Alkylation of substituted benzoic acids in a continuous flow microfluidic microreactor: Kinetics and linear free energy relationships

Article abstract of DOI:10.1021/op300085w

Alkylation of para-substituted benzoic acids by iodomethane using an organic superbase, 1,8-bis(tetramethylguanidino)naphthalene (TMGN) in DMF was chosen as a model reaction to test the quality of the control of experimental parameters in a continuous flow microfluidic reactor as it is expected to follow a perfect second order kinetics with a large dynamics by varying the substituents. These conditions may be directly used for the synthesis of natural product esters. Because TMGN reacts slowly with iodomethane, the three different mixing strategies between substrate, base and alkylating reagent were compared. The rate constants were determined for the reaction with a set of alkylating agents and in different solvents. In order to test the quality of the obtained data, temperature effect and free energy relationships, which are expected to follow predictable laws, were investigated. The kinetics vary over 6 orders of magnitude and follows a perfect Arrhenius law, allowing the determination of the energies, enthalpies, and entropies of activation. Finally, we established a Hammett linear relationship for a series of 16 substituted benzoic acids, leading to a reaction constant ρ of -0.65 for this reaction. The quality of the obtained kinetics allowed us to discuss the outliers. All kinetics were obtained with less than 0.5 mmol of substrate.

Full text of DOI:10.1021/op300085w

Article
pubs.acs.org/OPRD
Alkylation of Substituted Benzoic Acids in a Continuous Flow
Microfluidic Microreactor: Kinetics and Linear Free Energy
Relationships
Azarmidokht Gholamipour-Shirazi and Christian Rolando*
USR CNRS 3290, Miniaturisation pour la Synthes
̀
e, l’Analyse et la Proteo
́ ̀
mique and FR CNRS 2638 Institut Michel-Eugene
Chevreul, Universite
́
de Lille 1, Sciences et Technologies, 59655 Villeneuve d’Ascq, France
S
* Supporting Information
ABSTRACT: Alkylation of para-substituted benzoic acids by iodomethane using an organic superbase, 1,8-bis-
(tetramethylguanidino)naphthalene (TMGN) in DMF was chosen as a model reaction to test the quality of the control of
experimental parameters in a continuous flow microfluidic reactor as it is expected to follow a perfect second order kinetics with a
large dynamics by varying the substituents. These conditions may be directly used for the synthesis of natural product esters.
Because TMGN reacts slowly with iodomethane, the three different mixing strategies between substrate, base and alkylating
reagent were compared. The rate constants were determined for the reaction with a set of alkylating agents and in different
solvents. In order to test the quality of the obtained data, temperature effect and free energy relationships, which are expected to
follow predictable laws, were investigated. The kinetics vary over 6 orders of magnitude and follows a perfect Arrhenius law,
allowing the determination of the energies, enthalpies, and entropies of activation. Finally, we established a Hammett linear
relationship for a series of 16 substituted benzoic acids, leading to a reaction constant ρ of −0.65 for this reaction. The quality of
the obtained kinetics allowed us to discuss the outliers. All kinetics were obtained with less than 0.5 mmol of substrate.
acids without quaternization of their protected amine group
and using N,N′-dimethylformamide (DMF) as solvent.7a,b The
scope of this reaction has been extended to the alkylation of
crowded carboxylic acids using hexamethylphosphoramide
(HMPA) as solvent7c and has been used in several syntheses7d,e
including the synthesis of short-lived 11C propyl and butyl
esters.7f Kondo et al. demonstrated that the reaction of
tetramethylammonium benzoate salts with alkyl halides in
acetonitrile follows a second-order kinetics.8 Instead of using
cesium or tetraalkylammonium salts, Ono et al. used 1,8-
diazabicyclo[5.4.0]-undec-7-ene (DBU) as the base to
efficiently deprotonate benzoic acid in toluene at room
temperature.9a However, in these conditions the obtained
DBUH+I− salt is insoluble, and the resulting white slurry
precludes its use in a microsystem. In another study, Mal et al.
employed the same procedure for the O-methylation of various
carboxylic acids in acetone and in acetonitrile as solvent.9b
However, one of the most serious side reactions in these
syntheses is the alkylation of DBU by iodomethane. Barton et
al. reported that the hindered guanidine bases they synthesized
were much more stable toward alkylation.10a,b Barton’s bases
enable the alkylation of crowded carboxylic acid such as
adamantane-1-carboxylic acid even with a secondary alkyl
halide such as isopropyliodide.10c This reaction has been used
during the total synthesis of salinomycin.10d In order to apply
the results of this work to the synthesis of more complex
molecules such as phenol acids we used DMF as solvent, which
was a good solvent in carboxylate cesium or sodium salt
alkylation.11 Furthermore, the high polarity of DMF avoids the
INTRODUCTION
■
Chemistry in continuous flow microreactors has received
considerable attention over the past decade.1 In microreactors,
potentially explosive and hazardous reactions can be safely
conducted,2a−c short-lived intermediates can be trapped to
increase chemical yield,2d a cascade of reactions can be carried
out without the necessity of isolating intermediates, and
alternatively it is possible to use high-pressure and/or
-temperature conditions.2e Continuous flow microreactors
have found broad applications in multistep organic synthesis3a
and in the synthesis of complex natural products.3b−d One
more advantage of microreactors is that they provide an
opportunity for greener chemistry and faster process
development.3e−g Scale-up of microreactors can be easily
achieved by using multiple microreactors in parallel.4 The
efficiency of a given reaction in a microreactor compared to that
in batch relies critically on the mixing process of the
reagents.5a−e Most of the mixing evaluation reactions are
based on highly nonlinear reactions such as Bourne’s
reactions,5b−d and the iodine-iodate Villermaux−Dushman
reaction, which leads to complicated kinetics valid only in a
limited range of concentrations.5e Furthermore, these conven-
tional methods are based on spectrophotometric determination
of the products rather than using the isolated product yield.6
We were interested, therefore, to find and to study in a
continuous flow microreactor, a second-order reaction with
isolable products and variable rates by varying the substrate
without changing the reaction kinetics. The reaction of
substituted benzoate with iodomethane, which is a well-
known SN2 reaction, seems to fulfill these criteria. Chlorobenzyl
Merrifield resins (chloromethylated polystyrene−1% divinyl-
benzene) were efficiently alkylated by cesium salts of amino
Received: March 31, 2012
Published: April 17, 2012
© 2012 American Chemical Society
811
dx.doi.org/10.1021/op300085w | Org. Process Res. Dev. 2012, 16, 811−818

Organic Process Research & Development
Article
Figure 1. Experimental setup. It comprises a commercially available micromixer (NanoMixer) and a fused silica-based capillary tubular reactor.
