Use of lithium N,O-dimethylhydroxylamide as an efficient in situ protecting agent for aromatic aldehydes

Article abstract of DOI:10.1016/S0040-4020(02)00073-X

The use of lithium N,O-dimethylhydroxylamide as an alternative in situ protecting agent for aryl aldehydes with low ortho-directing properties has been evaluated and subsequently applied to two practical multi-step one-pot syntheses of developmental drug

Full text of DOI:10.1016/S0040-4020(02)00073-X

TETRAHEDRON
Pergamon
Tetrahedron 58 /2002) 1657±1666
Use of lithium N,O-dimethylhydroxylamide as an ef®cient in situ
protecting agent for aromatic aldehydes
²
Frank Roschangar,p Jennifer C. Brown, Bobby E. Cooley, Jr., Matthew J. Sharp
and Richard T. Matsuoka
Chemical DevelopmentÐSynthetic Chemistry, GlaxoSmithKline, Five Moore Drive, P.O. Box 13398,
Research Triangle Park, NC 27709, USA
Received 21 December 2001; accepted 18 January 2002
AbstractÐThe use of lithium N,O-dimethylhydroxylamide as an alternative in situprotecting agent for aryl aldehydes with low ortho-
directing properties has been evaluated and subsequently applied to two practical multi-step one-pot syntheses of developmental drug
candidate intermediates. q 2002 Elsevier Science Ltd. All rights reserved.
1. Introduction
as well as the strongly ortho-metal directing lithium
amide of N,N,N0-trimethylethylenediamine /LTMDA; 6)
/Fig. 1). Alternative in situprotection of aldehyde carbonyls
was achieved by use of titanium tetrakis/dialkylamides).4
The synthetic potential of a-amino lithium alkoxides 2 as
protecting groups, metalation-directing groups and
synthetic intermediates has been investigated and reviewed
/Scheme 1).1
Curiously, the readily available lithium N,O-dimethyl-
hydroxylamide /LDHA; 3) has not been investigated as an
in situ protecting group for aromatic aldehydes, although the
use of Weinreb amides has found widespread application for
ketone syntheses.5 In these reactions, the lack of over-
addition products was rationalized by a stable, metal-
chelated intermediate similar to 2. Lipshutz recently
exploited the stability of a-amino lithium alkoxide
complexes such as 2 for clean formylations of carbanions
with methoxy/methyl)formamide.6
R2
R1
N
Li
Li
O
O
1
R2
Ar
N
R1
Ar
H
H
2
Scheme 1.
Although the potential existed for a novel and potentially
useful application of lithium N,O-dimethylhydroxylamide
/3), our primary interest in the in situ aldehyde masking
methodology derived from the need to develop short and
cost-ef®cient routes to synthetic intermediates 10 and 11 of
GlaxoSmithKline drug candidates in preclinical develop-
ment /Fig. 2).
Comins pioneered the strategy of forming a-amino
alkoxides by addition of lithium dialkylamides 1 to
aromatic2 and heteroaromatic aldehydes,3 and demonstrated
that mild acidic treatment unveils the masked formyl
moiety. He introduced, amongst others, lithium 1-methyl-
piperazide /LNMP; 5), lithium morpholide /7), the weakly
basic lithium N-methyl-N-/2-pyridinyl)amide /LMPA; 4),
N
O
OMe
N
MeO
N
Li
N
N
Li
N
Li
N
Li
N
Li
N
Li
N
Li
7
8
9
3
4
5
6
Figure 1.
Keywords: N,O-dimethylhydroxylamine; a-amino lithium alkoxides; in situ protection of aromatic aldehydes.
Corresponding author. Tel.: 11-919-483-9664; fax: 11-919-315-8735; e-mail: fr83278@gsk.com
Present address: Department of Chemistry, Flanagan Building, East Carolina University, Greenville, NC 27858, USA.
p
²
0040±4020/02/$ - see front matter q 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020/02)00073-X

1658
F. Roschangar et al. / Tetrahedron 58 02002) 1657±1666
power. Benzaldehyde /14), the most fundamental aromatic
aldehyde, was chosen as model system for this initial screen
/Scheme 2),2b,9 in which a-amino lithium alkoxides 15 were
assessed in THF at 2208C.
In a typical experiment, lithium amides 3±9 were reacted
with benzaldehyde /14) to yield their corresponding
a-amino lithium alkoxides 15, which were subsequently
treated with 2.3 equiv. of n-BuLi for 1±3 h. If a particular
a-amino alkoxide moiety functioned not only as a
protecting group, but also as an ortho-directing group,
then ortho-lithiation would occur at a rate corresponding
to its ortho-directing power, to furnish a dianion which
could be alkylated by electrophiles. Therefore, to analyze
for ortho-directing power, the reactions were quenched with
iodomethane. The HPLC assessments of the experiments
were conducted using meticulous dilution techniques, and
showed predominantly three compounds: recovered 14,
2-methylbenzaldehyde /16), and 1-phenyl-1-pentanol /17)
/Table 1). 1-/2-Methylphenyl)-1-pentanol, the product of
both butyl addition and ortho-methylation, was only
detected by HPLC in a single experiment /entry 7; ca.
0.9% yield-in-solution) and thus omitted from the table.
We de®ned protective power /PP) as the sum of recovered
benzaldehydes: PPˆ% yield-in-solution /14)1% yield-in-
solution /16). The ortho-directing power /OP) was directly
derived from the amount of ortho-methylated benzalde-
hyde: OPˆ% yield-in-solution /16).
Figure 2.
Both compounds should be accessible by Suzuki coupling7
of boronic acids 12 and 13, respectively, with their corre-
sponding heteroaryl halide coupling partners. We intended
to use latent a-amino alkoxides in order to avoid the
common acetal protection±deprotection sequence of the
formyl moiety for the generation of the boronic acids. For
added practicality, we were interested in conducting this
chemistry at moderately low temperatures /240 to 2208C).
2. Results and discussion
We were pleased to discover that lithium N,O-dimethyl-
hydroxylamide /LDHA; 3) not only offered highly ef®cient
in situprotection of the aldehyde moiety, but also that its
corresponding a-amino lithium alkoxide displayed minimal
ortho-directing properties.
2.1. Ef®cacy screen of lithium amides
a-Amino lithium alkoxides are capable not only of protect-
ing the formyl moiety but also of directing subsequent
lithiation to the ortho-position, a fact which has been
exploited in synthetic applications.8 However, our intention
was to identify an ef®cient protecting agent which would
furnish an a-amino lithium alkoxide with low ortho-
directing propensity. The absence of published studies
quantifying these properties prompted us to initiate a
systematic screening of aryl aldehyde protecting agents
3±9.
