N-(Benzoyloxy)amines: An investigation of their thermal stability, synthesis, and incorporation into novel peptide constructs

Article abstract of DOI:10.1016/S0040-4020(03)00635-5

A series of N-benzoyloxyamines were pyrolyzed and their decomposition temperatures correlated well with the amine architecture's ability to stabilize a N-centered radical. A variety of amine substrates were treated with a biphasic mixture of benzoyl peroxide (BPO), CH₂Cl₂ and an aqueous carbonate buffer (at pH 10.5). Primary and secondary amines were successfully N-benzoyloxylated in good yield. Tertiary amines and BPO gave low yields of the corresponding N-oxide and complex product mixtures, presumably via radical decomposition. Electron deficient amines (such as fluorinated aliphatic amines, α-aminoacids, α-aminoesters, and α-aminoamides) were not N-benzoyloxylated under these conditions. Instead, N-benzoylation was observed with the fluorinated amines and the reaction was sensitive to temperature and the pH of the aqueous medium. A one-pot-two-step synthesis of N^α-FMOC-L-Leu-N^β-(benzoyloxy)-β-alanine ethyl ester, a peptide containing both an α- and a novel β-amino acid framework, was also developed.

Full text of DOI:10.1016/S0040-4020(03)00635-5

Tetrahedron 59 (2003) 4315–4325
N-(Benzoyloxy)amines: an investigation of their thermal stability,
synthesis, and incorporation into novel peptide constructs
*
Anne Nemchik, Valentina Badescu and Otto Phanstiel, IV
Department of Chemistry, University of Central Florida, P.O. Box 162366, Orlando, FL 32816-2366, USA
Received 27 September 2002; revised 16 April 2003; accepted 16 April 2003
Abstract—A series of N-benzoyloxyamines were pyrolyzed and their decomposition temperatures correlated well with the amine
architecture’s ability to stabilize a N-centered radical. A variety of amine substrates were treated with a biphasic mixture of benzoyl peroxide
(BPO), CH2Cl2 and an aqueous carbonate buffer (at pH 10.5). Primary and secondary amines were successfully N-benzoyloxylated in good
yield. Tertiary amines and BPO gave low yields of the corresponding N-oxide and complex product mixtures, presumably via radical
decomposition. Electron deficient amines (such as fluorinated aliphatic amines, a-aminoacids, a-aminoesters, and a-aminoamides) were not
N-benzoyloxylated under these conditions. Instead, N-benzoylation was observed with the fluorinated amines and the reaction was sensitive
to temperature and the pH of the aqueous medium. A one-pot-two-step synthesis of N a-FMOC-L-Leu-N b-(benzoyloxy)-b-alanine ethyl ester, a
peptide containing both an a- and a novel b-amino acid framework, was also developed. q 2003 Elsevier Science Ltd. All rights reserved.
1. Introduction
oxidation of primary and secondary amines (Scheme 1).5
Indeed, an aqueous carbonate buffer (pH 10.5)5 was shown
to enhance the selectivity of the BPO-mediated process and
resulted in a significant improvement upon earlier yields.5
The N-oxidation of amines is a chemical reaction that
creates at least one N–O bond. This bond offers access to
many types of nitrogen chemistries and functional groups.
For example, hydroxylamine (NH2OH) and its derivatives
can be converted synthetically into hydroxamates,1 hydrox-
amic acids2 and N-(hydroxy)-containing peptides.3 Early
studies by Milewska used benzoyl peroxide (BPO) to
directly convert a primary amine 1 into a N-(benzoyl-
oxy)amine (RNHOCOPh, 2, Scheme 1) in moderate yield.4
This method was heterogeneous and used BPO, anhydrous
Na2CO3, and CH2Cl2 mixtures. Using a biphasic approach,
an improved method was developed for the selective mono-
In recent years we have demonstrated the utility of the
N-(benzoyloxy)amines in the synthesis of a,b-unsaturated
hydroxamic acids (via acylation)6 and secondary amines
(via organoboranes).7 This work was then extended to
include entry to N-(hydroxy)-containing peptides3 and
siderophores8 (e.g. acinetoferrin6 and nannochelin A9
shown in Fig. 1). In this report the thermal stability of the
N-benzoyloxyamines (e.g. 2) and the synthetic scope of this
BPO method were investigated. In addition, the application
Scheme 1.
Keywords: thermal stability; amine oxidation; N-benzoyloxyamines; hydroxamic acids; b-peptides.
*
Corresponding author. Tel.: þ1-407-8235410; fax: þ1-407-8232252; e-mail: ophansti@mail.ucf.edu
0040–4020/03/$ - see front matter q 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0040-4020(03)00635-5

4316
A. Nemchik et al. / Tetrahedron 59 (2003) 4315–4325
2.1. Thermal studies
A thermal gravimetric analysis–differential thermal
analysis (TGA–DTA) study of a homologous series of
N-benzoyloxyamines is shown in Table 1. All compounds
gave positive DTA curves, which is in agreement with the
thermal decomposition of 6-12 being an exothermic
process. The DTA curve for each compound gave a maxima
at a particular temperature. The temperature associated with
this maxima was noted as the ‘decomposition temperature’
for the analyzed compound and is listed in Table 1. To the
best of our knowledge, this is the first report on the thermal
stabilities of these systems.
Figure 1. Acinetoferrin (R1) and Nannochelin A (R2).
of this method to peptide synthesis was demonstrated. Such
studies are important as they identify both the character-
istics of the amine architecture required for reaction and the
limitations of this technology.
Several trends are evident from Table 1. First, the
decomposition temperatures all fall within the 153–2188C
temperature range. The effect of a-carbon substitution
shows a 108C lower decomposition temperature for the
cyclohexyl derivative 7 (1688C) relative to the n-hexyl-
amine 6 (1788C). However, the trend towards lower
temperature upon further a-carbon substitution did not
hold true for the tert-octyl derivative 8, which had the
highest decomposition temperature (2188C) of the series.
N-Methylation generally resulted in lower decomposition
temperatures (see 9-12 in Table 1). Homolytic N–O bond
cleavage of 6-12 would generate both a benzoyloxy
radical (PhCOOz) and an aminyl radical (RR0Nz). In
general, radicals are electron-deficient species. Therefore,
2. Results and discussion
Previous studies probed the effect of substitution on the
a-carbon of the primary amine architecture in terms of
N-benzoyloxyamine yield. In general, 18 amines containing
bulky R groups (1, Scheme 1) gave higher yields of
N-(benzoyloxy)amines (Table 1).7 Secondary amines were
also readily converted in high yield.10 Having developed an
improved entry to the N-(benzoyloxy)amines (6-12) from
aliphatic amines, their thermal stability and potential as
nitrogen-radical precursors were investigated.