Figure 2. Second-order kinetics plot of benzoic acid alkylation by iodomethane in the presence of TMGN and in DMF at 20 °C.
formation of aggregates or strong ion pairs which complicate
the kinetics of the reaction.12 We chose 1,8-bis-
(tetramethylguanidino)naphthalene (N″,N′′′′′-1,8-naphthalene-
diyl-bis[N,N,N′,N′-tetramethyl]-guanidine, TMGN) as a base
as it is even less reactive than Barton’s bases toward
alkylation.13a Its backbone is the well-known ‘proton sponge’
1,8-bis(dimethylamino)naphtalene (DMAN) which does not
react with methyl iodide but which, unfortunately, is not basic
enough.13b,c The ionization constant (pKBH+) of TMGN in
acetonitrile (25.1) is higher than that of DBU (24.33).14a−c By
using the linear correlations between acidities in DMF and in
acetonitrile, pKBH+ of TMGN in DMF is estimated to range
from 16.4 to 17.5.14d pKa’s of benzoic acids in DMF are in the
range of 10.6 for 4-nitrobenzoic acid to 13.0 for 3,4-
dimethylbenzoic acid.14e Therefore, TMGN is able to fully
remove acidic hydrogen of all benzoic acids. Furthermore, in
those conditions either for DBU or TMGN the salt of the
protonated base cation with iodide remains soluble in DMF,
which is a requirement for experimental studies in micro-
reactors.
While several groups have developed their own continuous
flow microreactors dedicated to organic synthesis,15 the setup
we used in this study has the advantage of being based on
commercially available devices and thus can be reproduced
easily. We present here results obtained with this setup on the
kinetics of alkylation of substituted benzoic acid deprotonated
by TMGN, which is focused on the comparison of three mixing
strategies of the three reagents, TMGN, benzoic acid, and
iodomethane, which may be mixed in any combination, the
influence on the rate of the alkylating agents and of the solvent,
the temperature dependence of reaction rate, and finally the
linear free energy relationships for this reaction. The linear
correlation obtained shows that continuous flow microreactors
may be used in physical chemistry experiments with the
consumption of very small amounts of reagents.
RESULTS AND DISCUSSION
■
Fused glass capillary tubes with inherent microscale internal
dimensions provide modular and inexpensive building blocks
for the on-demand assembly of microfluidic reactors.16 All the
microfluidic experiments were carried out in a setup (Figure 1)
composed of two streams of reagent solutions in DMF
simultaneously delivered by a high-pressure syringe pump to
a micromixer followed by a fused silica-based capillary tubular
reactor which are all readily available commercial devices. The
two syringes containing the reagents are connected via 0.30 m
capillaries (i.d. 50 μm) to the micromixer. Because of the high
pressure drop along these inlet tubes, the liquid flows only in
one direction, and no backmixing occurs. The micromixer is
followed by a 3.0 m fused silica-based capillary (i.d. 75 μm)
which is the tubular reactor.17 The capillary tubular reactor is
kept at the desired temperature in a water bath. For the
micromixer, we utilized a commercially available multilaminat-
ing distributive micromixer chip.17a This mixer has been used
previously for time-resolved studies of protein conformation by
NMR.17b A similar PDMS device has also been used for the
controlled polymerization of N-carboxy anhydrides.17c It is one
of the more efficient mixing devices in this flow range.17d In our
case, slightly worse results were obtained with a simple
MicroTee filled with porous material.
812
dx.doi.org/10.1021/op300085w | Org. Process Res. Dev. 2012, 16, 811−818

Organic Process Research & Development
Article
The flow rates were varied from 5 μL·min−1 to 150 μL.min−1
to give residence times of 5 to 105 s (SI Table S1). This flow
rate range corresponds to a Bodenstein number Bo varying
from 90 to 1810. For these large Bo numbers, no dispersion
occurs, and plug flow is assumed in the capillary tubular
reactor.18 The concentrations of the remaining benzoic acid
and methyl benzoate were determined by GC/MS with
electron ionization (EI) after silylating the quenched reaction
mixture. To ensure the quality and integrity of the data
generated, we added three internal standards, one in each
syringe and another one in the collected sample. The absolute
average and the relative average deviation were 3% and less
than 0.5%, respectively (SI Tables S2 and S3). The conversion,
f, used in kinetics analysis, was calculated from the integrated
areas of benzoic acid methyl ester and silyl ester peak (f =
(peak area of methyl ester)/(peak area of methyl ester + peak
area of silyl ester)) since it is more reliable than the absolute
area of remaining benzoic acid or of produced methyl ester
alone. The kinetics constants k were determined graphically by
plotting the function 1/(1 − f), against residence time (t)
which resulted in a straight line with a slope equal to
k×[YC7H5O2 ]0, where [YC7H5O2 ]0 is the initial concen-
tration of the substituted benzoic acid. In all the experiments,
unless otherwise mentioned, the solution of benzoic acid and
base (TMGN) in DMF was in one syringe, whereas the
solution of the alkylating reagent, MeI, also in DMF, was kept
alone in the second syringe.
Figure 2 shows the results obtained for an initial
concentration of benzoic acid of 26.7 mM and molar ratios
of TMGN and iodomethane to benzoic acid of 1.0 and 1.1,
respectively. These concentrations have been applied for all the
experiments described in this paper. The straight line
relationship (n = 7; R2 = 0.999) which we observed up to a
conversion of 60% shows that the second-order kinetics not
only is observed at the initial stage of the reaction but also
remains verified up to a nearly preparative yield.
were repeated several times using different capillaries and
micromixer units and at different periods. From these data the
dispersion of results may be estimated to be around 4%.
However, the last combination (benzoic acid in one syringe,
TMGN and iodomethane in another syringe) does not follow a
clean second-order kinetics for residence times beyond 60 s and
gives a lower reaction constant nearly half of the previous value
(Table 1, entry c). Investigation by mass spectrometry and
NMR techniques show that the side reaction is the hydrolysis
of MeI as well as TMGN protonation by residual water.
Improved results were obtained using DMF dried overnight
over molecular sieves and using a small amount of molecular
sieve in each syringe (Table 1, entry d).
