Reduction of amine N-oxides by diboron reagents

Article abstract of DOI:10.1021/jo201192c

Facile reduction of alkylamino-, anilino-, and pyridyl-N-oxides can be achieved via the use of diboron reagents, predominantly bis- (pinacolato)- and in some cases bis(catecholato)diboron [(pinB)2 and (catB)₂, respectively]. Reductions occur upon simply mixing the amine N oxide and the diboron reagent in a suitable solvent, at a suitable temperature. Extremely fast reductions of alkylamino- and anilino-N-oxides occur, whereas pyridyl-N-oxides undergo slower reduction. The reaction is tolerant of a variety of functionalities such as hydroxyl, thiol, and cyano groups, as well as halogens. Notably, a sensitive nucleoside N-oxide has also been reduced efficiently. The different rates with which alkylamino- and pyridyl-N-oxides are reduced has been used to perform stepwise reduction of the N,N-dioxide of (S)-(-)-nicotine. Because it was observed that (pinB)2 was unaffected by the water of hydration in amine oxides, the feasibility of using water as solvent was evaluated. These reactions also proceeded exceptionally well, giving high product yields. In constrast to the reactions with (pinB)₂, triethylborane reduced alkylamino-N oxides, but pyridine N-oxide did not undergo efficient reduction even at elevated temperature. Finally, the mechanism of the reductive process by (pinB)₂ has been probed by ¹H and ¹¹B NMR. (Figure presented) ; 2011 American Chemical Society.

Full text of DOI:10.1021/jo201192c

ARTICLE
pubs.acs.org/joc
Reduction of Amine N-Oxides by Diboron Reagents
Hari Prasad Kokatla, Paul F. Thomson, Suyeal Bae, Venkata Ramana Doddi, and Mahesh K. Lakshman*
Department of Chemistry, The City College and The City University of New York, 160 Convent Avenue, New York,
New York 10031-9198, United States
S Supporting Information
b
ABSTRACT: Facile reduction of alkylamino-, anilino-, and pyridyl-N-oxides
can be achieved via the use of diboron reagents, predominantly bis-
(pinacolato)- and in some cases bis(catecholato)diboron [(pinB)2 and
(catB)2, respectively]. Reductions occur upon simply mixing the amine N-
oxide and the diboron reagent in a suitable solvent, at a suitable temperature.
Extremely fast reductions of alkylamino- and anilino-N-oxides occur, whereas
pyridyl-N-oxides undergo slower reduction. The reaction is tolerant of a
variety of functionalities such as hydroxyl, thiol, and cyano groups, as well as
halogens. Notably, a sensitive nucleoside N-oxide has also been reduced
efficiently. The different rates with which alkylamino- and pyridyl-N-oxides
are reduced has been used to perform stepwise reduction of the N,N0-dioxide
of (S)-(ꢀ)-nicotine. Because it was observed that (pinB)2 was unaffected by
the water of hydration in amine oxides, the feasibility of using water as solvent was evaluated. These reactions also proceeded
exceptionally well, giving high product yields. In constrast to the reactions with (pinB)2, triethylborane reduced alkylamino-N-
oxides, but pyridine N-oxide did not undergo efficient reduction even at elevated temperature. Finally, the mechanism of the
reductive process by (pinB)2 has been probed by 1H and 11B NMR.
’ INTRODUCTION
(PinB)2 and the corresponding catechol derivative [(catB)2]
have been briefly investigated for their abilities to reduce N-
oxides of pyridine, 4-phenylpyridine, and 4-methylmorpholine.
Although pyridine N-oxide was stated to be reduced in 90% yield
by (catB)2, no preparative use of these reagents has been
reported to date.33 Nevertheless, deoxygenation by diboron
reagents is anticipated to be an exceptionally mild and selective
process, with broad functional group compatibility. Both (catB)2
and (pinB)2 are commercially available. The latter is a compara-
tively less-expensive, stable compound, which requires no special
handling protocols. These considerations led us to explore the
use of diboron reagents for the reduction of amine N-oxides.
Amine N-oxides are frequently encountered in organic
synthesis,1 and often the chemical methodology calls for a
reduction of the N-oxide to the amine. A variety of methods
involving metals have been developed for the N-oxide to amine
conversion.2ꢀ21 There are fewer methods that do not rely on the
use of metals, such as processes involving sulfurous acid,22 SO2,23
sulfur monoxide,24 trimethyl(ethyl)amineꢀSO2 complex,25
PCl3,26 PPh3 at high temperatures,27 di-n-propyl sulfoxylate,28
CS2,29 bakers' yeast,30 and alcohols and a base.31
Many of these methods for N-oxide to amine conversion are
encumbered with functional group incompatibility problems, the
need for high temperatures and/or sealed tube techniques,
reagents that produce undesired side reactions, and difficult
access to reagents and/or catalysts. Thus, a mild and simple
method for the selective conversion of amine N-oxides to the
corresponding amines continues to be synthetically desirable.
In the course of recent work, we had observed an unusual
deoxygenation of O6-(benzotriazolyl)inosine and 20-deoxyino-
sine derivatives (a class of purine-benzotriazole ethers) upon
exposure to bis(pinacolato)diboron [(pinB)2] and Cs2CO3
(Scheme 1), and had proposed a plausible mechanism for the
conversion.32 The products from the deoxygenation reactions
are C-6 benzotriazolyl purine nucleoside derivatives, which were
isolated in good yields.32 In this reaction, it appears that the
oxygen atom attached to the benzotriazolyl moiety is transferred
to (pinB)2, which undergoes oxidation to (pinB)2O.