Table 1. Synthesis and TGA–DTA studies of N-benzoyloxyamines 6-12
#
Compound
Yield (%)
74b
Decomposition temperature (8C)
% mass lossa
5
6
178
7
8
9
82b
90b
63b
168
218
163
11
43
5
10
93
153
14
11
12
73b
86
167
158
2
2
a
At decomposition temperature.
From Ref. 7.
b

A. Nemchik et al. / Tetrahedron 59 (2003) 4315–4325
4317
Table 2. Products isolated from the thermal decomposition of 6 and 7 in the absence and presence of benzaldehyde
Starting material
Pyrolysis products (% yield)
with benzaldehyde (% yield)
a secondary aminyl radical (RR0Nz) is usually more stable
than its corresponding primary radical (RHNz). Since 10 and
12 had lower respective decomposition temperatures than 9
and 11, the DTA results suggest a direct relationship
between thermal decomposition temperature and the
stability of the resultant nitrogen radical center.
The TGA data tracked %mass loss as a function of
increasing temperature. Although large %mass losses
occurred at temperatures higher than the listed decompo-
sition temperature, the mass loss measured at the decompo-
sition temperature was typically very low, with the
exception of compound 8. This general observation was
rationalized by the relatively low temperatures involved
(,153–1788C) and the lag time needed to volatilize the
relatively high boiling amine, amide or imide products. The
high % mass loss observed with 8 (43%) is likely related to
its higher temperature of decomposition. The TGA traces of
both 6 and 7 (found in the Supporting Information) also
showed significant mass losses at this elevated temperature
(2188C) and were ,22 and 32%, respectively.
Table 3. Products isolated from thermal decomposition of 9 and 10 in the
absence and presence of benzaldehyde
Starting material
Pyrolysis products
(% yield)
with PhCHO
(% yield)
9a, trace
In order to demonstrate that amine radicals were generated
during the thermal decomposition of these N-benzoyloxy
substrates, several substrates (6, 7, 9 and 10) were
pyrolyzed. In every example (Tables 2 and 3), benzoic
acid was formed as a major product (presumably via
hydrogen atom abstraction by the benzoyloxy radical). The
fate of the amine radical was also monitored. As expected,
the thermal decompositions resulted in very complex
9b, 13%
1
mixtures as shown by the H NMR spectra and TLC of
the crude products. Nevertheless, several products were
isolated and characterized. The pyrolysis results with 6 and
7 are shown in Table 2. Analysis of the product architectures
suggested that N-(benzoyloxy)amines undergo N–O bond
cleavage to give aminyl radical formation.
Several experiments support the intermediacy of aminyl
radicals. As shown in Table 2, compound 6 was pyrolyzed
in refluxing CHCl3 (under N2) in the presence and absence
of excess benzaldehyde (PhCHO), a known radical trap.11a
The N-hexylbenzamide 6a was formed in 30% isolated yield
along with unreacted 6 and benzoic acid. Since benzoic acid
was formed during the pyrolysis of every N-benzoyl-
oxyamine tested, the discussion will focus on the fate of the
amine component. Pyrolysis in refluxing chloroform in the
presence of excess benzaldehyde generated the amide 6a

4318
A. Nemchik et al. / Tetrahedron 59 (2003) 4315–4325
Scheme 2.
(22%) and the symmetrical imide 6b (27%). Even though
the mass recoveries were low, the identification of imide 6b
is compelling evidence for radical intermediates as an ionic
pathway to 6b is unlikely. Likewise, thermolysis of
cyclohexyl derivative 7 gave the N-cyclohexylbenzamide
7a (11%). Pyrolysis of 7 in the presence of benzaldehyde
gave benzamide 7a (27%) and a benzoylated adduct 7b
(7%). In short, these product are readily accessed via radical
processes involving initial aminyl and later amidyl radicals.
transfer process has also been observed in studies with
N-benzoyloxy-substituted peptides.3
It is plausible that hydrogen atom abstraction by the initially
formed aminyl radical (generated from homolysis of 9)
could form benzyl amine in situ. As shown in Scheme 3, this
amine in turn could be N-benzoylated by 9 to form
N-benzylbenzamide 9a. Subsequent H atom abstraction
from 9a would generate a stable amidyl radical, which
could N-benzoylate via radical addition to 9 followed
by formation of imide 9b and loss of a stable nitroxyl
radical. The role of 9 as a benzoyl group donor is likely
replaced by benzaldehyde, when benzaldehyde is
present and would require a loss of an H atom for 9b
formation.
The N-(benzoyloxy)-t-octylamine 8 had the highest
decomposition temperature by DTA and the slowest
decomposition rate in solution. After pyrolysis for 42 h in
a sealed vessel at 1268C in the presence of CH3CN, 42% of
the unreacted 8 was recovered. Benzoic acid (50% yield)
was the only pure product isolated from the crude mixture.
Not surprisingly, the decomposition of 8 in the presence of
benzaldehyde under refluxing chloroform (e.g. 618C) also
had a very slow rate. Based upon the TLC and 1H NMR of
the crude mixture, the bulk of 8 remained after 72 h at
reflux. Therefore, the experiment was repeated at higher
temperature. Compound 8 and benzaldehyde (2 equiv.)
were dissolved in CDCl3 and placed in a sealed reaction
vessel and heated to 1708C. CAUTION: a violent explosion
took place after 9 h. The products were not determined. Due
to safety concerns, the reaction was not repeated and further
solution studies with 8 were abandoned.
Both imide 6b and 9b are formed in higher yields in the
presence of benzaldehyde as there are more benzoyl sources
present during these reactions. This finding is supported by
the fact that the product distribution is also time-dependent.
A time-course pyrolysis study with 6 in 2 equiv. of
benzaldehyde revealed that the ratio of amide 6a to imide
6b went from 1:10 at short reaction times (when copious
amounts of benzaldehyde still remained) to approximately
1:1 upon completion of the reaction (with only trace levels
of benzaldehyde remaining).
Interestingly, the N-methylbenzylamine adduct 10 gave
dealkylated products 9a and 10a (Table 3). Low levels of
benzaldehyde were also generated during the course of this
‘benzaldehyde-free’ pyrolysis (1H NMR singlet: ,10 ppm).