Effect of the Alkylating Reagent and Solvent. We next
investigated a set of alkylating reagents. For all alkylating agents
the reaction remained cleanly second order. The observed
reaction rate constants are displayed in Table 2. The k value
Table 2. Effect of alkylating reagent and solvent on the
reaction rate constant of benzoic acid alkylation in the
a
presence of TMGN
−
−
temperature
k
b
entry
solvent
DMF
alkylating reagent
(°C)
(mol−1·L·s−1
)
b
a
iodomethane
iodoethane
20
20
50
20
20
0.57 0.02
b
c
DMF
DMF
DMF
DMF
0.063 0.006
0.039 0.001
0.28 0.05
2-iodopropane
benzyl bromide
d
e
tert-butyl
bromoacetate
0.74 0.03
f
acetonitrile iodomethane
toluene iodomethane
20
20
0.074 0.01
g
−
a
Conditions: benzoic acid 26.7 mM and molar ratios of TMGN and
iodomethane to benzoic acid of 1.0 and 1.1, respectively, in HPLC
b
grade solvents used as received. Average and standard deviation from
two or three independent measurements.
There are two other combinations for introducing the three
reagents (benzoic acid, the base TMGN, and iodomethane)
into the micromixer. Since it might lead to different selectivity
in case of complex molecules, their kinetics were also recorded.
The two combinations in which reagents cannot react
irreversibly (i.e., benzoic acid and TMGN in one syringe and
iodomethane in another syringe or benzoic acid and iodo-
methane in one syringe and TMGN in another syringe) gave
quite similar rate constants 0.57 and 0.55 mol−1·L·s−1,
respectively (Table 1, entries a, b) close to the value obtained
in batch which is 0.64 mol−1·L·s−1. These three experiments
order observed for iodomethane, iodoethane and 2-iodopro-
pane, respectively 0.57, 0.063, 0.039 mol−1·L·s−1 (Table 2,
entries a, b, and c), follows the expected trend for methyl,
primary, and secondary alkyl halides in SN2 reactions.
No reaction was observed for 2-iodo-2-methylpropane
suggesting that elimination to 2-methylpropene is the main
reaction. Benzyl bromide and tert-butyl bromoacetate were very
reactive with k values of 0.28 and 0.74 mol−1·L·s−1, respectively
(Table 2, entries d and e). Roberts et al. reported that the
reactivity of tert-butyl bromoacetate was 2.2 times higher than
that of benzyl bromide in the alkylation of the cysteine thiol of
glutathione in water/DMSO (between 10 and 20%).19 Here
the observed ratio in DMF is 2.7. The effect of the solvent on
the reaction rate constants was also investigated and as
expected the reaction in acetonitrile is much slower than in
DMF with k values of 0.074 and 0.57 mol−1·L·s−1 respectively
(Table 2 entries a, f). The rate observed in acetonitrile is one-
third the value (0.26 mol−1·L·s−1) which can be obtained using
the Hammett equation given by Kondo et al.8d using the
preformed ion with tetramethylamonium as counterion. No
trace of the expected ester was observed by GC/MS when the
reaction was conducted in toluene at 20 °C (Table 2, entry g).
This result is at first glance surprising since Ono et al.9a
described the alkylation of benzoic acid by iodoethane using
DBU as base in toluene at room temperature. These conditions
(DBU, toluene) cannot be tested in our microdevice because
by reproducing the experiment in batch we observed that white
Table 1. Effect of reagent combination on the reaction rate
constant of the alkylation of TMGN deprotonated benzoic
a
acid by iodomethane in DMF at 20 °C
entry
syringe A
syringe B
iodomethane
k (mol−1·L·s−1
)
b
,
c
a
benzoic acid, TMGN
0.57 0.02
b,
d
b
c
benzoic acid, iodomethane TMGN
0.55 0.07
b
benzoic acid
benzoic acid
iodomethane, TMGN 0.24
e
d
iodomethane, TMGN 0.32
a
Conditions: benzoic acid 26.7 mM and molar ratios of TMGN and
b
iodomethane to benzoic acid of 1.0 and 1.1, respectively. HPLC
grade DMF used as received. Average and standard deviation from
three independent studies. Average and standard deviation from three
independent studies. DMF dried overnight on molecular sieves, with
c
d
e
a small amount of molecular sieves added in each syringe.
813
dx.doi.org/10.1021/op300085w | Org. Process Res. Dev. 2012, 16, 811−818

Organic Process Research & Development
Article
Figure 3. Arrhenius plot for the alkylation of benzoic acid deprotonated by TMGN in acetonitrile or DMF by iodomethane or iodoethane.
Table 3. Effect of alkylating agents and solvents on activation parameters of benzoic acid alkylation in the presence of TMGN
a
b
b
entry
solvent
DMF
alkylating agent
ΔH°⧧ (kJ·mol−1
)
ΔS°⧧ (J·mol−1·K−1
)
ΔG°⧧ (kJ·mol−1
)
Ea (kJ·mol−1
)
a
iodomethane
iodoethane
iodomethane
iodoethane
40.3
53.1
57.2
58.5
−112.2
−86.4
−72.7
−82.7
73.3
78.6
78.2
82.6
43.1
56.1
60.2
61.5
b
c
DMF
acetonitrile
acetonitrile
d
a
Conditions: benzoic acid 26.7 mM and molar ratios of TMGN and iodomethane to benzoic acid of 1.0 and 1.1, respectively, in HPLC grade
b
solvents used as received. Standard conditions: T = 20 °C, reagent concentrations = 1 mol·L−1.
slurry appears rapidly in the toluene solution. Furthermore, the
estimated half-life of the reaction from batch study at dilution
level of this work is higher than 10 h which precludes observing
it in a microfluidic device. In summary, the continuous flow
microfluidic reactor allowed us to screen rapidly and
quantitatively the reactivity of different substrates using less
than 0.5 mmol and to investigate various solvents.
Temperature Effect. The effect of the temperature for the
alkylation of benzoic acid itself by iodomethane in DMF was
then investigated in the range of 4−70 °C (SI, Table S4).
Kinetics for each temperature, even the highest (SI Figures S3−
S7), show no deviation for short residence time, which proves
that the thermal equilibrium is quickly achieved in the tubular
reactor. Otherwise, the kinetics would be slower at short
residence time for temperature above ambient temperature and
faster at temperature below ambient temperature.
The rate constants were used to construct an Arrhenius plot
(Figure 3) which is linear (R2 = 0.993) in the studied
temperature range (4−70 °C) and gives a value of 43.1
kJ·mol−1 for the activation energy. From these data the
enthalpy and the entropy of activation are estimated to be ΔH⧧
= 40.3 kJ·mol−1 and ΔS⧧ = −112.2 J·K−1·mol−1 leading to a
standard free energy of activation ΔG⧧ = 73.3 kJ·mol−1 at 293
K in standard conditions (c0 = 1 mol·L−1) (Table 3, entry a).
