Direct observation of the unstable intermediates in radical addition reaction by using an interfacing microchip combined with an NMR

Article abstract of DOI:10.1002/mrc.2087

Direct observation of the unstable intermediate in the radical addition reaction of the oxime ether 1 mediated by triethylborane (Et₃B) is described using ¹H and ¹¹B micro channeled cell for synthesis monitoring (MICCS), w

Full text of DOI:10.1002/mrc.2087

MAGNETIC RESONANCE IN CHEMISTRY
Magn. Reson. Chem. 2007; 45: 989–992
Published online in Wiley InterScience
(www.interscience.wiley.com) DOI: 10.1002/mrc.2087
Note
Direct observation of the unstable intermediates in
radical addition reaction by using an interfacing
microchip combined with an NMR
Masamichi Nakakoshi,1∗ Masafumi Ueda,2 Satoshi Sakurai,3 K. Asakura,3 Hiroshi Utsumi,3
Okiko Miyata,2 Takeaki Naito2 and Yutaka Takahashi3
1
Chemical Analysis Center, Yokohama National University, 79-5 Tokiwadai, Hodogaya-ku, Yokohama, 240-8501, Japan
Kobe Pharmaceutical University, Motoyamakita, Higashinada, Kobe 658-8558, Japan
JEOL Ltd. 1-2 Musashino 3-chome, Akishima, Tokyo, 196-8558,Japan
2
3
Received 18 April 2007; Revised 8 July 2007; Accepted 20 August 2007
Direct observation of the unstable intermediate in the radical addition reaction of the oxime ether 1
mediated by triethylborane (Et3B) is described using 1H and 11B micro channeled cell for synthesis
monitoring (MICCS), which was recently developed as an interfacing microchip for NMR. It was possible
that the signal of the intermediate was observed as a result of using MICCS technique with a standard
NMR instrument. This result supports the structure of the intermediate analyzed by diffusion-ordered
spectroscopy (DOSY) NMR method in a previous paper. The procedure of micro channeled cell for
synthesis monitoring-nuclear magnetic resonance (MICCS-NMR) was much easier than that of DOSY
method. It was proven that it could be applied to the reaction in an anhydrous condition. Copyright  2007
John Wiley & Sons, Ltd.
KEYWORDS: Nuclear magnetic resonance; MIcro channeled cell for synthesis monitoring; 11B; intermediate; oxime ether;
Et3B; radical reaction
this procedure, however the delay time between the reagents
introduction and the NMR measurement probably causes
problems in observing the short-term reaction progress and
the short-lived intermediates.
INTRODUCTION
Nuclear magnetic resonance (NMR) spectroscopy is used
as a routine method to determine the chemical structure of
organic compounds.1 Some researchers have requirements
to observe the progress of chemical reactions of organic
compounds. Analysis of reaction intermediates plays an
important role to elucidate the progress of chemical reactions.
As for a number of reactions, the structure analysis by NMR
is difficult because most of the reaction intermediates are
unstable.
We have reported the elucidation of the intermediate
structure in the radical reaction of the oxime ether 1 mediated
by triethylborane (Et3B) using 2D- and 3D-DOSY method.2
In this study, the intermediate was separately observed in
the reaction mixture. However, the preparation of reagent
solutions was complex. The reagent solutions were directly
mixed in an NMR sample tube under argon gas.
McGarrity et al. have reported rapid-injection NMR
methodology to observe the reactive intermediates in chem-
ical syntheses.3,4 The observation of reactive intermediates
with half-lives of around 100 ms at a concentration level
below 10ꢀ1 mol/l was possible with this methodology, and
it might be effective for these kinds of applications. A demerit
of this injection system was that it requires specially assem-
bled equipment on the NMR instrument. This hampered
normal organic chemists to easily adopt the methodology.
Kakuta et al. have reported a study regarding changes in
protein conformation using the combined technique with a
static mixer and a microcoil-based microfluidic NMR probe.5
This technique has the advantage to detect small volume
of samples. However, it is necessary to use a specially
assembled NMR probe for this study.
The direct mixing of chemical reagents in an NMR
sample tube was carried out to observe the progress of
chemical reactions and the unstable intermediates. In using
We have developed a new microchip, which was able to
monitor a reaction process by combination with a standard
NMR instrument.6 This device was named MICCS. MICCS-
NMR has several merits in the field of organic chemistry.
ŁCorrespondence to: Masamichi Nakakoshi, Chemical Analysis
Center, Yokohama National University, 79-5 Tokiwadai,
Hodogayaku, Yokohamashi, 240-8501, Japan.
E-mail: m-nakako@ynu.ac.jp
1. The mixing parts and the detection part are very close.
2. Three reagent solutions can be introduced at same time.
Copyright  2007 John Wiley & Sons, Ltd.