’ RESULTS AND DISCUSSION
The generally oxophilic nature of boron and the relatively
stable BꢀO bond (∼120ꢀ13034,35 kcal/mol) in comparison to
the BꢀB bond (∼68 kcal/mol35), and the large favorable
enthalpy associated with the formation of two BꢀO bonds
(ΔH ca. 180 kcal/mol33), all bode well for the use of diboron
reagents as reducing agents. Furthermore, trimethylamine N-
oxide has been known for some time to be an effective oxidant for
converting organoboranes to alcohols.36ꢀ39 With these data, we
conducted an NMR tube experiment, where 4-methylmorpho-
line N-oxide was exposed to a slight stoichiometric excess of
(pinB)2, in CDCl3 at room temperature. No additional
Received: June 8, 2011
Published: August 03, 2011
r
2011 American Chemical Society
7842
dx.doi.org/10.1021/jo201192c J. Org. Chem. 2011, 76, 7842–7848
|

The Journal of Organic Chemistry
ARTICLE
Scheme 1. Deoxygenation of O6-(Benzotriazol-1-yl)inosine
deoxygenation of 4-methylmorpholine N-oxide and 4-meth-
oxypyridine N-oxide was determined to be about 0.6 kcal/g by
calorimetry.
and 20-Deoxyinosine Using (pinB)2
With the information gleaned from NMR studies, we turned
our attention to the broader synthetic generality of the re-
ductive process. We were aware that alkylamine N-oxides
underwent rapid reactions, whereas pyridine N-oxides were
slower to react. One other issue remained to be addressed: the
formation of amine complexes with the boron-based bypro-
duct, (pinB)2O,33 which we also observed. The formation of
amine (pinB)2O adducts has been attributed to the increased
3
Lewis acidity at the boron center after oxidation of the BꢀB
bond.33 We reasoned that addition of a strongly coordinating
alkyldiamine should liberate the desired amine from the com-
plex, and we chose water-soluble ethylenediamine for the
workup protocol. The results of the reduction procedure are
presented in Table 1.
Some notable points emerge from the results in Table 1. N-
Oxides or their hydrates can be reduced, and water does not
seem to hamper this reaction by degradation of (pinB)2
(entries 3 and 5). Reactions of some amine N-oxides were
slow at room temperature, and heating accelerated the reduc-
tion. There does not appear to be an appreciable effect of
solvent, and solubility of reactants as well as reaction tempera-
ture may be the only considerations for solvent selection
(entries 4 and 11). Thiol and hydroxyl groups did not need
any specific caution or protection (entries 8 and 9). Halogens
are well tolerated, and steric crowding by two chlorine atoms
does not deter the reduction (entries 10 and 11). Electron-
deficient p-cyanopyridine N-oxide also underwent clean reduc-
tion (entry 12). The substituents in examples 10ꢀ12 could be
incompatible with some other reduction methods. N-Oxides of
N,N-diethylaniline, N-benzylpiperidine, and N-benzylmorpho-
line were all reduced rapidly (entries 13ꢀ15).
Figure 1. Evaluating the reduction of 4-methylmorpholine N-oxide
with (pinB)2 by NMR (at 500 MHz in CDCl3). Panel A: spectrum of
4-methylmorpholine N-oxide prior to addition of (pinB)2. Panel B:
immediately after addition of 1.0 mol equiv (pinB)2 (mixture was not
shaken). Panel C: immediately after shaking the reaction mixture. Panel
D: mixture of (pinB)2 and 1.5 mol equiv of 4-methylmorpholine N-
oxide. Panel E: spectrum of pure 4-methylmorpholine.
In the cases of pyridine, 2-picoline, 3-picoline, and 2,6-
lutidine N-oxides, although reductions with (pinB)2 proceeded
smoothly, product volatility made their separation from pina-
col-based byproducts somewhat more cumbersome. Use of
bis(catecholato)diboron [(catB)2] ameliorated the purification
problems (entries 16ꢀ19). In our hands, reactions with (catB)2
appeared slower as compared to (pinB)2, and (catB)2 seems to
be moisture-sensitive. As an example, in CH3CN, reduction of
pyridine N-oxide with stoichiometric (pinB)2 was complete
within 10 h at 70 °C. On the other hand, even with 1.5 mol
equiv of (catB)2, a comparable reaction was incomplete at 24 h
and a temperature of 120 °C was necessary. Nevertheless,
reductions can be conducted with (catB)2, and the steric bulk in
2,6-lutidine N-oxide was not an impediment. In prior work, a
difference has been noted in the reactivities of (pinB)2 and
(catB)2, but reduction of pyridine N-oxide with (catB)2 was
complete within 3 h at 25 °C, whereas reaction with (pinB)2
required 6 h at 70 °C.33
Next we wanted to test the deoxygenation of more complex
substrates. Given our interest in nucleoside functionalization, we
synthesized 20,30,50-tri-O-(tert-butyldimethylsilyl)adenosine
N1-oxide (1 in Scheme 2) using a known procedure.40 This
compound is significantly more fragile as compared to the
substrates in Table 1. Exposure of a 0.08 M solution of trisilyl
nucleoside 1 to (pinB)2 in either CH3CN at 70 °C, or in diglyme
at 120 °C, led to clean reduction, and trisilyl adenosine 2 was
isolated in 71% and 87% yield, respectively (Scheme 2).
precautions were taken, such as the use of inert atmosphere or
anhydrous conditions. On the basis of 1H NMR evaluation, this
reaction was extremely rapid (Figure 1) and complete reduction
took no more than 10 min at room temperature. In C6D6 this
reaction is reported to be complete within 1 h at 25 °C, leading to
a 74% isolated yield of a morpholine (pinB)2O adduct.33 Con-
3
tact of 4-methylmorpholine N-oxide with (pinB)2 at room
temperature, in the absence of solvent, led to an immediate,
violent reaction with evolution of fumes.
We then queried whether the hybridization of the nitrogen
atom made any difference on the course of the reaction. We
chose to conduct the reduction with 4-methoxypyridine
N-oxide hydrate because this compound was expected to be
reasonably reactive based on the anticipated reaction mechan-
ism (vide infra). Again, an experiment was conducted in an
NMR tube using CD3CN as solvent (based upon solubility of
reagents and the fact that the reaction mixture could be
heated, if necessary), and upon mixing 4-methoxypyridine
N-oxide hydrate and (pinB)2, a reaction was clearly evident.