These dealkylated materials (9a and 10a) are likely formed
via imine intermediates, which cleave to form the free
benzylamine or methylamine. These in turn may be
N-benzoylated by 10 (or newly generated benzaldehyde)
to form the observed adducts. Although speculative, this
proposed dealkylation mechanism is supported by the
observation of benzaldehyde (an imine cleavage product)
during the ‘benzaldehyde-free’ pyrolysis reaction and the
known disproportionation of aminyl radicals to amines and
imines.11b
As shown in Table 3, pyrolysis of N-benzyl derivative 9 in
refluxing CHCl3 gave both amide 9a and imide 9b. The fact
that 9b was isolated in the absence of benzaldehyde
suggested that a N-benzoyloxyamine could also act as a
benzoyl donor. A control experiment using phenethylamine
and N-benzoyloxyamine 11 revealed efficient O- to
N-benzoyl transfer to phenethylamine to generate N-(2-
phenylethyl)benzamide 13 in good yield (Scheme 2). This
Having a better understanding of the thermal properties of
6-12, we explored the scope and limitations of the synthetic
method to access the N-benzoyloxyamines in general.
2.2. Synthetic scope and limitations
Having already probed the role of steric factors in the BPO-
mediated process,7 we were interested in what electronic
perturbations were tolerated by this reaction. How was the
yield influenced by substituents adjacent to the amine
center? For example, would an electron withdrawing group
(EWG) on the a- or b-carbon adjacent to the amine center
reduce the yield of 2? In order to test this premise, several
diamines, fluorinated amines, a-aminoacids and a- and
Scheme 3.

A. Nemchik et al. / Tetrahedron 59 (2003) 4315–4325
4319
Scheme 4.
b-aminoacid derivatives were chosen as possible substrates
for N-oxidation (i.e. N-benzoyloxylation) with BPO.
Previous work with primary and secondary amines resulted
in high yields of the respective N-(benzoyloxy)amines.7,10
Initially, the optimized process was expected to allow for
selective oxidation of a primary amine in the presence of a
tertiary amine, e.g. N,N-dimethylpropanediamine (Fig. 2). It
was presumed that tertiary amines would be inert and would
not interfere with distal primary amine oxidation. To test
this premise, N,N-dimethyl-propanediamine was reacted
under the standard conditions7 with BPO. After consump-
tion of the starting amine, multiple products were visible by
TLC and column chromatography provided complex
mixtures. This suggested that the tertiary amine adducts
were unstable. A simpler model was chosen for further
investigation.
Beyond their role as models of electronically-perturbed
amines, novel aminoacids have obvious synthetic value.
Moreover, there are only limited entries to chiral
N-(hydroxy)aminoacid derivatives via direct amine oxi-
dation. Prior work by Danishefsky showed that aminoacid
derivatives (Method 1, Scheme 4) could be directly oxidized
to the corresponding N-(hydroxyl)amines with 2,2-
dimethyldioxirane.12 An alternative approach by
Chimiak,13 reacted a-aminoesters (Method 2, Scheme 4)
with p-anisaldehyde (14) to give the corresponding imine
15. The imine 15 was then oxidized with monoperphthalic
acid (MPP) to the corresponding oxaziridine 16, which gave
the N-(hydroxy)aminoacid 17 during heating with hydro-
chloric acid. The yields were typically 40%. The limitations
of these earlier methods include low yields and production
of the N-(hydroxyl)amine component in the uncapped, free
N-(hydroxyl)amine form. Subsequent acylation of this free
hydroxylamine form may be problematic due to selectivity
issues (involving N and O). In contrast, the N-(benzoyl-
oxy)amines are ‘O-capped’ systems, which can be
selectively N-acylated with acid chlorides.3,5
Tributylamine was selected due to its symmetry and
relatively high molecular weight for ease in isolation of
reactant products. As shown in Figure 2, tributylamine was
reacted with BPO under the standard conditions. Multiple
products were observed by TLC and the crude decomposed
upon standing overnight. The only product isolated in an
appreciable quantity was tributylamine N-oxide 18 (9%).
This suggested that the N-(benzoyloxy)amine 19 was
generated in situ and converted to the more stable N-oxide
or further decomposed. Separate synthesis of N-(benzoyl-
oxy)-N,N-dibutylamine 20 and 1H NMR comparison
suggested that none of the major products were generated
by N-dealkylation of 19 (a pathway demonstrated in our
earlier pyrolysis studies with 10). Indeed, an earlier report
by Buckley was consistent with our findings and noted that
the tertiary N-(benzoyloxy)amines decomposed via
radicals.10,14,15 Therefore, one potential limitation in this
BPO-process is the presence of tertiary amine centers,
which give rise to radical chemistry at room temperature.
With primary, secondary, and tertiary amines characterized,
the next amine series to be evaluated were the electron-poor
fluorinated amines. In general, fluorinated amines typically
contain fluorine atoms in the b-position (or even more distal
Scheme 5. Reagents: (a) BPO, CH2Cl2, aq. buffer pH 10.5, rt; (b) BPO, aq.
buffer pH 10.5, CHCl3, reflux for 48 h.

4320
A. Nemchik et al. / Tetrahedron 59 (2003) 4315–4325
Figure 2. Tertiary amine oxidation products.
to N) due to the instability of a-fluorinated amines. As
shown in Scheme 5, the respective amines, 1H,1H-
heptafluorobutylamine 21 and 2,2,2-trifluoroethylamine
22, were treated with BPO in a biphasic solution (CH2Cl2/
aqueous buffer pH¼10.5) at room temperature.10 1H,1H-
heptafluorobutylamine did not react after 72 h under these
conditions. The reaction was repeated in refluxing CHCl3
and N-(1H,1H-heptafluorobutyl)benzamide 23 was isolated
in 40% yield. In contrast, the reaction of 22 with BPO
(biphasic conditions) at 258C gave N-(2,2,2-trifluoroethyl)-
benzamide 24 (Scheme 5) in 67% yield. Neither fluorinated
alkylamine provided the desired N-(benzoyloxy) adduct.
Instead, amide products were isolated and assigned by their
respective IR and 1H NMR spectra. For example, N-(2,2,2-
trifluoroethyl)benzamide showed a characteristic band at
1678 cm21 in the IR spectrum; whereas the N-benzoyloxy
‘ester’ carbonyl usually appears at 1720 cm21. The 1H
NMR spectrum of N-(2,2,2-trifluoroethyl)benzamide
exhibited a multiplet at d 4.00 ppm and a broad singlet at
d 6.80 ppm, which were indicative of the amide CH2 and
NH, respectively. Based upon these results, the acylation
pathway is clearly favored for the fluorinated alkylamines,
whereas the N-oxidation pathway is preferred with N-alkyl-
amines under the same conditions. The putative mechanistic
pathways are illustrated in Figure 3.
Previous work in our laboratories showed that the
N-oxidation pathway to form 2 in Figure 3 was pH-
dependent and favored at pH 10.5 with aliphatic amines.5,7
Perhaps not surprisingly, this optimum pH value is in the
same range as the pKa for the ammonium ions of
alkylamines.5 Moreover, the aliphatic 18 amines also
generated their corresponding amides (3, Fig. 3) at higher
pH (11–12), in lieu of the N-benzoyloxyamine product 2.