990
M. Nakakoshi et al.
was 5.0 s, acquisition time was about 5.5 s). For 11B MICCS-
NMR, single pulse with 1H decoupling was used and number
3. The mixing ratios of reagent solutions and the reaction
temperature were easily controlled by adjusting the flow
rates of reagent delivery system and probe temperature.
4. When the reaction conditions were stable, the reaction
mixture of stable state was continuously delivered from
the mixing part to the detection part.
°
of scans were 32. The parameter of the 90 pulse width was
20 µs, of the repetition time was about 1.0 s (relaxation delay
was about 0.5 s, acquisition time was about 0.5 s). Chemical
solutions were delivered into micro-channels using syringe
pumps (MD1001 with the BS-MD1200, Bioanalytical Systems
Inc., West Lafayette, IN). Syringes and MICCS were coupled
with fused-silica capillary tubings (150 µm ID ð 375 µm OD)
purchased from GL Science (Tokyo, Japan).
Acetone-d6, purchased from Sigma-Aldrich Inc. (St Louis,
MO), was used as an external lock. It was injected into the
gap between the outside wall of MICCS and the inside wall
of the 5-mm O.D. sample tube (MICCS was inserted into the
sample tube.). CH2Cl2 and methanol were purchased from
Tokyo Chemical Industry Co. Ltd. (Tokyo, Japan). CH2Cl2
was dehydrated grade. The design of flow route of MICCS
and the schematic diagram of MICCS-NMR are shown in
Fig. 1.
CH2Cl2 solutions of oxime ether 1 (1 mol/l) and triethylb-
orane (2) (1 mol/l) were introduced into the two inlet ports of
MICCS by syringe pumps. Both solutions were mixed at the
1st Y-shaped channel in Fig. 1, and a radical addition reaction
occurred. CH2Cl2 solution of methanol was introduced into
the 3rd inlet port and was mixed with reaction mixture at the
2nd Y-shaped channel to quench the reaction. The reaction
temperature was not controlled (room temperature), because
temperature control is not necessary in this reaction.8 The
flow rates of the above three solutions were changed while
consecutive measurements were taken as shown in Table 1.
The flow-rate ratio of 1/2 was consecutively changed from
10/0.5 to 1.0/5.0 µl/ min. At the time, when flow-rate ratio
of 1/2 is 5.0/5.0 µl/ min, the intensity of the intermediate
According to these merits, this technique has the
following advantages over other methods.
1. Detection of short-lived intermediates is possible.
2. Two-step chemical reactions can be observed.
3. Examination of the reaction conditions is very easy by a
real-time monitoring with NMR.
4. Long-time integration to observe small amount of prod-
ucts and the unstable intermediates would be possible.
From these characteristics, the direct observation of the
reaction progress, which is the changes in the raw materials,
unstable intermediates and products, would be possible by
the measurement of the nucleus that has short relaxation
time (e.g. 11B). In this paper, 1H and 11B MICCS-NMR were
used to observe the intermediate in this reaction to evaluate
the applicable fields of MICCS-NMR technology. The results
were compared to those of DOSY method of the previous
paper.2
EXPERIMENTAL
1H and 11B MICCS-NMR were carried out using a JEOL JNM-
ECA-500 (Tokyo, Japan) equipped with a TH5FG probe.
WET7 was used for solvent suppression in 1H MICCS-NMR
and the number of scans were 16, The parameters for 1H
°
measurement were as follows: The 90 pulse width was
13.5 µs, repetition time was about 10.5 s (relaxation delay
(a)
Fused-silica capillary
(b)
Exhaust
XY alignment
guide
MICCS
5 mmφ sample tube
D-solvent for NMR lock
Figure 1. (a) The design of flow route of MICCS and (b) the schematic diagram of MICCS-NMR.
Copyright  2007 John Wiley & Sons, Ltd.
Magn. Reson. Chem. 2007; 45: 989–992
DOI: 10.1002/mrc

