Pd/C (6.5 mg) and ethanol
(20 mL). The mixture was stirred under a hydrogen balloon
for 12 h. The reaction mixture was filtered through Celite
and concentrated to afford 55 mg (95%) of a light brown
solid. MS m/z: 241 (Mϩ1). This material (25 mg,
0.104 mmol) was dissolved in 1.5 mL of CH2Cl2. Acetyl
chloride (16.3 mg, 0.208 mmol) and Hunig’s base (27 mg,
0.208 mmol) were added by syringe, and the reaction
mixture was stirred at room temperature for 12 h. The
reaction mixture was treated with sat. NaHCO3, and
extracted with CH2Cl2 three times. The combined organic
layers were dried over anhydrous Na2SO4, filtered, and
concentrated in vacuo. The resulting crude oil was purified
by flash chromatography on silica gel with MeOH/CH2Cl2/
NH4OH(4:95:1) as eluant to afford 18 mg (61%) of an off-
white solid. MS m/z: 283 (Mϩ1).
Analytical data
6-Methyl-3-nitro-5H, 7H-dibenzo[b,g]1,5-oxazocine (6a).
Light yellow solid. mp 97–99ЊC; 1H NMR (CD2Cl2) d 8.14
(s, 1H), 8.00 (s, 1H), 7.33–7.19 (m, 5H), 4.7–2.0 (m, 7H)
13C NMR (CD2Cl2) d 133.2, 130.4, 125.6, 121.0, 6.12, 53.7,
42.8. HRMS (FAB) calcd for C15H14N2O3: m/zˆ271.1083
(MHϩ), found: 271.1095. Anal. calcd for C15H14N2O3: C,
66.66; H, 5.22; N, 10.36. Found: C, 66.55; H, 5.11; N, 10.36.
(4-Chlorophenyl)-N-(6-methyl(5H, 7H-benzo[b]benzo[3,4-
g]1,5-oxazocin-3-yl))carboxamide (7b). The reaction
conditions were analogous to the synthesis of 7a, except
4-chlorobenzoyl chloride (36.4 mg, 0.208 mmol) was used
as the acylating agent. The resulting crude oil was purified
by flash chromatography on silica gel with EtOAc/hexane
(20:80) as eluant to afford 15 mg (38.1%) of an off-white
solid. MS m/z: 379, 381 (Mϩ1).
3-Nitro-6-prop-2-enyl-5H, 7H-dibenzo[b,g]1,5-oxazocine
(6b). Yellow oil. 1H NMR (CD2Cl2) d 8.15 (dd,
J1ˆ8.8 Hz, J2ˆ2.8 Hz, 1H), 7.98 (d, Jˆ2.8 Hz, 1H), 7.5–
7.2 (m, 5H), 5.90 (m, 1H), 5.23 (d, Jˆ6.8 Hz, 2H), 4.65–
2.95 (m, 6H); 13C NMR (CD2Cl2) d 136.7, 133.1, 130.4,
129.9, 125.8, 125.6, 120.6, 118.6, 58.1, 51.7, 51.0; HRMS
(FAB) calcd for C17H16N2O3: m/zˆ297.1239 (MHϩ), found:
297.1251. Anal. calcd for C17H16N2O3: C, 68.91; H, 5.44; N,
9.45. Found: C, 68.70; H, 5.44; N, 9.44.
Solid phase synthesis of dibenzo[b,g]1,5-oxazocines
6a–6i
3-Nitro-6-benzyl-5H, 7H-dibenzo[b,g]1,5-oxazocine (6c).
Yellow needles. mp 149–152ЊC; 1H NMR (CD2Cl2) d 8.16
(dd, J1ˆ8.8 Hz, J2ˆ2.8 Hz, 1H), 7.9–7.0 (m, 11H), 4.7–3.1
(m, 6H); 13C NMR (CD2Cl2) d 138.9, 132.7, 130.0, 129.1,
128.8, 127.6, 125.2, 120.0, 58.8, 55.8, 51.2; HRMS (FAB)
calcd for C21H18N2O3: m/zˆ347.1396 (MHϩ), found:
347.1409. Anal. calcd for C21H18N2O3: C, 72.82; H, 5.24;
N, 8.09. Found: C, 72.67; H, 5.29; N, 8.12.