Fluorinated alkylamines are weaker bases than alkylamines.
For example, the ammonium salt of 2,2,2-trifluoroethyl-
amine has a pKa¼5.7,16 whereas the ethylamine counterpart
Figure 3. Possible mechanistic pathways for the reaction of fluorinated alkylamines with BPO.

A. Nemchik et al. / Tetrahedron 59 (2003) 4315–4325
4321
Scheme 6.
Scheme 7.
has a pKa of 10.81.17 Based upon the earlier pH dependence
of the BPO method, it seemed likely that lowering the pH of
the water phase closer to the pKa of the fluorinated amines
might influence the mechanism of this process and favor the
N-oxidation pathway.
were not oxidized by BPO illustrated another limitation of
this method. However, this finding also suggested that
selective N-oxidation could be attained with substrates
containing both electron rich and electron poor amine
centers (e.g. lysine or ornithine).
The reaction of 2,2,2-trifluoroethylamine with BPO under
biphasic conditions at different pH values was performed.
This was accomplished by using various aqueous buffer
solutions. At pH¼7.3 and 8.3, no reaction occurred after
stirring overnight. At pH 9.3 traces of N-(2,2,2-trifluoro-
ethyl)benzamide 24 were detected. At pH 10.5 the amide 24
was isolated in 67% yield. This study showed that having a
lower pH value of the aqueous solution (closer with the pKa
of the ammonium ion of 2,2,2-trifluoroethylamine) did not
facilitate the N-oxidation pathway. This is in direct contrast
to the pH dependency of aliphatic (non-fluorinated) amine
systems.
Although the a-aminoacid substrates were unreactive, their
b-derivatives do allow access to N-hydroxy-containing
peptides. Prior work demonstrated the clean acylation of
N-(benzoyloxy)amines with acid chlorides.3 Moreover,
Carpino’s FMOC-protected aminoacid chloride19 was
shown to acylate N-(benzoyloxy)amines without racemiza-
tion.3 Recognizing that these two technologies could be
combined to form b-aminoacid containing peptides, a
Attempts to apply this biphasic technology to the oxidation
of a-aminoacid derivatives were not successful (Scheme 6).
The respective systems of L-phenylalanine, L-phenylalanine
benzyl ester, glycinamide and L-phenylalanine-L-leucine
amide were each treated with BPO, following the general
procedure.7 No reaction occurred after 48 h in each case.
These findings were rationalized by relating the amine’s
propensity for oxidation to the electron richness and
nucleophilicityoftheaminecenter.18 Onepossibleexplanation
is that the presence of the carbonyl group on the a-carbon
inductively withdraws electron densityfromthe nitrogen atom,
makingit less available fora nucleophilic attack onthe peroxy-
oxygen of BPO (see oxidation pathway—Fig. 3). This
resistance to N-oxidation is similar to the fluorinated amines,
which are also electron-poor amine systems.
To further illustrate that the oxidation process is strongly
dependent upon the electron density of the starting amine,
b-alanine ethyl ester 25 was treated with BPO under
biphasic conditions at room temperature.7,10 In this case,
N-(benzoyloxy)-b-alanine ethyl ester 26 was isolated in
76% (Scheme 7).3 Therefore, simple insertion of an
additional methylene unit was sufficient to regain sufficient
amine nucleophilicity. The fact that electron poor amines
Scheme 8. Reagents: (a) SOCl2, reflux, followed by concentration; (b) BPO,
CH2Cl2, aq. carbonate buffer, pH 10.5; (c) CH2Cl2, aq. carbonate buffer, pH
10.5.

4322
A. Nemchik et al. / Tetrahedron 59 (2003) 4315–4325
convergent approach involving a one-pot, two-step syn-
thesis was developed. As shown in Scheme 8, FMOC-L-
leucine 27 was converted to its acid chloride 28 in refluxing
thionyl chloride, followed by concentration under high
vacuum.3,19 Conversion was monitored by the IR carbonyl
resonance (27: 1717 cm21, 28: 1789 cm21). The crude acid
chloride acylated 26 to give the final hybrid peptide 29 in
85% isolated yield, for an overall yield of 61% from the
starting amine, 25.
(unless otherwise indicated), in an inert atmosphere.
Benzaldehyde was distilled prior to use. Mass spectra
were obtained typically using the fast atom bombardment
(FAB) technique in magnetic scan mode. Masses from the
matrix (e.g. meta-nitrobenzylalcohol) were used as
reference points.
4.2. TGA–DTA studies
Respective samples of 6-12 were placed into a sample
holder and TGA–DTA analysis were performed with a STD
2960 instrument, where the flow rate of the purge N2 gas is
96 mL/min. A controlled heating rate (at 108C/min) up to
3008C was used.
3. Conclusions
By investigating the thermal behavior of a series of
N-(benzoyloxy)amines, several trends became apparent.
First, N-(benzoyloxy)amines derived from 28 amines (such
as 10) undergo both N–O and N–C bond cleavage and give
rise to very complex product mixtures. Second, N-methyl-
ation of a N-benzoyloxyamine resulted in a lower
decomposition temperature by TGA–DTA analysis (e.g. 9
and 10, 11 and 12). This is consistent with a more facile
homolytic process giving rise to the more stable 28 aminyl
radical. Third, both aminyl and amidyl radicals are
generated during the thermal decomposition of N-(benzoyl-
oxy)amines. In terms of synthetic limitations, BPO-
mediated amine oxidation was problematic for both tertiary
and electron deficient amines. Tertiary amines were
oxidized, but rapidly decomposed. Both the fluoroamine
systems and the a-aminoacid derivatives did not undergo
N-oxidation using the biphasic conditions employed
successfully with 18 and 28 N-alkylamines.7 In short, the
electron density of the amine nitrogen was shown to
strongly influence the BPO-mediated oxidation process.
4.3. General procedure for amine oxidation
A solution of benzoyl peroxide (BPO, 1 equiv.) in CH2Cl2
(5 mL/mmol BPO) was added quickly to a mixture of the
amine (1 equiv.) and a pH 10.5 aqueous buffer solution
(5 mL/mmol amine) at room temperature. [Note: using
1–1.1 equiv. of BPO facilitated the isolation of the
intermediate N-(benzoyloxy)amines, which often had Rf
values close to that of BPO.]7 TLC was used to monitor the
consumption of the starting amine, using both UV and
iodine vapor staining for visualization as needed. After the
reaction was complete, the aqueous layer was extracted
three times with fresh CH2Cl2. The organic layers were
combined, dried over anhydrous Na2SO4, filtered, and
concentrated. The crude residue was subjected to flash
column chromatography for purification of the products. In
this manner, compounds 6-9 and 11 were prepared by
published methods.7
4.4. General pyrolysis procedure
Lastly, a new method was developed to access the
N-(hydroxy)-b-aminoacids in an O-acylated form con-
ducive for subsequent N-acylation. This approach should
allow for the synthesis of a new class of N-hydroxyamide-
containing peptides with unique functionality. In con-
clusion, the diverse chemistries available to the N-benzoyl-
oxyamines make them attractive systems for future study.