However, the reaction was much slower than that of 4-methyl-
morpholine N-oxide and reached completion within 20 h at
room temperature. In comparison to 4-methylmorpholine N-
oxide, a similar slower reduction of pyridine N-oxide by
(pinB)2 has been noted.33 The exotherm associated with the
7843
dx.doi.org/10.1021/jo201192c |J. Org. Chem. 2011, 76, 7842–7848

The Journal of Organic Chemistry
ARTICLE
Table 1. Generality of Amine N-Oxide Reduction with Diboron Reagentsa
a Reactions were conducted at 1 M amine N-oxide concentration. b Where reported, yield is of isolated and purified product. c NMR tube experiment
showed complete disappearance of the N-oxide resonance(s) and appearance of product resonance(s). Some chemical shift differences and/or signal
broadening was observed, likely due to coordination of the amine to (pinB)2O in the reaction mixture (as reported in ref 33). d Reaction was incomplete.
e Bis(catecholato)diboron: (catB)2.
Scheme 2. Reduction of 20,30,50-Tri-O-(tert-
Scheme 3. Reduction of Nicotine N-Oxides with (pinB)2
butyldimethylsilyl)adenosine N1-Oxide Using (pinB)2
(S)-(ꢀ)-Nicotine presents an interesting case study be-
cause it contains alkylamino and pyridyl nitrogen atoms. Both
the N-oxide 4 and the N,N0-dioxide 5 were synthesized using
known procedures.41,42 These compounds gave us excellent
models to test selective as well as comprehensive N-oxide
reduction (Scheme 3). Reaction of nicotine N-oxide 4 with 1
mol equiv of (pinB)2 gave nicotine (3) in 95% yield within 10
min, at room temperature. On the other hand, reduction of the
N,N0-dioxide 5 with 1 mol equiv of (pinB)2 led exclusively to
N-oxide 630,42 in 96% yield, within 10 min at room tempera-
ture. By contrast, reduction of 5 to 6 has previously been
conducted using SO2/EtOH, in an overnight reaction.43 The
present method provides straightforward, scalable access to
the expensive N-oxide 6 (commercially ca. $1ꢀ3/mg) from
readily prepared N,N0-dioxide 5.
Finally, reaction of N,N0-dioxide 5 with 2 mol equiv of (pinB)2
showed formation of N-oxide 6 within 10 min at room tempera-
ture, and complete reduction to (S)-(ꢀ)-nicotine (3) was then
7844
dx.doi.org/10.1021/jo201192c |J. Org. Chem. 2011, 76, 7842–7848

The Journal of Organic Chemistry
ARTICLE
Table 2. Use of Water as Reaction Medium for Reductions
a
Using (pinB)2
a Reactions were conducted at 1 M amine N-oxide concentration
in water. b Yields reported are of isolated and purified products.
Figure 2. Monitoring the reaction between 4-methoxypyridine N-oxide
hydrate and (pinB)2 in CD3CN, by NMR.
achieved within 1 h, by heating the reaction mixture to 70 °C
(93% yield). In contrast to the workup procedure with the
simpler amines, which required addition of ethylenediamine to
completely remove boron-containing byproducts, reactions of 4
and 5 did not require this step. The chiral center adjacent to the
pyrrolidine nitrogen atom remains unaffected in the reduction, as
seen from the optical rotation of the total reduction product 3 in
comparison to the (S)-(ꢀ)-nicotine sample used for this work.
Use of Water as Reaction Medium. Because the water of
hydration in the amine N-oxides did not seem to affect the
reductions by (pinB)2 (entries 3 and 5 in Table 1; also see the
NMR studies described below), we queried whether water could
be used as a reaction medium. Results from the reactions of five
amine N-oxides in water are shown in Table 2. Not only did these
reactions prove comparable to those accomplished in organic
solvents, but also product isolation was just as readily achieved.
Comparison of Reductions by Triethylborane and (pinB)2.
Although pinBꢀH has been used for preparing (pinB)2O by
reduction of trimethylamine N-oxide,44 its utility as a reducing
reagent is limited by functional group compatibility issues and by
moisture sensitivity considerations. Further, pinBꢀOH initially
formed in the reaction of pinBꢀH and the amine N-oxide
consumes 1 equiv of pinBꢀH, releasing hydrogen.44 Formation
of hydrogen by reaction of BꢀH containing organoboranes and
amine N-oxides is known.36,37 Thus, minimally 2 mol equiv of
pinBꢀH will be required for reducing amine oxides. In principle,
trialkylboranes can be used to reduce amine N-oxides, as has been
demonstrated with trimethylamine N-oxide.36ꢀ39 Although 3
mol of a N-oxide can be reduced per 1 mol of a trialkylborane, it
has been shown that the rate for each sequential BꢀC oxidation
step is slower than the previous one.37 For example, reaction of
2 mol equiv of trimethylamine N-oxide with 1 mol equiv of n-
Bu3B is rapid.45 However, reaction with the third molar equiva-
lent of the N-oxide required 24 h at reflux.45 On the basis of these
results, we conducted several experiments. First, we assessed
selectivity of the reduction by exposing a solution of nicotine N,
N0-dioxide 5 in CH3CN to 0.3 mol equiv of commercially
available Et3B, at room temperature. This reaction yielded
exclusively N-oxide 6 in 53% yield. Because only two alkylamine
N-oxides can be expected to undergo reduction per Et3B, at room
temperature,45 in theory the maximal yield should be 60%.
