Borohydride exchange resin as an alternative method for deuterium labelling: selective reduction of aryl bromides

Article abstract of DOI:10.1002/jlcr.520

Transition metal-activated borohydride exchange resin and borodeuteride exchange resin have been prepared and used for the selective reductive deuteration of aryl bromides. This methodology, which utilizes deuterated methanol as the primary source of labe

Full text of DOI:10.1002/jlcr.520

JOURNAL OF LABELLED COMPOUNDS AND RADIOPHARMACEUTICALS
J Labelled Cpd Radiopharm 2001; 44: 1025–1033.
DOI: 10.1002/jlcr.520
Research Article
Borohydride exchange resin as an alternative
method for deuterium labelling:
selective reduction of aryl bromides
Scott W. Landvatter*, Douglas J. Schauery, Keith T. Garnes,
z
James F. Mack and Lewis B. KillmerJr.
GlaxoSmithKline, Isotope Chemistry, PO Box 1539, King of Prussia,
PA 19406, USA
Summary
Transition metal-activated borohydride exchange resin and borodeuteride
exchange resin have been prepared and used for the selective reductive
deuteration of aryl bromides. This methodology, which utilizes deuterated
methanol as the primary source of label, provides an alternative method for the
introduction of deuterium. Investigation of which bond in the solvent donates
the deuterium and model reductions on ortho-bromo-N;N-dimethylbenzamide
suggest that an organocopper(III) intermediate is most likely involved. The
immobilized hydride on resin also provides an easy workup and gives products
of high purity. This methodology is far more selective than typical catalytic
deuteration and, in most cases, provides products with high levels of deuterium
incorporation. Copyright # 2001 John Wiley & Sons, Ltd.
Key Words: labelling; reduction; dehalogenation; resins
Introduction
Borohydride exchange resin (BER)1 is a quaternary ammonium
borohydride supported on an ion exchange resin (typically Amberlite
*Correspondence to: S. W. Landvatter, GlaxoSmithKline, Isotope Chemistry, P. O. Box 1539,
King of Prussia, Pennsylvania 19406, USA. E-mail: scott w landvatter@gsk.com
y Co-op student, Drexel University, Philadelphia, Pennsylvania U.S.A.
z Analytical Sciences Department.
Copyright # 2001 John Wiley & Sons, Ltd.
Received 18 June 2001
Revised 2 August 2001
Accepted 26 August 2001

1026
S. W. LANDVATTER ET AL.
IRA-400). BER is a milder reducing agent than sodium borohydride2–4
and allows for workup by simple filtration.
BER reactivity can be modified via addition of a transition metal salt
which greatly increases its reducing power. These salts include
8
Ni(OAc)2,5–7 CuSO4 and Pd.9 Modified BER resin has been used for
selective olefin reductions,5 alkyne to alkene reductions,10 aryl nitro
reductions to anilines11 and reductive aminations.12
Labelling an aromatic ring with deuterium is typically accomplished
by dehalogenation using deuterium or tritium gas over a Pd catalyst, or
by isotopic exchange with labelled water or deuterium or tritium gas.13
Often, these reductions need to be selective since many compounds
of interest bear chlorine. Copper sulfate-activated BER, in particular,
is thus of interest for isotopic labelling due to its selectivity in
reducing bromoaromatics in the presence of chloroaromatics such as
4-bromochlorobenzene to chlorobenzene.8
Narisada14 has previously noted that a solution phase copper(I)
chloride–sodium borohydride system can effect labelling when deuter-
ated solvent is used. Since the primary source of the proton in these
BER reductions must also be from the solvent rather than from the
resin, use of a deuterated solvent would provide a deuteron for
quenching the active intermediate. This would thus offer an alternative
way to introduce a deuterium label into a molecule.
Results and discussion
Preparation of BER and BDR
The BER is easily prepared by treating the chloride form of Amberlite
IRA-400 with a 1 M aqueous solution of sodium borohydride as
described by Yoon.15 Borodeuteride resin (BDR) was prepared in a
similar manner by treating the chloride form of Amberlite IRA-400 with
a 1 M aqueous solution of sodium borodeuteride (in H2O, not D2O).
Excess borohydride or borodeuteride was rinsed from the resin with
water and the resin was dried under vacuum.
Model reductive deuterations
A thorough examination of the labelling capability of the BER/BDR–
copper sulfate system was carried out using 4-bromobenzene derivatives
as model substrates (Scheme 1). The data are summarized in Table 1.
Copyright # 2001 John Wiley & Sons, Ltd.
J Labelled Cpd Radiopharm 2001; 44: 1025–1033