Compounds 6 and 9 were initially pyrolyzed in CHCl3 with
and without benzaldehyde. For pyrolyses with benzal-
dehyde, benzaldehyde (0.181 g, 1.8 mmol) and the
N-(benzoyloxy)amine (0.9 mmol) were dissolved in
CHCl3 (7 mL). In experiments without added benzaldehyde,
the N-(benzoyloxy)amine (0.9 mmol) was dissolved in
CHCl3 (7 mL). The respective solution was refluxed under
a N2 atmosphere. Compounds 7, 8, and 10 were pyrolyzed in
CH3CN at 1268C due to the slow rates observed in CHCl3.
The pyrolyses in CH3CN were also conducted in the
presence and absence of benzaldehyde using the molar
ratios listed above in CHCl3. The conversion of the
N-(benzoyloxy)amine was monitored by TLC and 1H
NMR. Multiple new spots were visible by TLC. The
mixture was concentrated under high vacuum and the
4. Experimental
4.1. Materials and methods
All reagents were used without purification unless otherwise
1
noted. H and 13C NMR spectra were recorded at 300 and
75 MHz, respectively. The fluorinated amine starting
materials were generously provided by SynQuest Labora-
tories, Inc. of Gainesville, FL. The mass spectra data were
acquired at the University of Florida by Dr David Powell.
The pH 10.5 buffer solution was prepared by adding 1.5N
aqueous NaOH to aqueous NaHCO3. The TLC and column
chromatography solvents are expressed in vol%. Ammonia-
containing solutions were prepared by measuring out
concentrated aqueous NH4OH in the listed vol%. Silica
gel (32–63 mm) was used for flash chromatography.
Storage recommendations for the oxidized amines are 08C
under argon (months) or as an HCl salt under the same
conditions (years). All the thermal decompositions were
performed in a sealed glass reactor in dry acetonitrile
1
residue was dissolved in CDCl3. An H NMR spectrum of
the crude mixture was used to quantify the reaction mixture
via an internal standard, C2H2Cl4 (47.0 mg, 0.28 mmol).
The standard was added to the crude and gave a quantifiable
1H NMR singlet at 5.90 ppm. The crude was then subjected
to flash column chromatography (often eluting with 100%
CHCl3). Several products were isolated (typically the amide
1
and imide) along with benzoic acid. Using the above H
NMR integration information, the respective yields were
determined. The quantified NMR signals were well-
resolved from other pyrolysis products and easily identified
by their chemical shifts.

A. Nemchik et al. / Tetrahedron 59 (2003) 4315–4325
4323
4.4.1. N-Hexylbenzamide 6a.20 Rf¼0.3 in 100% CHCl3; 1H
NMR (CDCl3): d 7.77 (m, 2H, aromatic), 7.41 (m, 3H,
aromatic), 6.30 (br s, 1H), 3.4 (q, 2H), 1.60 (m, 2H), 1.30
(m, 6H), 0.90 (t, 3H); 13C NMR (CDCl3): d 167.5, 134.8,
131.1, 128.4, 126.8, 40.1, 31.5, 29.5, 27.6, 22.5, 14.0.
added quickly to a mixture of methylphenethylamine
(0.543 g, 4 mmol) in 30 mL of an aqueous buffer solution
(pH¼10.5) at room temperature. The disappearance of the
starting material was monitored by TLC (2% NH4OH/
MeOH, Rf¼0.35). After the reaction was complete, the
organic layer was separated and the water layer was
extracted twice with 25 mL of CH2Cl2. The organic layers
were combined, dried over anhydrous Na2SO4, filtered, and
concentrated to give the crude product. The product mixture
was subjected to flash column chromatography, eluting with
20% EtOAc/hexane, to isolate the N-(benzoyloxy)-methyl-
phenethylamine 12 (0.88 g, 86%).
4.4.2. N-Hexyldibenzamide 6b. Rf¼0.40 in 100% CHCl3;
1H NMR (CDCl3): d 7.41 (m, 4H, aromatic), 7.20 (m, 6H,
aromatic), 4.00 (t, 2H), 1.80 (m, 2H), 1.40 (m, 2H), 1.36 (m,
4H), 0.90 (t, 3H); 13C NMR: d 175.4, 137.9, 132.7, 129.7,
129.2, 48.5, 32.5, 30.0, 27.8, 23.5, 15.0.
4.4.3. N-Cyclohexylbenzamide 7a.20 Rf¼0.35 in 100%
1
CHCl3; H NMR (CDCl3): d 7.75 (m, 2H, aromatic), 7.45
(m, 3H, aromatic), 6.0 (br, s, 1H), 4.0 (m, 1H), 1.62 (m,
10H).
Compound 12. Rf¼0.47 in 20% EtOAc/hexane. IR (CDCl3):
3154, 2968, 1732, 1602, 1452, 1261, 1081, 1061,
1
1025 cm21; H NMR (CDCl3): d 8.01 (m, 2H, aromatic),
7.35 (m, 8H, aromatic), 3.22 (t, 2H), 2.90 (m, 5H); 13C NMR
(CDCl3) d 165.1, 139.2, 133.0, 129.4, 129.1, 128.6, 128.4,
128.3, 126.2, 62.6, 47.0, 33.7. Anal/calcd for C16H17NO2:
C,75.27; H, 6.71; N, 5.48. Found: C, 75.23; H, 6.68; N, 5.44.
4.4.4.
N-(Benzoyloxy)-N-cyclohexylbenzamide
7b.
1
Rf¼0.42 in 100% CHCl3; H NMR (CDCl3): d 8.0 (m,
2H, aromatic), 7.59 (m, 2H, aromatic), 7.35 (m, 6H,
aromatic) 4.35 (m, 1H), 1.65 (m, 10H). High resolution
mass spectrum: theory for C20H21NO3 (Mþ1)¼324.1599,
found Mþ1¼324.1556.
4.4.10. Reaction of phenethylamine with 11 to give 13.
Phenethylamine (0.050 g, 0.413 mmol) and N-(benzoyl-
oxy)phenethylamine7 (0.099 g, 0.413 mmol) were dissolved
in CH3CN (10 mL). The solution was stirred under nitrogen.
