Synthetic use of the primary kinetic isotope effect in hydrogen atom transfer: Generation of α-aminoalkyl radicals

Article abstract of DOI:10.1039/c0ob00205d

The extent to which deuterium can act as a protecting group to prevent unwanted 1,5-hydrogen atom transfer to aryl and vinyl radical intermediates was examined in the context of the generation of α-aminoalkyl radicals in a pyrrolidine ring. Intra- and int

Full text of DOI:10.1039/c0ob00205d

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Synthetic use of the primary kinetic isotope effect in hydrogen atom transfer:
generation of a-aminoalkyl radicals†
Mark E. Wood,*a Sabine Bissiriou,a Christopher Lowe,b Andrew M. Norrish,a Katell Se´ne´chal,a
Kim M. Windeatt,a Simon J. Colesc and Michael B. Hursthousec
Received 4th June 2010, Accepted 5th July 2010
DOI: 10.1039/c0ob00205d
The extent to which deuterium can act as a protecting group to prevent unwanted 1,5-hydrogen atom
transfer to aryl and vinyl radical intermediates was examined in the context of the generation of
a-aminoalkyl radicals in a pyrrolidine ring. Intra- and intermolecular radical trapping following
hydrogen atom transfer provides an illustration of the use of the primary kinetic isotope effect in
directing the outcome of synthetic C–C bond-forming processes.
Introduction
Results and discussion
Traditionally, the observation of a primary kinetic isotope effect
(or lack thereof) has provided an excellent, straightforward probe
for the examination of reactions whose mechanisms involve the
cleavage of a carbon–hydrogen bond. The magnitude of the
isotope effect observed gives an estimate of the extent of hydrogen
atom transfer in the transition state, thus providing insight into
the rate-determining step of the reaction.1 In some instances, the
substitution of a hydrogen atom with deuterium may even be
sufficient to divert a reaction predominantly along an alternative,
normally minor pathway.2 There are, however, comparatively few
examples of this phenomenon being applied to routine synthetic
procedures.
Our earlier studies into the preparation of highly substituted
b-amino alcohols utilised a radical-based synthetic pathway, in
which the key step was the generation of a 1,3-oxazolidine-derived
C4 a-aminoalkyl radical 1 via a 1,5-hydrogen atom transfer step
(Scheme 1).9
Scheme 1 a-Aminoalkyl radical generation by 1,5-hydrogen atom
Clayden et al. have shown that the replacement of hydrogen
with deuterium at positions of high kinetic acidity in benza-
mides can result in selectivity in the regiochemical course of
deprotonation reactions with organolithium reagents, a deuterium
atom effectively acting as a “protecting group” for a carbon–
hydrogen bond.3 Prior to this, Miyano reported the use of
deuterium to invert the exo/endo selectivity in the generation
of an alkene by a thionyl chloride-induced b-elimination of
a 1-methylcyclobutanol moiety, as a key step in the synthesis
of 18-oxoprogesterone,4 and both Vedejs and Little5 and Dan-
ishefsky et al.6 have also used deuterium to prevent unwanted
intramolecular deprotonation by key intermediate organolithium
species (originally intended to act as nucleophiles) in cyclisation
procedures.
transfer.
During the course of this work, we encountered a problem
with the regioselectivity of hydrogen atom transfer in the remote
functionalisation process. Whilst attempting to maximise the
stereoselectivity of the trapping of the intermediate radicals 1 at
C-4, using the principle of “self-regeneration of stereocentres,”10
we examined the radical reactions of a racemic, aldehyde-derived
1,3-oxazolidine 2 (with a large difference in C2 substituent
size), prepared from 2-(2-iodobenzylamino)ethanol 39 and 4-
pyridinecarboxaldehyde 4 (Scheme 2). Although the aldehyde-
derived 1,3-oxazolidines such as 2 were found to be more robust to-
wards hydrolytic cleavage (particularly during the commonly non-
trivial chromatographic purification of radical reaction products)
than the ketone-derived species used in earlier work,9 this substrate
obviously introduced the issue of C2 vs. C4 regioselectivity in
the hydrogen atom transfer, occurring after initial aryl radical
generation.
Herein, we discuss the background to and provide the full
experimental details of some of our recent studies7 into the extent
to which deuterium can be used to protect carbon–hydrogen bonds
in radical procedures involving hydrogen atom transfer.8
aSchool of Biosciences, University of Exeter, Geoffrey Pope Building, Stocker
Road, Exeter, UK EX4 4QD. E-mail: m.e.wood@exeter.ac.uk; Tel: +44
1392 263450
bUCB Celltech, 208 Bath Road, Slough, Berkshire, UK SL1 3WE
cEPSRC X-ray Crystallography Service, Department of Chemistry, Univer-
sity of Southampton, Highfield, Southampton, UK SO17 1BJ
† CCDC reference number 291244. For crystallographic data in CIF or
other electronic format see DOI: 10.1039/c0ob00205d
Scheme 2 Reagents and conditions: MgSO4, THF, rt, 18 h (85%).
Org. Biomol. Chem., 2010, 8, 4653–4665 | 4653
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
On reducing the aryl iodide moiety in 2 with tributyltin hydride
(under standard conditions with AIBN as initiator), only two
heterocyclic products could be isolated after repeated chromatog-
raphy of the crude reaction product, required to remove all tin-
containing residues. (The remainder of the material appeared to
be a complex mixture of degradation products.) These products
were identified as the C2-linked dimeric bioxazolidine 5 and the
dihydrooxazole 6, isolated in very low (8 and 3%, respectively)
yields (Scheme 3).
from single electron reduction of a 2-aryl-2-oxazolinium salt 10
(Scheme 4).11 Our observation of apparently exclusive formation
of the meso-diastereoisomer of the dimer 5 is in contrast to the
stereoselectivity proposed by Shono et al. (Scheme 4), although
the reasons for this stereoselectivity are unclear. Indeed, it is not
unfeasible that we were only able to isolate the minor diastereoiso-
mer of 5, with the major 2R,2¢R/2S,2¢S racemic mixture possibly
being labile under the reaction and/or purification conditions. The
1H NMR spectrum of 5 also revealed a remarkable ca. 2 ppm
chemical shift difference between the diastereotopic methylene
group protons at the benzylic and oxazolidine C4/C5 positions,
as can be seen in the portion of the spectrum shown in Fig. 3.
Scheme 3 Reagents and conditions: Bu3SnH (4.1 equiv.), AIBN (0.1
equiv.), C6H6, 80 ◦C, 18 h (5, 8%; 6, 3%).
Scheme 4 Reductive coupling of 2-aryl-2-oxazolinium salts.11
Both products had clearly arisen from the formation of the
highly stabilised C2 radical intermediate 7 (Fig. 1) by either b-
elimination of a benzyl radical for the formation of 6 or radical–
radical combination in the case of 5. No products could be
identified as being derived from the required intermediate C4
radical 8 (Fig. 1).
Fig. 1 Intermediate a-aminoalkyl radicals derived from 2.
Despite the small quantity of material obtained, we were
fortunate enough to be able to obtain an X-ray crystal structure
of dimer 5, thus providing confirmation of the product structure
(Fig. 2).
Fig. 3 400 MHz 1H NMR spectrum of 5.
From previous studies,9 we were confident that should the
desired intermediate C4 a-aminoalkyl radical 8 be formed during
the hydrogen atom transfer step, we should be able to trap it with
an appropriate radicalphile. To this end, we therefore carried out a
tributyltin hydride reduction of 2 in the presence of a 3-fold excess
of tert-butyl acrylate (Scheme 5).
Scheme 5 Reagents and conditions: tert-butyl acrylate (3.0 equiv.),
Bu3SnH (3.5 equiv.), AIBN (0.5 equiv.), C6H6, 80 ◦C, 23 h, (59%).
Fig. 2 ORTEP representation of 5; ellipsoids drawn at 30% probability
level.
Surprisingly, this reaction gave the dihydrooxazole 6 as the
only isolable product, in a 59% yield, again suggesting efficient
and apparently exclusive formation of the undesired C2 radical
Interestingly, Shono et al. have demonstrated an essentially
identical dimerisation of oxazolidine C2 radicals such as 9, derived
4654 | Org. Biomol. Chem., 2010, 8, 4653–4665
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
intermediate 7. At this stage, therefore, we were faced with a
difficult regiochemical problem to solve in our key hydrogen atom
transfer step.
with 2-(2-iodobenzylamino)ethanol 3 under previously developed
conditions,9 gave 12 in an excellent (89%) yield (Scheme 8).
During studies ultimately leading to the successful total synthe-
sis of fredericamycin A, an antibiotic with antitumour activity,
Clive et al. experienced similar problems with unwanted 1,6-
hydrogen atom transfer from an aryl methyl ether (used as a phenol
protecting group) to a vinyl radical 11, generated from a 5-exo dig
spirocyclisation (Scheme 6).12
Scheme 8 Reagents and conditions: MgSO4, THF, rt, 18 h (89%).
In contrast to the results obtained with 1,3-oxazolidine substrate
2, on subjecting 12 to reduction with tributyltin hydride in the
presence of a 3-fold excess of tert-butyl acrylate, this time with
initiation using triethlyborane,15 we obtained the desired C-4
alkylated derivative 17 as the sole heterocyclic species from the
crude reaction product (Scheme 9). This was obtained in a poor
(13%) yield as a 6.5 : 1 ratio of inseparable diastereoisomers
but most importantly, neither the dihydrooxazole 6 nor the
bioxazolidine dimer 5 appeared to be produced, suggesting that
hydrogen atom transfer to give the unwanted C2 radical 7 had
been significantly suppressed as required.
Scheme 6 Unwanted 1,6-hydrogen atom transfer in fredericamycin A
synthesis.12
Deuteration of this methyl ether was found to retard this
hydrogen atom transfer sufficiently to ensure that the intermediate
vinyl radical 11 was trapped with triphenyltin hydride before
the problematic side reaction could occur.13 Encouraged by this
observation, we decided to investigate the reactivity of a C2
deuterated version of our 1,3-oxazolidine radical precursor 12
(Fig. 4).
Scheme 9 Reagents and conditions: tert-butyl acrylate (3.0 equiv.),
Bu3SnH (3.5 equiv.), Et3B (0.2 equiv.), C6H6, rt, 18 h (13%).
Although at this stage we had identified an alternative solution
to our problem of regioselectivity of hydrogen atom transfer
in 1,3-oxazolidines by using 2-iodobenzamide derivatives,16 the
importance of a-aminoalkyl radicals in synthesis17 prompted us
to undertake a more detailed, fundamental investigation into the
effectiveness of deuterium in controlling the selectivity of this
process. It is well established that quantum mechanical tunnelling
can provide the principal mechanism for hydrogen atom transfer,
resulting in much higher kinetic isotope effects than would have
been predicted from a simple zero-point energy model, adding to
the potential synthetic utility of such an approach.18 We therefore
felt it appropriate and important to assess the efficiency of
deuterium as a “blocking group” in synthetically useful scenarios
in a systematic manner.7
Fig. 4 C2 deuterated 1,3-oxazolidine radical precursor.