In continuation of our initial screen, we evaluated the most
ef®cient non-ortho-directing in situprotecting gropus,
namely LDHA /3), lithium 1-methylpiperazide /LNMP;
5), and lithium morpholide /7), at different temperatures
/0 and 2408C in THF; entries 2, 3, 9, 10, 16, 17) and with
two different solvents /DME and TBME at 2208C; entries
5, 6, and 12, 13). As a result of this study, the ef®cacy of
lithium amides as non-ortho-directing benzaldehyde /14)
protecting agents, represented by the % yield-in-solution
of 14, at temperatures between 0 and 2408C, can be ranked
in the following order:
In particular, we were interested in a head-to-head com-
parison of LDHA /3) with literature-precedent lithium
amides in regards to protective as well as ortho-directing
R1
LDHA /3)$lithium morpholide /7)<LNMP /5).lithium
diethylamide /9)@lithium 2-methoxy-N-methylethanamide
/8).LTMDA /6)@LMPA /4).
OLi
O
Li
N
R2
R2
H
N
R1
Notably, compared to amides 5 and 7, the use 3 provided
signi®cantly higher recovery of 14 at 08C /entries 3, 10 and
17) or upon prolonged exposure to excess n-BuLi at 2208C
/3 h vs. 1 h; entries 4 and 11). In addition, the LDHA /3)-
derived a-amino lithium alkoxide displayed the overall
lowest OP properties /entries 1±6), contrasting, as expected,
the a-amino lithium alkoxides derived from 6 and 8, which
demonstrated good PP /entries 14, 18) and concurrently
strong OP. This, of course, rendered the latter two parent
amides less attractive for our purposes as non-ortho-direct-
ing protecting agents. Finally, a brief look at solvent effects
/THF vs DME vs TBME; entries 1, 5, 6, and 8, 12, 13)11
14
15
O
O
H
OH
14
n-BuLi; MeI;
AcOH, 0 oC;
piperidine, 20 oC
17
H
16
Scheme 2.

F. Roschangar et al. / Tetrahedron 58 02002) 1657±1666
1659
Table 1. Composition of mixtures for the reaction 14!15!14116117 for various lithium amides
Entry
Lithium amide
Exposure to
n-BuLi /h)/
temperature /8C)
Corrected %
yield-in-solution
of 14a
Corrected %
yield-in-solution of 16a
/ˆ% directing power)
Corrected %
yield-in-solution
of 17a
% Protective
power
1
1/220
97.8
0.2
0.0
98.0
2
3
1/240
1/0
3/220
1/220
1/220
99.4
90.3
93.9
96.9
92.1
0.0
0.9
0.2
0.7
0.1
0.4
0.0
0.0
0.0
2.1
99.4
91.2
94.1
97.6
92.2
4
5b
6c
7
8
1/220
13.0
0.2
83.4
13.2
1/220
92.4
0.7
0.0
93.1
9
1/240
1/0
3/220
1/220
1/220
94.3
78.9
84.7
90.4
93.7
0.1
3.0
1.5
2.6
0.0
0.0
0.0
0.0
1.1
0.0
94.4
81.8
86.2
93.0
93.7
10
11
12b
13c
14
1/220
58.8
35.1
0.0
93.9
15
1/220
95.4
0.8
0.5
96.2
16
17
1/240
1/0
94.8
78.7
0.1
3.0
0.0
0.0
94.9
81.7
18
19
1/220
1/220
67.4
86.7
17.4
0.5
0.0
0.0
84.7
87.2
Reactions carried out in THF and quenched with MeI followed by AcOH unless otherwise noted. Temperature was not allowed to rise more than 58C during
reagent additions. % Yield-in-solution determined by HPLC.10
Corrected for HPLC response factors.
a
b
Reaction carried out in 1,2-dimethoxyethane /DME).
Reaction carried out in tert-butyl methyl ether /TBME).
c
indicated slightly elevated ortho-directing aptitudes of the
a-amino lithium alkoxides in DME vs THF. Although
TBME provided positive results, the use of this solvent
was deemed less appropriate due to a signi®cant degree of
heterogeneity observed during these experiments.
available 2-furaldehyde /18). In an earlier disclosure, we
described the challenges associated with the development
of a practical one-pot synthesis of 5-/4-{3-chloro-4-[/3-
¯uorobenzyl)oxy]anilino}-6-quinazolinyl)-2-furaldehyde
/10) via palladium-mediated Suzuki coupling of in situ-
generated 5-/diethoxymethyl)-2-furylboronic acid with
readily available N-{3-chloro-4-[/3-¯uorobenzyl)oxy]-
phenyl}-6-iodo-4-quinazolinamine /23).13 Although signi®-
cant cost-savings had been achieved with this procedure, we
felt that replacement of 2-/diethoxymethyl)furan with
inexpensive 18 could deliver an even more economic
approach to 5-formyl-2-furylboronic acid /12).14
2.2. Application to 2-furaldehyde
Encouraged by these preliminary ®ndings, we proceeded to
examine the practicality and versatility of LDHA /3) as an
in situaryl aldehyde protecting agent with two Glaxo-
SmithKline preclinical drug candidates. The ®rst application
entailed the development of a convenient one-pot synthesis
of the functionalized pentacyclic core 10 /Fig. 2) of
compounds in GlaxoSmithKline's erbB family of protein
tyrosine kinase inhibitors,12 starting from commercially
2.2.1. Evaluation of selected secondary lithium amides
for in situ protection of 2-furaldehyde. We commenced
on this project by evaluating the four most ef®cient

1660
F. Roschangar et al. / Tetrahedron 58 02002) 1657±1666
R1
R2
ments /see Section 2.1). However, this study targeted the
Li
N
LiO
clean formation of 5-methyl-2-furaldehyde /20). We found
that the a-amino alkoxides 19, including the one derived
from strongly ortho-metal directing LTMDA /6), func-
tioned as protecting and not as ortho-directing groups.
This observation agreed with studies by Comins that
metalations of 18 via a-amino alkoxides occur exclusively
at the 5-position, regardless of the amine component.3b The
three major products of the reactions were 5-methyl-2-fural-
dehyde /20), recovered 18, and 1-/5-methyl-2-furyl)-1-penta-
nol /21). The results of these experiments are shown in
Table 2.
OHC
O
R1
N
O
-20 oC
R2
18
19
OHC
18
n-BuLi, -20 oC;
MeI, -20 oC;
O
O
Bu
O
HO
21
AcOH, 0 oC;
piperidine, 20 oC
OHC
20
We were delighted to ®nd that lithium N,O-dimethyl-
hydroxylamide /LDHA; 3) excelled as a protecting agent
in this series of experiments, yielding 91% of 5-methyl-2-
furaldehyde /20) in THF at 2208C, a result clearly superior
to that of its closest competitors, LNMP /5; 79% of 20; entry
6), and lithium morpholide /7, 74% of 20; entry 10). Lower-
ing the reaction temperature to 2408C seemed to have no
signi®cant effect /entry 3), but increasing it to 08C proved
detrimental /entry 4). As with the benzaldehyde series, the
use of THF as the solvent consistently provided improved
purity pro®les and better conversion compared to DME for
all tested protecting groups.