Second, we conducted a reaction of N-benzylpiperidine N-oxide
with 1 mol equiv of Et3B, also in CH3CN. This reaction was
complete within 10 min, affording N-benzylpiperidine in 83%
yield. Finally, with 1 mol equiv of Et3B, practically no reduction
of pyridine N-oxide was observed at room temperature in
CH3CN, and just a trace of pyridine was observed by TLC, after
heating at 80 °C for 20 h. It has been reported that in boiling
xylene, only a single BꢀC bond of an alkylborane is oxidized by
4-methylpyridine N-oxide.37 Thus, reactions of alkylboranes with
N-oxides of alkylamines are facile in comparison to reactions with
pyridine N-oxides. On the basis of these collective data, (pinB)2
appears to be a reagent endowed with broad N-oxide reducing
ability as well as selectivity.
Evaluation of the Reaction Mechanism. To gain insight into
the mechanism of this reaction, we conducted some 1H and 11
B
NMR experiments. A 0.28 M solution of 4-methoxypyridine
N-oxide in CD3CN was exposed to 1.05 mol equiv of (pinB)2 in
an NMR tube. Figure 2, panel A, shows that within 10 min at
room temperature, a fast reaction ensued. Continued monitoring
of this reaction showed that substantial reduction was complete
within 125 min, and finally no trace of the N-oxide was detectable
at 20 h. The disappearance of the OCH3 resonance in 4-methox-
ypyridine N-oxide (δ = 3.90 ppm) and formation of the
corresponding resonance in 4-methoxypyridine (δ = 3.89 ppm)
was also observed (not shown in Figure 2, panel A). The
7845
dx.doi.org/10.1021/jo201192c |J. Org. Chem. 2011, 76, 7842–7848

The Journal of Organic Chemistry
ARTICLE
Scheme 4. Plausible Mechanism for the N-Oxide Reduction
resonances of (pinB)2 and its oxidation product appear at δ =
1.25ꢀ1.26 ppm but are not particularly informative.
conducted in water as reaction medium. In recent years,
heterocyclic N-oxides have proven to be excellent substrates
for metal-catalyzed CꢀH bond activation processes, wherein
the N-oxide moiety is finally reduced using Pd/HCO2NH4,
Zn/NH4Cl, or PCl3.50,51 In this context, the present approach
provides an excellent, mild, and functional group-compatible
method for the deoxygenation. Also, deoxygenative functiona-
lization of pyridines has been recently reported.52 Thus, various
natural products containing both pyridyl and alkylamino
groups can be functionalized at the pyridine ring via the
selectivity offered by this route. All these factors render this
method attractive for wide ranging applications in organic
synthesis, and our efforts are focused on tapping its utility for
more complex functionalizations.
11B NMR (Figure 2, panel B) showed disappearance of the
resonance at δ = 30.6 ppm corresponding to (pinB)2 and
formation of a single new entity with a resonance at δ = 22.5
ppm. From the 11B NMR chemical shift, it is plausible that the
oxidation product formed is (pinB)2O because the chemical
shift is consistent for a boron bonded to an oxygen atom.46 For
comparison, authentic (pinB)2O was synthesized as reported,44
and its 11B resonance was recorded in several solvents.47 The
chemical shift of the boron-containing product from the
deoxygenation reaction is consistent with (pinB)2O. Because
it is not of particular relevance to the chemical methodology,
we have not determined whether this product is or contains
hydrolyzed pinBꢀOH. The literature indicates very similar
chemical shifts for (pinB)2O and pinBꢀOH,44,48,49 and amine
can form a (pinB)2O adduct.33
’ EXPERIMENTAL SECTION
To investigate by NMR whether water in the N-oxide hydrate
had any influence on the reaction course, the reduction of
4-phenylpyridine N-oxide (which was not a hydrate) was also
monitored by 1H and 11B NMR. The results from the 11B NMR
experiment in this case were comparable to those from the
reduction of 4-methoxypyridine N-oxide hydrate.
General Experimental Considerations. Thin layer chromatog-
raphy was performed on 250 μm silica plates, and column chromato-
graphic purifications were performed on 200ꢀ300 mesh silica gel.
MeCN was distilled over CaH2, and 1,4-dioxane was distilled over
LiAlH4 and then over Na. All other reagents were obtained from
commercial sources. N-Oxides in entries 1ꢀ6, 8, 9, 11, 12, 16ꢀ19 were
obtained from commercial suppliers and were used without further
purification, whereas N-oxides in entries 7, 10, 13, 14, 15 were
synthesized. Trisilyl adenosine N1-oxide 1 was prepared via a known
route (ref 40). Nicotine N-oxide 4 and the N,N0-dioxide 5 were
synthesized by reported routes (ref 41). 1H NMR spectra were recorded
at 500 MHz in CDCl3 or DMSO-d6 and are referenced to the residual
protonated solvent resonance. All products are known compounds and,
except for N-benzylmorpholine, are commercially available. Character-
ication data of N-benzylpiperidine and N-benzylmorpholine are re-
ported in the literature.53,54
General Procedure for Synthesis of Amine N-Oxides. To a
stirred solution of amine (5.5ꢀ6.7 mmol) in CHCl3 (2 mL) was added
70% m-CPBA (1 mol equiv), portionwise at 0 °C. The resulting mixture
was stirred at room temperature for 12 h, at which time complete
comsumption of starting material was observed by TLC. The reaction
mixture was diluted with CHCl3, and solid K2CO3 (4 mol equiv) was
added. The resulting mixture was stirred for an additional 10 min. The
solid was separated by filtration, and the filtrate was dried over Na2SO4
and concentrated under reduced pressure to afford the amine N-oxide in
yields of 86ꢀ90%.
On the basis of the 11B NMR experiments, where only two
boron resonances ascribable to (pinB)2 and a final product were
discernible, a plausible mechanism is shown in Scheme 4 that is
consistent with the one proposed previously.33 An NꢀOꢀB
bond is formed by reaction of the N-oxide at a boron center, and
in a subsequent step (pinB)2O is produced with liberation of the
amine that can remain bound to the boron. The 1H NMR results,
where only two sets of resonances were observed during the
course of the reduction, also support this proposal. Consistent
with the mechanism proposed, the highly electron-depleted p-
nitropyridine N-oxide was a recalcitrant substrate, whereas the
electron-rich p-methoxypyridine N-oxide underwent reaction at
room temperature.