In order to access 12, a-deutero-4-pyridinecarboxaldehyde 13
was firstly prepared by a three-step process. Isonicotinic acid
14 was converted into an active ester 15 by an EDCI-mediated
coupling with N-hydroxysuccinimide.14 Reduction of 15 with
an excess of sodium borodeuteride gave the corresponding di-
deuterated alcohol 16 which was then oxidised to the required
aldehyde 13 under standard Swern conditions (Scheme 7).
Our interest in the generation of a-aminoalkyl radicals
via 1,5-hydrogen atom transfer highlighted studies carried out
by Undheim et al. which investigated the radical-based a-
alkylation of a number of nitrogen-containing heterocycles. Of
particular relevance was the remote functionalisation of N-(2-
iodobenzyl)pyrrolidine 18 shown in Scheme 10.19 Our initial
investigations were aimed at assessing the degree of selectivity
(and hence, the extent of the kinetic isotope effect) between 1,5-
hydrogen atom transfer and the competitive deuterium transfer in
the di-deuterated analogue 19 of Undheim et al.’s radical reaction
substrate 18 (Scheme 11).
Scheme 7 Reagents and conditions: (a) N-hydroxysuccinimide, EDCI,
DMF, 0 ◦C, 2 h, (60%); (b) NaBD4 (4.0 equiv.), EtOH, rt, 18 h (98%);
(c) DMSO (2.2 equiv.), (COCl)2 (1.1 equiv.), CH2Cl2, -60 ◦C, 2 min, then
16, -60 ◦C, 15 min, then Et3N (5 equiv.), -60 ◦C, 5 min, then warm to rt
(55%).
The overall yield of 13 from isonicotinic acid 14 was 32%, with
the level of deuteration being at least 98%. Condensation of 13
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 4653–4665 | 4655
©

View Article Online
excess of Hu¨nig’s base, 19 was isolated in a 33% overall yield from
23 (Scheme 13).
Scheme 10 Reagents and conditions: methyl methacrylate (3.2 equiv.),
Bu3SnH (2.0 equiv.), AIBN (0.1 equiv.), C6H6, 80 ◦C, 9 h (66%).20
Scheme 13 Reagents and conditions: (a) LiAlD4 (2.5 equiv.), THF, 0 ◦C
then 67 ◦C, 14 h, then HCl, H2O (62%); (b) 2-iodobenzyl bromide (1.1
equiv.), iPrNEt2 (4.0 equiv.), CH3CN, rt, 14 h (53%).
19 was then subjected to reduction reactions using 1.3 equiv.
of both tributyltin hydride (Scheme 14) and tributyltin deuteride
(Scheme 15), at three different temperatures (80, 25 and -50 ◦C),
in order to estimate the extent to which the primary kinetic
isotope effect could determine the regioselectivity of the hydrogen
atom transfer. All reactions were carried out at our standardised
substrate concentration of 50–60 mM, with benzene as solvent for
the reactions at 80 and 25 ◦C and fluorobenzene at -50 ◦C. With
AIBN as initiator, thermal decomposition at 80 ◦C was replaced
by photochemical radical generation at the lower temperatures,
using a medium pressure mercury vapour lamp. Chromatographic
purification of the reduced products 24–27 was carried out on silica
gel containing potassium fluoride (10% w/w),20 which was found
to significantly improve the isolated yields when compared with the
tributyltin deuteride reduction of 18, and this purification method
was also used very successfully in later experiments. Table 1
summarises the results of these studies.‡
Scheme 11 1,5-Hydrogen vs. deuterium atom transfer.
Having found in earlier studies with 1,3-oxazolidine substrates
that the rate of 1,5-hydrogen atom transfer was sufficiently high
that slow addition of tributyltin hydride and the radical initiator
was not necessary,9 we firstly confirmed that this was also the case
in the pyrrolidine system 18 by carrying out its reduction with
tributyltin deuteride. (Scheme 12; the reduction was carried out
with 2.0 equiv. of tributyltin deuteride at a substrate concentration
of 58 mM.)
Scheme 14 Reagents and conditions: Bu3SnH (1.3 equiv.), AIBN (0.7–2.4
equiv.), C6H6 or C6H5F, heat or UV irradiation, 80, 25 or -50 ◦C, 6–12 h.
Scheme 12 Reagents and conditions: Bu3SnD (2.0 equiv.), AIBN (2.3
equiv.), C6H6, 80 ◦C, 6 h (23%; 20 : 21, 1 : 19).
As is often the case, the repeated chromatography required
for removal of the organotin residues from the reaction products
resulted in a disappointing (23%) final isolated yield of the 20/21
product mixture. Fortunately, the product ratios could also be
2
verified if required using H NMR spectroscopy on the crude
reaction product, the 1 : 19 ratio of 20 : 21 being confirmed before
and after chromatography.‡ With 95% of the isolable reaction
product originating from 1,5-hydrogen atom transfer, the high
efficiency of this process was confirmed and similar conditions
were used for all future radical reactions.
Scheme 15 Reagents and conditions: Bu3SnD (1.3 equiv.), AIBN (2.1–2.4
equiv.), C6H6 or C6H5F, heat or UV irradiation, 80, 25 or -50 ◦C, 6–12 h.
The required di-deuterated substrate 19 was prepared using 2,2-
dideuteropyrrolidine, isolated as its hydrochloride salt 22 from
a lithium aluminium deuteride reduction of 2-pyrrolidinone 23.
After reaction with 2-iodobenzyl bromide in the presence of an
The ambiguity in terms of the possible origins of compounds
24 (a product of direct reduction and/or 1,5-hydrogen atom
transfer?) and 27 (a product of direct reduction and/or 1,5-
deuterium atom transfer?) of course does not allow an accurate
estimation of the actual kH/kD values from these experiments.
The general preference for hydrogen atom transfer over deuterium
atom transfer was, however, clear, as was the apparent increased
1
2
‡ Product ratios were determined by a combination of H and H NMR
spectroscopy and mass spectrometry as appropriate.
4656 | Org. Biomol. Chem., 2010, 8, 4653–4665
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Table 1 Product ratios and yields for reduction of 19‡
Table 2 Product ratios and yields for acrylonitrile trapping experiments‡
Entry
Bu3SnH/D
T/◦C
Products
Ratio
Yield (%)
Entry
T/◦C
Ratio 29 : 30
Yield (%)
1
2
3
4
5
6
H
H
H
D
D
D
80
25
-50
80
25
-50
24, 25
24, 25
24, 25
26, 27
26, 27
26, 27
3 : 1
55
58
66
58
71
55
1
2
3
80
25
-50
3.2 : 1
4.5 : 1
5.7 : 1
15
23
56
12 : 1
49 : 1
1 : 2
7 : 1
7 : 1
To further extend these studies to include intramolecular a-
aminoalkyl radical trapping, an a-dideuterated analogue 31 of
the radical precursor used by Robertson et al. for the synthesis of
( )-heliotridane22 was then prepared. This substrate contains an
appropriately positioned vinyl group for a 5-exo-trig cyclisation to
follow 1,5-hydrogen (or deuterium) atom transfer to intermediate
vinyl radical 32 (Scheme 18).
selectivity for hydrogen transfer as the reaction temperature was
lowered.
To investigate the degree to which deuterium “protection” of
a carbon–hydrogen bond can be used to direct hydrogen atom
transfer in a synthetically valuable carbon–carbon bond forming
procedure, we chose the a-alkylation of the pyrrolidine moiety19 in
radical precursor 18 as our starting point. Acrylonitrile was chosen
as the radicalphile on the basis of its high relative rate for trapping
of alkyl radicals compared with other electron-deficient alkenes
such as acrylate esters.21 At a substrate concentration of 58 mM,
18 was reduced with tributyltin hydride in the presence of a 4.9-
fold excess of acrylonitrile, with thermal AIBN initiation at 80 ◦C
(Scheme 16). The a-alkylated derivative 28 was isolated as the
only identifiable pyrrolidine-containing product, in a synthetically
viable 56% yield.
Scheme 16 Reagents and conditions: Bu3SnH (2.0 equiv.), acrylonitrile
(4.9 equiv.), AIBN (2.1 equiv.), C6H6, 80 ◦C, 6 h (56%).
The same methodology was then transferred to di-deuterated
substrate 19. Again the temperatures chosen were 80, 25 and -50
◦C, with a change of solvent to fluorobenzene for the -50 ◦C
experiment and photochemical AIBN initiation at both 25 and
-50 ◦C (Scheme 17).
Scheme 18 Intramolecular radical trapping after 1,5-hydrogen or deu-
terium transfer.
31 was prepared by the N-alkylation of 2,2-dideuteropyrrolidine
hydrochloride 22 with 3-bromobut-3-enyl methanesulfonate 3323
(prepared from 3-bromobut-3-en-1-ol 3424), in the presence of
potassium carbonate (Scheme 19).
Scheme 17 Reagents and conditions: Bu3SnH (2.0 equiv.), acrylonitrile
(5.0–5.8 equiv.), AIBN (2.0–2.4 equiv.), C6H6 or C6H5F, heat or UV
irradiation, 80, 25 or -50 ◦C, 6 h.
◦
The results, summarised in Table 2, show a clear trend, with
the maximum primary kinetic isotope effect of kH/kD = 5.7 being
observed at -50 ◦C (entry 3) and the value dropping to 3.2 at 80 ◦C
(entry 1).‡ The deviation from the normally expected maximum
value of kH/kD ~ 7–8 is probably best rationalised by invoking a
later, more “product-like” transition state for the hydrogen transfer
process, stabilised by the adjacent nitrogen atom.
At 80 C, with an initial substrate concentration of 12 mM in
benzene, 31 was reduced with a solution of tributyltin hydride
and AIBN (also in benzene) added slowly over 1 h. Thiophenol
was added to aid product purification22 and the two di-deuterated
products 35 and 36 were isolated in a 5.7 : 1 ratio as their hy-
drobromide salts (30% overall yield, Scheme 20).‡ Unfortunately,
attempts to carry out the reaction at lower temperatures with
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 4653–4665 | 4657
©

View Article Online
Experimental
General experimental
Melting points were determined with a Gallenkamp MPD350
apparatus and are uncorrected.
Infrared spectra were recorded using a Nicolet Magna 550 spec-
trometer. Generally, only major absorbances are quoted, using the
abbreviations: (for intensity) w, weak; m, medium; and s, strong,
with br indicating a broad peak. Thin film samples were produced
by evaporation of a dilute chloroform or dichloromethane solution
of the sample on a sodium chloride plate.
Scheme 19 Reagents and conditions: (a) CH3SO2Cl (1.2 equiv.), Et3N (1.0
equiv.), CH2Cl2, rt, 1 h (90%); (b) 22 (1.2 equiv.), K2CO3 (3.6 equiv.),
CH3CN, 82 ◦C, 24 h (28%).
1H NMR spectra were recorded at 300 and 400 MHz using
Bruker ACF300 and Avance DRX400 spectrometers, respectively.
Chemical shifts are quoted in ppm relative to tetramethylsilane, the
residual solvent peak being used for referencing purposes where
possible. Coupling constants are quoted to the nearest 0.5 Hz with
peak multiplicities for single resonances being labelled as: s, singlet;
d, doublet; t, triplet; q, quartet; m, unresolved multiplet; AB, AB
pattern; AA¢BB¢, AA¢BB¢ pattern and br, broad. Where delin-
eation of individual resonances is not possible, the relevant peaks
are grouped together, over a chemical shift range, as complex.