Scheme 3.
protecting groups of the benzaldehyde study /3, 5, 7, and 9)
with 2-furaldehyde /18) in regards to overall ef®cacy /ˆ%
yield-in-solution of 20; Scheme 3).
Assessment of a-amino lithium alkoxides 19 in THF and
DME at temperatures ranging from 0 to 2408C was
conducted in the same fashion as the benzaldehyde experi-
Table 2. Composition of mixtures for the reaction 18!19!18120121 for selected lithium amides
Entry
Lithium amide
Solvent
Corrected %
yield-in-
solution of
20a
Corrected %
yield-in-
solution of
18a
Corrected %
yield-in-
solution of
21a
1
DME
72.2
2.8
0.7
2
THF
THF
THF
90.7
90.5
56.9
1.4
5.4
2.3
1.1
0.0
0.4
3b
4c
5
6
7
DME
THF
68.6
79.2
38.6
4.6
2.0
6.3
0.5
0.9
0.8
DME
8
THF
40.0
2.2
0.5
9
DME
THF
DME
THF
69.8
74.3
43.3
57.7
5.0
3.6
7.4
2.0
0.4
0.1
1.0
0
10
11
12
a-Amino alkoxides were exposed to 2.3 equiv. of n-BuLi at 2208C for 1 h unless otherwise noted, and the reactions were quenched with MeI/AcOH.
Temperature was not allowed to rise more than 58C during reagent additions. % Yield-in-solution determined by HPLC.10,15
Corrected for HPLC response factors.
a
b
Reaction carried out at 2408C.
Reaction carried out at 08C.
c

F. Roschangar et al. / Tetrahedron 58 02002) 1657±1666
1661
2.2.2. Safety assessment. At this point, the thermal stability
of the new protecting agent 3 and its corresponding proto-
nated versions needed to be assessed,16 since we intended
not only to generate 5-formyl-2-furylboronic acid /12), but
also to directly advance this compound into the Suzuki
coupling step which would likely require elevated tempera-
tures. We found that N,O-dimethylhydroxylamine hydro-
chloride exhibited a broad melt at ca. 1108C followed by a
large exotherm ranging from 146±2398C with a peak energy
yield of 1.4 kJ/g. Assessment of the corresponding free base
under pseudo-adiabatic conditions resulted in a thermal
runaway event beginning at temperatures above ca. 508C.
Based on this assessment, we targeted a maximum operating
temperature of 208C during operations involving N,O-
dimethylhydroxylamine. Consequently, a low temperature
/,208C) vacuum distillation was used to remove the vola-
tile17 N,O-dimethylhydroxylamine following the boronic
acid generation and prior to the Suzuki reaction.
describing its preparation. These reported syntheses
typically suffer from low temperature requirements,
capricious reproducibility, tedious workup, as well as
unsuitably low purity and isolated yields /26±45%). Fortu-
nately, by applying a protocol similar to the one we had
developed for the synthesis of 5-/diethoxymethyl)-2-furyl-
boronic acid from 2-/diethoxymethyl)furan,13 we achieved
rapid success and found that formation of 5-formyl-2-furyl-
boronic acid /12) was ef®cient in both DME and THF /91
and 86% uncorrected yield-in-solution by HPLC at
lˆ280 nm, respectively). Due to the known dif®culties
associated with the isolation of this boronic acid, the
compound was directly advanced to the Suzuki coupling
step. Prior to this, we conducted a vacuum-assisted removal
of ca. one-third of the crude boronic acid solution's total
volume while maintaining the internal temperature below
208C, which ensured the safe purge of N,O-dimethyl-
hydroxylamine /see Section 2.2.2). The Suzuki coupling
with N-{3-chloro-4-[/3-¯uorobenzyl)oxy]phenyl}-6-iodo-
4-quinazolinamine /23) was cleaner and more facile in THF
and required only 4 mol% of dichloro[1,10-bis/diphenyl-
phosphino)-ferrocene]palladium/II) [PdCl2/dppf)] catalyst
vs. 8 mol% in DME. In summary, using THF as the reaction
solvent, we obtained quinazolinyl-2-furaldehyde 10 in a
one-pot operation starting from readily available 2-furalde-
hyde /18) after silica gel ¯ash column chromatography in
95% yield /based on 6-iodo-4-quinazolinamine 23 input).
2.2.3. One-pot synthesis of an anticancer drug candidate
intermediate. Encouraged by the favorable results of the
2-furaldehyde /18) study, and considering the thermal
instability data of N,O-dimethylhydroxylamine, we decided
to deploy LDHA /3) in a one-pot synthesis of the advanced
erbB
intermediate
quinazolinyl-2-furaldehyde
10
/Scheme 4).18 If successful, this sequence would involve
the combination of four distinct chemical transformations,
namely: /1) the protection of 18 as a-amino lithium
alkoxide 22, /2) its conversion to the boronate ester, /3)
the unmasking of the aldehyde and boronate ester moieties
to furnish 5-formyl-2-furylboronic acid /12), and /4) the
®nal Suzuki coupling step. Although Suzuki cross-coupling
reactions with in situgenerated boronic acids have been
described by us13 and others,19 we found no literature pre-
cedent for the combination of boronic acid syntheses via in
situ aryl aldehyde protection methodology with Suzuki
couplings in a one-pot operation.
2.3. Application to 4-bromo-2-¯uorobenzaldehyde
Motivated by the ef®cient application of lithium N,O-
dimethylhydroxylamide /LDHA; 3) to the 2-furaldehyde
/18) process, we set out to explore the versatility of the
new protecting agent with a second GlaxoSmithKline
project. This effort was aimed at a rapid one-pot assembly
of the bicyclic compound 11 /Fig. 2), a key intermediate to
novel insulin secretion modulator23 lead compounds, start-
ing from commercially available 5-bromo-2-¯uorobenzal-
dehyde /24). Although the reaction sequence leading to
formation of 4-¯uoro-3-formylphenylboronic acid /13)
may appear similar to the 2-furaldehyde /18) process at
®rst glance, the chemistry required stability of lithium /5-
bromo-2-¯uorophenyl)[methoxy/methyl)-amino]methoxide
/25; R1ˆMe, R2ˆOMe) towards halogen±metal exchange
conditions. Literature precedent indicated the viability of
this approach as Borchardt had previously utilized Comins'
methodology for a one-pot synthesis of hydroxybenzalde-
hydes, phthalaldehydic acids as well as phthalides via
bromine±lithium exchange of in situ protected bromo-
benzaldehydes.24 If successful, this second study would
signi®cantly broaden the versatility of LDHA /3) as a in
situprotecting agent for aryl aldehydes.