Kondo et al. found ΔH⧧ = 61.9 kJ·mol−1 and ΔS⧧ = −66.0
J·K−1·mol−1 for the alkylation of tetramethyl ammonium
benzoate salt by iodoethane in acetonitrile.8c
In order to be able to compare our data with Kondo’s results
we conducted three experiments by changing the solvent-
alkylating agent pairs: (i) DMF, iodoethane; (ii) acetonitrile,
iodomethane, and (iii) acetonitrile, iodoethane (Table 3,
entries b, c, d, respectively). For the alkylation of benzoic
acid by iodoethane in acetonitrile our results ΔH⧧⧧ = 58.5
kJ·mol−1 and ΔS⧧ = −82.7 J·K−1·mol−1 (Table 3, entry d) are in
agreement with Kondo’s data quoted above. The slightly lower
enthalpy of activation (3.5 kJ·mol−1) may be due to a looser ion
pair as the charge is more hidden in protonated TMGN. As
expected, the enthalpy of activation is always smaller for the
same reaction in DMF than in acetonitrile (Table 3 entries a
and c for alkylation by MeI, and b and d for alkylation by EtI).
The behavior of iodomethane in DMF is singular with a much
lower enthalpy of activation ΔH⧧ = 40.3 kJ·mol−1 and a more
negative entropy of activation ΔS⧧= −112.2 J·K−1·mol−1 (Table
3, entry a). These results show that continuous flow
microfluidic devices enable us to determine activation
parameters based on a very broad range of rate constants
covering 6 orders of magnitude using a very small amount of
substrates as only less than 0.5 mmol of benzoic acid were used
here per temperature.
Substituent Effect. We then used our setup to determine
the reaction rate of substituted benzoic acids over a wide range
of Hammett σ constant values (−0.66 to 0.78). A summary of
the results is presented in Table 4. Most of the benzoic acids
used were para-subsituted, but some more functionalized
benzoic acids were also studied including meta,para- and
ortho,meta-disubstituted benzoic acids.14e,20Hammett σ values
for para-substituted benzoic acid were obtained from the
classical Jaffe’s or Hansch’s tables.20a,b Hammett σ values for
ortho-substituents were obtained by multiplying those of para
values with 0.65.20c For 2,3- and 2,4-dimethoxy benzoic acids, σ
constants were obtained by summing the corresponding
ortho,meta and ortho,para values.20d Values of pKa in DMF
for dimethoxy substituents were obtained from pKa values in
DMSO by Exner et al.20e using the equation proposed by
814
dx.doi.org/10.1021/op300085w | Org. Process Res. Dev. 2012, 16, 811−818

Organic Process Research & Development
Article
acids alkylation and the values of σ for most para-substituted
benzoic acids. The obtained ρ Hammet constant reaction value
for the data marked with the ○ symbol is −0.65 (n = 8, R2 =
0.995). A ρ value of −0.92 for alkylation of substituted
tetramethylammonium benzoate with iodomethane in acetoni-
trile has already been reported by Kondo et al. (n = 4, R2 =
0.9895).8d In DMF as in acetonitrile, the Hammett alkylation
Table 4. Reaction rate constants for the alkylation of
substituted benzoic acids by iodomethane in the presence of
TMGN and in DMF at 20 °C
k
entry
substituent
σ
pKa in DMF
k (mol−1·L·s−1
)
a
a
a
f
a
b
c
d
e
f
4-NO2
0.778
0.628
0.276
10.6
0.175 0.004
0.249 0.002
0.72 0.04
g
4-CN
11.02
h
4-I
11.65
DMF
ACN
constant ρ is much smaller (ρ
= −0.65, ρ
=
=
Alkylation
Alkylation
DMF
b
j
3,5-(OCH3)2
4-Br
0.24
11.84
0.53 0.05
−0.92) than the Hammett ionization constant ρ (ρ
Ionization
a
a
a
f
l
0.232
0.227
0.062
0.00
11.6
0.65 0.10
ACN
Ionization
−2.29; ρ
= −2.49) showing that compensation of
f
l
4-Cl
11.5
0.46 0.07
solvation effects is taking place during the alkylation. Indeed,
during ionization the system is going from neutral to charged
benzoate, whereas during alkylation the system is going from
localized charge on benzoate anion to delocalized charges in the
transition step which leads to a smaller difference in the
transition states. The Hammett alkylation constant ρ is smaller
in DMF than in acetonitrile as expected due to the higher
reactivity in DMF than in acetonitrile (Table 2).
The quality of the kinetics data pushed us to find a
correlation including more substituted benzoic acids. Our first
trial was based on using Hammett substituent constant σ−
values20b but we found that the nonlinearity was much more
pronounced (SI Table S5 and Figure S1). Several correlations
based on quantum-calculated descriptors were also investigated
and proved to be unsuccessful. Hollingsworth et al.21a
i
g
h
i
4-F
11.84
0.56 0.01
f
l
4-H
12.3
0.57 0.02
c
,
d
j
2,3-(OCH3)2
3,4-(OCH3)2
4-CH3
−0.054
12.01
1.95 0.35
0.75 0.03
0.47 0.05
0.45 0.04
0.90 0.06
0.32 0.04
1.18 0.10
0.31 0.02
0.31 0.02
b
j
j
−0.15
12.50
a
f
k
l
−0.170
12.6
b
f
3,4-(CH3)2
4-OCH3
4-OH
−0.24
13.0
a
a
i
m
n
o
p
q
−0.268
−0.357
−0.442
−0.660
12.78
i
13.25
c
,
d
j
2,4-(OCH3)2
4-NH2
12.58
a
e
f
13.96
+
(4-NH3 )
0.600
a
b
c
From references 20a and b. From reference 20h. Calculated using
d
σortho = 0.65 × σpara based on reference 20c. Sum of the
corresponding ortho-, meta- and para constants based on reference
e
f
g
20d. From reference 20b. From reference 14e. From reference 20i.
h
i
j
From reference 20g. From reference 20j. Calculated using pKa
demonstrated that calculated Lowdin charges are effective
̈
values in DMSO from reference 20e and pKa(DMSO) to pKa(DMF)
parameters for the description of benzoic acids pKa values in
water. Thus, we tried to use them (SI Table S6) to correlate
reaction rate constants using the six substituents included in
our set, but they did not yield better linear correlations.