The procedure reported below for the synthesis of 3-nitro-6-
prop-2-enyl-5H, 7H-dibenzo[b,g]1,5-oxazocine (6b) is a
representative example of the solid-phase synthesis of
dibenzo[b,g]1,5-oxazocines 6a–6i.
Deprotected b-Ala Wang resin (8, 100 g, 0.8 mmol/g load-
ing, available from NovaBiochem) was treated with a
mixture of trimethyl orthoformate (480 mL), glacial
AcOH (20 mL), and 2-fluoro-5-nitrobenzaldehyde (1,
43 g, 250 mmol). The slurry was gently stirred under
nitrogen for 3 h. The resin was filtered and washed with
anhydrous CH2Cl2 followed by DMF. The resulting immo-
bilized imine of 2-fluoro-5-nitrobenzaldehyde (100 g) was
treated with a solution of NaBH(OAc)3 (106 g, 500 mmol)
and 5 mL of glacial AcOH in 500 mL of anhydrous DMF.
The slurry was gently stirred under nitrogen for another 5 h.
The resulting resin 9 (100 g) was filtered, washed twice with
MeOH, DCM, DMF, dioxane and Et2O, and treated with a
mixture of salicylaldehyde (3a, 31 g, 250 mmol), and
NaBH(OAc)3 (106 g, 500 mmol) in 500 mL of anhydrous
DMF for an additional 8 h. The resin was filtered, washed
twice with MeOH, CH2Cl2, DMF, dioxane, and Et2O, and
dried in vacuo to afford the precursor 10 for the intramole-
cular cyclization. 2 g of the resultant resin was shaken with
50 mL of a 5% solution of DBU in DMF at room tempera-
ture for 12 h. The resin was filtered and treated with 10%
AcOH in DMF to remove excess DBU. The resin was then
washed with DMF, MeOH, CH2Cl2, and dried in vacuo. The
dried resin was treated with a 1 M solution of allyl bromide
6-[(4-Chlorophenyl)methyl]-3-nitro-5H, 7H-dibenzo[b,g]-
1,5-oxazocine (6d). Light orange powder. mp 165–168ЊC;
1H NMR (CD2Cl2) d 8.16 (dd, J1ˆ8.8 Hz, J2ˆ2.8 Hz, 1H),
7.9–6.9 (m, 10H), 4.7–3.1 (m, 6H); 13C NMR (CD2Cl2) d
133.0, 130.9, 130.5, 129.3, 125.7, 120.4; HRMS (FAB)
calcd for C21H17ClN2O3: m/zˆ381.1006 (MHϩ), found:
381.0995. Anal. calcd for C21H17ClN2O3: C, 66.23; H,
4.50; N, 7.36: Cl, 9.31. Found: C, 66.25; H, 4.57; N, 7.31;
Cl, 9.37.
1-Methoxy-9-nitro-6-prop-2-enyl-5H, 7H-dibenzo[b,g]1,5-
1
oxazocine (6e). Light yellow solid. mp 121–123ЊC; H
NMR (CD2Cl2) d 8.12 (dd, J1ˆ8.8 Hz, J2ˆ2.8 Hz, 1H),
7.91 (d, Jˆ2.8 Hz, 1H), 7.35 (dd, J1ˆ6.8 Hz, J2ˆ2.8 Hz,
1H), 7.2–7.0 (m, 3H), 5.79 (m, 1H), 5.14 (d, Jˆ8.8 Hz,
2H), 4.5–2.7 (m, 6H); 13C NMR (CD2Cl2) d 138.1, 131.3,
129.2, 127.7, 126.8, 125.6, 122.8, 121.8, 119.7, 114.8, 59.7,
58.1, 53.7, 52.8; HRMS (FAB) calcd for C18H18N2O4:
m/zˆ327.1345 (MHϩ), found: 327.1349. Anal. calcd for
C18H18N2O4: C, 66.25; H, 5.56; N, 8.58. Found: C, 66.17;
H, 5.59; N, 8.51.