13C NMR spectra were recorded at 75 and 100 MHz using
Bruker ACF300 and Avance DRX400 spectrometers, respectively.
Chemical shifts are quoted in ppm relative to tetramethylsilane, the
solvent peak being used for referencing. Where necessary, DEPT
editing was used for assignment purposes.
Scheme 20 Reagents and conditions: (a) Bu3SnH (2.7 equiv.), AIBN (0.2
equiv.), C6H6, 80 ◦C, 3 h; (b) PhSH (3.2 equiv.) (30% over 2 steps).
photochemical initiation did not give any of the desired products,
but the result obtained at 80 ◦C shows a significant preference
for hydrogen rather than deuterium transfer to intermediate vinyl
radical 32 with an apparent value of kH/kD = 5.7.
A small quantity of the non-cyclised reduction product 37
(Fig. 5) was also produced (9% yield) during this reaction and
this could not be separated from the 35/36 mixture.
2H NMR spectra were recorded at 61.4 MHz using a Bruker
DRX400 spectrometer. Where possible, the natural abundance
deuterium resonance from the protonated solvent was used for
referencing.
Mass spectra were obtained using Thermoquest Finnigan
Trace 2000 GC-MS and Micromass GCT instruments or by the
EPSRC National Mass Spectrometry Service Centre, Swansea,
UK. Ionisation modes are labelled as: EI, electron impact; CI,
chemical ionisation and ES+, positive ion electrospray.
Fig. 5 1-(But-3-enyl)-2,2-dideuteropyrrolidine hydrobromide.
X-Ray crystallographic data were recorded on a Nonius Kap-
paCCD diffractometer driven by COLLECT25 and DENZO26
software. The structure was determined using the direct methods
procedure in SHELXS-97 and refined by full-matrix least squares
on F2 using SHELXL-97.27 Full details of data collection and
structure determination have been deposited with the Cambridge
Crystallographic Data Centre with deposition number CCDC
291244.†
Analytical thin layer chromatography was carried out using
Merck Kieselgel 60 F254, coated on aluminium plates, with
visualisation of spots by quenching of UV (254 nm) fluorescence or
staining with iodine or alkaline potassium permanganate solution
as appropriate. Silica gel with particle size 40–63 mm was used for
flash chromatography.
Conclusions
Successful implementation of 1,5-hydrogen atom transfer in the
remote activation of carbon centres in synthetic methodologies
involving radical intermediates, requires high regioselectivity. We
have shown that for the generation of a-aminoalkyl radicals by
such a process, deuterium can be used to block this atom transfer
to a degree that could potentially be synthetically useful. The
estimated values of kH/kD observed in these experiments suggest
the likelihood of a later, product radical-like transition state for
this process.
Solvents and reagents were used as commercially supplied or,
when necessary, purified using standard procedures detailed in
Purification of Laboratory Chemicals, 3rd edn, W. L. F. Armarego
and D. D. Perrin, Pergamon Press, Oxford, 1998. The fraction of
light petroleum ether boiling in the range 40 to 60 C is referred
to as “petroleum ether.”
A Bu¨chi R110 Rotovapor was used for the removal of solvents
under reduced pressure, with a water or dry ice condenser being
used as appropriate.
In order to use such a “protecting group” strategy, based on
the primary kinetic isotope effect, either (i) the products which
would ultimately arise from the different hydrogen atom transfer
processes would need to be different in structure and, therefore,
separable; (ii) the blocking deuterium atom(s) would need to be
incorporated into a removable protecting group (as exemplified by
the earlier studies of Clive et al.13); or (iii) the deuterium atoms
would need to be readily exchangeable with hydrogen if isotopic
labelling of the final product were not desired.
◦
4658 | Org. Biomol. Chem., 2010, 8, 4653–4665
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
3-(2-Iodobenzyl)-2-(pyridin-4-yl)oxazolidine (2)
2774 (m) (C–H), 1653 (s), 1601 (s), 1558 (m), 1497 (w), 1412 (s),
1367 (s), 1265 (s), 1217 (s), 1094 (m), 1080 (m), 1065 (m), 995
(w), 976 (w), 945 (s), 905 (w), 837 (m), 754 (s), 676 (s) and 667
(s); dH (400 MHz; CDCl3) 4.09 (2H, t, J = 9.5, CH2N), 4.46
(2H, t, J = 9.5, CH2O) and 7.77, 8.70 (2 ¥ 2H, AA¢BB¢, J = 6,
pyridineCH); dC (100 MHz; CDCl3) 55.1 (CH2N), 68.0 (CH2O),
121.9 (pyridineCH), 135.1 (pyridine ipsoC), 150.2 (pyridineCH)
and 163.0 (N = C–O); m/z (EI) 148 (M+, 37%), 118 (100), 91 (32),
4-Pyridinecarboxaldehyde 4 (610 ml, 6.4 mmol) was added to a
stirred solution of 2-(2-iodobenzylamino)ethanol9 3 (2.00 g, 7.2
mmol) in tetrahydrofuran (50 cm3) containing anhydrous mag-
nesium sulfate (2.00 g, 16.6 mmol), under a nitrogen atmosphere.
After stirring for 18 h, the reaction mixture was filtered, the solvent
was evaporated in vacuo and the residue dissolved in ethyl acetate
(30 cm3). The resulting solution was washed with aqueous sodium
metabisulfite solution (20% w/v, 5 ¥ 10 cm3), dried (magnesium
sulfate), filtered and evaporated in vacuo to give 2 (2.25 g, 85%)
as a yellow oil (Found MH+ (CI) 367.0299, C15H16IN2O requires
367.0307); nmax(thin film)/cm-1 3112–2767 (m) (C–H), 1599 (s),
1562 (w), 1466 (m), 1437 (m), 1410 (s), 1385 (w), 1342 (m), 1317
(m), 1234 (m), 1200 (w), 1159 (w), 1124 (m), 1101 (m), 1063 (s),
1036 (s), 1013 (s), 993 (m), 904 (w), 814 (m), 784 (m), 752 (s) and
638 (m); dH (400 MHz; CDCl3) 2.88 (1H, m, NCHHCH2), 3.08
(1H, m, NCHHCH2), 3.76, 3.86 (2 ¥ 1H, AB, J = 14, ArCH2),
3.96 (1H, m, CHHO), 4.07 (1H, m, CHHO), 5.27 (1H, s, NCHO),
6.95 (1H, dt, J = 1.5 and 7.5, arylCH), 7.31 (1H, dt, J = 1 and 7.5,
arylCH), 7.41–7.47 (3H, complex, arylCH and 2 x pyridineCH),
7.83 (1H, dd, J = 1 and 8, arylCH) and 8.56 (2H, part of AA¢BB¢,
J = 6, pyridineCH); dC (100 MHz; CDCl3) 51.2 (NCH2CH2),
61.7 (CH2O), 64.6 (ArCH2), 95.4 (NCHO), 100.1 (arylCI), 122.2
(pyridineCH), 128.2, 129.1, 130.2, 139.7 (arylCH), 140.5 (aryl
ipsoC) and 149.8 (pyridineCH); m/z (EI) 366 (M+, 3%), 288 (100),
217 (85), 149 (12), 135 (19), 119 (25), 91 (68), 90 (66), 89 (43), 78
(32, 77 (22), 65 (33), 63 (25), 51 (40) and 39 (26); m/z (CI) 367
(MH+, 11%), 241 (21), 149 (100), 134 (12), 108 (31), 107 (23), 106
(17), 94 (43), 80 (19) and 72 (9).
+
84 (35), 78 (44), 51 (60), 49 (53) and 40 (35); m/z (CI) 166 (MNH4 ,
8%) and 149 (MH+, 100).
Crystal structure determination for compound 5
˚
Crystal data at 120(2) K with Mo-Ka (l = 0.71073 A). 5: M =
478.58, monoclinic, P21/c, a◦= 8.9276(2), b = 14.35482(2), c =
3
˚
˚
10.1033(4) A, b = 109.570(2) , V = 1236.42(7) A , Z = 2, 8639
measured reflections, 2801 unique reflections (Rint = 0.0467), R =
0.0412, wR = 0.1158.
2-(Pyridin-4-yl)-4,5-dihydrooxazole (6)
tert-Butyl acrylate (240 ml, 1.64 mmol) was added to a stirred so-
lution of 3-(2-iodobenzyl)-2-(pyridin-4-yl)oxazolidine 2 (200 mg,
0.55 mmol) in degassed benzene (10 cm3) and the resulting
solution was heated to reflux under a nitrogen atmosphere.
A solution of tributyltin hydride (510 ml, 1.90 mmol) and
2,2¢-azobisisobutyronitrile (44.8 mg, 0.27 mmol) in degassed
benzene (10 cm3) was added dropwise (using a syringe pump)
over a period of 5 h, after which time heating under reflux
was continued for a further 18 h. The solvent was removed
in vacuo and the resulting residue was purified by flash chromatog-
raphy on silica gel (ethyl acetate) to give 6 (47.6 mg, 59%) as an
orange oil. Data as reported above.
3,3¢-Dibenzyl-2,2¢-di(pyridin-4-yl)-2,2¢-bioxazolidine (5)
Tributyltin hydride (840 ml, 3.1 mmol) was added to a stirred,
degassed solution of 3-(2-iodobenzyl)-2-(pyridin-4-yl)oxazolidine
2 (275.6 mg, 0.75 mmol) and 2,2¢-azobisisobutyronitrile (12.0 mg,
73.1 mmol) in benzene (50 cm3). After stirring under reflux under a
nitrogen atmosphere for 18 h, the solvent was removed in vacuo and
the crude product was purified by flash chromatography on silica
gel (50% petroleum ether–50% ethyl acetate) to give 5 (15.1 mg,
8%) as a pale yellow, crystalline solid (Found MH+ (CI) 479.2439,
C30H31N4O2 requires 479.2447); nmax(thin film)/cm-1 3127–2783
(m) (C–H), 1734 (m), 1635 (s), 1549 (w), 1497 (w), 1456 (m), 1426
(m), 1410 (m), 1369 (w), 1281 (s), 1217 (w), 1124 (w), 1065 (m),
833 (w), 756 (s), 700 (m) and 667 (m); dH (400 MHz; CDCl3) 2.14
(2H, m, 2 ¥ NCHHCH2), 2.68 (2H, 2 ¥ CHHO), 3.18, 4.84 (2
¥ 2H, AB, J = 13, 2 ¥ PhCH2), 3.78 (2H, m, NCHHCH2), 4.07
(2H, m, CHH2O), 6.83–6.88 (4H, complex, phenylCH), 7.13–7.18
(6H, complex, phenylCH) and 7.76, 8.71 (2 ¥ 4H, AA¢BB¢, J =
6, pyridineCH); dC (100 MHz; CDCl3) 48.3 (NCH2CH2), 53.6
(CH2O), 66.1 (PhCH2), 99.2 (NCO), 124.3 (pyridineCH), 126.8,
127.7, 128.2 (phenylCH), 139.2 phenyl ipsoC), 148.0 (pyridine
ipsoC) and 149.0 (pyridineCH); m/z (EI) 239 (3%), 162 (18), 148
(14), 118 (27), 106 (21), 91 (100), 84 (24), 78 (30), 65 (19), 51
(39), 49 (41) and 41 (23); m/z (CI) 479 (MH+, 1%), 241 (23), 239
(15), 149 (100), 134 (18), 108 (18), 106 (17), (94 (15) and 80 (25).