No reaction occurred after 24 h at room temperature by
TLC. The solution was heated to reflux (818C). The starting
material N-(benzoyloxy)phenethylamine was consumed
after 6 h as shown by TLC (Rf¼0.45 in 100% CHCl3).
The crude product was subjected to prep TLC (100%
CHCl3) to give N-(phenethyl)benzamide (66 mg,
0.293 mmol) in 71% yield. N-(Phenethyl)-benzamide, 13:
1H NMR (CDCl3): d 7.60–7.10 (m, 10H), 6.08 (br s, 1H),
3.61 (q, 2H), 2.85 (t, 2H).
4.4.5. N-Benzylbenzamide 9a.20 Rf¼0.23 in 20% EtOA-
c/hexane; 1H NMR (CDCl3) d 7.80 (m, 2H, aromatic), 7.40
(m, 8H, aromatic), 6.42 (br, s, 1H), 4.62 (d, 2H). High
resolution mass spectrum: theory for C14H13NO
(Mþ1)¼212.1075, found Mþ1¼212.1021.
4.4.6. N-Benzyldibenzamide 9b. 1H NMR (CDCl3): d 7.80
(m, 4H, aromatic), 7.41 (m, 6H, aromatic), 7.20 (m, 5H,
aromatic), 5.22 (s, 2H).
4.4.7. N-Methylbenzamide 10a.20 Rf¼0.3 in 20% EtOA-
c/hexane; 1H NMR (CDCl3) d 7.70 (m, 2H, aromatic), 7.40
(m, 3H, aromatic), 6.1 (br, s, 1H), 2.95 (d, 3H).
4.4.11. Tributylamine N-oxide 18.20 Tributylamine
(0.45 g, 2.4 mmol) was reacted with BPO (0.65 g,
2.7 mmol) using the general procedure. After stirring
overnight, the reaction was worked up. The crude was
subjected to flash chromatography using the following
gradient: 5% EtOH/CHCl3 to 60% EtOH/CHCl3 and then
3% NH4OH/MeOH. A complex mixture of products was
recovered. The N-oxide was isolated in 9% yield (0.05 g).
TLC with 3% NH4OH/MeOH (Rf¼0.45) detected the
N-oxide by iodine vapor.
4.4.8. N-(Benzoyloxy)benzylmethylamine 10. A solution
of BPO (0.96 g, 4 mmol) in CH2Cl2 (30 mL) was added
quickly to a mixture of benzylmethylamine (0.484 g,
4 mmol) in 30 mL of an aqueous buffer solution
(pH¼10.5) at room temperature. The disappearance of the
starting material was monitored by TLC (2% NH4OH/
MeOH, Rf¼0.30). After the reaction was complete, the
organic layer was separated and the water layer was
extracted twice with 25 mL of CH2Cl2. The organic layers
were combined, dried over anhydrous Na2SO4, filtered, and
concentrated to give the crude product. The product mixture
was subjected to flash column chromatography, eluting with
20% EtOAc/hexane, to isolate the N-(benzoyloxy)benzyl-
methylamine 10 (0.90 g, 93%).
Compound 18. 1H NMR (CDCl3): d 3.34 (m, 6H), 1.72 (m,
6H), 1.36 (m, 6H), 0.93 (t, 9H); high resolution mass
spectrum (FAB) theory for (C12H28NO) Mþ1¼202.2171,
found Mþ1¼202.2149. Note: tributylamine gave the
following: 1H NMR (CDCl3): d 2.38 (t, 6H), 1.40 (m,
6H), 1.28 (m, 6H), 0.90 (t, 9H).
Compound 10. Rf¼0.37 in 20% EtOAc/hexane. IR (CDCl3)
4.4.12. N-(Benzoyloxy)-dibutylamine 20. Dibutylamine
(0.21 g, 1.6 mmol) was reacted with BPO (0.40 g,
1.6 mmol) using the general procedure. Disappearance of
the starting amine was monitored by TLC (CHCl3). Since
the amine was not fully consumed after a day, an additional
amount of BPO (0.17 g, 0.71 mmol) was added. Upon
completion, the reaction was worked up and the crude
product was subjected to flash chromatography (30%
hexane/CHCl3 to 10% hexane/CHCl3). The oxidized
amine (Rf¼0.24 in 10% hexane/CHCl3) was isolated in
3155, 2900, 1735, 1602, 1452, 1263, 1093, 1059,
1
1025 cm21; H NMR (CDCl3) d 7.90 (m, 2H, aromatic),
7.35 (m, 8H, aromatic), 4.15 (s, 2H), 2.9 (s, 3H); 13C NMR
(CDCl3) d 164.9, 135.6, 132.9, 129.4, 129.3, 128.3, 128.2,
127.7, 65.0, 46.0. Note: there was significant overlap in the
129–128 ppm range.
4.4.9. N-(Benzoyloxy)methylphenethylamine 12. A sol-
ution of BPO (0.96 g, 4 mmol) in CH2Cl2 (30 mL) was

4324
A. Nemchik et al. / Tetrahedron 59 (2003) 4315–4325
94% yield (0.38 g). 1H NMR (CDCl3) d 8.03 (d, 2H), 7.58 (t,
1H), 7.44 (t, 2H), 2.98 (t, 4H), 1.59 (m, 4H), 1.37 (m, 4H),
0.92 (t, 6H). Anal. calcd for C15H23O2N: C, 72.25; H, 9.30;
N, 5.62. Found: C, 71.95; H, 9.22; N, 5.56.
aromatic), 7.45 (m, 3H, aromatic), 6.40 (broad s, 1H,
NH), 4.22 (m, 2H, CF2CH2N).
4.4.15. N-(2,2,2-Trifluoroethyl)benzamide 24.20
A
solution of BPO (0.960 g, 4 mmol, dissolved in 20 mL
CH2Cl2) was added quickly at room temperature to a
vigorously stirred mixture of 2,2,2-trifluoroethylamine
hydrochloric acid salt (0.484 g, 4 mmol) and a pH 10.5
carbonate buffer solution (20 mL). The 2,2,2-trifluoroethyl-
amine was consumed after 15 h as shown by TLC (Rf¼0.21
in 2% NH4OH/MeOH). The organic layer was separated.
The aqueous layer was extracted twice with CH2Cl2. The
organic layers were combined, dried over anhydrous
Na2SO4, filtered, and concentrated to give the crude
product. The residue was subjected to flash column
chromatography, eluting with 100% methylene chloride to
give 2,2,2-trifluoroethylbenzamide 24 as a white crystalline
solid (0.55 g, 67%).