’ CONCLUSIONS
We have described a facile reduction method for alkylamino-,
anilino-, and pyridyl-N-oxides. The reactions rely on a simple
exposure of the N-oxide to (pinB)2, in a suitable solvent, and at a
suitable temperature to attain a reasonable reaction rate. The
reductions are generally mild, require no specialized handling, are
compatible with a wide range of functionalities, and can be
conducted with stoichiometric (pinB)2. Product isolation is also
straightforward. In addition to the broad range of amine oxides
that can be reduced, (pinB)2 offers selectivity in reduction. Thus,
an alkylamino-N-oxide can be reduced in the presence of a
pyridyl-N-oxide, as demonstrated with the N,N0-dioxide of
(S)-(ꢀ)-nicotine. Water of hydration does not impact the
reductions, and notably reductions of amine N-oxides can be
General Procedure for the Reduction of Amine N-Oxides.
Reactions were conducted on a 0.5ꢀ1.0 mmol scale. Caution: These
reductions, particularly those of alkylamine N-oxides, can be quite
vigorous and exothermic. Hence, appropriate precautions should be
taken while conducting them.
For High Boiling Amines. A 1 M solution of amine N-oxide in the
appropriate solvent (see Table 1) was stirred in an oven-dried reaction
vial. (PinB)2 (1 mol equiv) was added, the vial was flushed with nitrogen
gas, and the mixture was stirred at the appropriate temperature (see
Table 1 for details). When TLC indicated the reaction to be complete,
7846
dx.doi.org/10.1021/jo201192c |J. Org. Chem. 2011, 76, 7842–7848

The Journal of Organic Chemistry
ARTICLE
ethylenediamine (20 mol equiv) was added to the mixture, and the
stirring was continued for 1 h at room temperature. Then the reaction
mixture was diluted with water (10 mL) and extracted with Et2O (3 ꢁ
10 mL). The organic layer was dried over Na2SO4 and concentrated
under reduced pressure. The crude material was then purified via a silica
gel column using EtOAc/hexanes as eluting solvent.
2.78ꢀ2.69 (m, 1H), 2.63ꢀ2.54 (m, 1H), 2.31ꢀ2.25 (m, 1H),
2.10ꢀ2.03 (m, 1H). [R]25D = ꢀ73.5 (c = 2 mg/mL, CHCl3).
Nicotine N,N0-Dioxide 5.41 Synthesized by the procedure reported
in ref 41. 1H NMR (500 MHz, CDCl3): δ 8.35 (br s, 1H, Ar-H), 8.22
(d, 1H, Ar-H, J = 6.3 Hz), 7.84 (d, 1H, Ar-H, J = 7.8 Hz), 7.30 (t, 1H,
Ar-H, J = 7.0 Hz), 4.14 (dd, 1H, J = 7.6, 11.2 Hz), 3.85 (t, 1H, J = 8.7
Hz), 3.62 (app q, 1H, Japp ∼ 9.8 Hz), 3.06 (s, 3H, CH3), 2.70ꢀ2.52
(m, 2H), 2.35ꢀ2.29 (m, 1H), 2.12ꢀ2.03 (m, 1H). [R]25D = +109.3
(c = 5 mg/mL, CHCl3).
For Low Boiling Amines. A 1 M solution of amine N-oxide in MeCN
was stirred in an oven-dried reaction vial. (CatB)2 (1.5 mol equiv) was
added, the vial was flushed with nitrogen gas, and the mixture was stirred
at 120 °C. When TLC indicated the reaction to be complete, ethylene-
diamine (20 mol equiv) was added to the mixture, and the stirring was
continued for 1 h at room temperature. Then the reaction mixture was
diluted with water (10 mL) and extracted with Et2O (3 ꢁ 10 mL). The
organic layer was extracted with 1 N HCl. The aqueous layer was
separated and carefully neutralized with 1 N NaOH. The free amine was
extracted into Et2O. The organic layer was dried over Na2SO4 and
concentrated under reduced pressure. The crude material was then
purified via a silica gel column using EtOAc/hexanes as eluting solvent.
General Procedure for the Reduction of Amine N-Oxides
in Water. To a 1 M solution of amine N-oxide in water was added
(pinB)2 (1 mol equiv) with stirring. The mixture was stirred at the
appropriate temperature (see Table 2 for details). When TLC indicated
the reaction to be complete, ethylenediamine (20 mol equiv) was added
to the mixture, and the stirring was continued for 1 h at room
temperature. Then the reaction mixture was diluted with water
(10 mL) and extracted with Et2O (3 ꢁ 10 mL). The organic layer
was dried over Na2SO4 and concentrated under reduced pressure. The
crude material was then purified via a silica gel column using EtOAc/
hexanes as eluting solvent.
Reduction of Nicotine N-Oxide 4 to (S)-(ꢀ)-Nicotine (3). In an oven-
dried reaction vial equipped with a stirring bar was placed the nicotine N-
oxide 4 (50.0 mg, 0.280 mmol) in CH3CN (0.25 mL). (PinB)2 (71.0 mg,
0.280 mmol) was added, the vial was flushed with nitrogen gas, and the
mixture was stirred at the room temperature. After 10 min, TLC
indicated the reaction to be complete. The reaction mixture was diluted
with water (10 mL) and extracted with Et2O (3 ꢁ 10 mL). The organic
layer was dried over Na2SO4 and concentrated under reduced pressure.
The crude material was then purified via a silica gel column using 80%
Et2O in hexanes as eluting solvent to give 43 mg (95% yield) of 3 as a
yellowish liquid. For 1H NMR data see reduction of 5 to 3 below.