X. Ouyang et al. / Tetrahedron 56 (2000) 2369–2377
2375
2-Methoxy-9-nitro-6-prop-2-enyl-5H, 7H-dibenzo[b,g]1,5-
oxazocine (6f). Light yellow solid. mp 119–122ЊC; H
(50 mL) in DMF for 8 h. The resultant quaternized amine
on solid support was filtered, washed with MeOH, CH2Cl2,
and DMF, and treated with 5% Hunig’s base in CH2Cl2
(100 mL) for 2 h. The filtrate was collected and washed
twice with a saturated NaHCO3 solution. The extract was
dried over MgSO4, filtered, and co-evaporated twice with
MeOH to afford the desired products as solids (HPLC
purity: 100%). The compounds were further purified by
flash chromatography on silica gel with EtOAc/hexane
(20:80) as eluent to afford the analytically pure 7a–7h in
20–78% yields.
1
NMR (CD2Cl2) d 8.13 (dd, J1ˆ8.8 Hz, J2ˆ2.8 Hz, 1H),
7.96 (d, Jˆ2.8 Hz, 1H), 7.32 (s, 1H), 7.10 (s, 1H), 6.96 (d,
Jˆ8.0 Hz, 1H), 6.78 (s, 1H), 5.89 (m, 1H), 5.18 (d,
Jˆ9.2 Hz, 2H), 4.5–2.8 (m, 6H), 3.91 (s, 3H); 13C NMR
(CD2Cl2) d 161.4, 136.8, 133.7, 129.9, 125.6, 121.0, 118.6,
117.2, 111.3, 105.9, 58.1, 56.3, 51.1; HRMS (FAB) calcd
for C18H18N2O4: m/zˆ327.1345 (MHϩ), found: 327.1342.
Anal. calcd for C18H18N2O4: C, 66.25; H, 5.56; N, 8.58.
Found: C, 66.10; H, 5.55; N, 8.50.
9-Methoxy-3-nitro-6-prop-2-enyl-5H, 7H-dibenzo[b,g]1,5-
oxazocine (6g). Yellow thick oil. 1H NMR (CD2Cl2) d 8.11
(dd, J1ˆ8.8 Hz, J2ˆ2.8 Hz, 1H), 7.97 (d, Jˆ2.8 Hz, 1H),
7.5–6.7 (m, 4H), 5.90 (m, 1H), 5.22 (d, Jˆ6.8 Hz, 2H),
4.58–2.79 (m, 6H), 3.78 (s, 3H); 13C NMR (CD2Cl2) d
155.9, 135.0, 128.5, 126.3, 123.8, 122.7, 119.8, 119.2,
118.6, 115.8, 114.7, 113.5, 56.5, 54.6, 50.4, 49.6; HRMS
(FAB) calcd for C18H18N2O4: m/zˆ327.1345 (MHϩ), found:
327.1340. Anal. calcd for C18H18N2O4: C, 66.25; H, 5.56; N,
8.58. Found: C, 66.00; H, 5.56; N, 8.54.
Analytical data
N-(6-Methyl-5H, 7H-benzo[b]benzo[3,4-g]1,5-oxazocin-
3-yl)acetamide (7a). Pale yellow solid. mp 71–73ЊC; H
1
NMR (CD2Cl2) d 8.03 (s, 1H), 7.43 (dd, J1ˆ8.6 Hz,
J2ˆ2.5 Hz, 1H), 7.3–7.0 (m, 6H), 4.7–1.9 (m, 7H), 2.08
(s, 3H); 13C NMR (CD2Cl2) d 171.3, 162.4, 158.5, 137.7,
134.8, 132.1, 127.5, 126.1, 125.2, 123.6, 122.2, 62.2, 44.7,
39.5, 26.8; HRMS (FAB) calcd for C17H18N2O2:
m/zˆ283.1447 (MHϩ), found: 283.1446. Anal. calcd for
C17H18N2O2: C, 72.32; H, 6.43; N, 9.92. Found: C, 72.21;
H, 6.32; N, 9.69.