We were already aware that it would be challenging to
synthesize the seemingly simple furylboronic acid 12,20
following reference to publications21 and patents22
2.3.1. Evaluation of selected secondary lithium amides
for in situ protection of 4-bromo-2-¯uorobenzaldehyde.
We initially evaluated the four most ef®cient protecting
agents of the benzaldehyde study /3, 5, 7, and 9) with
5-bromo-2-¯uorobenzaldehyde /24) in regards to overall
ef®cacy /ˆ% yield-in-solution of 13; Scheme 5).
During these studies, we found that the lithiated benzene
derived from a-amino lithium alkoxides 25 was inherently
unstable and tended to decompose rapidly, even at
Scheme 4.

1662
F. Roschangar et al. / Tetrahedron 58 02002) 1657±1666
controlled addition of 3.0 equiv. of tert-butyllithium. We
R1
R2
F
F
OLi
assessed the ef®cacy of a-amino lithium alkoxides 25 in
DME and THF at 220 and 2408C by determining the
recovery of boronic acid 13. The two most signi®cant
side-products of these transformations were 2-¯uorobenzal-
dehyde /26) and recovered 24. The results of this study are
displayed in Table 3.
Li
N
CHO
R2
N
R1
-20 oC
24
25
Br
Br
F
We determined that lower temperatures provided signi®-
cantly better results /240 vs 2208C; entries 1 vs 2), and
that DME furnishes increased amounts of 26 /23 vs 8%;
entries 3 vs 1). Interestingly, unreacted 24 was only
observed as part of the product mixture when lithium
N,O-dimethylhydroxylamide /LDHA; 3) was used as
protecting agent. As in the previous two studies, 3 was at
least as ef®cacious as LNMP /5) and lithium morpholide /7)
and cleanly delivered 4-¯uoro-3-formylphenylboronic acid
/13) in THF at 2408C.
CHO
1. B(OiPr)3, -40 oC
2. tert-BuLi, -40 oC
F
CHO
Br
F
24
CHO
3. AcOH, 20 oC
4. H2O, 20 oC
H
26
13
B(OH)2
Scheme 5.
2.3.2. One-pot synthesis of a diabetes drug candidate
intermediate. We were now prepared to target the employ-
ment of LDHA /3) in an one-pot synthesis of the key insulin
secretion modulator intermediate 2-¯uoro-5-/3-nitro-2-
pyridinyl)benzaldehyde /11) /Scheme 6). The synthetic
sequence would involve the combination of four distinct
chemical transformations, including the protection±depro-
tection sequence of the formyl moiety, generation of the
boronic acid, and the Suzuki coupling.
temperatures as low as 2788C. Therefore, triisopropyl-
borate was selected as the electrophile, since this compound
could be present in the reaction mixture during addition of
the halogen±metal exchange reagent at low temperatures,
thereby reducing decomposition by instant trapping of the
lithiated species. While n-butyllithium at 2788C did afford
excellent reaction outcomes and clean formation of
4-¯uoro-3-formylphenylboronic acid /13), at temperatures
more suitable for manufacturing /2408C and above)
addition of n-butyllithium to triisopropylborate competed
effectively with the desired halogen±metal exchange
process. The issue could be alleviated by substituting tert-
butyllithium for n-butyllithium. Therefore, in this sequence
of experiments, the a-amino lithium alkoxides /25) were
treated with 3.0 equiv. of triisopropylborate followed by
Since we had already overcome the hurdle of preparing
phenylboronic acid 13 /see Section 2.3.1), we could focus
our immediate attention on the in situ advancement of this
material to the Suzuki coupling step. For safety reasons /see
Section 2.2.2), N,O-dimethylhydroxylamine was removed
in vacuo prior to the palladium-mediated cross coupling.
We discovered that EtOH was unsuitable as a Suzuki
Table 3. Composition of mixtures for the reaction 24!25!13124126 for selected lithium amides
Entry
Lithium amide
Temperature
/8C)
Normalized
% yield-in-
Corrected
% yield-in-
Corrected
% yield-in-
solution of 13a
solution of 24b
solution of 26b
1
240
80.5
0.3
7.5
2
220
240
67.7
50.3
7.4
9.2
8.0
23.4
3c
4
240
75.7
0
8.7
5
6
240
240
73.1
38.5
0
0
7.2
5.5
Reactions carried out in THF and quenched with AcOH unless otherwise noted. Temperature was not allowed to rise more than 58C during reagent additions. %
Yield-in-solution determined by HPLC.10
a
Normalized using 2-¯uorobenzaldehyde /26) as an external standard.25
Corrected for HPLC response factors.
Reaction carried out in DME.
b
c

F. Roschangar et al. / Tetrahedron 58 02002) 1657±1666
1663
Cl
F
O2N
F
N
CHO
CHO
27
PdCl2(dppf), Na2CO3
O2N
N
H2O, IPA, toluene, 73 oC
B(OH)2
13
11
(80%)
Scheme 6.
reaction solvent additive due to its propensity to add directly
to the highly electron-de®cient 2-chloro-3-nitropyridine
/27), delivering notable amounts of the undesired 2-
ethoxy-3-nitropyridine. This issue was circumvented by
the use of the solvent IPA. Furthermore, PdCl2/dppf) in
combination with aqueous sodium carbonate, toluene and
IPA were identi®ed as appropriate conditions to realize
excellent conversion. We thus attained 2-pyridinylbenzal-
dehyde 11 in a one-pot operation starting from commer-
cially available 5-bromo-2-¯uorobenzaldehyde /24) in
80% yield after silica gel ¯ash column chromatography
/based on 2-chloro-3-nitropyridine /27) input).
DSC 2910 and a Mettler Toledo DSC 2E, and are
uncorrected. IR spectra were recorded on a Nicolet
20DXC FT-IR spectrometer. 1H NMR and 13C spectra
were obtained in DMSO-d6 with Varian INOVA 300 and
Varian INOVA 400 NMR instruments, respectively, and
chemical shifts are reported in d values /ppm) relative to
the internal reference of DMSO-d6 /d 2.49 for 1H and d 39.5
for 13C spectra). Exact mass data were obtained on a Micro-
mass LCT time-of-¯ight mass spectrometer. Elemental
analyses were performed by Atlantic Microlab. N,O-
Dimethylhydroxylamide hydrochloride, 2-chloro-3-nitro-
pyridine, 5-bromo-2-¯uorobenzaldehyde and triisopropyl-
borate were purchased from Avocado, iodomethane from
Fluka, N-/2-methoxyethyl)methylamine from TCI America,
acetic acid and PdCl2/dppf) from Alfa Aesar, MeOH, IPA,
acetonitrile and toluene from EM Science, tert-butyllithium
/as a 2.0 solution in heptane) from FMC Lithium, and n-
butyllithium /as a 2.5 solution in hexanes), THF, DME,
EtOH, 2-/methylamino)pyridine, N,N,N0-trimethylethylene-
diamine, 1-methylpiperazine, morpholine, diethylamine,
triethylamine, the sodium carbonate solution /1.016 M in
water) and 2-furaldehyde from Aldrich. All reagents were
used without puri®cation or degassing, and all reactions
were performed under N2.