Reduction of Nicotine N,N0-Dioxide 5 to Nicotine N-Oxide 6. In an
oven-dried reaction vial equipped with a stirring bar was placed the
nicotine N,N0-dioxide 5 (50.0 mg, 0.257 mmol) in CH3CN (0.25 mL).
(PinB)2 (65.0 mg, 0.257 mmol) was added, the vial was flushed with
nitrogen gas, and the mixture was stirred at the room temperature.
After 10 min, TLC indicated the reaction to be complete. The reaction
mixture was directly loaded onto a silica gel column and purified using
10% MeOH in Et2O as eluting solvent to give 45 mg (96% yield) of 6 as a
yellowish liquid. [R]25D = ꢀ146.7 (c = 3.5 mg/mL, CHCl3). Rf (SiO2/
40%MeOHinEtOAc) = 0.25. 1H NMR (500 MHz, CDCl3): δ 8.24 (brs,
1H, Ar-H), 8.10 (d, 1H, Ar-H, J = 5.8 Hz), 7.26 (m, 1H, Ar-H), 7.21 (app
triplet, 1H, Ar-H, Japp ∼ 7.0 Hz), 3.21 (t, 1H, J = 7.8 Hz), 3.08 (t, 1H, J =
7.8 Hz), 2.33 (app q, 1H, Japp ∼ 8.7 Hz), 2.29ꢀ2.20 (s, 3H, CH3
superimposed on m, 1H), 1.98ꢀ1.89 (m, 1H), 1.85ꢀ1.75 (m, 1H),
1.71ꢀ1.64 (m, 1H).
20,30,50-Tri-O-(tert-butyldimethylsilyl)adenosine N1-Oxide (1).40 In
an oven-dried 10 mL round-bottom flask, equipped with a stirring bar,
was placed a solution of 20,30,50-tri-O-(tert-butyldimethylsilyl)adenosine
(2, 200 mg, 0.32 mmol) in MeOH (5 mL). To this well-stirred mixture
was added 70% m-CPBA (79 mg, 1.4 mol equiv), the vial was flushed
with nitrogen gas, and the mixture was stirred at room temperature.
After 12 h, TLC indicated the reaction to be complete. The mixture was
diluted with CHCl3 (30 mL) and washed with aqueous NaHCO3 (3 ꢁ
20 mL). The organic layer was dried over Na2SO4, filtered, and
evaporated under reduced pressure. The crude material was purified
via a silica gel column using 10% MeOH in EtOAc to give 1 as a brown
solid (179 mg, 87% yield). 1H NMR (500 MHz, CDCl3): δ 8.68 (s, 1H,
Ar-H), 8.30 (s, 1H, Ar-H), 5.98 (d, 1H, H-10, J = 5.4 Hz), 4.57 (t, 1H,
H-20, J = 4.4 Hz), 4.28 (t, 1H, H-30, J = 3.9 Hz), 4.27 (q, 1H, H-40, J = 3.1
Hz), 4.00 (dd, 1H, H-5, J = 3.9, 11.5 Hz), 3.78 (dd, 1H, H-50, J = 2.4, 11.5
Hz), 0.96, 0.92, and 0.80 (3s, 27H, t-Bu), 0.15, 0.13, 0.10, 0.09, ꢀ0.03,
and ꢀ0.23 (6s, 18H, SiCH3). HRMS (ESI) calcd for C28H56N5O5Si3
[M + H]+ 626.3584, found 626.3608.
Reduction of Nicotine N,N0-Dioxide 5 to (S)-(ꢀ)-Nicotine (3). In an
oven-dried reaction vial equipped with a stirring bar was placed the
nicotine N,N0-dioxide 5 (50.0 mg, 0.257 mmol) in CH3CN (0.25 mL).
(PinB)2 (130.0 mg, 0.514 mmol) was added, the vial was flushed with
nitrogen gas, and the mixture was stirred at the room temperature for 10
min and then heated to 70 °C for 1 h. TLC indicated the reaction to be
complete. The reaction mixture was directly loaded onto a silica gel
column and purified using 80% Et2O in hexanes as eluting solvent to
yield 39 mg (93% yield) of 3 as a yellowish liquid. [R]25D = ꢀ165.5 (c =
1.5 mg/mL, CHCl3). For the commercial sample of nicotine used for
1
this chemistry [R]25 = ꢀ167.2 (c = 1.5 mg/mL, CHCl3). H NMR
D
(500 MHz, CDCl3): δ 8.55 (d, 1H, Ar-H, J = 1.9 Hz), 8.51 (dd, 1H, Ar-
H, J = 1.4, 4.8 Hz), 7.71 (dt, 1H, Ar-H, J = 1.9, 7.8 Hz), 7.24 (dd, 1H, Ar-
H, J = 4.8, 7.8 Hz), 3.24 (t, 1H, J = 7.8 Hz), 3.07 (t, 1H, J = 8.3 Hz), 2.30
(app q, 1H, Japp ∼ 8.7 Hz), 2.23ꢀ2.16 (s, 3H, CH3 superimposed on m,
1H), 1.98ꢀ1.91 (m, 1H), 1.85ꢀ1.78 (m, 1H), 1.74ꢀ1.70 (m, 1H).
Reduction of 20,30,50-Tri-O-(tert-butyldimethylsilyl)adenosine N1-
Oxide (1). In an oven-dried reaction vial equipped with a stirring bar
was placed the adenosine N-oxide 1 (50.0 mg, 0.08 mmol) in diglyme
(1 mL). (PinB)2 (20.0 mg, 0.08 mmol) was added, the vial was flushed
with nitrogen gas, and the mixture was stirred at 120 °C. After 2 h, TLC
indicated the reaction to be complete. Ethylenediamine (20 mol
equiv) was added to the mixture, and the stirring was continued for
1 h atroom temperature. The mixture was dilutedwithwater (10 mL) and
extracted with CH2Cl2 (3 ꢁ 10 mL). The organic layer was dried over
Na2SO4 and concentrated under reduced pressure. The crude material
was then purified via a silica gel column using 30% EtOAc in hexanes as
eluting solvent to give 42 mg (87% yield) of 2 as a white solid.