A second fraction contained 2-(pyridin-4-yl)-4,5-dihydrooxazole
6 (3.8 mg, 3%), obtained as an orange oil (Found MH+ (ES+)
149.0716, C8H9N2O requires 149.0715); nmax(thin film)/cm-1 3128–
2,5-Dioxopyrrolidin-1-yl isonicotinate (15)14
N-(3-Dimethylaminopropyl)-N¢-ethylcarbodiimide
hydrochloride (5.00 g, 26.1 mmol) was added to a stirred
solution of isonicotinic acid 14 (3.21 g, 26.1 mmol) in
N,N-dimethylformamide (60 cm3) at 0 ◦C, followed by N-
hydroxysuccinimide (3.00 g, 26.1 mmol). After stirring at this
temperature for 2 h, the reaction mixture was poured into
ethyl acetate (70 cm3) and the resulting solution was washed
successively with water (30 cm3), saturated aqueous sodium
bicarbonate solution (30 cm3) and saturated aqueous sodium
chloride solution (30 cm3). The separated organic phase was then
dried (magnesium sulfate), filtered and evaporated in vacuo to
give 15 (3.47 g, 60%) as a white, crystalline solid (Found MH+
◦
(ES+) 221.0560, C10H9N2O4 requires 221.0562); mp 115–120 C;
n
max(KBr disc)/cm-1 3331–2806 (s) (C–H), 1782 (m) (C O), 1716
(s) (C O), 1614 (w), 1521 (w), 1475 (w), 1412 (m), 1336 (w),
1306 (m), 1215 (s), 1078 (s), 1027 (s), 944 (m), 930 (m) and 856
(m); dH (400 MHz; CDCl3) 2.93 (4H, s, CH2CH2) and 7.94, 8.88
(2 ¥ 2H, AA¢BB¢, J = 6, pyridineCH); dC (100 MHz; CDCl3) 25.7
(CH2CH2), 123.2 (pyridineCH), 132.6 (pyridine ipsoC), 151.0
(pyridineCH), 160.8 (ester C = O) and 168.6 (imide C = O); m/z
(EI) 220 (MH+, 1%), 106 (100), 78 (79), 51 (63), 50 (33) and 42
(9); m/z (CI) 221 (MH+, 88%), 141 (12), 138 (51), 124 (100), 123
(32), 108 (8), 91 (16), 80 (41) and 74 (9).
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 4653–4665 | 4659
©

View Article Online
1,1-Dideutero(pyrid-4-yl)methanol (16)
a dark orange oil (Found MH+ (ES+) 368.0366, C15H15DIN2O
requires 368.0370); nmax(thin film)/cm-1 3119–2741 (m) (C–H),
1599 (s), 1560 (m), 1466 (m), 1437 (m), 1408 (m), 1273 (m), 1234
(m), 1157 (m), 1065 (s), 906 (w), 818 (m), 787 (m), 752 (s) and
621 (m); dH (400 MHz; CDCl3) 2.84–2.93 (1H, m, CHHN), 3.03
(1H, m, CHHN), 3.76, 3.86 (2 ¥ 1H, AB, J = 14, ArCH2), 3.96
(1H, ca q, J = 6, CHHO), 4.06 (1H, ca q, J = 6, CHHO), 6.95
(1H, dt, J = 2 and 7, arylCH), 7.31 (1H, t, J = 7, arylCH) and
7.40–8.92 (6H, complex, arylCH and pyridineCH); dD (61.4 MHz;
CCl4) 5.17 (s, CD); dC (100 MHz; CDCl3) 51.2 (NCH2CH2), 61.7
(CH2CH2O), 64.6 (ArCH2), 95.0 (m, CD), 100.1 (arylCI) 122.2,
128.2, 129.1, 130.2, 139.7 (arylCH and pyridineCH) 140.6 (aryl
ipsoCCH2), 149.0 (pyridine ipsoC) and 149.8 (pyridineCH); m/z
(EI) 367 (M+, 4%), 289 (100), 217 (71), 120 (19), 91 (31), 90 (46),
89 (30) and 51 (22); m/z (CI) 368 (MH+, 84%), 242 (43) and 149
(100).
2,5-Dioxopyrrolidin-1-yl isonicotinate 15 (1.00 g, 4.6 mmol) was
added to a stirred solution of sodium borodeuteride (98 atom%
D, 760 mg, 18.2 mmol) in ethanol at room temperature and
stirring was continued at this temperature for 18 h. The solvent
was removed in vacuo and the crude product was partitioned
between water (40 cm3) and ethyl acetate (30 cm3), the separated
aqueous phase being extracted with ethyl acetate (2 ¥ 30 cm3). The
combined organic extracts were dried (magnesium sulfate), filtered
and evaporated in vacuo to give 16 (500 mg, 98%) as a yellow
oil (Found MH+ (CI) 112.0729, C6H6D2NO requires 112.0731);
n
max(thin film)/cm-1 3315 (br, m) (O–H), 3086–2806 (m) (C–H),
1608 (m), 1417 (s), 1313 (w), 1216 (s), 1113 (w), 1061 (w), 1006
(w), 921 (w), 758 (s) and 688 (s); dH (400 MHz; CDCl3) 2.72 (1H,
br s, OH) and 7.29, 8.51 (2 ¥ 2H, AA¢BB¢, J = 6, pyridineCH); dD
(61.4 MHz; CCl4) 4.62 (s, CD2OH); dC (100 MHz; CDCl3) 121.2
(pyridineCH), 149.5 (pyridineCH) and 150.4 (pyridine ipsoC);
m/z (EI) 111 (M+, 100%), 110 (92), 109 (31), 82 (56), 81 (94), 51
tert-Butyl-3-(3-benzyl-2-deutero-2-(pyridin-4-yl)oxazolidin-4-
yl)propanoate (17)
(53) and 40 (19); m/z (CI) 129 (MNH4 , 3%), 112 (MH+, 100), 96
+
(66), 95 (14) and 80 (7).
Tributyltin hydride (510 ml, 1.90 mmol) was added to a stirred solu-
tion of 3-(2-iodobenzyl)-2-deutero-2-(pyridin-4-yl)oxazolidine 12
(200 mg, 0.55 mmol) and tert-butyl acrylate (240 ml, 1.64 mmol)
in benzene (40 cm3), under a nitrogen atmosphere. Triethylborane
(1 M solution in tetrahydrofuran, 100 ml, 0.10 mmol) was added
and the reaction mixture was stirred at room temperature for 18 h.
The solvent was removed in vacuo and the resulting residue was
purified by flash chromatography on silica gel (40% petroleum
ether–60% ethyl acetate) to give 17 as a 6.5 : 1 ratio of insepa-
rable diastereoisomers (26 mg, 13%) as a yellow syrup (Found
MH+ (ES+) 370.2250, C22H28DN2O3 requires 370.2241); nmax(thin
film)/cm-1 3092–2773 (m) (C–H), 1726 (s) (C O), 1593 (w), 1560
(w), 1456 (w), 1048 (w), 1367 (m), 1257 (m), 1153 (s), 1065 (w), 993
(w), 847 (w), 756 (s), 700 (w) and 667 (w); dH (400 MHz; CDCl3;
major diastereoisomer only) 1.41 (9H, s, (CH3)3C), 1.58 (2H, ca
a-Deutero-4-pyridinecarboxaldehyde (13)
A solution of dimethyl sulfoxide (570 ml, 8.0 mmol) in
dichloromethane (10 cm3) was added slowly, over a period of
5 min, to a stirred solution of oxalyl chloride (360 ml, 4.1 mmol) in
dichloromethane (10 cm3) at -60 ◦C, under a nitrogen atmosphere.
After stirring at this temperature for a further 2 min, a solution
of 1,1-dideutero(pyrid-4-yl)methanol 16 (410 mg, 3.7 mmol) in
dichloromethane (10 cm3) was added slowly and after a further
15 min of stirring, triethylamine (2.58 cm3, 18.5 mmol) was also
added slowly. Stirring was continued at -60 ◦C for a further
5 min and then the reaction mixture was allowed to attain room
temperature before the solvent was removed in vacuo. The crude
product was purified by flash chromatography on silica gel (40%
dichloromethane–60% ethyl acetate) to give 13 (220 mg, 55%) as
an orange oil (Found MH+ (CI) 109.0510, C6H5DNO requires
109.0512; nmax(thin film)/cm-1 3388–2791 (m) (C–H), 1697 (s)
(C O), 1675 (s), 1603 (s), 1563 (s), 1410 (s), 1388 (s), 1326 (w),
1302 (w), 1256 (m), 1223 (s), 1207 (s), 1196 (s), 1105 (s), 1061
(s), 992 (w), 945 (w), 889 (w), 854 (w), 789 (s), 722 (w), 667 (w)
and 639 (s); dH (400 MHz; CDCl3) 7.78, 8.86 (2 ¥ 2H, AA¢BB¢,
J = 6, pyridineCH); dD (61.4 MHz; CCl4) 10.1 (s, C(O)D); dC
(100 MHz; CDCl3) 122.1 (pyridineCH), 141.4 (pyridine ipsoC),
151.2 (pyridineCH) and 191.1 (t, J = 30, C(O)D); m/z (EI) 108
(M+, 100%); 106 (43), 86 (19), 84 (32), 78 (50), 53 (40), 51 (94), 50
t
t
q, J = 7, CH2CH2CO2 Bu), 2.23 (2H, ca dt, J = 3, 7, CH2CO2 Bu),
3.20 (1H, m, NCH), 3.64 (1H, ca dd, J = 6, 8, CHHO), 3.85 (2H, s,
PhCH2), 4.15 (1H, ca dd, J = 7.5 and 8, CHHO) and 7.18–8.48
(9H, complex, phenylCH and pyridine CH); dC (100 MHz; CDCl3;
t
major diastereoisomer) 28.1 ((CH3)3C), 29.1 (CH2CH2CO2 Bu),
t
32.0 (CH2CO2 Bu), 58.7 (PhCH2), 63.2 (NCH), 70.6 (CH2O), 80.4
((CH3)3C), 127.5, 128.4, 129.2 (phenylCH), 138.1 (phenyl ipsoC)
and 172.5 (C = O) (note: NCD and pyridine carbon signals not
visible owing to dilute sample and line broadening, respectively);
dC (100 MHz; CDCl3; minor diastereoisomer) 28.1 ((CH3)3C),
t
t
29.7 (CH2CH2CO2 Bu), 32.6 (CH2CO2 Bu), 51.5 (PhCH2), 60.6
(NCH), 68.8 (CH2O) and 127.3, 128.4, 128.8 (phenylCH) (note:
other signals not visible owing to dilute sample); dD (61.4 MHz;
CCl4) 5.05 (0.87D, CD, major diastereoisomer) and 5.27 (0.13D,
CD, minor diastereoisomer); m/z (EI) 369 (M+, 1%), 240 (33), 91
(100) and 57 (34); m/z (CI) 370 (MH+, 58%), 190 (10), 106 (20),
95 (100) and 80 (35).
+
(52), 49 (70), 44 (42) and 43 (31); m/z (CI) 126 (MNH4 , 10%),
109 (MH+, 100), 95 (35) and 80 (10).