BOROHYDRIDE EXCHANGE RESIN REDUCTION OF ARYL BROMIDES 1027
R
R
BDR/CuSO4
CD3OD
Br
D
1 (R=Cl)
2 (R=OMe)
3 (R=CN)
5 (R=Cl)
6 (R=OMe)
7 (R=CN)
4 (R=CO2Et)
8 (R=CO2Et)
Scheme 1.
Table 1. Reduction of 4-bromobenzenes 1–4 with BER–CuSO4 or BDR–CuSO4
in methanol or methanol-d4
Resin
Solvent
R
% d0
% d1
1. BER
2. BER
3. BER
4. BER
5. BDR
6. BDR
7. BDR
8. BDR
9. BDR
10. BDR
11. BDR
12. BDR
CD3OD
CD3OD
CD3OD
CD3OD
CD3OD
CD3OD
CD3OD
CD3OD
CH3OH
CH3OH
CH3OH
CH3OH
Cl
OMe
CN
CO2Et
Cl
OMe
CN
CO2Et
Cl
30
34
17
33
15
8
5
5
95
96
95
95
70
66
83
67
85
92
95
95
5
OMe
CN
CO2Et
4
5
5
Entries 1–4 represent our initial attempts at the reduction using BER
and deuterated methanol as solvent. Yields were moderate until the
resin was extracted with methylene chloride which gave yields of
90–100%. All four substrates were selectively reduced and had
deuterium incorporations of 0.66–0.83 d/molecule. However, this
incorporation was substantially less than theoretical and indicated that
there must be a significant source of isotopic dilution which occurred
during reduction.
One potential source of isotopic dilution was the resin itself.
Switching to BDR and still using methanol-d4 as solvent increased
deuterium incorporation to 0.85–0.96 d/molecule (entries 5–8). It thus
appeared that the hydride protons on the resin itself were entering into
the reduction.
Based on entries 1–4, it was clear that deuterium was primarily
coming from the solvent, although a still significant portion was derived
from the borodeuteride on the resin (entries 5–8). Entries 9–12
confirmed this where the only source of deuterium was the resin itself.
Copyright # 2001 John Wiley & Sons, Ltd.
J Labelled Cpd Radiopharm 2001; 44: 1025–1033

1028
S. W. LANDVATTER ET AL.
Table 2. Reductive deuteration of various aryl bromides
R
R
B
O4
CD3OD
65oC, 1 hr
Br
D
X1
X2
R
X1
BrD
X2
Average d/moleculea
b
OMe
H
1.88
CONMe2
CONMe2
0.86
a Isotope incorporation was determined by mass spectroscopy. Yields were >95% in all cases and
were determined by HPLC.
b 88.7% d2, 10.9% d1, 0.4% d0.
A deuterium incorporation of 0.05 d/molecule was still observed with
BDR in methanol.
We also investigated the reduction of 3,4-dibromoanisole and
2-bromo-N;N-dimethylbenzamide (Table 2). Deuterium incorporation
was again quite high.
Mechanism
Some reports invoke a radical based mechanism for BER halide
reductions.8 Under these conditions, 1-chlorooctane is not reduced but
1-bromooctane is. Reduction of bromocyclohexane gives a small
amount of cyclohexene, and reduction of benzyl chloride gives an 8%
yield of 1,2-diphenylethane along with the expected high yield of toluene
(83%). However, these observations were for BER reductions of
aliphatic systems. Narisada, on the other hand, in his work on copper(I)
chloride/sodium borohydride reductions in solution, has suggested the
intermediacy of a copper hydride.14 We chose to evaluate the
mechanism by examining the reductive deuteration of 4-bromochlor-
obenzene (1) with coppersulfate-activated BER orBDR in methanol- d4
to give 4-deuterochlorobenzene (5).
All possible sources of the deuterium label in the final product were
assessed by using CD3OD, CD3OH, CH3OD and CH3OH as solvent.
This enabled the deuterium donating capabilities of the C–D and O–D
bonds to be separately evaluated.
A radical based mechanism would predict the HOCD2–D bond, and
not the CD3O–D bond, as the primary donator of deuterium to the final
product. This order is predicted based on the bond dissociation energies
Copyright # 2001 John Wiley & Sons, Ltd.
J Labelled Cpd Radiopharm 2001; 44: 1025–1033