’ ASSOCIATED CONTENT
Supporting Information. 1H NMR spectra of all N-
S
b
1
1
oxides and their reduction products, as well as the Hꢀ H
COSY spectrum for 1. This material is available free of charge
via the Internet at http://pubs.acs.org.
Nicotine N-Oxide 4.41 Synthesized by the procedure reported in
ref 41. 1H NMR (500 MHz, CDCl3): δ 8.63 (dd, 1H, Ar-H, J = 1.4, 4.8
Hz), 8.58 (d, 1H, Ar-H, J = 1.4 Hz), 8.29 (d, 1H, Ar-H, J = 7.8 Hz), 7.36
(dd, 1H, Ar-H, J = 4.8, 7.8 Hz), 4.24 (dd, 1H, J = 7.3, 11.7 Hz), 3.78 (t,
1H, J = 8.7 Hz), 3.63 (app q, 1H, Japp ∼ 9.9 Hz), 2.98 (s, 3H, CH3),
’ AUTHOR INFORMATION
Corresponding Author
*Tel (212) 650-7835; fax (212) 650-6107; e-mail: lakshman@
sci.ccny.cuny.edu.
7847
dx.doi.org/10.1021/jo201192c |J. Org. Chem. 2011, 76, 7842–7848

The Journal of Organic Chemistry
ARTICLE
’ ACKNOWLEDGMENT
Bifunctional Lewis Acid Reactivity of Diol-Derived Diboron Reagents.
In Group 13 Chemistry/From Fundamentals to Applications; Shapiro, P. J.;
Atwood, D. A., Eds.; ACS Symposium Series 822; American Chemical
Society: Washington, DC, 2002; pp 70ꢀ87.
Infrastructural support at CCNY was provided by NIH/NCRR/
RCMI Grant G12 RR03060. PFT was supported by Ruth L.
Kirschstein Research Service Award 5F31 GM082025-02 from
the NIH as well as the NIH RISE program. We are grateful to
AllyChem for a generous sample of bis(pinacolato)diboron and
bis(catecholato)diboron. We thank Dr. P. Pradhan (CCNY) for
assistance with the NMR experiments, Prof. B. Zajc (CCNY) for
important discussions, and Ms. A. Joshi-Pangu and Prof. M. Biscoe
(CCNY) for help with and use of their calorimeter.
(34) Larkin, J. D.; Bhat, K. L.; Markham, G. D.; James, T. D.; Brooks,
B. R.; Block, C. W. J. Phys. Chem. A 2008, 112, 8446–8454.
(35) Sana, M.; Leroy, G.; Wilante, C. Organometallics 1991,
10, 264–270.
(36) K€oster, R.; Morita, Y. Angew. Chem., Int. Ed. 1966, 5, 580.
(37) K€oster, R.; Morita, Y. Liebigs Ann. Chem. 1967, 704, 70–90.
(38) K€oster, R.; Arora, S.; Binger, P. Angew. Chem., Int. Ed. 1969,
8, 205.
(39) Kabalka, G. W.; Hedgecock, H. C., Jr. J. Org. Chem. 1975, 40,
1776–1779.
(40) Tsunoda, H.; Ohkubo, A.; Taguchi, H.; Seio, K.; Sekine, M.
J. Org. Chem. 2008, 73, 1217–1224.
(41) Uwai, K.; Sato, H.; Kazakami, N.; Matsuzaki, H.; Takeshita, M.
Arkivoc 2003, viii, 211–219.
(42) Phillipson, J. D.; Handa, S. S. Phytochemistry 1975, 14,
2683–2690.
(43) Johnson, A. W.; King, T. J.; Turner, J. R. J. Chem. Soc.
1958, 3230–3231.
(44) Hawkeswood, S.; Stephan, D. W. Dalton Trans.
2005, 2182–2187.
(45) Soderquist, J. A.; Najafi, M. R. J. Org. Chem. 1986, 51,
1330–1336.
’ REFERENCES
(1) Albini, A. Synthesis 1993, 263–277.
(2) (a) Balicki, R. Synthesis 1989, 645–646. (b) Berson, J. A.; Cohen,
T. J. Org. Chem. 1955, 20, 1461–1468.
(3) Balicki, R.; Cybulski, M.; Maciejewski, G. Synth. Commun. 2003,
33, 4137–4141.
(4) Jousseaume, B.; Chanson, F. Synthesis 1987, 55–57.
(5) Zhang, Y.; Lin, R. Synth. Commun. 1987, 17, 329–332.
(6) Ilanakumaran, P.; Chandrasekaran, S. Tetrahedron Lett. 1995,
36, 4881–4882.
(7) Yoo, B. W.; Choi, J. W.; Yoon, C. M. Tetrahedron Lett. 2005,
47, 125–126.
(8) Yoo, B. W.; Park, M. C. Synth. Commun. 2008, 38, 1646–1650.
(9) Yadav, J. S.; Subba Reddy, B. V.; Murlidhar Reddy, M. Tetra-
hedron Lett. 2000, 41, 2663–2665.
(46) Laitar, D. S.; M€uller, P.; Sadighi, J. P. J. Am. Chem. Soc. 2005,
127, 17196–17197.
(47) Boron resonance of (pinB)2 in the following solvents: benzene-
d6 δ = 34.2 ppm, CD3CN δ = 30.6 ppm, CDCl3 δ = 30.6 ppm, and THF-
d8 δ = 31.7 ppm. Boron resonance of (pinB)2O in the following solvents:
benzene-d6 δ = 22.2 ppm (21.6 in ref 41), CD3CN δ = 22.4 ppm, CDCl3
δ = 21.7 ppm, and THF-d8 δ = 22.5 ppm. Signals were referenced to
(10) Ilias, M.; Barman, D. C.; Prajapati, D.; Sandhu, J. S. Tetrahedron
Lett. 2002, 43, 1877–1879.