3. Conclusion
In summary, we have demonstrated lithium N,O-dimethyl-
hydroxylamide /LDHA; 3) to be a highly ef®cient and
weakly ortho-directing in situprotecting agent for aryl
aldehydes at temperatures convenient for large-scale
processing. In all three studies evaluating its potential,
amide 3 was at a minimum as ef®cient as its two most
effective literature-known counterparts, LNMP /5) and
lithium morpholide /7). Its corresponding a-amino lithium
alkoxide was shown to be stable to excess base and
halogen±metal exchange conditions. Furthermore, the new
protecting agent proved highly useful for the generation of
boronic acids 12 and 13, which were successfully employed
in situ for subsequent Suzuki reactions in four-step one-pot
operations towards the synthesis of developmental drug
candidate intermediates. To our knowledge, these are the
®rst examples of combining boronic acid syntheses via in
situ aryl aldehyde protection methodology with Suzuki
couplings in one pot. The shortcoming of LDHA /3),
namely the requirement of two equivalents of n-butyl-
lithium for its generation from commercially available
N,O-dimethylhydroxylamine hydrochloride, is offset by
the exceptional ef®cacy of the reagent. Moreover, the low
boiling point of N,O-dimethylhydroxylamine17 allows for
facile removal of the reagent from the reaction mixture by
low-temperature vacuum distillation, thus alleviating the
thermal hazards associated with its handling. In conclusion,
this reagent should prove to be a useful addition to the
arsenal of in situaryl aldehyde protecting groups. Further
applications of lithium amide 3 to current GlaxoSmithKline
projects are ongoing.
4.2. Ef®cacy screen of lithium amides
4.2.1. Benzaldehyde 214) seriesÐtypical procedure.26 To
a 3-necked, 100 mL round-bottom ¯ask equipped with an
electrical overhead stirrer and internal thermocouple was
added the amine /5.87 mmol) and solvent /THF or DME;
30 mL). The ¯ask was cooled to the desired internal
temperature using a Cryocool-controlled IPA bath. n-Butyl-
lithium /2.5 M solution in hexanes; 4.70 mL, 11.75 mmol
for generation of 3; 2.35 mL, 5.87 mmol for the generation
of 4±9) was added at a rate to maintain the internal tempera-
ture within 108C of the setpoint. The mixture was then stir-
red for 30 min, followed by addition of benzaldehyde /14;
0.50 mL, 4.89 mmol) and stirring for 15 min. n-Butyl-
lithium /2.5 M solution in hexanes; 4.50 mL, 11.26 mmol)
was added, and the mixture was stirred for 1±3 h. Iodo-
methane /1.53 mL, 24.47 mmol) was added, the mixture
was stirred for 30 min, and then treated with acetic acid
/0.34 mL, 5.87 mmol). The mixture was warmed to 08C,
at which time piperidine /2.43 mL, 24.47 mmol) was
added to quench the excess iodomethane. The resulting
exotherm rapidly increased the internal temperature to ca.
208C. The mixture was diluted successively with CH3CN
/40 mL), MeOH /5.00 mL) and DI water /5.00 mL). A
5.00 mL sample was removed from the resulting clear
yellow-orange solution and diluted with DI water
/5.00 mL) and CH3CN /40 mL) to a total volume of
4. Experimental
4.1. General
Melting points were determined using a TA Instruments

1664
F. Roschangar et al. / Tetrahedron 58 02002) 1657±1666
50 mL in a volumetric ¯ask. This solution was used directly
for HPLC analysis. While benzaldehyde /14) and 2-methyl-
benzaldehyde /16) were evaluated at 250 nm, 1-phenyl-1-
pentanol /17) was evaluated at 220 nm /corresponding to
their lmax values). The % yield-in-solution obtained for
each of the three products was corrected for response factors
derived from standard solutions of known concentrations.27
volumetric ¯ask. This solution was used directly for
HPLC analysis. The % yield-in-solution thus obtained for
4-¯uoro-3-formylphenylboronic acid /13) was calculated
using 2-¯uorobenzaldehyde /26) as an external standard.
This compound, as well as the side-products, 5-bromo-2-
¯uorobenzaldehyde /24) and 2-¯uorobenzaldehyde /26),
were evaluated at 245 nm /lmax for these materials). The
% yield-in-solution of the latter two products was corrected
for response factors derived from standard solutions of
known concentrations.30
4.2.2. 2-Furaldehyde 218) seriesÐtypical procedure.26
The secondary lithium amides were generated according
to the method described in Section 4.2.1, using 6.24 mmol
of amine, 30 mL of solvent /THF or DME) and the corre-
sponding amount of n-butyllithium /2.5 M solution in
hexanes; 5.00 mL, 12.49 mmol for generation of 3;
2.50 mL, 6.24 mmol for the generation of 5±7 and 9). The
mixture was treated with 2-furaldehyde /18; 0.43 mL,
5.20 mmol) and then stirred for 15 min. n-Butyllithium
/2.5 M solution in hexanes; 4.80 mL, 11.96 mmol) was
added and the mixture was stirred for 1 h. Iodomethane
/1.62 mL, 26.00 mmol) was added, the mixture was stirred
for 30 min, and then treated with acetic acid /0.36 mL,
6.24 mmol). The mixture was warmed to 08C, at which
time piperidine /2.57 mL, 26.00 mmol) was added to
quench the excess iodomethane. The resulting exotherm
rapidly increased the internal temperature to ca. 208C. In
preparation for the HPLC assay, the mixture was diluted
as described in Section 4.2.1. While 2-furaldehyde /18)
and 5-methyl-2-furaldehyde /20) were evaluated at
280 nm, 1-/5-methyl-2-furyl)-1-pentanol /21) was evalu-
ated at 220 nm /corresponding to their lmax values). The
% yield-in-solution obtained for each of the three products
was corrected for response factors derived from standard
solutions of known concentrations.28
4.3. Practical applications
4.3.1. 5-24-{3-Chloro-4-[23-¯uorobenzyl)oxy]anilino}-6-
quinazolinyl)-2-furaldehyde 210). In a 100 mL, 3-necked
round-bottom ¯ask equipped with an overhead stirrer and
internal thermocouple, N,O-dimethylhydroxylamine hydro-
chloride /623 mg; 6.26 mmol) was suspended in THF
/30 mL), and the ¯ask was cooled to 2208C /internal
temperature; Cryocool-controlled IPA bath). n-Butyllithium
/2.5 M solution in hexane; 5.00 mL, 12.49 mmol) was
added dropwise while maintaining the internal temperature
below 2108C. The reaction mixture was stirred at 2208C
for 30 min before adding 2-furaldehyde /0.43 mL;
5.20 mmol) at a rate to maintain the temperature below
2158C. n-Butyllithium /2.5 M solution in hexanes;
2.70 mL, 6.76 mmol) was added dropwise to this solution
while maintaining the internal temperature below 2158C.