3-(2-Iodobenzyl)-2-deutero-2-(pyridin-4-yl)oxazolidine (12)
a-Deutero-4-pyridinecarboxaldehyde 13 (210 mg, 1.9 mmol) was
added to a stirred solution of 2-(2-iodobenzylamino)ethanol9
3 (440 mg, 1.6 mmol) in tetrahydrofuran (30 cm3), containing
anhydrous magnesium sulfate (2.00 g, 16.3 mmol), under a
nitrogen atmosphere. After stirring for 18 h at room temperature,
the reaction mixture was filtered, the solvent evaporated and the
residue was purified by flash chromatography on silica gel (40%
dichloromethane–60% ethyl acetate) to give 12 (520 mg, 89%) as
N-(2-Iodobenzyl)pyrrolidine 1819
2-Iodobenzyl bromide (1.75 g, 5.9 mmol) and N,N-
diisopropylethylamine (4.00 cm3, 23.0 mmol) were added to a
stirred solution of pyrrolidine (560 ml, 6.7 mmol) in acetonitrile
(16 cm3). After stirring for 48 h, the solvent was removed in vacuo
4660 | Org. Biomol. Chem., 2010, 8, 4653–4665
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
and the residue was partitioned between ethyl acetate (16 cm3)
and water (24 cm3). The separated organic phase was washed
with water (2 ¥ 24 cm3) and aqueous sodium chloride solution
(saturated, 24 cm3) before drying (magnesium sulfate), filtering and
evaporation of the solvent in vacuo. The residue was purified by
flash chromatography on silica gel (gradient from 95% petroleum
ether–5% ethyl acetate to 80% petroleum ether–20% ethyl acetate)
to give 18 (1.38 g, 82%) as a colourless oil (Found MH+ (CI)
288.0244, C11H15IN requires 288.0244); nmax(thin film)/cm-1 3086–
2644 (s) (C–H), 1697 (w), 1584 (w), 1459 (m), 1436 (m), 1374 (w),
1348 (m), 1326 (w), 1239 (w), 1201 (w), 1147 (w), 1129 (w) 1012 (s),
881 (w), 749 (s) and 650 (w); dH (300 MHz; CDCl3) 1.81 (4H, m,
CH2CH2CH2CH2), 2.60 (4H, m, N(CH2)2), 3.70 (2H, s, ArCH2),
6.93 (1H, ca dt, J = 3.0, 7.5, arylCH), 7.32 (1H, ca dt, J = 1.5, 6.0,
arylCH), 7.47 (1H, ca dd, J = 3.0, 6.0, arylCH) and 7.82 (1H, dd, J
= 1.5, 7.5, arylCH); dC (75 MHz; CDCl3) 23.6 (CH2CH2CH2CH2),
54.1 (N(CH2)2), 64.4 (ArCH2), 100.1 (arylCI), 128.0, 128.4, 129.8,
139.2 (arylCH) and 141.6 (aryl ipsoC); m/z (CI) 288 (MH+, 73%),
162 (100), 70 (75) and 52 (23). All data consistent with literature
values.19
ing solution was extracted with dichloromethane (4 ¥ 10 cm3) and
the combined extracts were washed with hydrochloric acid (2 M, 3
¥ 30 cm3). The combined aqueous extracts were evaporated to dry-
ness in vacuo to give 22 (0.80 g, 62%) as a pale brown, hygroscopic
solid (Found M–HCl (EI) 73.0864, C4H7D2N requires 72.0860);
n
max(thin film)/cm-1 3600–3200 (s) (N–H), 3138–2488 (s) (C–H, C–
D), 1664 (m), 1460 (w), 1304 (w), 1129 (w), 1080 (w) and 1030 (w);
dH (300 MHz; CDCl3) 1.95–2.13 (4H, complex, NCH2CH2CH2),
3.29 (2H, br m, NCH2) and 9.63 (2H, br s, +NH2); dC (75 MHz;
CDCl3) 24.6 (NCD2CH2), 24.4 (NCH2CH2), 44.5 (CD2) and 45.0
(NCH2); dD (61.4 MHz; H2O) (s, 3.18 CD2); m/z (EI) 73 (M–HCl,
100%).
N-(2-Iodobenzyl)-2,2-dideuteropyrrolidine (19)
2-Iodobenzyl bromide (2.10 g, 7.1 mmol) and N,N-
diisopropylethylamine (4.44 cm3, 25.5 mmol) were added to
a stirred solution of 2,2-dideuteropyrrolidine hydrochloride 22
(700 mg, 6.4 mmol) in acetonitrile (20 cm3). After stirring for
14 h, the solvent was removed in vacuo and the residue was
partitioned between ethyl acetate (20 cm3) and water (20 cm3) and
the separated organic phase was washed with water (2 ¥ 20 cm3)
and aqueous sodium chloride solution (saturated, 20 cm3). After
drying (magnesium sulfate), filtering and evaporation of the ethyl
acetate in vacuo, the residue was purified by flash chromatography
on silica gel (gradient from 95% petroleum ether–5% ethyl acetate
to 80% petroleum ether–20% ethyl acetate) to give 19 (970 mg,
53%) as a yellow oil (Found MH+ (CI) 290.0370, C11H13D2NI
requires 290.0369); nmax(thin film)/cm-1 3120–2642 (m) (C–H and
C–D), 1583 (w), 1562 (m), 1460 (m), 1436 (m), 1363 (m), 1328
(m), 1305 (m), 1229 (w), 1205 (w), 1170 (m), 1134 (m), 1116 (m),
1044 (w), 1012 (s), 749 (s) and 651 (w); dH (300 MHz; CDCl3)
1.82 (4H, complex, CH2CH2CD2), 2.58 (2H, m, NCH2CH2), 3.67
(2H, s, ArCH2), 6.94 (1H, ca dt, J = 3.0, 7.5, arylCH), 7.31 (1H,
ca dt, J = 1.5, 6.0, arylCH), 7.44 (1H, ca dd, J = 3.0, 6.0, arylCH)
and 7.82 (1H, dd, J = 1.5, 7.5, arylCH); dC (75 MHz; CDCl3) 23.4,
23.6 (CH2CH2CD2), 53.3 (CD2), 54.1 (NCH2CH2), 64.4 (ArCH2),
100.1 (arylCI), 128.0, 128.4, 129.8, 139.2 (arylCH) and 141.7 (aryl
ipsoC); dD (61.4 MHz; H2O) (s, 3.18 CD2); 2.65 (s, CD2); m/z
(EI) 289 (M+, 23%), 217 (22), 162 (12), 134 (12), 127 (10), 105
(10), 92 (30), 91 (50), 86 (100), 72 (18), 63 (15), 57 (12) and 44
(59); m/z (CI) 290 (MH+, 81%), 164 (85), 72 (100), 52 (61) and
44 (56).
N-Benzyl-2-deuteropyrrolidine (20) and
N-(2-deuterobenzyl)pyrrolidine (21)
Tributyltin deuteride (98 atom% D, 508 mg, 1.74 mmol) and 2,2¢-
azobisisobutyronitrile (81 mg, 0.49 mmol) were added to a stirred,
degassed solution of N-(2-iodobenzyl)pyrrolidine 18 (250 mg,
0.87 mmol) in benzene (15 cm3), under a nitrogen atmosphere.
The reaction mixture was heated to reflux and three further
aliquots of 2,2¢-azobisisobutyronitrile (81 mg, 0.49 mmol) were
added at 1, 3.5 and 5 h. Reflux was continued for a further
1 h, after which time, the solvent was removed in vacuo and
the residue was purified by flash chromatography on silica gel
containing potassium fluoride (20% w/w) (gradient from 100%
dichloromethane to 80% dichloromethane–20% ethyl acetate) to
give a mixture of 20 and 21 (65 mg, 23%) in a 19 : 1 ratio as a
colourless oil (Found MH+ (ES+) 163.1340, C11H15DN requires
163.1340); nmax(thin film)/cm-1 3073–2503 (s) (C–H, C–D), 1667
(w), 1496 (w), 1456 (m), 1214 (w), 754 (w) and 701 (m); dH
(300 MHz; CDCl3) 1.96–2.05 (4H, complex, CHDCH2CH2),
2.96–3.06 (3H, complex, CH2NCHD), 4.00 (2H, s, PhCH2), 7.36–
7.42 (3H, complex, phenylCH) and 7.53–7.61 (2H, complex,
phenylCH); dC (75 MHz; CDCl3) 23.1, 23.2 (CHDCH2CH2),
53.00 (CHD), 53.3 (NCH2CH2), 129.0, 130.2 (phenylCH) and
132.8 (phenyl ipsoC); dD (61.4 MHz; CH2Cl2) 2.87 (0.95D, s,
NCHD) and 7.52 (0.05D, s, phenylCD) m/z (EI) 162 (M+, 20%),
161 ((M–H)+, 45), 92 (17), 91 (100), 85 (49), 77 (15), 71 (30), 65
(37), 43 (42) and 42 (31); m/z (CI) 163 (MH+, 100%), 108 (4), 91
(5), 85 (4), 73 (15) and 71 (6).
N-Benzyl-2,2-dideuteropyrrolidine (24) and
N-(2-deuterobenzyl)-2-deuteropyrrolidine (25)
Reaction at 80 ◦C. Tributyltin hydride (240 ml, 0.89 mmol)
was added to a degassed solution of N-(2-iodobenzyl)-2,2-
dideuteropyrrolidine 19 (200 mg, 0.69 mmol) in benzene (15 cm3)
under a nitrogen atmosphere. The reaction mixture was heated
to reflux and 2,2¢-azobisisobutyronitrile (20 mg, 0.12 mmol) was
added. Reflux was continued for a further 12 h, during which
time, three further aliquots of 2,2¢-azobisisobutyronitrile (10 mg,
0.12 mmol) were added after 1.5, 3 and 10 h. The solvent
was removed in vacuo and the residue was purified by flash
chromatography on silica gel containing potassium fluoride (10%
w/w) (gradient from 100% petroleum ether to 80% petroleum
ether–20% ethyl acetate) to give a mixture of 24 and 25 in a
2,2-Dideuteropyrrolidine hydrochloride (22)
Lithium aluminium deuteride (98 atom% D, 1.23 g, 29.3 mmol)
was added to a stirred solution of 2-pyrro◦lidinone 23 (1.00 g,
11.8 mmol) in tetrahydrofuran (15 cm3) at 0 C, under a nitrogen
atmosphere. The reaction mixture was heated at reflux for 14 h,
after which time the solution was cooled in an ice bath and
water (1.23 cm3), aqueous sodium hydroxide solution (15% w/v,
1.23 cm3) and water (3.7 cm3) were added sequentially. The result-
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 4653–4665 | 4661
©

View Article Online
3 : 1 ratio (62 mg, 55%) as a colourless oil; nmax(thin film)/cm-1
3500–3000 (s) (C–D), 2970–2900 (s) (C–H), 1642 (m), and 1475
(m); dH (300 MHz; CDCl3) 1.70 (4H, m, NCH2CH2CH2), 2.45
(2.25H, m, NCH2, NCHD), 3.62 (2H, s, PhCH2) and 7.13–
7.35 (4.75H, complex, phenylCH); dC (75 MHz; CDCl3) 22.3,
22.5 (NCH2CH2CH2), 58.3 (NCH2CH2), 65.0 (PhCH2), 128.2,
128.8 (phenylCH), 131.7 (phenyl ipsoC) and 132.6 (phenylCH);
dD (61.4 MHz; CHCl3) 2.53 (1.73D, s, NCD2, NCHD) and 7.40
(0.27D, s, phenylCD); m/z (EI) 163 (M+, 54%), 162 (72), 161 (12),
92 (42), 91 (100), 86 (58), 85 (39), 72 (48) and 65 (58); m/z (CI)
164 (MH+, 100%), 163 (25) and 62 (14).