11-Nitro-8-prop-2-enyl-7H, 9H-benzo[g]naphtho[2,1-b]1,5-
1
oxazocine (6h). Light yellow solid. mp 146–148ЊC; H
NMR (CD2Cl2) d 8.07 (dd, J1ˆ9.0 Hz, J2ˆ2.8 Hz, 1H),
8.01 (d, Jˆ9.0 Hz, 1H), 7.85 (d, Jˆ2.8 Hz, 1H), 7.78 (d,
Jˆ9.0 Hz, 2H), 7.50 (t, Jˆ7.6 Hz, 1H), 7.41 (t, Jˆ7.6 Hz,
1H), 7.25 (t, Jˆ8.0 Hz, 2H), 5.92 (m, 1H), 5.20 (t,
Jˆ7.6 Hz, 2H), 3.99 (s, 2H), 3.59 (s, 2H), 3.03 (s, 2H);
13C NMR (CD2Cl2) d 136.6, 133.6, 132.5, 130.9, 130.3,
129.2, 127.8, 126.2, 125.6, 124.7, 121.2, 120.7, 120.4,
118.9, 58.7, 52.0, 47.4; HRMS (FAB) calcd for
C21H18N2O3: m/zˆ347.1396 (MHϩ), found: 347.1393.
Anal. calcd for C21H18N2O3: C, 72.82; H, 5.24; N, 8.09.
Found: C, 72.54; H, 5.28; N, 7.98.
(4-Chlorophenyl)-N-(6-methyl(5H,7H-benzo[b]benzo[3,4-
g]1,5-oxazocin-3-yl))carboxamide (7b). Pale solid. mp
1
189–191ЊC; H NMR (CD2Cl2) d 7.69 (d, Jˆ8.6 Hz, 3H),
7.44 (dd, J1ˆ8.6 Hz, J2ˆ2.5 Hz, 1H), 7.36 (d, Jˆ8.6 Hz,
2H), 7.27 (s, 2H), 7.17 (t, Jˆ6.8 Hz, 2H), 6.98 (m,
2H), 4.58–3.51 (m, 4H), 2.14–1.88 (m, 3H). 13C
NMR (d6-DMSO) d 162.4, 157.8, 153.7, 134.6, 133.7,
131.7, 130.2, 127.8, 127.5, 126.6, 125.9, 123.0, 122.0,
120.9, 119.5, 117.3, 58.0, 50.9, 35.0. HRMS (FAB)
calcd
for
C22H19ClN2O2:
m/zˆ283.1447379.1213
(MHϩ), found: 379.1222. Anal. calcd for C22H19ClN2O2:
C, 69.75; H, 5.05; N, 7.39. Found: C, 69.57; H, 5.02; N,
7.34.
9-Bromo-3-nitro-6-prop-2-enyl-5H, 7H-dibenzo[b,g]1,5-
1
oxazocine (6i). Yellow solid. mp 123–126ЊC; H NMR
(CD2Cl2) d 8.06 (dd, J1ˆ9.0 Hz, J2ˆ2.8 Hz, 1H), 7.91 (d,
Jˆ2.8 Hz, 1H), 7.35 (dd, J1ˆ6.8 Hz, J2ˆ2.8 Hz, 1H), 7.2–
7.0 (m, 3H), 5.79 (m, 1H), 5.14 (d, Jˆ8.6 Hz, 2H), 4.5–2.7
(m, 6H); 13C NMR (CD2Cl2) d 136.4, 135.6, 133.2, 129.9,
125.7, 124.5, 122.6, 121.1, 120.4, 118.8, 58.1, 51.2;
HRMS (FAB) calcd for C17H15BrN2O3: m/zˆ375.0345
(MHϩ), found: 375.0355. Anal. calcd for C17H15BrN2O3:
C, 54.42; H, 4.03; N, 7.47. Found: C, 54.27; H, 4.12; N,
7.41.