Direct observation in organic reaction by MICCS-NMR 991
was expected to be maximum. After that, by adding CH2Cl2
solution of methanol, it was expected that the intermediate
would change to the final product.
On the basis of our prior work, the pathway of this
reaction was proposed in Scheme 1. Structure of the
intermediate 4 was proposed by 2D- and 3D-DOSY method
as described in the literature.2
materials, the unstable intermediates andthe products, was
observed by using MICCS-NMR technique. Complex signals
at υ 0.5 ꢀ 1.0 ppm corresponded to the ethyl group bonded
to boron element. Since MICCS-NMR gives high-resolution
spectra as described in the literature,6 the signals of ethyl
group of the raw material 2 and the intermediate 4 could be
distinguished.
RESULT AND DISCUSSION
A 2D-display of the 1H-NMR spectra is shown in Fig. 2.
Sliced spectrum at the flow rates of 1 and 2 of 5 µl/ min
each and the expected structure of the intermediate 4 are
shown in Fig. 3. Table 2 shows the assignment of the 1H
signal for the structure 4. The spectrum shown in Fig. 3
is consistent with the structure of the intermediate for 4,
as reported in the literature.2 Three characteristic signals
of 4 were observed at υ 1.8, υ 4.0 and υ 4.5 ppm. In
Fig. 2, the spectra of 4, which contain these three signals,
were clearly observed at the spectra # 3–7. And then,
by adding methanol, these three signals disappeared and
different signals at υ 1.4 and υ 4.5 ppm were observed at
the spectra # 8–11. These spectra were consistent with the
structure of the final product 3. We could obtain not only the
structural information of 4 but also that of 3 at single analysis.
Therefore, the reaction progress, which is changes in the raw
Table 1. The flow rates of the three reagent solutions in 1
H
NMR experiment
Spectrum
no.
Oxime ether 1
(µl/ min)
2
CH3OH
(µl/ min)
(µl/ min)
Figure 2. 2D-display of the 1H-NMR spectra.
1
2
3
4
5
6
7
8
10.0
9.0
8.0
7.0
6.0
5.0
5.0
4.5
4.0
3.5
2.0
1.0
0.5
1.0
2.0
3.0
4.0
5.0
5.0
4.5
4.0
3.5
4.0
5.0
0.0
0.0
0.0
0.0
0.0
0.0
0.5
1.0
2.0
3.5
2.0
5.0
Table 2. The 1H-NMR assignment of the intermediate 4
No.
ppm
1
2
3.97(1H,dd,J D 8.9,6.1Hz)
1.79(2H, m)
3
0.72(3H)
10
4.51(1H,d,J D 9.8) and 4.57(1H,d,J D 9.8)
0.93(3H) and 0.76(3H)
0.80(3H) and 0.72(3H)
3.58(3H,s)
9
100
200
OCH3
Phenyl
10
11
12
7.12–7.18(5H)
O
O
H
Et3B(2),CH2Cl2,rt
NOBn
N
Bn
H CO
3
H3CO
O
H
3
Et
H
1
H2O
Et3B
Et
.
O2
Et
.
O
Et3B
BEt3
O
O
BEt2
BEt3
NOBn
.
NOBn
NOBn
H3CO
H3CO
H3CO
H
Et
H
Et
H
A
B
4
Scheme 1. Proposed the mechanism on the radical reaction of Et3B and oxime ether 1.
Copyright  2007 John Wiley & Sons, Ltd.
Magn. Reson. Chem. 2007; 45: 989–992
DOI: 10.1002/mrc