heated to reflux and 2,2¢-azobisisobutyronitrile (50 mg, 0.31 mmol)
was added. Reflux was continued for a further 12 h, during
which time three further aliquots of 2,2¢-azobisisobutyronitrile
(50 mg, 0.31 mmol) were added after 1, 3.5 and 10 h. The
solvent was removed in vacuo and the residue was purified by flash
chromatography on silica gel containing potassium fluoride (10%
w/w) (gradient from 100% petroleum ether to 80% petroleum
ether–20% ethyl acetate) to give a mixture of 26 and 27 in a
1 : 2 ratio (56 mg, 58%) as a colourless oil; nmax(thin film)/cm-1
3500–3000 (w) (C–D), 2990–2900 (s) (C–H), 1665 (m) and 1494
(m); dH (300 MHz; CDCl3) 1.73 (4H, NCHDCH2CH2), 2.43
(1.70H, m, NCHD, NCH2), 3.54 (2H, s, PhCH2) and 7.15–7.29
(4.30H, complex, phenylCH); dC (75 MHz; CDCl3) 23.7, 23.8
(NCH2CH2CH2), 54.2 (NCD2), 54.6 (NCHDCH2), 62.5 (PhCH2),
127.4, 128.5, 128.7, 129.4 (phenylCH), 139.5 (phenyl ipsoC); dD
(61.4 MHz; CHCl3) 2.45 (2.30D, s, NCD2, NCHD) and 7.28
(0.70D, s, phenylCD); m/z (EI) 165 (MH+, 20%), 164 (M+, 32),
163 ((M–H)+, 100), 162 (20), 136 (8), 135 (20), 134 (8), 92 (38), 91
(84), 86 (62), 73 (42), 72 (49) and 65 (50).
Reaction at 25 ◦C. Tributyltin hydride (300 ml, 1.12 mmol)
and 2,2¢-azobisisobutyronitrile (70 mg, 0.43 mmol) were
added to a stirred, degassed solution of N-(2-iodobenzyl)-2,2-
dideuteropyrrolidine 19 (250 mg, 0.86 mmol) in benzene (15 cm3),
under a nitrogen atmosphere at 25 ◦C. The reaction mixture was
irradiated with a medium pressure mercury vapour lamp for 6 h,
with three further aliquots of 2,2¢-azobisisobutyronitrile (70 mg,
0.43 mmol) being added after 1, 3.5 and 5 h. The solvent was
removed in vacuo and the residue was purified by flash chro-
matography on silica gel containing potassium fluoride (10% w/w)
(gradient from 100% petroleum ether to 80% petroleum ether–20%
ethyl acetate) to give a mixture of 24 and 25 in a 12 : 1 ratio (81 mg,
58%) as a colourless oil; nmax(thin film)/cm-1 as reported above;
Reaction at 25 ◦C. Tributyltin deuteride (98 atom% D, 250 ml,
0.93 mmol) and 2,2¢-azobisisobutyronitrile (70 mg, 0.43 mmol)
were added to a stirred, degassed solution of N-(2-iodobenzyl)-
2,2-dideuteropyrrolidine 19 (200 mg, 0.69 mmol) in benzene
(15 cm3), under a nitrogen atmosphere at 25 ◦C. The reaction mix-
ture was irradiated with a medium pressure mercury vapour lamp
for 6 h, with three further aliquots of 2,2¢-azobisisobutyronitrile
(70 mg, 0.43 mmol) being added after 1, 3.5 and 5 h. The solvent
was removed in vacuo and the residue was purified by flash chro-
matography on silica gel containing potassium fluoride (10% w/w)
(gradient from 100% petroleum ether to 80% petroleum ether–20%
ethyl acetate) to give a mixture of 26 and 27 in a 7 : 1 ratio (81 mg,
71%) as a colourless oil; nmax(thin film)/cm-1 as reported above;
d
H (300 MHz; CDCl3) 1.73 (4H, m, NCH2CH2CH2), 2.45 (2.07H,
m, NCH2, NCHD), 3.55 (2H, s, PhCH2) and 7.15–7.28 (4.93H,
complex, phenylCH); dC (75 MHz; CDCl3) as reported above;
dD (61.4 MHz; CHCl3) 2.85 (1.92D, s, NCD2, NCHD) and 7.65
(0.08D, s, phenylCD); m/z (EI) 162 ((M–H)+, 100%), 134 (29), 92
(8), 91 (73), 85 (68), 71 (85) and 65 (66).
Reaction at -50 ◦C. Tributyltin hydride (250 ml, 0.93 mmol)
and 2,2¢-azobisisobutyronitrile (70 mg, 0.43 mmol) were
added to a stirred, degassed solution of N-(2-iodobenzyl)-2,2-
dideuteropyrrolidine 19 (210 mg, 0.73 mmol) in fluorobenzene
(12 cm3), under a nitrogen atmosphere, at -50 ◦C. The reac-
tion mixture was irradiated with a medium pressure mercury
vapour lamp for 6 h, with three further aliquots of 2,2¢-
azobisisobutyronitrile (70 mg, 0.43 mmol) being added after 1,
3.5 and 5 h. The solvent was removed in vacuo and the residue
was purified by flash chromatography on silica gel containing
potassium fluoride (10% w/w) (gradient from 100% petroleum
ether to 80% petroleum ether–20% ethyl acetate) to give a mixture
of 24 and 25 in a 49 : 1 ratio (78 mg, 66%) as a colourless oil;
d
H (300 MHz; CDCl3) 1.71 (4H, m, NCH2CH2CH2), 2.43 (1.15H,
m, NCHD, NCH2), 3.56 (2H, s, PhCH2) and 7.16–7.28 (4.85H,
complex, phenylCH); dC (75 MHz; CDCl3) as reported above;
dD (61.4 MHz; CHCl3) 2.60 (2.87D, s, NCD2, NCHD) and 7.65
(0.13D, s, phenylCD); m/z (EI) 164 (MH+, 69%), 162 ((M–H)+,
75), 162 (27), 92 (72), 91 (100), 87 (75), 73.2 (43), 72 (38) and 65
(66).
Reaction at -50 ◦C. Tributyltin hydride (200 ml, 0.74 mmol)
and 2,2¢-azobisisobutyronitrile (70 mg, 0.43 mmol) were
added to a stirred, degassed solution of N-(2-iodobenzyl)-2,2-
dideuteropyrrolidine 19 (200 mg, 0.69 mm◦ol) in fluorobenzene
(12 cm3), under a nitrogen atmosphere at -50 C. The reaction mix-
ture was irradiated with a medium pressure mercury vapour lamp
for 6 h, with three further aliquots of 2,2¢-azobisisobutyronitrile
(70 mg, 0.43 mmol) being added after 1, 3.5 and 5 h. The solvent
was removed in vacuo and the residue was purified by flash
chromatography on silica gel containing potassium fluoride (10%
w/w) (gradient from 100% petroleum ether to 80% petroleum
ether–20% ethyl acetate) to give a mixture of 26 and 27 in 7 : 1 ratio
(62 mg, 55%) as a colourless oil; nmax(thin film)/cm-1 as reported
above; dH (300 MHz; CDCl3) 1.73 (4H, m, NCH2CH2CH2), 2.44
(1.15H, m, NCHD, NCH2), 3.55 (2H, s, PhCH2) and 7.15–7.28
(4.85H, complex, phenylCH); dC (75 MHz; CDCl3) as reported
above; dD (61.4 MHz; CHCl3) 2.58 (2.88D, s, NCD2, NCHD) and
7.46 (0.12D, s, phenylCD); m/z (EI) 163 ((M–H)+, 30%), 162 (59),
92 (6), 91 (79), 86 (100), 72 (90) and 65 (58).
n
max(thin film)/cm-1 as reported above; dH (300 MHz; CDCl3)
1.70 (4H, m, NCH2CH2CH2), 2.48 (2.00H, m, NCH2, NCHD),
3.55 (2H, s, PhCH2) and 7.15–7.28 (5.00H, complex, phenylCH);
dC (75 MHz; CDCl3) as reported above; dD (61.4 MHz; CHCl3)
2.75 (1.98D, s, NCD2, NCHD) and 7.60 (0.02D, s, phenylCD); m/z
(EI) 162 ((M–H)+, 20%), 92 (10), 91 (30), 73 (30) and 57 (100); m/z
(CI) 164 (MH+, 20%), 153 (30) and 113 (58).
N-Benzyl-2,2-dideutero-5-deuteropyrrolidine (26) and
N-(2-deuterobenzyl)-2,2-dideuteropyrrolidine (27)
Reaction at 80 ◦C. Tributyltin deuteride (98 atom% D, 210 ml,
0.78 mmol) was added to a degassed solution of N-(2-iodobenzyl)-
2,2-dideuteropyrrolidine 19 (170 mg, 0.59 mmol) in benzene
(12 cm3), under a nitrogen atmosphere. The reaction mixture was
4662 | Org. Biomol. Chem., 2010, 8, 4653–4665
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
3-(N-Benzylpyrrolidin-2-yl)propanenitrile (28)
7.42 (4.76H, complex, phenylCH both products); dC (100 MHz;
CDCl3; with DEPT editing) 13.6 (CH2CN), 22.7 (CH2CH2CN),
29.8 (pyrrolidine NCD2CH2), 30.1 (pyrrolidine NCHCH2), 54.7
(pyrrolidine NCH2 minor product), 59.2 (PhCH2), 62.7 (NCH)
and 127.4, 128.7, 129.1 (phenylCH); dD (61.4 MHz; CH2Cl2) 2.15
(0.76D, s, pyrrolidine NCDaDb major product), 2.60 (0.24D, s,
NCD minor product), 2.85 (0.76D, s, pyrrolidine NCDaDb major
product) and 7.35 (0.24D, s, phenylCD minor product); m/z (EI)
216 (M+, 1%), 162 (98), 92 (37), 91 (100), 65 (17), 54 (12) and 41
(22); m/z (CI) 217 (MH+, 100%), 162 (8), 127 (3) and 52 (1).