(4-Chlorophenyl)-N-(11-methoxy-6-prop-2-enyl(5H,7H-
benzo[b]benzo[3,4-g]1,5-oxazocin-3-yl))carboxamide
(7c). Pale solid. mp 149–152ЊC; 1H NMR (CD2Cl2) d 8.15
(s, 1H), 7.80 (d, Jˆ8.6 Hz, 2H), 7.54 (dd, J1ˆ8.6 Hz,
J2ˆ2.8 Hz, 1H), 7.44 (d, Jˆ8.6 Hz, 2H), 7.36 (s, 1H),
7.05 (t, Jˆ7.6 Hz, 1H), 6.96 (d, Jˆ8.6 Hz, 1H), 6.74 (d,
Jˆ6.8 Hz, 1H), 5.90 (m, 1H), 5.16 (d, Jˆ6.8 Hz, 2H),
3.96 (s, 6H), 4.58–2.28 (m, 3H); 13C NMR (CD2Cl2) d
165.4, 138.6, 137.1, 134.2, 129.6, 129.4, 125.4, 124.6,
121.7, 117.9, 113.2, 58.6, 56.6, 53.2; HRMS (FAB) calcd
for C25H23ClN2O3: m/zˆ435.1475 (MHϩ), found: 435.1495.
Anal. calcd for C25H23ClN2O3: C, 69.04; H, 5.33; N, 6.44.
Found: C, 69.00; H, 5.35; N, 6.52.
Solid phase synthesis of dibenzo[b,g]1,5-oxazocines
7a–7h
The procedure reported below is typical for the solid-phase
synthesis of dibenzo[b,g]1,5-oxazocines 7a–7h.
(4-Chlorophenyl)-N-(10-methoxy-6-prop-2-enyl(5H,7H-
benzo[b]benzo[3,4-g]1,5-oxazocin-3-yl))carboxamide
(7d). Pale solid. mp 147–150ЊC; 1H NMR (CD2Cl2) d 8.23
(s, 1H), 7.81 (d, Jˆ8.0 Hz, 2H), 7.60 (dd, J1ˆ8.6 Hz,
J2ˆ2.2 Hz, 1H), 7.44 (d, Jˆ8.6 Hz, 4H), 7.02 (d,
Jˆ8.0 Hz, 1H), 6.65 (dd, J1ˆ8.6 Hz, J2ˆ2.8 Hz, 1H), 5.90
(m, 1H), 5.18 (d, Jˆ6.8 Hz, 2H), 3.96 (s, 4H), 4.58 (s, 1H),
3.65 (s, 2H), 2.56 (m, 2H); 13C NMR (CD2Cl2) d 165.4,
138.6, 137.1, 134.2, 129.6, 129.4, 125.4, 124.6, 121.7,
117.9, 113.2, 58.6, 56.6, 53.2; HRMS (FAB) calcd for
The resin resulting from the DBU cyclization of 10 (2 g)
was treated with 25 mL of a 1 M solution of SnCl2·H2O in
DMF for 12 h at room temperature. After filtration, the resin
was washed with MeOH, CH2Cl2, DMF, and treated with a
mixture of acylating agent and Hunig’s base (0.2 M solution
for each component, 25 mL overall volume) in CH2Cl2 for
4 h. The resin was filtered, washed with MeOH, CH2Cl2,
DMF, and treated with a 1 M solution of allyl bromide

2376
X. Ouyang et al. / Tetrahedron 56 (2000) 2369–2377
C25H23ClN2O3: m/zˆ435.1475 (MHϩ), found: 435.1487.