4.4.13. Reaction of 22 and BPO using different aqueous
buffer solutions. A solution of BPO (0.960 g, 4 mmol,
dissolved in 20 mL CH2Cl2) was added quickly at room
temperature to a vigorously stirred mixture of 2,2,2-
trifluoroethylamine hydrochloric acid salt (0.484 g,
4 mmol) and 20 mL of either a pH 7.3 buffer solution
(50 mL of 0.1 M KH2PO4 and 41.1 mL of 0.1 M NaOH
mixed together and diluted with deionized water to a final
volume of 100 mL) or a pH 8.3 buffer solution (50 mL of
0.025 M borax and 17.7 mL of 0.1 M HCl mixed together
and diluted with deionized water to a final volume of
100 mL) or a pH 9.3 buffer solution (50 mL of 0.025 M
borax and 3.6 mL of 0.1 M NaOH mixed together and
diluted with deionized water to a final volume of 100 mL).
The organic layer was checked by TLC. After 40 h there
was no evidence of product formation at pH 7.3 or 8.3 (the
only spot visible by UV and iodine detection was BPO,
Rf¼0.6 in 100% CHCl3). After 24 h a trace of 2,2,2-
trifluoroethylbenzamide 24 was detected in the pH 9.3
reaction. Overall, there was no evidence of N-(benzoyloxy)-
2,2,2-trifluoroethylamine formation.
Compound 24. Rf¼0.23 in 100% methylene chloride. IR
(CDCl3) 3460, 1679, 1523, 1490, 1391, 1260, 914,
1
748 cm21; H NMR (CDCl3): d 7.69 (m, 2H, aromatic),
7.35 (m, 3H, aromatic), 6.89 (s, 1H, NH), 4.00 (m, 2H,
CF3CH2N).
4.4.16. N-(Benzoyloxy)-b-alanine ethyl ester 26. A
solution of BPO (1.21 g, 5 mmol) dissolved in 30 mL
CH2Cl2 was added quickly at room temperature to a
vigorously stirred mixture of b-alanine ethyl ester HCl salt
(0.768 g, 5 mmol) and 30 mL of a pH 10.5 carbonate buffer
solution. The starting material was consumed after 10 h as
shown by TLC (Rf¼0.23 in 2% NH4OH/MeOH). The
organic layer was separated. The aqueous layer was
extracted twice with CH2Cl2. The organic layers were
combined, dried over anhydrous Na2SO4, filtered, and
concentrated to give the crude product. The residue was
subjected to flash column chromatography, eluting with
100% methylene chloride to give N-(benzoyloxy)-b-alanine
ethyl ester 26 as a colorless oil (0.85 g, 76%). Rf¼0.35 in
100% methylene chloride. IR (CDCl3) 3240, 2985, 1725,
4.4.14. N-(1H,1H-Heptafluorobutyl)benzamide 23. A
solution of BPO (0.484 g, 2 mmol, dissolved in 10 mL
CH2Cl2) was added quickly at room temperature to a
vigorously stirred mixture of 1H,1H-heptafluorobutylamine
(0.398 g, 2 mmol) and 10 mL of a pH 10.5 carbonate buffer
solution. Since it was difficult to monitor the consumption of
the 1H,1H-heptafluorobutylamine by TLC (it is not visible
by iodine or UV detection), the organic layer was checked
by TLC for the formation of the expected UV active
products. After 58 h at room temperature, BPO was the
only visible spot on the TLC plate (Rf¼0.6 in 100%
chloroform).
1
The reaction was run again in deuterated chloroform. BPO
(0.242 g, 1 mmol) was dissolved in CDCl3 (7 mL) and
1H,1H-heptafluorobutylamine (0.199 g, 1 mmol) was added
to the mixture. A 1H NMR spectrum was taken at time zero.
1H,1H-Heptafluorobutylamine gave a characteristic triplet
at d 3.50 ppm (CF2CH2N). After 24 h at room temperature,
the TLC showed only BPO (Rf¼0.6 in 100% chloroform).
There was no evidence of product formation by TLC or 1H
NMR. Therefore, the reaction mixture was heated to reflux
(CDCl3 bp¼60.98C, 1H,1H-heptafluorobutylamine
bp¼728C). After another 24 h under reflux, TLC showed
only one new UV and iodine active spot (Rf¼0.32 in 20%
EtOAc/hexane). The 1H NMR spectrum of the crude
indicated that the starting material 1H,1H-heptafluorobutyl-
amine was not totally consumed and two multiplets were
observed at d 3.85 and 4.20 ppm. The crude product was
subjected to flash chromatography. Only N-(1H,1H-hepta-
fluorobutyl)-benzamide 23 (Rf¼0.32 in 20% EtOAc/
hexane) was isolated in a 40% yield (based on approximate
70% conversion of the starting material). The compound,
which gave rise to the other multiplet at d 3.85 ppm, was not
isolated. 23: IR (CDCl3): 3461, 1682, 1520, 1489, 1262,
1602, 1452, 1270, 1194, 1026, 926, 720 cm21; H NMR
(CDCl3): d 8.03 (m, 2H, aromatic), 7.45 (m, 3H, aromatic),
4.18 (q, 2H, OCH2), 3.45 (t, 2H, NHCH2), 2.65 (t, 2H,
CH2CO), 1.27 (t, 3H, CH3). Anal. calcd for C12H15O4N: C,
60.75; H, 6.37; N, 5.90. Found: C, 60.85; H, 6.34; N, 5.90.
13C NMR (CDCl3): d 171.74, 166.56, 133.36, 129.31,
128.45, 128.17, 60.76, 47.78, 32.15, 14.09. High resolution
mass spectrum (FAB) theory for (C12H15O4N)
Mþ1¼238.1000; found: Mþ1¼238.1080.
4.4.17. S-3-{Benzoyloxy-[2-(9H-fluoren-9-ylmethoxycar-
bonylamino)-4-methyl-pentanoyl]-amino}-propionic
acid ethyl ester, 29. A solution of BPO (0.79 g, 3.3 mmol)
in CH2Cl2 (17 mL) was quickly added to a stirring solution
of b-alanine ethyl ester hydrochloride salt (0.50 g,
3.2 mmol) in pH 10.5 buffer (17 mL). The consumption of
the b-aminoester was monitored by TLC (2% NH4OH/
MeOH, Rf¼0.23). The N-oxidized b-aminoester 26 was
generated in 72% yield as evidenced by 1H NMR integration
of the crude.