992
M. Nakakoshi et al.
Figure 3. Sliced 1H-NMR spectrum of the intermediate 4.
A 2D-display of the 11B-NMR spectra was shown in Fig. 4.
The total flow rate of the two solutions was 4.0 µl/ min. The
flow-rate ratio of the two solutions was changed during
measurement (v (1)/v (2); 4/0, 3/1, 2.5/1.5, 2.0/2.0, 1.5/2.5,
1/3 and 0/4). Methanol was not used in this experiment.
The intensity of the signals at υ 86.6 ppm assigned to
triethylborane (2) was increased with increasing the flow
rate. The intensity of the signals at υ 53.9 ppm associated
to 4 was maximum at the flow rates of 2 µl/ min each for
1 and 2. This chemical shift was also consistent with the
structure of 4. The signals at υ 60.0 ppm would be assigned
to the decomposed compound of 2. Because of the merits
of MICCS-NMR (# 1 and # 4) as described above, 11B-NMR
analysis was possible in this study. On the other hand, 11B-
NMR could not analyze in the DOSY method, since 11B
nucleus has short relaxation time.
Figure 4. 2D-display of the 11B-NMR spectra.
the intermediate elucidated by 2D- and 3D-DOSY method
as described in previous paper. The reagent preparation
of MICCS-NMR was much easier compared to that of
DOSY. MICCS-NMR could be expected to be a commercially
available tool to analyze the reaction intermediates and the
reaction progress in a chemical reaction.
Acknowledgements
This work was supported by ‘‘Project of Micro-Chemical Technology
for Production, Analysis and Measurement Systems’’ of New Energy
and Industrial Technology Development Organization (NEDO),
Japan.
The results using MICCS method were obtained much
easier than DOSY method. In DOSY method, the reagent
solutions were carefully prepared and mixed in an NMR
sample tube under argon gas to avoid the decomposition
of the intermediate. On the other hand, in MICCS method,
the reagent solutions were prepared under the atmospheric
condition. And the required time for MICCS-NMR measure-
ment was shorter than that for DOSY. This device can be
used under the anhydrous condition in the organic reaction.
Moreover, the structural information of the intermediate
could be obtained even in the multiple nuclear NMR.
REFERENCES
1. Gunther H. NMR Spectroscopy; Basic Principles, Concepts and
Applications in Chemistry (2nd edn). John Wiley & Sons. 1996;
221.
2. Nakakoshi M, Ueda M, Sakurai S, Miyata O, Sugiura M, Naito T.
Magn. Reson. Chem. 2006; 44: 80.
3. McGarrity JF, Prodolliet J, Smyth T. Org. Magn. Reson. 1981; 17:
59.
4. McGarrity JF, Prodolliet J. J. Org. Chem. 1984; 49: 4465.
5. Kakuta M, Jayawickrama DA, Wolters AM, Manz A, Sweedler JV.
Anal. Chem. 2003; 75: 956.
6. Takahashi Y, Nakakoshi M, Sakurai S, Akiyama Y, Suematsu H,
Utsumi H, Kitamori T. Anal. Sci. 2007; 23: 395.
7. Smallcombe SH, Patt SL, Keifer PA. J. Magn. Reson. 1995; 117A:
295.
8. Miyabe H, Ueda M, Yoshioka N, Yamakawa K, Naito T.
Tetrahedron 2000; 56: 2413.
CONCLUSION
The direct observation of the unstable intermediate in
radical addition reaction was succeeded using a 1H- and
11B MICCS-NMR. These results support the structure 4 for
Copyright  2007 John Wiley & Sons, Ltd.
Magn. Reson. Chem. 2007; 45: 989–992
DOI: 10.1002/mrc