Anal. calcd for C25H23ClN2O3: C, 69.04; H, 5.33; N, 6.44.
Found: C, 69.0; H, 5.35; N, 6.38.
18, 10150. (e) Kuo, C. Biophys. J. 1998, 75, 2845. (f) Ocharan,
M. E.; Asbun, J.; Valencia, I.; Castillo, C.; Castillo, E. F. Proc.
West. Pharmacol. Soc. 1998, 41, 141; Gallego, M.; Casis, L.;
Casis, O. Eur. J. Pharmacol. 1998, 360, 113. (g) Jagadeesh, S.
R.; Subhash, M. N. Biol. Psychiatry 1998, 44, 617. (h) Adali, O.;
Carver, G. C.; Philpot, R. M. Arch. Biochem. Biophys. 1998, 358,
92. (i) Casis, O.; Sanchez-Chapula, J. A. J. Cardiovasc. Pharma-
col. 1998, 32, 521. (j) Zhu, J.; Mix, E.; Bengtsson, B.-O.; Link, H.
J. Peripher. Nerv. Syst. 1997, 2, 30. (k) Muraoka, S.; Kamei, K.;
Muneoka, K.; Takigawa, M. J. Neurochem. 1998, 71, 1709. (l)
Boyer, P.-A.; Skolnick, P.; Fossom, L. H. J. Mol. Neurosci.
1998, 10, 219. (m) Wuonola, M.; Palfreyman, M. G.; Motohashi,
N.; Kawase, M.; Gabay, S.; Gupta, R. R.; Molnar, J. Anticancer
Res. 1998, 18, 337. (n) Pajeva, I.; Wiese, M. J. Med. Chem. 1998,
41, 1815.
2. (a) Joergensen, T. K.; Andersen, K. E.; Andersen, H. S.;
Hohlweg, R.; Madsen, P.; Olsen, U. B. Dutch Patent 96-DK 138
960401; CA 1996, 713004. (b) Ohgoh, T.; Tanaka, S.; Fujimoto,
M.; Kitahara, A. Yakugaku Zassi 1977, 97, 24. (c) Gellatly, R. P.;
Ollis, W. D.; Sutherland, I. O. J. Chem. Soc., Perkin Trans. 1 1976,
913. (d) Tanaka, S.; Hashimoto, K. Yakugaku Zassi 1973, 93,
1003.
3. (a) Tanaka, S.; Hashimoto, K. Japan Patent JP 690908; CA
1972, 59680. (b) Tanaka, S.; Hashimoto, K.; Watanabe, H.;
Sakaguchi, K. Chem. Pharm. Bull. 1973, 21, 1683. (c) Tanaka,
S., Hashimoto, K. US Patent 1974, 3,803,143. (d) Lieb, F.; Eiter,
K. Liebigs Ann. Chem. 1976, 14, 203. (e) Tanaka, S.; Ogata, Y.
Yakugaku Zassi 1972, 93, 1003.
4. For selected publications on SNAr see, for example: (a)
Terrier, F. Nucleophilic Aromatic Displacement: The Role of
the Nitro Group; VCH: New York, 1991, Chapter 1. (b)
Artamkina, G. A.; Egorov, M. P.; Beletskaya, I. P. Chem. Rev.
1982, 82, 427.
5. (a) Boger, D. L.; Yohannes, D.; Zhou, J.; Patane, M. A. J. Am.
Chem. Soc. 1993, 115, 3420. (b) Boger, D. L.; Borzilleri, R. M.;
Nukui, S.; Beresis, R. T. J. Org. Chem. 1997, 62, 4721. (c) Zhu, J.
Syn Lett. 1997, 133 and references cited therein. (d) Gonzalez, G. I.;
Zhu, J. P. J. Org. Chem. 1997, 62, 7544. (e) Evans, D. A.; Barrow,
J. C.; Watson, P. S.; Ratz, A. M.; Dinsmore, C. J.; Evrard, D. A.;
DeVries, K. M.; Ellman, J. A.; Rychnowsky, S. D.; Lacour, J.