After complete addition, the pale yellow mixture was
allowed to stir at 2208C for 75 min. Triisopropylborate
/1.84 mL, 7.81 mmol) was added dropwise, maintaining
the internal temperature below 2158C. The reaction
mixture was stirred for 15 min before adding acetic acid
/0.75 mL, 13.00 mmol) and warming to 158C. DI water
/470 mL, 26.02 mmol) was added, which resulted in an
exotherm that caused the internal temperature to rise to
ca. 208C. The resulting pale yellow slurry was stirred for
10 min. Approximately 10±15 mL of volatiles were
removed under vacuum, and the mixture was diluted to its
original solvent level with THF. The crude slurry of
5-formyl-2-furylboronic acid /12) was analyzed for purity
by HPLC /in CH3CN plus one drop DI water at lˆ280 nm;
85.7%) and then directly carried forward to the Suzuki
coupling reaction.
4.2.3. 5-Bromo-2-¯uorobenzaldehyde 224) seriesÐtypi-
cal procedure.26 The secondary lithium amides were gener-
ated according to the method described in Section 4.2.1,
using 6.24 mmol of amine, 25 mL of solvent /THF or
DME) and the corresponding amount of n-butyllithium
/2.5 M solution in hexanes; 5.00 mL, 12.49 mmol for
generation of 3; 2.50 mL, 6.24 mmol for the generation of
5, 7, and 9). The mixture was treated with 5-bromo-2-¯uoro-
benzaldehyde /24; 5.20 mmol, 6.41 mL of 16.8 wt% solu-
tion in THF or DME)29 and then stirred for 15 min.
Triisopropylborate /3.67 mL 15.60 mmol) was added,
resulting in a white precipitate. During addition of tert-
butyllithium /2.0 M solution in heptane; 7.80 mL,
15.60 mmol), the mixture brie¯y turned bright yellow
upon contact with the base. Upon completion of addition,
acetic acid /0.895 mL, 15.60 mmol) was added, and the
yellow mixture was allowed to stir for 5 min. The mixture
was warmed to 158C, at which time DI water /0.47 mL,
26.02 mmol) was added to hydrolyze the boronic ester.
The resulting exotherm increased the internal temperature
to ca. 208C. The mixture was stirred for 30 min and then
diluted with CH3CN /40 mL), MeOH /5.00 mL) and DI
water /5.00 mL). In contrast to the previous sets of experi-
ments /see Sections 4.2.1 and 4.2.2), a yellow-orange slurry
containing a white inorganic precipitate was formed. Thus,
stirring of the reaction vessel was suspended to allow for
settlement of the particles. A 5.00 mL sample was extracted
from the supernatant and diluted with DI water /5.00 mL)
and CH3CN /40 mL) to a total volume of 50 mL in a
To the crude boronic acid solution /4.46 mmol of 12
assuming 85.7% conversion) was added N-{3-chloro-4-
[/3-¯uorobenzyl)oxy]phenyl}-6-iodo-4-quinazolinamine /23)
/1.12 g, 2.23 mmol). The resulting mixture was treated
successively with EtOH /15 mL), NEt3 /930 mL,
6.69 mmol), and dichloro[1,10-bis/diphenylphosphino)-
ferrocene]palladium/II) /76 mg, 0.0892 mmol), heated to
608C /internal temperature), and stirred at this temperature
until completion of reaction was determined by HPLC
/using CH3CN plus one drop DI water at lˆ220 nm). The
reddish black mixture was cooled to 208C, directly loaded
onto silica gel and puri®ed by ¯ash column chromatography
Ê
/silica gel, 200±400 mesh, 60 A) using 80% EtOAc±hexane
containing 0.1% /v/v) NEt3. The eluants were concentrated
in vacuo to afford 0.96 g /95% yield) of 5-/4-{3-chloro-4-
[/3-¯uorobenzyl)oxy]anilino}-6-quinazolinyl)-2-furaldehyde
/10) as a yellow crystalline solid. An analytically pure
sample was obtained after re-slurry of the compound in

F. Roschangar et al. / Tetrahedron 58 02002) 1657±1666
1665
hot EtOAc /608C). Mp 229±2318C. IR /neat): nˆ3399,
NO2)vCH), 8.02 /dd, 1H, Jˆ6.6, 2.5 Hz, CF±C/±CHO)±
CH), 7.92 /ddd, 1H, Jˆ8.6, 5.0, 2.5 Hz, CF±CHvCH), 7.74
/dd, 1H, Jˆ8.3, 4.7 Hz, CvN±CHvCH), 7.54 /dd, 1H,
Jˆ10.2, 8.6 Hz, CF±CHvCH). 13C NMR /100 MHz):
dˆ187.4 /d, Jˆ4.6 Hz), 163.7 /d, Jˆ260.2 Hz), 152.9,
149.6, 145.6, 136.3 /d, Jˆ9.9 Hz), 133.6 /d, Jˆ3.8 Hz),
133.4, 129.1 /d, Jˆ2.3 Hz), 124.2, 123.8 /d, Jˆ9.2 Hz),
117.3 /d, Jˆ21.4 Hz). HRMS /ES pos.): m/z calcd for
1
1672, 1490, 1382, 1259, 777 cm21. H NMR /300 MHz):
dˆ10.09 /s, 1H, NH), 9.66 /s, 1H, CHO), 8.94 /s, 1H,
NvCH±N), 8.58 /s, 1H, furan-C±CHvC±C/±
NHAr)vN), 8.29 /d, 1H, Jˆ8.7 Hz, furan-CvCH±CH),
7.97 /d, 1H, Jˆ1.9 Hz, Cl±CvCH), 7.84 /d, 1H,
Jˆ8.7 Hz, furan-CvCH±CH), 7.74 /d, 1H, Jˆ3.7 Hz,
OHC±CvCH±CH), 7.71 /dd, 1H, Jˆ8.8, 1.9 Hz, ArHN±
CvCH±CH), 7.46 /q, 1H, Jˆ8.3 Hz, F±CvCH±CH), 7.40
/d, 1H, Jˆ3.7 Hz, OHC±CvCH±CH), 7.23±7.37 /m, 3H,
1
C12H8FN2O3 /M1H1): 247.0519. Found: 247.0510.
Anal. calcd for C12H7FN2O3: C, 58.54; H, 2.87; F, 7.72;
N, 11.38. Found: C, 58.12; H, 3.00; F, 7.59; N, 11.28.