BOROHYDRIDE EXCHANGE RESIN REDUCTION OF ARYL BROMIDES 1029
Table 3. Source of deuterium in 4-bromochlorobenzene (1) reductions
Resin
Solvent
Ave. D incorporation
% d0
% d1
1
2
3
4
5
6
7
BDR
BDR
BDR
BDR
BER
BER
BER
CD3OD
CD3OH
CH3OD
CH3OH
CD3OD
CD3OH
CH3OD
0.85
0.11
0.86
0.05
0.70
0.02
0.65
15
89
14
95
30
98
35
85
11
86
5
70
2
65
of the two species. The HOCH2–H bond dissociation energy is 92 kcal/
mol16 whereas the CH3O–H bond dissociation energy is 102 kcal/mol.17
Cohen18 observed that the expected selectivity from the radical
intermediate in the copper(I) induced aryldiazonium decomposition of
diazobenzamides. Labelling studies demonstrated that the radical
intermediate abstracted a hydrogen from the methyl group of methanol.
An ion based mechanism, on the otherhand, should show predominant
deuterium contribution from the acidic O–D deuteron. Our results,
which are summarized in Table 3, clearly show the O–D deuteron to be
the primary source of label.
Deuterium incorporation is virtually the same regardless of whether
CD3OH orCH OH was used as solvent with BDR resin (entries 2 and
3
4). In the experiment, where the only possible source of label was the
CD3 group (entry 6), deuterium incorporation was only 0.02 d/molecule.
In contrast, deuterium incorporation was always high when an acidic
O–D group was present (entries 1, 3, 5 and 7). Thus, the primary source
of label is the O–D bond with a small amount of label deriving from the
BDR resin itself (entry 4).
More direct evidence for an ion based, rather than a radical based
mechanism, was obtained by studying the BDR reduction of ortho-
bromo-N;N-dimethylbenzamide (9). It is known that a radical generated
in the ortho position of N;N-dimethylbenzamide will lead to formation
of N-methylbenzamide (10) via hydrogen abstraction from one of the
methyl groups (Scheme 2).19
However, reduction of ortho-bromo-N; N-dimethylbenzamide (9) with
BDR/CuSO4 in methanol-d4 gave, as the sole product, 2-deutero-N;N-
dimethylbenzamide (11). Deuterium incorporation was 0.86 d/molecule.
None of the monomethyl benzamide (10) was observed (Scheme 3).
One mechanism consistent with these observations is shown below
(Scheme 4). Reduction of CuII to CuI is followed by oxidative-addition
Copyright # 2001 John Wiley & Sons, Ltd.
J Labelled Cpd Radiopharm 2001; 44: 1025–1033

1030
S. W. LANDVATTER ET AL.
Br
CuCl
.
CH3
N
CH3
H
N
N
CH3
CH3
CH3
O
O
O
9
10
Scheme 2.
BDR/CuSO4
X
CD3OD
D
O
BDR/CuSO4
CD3OD
H
O
Br
CH3
N
H
N
CH3
N
CH3
CH3
CH3
O
10
9
11
Scheme 3.
Br
Br
Br
H+
Cu
Cu
D
H
Red-
elim
Red-
elim
"CuD"
BDR
-DBr
H
-HBr
CuII
H
D
CH3OD
Cu
CH3OH
Scheme 4.
to the aryl halide to give a CuIII intermediate. This CuIII intermediate
could undergo either reductive elimination or decomposition to an aryl
cuprate, which then reacts with solvent. Direct reaction with the CuI
species with CuI and CuBrin the absence of hydirde showed no
reduction. It has been suggested that this CuI intermediate, in solution
reactions, is copper deuteride.14 If copper deuteride were formed as the
key intermediate, oxidative addition followed by the loss of DBr would
give an aryl cuprate that would rapidly be quenched by the O–D proton
of the solvent. A small amount of labelled product could also be derived
by reductive elimination from the CuIII intermediate. This would
explain why a small amount of label seems to derive solely from the
resin and not the solvent (Table 1). H–D exchange from the CuIII
intermediate would also be expected to occur.
Copyright # 2001 John Wiley & Sons, Ltd.
J Labelled Cpd Radiopharm 2001; 44: 1025–1033

BOROHYDRIDE EXCHANGE RESIN REDUCTION OF ARYL BROMIDES 1031
Cohen20 and Meyers21 has made similarobservations in the Ullman
coupling of aryl bromides where the likely mechanism involves initial
addition of a copper(I) halide to the aryl bromide producing an aryl
copper(III) intermediate. They have shown this intermediate to be
sensitive to protonation. Such is the case in these BDR/CD3OD
reductions where the large excess of CD3OD protonates the organo-
copper(III) intermediate giving the observed deuterium labelled
products.
Conclusion
In conclusion, copper sulfate-activated borodeuteride exchange resin
reductions, when carried out in methanol-d4, selectively reduces
bromoaromatics with high incorporation of deuterium. This represents
an easy alternative method for the introduction of deuterium into
aromatic molecules. The method is rapid, mild, highly selective and is
compatible with a wide variety of functional groups. Simple filtration
gives products of high purity.
Experimental
Amberlite IRA-400 resin (40 mesh), sodium borohydride and sodium
borodeuteride were purchased from Aldrich. CD3OD was purchased
from Isotec and was 99.96 atom % D. CH3OD was purchased from
Aldrich and was 99.5 atom % D. CD3OH was purchased from Aldrich
and was 99 atom % D. Reduction experiments were conducted using 5
equivalents (based on boron) of BER or BDR with the resin first treated
with 0.05 mmol CuSO4 ꢀ 5H2O/0.1 mmol aryl bromide for 2 minutes at
room temperature in either methanol or methanol-d4 (1 mL/0.1 mmol
aryl bromide). After 2 minutes the aryl bromide was added all at once
(eitherneat oras a solution in methanol ormethanol-d 4). Reactions
were monitored and yields determined by either HPLC or GC
(mesitylene as an internal standard). All product identities were
1
confirmed by H-NMR and HPLC, GC orTLC. GC-MS (Finnigan)
with chemical ionization (ammonia) was used to determine deuterium
incorporation.
Copyright # 2001 John Wiley & Sons, Ltd.
J Labelled Cpd Radiopharm 2001; 44: 1025–1033