J. Am. Chem. Soc. 1997, 119, 3419. (f) Evans, D. A.;
Dinsmore, C. J.; Ratz, A. M.; Evrard, D. A.; Barrow, J. C.
J. Am. Chem. Soc. 1997, 119, 3417. (g) Kiselyov, A. S.;
Eisenberg, S., Luo, Y. Tetrahedron 1998, 54, 10 635. (h)
Ouyang, X.; Kiselyov, A. S. Tetrahedron 1999, 55, 8295.
(i) Goldberg, M.; Smith, L.; Tamayo, N.; Kiselyov, A. S.
Tetrahedron 1999, 55, 13 887. (j) Islas-Gonzalez, G.; Zhu, J.
J. Org. Chem. 1999, 64, 914.
(4-Chlorophenyl)-N-(9-methoxy-6-prop-2-enyl(5H,7H-
benzo[b]benzo[3,4-g]1,5-oxazocin-3-yl))carboxamide
(7e). Light yellow solid. mp 127–130ЊC; 1H NMR (CD2Cl2)
d 7.75 (s, 1H), 7.68 (d, Jˆ8.6 Hz, 2H), 7.43 (d, Jˆ8.6 Hz,
1H), 7.36 (d, Jˆ8.6 Hz, 2H), 7.23 (b, 2H), 6.69 (d,
Jˆ7.6 Hz, 1H), 6.54 (s, 1H), 5.78 (m, 1H), 5.05 (d,
Jˆ9.6 Hz, 2H), 4.5–2.6 (m, 6H), 3.64 (s, 3H); 13C NMR
(CD2Cl2) d 163.5, 155.2, 136.9, 135.4, 133.4, 132.4, 127.9,
127.6, 123.1, 121.9, 120.2, 118.7, 116.2, 112.7, 56.8,
54.5, 521.0, 51.5; HRMS (FAB) calcd for C25H23ClN2O3:
m/zˆ435.1475 (MHϩ), found: 435.1487. Anal. calcd for
C25H23ClN2O3: C, 69.04; H, 5.33; N, 6.44. Found: C,
69.06; H, 5.47; N, 6.37.
N-(8-Prop-2-enyl-7H,9H-benzo[3,4-g]naphtho[2,1-b]1,5-
oxazocin-11-yl)acetamide (7f). White solid. mp 227–
1
230ЊC; H NMR (d6-DMSO) d 9.88 (s, 1H), 8.14 (s, 1H),
7.90 (d, Jˆ8.6 Hz, 2H), 7.57–7.44 (m, 4H), 7.29 (d,
Jˆ2.4 Hz, 2H), 5.89 (m, 1H), 5.18 (d, Jˆ7.6 Hz, 2H),
4.03–2.98 (m, 6H), 2.01 (s, 3H); 13C NMR (d6-DMSO) d
168.3, 136.4, 133.2, 131.0, 130.1, 128.7, 127.0, 125.1,
124.0, 120.2, 118.1, 24.2; HRMS (FAB) calcd for
C23H22N2O2: m/zˆ359.1759 (MHϩ), found: 359.1746.
Anal. calcd for C23H22N2O2: C, 77.07; H, 6.19; N, 7.82.
Found: C, 77.09; H, 6.13; N, 7.75.
N-(9-Bromo-6-benzyl-5H,7H-benzo[b]benzo[3,4-g]1,5-
oxazocin-3-yl)acetamide (7g). White solid. mp 192–
195ЊC; 1H NMR (d6-DMSO) d 9.89 (s, 1H), 7.51–
7.46 (m, 3H), 7.36–7.12 (m, 8H), 7.36–7.24 (m, 7H),
4.54–3.15 (m, 6H), 1.99 (s, 3H); 13C NMR (d6-DMSO)
d 166.6, 137.0, 134.6, 133.5, 132.7, 130.5, 126.8, 126.7,
125.4, 121.0, 120.8, 118.3, 117.6, 55.0, 51.3, 22.3;
HRMS (FAB) calcd for C23H21BrN2O2: m/zˆ437.0865
(MHϩ), found: 437.0862. Anal. calcd for C23H21BrN2O2:
C, 63.17; H, 4.84; N, 6.41. Found: C, 63.17; H, 4.86; N,
6.46.