Tributyltin hydride (470 ml, 1.75 mmol) and 2,2¢-
azobisisobutyronitrile (85 mg, 0.52 mmol) were added to a
stirred, degassed solution of N-(2-iodobenzyl)pyrrolidine 18
(250 mg, 0.87 mmol) and acrylonitrile (280 ml, 4.25 mmol) in
benzene (15 cm3), under a nitrogen atmosphere. The reaction
mixture was heated under reflux for 6 h with three further
aliquots of 2,2¢-azobisisobutyronitrile (70 mg, 0.43 mmol)
being added after 1, 3.5 and 5 h. The solvent was removed in
vacuo and the residue was purified by flash chromatography on
silica gel containing potassium fluoride (20% w/w) (gradient
from 90% dichloromethane–10% petroleum ether to 80%–20%
dichloromethane) to give 28 (104 mg, 56%) as a colourless
oil (Found MH+ (EI+) 215.1544, C14H19N2 requires 215.1543);
Reaction at 25 ◦C. Tributyltin hydride (460 ml, 1.71 mmol)
and 2,2¢-azobisisobutyronitrile (70 mg, 0.43 mmol) were
added to a stirred, degassed solution of N-(2-iodobenzyl)-2,2-
dideuteropyrrolidine 19 (250 mg, 0.87 mm◦ol) and acrylonitrile
(280 ml, 4.25 mmol) inbenzene (15 cm3), at 25 C. Thereactionmix-
ture was irradiated with a medium pressure mercury vapour lamp
for 6 h, with three further aliquots of 2,2¢-azobisisobutyronitrile
(70 mg, 0.43 mmol) being added after 1, 3.5 and 5 h. The
solvent was removed in vacuo and the residue was purified
by flash chromatography on silica gel containing potassium
fluoride (10% w/w) (gradient from 100% petroleum ether to 80%
petroleum ether–20% ethyl acetate) to give a mixture of 29 and
30 in a 4.5 : 1 ratio (44 mg, 23%) as a colourless oil (Found
MH+ (ES+) 217.1669, C14H17D2N2 requires 217.1668); nmax(thin
film)/cm-1 as reported above; dH (300 MHz; CDCl3) 1.50 (1H
m, pyrrolidine NCH(orD)CHaHb both products), 1.61–1.81 (3H,
complex, pyrrolidine NCH(orD)2CHaHb, CH2CH2CN both prod-
ucts), 1.85–2.05 (2H, complex, pyrrolidine NCH(orD)CHaHb,
pyrrolidine NCH(or D)2CHaHb both products), 2.19 (0.18H, q, J
= 6.5, pyrrolidine NCHaHb minor product), 2.34, 2.46 (2 ¥ 1H, 2 ¥
m, CH2CN both products), 2.62 (0.82H, m, NCH major product),
2.93 (0.18H, m, pyrrolidine NCHaHb minor product), 3.28, 3.92
(2 ¥ 1H, AB, J = 12, PhCH2 both products) and 7.07–7.42 (4.82H,
complex, phenylCH both products); dC (75 MHz; CDCl3) 13.1
(CH2CN), 22.2 (CH2CH2CN), 29.4 (pyrrolidine NCD2CH2), 29.6
(pyrrolidine NCHCH2), 53.4 (pyrrolidine NCH2 major product),
54.2 (pyrrolidine NCH2 minor product), 58.8 (PhCH2), 62.2
(NCH), 120.2 (CN), 126.9, 128.2, 128.6 (phenylCH) and 139.2
(phenyl ipsoC); dD (61.4 MHz; CH2Cl2) 2.14 (0.82D, s, pyrrolidine
NCDaDb major product), 2.60 (0.18D, s, NCD minor product),
2.85 (0.82D, s, pyrrolidine NCDaDb major product) and 7.35
(0.18D, s, phenylCD minor product); m/z (EI) 216 (M+, 1%), 215
((M–H)+, 3), 162 (53), 92 (30), 91 (100) and 65 (11); m/z (CI) 217
(MH+, 100%), 162 (10), 127 (18) and 106 (21).
n
max(thin film)/cm-1 3122–2726 (s) (C–H), 2244 (m) (CN),
1686 (m), 1530 (w), 1495 (m), 1453 (s), 1425 (m), 1369 (m),
1210 (w), 1154 (w), 1125 (m), 1075 (w), 740 (m) and 700 (m);
dH (300 MHz; CDCl3) 1.47 (1H, m, pyrrolidine NCHCHaHb),
1.59–1.79 (3H, complex, pyrrolidine NCH2CHaHb, CH2CH2CN),
1.89–2.06 (2H, complex, pyrrolidine NCHCHaHb, pyrrolidine
NCH2CHaHb), 2.19 (1H, ca q, J = 6.5, pyrrolidine NCHaHb), 2.41
(2H, m, CH2CN), 2.62 (1H, m, NCH), 2.93 (1H, m, pyrrolidine
NCHaHb), 3.29, 3.93 (2 ¥ 1H, AB, J = 12.0, PhCH2) and
7.20–7.38 (5H, complex, phenylCH); dC (75 MHz; CDCl3) 13.1
(CH2CN), 22.4 (CH2CH2CN), 29.4 (NCH2CH2), 29.6 (CHCH2),
53.4 (pyrrolidine NCH2), 58.9 (PhCH2), 120.3 (CN), 127.0, 128.3,
128.6 (phenylCH) and 139.2 (phenyl ipsoC); m/z (CI) 215 (MH+,
100%), 160 (14), 123 (10) and 58 (15).
3-(N-Benzyl-2,2-dideuteropyrrolidin-2-yl)propanenitrile (29) and
3-(N-(2-deuterobenzyl)-2-deuteropyrrolidin-2-yl)propanenitrile
(30)
Reaction at 80 ◦C. Tributyltin hydride (460 ml, 1.71 mmol)
and 2,2¢-azobisisobutyronitrile (70 mg, 0.43 mmol) were
added to a stirred, degassed solution of N-(2-iodobenzyl)-2,2-
dideuteropyrrolidine 19 (250 mg, 0.87 mmol) and acrylonitrile
(280 ml, 4.25 mmol) in benzene (15 cm3), under a nitrogen
atmosphere. The reaction mixture was heated under reflux for
6 h with three further aliquots of 2,2¢-azobisisobutyronitrile
(70 mg, 0.43 mmol) being added after 1, 3.5 and 5 h. The
solvent was removed in vacuo and the residue was purified by
flash chromatography on silica gel containing potassium fluoride
(20% w/w) (gradient from 90% dichloromethane–10% petroleum
ether to 80%–20% dichloromethane) to give a mixture of 29 and
30 in a 3.2 : 1 ratio (29 mg, 15%) as a colourless oil (Found
MH+ (ES+) 217.1667, C14H17D2N2 requires 217.1668); nmax(thin
film)/cm-1 3104–2692 (s) (C–H, C–D), 2245 (m) (CN), 1604 (w),
1495 (m), 1453 (s), 1425 (m), 1368 (m), 1201 (m), 1181 (m), 1119
(m), 1028 (w), 739 (m) and 701 (m); dH (400 MHz; CDCl3) 1.51 (1H
m, pyrrolidine NCH(orD)CHaHb both products), 1.57–1.79 (3H,
complex, pyrrolidine NCH(orD)2CHaHb, CH2CH2CN both prod-
ucts), 1.82–2.03 (2H, complex, pyrrolidine NCH(orD)CHaHb,
pyrrolidine NCH(or D)2CHaHb both products), 2.19 (0.24H, q,
J = 6.5, pyrrolidine NCHaHb minor product), 2.33, 2.45 (2 ¥
1H, 2 ¥ m, CH2CN both products), 2.62 (0.74H, m, NCH major
product), 2.91 (0.24H, m, pyrrolidine NCHaHb minor product),
3.28, 3.91 (2 ¥ 1H, AB, J = 12, PhCH2 both products) and 7.07–
Reaction at -50 ◦C. Tributyltin hydride (390 ml, 1.45 mmol)
and 2,2¢-azobisisobutyronitrile (70 mg, 0.43 mmol) were
added to a stirred, degassed solution of N-(2-iodobenzyl)-2,2-
dideuteropyrrolidine 19 (210 mg, 0.73 mmol) and acrylonitrile
(280 ml, 4.25 mmol) in fluorobenzene (13 cm3), at -50 ◦C. The
reaction mixture was irradiated with a medium pressure mer-
cury vapour lamp for 6 h, with three further aliquots of 2,2¢-
azobisisobutyronitrile (70 mg, 0.43 mmol) being added after 1,
3.5 and 5 h. The solvent was removed in vacuo and the residue
was purified by flash chromatography on silica gel containing
potassium fluoride (10% w/w) (gradient from 100% petroleum
ether to 80% petroleum ether–20% ethyl acetate) to give a mixture
of 29 and 30 in a 5.7 : 1 ratio (89 mg, 56%) as a colourless oil (Found
MH+ (ES+) 217.1667, C14H17D2N2 requires 217.1668); nmax(thin
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 4653–4665 | 4663
©

View Article Online
film)/cm-1 as reported above; dH (300 MHz; CDCl3) 1.51 (1H
m, pyrrolidine NCH(orD)CHaHb both products), 1.64–1.81 (3H,
complex, pyrrolidine NCH(orD)2CHaHb, CH2CH2CN both prod-
ucts), 1.87–2.06 (2H, complex, pyrrolidine NCH(orD)CHaHb,
pyrrolidine NCH(or D)2CHaHb both products), 2.21 (0.15H, q, J
= 6.5, pyrrolidine NCHaHb minor product), 2.34, 2.47 (2 ¥ 1H, 2 ¥
m, CH2CN both products), 2.63 (0.85H, m, NCH major product),
2.93 (0.15H, m, pyrrolidine NCHaHb minor product), 3.29, 3.94
(2 ¥ 1H, AB, J = 12, PhCH2 both products) and 7.17–7.35 (4.85H,
complex, phenylCH both products); dC (75 MHz; CDCl3) 13.1
(CH2CN), 22.2 (CH2CH2CN), 29.4 (pyrrolidine NCD2CH2), 29.6
(pyrrolidine NCHCH2), 54.3 (pyrrolidine NCH2 minor product),
58.8 (PhCH2), 62.3 (NCH), 120.3 (CN), 127.0, 128.3, 128.7
(phenylCH) and 139.3 (phenyl ipsoC) - major product NCD and
other signals or minor product not observed; dD (61.4 MHz;
CH2Cl2) 2.15 (0.85D, s, pyrrolidine NCDaDb major product),
2.59 (0.15D, s, NCD minor product), 2.85 (0.85D, s, pyrrolidine
NCDaDb major product) and 7.35 (0.15D, s, phenylCD minor
product); m/z (EI) 216 (M+, 3%), 215 ((M–H)+, 5), 162 (100), 92
(24), 91 (83), 65 (10) and 40 (39); m/z (CI) 217 (MH+, 100%), 162
(10), 127 (8), 106 (6), 52 (18) and 77 (7).
to 95% chloroform–5% methanol) to give 31 (175 mg, 28%) as
a colourless oil (Found MH+ (79Br, CI) 206.0505, C8H1379BrD2N
requires 206.0508); nmax(thin film)/cm-1 3022–2338 (s) (C–H and
C–D), 1672 (w), 1629 (m) (C C), 1448 (w), 1317 (w), 1292 (w),
1240 (w), 1158 (m) and 913 (w); dH (300 MHz; CDCl3) 1.83 (4H, br
complex, CD2CH2CH2), 2.59–2.79 (6H, br complex, pyrrolidine
NCH2 and C(Br)CH2CH2), 5.46 (1H, s, CHH C) and 5.68
(1H, s, CHH = C); dC (75 MHz; CDCl3) 23.1, 23.4 (CD2CH2CH2),
40.5 (C(Br)CH2CH2), 53.3 (CD2), 54.1 (pyrrolidine NCH2), 54.5
(C(Br)CH2CH2N), 118.0 (CH2 C) and 131.5 (CH2 = C); dD
(61.4 MHz; CH2Cl2) 3.20 (CD2); m/z (CI) 208 (MH+ (81Br), 100%),
206 (MH+ (79Br), 98), 128 (25), 86 (24), 74 (10) and 52 (10).