Molecular electrostatic potential minimum Vmin is another
descriptor used to quantify substituent effects in benzene21b
and benzoic acids.21c Again, we observed that some benzoic
acids fail to give a well-fitting regression line (SI, Table S7 and
Figure S2).
k
correlation from reference 14e. Average and standard deviation from
l
2 or better independent measurements. From 3 independent
measurements.
Maran et al.14e The reported ρ values for the Hammett plot of
pKa of benzoic acid in DMF vs σ constants are −2.36 (n = 8)
and −2.49 (n = 13) from Kolthoff et al. and Bartnicka et al.
respectively.20f,g In this work we obtained a ρ value of −2.29 for
16 substituents from a compilation of literature data.
The Hammett plot of log(kY/k0) versus σ constant, displayed
in Figure 4, shows that there is a good correlation between the
logarithm of relative rate constants of the substituted benzoic
If we look back to the data, Figure 4 shows that outliers can
be divided into three groups: bulky halogens (4-I and 4-Br),
alkyl substituents (4-Me and 3,4-Me2), and benzoic acids
bearing two reactive sites (4-NH2 and 4-OH). Deviation of p-
Figure 4. Log10 of the rate constant for the alkylation of benzoic acids by iodomethane versus Hammett substituent constant, σ, in the presence of
○
TMGN and in DMF at 20 °C. Data represented with symbols are taken into account in the Hammett linear free energy correlation, and Δ
symbols refer to the outliers.
815
dx.doi.org/10.1021/op300085w | Org. Process Res. Dev. 2012, 16, 811−818

Organic Process Research & Development
Article
Figure 5. Difference between observed log(k/k0) and predicted log(k/k0) for Hammett correlation outliers versus the Kamlet−Taft π* parameter
substituent incremental values for aromatic compounds.
amino- and hydroxybenzoic acids from normal Hammett
behavior has already been reported in the literature. p-
Aminobenzoic acid may exist as a zwitterion in the solution
the solvent ability to accept a proton (donate an electron pair)
in a solute-to-solvent hydrogen bond.
This equation does not depend on π* solvent dipolarity/
polarizability parameter which measures the ability of the
solvent to stabilize a charge or a dipole by virtue of its dielectric
effect. The halogen-substituted benzene series have a higher π*
than unsubsituted benzene (π* = 0.59), arranged in ascending
order of size from fluorine to iodine (π*F = 0.62, π*Cl = 0.71,
π*Br = 0.79, π*I = 0.81), but the π* values of alkyl substituents
are lower than that of unsubsituted benzene (π*Me = 0.55),
whereas their α and β parameters are nearly constant.24a A less
complete list of values is available for para-substituted benzoic
acid but follows the same trends: π*H = 0.74, π*Cl = 0.74, π*Br
= 0.79, and π*Me = 0.70.24a This variation clearly is in line with
our observation that bulky halogens (4-I, 4-Br) are more
reactive, and that alkyl substituents (4-Me, 3,4-Me) are less
reactive than expected. Figure 5 shows the good correlation
obtained for the difference between the observed log(k/k0) and
the value log(k/k0) predicted by the Hammett equation using
the ρ = 0.65 value we found based on the π* incremental values
calculated according to Hickey et al. tabulated values.24b It must
be pointed out that between σ = −0.3 to σ = 0.3, where the
outliers are located, the π* additive values are correlated to σ,
whereas Vi/100, α and β increment are independent (see SI,
Figures 8−11). Unfortunately, the Kamlet−Taft parameter
substituent incremental values for aromatic compounds are not
precise enough; for example, they do not include substituent
position, which precludes a quantitative treatment. Such a
stabilization of the ground state versus the transition state has
been proposed for interpreting the curved Hammett relation-
ship during the reaction of various nucleophiles with
substituted aryl benzoates.25 Clearly the influence of π* solvent
dipolarity/polarizability parameter on the reactivity of benzoic
acids reactivity in DMF deserves further studies.
leading to protonation of the amino site.22a When we use σNH
+
3
instead of σNH , the point displaces much closer to the fitted
2
straight line (Figure 4). McMahon and Kebarle have shown
that in the gas phase the lowest-energy anion derived from p-
hydroxybenzoic acid appears to be p-carboxyphenoxide ion
rather than p-hydroxybenzoate, because phenoxide ion receives
resonance stabilization while no equivalent stabilization by the
OH group is available to the p-hydroxybenzoate anion.22b No
methoxy benzoic acid or its methyl ester was detected by GC/
MS, but the existence of this p-carboxyphenoxide ion in
conjugated form is in agreement with the reactivity which is
lower than expected from Hammett’s correlation of p-
hydroxybenzoic acid.
A brief survey of the data shows that bulky halogens (4-I, 4-
Br) are more reactive and alkyl substituents (4-Me, 3,4-Me2)
are less reactive than expected. Several authors, such as Herbst
and Jacox,23a Kochai and Hammond,23b Kloosterziel and
Backer,23c Miron and Hercules,23d have already described the
abnormal behavior for p-methyl-substituted benzene in differ-
ent reactions that they attribute to the strong sensitivity of the
hyperconjugation effect to the solvent. Nagarajan et al. studied
the rate of deprotonation of the 2-methyl group in 1,2,3-
trimethylpyrazinium ion by benzoates in D2O.23e They
observed a deviation from Bronsted’s equation for o-
halobenzoate which is increasing with the group size, i.e. I >
Br > Cl > F. This led us to suspect that these variations may be
due to the solvation effects. Bartnicka et al.20g showed that the
Hammett acidity reaction constant for benzoic acid ionization
is well correlated by Kamlet and Taft solvatochromic
parameters for the solvent,24a with a good confidence for 10
very different solvents:
CONCLUSION
Solvent
= −(0.898 0.198)αSolvent + (0.916 0.427)
βSolvent + (1.790 0.241)
■
ρ
Ionization
A microfluidic setup based on commercially available devices
was developed to study the kinetics of reactions in continuous
flow mode, and the results were compared to the results
obtained in batch mode. We chose as the model reaction,
benzoic acid alkylation by iodomethane in DMF using an
organic superbase TMGN for deprotonation because this
where αSolvent and βSolvent are the Kamlet−Taft parameters
which describe respectively the ability of a solvent to donate a
proton in a solvent-to-solute hydrogen bond and a measure of
816
dx.doi.org/10.1021/op300085w | Org. Process Res. Dev. 2012, 16, 811−818

Organic Process Research & Development
Article
reaction is synthetically useful and is a good candidate for
physical chemistry correlations. As expected, this reaction
follows a very clean second-order kinetics which is preserved up
to complete conversion in the continuous flow microreactor.