N-(11-Fluoro-6-benzyl-5H,7H-benzo[b]benzo[3,4-g]1,5-
oxazocin-3-yl)acetamide (7h). White solid. mp 185–187ЊC
1H NMR (d6-DMSO) d 9.93 (s, 1H), 7.52 (d, Jˆ7.6 Hz, 1H),
7.44 (s, 1H), 7.36–7.24 (m, 7H), 7.08 (dd J1ˆ7.6 ,
J2ˆ5.2 Hz, 1H), 6.77 (s, 1H), 4.58–3.25 (m, 6H), 2.03 (s,
3H); 13C NMR (d6-DMSO) d 168.5, 139.0, 136.6,
135.9, 132.4, 128.7, 128.6, 127.8, 127.3, 125.8, 122.9,
120.2, 116.9, 116.8, 57.3, 53.2, 52.6, 24.2; HRMS
(FAB) calcd for C23H21FN2O2: m/zˆ377.1665 (MHϩ),
found: 371.1669. Anal. calcd for C23H21FN2O2: C,
73.39; H, 5.62; N, 7.44. Found: C, 73.33; H, 5.58; N,
7.45.
6. Ouyang, X.; Tamayo, N.; Kiselyov, A. S. Tetrahedron 1999, 55,
2827.
7. (a) Beugelmans, R.; Singh, G. P.; Bois-Choussy, M.; Castanet,
J.; Zhu, J. J. Org. Chem. 1994, 59, 5535. (b) Zhu, J.; Beugelmans,
R.; Bourdet, S.; Chastanet, J.; Roussi, G. J. Org. Chem. 1995, 60,
6389. (c) Boger, D. L.; Borzilleri, R. M.; Nukui, S.; Beresis, R. T.
J. Org. Chem. 1997, 4721.
8. Laib, T.; Zhu, J. Tetrahedron Lett. 1999, 40, 83.
9. (a) Beugelmans, R.; Bois-Choussy, M.; Vergne, C.; Bouillon,
J. P.; Zhu, J. J. Chem. Soc., Chem. Commun. 1996, 1029. (b) Bois-
Choussy, M.; Vergne, C.; Neuville, L.; Beugelmans, R.; Zhu, J.
Tetrahedron Lett. 1997, 38, 5795.
10. Ouyang, X.; Armstrong, R. W.; Murphy, M. J. Org. Chem.
1998, 63, 1027.
References
1. (a) Mueller, A. L.; Moe, S. T. US Patent 1997, 873011. (b)
Vermeulin, N. M. J.; O’Day, C. L.; Webb, H. K.; Burns, M. R.;
Bergstrom, D. E. US Patent 1998, 85538. (c) Averback, P.;
Ghanbari, H.; Beheshti, I.; Morse, D. US Patent 1997, 858343.
(d) Haddjeri, N.; Blier, P.; De Montigny, C. J. Neurosci. 1998,
11. Initially, we attempted the reductive amination reaction of

X. Ouyang et al. / Tetrahedron 56 (2000) 2369–2377
2377
salicylaldehyde 3a (instead of 1), and immobilized b-alanine 8 in
DMF in the presence of 1% of AcOH. However, treatment of the
resulting resin with 20% TFA in CH2Cl2 afforded the expected
amino acid in only a 7% yield. Additionally, attempts to conduct
the second reductive amination reaction with 2-fluoro-5-nitro-
benzaldehyde (1) were unsuccessful under a variety of experi-
mental conditions. However, changing the reductive amination
sequence proved to be beneficial.