1032
S. W. LANDVATTER ET AL.
Preparation of BDR
Amberlite IRA-400 (Cl- form, 100 g) was washed with water, ethanol
and then water. The wet resin was stirred with a solution of 19 g of
sodium borodeuteride in 500 ml water for 60 min. The mixture was
filtered and the borodeuteride resin washed with water to remove
residual sodium borodeuteride (monitored by addition of the filtrate to
dilute hydrochloric acid). The resin was dried under vacuum at 568C.
Borohydride exchange resin (BER) was prepared in the same manner
with sodium borohydride.
Typical reduction
The experiments were conducted using 5 equivalents (based on boron)
of BER or BDR. Reduction of 4-bromoanisole is representative. BDR
resin (162 mg) was treated with 12.5 mg CuSO4 ꢀ 5H2O (0.5 equivalents/
0.1 mmol aryl bromide) for 2 min at room temperature in 1 ml of
methanol-d4. Vigorous bubbling was seen and the resin turned black.
After2 min, the bubbling had ceased and 4-bormoanisole (18.7 mg,
12 ml, 0.1 mmol) was added. The reaction was heated at 658C for1 h.
The reaction was filtered and concentrated under vacuum. This
procedure gave 10 mg of anisole (93%). Proton NMR was identical
with an authentic standard except for the absence of signals where
deuterium was incorporated. Deuterium incorporation was determined
by GC/MS (Table 1).
References
1. Gibson HW, Bailey CF. J Chem Soc Chem Commun 1977; 815.
2. Yoon NM, Park KB, Gyoung YS. Tetrahedron Lett 1983; 24: 5367–5370.
3. Gyoung YS, Yoon NM, Jeon DH. Bull Korean Chem Soc 1987; 8:
162–165.
4. Salunkhe MM, WadgaonkarPP, SagarAD.
967–968.
Eur Polym J 1994; 30:
5. Choi J, Yoon NM. Synthesis 1996; 597–599.
6. Yoon NM, Lee HJ, Ahn JH, Choi J. J Org Chem 1994; 59: 4687–4688.
7. Yoon NM, Choi J, Lee HJ. Bull Korean Chem Soc 1993; 14: 543–545.
8. Sim TB, Yoon NM. Bull Chem Soc Jpn 1997; 70: 1101–1107.
9. Yoon NM, Lee HW, Choi J, Lee HJ. Bull Korean Chem Soc 1993; 14:
281–283.
Copyright # 2001 John Wiley & Sons, Ltd.
J Labelled Cpd Radiopharm 2001; 44: 1025–1033

BOROHYDRIDE EXCHANGE RESIN REDUCTION OF ARYL BROMIDES 1033
10. Choi J, Yoon NM. Tetrahedron Lett 1996; 37: 1057–1060.
11. Yoon NM, Choi J. Synlett 1993; 135–136.
12. Yoon NM, Kim EG, Son HS, Choi J. Syn Commun 1993; 23: 1595–1599.
13. Shu AYL, Chen W, Heys JR. J Organometal Chem 1996; 524: 87–93.
14. Narisada M, Horibe I, Watanabe F, Takeda K. J Org Chem 1989; 54:
5308–5313.
15. Sim TB, Choi J, Joung MJ, Yoon NM. J Org Chem 1997; 62: 2357–2361.
16. Kerr JA. Chem Rev 1966; 66: 465–500.
17. Benson, SW. J Chem Ed 1965; 42: 502–518.
18. Cohen T, Smith KW, Swerdloff MD. J Am Chem Soc 1971; 93: 4303–4304.
19. Cohen T, Wood J, Dietz AG. Tetrahedron Lett 1974; 3555–3558.
20. Cohen T, Cristea, I. J Am Chem Soc 1976; 98: 748–753.
21. Nelson DN, Meyers AI. J Org Chem 1994; 59: 2655–2658.
Copyright # 2001 John Wiley & Sons, Ltd.
J Labelled Cpd Radiopharm 2001; 44: 1025–1033