The use of an organic superbase allowed us to try the three
possible reagent combinations: (i) benzoic acid + MeI, TMGN;
(ii) benzoic acid + TMGN, MeI; but also (iii) TMGN + MeI,
benzoic acid, as the proton sponge base is only slowly alkylated
in our conditions. The first two combinations exhibited nearly
the same kinetics as the batch values, whereas the last one was
shown to be very sensitive to residual water. The rate constants
for the reaction between different alkylating agents and benzoic
acid and in different solvents were also determined. This setup
enabled us to study the effect of temperature on this reaction
during which the variations of rate constant cover 6 orders of
magnitude. The plot of the conversion versus flow rate was
found to be linear at all temperatures, which proves that the
thermal equilibrium is rapidly established. From this data,
energy, enthalpy, and entropy of activation of benzoic acid
alkylation by MeI in DMF are estimated to be 43.1 kJ·mol−1,
40.2 kJ·mol−1, and −112.2 J·K−1·mol−1, respectively. The
activation parameters obtained in acetonitrile are in agreement
with previously published values. Finally, the alkylation kinetics
of a series of para-substituted benzoic acids was studied. Their
reactivities are well correlated with Hammett reaction constant
of −0.65. The quality of the data allowed us to ascertain the
origin of the deviations which were explained using the Kamlet
and Taft solvatochromic parameters. It must be pointed out
that reaction rates were measured consuming less than 0.5
mmol of substrate per condition. The very good correlations
obtained for Arrhenius plot and Hammett free energy
relationships demonstrate that capillary continuous flow
microreactors, well-known for their synthetic application, also
provide sound physical chemistry data. These data are now
used to develop the selective alkylation of multifunctional
natural products in our laboratory.
standards, 1,3,5-trimethoxybenzene (TMB) and 1,4-dimethox-
ybenzene (DMB), were dissolved in reagents inlet streams. 4-
iodoanisole is another internal standard that was added to the
samples to control the analysis, just before being analyzed.
Batch kinetics studies were conducted in a 10 mL vial
containing the required volume solution of acid and base in
DMF. To begin the reaction, a stoichiometric volume of
alkylating reagent was injected as rapidly as possible. During the
experiments samples of 240 μL volume, were withdrawn at
different times and were quenched. Quenching medium,
dilution and derivatization, as well as the analysis method,
were exactly those that were applied in continuous method.
All samples were analyzed on an ion trap mass spectrometer
using electron ionization (EI, 70 eV) fitted with a gas
chromatograph equipped with a split/splitless injector and an
autosampler. Separations were accomplished using a 60 m ×
0.25 mm column coated with a 5% diphenyl/95% dimethyl
polysiloxane film of 0.50 μm thickness. Liquid injections of 1
μL were introduced into the injector heated at 250 °C with a
50:1 split ratio and a mobile phase (helium) flow rate of 1 mL/
min. All analyses were carried out using a linear temperature
program from 50 °C to 250 °C at 10 °C/min followed by a
plate at 250 °C for 10 min. The mass spectrometer was scanned
from 40 to 400 (m/z).
ASSOCIATED CONTENT
* Supporting Information
Flow rates used in kinetics studies and their corresponding
■
S
residence times, results of a typical GC/MS quantification
experiment and statistical analysis, tables of calculated Lowdin
̈
charges and relative Vmin values for different substituted benzoic
acids, benzoic acid alkylation by iodomethane rate constant
logarithm versus σ− and Vmin, second-order kinetics plot of
benzoic acid alkylation by iodomethane at different temper-
atures, and plot of Kamlet−Taft Vi, α, β, and π* parameter
substituent incremental values versus Hammett σ constant.
This material is available free of charge via the Internet at
http://pubs.acs.org.
EXPERIMENTAL SECTION
■
Materials. All chemicals were in the highest purity available
and were used as received without further purification. For
some experiments (specified in the text), DMF was dried over
3 Å pore freshly activated molecular sieves.
AUTHOR INFORMATION
Corresponding Author
*Christian.Rolando@univ-lille1.fr
■
Methods. The high pressure syringe pump (pumping force
up to 1926 N) was fitted with two 8 mL stainless steel syringes
which are driven simultaneously. The two syringes, containing
the reagents, are connected via 0.30 m capillaries with internal
diameter smaller (i.d. 50 μm) than the one of the capillary
reactor to the micromixer. As micromixer, we utilized a
commercially available multilaminating distributive micromixer
chip. Capillary reactor internal diameter and length were 75 μm
and 300 cm respectively. In order to provide heating (up to 70
°C) or cooling (down to 4 °C) the capillary tubular reactor was
immersed in a water bath equipped with a thermostat or an ice/
water bath.
For different flow rates, 240 μL sample of reaction medium
was collected at the outlet of the tubular reactor and quenched
in a mixture of 400 μL of dichloromethane and 50 μL of formic
acid. For each flow rate, two samples were taken directly and
analyzed independently. A volume of 30 μL of the taken sample
was diluted by 400 μL of iodoanisole in dichloromethane
solution. This sample, was later derivatized by 50 μL of BSTFA
and 20 μL of pyridine and was kept overnight for GC/MS
analysis. To control the system performance, two internal
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
The Mass Spectrometry facilities used in this study were funded
by the European Community (FEDER), Reg
■
́
ion Nord-Pas de
de Lille 1, Sciences
Calais (France), CNRS, and the Universite
́
et Technologies. A.G. acknowledges the scholarship by The
French Ministry of Foreign affairs. The authors thank Anne-
Sophie Lacoste and Geoffrey Vauvy for their technical
assistance and Dr. Maria van Agthoven for her help with
editing this manuscript.
REFERENCES
■
(1) (a) Yoshida, J.-i.; Kim, H.; Nagaki, A. ChemSusChem 2011, 4,
331. (b) Mak, X. Y.; Laurino, P.; Seeberger, P. H. Beilstein J. Org.
Chem. 2009, 5. (c) Tanaka, K.; Fukase, K. Org. Process Res. Dev. 2009,
13, 983. (d) Geyer, K.; Gustafsson, T.; Seeberger, P. H. Synlett 2009,
2382. (e) Wiles, C.; Watts, P. Chem. Commun. 2011, 47, 6512.