ArHN±CvCH±CH,
F±CvCH±CHvCH,
F±C±
CHvC/±CH2)), 7.18 /bt, 1H, Jˆ8.3 Hz, F±CvCH±CH),
5.25 /2H, Ar-O±CH2). 13C NMR /100 MHz): dˆ177.9,
162.2 /d, Jˆ244.2 Hz), 157.8, 157.6, 155.3, 152.1, 150.2,
150.0, 139.6 /d, Jˆ6.9 Hz), 132.8, 130.6 /d, Jˆ8.4 Hz),
129.6, 128.8, 126.3, 125.8, 124.6, 123.3 /d, Jˆ3.1 Hz),
122.8, 121.0, 119.5, 115.3, 114.7 /d, Jˆ20.6 Hz), 114.2,
114.0 /d, Jˆ21.4 Hz), 109.8, 69.4. HRMS /ES pos.): m/z
Acknowledgements
We thank Ailette Aguila, Joanne E. Anderson, Roy C.
Flanagan, Bill C. Hinkley, N. Peter Kitrinos, Matthew I.
Lochansky, and Wendy L. White for technical support,
and John A. Corona, Roman Davis, Francis L. DeMartin,
Michael T. Martin, John Roberts, Tom D. Roper, and Maria
F. Tymoschenko for helpful discussions.
calcd for C26H18ClFN3O3 /M1H1): 474.1021. Found:
1
474.1007. Anal. calcd for C26H17ClFN3O3: C, 65.90; H,
3.62; Cl, 7.48; F, 4.01; N, 8.87. Found: C, 65.66; H, 3.56;
Cl, 7.42; F, 3.89; N, 8.88.
4.3.2. 2-Fluoro-5-23-nitro-2-pyridinyl)benzaldehyde 211).
4-Fluoro-3-formylphenylboronic acid /13) was prepared
according to the procedure described in Section 4.2.3,
using N,O-dimethylhydroxylamine hydrochloride /622 mg;
6.25 mmol) and THF /25 mL) at 2408C. After warming the
boronic ester solution to 158C, it was treated with 1.016 M
aqueous Na2CO3 solution /10.2 mL, 10.40 mmol). The
resulting exotherm increased the internal temperature to
258C. The mixture was stirred for 1 h and then diluted
with IPA /15 mL), followed by vacuum-assisted removal
of approximately 15 mL of volatiles. The crude slurry of
phenylboronic acid 13 was analyzed for purity by HPLC
/in CH3CN plus one drop DI water at lˆ220 nm; 93.8%)
and directly carried forward into the Suzuki coupling step.
References
1. Comins, D. L. Synlett 1992, 615.
2. /a) Comins, D. L.; Brown, J. D.; Mantlo, N. B. Tetrahedron
Lett. 1982, 39, 3979. /b) Comins, D. L.; Brown, J. D.
Tetrahedron Lett. 1981, 22, 4213.
3. /a) Comins, D. L.; Killpack, M. O. J. Org. Chem. 1990, 55, 69.
/b) Comins, D. L.; Killpack, M. O. J. Org. Chem. 1987, 52,
104.
4. Reetz, M. T.; Wenderoth, B.; Peter, R. J. Chem. Soc., Chem.
Commun. 1983, 406.
5. /a) Mentzel, M.; Hoffmann, H. M. R. J. Prakt. Chem. 1997,
339, 517. /b) Nahm, S.; Weinreb, S. M. Tetrahedron Lett.
1981, 22, 3815.
6. Lipshutz, B. H.; Pfeiffer, S. S.; Chrisman, W. Tetrahedron
Lett. 1999, 40, 7889.
To the 100 mL, 3-necked round-bottom ¯ask containing the
crude solution of 4-¯uoro-3-formylboronic acid /13;
4.87 mmol assuming 93.8% conversion) were added, at
ambient temperature, 2-chloro-3-nitropyridine /0.65 g,
4.06 mmol), dichloro[1,10-bis/diphenylphosphino)-ferro-
cene]palladium/II) dichloromethane adduct /100 mg,
0.12 mmol), and toluene /15 mL). The orange slurry was
heated to 738C /internal temperature; re¯ux) and stirred
until completion of reaction was indicated by HPLC
/using CH3CN plus one drop DI water at lˆ220 nm). Addi-
tion of a second batch of the PdCl2/dppf) catalyst /100 mg,
0.12 mmol) after ca. 2±3 h was required to complete the
reaction. The reddish black mixture was cooled to 208C,
directly loaded on silica gel and puri®ed by ¯ash column
7. /a) Stanforth, S. P. Tetrahedron 1998, 54, 263. /b) Suzuki, A.
Pure Appl. Chem. 1994, 66, 213. /c) Miyaura, N.; Yanagi, T.;
Suzuki, A. Synth. Commun. 1981, 11, 513.
8. See, for example: /a) Comins, D. L.; Nolan, J. M. Org. Lett.
2001, 3, 4255. /b) Hong, H.; Comins, D. L. J. Org. Chem.
1996, 61, 391. /c) Kelly, T. R.; Kim, M. H. J. Org. Chem.
1992, 57, 1593. /d) Haseltine, J. N.; Paz Cabal, M.; Mantlo,
N. B.; Iwasawa, N.; Yamashita, D. S.; Coleman, R. S.;
Danishefsky, S. J.; Schulte, G. K. J. Am. Chem. Soc. 1991,
113, 3850.
9. Benzaldehyde /14) and substituted benzaldehydes have been
evaluated in ortho-lithiation studies of various a-amino
alkoxides: Comins, D. L.; Brown, J. D. J. Org. Chem. 1984,
49, 1078.
Ê
chromatography /silica gel, 200±400 mesh, 60 A) using
30% EtOAc±hexane containing 0.1% /v/v) NEt3. The
eluants were concentrated in vacuo to furnish 0.80 g /80%
yield) of 2-¯uoro-5-/3-nitro-2-pyridinyl)benzaldehyde /11)
as a pale beige crystalline solid. An analytically pure sample
was obtained after recrystallization from warm toluene±
hexane /1:1; 608C). Mp 151±1528C. IR /neat): nˆ3077,
1684, 1604, 1515, 1494, 1446, 1400, 1350, 1262, 1225,
10. HPLC column: Phenomenex Luna C18/2) 50 mm£2.0 mm
3 mm. Mobile phase: Aˆ0.05% /v/v) TFA in H2O,
Bˆ0.05% /v/v) TFA in MeCN. Gradient pro®le: 0±95% B
over 8 min. 1.0 mL/min ¯ow-rate with injection volume of
1.0 mL.
11. The effect of donor solvents such as THF and DME on
aggregation and reactivity of organolithium species has been
investigated: /a) Streitwieser, A.; Juaristi, E.; Kim, Y.-J.;
Pugh, J. K. Org. Lett. 2000, 2, 3739. /b) Reich, H. J.; Green,
D. P.; Medina, M. A.; Goldenberg, W. S.; Gudmundsson,
1175, 1118, 843, 766 cm21
.