N-FMOC-L-Leucine 27 (1.27 g, 3.6 mmol) was refluxed for
4 h in an excess of freshly distilled thionyl chloride (4.5 mL,
1230, 918 cm21
;
1H NMR (CDCl3) d 7.79 (m, 2H,

A. Nemchik et al. / Tetrahedron 59 (2003) 4315–4325
4. Milewska, M. J.; Chimiak, A. Synthesis 1990, 233–234.
4325
59 mmol) dissolved in CH2Cl2 (25 mL). The reaction was
monitored by the conversion of the carbonyl IR stretch from
1717 to 1789 cm21. The acid chloride 28 was concentrated
down to provide a residue. Benzene was added and the
solution was re-concentrated under vacuum in order to
remove the remaining SOCl2. The crude was then
transferred to the oxidized alanine mixture 26 with a
minimum of CH2Cl2. The reaction was stirred at room
temperature for 2 days. The aqueous phase was extracted
three times with fresh CH2Cl2. The organic layers were
combined, dried over anhydrous Na2SO4, filtered, and
concentrated to give the crude product. The residue was
subjected to flash chromatography (1% EtOH/1% EtOAc/
CHCl3, Rf¼0.34). The peptide 29 was isolated in 85% yield.
5. Wang, Q. X.; King, J.; Phanstiel, O., IV. J. Org. Chem. 1997,
62, 8104–8108.
6. Wang, Q. X.; Phanstiel, O., IV. J. Org. Chem. 1998, 63,
1491–1495.
7. Phanstiel, O., IV; Wang, Q. X.; Powell, D. H.; Ospina, M. P.;
Leeson, B. A. J. Org. Chem. 1999, 64, 803–806.
8. Guo, H.; Naser, S. A.; Ghobrial, G.; Phanstiel, O., IV. J. Med.
Chem. 2002, 45, 2056–2063.
9. Bergeron, R. J.; Phanstiel, O., IV. J. Org. Chem. 1992, 57,
7140–7143.
10. Badescu, V. Synthesis and Thermal Study of a Series of
N-Benzoyloxyamines and the Scope and Limitations of an
Amine Oxidation Process. Masters Thesis, Department of
Chemistry, University of Central Florida, 2000.
11. (a) Gitchell, A.; Simonaitis, R.; Heiklen, J. The Inhibition of
Photochemical Smog. J. Air Pollut. Control Assoc. 1974, 24,
357–361. (b) Mackay, D.; Waters, W. A. J. Chem. Soc. C
1966, 813.
1
The overall yield was 61% for the 2 steps. 29: H NMR
(CDCl3): d 8.08 (d, 2H), 7.73 (d, 2H), 7.63 (t, 1H), 7.57 (t,
2H), 7.47 (t, 2H), 7.36 (t, 2H), 7.28 (t, 2H), 5.41 (d, 1H),
4.63 (m, 1H), 4.38 (d, 2H), 4.18 (m, 3H), 4.04 (q, 2H), 2.71
(t, 2H), 1.61 (m, 1H), 1.50 (m, 2H), 1.20 (t, 3H), 0.85 (d,
3H), 0.70 (br s, 3H); 13C NMR with DEPT analysis
(CDCl3): d 173.65 (CvO), 171.00 (CvO), 164.12 (CvO),
156.15 (CvO), 144.14 (quaternary C), 143.92 (quaternary
C), 141.44 (quaternary C), 134.89 (CH), 130.30 (CH),
129.57 (CH), 129.13 (CH), 128.74 (CH), 127.88 (CH),
127.26 (CH), 126.48 (CH), 125.34 (CH), 120.16 (CH),
67.32 (CH2), 61.20 (CH2), 50.07 (CH), 47.52 (CH), 44.83
(CH2), 42.57 (CH2), 32.66 (CH2), 24.86 (CH), 23.73 (CH3),
21.86 (CH3), 14.49 (CH3). High resolution mass spectrum
(FAB) theory for (C33H37O7N2) Mþ1¼573.2601; found:
Mþ1¼573.2593. Anal/calcd for C33H36O7N2·1H2O: C,
67.10; H, 6.48; N, 4.74; found: C, 67.45; H, 6.20; N, 4.75.
Optical rotation [a]D¼þ258 (c¼3.8, CDCl3, 238C).
12. Wittman, M. D.; Halcomb, R. L.; Danishefsky, S. J. J. Org.
Chem. 1990, 55, 1981–1983.
13. Pokonski, T.; Chimiak, A. Tetrahedron Lett. 1974, 28,
2453–2456.
14. Buckley, D.; Dunstan, S.; Henbest, H. B. J. Chem. Soc. 1957,
4901–4904.
15. Geiger, W. B. J. Org. Chem. 1958, 23, 298–299.
16. Koppel, I.; Prodel, A.; Koppel, J. Org. React. (Tartu) 1990,
27(1–2), 74–86.
17. McMurry, J. Organic Chemistry; 4th ed. Brooks and Cole:
New York, 1996; pp 942–943.
18. Carey, F. A.; Sundberg, R. J. Advanced Organic Chemistry,
Part A: Structure and Mechanisms; 4th ed. Kluwer Academic:
New York, 2000; p 293 and references therein.
19. Carpino, L. A.; Cohen, B. J.; Stephens, K. E., Jr.; Sadat-
Aalaee, S. Y.; Tien, J.; Langridge, D. C. J. Org. Chem. 1986,
51, 3732–3734.
Acknowledgements
The authors thank Dr Adam Alty and SynQuest Labora-
tories for their generous gift of the fluorinated amine starting
materials. In related experiments Omar Shami generated the
N-benzoyloxyamine HCl salts and conducted their long-
term stability studies. The stable HCl salts allow for the
long-term storage of these materials. Dr Richard Gardner
assisted with the characterization of peptide 29.
20. These are known compounds with matching literature NMR
data. N-Hexylbenzamide 6a: Katritzky, A. R.; Drewniak, M.
J. Chem. Soc., Perkin Trans. 1 1988, 2339–2344. N-Cyclo-
hexyl-benzamide 7a: Kunishima, M.; Kawachi, C.; Hioki, K.;
Terao, K.; Tani, S. Tetrahedron 2001, 57, 1551–1558.
N-Benzylbenzamide 9a: Wan, Y.; Alterman, M.; Larhed,
M.; Hallberg, A. J. Org. Chem. 2002, 67, 6232–6235.
N-Methylbenzamide 10a: Hajipour, A. R.; Ghasemi, M.
Indian J. Chem. 2001, 504–507. Tributylamineoxide 18:
Ferrer, M.; Sanchez-Baeza, F.; Messeguer, A. Tetrahedron
1997, 53, 15877–15888. N-(2,2,2-Trifluoroethyl)-benzamide
24: Hegarty, A. F.; Eustace, S. J.; Tynan, N. M.; Pham-Tran,
N.-N.; Nguyen, M. T. J. Chem. Soc., Perkin Trans. 2 2001,
1239–1246.
References
1. Bauer, L.; Exner, O. Angew. Chem. Int. Ed. Engl. 1974, 13,
376–384.
2. Yale, H. L. Chem. Rev. 1943, 33, 209–256.
3. Wang, L.; Phanstiel, O., IV. J. Org. Chem. 2000, 65,
1442–1447.