(f) Watts, P.; Haswell, S. J. Chem. Soc. Rev. 2005, 34, 235. (g) Haswell,
S. J.; Middleton, R. J.; O’Sullivan, B.; Skelton, V.; Watts, P.; Styring, P.
817
dx.doi.org/10.1021/op300085w | Org. Process Res. Dev. 2012, 16, 811−818

Organic Process Research & Development
Article
Chem. Commun. 2001, 391. (h) Jas, G.; Kirschning, A. Chem.Eur. J.
2003, 9, 5708.
W. R. S.; Winterman, D. R. Chem. Commun. 1968, 723. (c) Alder, R.
W.; Goode, N. C. J. Chem. Soc., Chem. Commun. 1976, 108.
(14) (a) Przybylski, P.; Gierczyk, B.; Schroeder, G.; Zundel, G.;
Brzezinski, B.; Bartl, F. J. Mol. Struct. 2007, 844−845, 157. (b) Novak,
I.; Wei, X.; Chin, W. S. J. Phys. Chem. A 2001, 105, 1783. (c) Ishikawa,
T.; Ed. Superbases for Organic Synthesis: Guanidines, Amidines,
Phosphazenes and Related Organocatalysts; 2009; p 326. (d) Daasbjerg,
K. Acta Chem. Scand. 1995, 49, 878. (e) Maran, F.; Celadon, D.;
Severin, M. G.; Vianello, E. J. Am. Chem. Soc. 1991, 113, 9320.
(15) (a) Hartman, R. L.; Jensen, K. F. Lab Chip 2009, 9, 2495.
(b) Jaehnisch, K.; Hessel, V.; Loewe, H.; Baerns, M. Angew. Chem., Int.
Ed. 2004, 43, 406. (c) McMullen, J. P.; Jensen, K. F. Annu. Rev. Anal.
Chem. 2010, 3, 19.
(2) (a) Struempel, M.; Ondruschka, B.; Daute, R.; Stark, A. Green
Chem. 2008, 10, 41. (b) Irfan, M.; Glasnov, T. N.; Kappe, C. O. Org.
Lett. 2011, 13, 984. (c) Bartrum, H. E.; Blakemore, D. C.; Moody, C.
J.; Hayes, C. J. Chem.Eur. J. 2011, 17, 9586. (d) Tomida, Y.; Nagaki,
A.; Yoshida, J.-i. J. Am. Chem. Soc. 2011, 133, 3744. (e) Razzaq, T.;
Glasnov, T. N.; Kappe, C. O. Eur. J. Org. Chem. 2009, 1321.
(3) (a) Ahmed-Omer, B.; Brandt, J. C.; Wirth, T. Org. Biomol. Chem.
2007, 5, 733. (b) Baumann, M.; Baxendale, I. R.; Brasholz, M.;
Hayward, J. J.; Ley, S. V.; Nikbin, N. Synlett 2011, 2011, 1375. (c) Li,
P.; Buchwald, S. L. Angew. Chem. 2011, 123, 6520. (d) Riva, E.;
Rencurosi, A.; Gagliardi, S.; Passarella, D.; Martinelli, M. Chem.Eur.
J. 2011, 17, 6221. (e) Hessel, V.; Kralisch, D.; Krtschil, U. Energy
Environ. Sci. 2008, 1, 467. (f) Mason, B. P.; Price, K. E.; Steinbacher, J.
L.; Bogdan, A. R.; McQuade, D. T. Chem. Rev. 2007, 107, 2300.
(g) Wirth, T. ChemSusChem 2012, 5, 215 and all the articles in this
special issue for Flow Chemistry.
(16) Lin, W.-Y.; Wang, Y.; Wang, S.; Tseng, H.-R. Nano Today 2009,
4, 470.
(17) (a) Bessoth, F. G.; deMello, A. J.; Manz, A. Anal. Commun.
1999, 36, 213. (b) Kakuta, M.; Jayawickrama, D. A.; Wolters, A. M.;
Manz, A.; Sweedler, J. V. Anal. Chem. 2003, 75, 956. (c) Honda, T.;
Miyazaki, M.; Nakamura, H.; Maeda, H. Lab Chip 2005, 5, 812.
(d) Nguyen, N.-T.; Wu, Z. J. Micromech. Microeng. 2005, 15, R1.
(18) Kockmann, N. Transport Phenomena in Micro Process Engineer-
ing; Springer: New York, 2008.
(4) (a) Calabrese, G. S; Pissavini, S. AIChE J. 2011, 57, 828.
(b) Hessel, V. Chem. Eng. Technol. 2009, 32, 1655.
(5) (a) Valera, F. E.; Quaranta, M.; Moran, A.; Blacker, J.; Armstrong,
A.; Cabral, J. T.; Blackmond, D. G. Angew. Chem., Int. Ed. 2010, 49,
2478. (b) Bałdyga, J.; Bourne, J. R. Turbulent Mixing and Chemical
Reactions; Wiley: New York., 1999; p 867. (c) Baldyga, J.; Bourne, J.
R.; Walker, B. Can. J. Chem. Eng. 1998, 76, 641. (d) Dolman, S. J.;
Nyrop, J. L.; Kuethe, J. T. J. Org. Chem. 2011, 76, 993. (e) Koelbl, A.;
Kraut, M.; Schubert, K. AIChE J. 2008, 54, 639.
(19) Roberts, D. W.; Schultz, T. W.; Wolf, E. M.; Aptula, A. O. Chem.
Res. Toxicol. 2010, 23, 228.
(20) (a) Jaffe, H. H. Chem. Rev. 1953, 53, 191. (b) Hansch, C.; Leo,
A.; Taft, R. W. Chem. Rev. 1991, 91, 165. (c) Beteringhe, A. Cent. Eur.
J. Chem. 2005, 3, 585. (d) Verma, M.; Chaudhry, A. F.; Fahrni, C. J.
Org. Biomol. Chem. 2009, 7, 1536. (e) Exner, O.; Fiedler, P.;
Budesinsky, M.; Kulhanek, J. J. Org. Chem. 1999, 64, 3513. (f) Kolthof
f, I. M.; Chantooni, M. K., Jr. J. Am. Chem. Soc. 1971, 93, 3843.
(g) Bartnicka, H.; Bojanowska, I.; Kalinowski, M. K. Aust. J. Chem.
1992, 46, 31. (h) Seaton, C. C.; Chadwick, K.; Sadiq, G.; Guo, K.;
Davey, R. J. Cryst. Growth Des. 2009, 10, 726. (i) Jover, J.; Bosque, R.;
Sales, J. QSAR Comb. Sci. 2008, 27, 563. (j) Ludwig, M.; Baron, V.;
Kalfus, K.; Pytela, O.; Vecera, M. Collect. Czech. Chem. Commun. 1986,
51, 2135.