(11) Barton, D. H. R.; Fekih, A.; Lusinchi, X. Tetrahedron Lett. 1985,
26, 4603–4606.
(12) Malinowski, M. Synthesis 1987, 8, 732–734.
(13) Chandrasekhar, S.; Reddy, C. R.; Rao, R. J.; Rao, J. M. Synlett
2002, 349–351.
(14) Han, J. H.; Choi, J. W.; Choi, K. I.; Kim, J. H.; Yoon, C. M.; Yoo,
H. W. Bull. Korean Chem. Soc. 2005, 26, 1921–1922.
(15) Alonso, F.; Radivoy, G.; Yus, M. Tetrahedron 2000,
56, 8673–8678.
(16) Han, J. H.; Choi, K. I.; Kim, J. H.; Yoo, B. W. Synth. Commun.
2004, 34, 3197–3201.
(17) Saini, A.; Kumar, S.; Sandhu, J. S. Synlett 2006, 395–398.
(18) Kano, S.; Tanaka, Y.; Sugino, E.; Hibino, S. Synthesis
1980, 695–697.
BF3 Et2O.
3
(48) (a) Bettinger, H. F.; Filthaus, M.; Bornemann, H.; Oppel, I. M.
Angew. Chem., Int. Ed. 2008, 47, 4744–4747(reported 11B resonance of
pinBꢀOH at δ = 22.5 ppm in benzene-d6). (b) Salomon, M. A.; Braun,
T.; Penner, A. Angew. Chem., Int. Ed. 2008, 47, 8867–8871 (reported 11
resonance of pinBꢀOH at δ = 21.6 ppm in PhMe-d8).
B
(49) Because of the similarities in the chemical shifts of (pinB)2O
and pinB-OH, we cannot definitively say that the former alone is formed
in the previously described deoxygenation of nucleoside benzotriazol-1-
yl ethers (ref 32). However, initial formation of (pinB)2O is entirely
plausible in that chemistry.
(50) See for examples: (a) Campeau, L.-C.; Stuart, D. R.; Leclerc,
J.-P.; Bertrand-Leperle, M.; Villemure, E.; Sun, H.-Y.; Lasserre, S.;
Guimond, N.; Lecavallier, M.; Fagnou, K. J. Am. Chem. Soc. 2009,
131, 3291–3306. (b) Huestis, M. P.; Fagnou, K. Org. Lett. 2009,
11, 1357–1360. (c) Campeau, L.-C.; Bertrand-Leperle, M.; Leclerc,
J.-P.; Villemure, E.; Gorelsky, S.; Fagnou, K. J. Am. Chem. Soc. 2008,
130, 3276–3277. (d) Campeau, L.-C.; Rousseaux, S.; Fagnou, K. J. Am.
Chem. Soc. 2005, 127, 18020–18021.
(19) Yoon, N. M. Pure Appl. Chem. 1996, 68, 843–848.
(20) Wang, Y.; Espenson, J. H. Org. Lett. 2000, 2, 3525–3526.
(21) Kumar, S.; Saini, A.; Sandhu, J. S. Tetrahedron Lett. 2005,
46, 8737–8739.
(22) Hayashi, E.; Ijima, Ch. Yakugaku Zasshi 1962, 82, 1093–1102.
(23) Daniher, F. A.; Hackley, B. E., Jr. J. Org. Chem. 1966,
31, 4267–4268.
(24) Bonini, B. F.; Maccagnani, G.; Mazzanti, G.; Pedrini, P. Tetra-
hedron Lett. 1979, 20, 1799–1800.
(25) Olah, G. A.; Arvanaghi, M.; Vankar, Y. D. Synthesis
1980, 660–661.
(51) Gong, X.; Song, G.; Zhang, H.; Li, X. Org. Lett. 2011,
13, 1766–1769.
(52) (a) Londregan, A. T.; Jennings, S.; Wei., L. Org. Lett. 2011,
13, 1840–1843. (b) Londregan, A. T.; Jennings, S.; Wei., L. Org. Lett.
2010, 12, 5254–5257.
(53) Ferrer, M.; Sꢀanchez-Baeza, F.; Messeguer, A. Tetrahedron 1997,
53, 15877–15888.
(26) Lutz, W. B.; Lazarus, S.; Klutchko, S.; Meltzer, R. I. J. Org. Chem.
1964, 29, 1645–1647.
(27) Howard, E., Jr.; Olszewski, W. F. J. Am. Chem. Soc. 1959,
81, 1483–1484.
(54) Rosenau, T.; Schmid, P.; Kosma, P. Tetrahedron 2005, 61,
3483–3487.
(28) Kagami, H.; Motoki, S. J. Org. Chem. 1978, 43, 1267–1268.
(29) Yoshimura, T.; Asada, K.; Oae, S. Bull. Chem. Soc. Jpn. 1982,
55, 3000–3003.
(30) Takeshita, M.; Yoshida, S. Heterocycles 1990, 30, 871–874.
(31) Bjørsvik, H.-R.; Gambarotti, C.; Jensen, V. R.; Gonꢀzalez, R. R.
J. Org. Chem. 2005, 70, 3218–3224.
(32) Bae, S.; Lakshman, M. K. J. Org. Chem. 2008, 73, 1311–1319.
(33) Carter, C. A. G.; John, K. D.; Mann, G.; Martin, R. L.; Cameron,
T. M.; Baker, R. T.; Bishop, K. L.; Broene, R. D.; Westcott, S. A.
7848
dx.doi.org/10.1021/jo201192c |J. Org. Chem. 2011, 76, 7842–7848