1H NMR /300 MHz):
dˆ10.25 /s, 1H, CHO), 8.95 /dd, 1H, Jˆ4.7, 1.4 Hz,
CvN±CH), 8.53 /dd, 1H, Jˆ8.3, 1.4 Hz, NvC±C/±

1666
F. Roschangar et al. / Tetrahedron 58 02002) 1657±1666
È
B.O.; Dykstra, R. R.; Philips, N. H. J. Am. Chem. Soc. 1998,
120, 7201. /c) Remenar, J. F.; Lucht, B. L.; Collum, D. B.
J. Am. Chem. Soc. 1997, 119, 5567.
21. /a) Starling, S. M.; Raslan, D. S.; de Oliveira, A. B. Synth.
Commun. 1998, 28, 1013. /b) Bracher, F.; Hildebrand, D.
Liebigs Ann. Chem. 1992, 1315. /c) Florentin, D.; Roques,
B. P.; Fournie-Zaluski, M. C. Bull. Soc. Chim. Fr. 1976, 13,
1999. /d) Florentin, D.; Roques, B. C. R. Acad. Sci. Paris, Ser.
C 1970, 270, 1608.
12. These compounds are potential targets for the treatment of
solid tumors: /a) Lackey, K. E.; Af¯eck, K.; Allen, P.;
Alligood, K. J.; Cai, Z.; Carter, M. C.; Cockerill, G. S.;
Crosby, R. M.; Dickerson, S.; Frye, S. V.; Gaul, M.; Gauthier,
C.; Gilmer, T. M.; Glennon, K.; Grif®n, R.; Guntrip, S. B.;
Guo, Y.; Johnson, N. W.; Keith, B. R.; Knight, W. B.; Luzzio,
M. J.; Mook, R. A.; Mullin, R. J.; Murray, D. M.; Rusnak, D.
W.; Tadepalli, S. M.; Sinhababu, A. K.; Smith, K. J.; Wood, E.
R.; Zhang, Y. -M. Proceedings of the American Association of
Cancer Research, 92nd Annual Meeting, 2001. /b) Cockerill,
G. S.; Lackey, K. E. International Patent WO-0104111, 2001.
13. McClure, M. S.; Roschangar, F.; Hodson, S. J.; Millar, A.;
Osterhout, M. H. Synthesis 2001, 1681.
22. Guerry, P.; Jolidon, S.; Masciadri, R.; Stalder, H.; Then, R.
International Patent WO-9616046, 1996, p 29.
23. Insulin secretion modulators have been demonstrated as
promising targets for type-2 diabetes and obesity therapy.
24. Sinhababu, A. K.; Borchardt, R. T. J. Org. Chem. 1983, 1941,
2356.
25. Due to dif®culties associated with the isolation of 4-¯uoro-3-
formylphenylboronic acid /13), an HPLC standard solution of
known concentration could not be prepared.
26. All reagent additions, following generation of the lithium
amides, were conducted at such a rate not permitting the
internal temperature to rise more than 58C.
27. Benzaldehyde /14), 2-methylbenzaldehyde /16), and 1-
phenyl-1-pentanol /17) reference solutions were obtained by
diluting 0.50 mL of the respective commercial compound with
CH3CN /90 mL), extracting a 5.00 mL sample, and diluting it
with CH3CN /45 mL) in a 50 mL volumetric ¯ask.
14. The cost-savings of boronic acid generation using the in situ
protection methodology vs starting from commercial 2-
/diethoxymethyl)furan would be signi®cant if both approaches
were equally ef®cient. Excluding labor and overhead costs, we
estimated ca. 80% savings for the in situgeneration of 12, with
material costs decreasing from ca. $1200/mol to ca. $250/mol.
The material costs were based on the following catalog prices:
2-furaldehydeÐ$2.21/mol /Aldrich); 2-/diethoxymethyl)-
furanÐ$1001.51/mol /Aldrich); N,O-dimethylhydroxyl-
amine hydrochlorideÐ$69.22/mol /TCI America).
28. 2-Furaldehyde /18) and 5-methyl-2-furaldehyde /20)
reference solutions were obtained by diluting 0.43 and
0.52 mL of the respective commercial compound with
CH3CN /90 mL), extracting a 5.00 mL sample, and diluting
it with CH3CN /45 mL) in a 50 mL volumetric ¯ask. The 1-/5-
methyl-2-furyl)-1-pentanol /21) marker was obtained by
reacting 2-furaldehyde /0.43 mL; 5.20 mmol) with n-butyl-
lithium /2.5 M solution in hexanes; 2.50 mL, 6.24 mmol) in
THF /30 mL) at 2788C. The reaction was quenched with
AcOH /0.39 mL, 6.76 mmol), warmed to 208C, diluted with
CH3CN /40 mL), DI water /5.00 mL) and MeOH /5.00 mL).
A 5.00 mL sample was extracted and diluted with DI water
/5.00 mL) and CH3CN /40 mL) in a 50 mL volumetric ¯ask.
29. Preparation of this solution prevented solidi®cation of the
reagent during addition.
15. The unaccounted % yield-in-solution was attributed to
decomposition as indicated by HPLC.
16. The thermal stability of N,O-dimethylhydroxylamine hydro-
chloride and its free-base were assessed using a Mettler
DSC12E instrument.
17. The boiling point of N,O-dimethylhydroxylamine has been
determined as ca. 428C at 760 Torr. See, for example:
Shirayev, A.; Iin, P. K. T.; Moiseev, I. K. Synthesis 1997, 38.
18. McClure, M. S.; Osterhout, M. H.; Roschangar, F.; Sacchetti,
M. J. International Patent Application WO-0202552, 2002.
19. /a) Alvarez, R.; Dominguez, B.; De Lera, A. R. Synth.
Commun. 2001, 31, 2083. /b) O'Neill, B. T.; Yohannes, D.;
Bundesmann, M. W.; Arnold, E. P. Org. Lett. 2000, 2, 4201.
30. 5-Bromo-2-¯uorobenzaldehyde /24) and 2-¯uorobenzalde-
hyde /26) reference solutions were obtained by diluting
2.98 mL of a 16.8 wt% solution of 24 in THF and 0.50 mL
of 26, respectively, with CH3CN /85 and 90 mL, respectively),
extracting a 5.00 mL sample from these solutions, and diluting
each sample with CH3CN /45 mL) in a 50 mL volumetric
¯ask.
Â
Â
/c) Baudoin, O.; Guenard, D.; Gueritte, F. J. Org. Chem. 2000,
65, 9268. /d) Andersen, N. G.; Maddaford, S. P.; Keay, B. A.
J. Org. Chem. 1996, 61, 9556. /e) Maddaford, S. P.; Keay,
B. A. J. Org. Chem. 1994, 59, 6501. /f) Cristofoli, W. A.;
Keay, B. A. Tetrahedron Lett. 1991, 32, 5881.
20. Although furylboronic acid 12 was available from a single
chemical supplier, its use was considered cost-prohibitive
/ca. $26,000/mol, available from Frontier Scienti®c).