(21) (a) Hollingsworth, C. A.; Seybold, P. G.; Hadad, C. M. Int. J.
Quantum Chem. 2002, 90, 1396. (b) Suresh, C. H.; Gadre, S. R. J. Phys.
Chem. A 2007, 111, 710. (c) Sayyed, F. B.; Suresh, C. H. New J. Chem.
2009, 33, 2465.
(22) (a) Wiberg, K. B. J. Org. Chem. 2002, 67, 4787. (b) McMahon,
T. B.; Kebarle, P. J. Am. Chem. Soc. 1977, 99, 2222.
(23) (a) Herbst, R. L., Jr.; Jacox, M. E. J. Am. Chem. Soc. 1952, 74,
3004. (b) Hammond, G. S.; Reeder, C. E.; Fang, F. T.; Kochi, J. K. J.
Am. Chem. Soc. 1958, 80, 568. (c) Kloosterziel, H.; Backer, H. J. J. Am.
Chem. Soc. 1952, 74, 5806. (d) Miron, R. R.; Hercules, D. M. Anal.
Chem. 1961, 33, 1770. (e) Nagarajan, K.; Lee, T. W. S.; Perkins, R. R.;
Stewart, R. Can. J. Chem. 1986, 64, 1090.
(24) (a) Kamlet, M. J.; Doherty, R. M.; Abraham, M. H.; Marcus, Y.;
Taft, R. W. J. Phys. Chem. 1988, 92, 5244. (b) Hickey, J. P.; Passino-
Reader, D. R. Environ. Sci. Technol. 1991, 25, 1753.
(6) (a) Daridon, A.; Sequeira, M.; Pennarun-Thomas, G.; Dirac, H.;
Krog, J. P.; Gravesen, P.; Lichtenberg, J.; Diamond, D.; Verpoorte, E.;
de Rooij, N. F. Sens. Actuators, B 2001, B76, 235. (b) Mozharov, S.;
Nordon, A.; Littlejohn, D.; Wiles, C.; Watts, P.; Dallin, P.; Girkin, J. M.
J. Am. Chem. Soc. 2011, 133, 3601. (c) Baldwin, R. P.; Roussel, T. J.,
Jr.; Crain, M. M.; Bathlagunda, V.; Jackson, D. J.; Gullapalli, J.;
Conklin, J. A.; Pai, R.; Naber, J. F.; Walsh, K. M.; Keynton, R. S. Anal.
Chem. 2002, 74, 3690. (d) Salmon, J.-B.; Dubrocq, C.; Tabeling, P.;
Charier, S.; Alcor, D.; Jullien, L.; Ferrage, F. Anal. Chem. 2005, 77,
3417. (e) Kerby, M. B.; Legge, R. S.; Tripathi, A. Anal. Chem. 2006, 78,
8273. (f) Keybl, J.; Jensen, K. F. Ind. Eng. Chem. Res. 2011, 50, 11013.
(7) (a) Gisin, B. F. Helv. Chim. Acta 1973, 56, 1476. (b) Wang, S.-S.;
Gisin, B. F.; Winter, D. P.; Makofske, R.; Kulesha, I. D.; Tzougraki, C.;
Meienhofer, J. J. Org. Chem. 1977, 42, 1286. (c) Shaw, J. E.; Kunerth,
D. C. J. Org. Chem. 1974, 39, 1968. (d) Kruizinga, W. H.; Kellogg, R.
M. J. Am. Chem. Soc. 1981, 103, 5183. (e) Hencken, C. P.; Genna, D.
T.; Siegler, M. A.; Posner, G. H. J. Org. Chem. 2011, 76, 5149.
(f) Eriksson, J.; Antoni, G.; Langstroem, B. J. Labelled Compd.
Radiopharm. 2006, 49, 1105.
(8) (a) Kondo, Y.; Fujiwara, T.; Hayashi, A.; Kusabayashi, S.; Takagi,
T. J. Chem. Soc., Perkin Trans. 2 1990, 741. (b) Kondo, Y.; Sugitani,
W.; Tokui, M.; Takagi, T. J. Chem. Soc., Perkin Trans. 2 1995, 1049.
(c) Kondo, Y.; Tsukamoto, T.; Kimura, N. J. Chem. Soc., Perkin Trans.
2 1997, 1765. (d) Kondo, Y.; Urade, M.; Yamanishi, Y.; Chen, X. J.
Chem. Soc., Perkin Trans. 2 2002, 1449.
(9) (a) Ono, N.; Yamada, T.; Saito, T.; Tanaka, K.; Kaji, A. Bull.
Chem. Soc. Jpn. 1978, 51, 2401. (b) Mal, D.; Jana, A.; Ray, S.;
Bhattacharya, S.; Patra, A.; De, S. R. Synth. Commun. 2008, 38, 3937.
(10) (a) Barton, D. H. R.; Elliott, J. D.; Gero, S. D. J. Chem. Soc.,
Perkin Trans. 1 1982, 2085. (b) Barton, D. H. R.; Elliott, J. D.; Gero, S.
D. J. Chem. Soc., Chem. Commun. 1981, 1136. (c) Ishikawa, T.;
Kumamoto, T. Synthesis 2006, 737. (d) Kocienski, P. J.; Brown, R. C.
D.; Pommier, A.; Procter, M.; Schmidt, B. J. Chem. Soc., Perkin Trans. 1
1998, 9.
(25) Um, I.-H.; Han, H.-J.; Ahn, J.-A.; Kang, S.; Buncel, E. J. Org.
Chem. 2002, 67, 8475.
(11) (a) Bouktaib, M.; Atmani, A.; Rolando, C. Tetrahedron Lett.
2002, 43, 6263. (b) Bouktaib, M.; Lebrun, S.; Atmani, A.; Rolando, C.
Tetrahedron 2002, 58, 10001. (c) Kajjout, M.; Rolando, C. Tetrahedron
2011, 67, 4731.
(12) (a) Alexander, R.; Ko, E. C. F.; Parker, A. J.; Broxton, T. J. J. Am.
Chem. Soc. 1968, 90, 5049. (b) Parker, A. J. Chem. Rev. 1969, 69, 1.
(13) (a) Raab, V.; Kipke, J.; Gschwind, R. M.; Sundermeyer, J.
Chem.Eur. J. 2002, 8, 1682. (b) Alder, R. W.; Bowman, P. S.; Steele,
818
dx.doi.org/10.1021/op300085w | Org. Process Res. Dev. 2012, 16, 811−818