( )-5,5-Dideutero-1-methylhexahydro-(1H-pyrrolizine)
hydrobromide (35) and ( )-1,7a-dideutero-1-methyl-hexahydro-
1H-pyrrolizine hydrobromide (36)
A degassed solution of tributyltin hydride (250 ml, 0.93 mmol) and
2,2¢-azobisisobutyronitrile (6 mg, 0.04 mmol) in benzene (12 cm3)
was added dropwise (via a syringe pump) to a degassed solution
of 1-(3-bromobut-3-enyl)-2,2-dideuteropyrrolidine 31 (70 mg,
0.34 mmol) and 2,2¢-azobisisobutyronitrile (6 mg, 0.04 mmol) in
benzene (28 cm3), at reflux under a nitrogen atmosphere, over a
period of 1 h. After a further 2 h at reflux, the solvent was removed
in vacuo and thiophenol (112 ml, 1.09 mmol) was added to the
resulting residue. The crude product thus obtained was purified
by flash chromatography on silica gel (75% petroleum ether–
25% diethyl ether, then a gradient from 99% dichloromethane–
1% methanol to 95% dichloromethane–5% methanol) to give
a mixture of 35 and 36 in a 5.7 : 1 ratio (21 mg, 30%) as a
colourless syrup (Found (MH–Br)+ (ES+) 128.1402, C8H14D2N
requires 128.1403; nmax(thin film)/cm-1 3436 (br, m) (+N–H), 3122–
2417 (s) (C–H), 1641 (w), 1460 (m), 1385 (w) and 1033 (w); dH
(300 MHz; CDCl3) 1.07 (3H, d, J = 7.0, CH3), 1.60–1.80 (2H,
complex, C(2)H and C(7)H), 1.89–2.21 (4H, complex, C(2)H,
C(7)H and C(6)H2), 2.50 (0.85H, m, C(1)H), 2.69 (0.15H, m,
C(5)H), 3.01 (1H, dd, J = 7, 10, C(3)H), 3.57 (1H, m, C(3)H), 3.92
(0.15H, m, C(5)H) and 4.24 (0.85H, ca q, J = 9, C(7a)H) (Note:
The 1H NMR spectrum showed the presence of a small quantity
(ca 9%) of 1-(but-3-enyl)-2,2-dideuteropyrrolidine hydrobromide
37, with distinguishable resonances at 5.16 (1H, d, J = 8, CH CH2
cis), 5.17 (1H, d, J = 16, CH CH2 trans) and 5.73 (1H, m, CH
= CH2)); dD (61.4 MHz; CH2Cl2) 2.47 (0.15D, s, C(1)D), 2.70
(0.85D, s, C(5)D), 3.88 (0.85D, s, C(5)D) and 4.20 (0.15D, s,
C(7a)D); dC (75 MHz; CDCl3) 13.4 (CH3), 25.6 (C(6)), 25.7 (C(7)),
30.6 (C(2)), 34.6 (C(1)), 53.7 (C(3)) and 70.2 (C(7a)) (Note: C(5)D2
not distinguishable.); m/z (CI) 128 (MH+ (–Br), 100%), 58 (10),
52 (51) and 44 (18).
3-Bromobut-3-enyl methanesulfonate (33)23
Methanesulfonyl chloride (1.05 cm3, 13.6 mmol) was added
dropwise to a stirred solution of 3-bromobut-3-en-1-ol 34
(1.70 g, 11.3 mmol) and triethylamine (1.82 cm3, 13.6 mmol) in
dichloromethane (38 cm3) at 0 ◦C. After stirring for a further 1 h
at this temperature, water (38 cm3) was added and the separated
aqueous phase was extracted with dichloromethane (3 ¥ 22 cm3).
The combined organic extracts were dried (magnesium sulfate),
filtered and evaporated in vacuo to give 33 (2.33 g, 90%) as a pale
yellow oil (Found MH+ (79Br, CI) 228.9457, C5H1079BrO3S requires
228.9534); nmax(thin film)/cm-1 3037–2832 (w) (C–H), 1632 (m)
(C C), 1468 (w), 1414 (w), 1355 (s), 1265 (w), 1227 (w), 1176 (s),
1145 (m), 1057 (w), 987 (m), 964 (m), 910 (m), 805 (m) and 734 (w);
dH (300 MHz; CDCl3) 2.81 (2H, t, J = 6.0, CH2CH2O), 3.00 (3H, s,
CH3), 4.35 (2H, t, J = 6.0, CH2O), 5.54 (1H, d, J = 1.2, CHH C)
and 5.72 (1H, d, J = 1.2, CHH = C); dC (75 MHz; CDCl3) 37.2
(CH3), 40.7 (CH2CH2O), 66.7 (CH2O), 120.4 (CH2 C) and 127.5
(CH2 = C); m/z (CI) 231 (MH+ (81Br), 3%), 229 (MH+ (79Br), 3),
149 (13), 135 (98), 133 (100), 115 (21), 97 (12), 85 (10), 79 (10),
71 (4), 53 (7) and 51 (7). (Note: Although the preparation of 33
has been reported,23 no isolation method or characterisation was
presented.)
1-(3-Bromobut-3-enyl)-2,2-dideuteropyrrolidine (31)
Anhydrous potassium carbonate (1.000 g, 7.24 mmol) and 3-
bromobut-3-enyl methanesulfonate 33 (686 mg, 3.00 mmol) were
added sequentially to a stirred solution of 2,2-dideuteropyrrolidine
hydrochloride 22 (395 mg, 3.60 mmol) in dry acetonitrile (15 cm3).
After heating the stirred reaction mixture under reflux for 12 h,
additional anhydrous potassium carbonate (500 mg, 3.62 mmol)
was added and reflux was continued for a further 12 h. The
solid was filtered off and washed with dichloromethane (2 ¥
15 cm3) and the combined filtrate/washings were evaporated to
dryness in vacuo. The residue produced was purified by flash
chromatography on silica gel (gradient from 100% chloroform
Acknowledgements
For funding, the authors gratefully acknowledge the support
of EPSRC (Project Studentship to S. B. and DTA funding to
K. M. W.) and UCB Celltech (CASE funding to K. M. W.). Dr
Jeremy Roberston (Oxford, UK) generously provided essential
advice on the preparation and isolation of ( )-heliotridane. For
NMR studies, we thank the late Dr Vladimir Sik, Dr Ivan Prokes
and Dr Stephen J. Simpson (all Exeter, UK) and for mass spectra,
4664 | Org. Biomol. Chem., 2010, 8, 4653–4665
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
the EPSRC National Mass Spectrometry Service Centre (Swansea,
UK) and Mr Eric Underwood (Exeter, UK).
Middleton, C. J. Nichols, S. R. Richardson and P. G. Vernon, J. Am.
Chem. Soc., 1994, 116, 11275.
13 D. L. J. Clive, A. Khodabocus, M. Cantin and Y. Tao, J. Chem. Soc.,
Chem. Commun., 1991, 1755; D. L. J. Clive, M. Cantin, A. Khodabocus,
X. Kong and Y. Tao, Tetrahedron, 1993, 49, 7917.
14 H. Tsumoto, C. Murata, N. Miyata, R. Taguchi and K. Kohda, Chem.
Pharm. Bull., 2003, 51, 1399. Other than melting point for which there
was a discrepancy, our analytical data for 15 matched those reported
in; J. B. Christensen, Molecules, 2001, 6, 47.
Notes and references
1 T. H. Lowry and K. S. Richardson in Mechanism and Theory in Organic
Chemistry, HarperCollins, New York, 3rd edn, 1987, 232.
2 J. E. Baldwin, R. M. Adlington, R. T. Aplin, N. P. Crouch, G. Knight
and C. J. Schofield, J. Chem. Soc., Chem. Commun., 1987, 1651.
3 J. Clayden, J. H. Pink, N. Westlund and F. X. Wilson, Tetrahedron Lett.,
1998, 39, 8377.
15 K. Nozaki, K. Oshima and K. Utimoto, Tetrahedron, 1989, 45, 923.
16 M. E. Wood, M. J. Penny, J. S. Steere, P. N. Horton, M. E. Light and
M. B. Hursthouse, Chem. Commun., 2006, 2983.
4 M. Miyano, J. Org. Chem., 1981, 46, 1854.
17 C. J. Easton, Chem. Rev., 1997, 97, 53.
5 E. Vedejs and J. Little, J. Am. Chem. Soc., 2002, 124, 748.
6 G. B. Dudley, S. J. Danishefsky and G. Sukenick, Tetrahedron Lett.,
2002, 43, 5605.
7 M. E. Wood, S. Bissiriou, C. Lowe and K. M. Windeatt, Org. Biomol.
Chem., 2008, 6, 3048.
18 G. Brunton, D. Griller, L. R. C. Barclay and K. U. Ingold, J. Am.
Chem. Soc., 1976, 98, 6803; G. Brunton, J. A. Gray, D. Griller, L. R. C.
Barclay and K. U. Ingold, J. Am. Chem. Soc., 1978, 100, 4197.
19 L. Williams, S. E. Booth and K. Undheim, Tetrahedron, 1994, 50, 13697.
20 D. C. Harrowven and I. L. Guy, Chem. Commun., 2004, 1968.
21 A. Citterio, A. Arnoldi and F. Minisci, J. Org. Chem., 1979, 44, 2674.
22 J. Robertson, M. A. Peplow and J. Pillai, Tetrahedron Lett., 1996, 37,
5825.
23 A. Padwa and A. G. Waterson, Tetrahedron, 2000, 56, 10159.
24 T. J. Barton, J. Lin, S. Ijadi-Maghsoodi, M. D. Power, X. Zhang, Z. Ma,
H. Shimizu and M. S. Gordon, J. Am. Chem. Soc., 1995, 117, 11695.
25 R. Hooft, Collect: Data Collection Software, Nonius B.V., Delft, The
Netherlands, 1998.
8 J. Robertson, J. Pillai and R, K. Lush, Chem. Soc. Rev., 2001, 30,
94.
9 R. Gosain, A. M. Norrish and M. E. Wood, Tetrahedron Lett., 1999,
40, 6673; R. Gosain, A. M. Norrish and M. E. Wood, Tetrahedron,
2001, 57, 1399.
10 D. Seebach, A. R. Sting and M. Hoffmann, Angew. Chem., Int. Ed.
Engl., 1996, 35, 2708.
11 T. Shono, N. Kise, R. Nomura and A. Yamanami, Tetrahedron Lett.,
1993, 34, 3577.
12 D. L. J. Clive, Y. Tao, A. Khodabocus, Y.-J. Wu, A. G. Angoh, S. M.
Bennett, C. N. Boddy, L. Bordeleau, D. Kellner, G. Gleiner, D. S.
26 Z. Otwinowski and W. Minor, Methods Enzymol., 1997, 276, 307.
27 G. M. Sheldrick, Acta Crystallogr., Sect. A: Found. Crystallogr., 2008,
64, 112.
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 4653–4665 | 4665
©