Chinese Journal of Chemical Engineering 23 (2015) 11231130

Contents lists available at ScienceDirect

Chinese Journal of Chemical Engineering

journal homepage: www.elsevier.com/locate/CJChE

Catalysis, Kinetics and Reaction Engineering

Evaluation of an improved epichlorohydrin synthesis from

dichloropropanol using a microchemical system
Yangcheng Lu , Rui Wang, Jisong Zhang, Qianru Jin, Guangsheng Luo
State Key Laboratory of Chemical Engineering, Department of Chemical Engineering, Tsinghua University, Beijing 100084, China

a r t i c l e

i n f o

Article history:
Received 1 August 2014
Received in revised form 28 September 2014
Accepted 26 October 2014
Available online 11 March 2015
Keywords:
Epichlorohydrin
Dichloropropanol
Microchemical system
Synthesis
Process decoupling

a b s t r a c t
Synthesizing epichlorohydrin (ECH) from dichloropropanol (DCP) is a complicated reaction due to the partial
decomposition of ECH under harsh conditions. A microchemical system can provide a feasible platform for
improving this process by conducting a separation once full conversion has been achieved. In this work, referring
to a common DCP feed used in industry, the reaction performance of mixed DCP isomers with NaOH in the
microchemical system on various time scales was investigated. The operating window for achieving high conversion and selectivity was on a time scale of seconds, while the side reactions normally occurred on a time scale of
minutes. Plenty of Cl ions together with a high temperature were proved to be critical factors for ECH hydrolysis.
A kinetic study of alkaline mediated ECH hydrolysis was performed and the requirements for an improved ECH
synthesis were proposed by combining quantitative analysis using a simplied reaction model with experimental results on the time scale of minutes. Compared with the conventional distillation process, this new strategy for
ECH synthesis exploited microchemical system and decoupled the reaction and separation with potentials of
higher productivity and better reliability in scaling up.
2015 The Chemical Industry and Engineering Society of China, and Chemical Industry Press. All rights reserved.

1. Introduction
Epichlorohydrin (ECH) is an important intermediate in the production of epoxide resins, chlorohydrin rubbers and other organic products
[13]. There have been several industrial manufacturing processes to
produce ECH such as propylene chlorination and glycerol chlorination
[47]. One of the most common routes for the synthesis of ECH is
through the elimination of hydrogen chloride from dichloropropanol
(DCP). Fig. 1 illustrates this process, in which the desired ring closure is
generally followed with hydrolysis as the major side reaction [8]. The
immediate removal of ECH is necessary for limiting further hydrolysis.
Dichloropropanol has two isomers: 1,3-dichloro-2-propanol (1,3DCP) and 2,3-dichloro-1-propanol (2,3-DCP). A 1:2 ratio is normally
obtained from the chlorohydrination of allyl chloride. Due to inductive
effects and steric hindrance, the reaction rate of the elimination of
hydrogen chloride from 1,3-DCP was reported to be 20 times faster
than that from 2,3-DCP [8,9]. Industrial ECH synthesis aims to convert
2,3-DCP sufciently while avoiding subsequent hydrolysis. In this regard, the reactive distillation column is conventionally used for ECH synthesis, in which ECH can be stripped with steam once being produced.
Carr et al. [8,9] determined the reaction kinetics for a complete description of ECH production process using lime milk as alkaline and provided
Supported by the National Natural Science Foundation of China (21036002, 21176136)
and the National Science and Technology Support Program of China (2011BAC06B01).
Corresponding author.
E-mail address: luyc@tsinghua.edu.cn (Y. Lu).

a model for simulating a multistage unit coupling reaction with distillation. Ma et al. [10] investigated the apparent kinetics of ring closure and
hydrolysis using sodium hydroxide solution as alkaline. They indicated
that raising the temperature favored the desired reaction pathway if
the residence time of ECH was sufciently short. All of the analyses or
simulations in these studies were based on the assumptions of ideal
feed mixing and ow as well as instantaneous ECH separation. However,
actual reaction output in production was frequently different from the
expected results for several reasons. First, the apparent kinetic parameters determined in a conventional reactor are usually case-dependent
(Ma et al. [9] even obtained two different kinetic equations according
to different temperature ranges). Second, in a cascade reactive distillation column, it is impossible to realize exact idealities in mixing, twophase counter-ow or instantaneous phase equilibrium [1113]. Thus,
it is still a challenge to control and enhance DCP conversion effectively.
In recent years, microreactors have shown numerous advantages
over traditional reactors, such as enhanced mixing and mass transport,
controlled hydrodynamic ow and good inherent safety [1416]. They
have been used as effective tools for intensifying fast reactions dependent on careful controls. Examples included rapid precipitation for
nanoparticle preparation [17,18], and fast and strongly exothermic reactions for chemical intermediates synthesis [1921]. Microreactors
have the potential to carry out an improved ECH synthesis.
In our previous work, the kinetic parameters of dehydrochlorination
of 1,2-DCP were determined in a microchemical system consisting of a
micromixer, a delay loop and a microneutralizer [22]. However, an
industrial case using microchemical system as the core unit may need

http://dx.doi.org/10.1016/j.cjche.2014.10.021
1004-9541/ 2015 The Chemical Industry and Engineering Society of China, and Chemical Industry Press. All rights reserved.

1124

Y. Lu et al. / Chinese Journal of Chemical Engineering 23 (2015) 11231130

Fig. 1. Synthesis of ECH. (a) Desired reaction; (b) a typical following side reaction.

more considerations and evaluations because the starting material is a

mixture of DCP isomers and the ECH product must eventually be separated from reaction system instead of inhibiting the hydrolysis temporarily. In this work, the reaction performance of mixed DCP (1,3-DCP/
2,3-DCP: 1/2) aqueous solution with sodium hydroxide solution in the
microchemical system on various time scales was investigated and the
kinetic characteristics of varieties of side reactions were explored to
propose proper reaction conditions and to schedule the integration of
reaction and separation for ECH synthesis enhancement. Compared
with the traditional reactive distillation process, this new strategy for
ECH synthesis exploited microchemical system and decoupled the reaction and separation, therefore, higher productivity and better reliability
in scaling up were worth expectation.
2. Experimental
2.1. Reagents
1,3-Dichloro-2-propanol (1,3-DCP, 99% in mass) and 2,3-dichloro-1propanol (2,3-DCP, 97% in mass) were purchased from J&K Scientic
Ltd. (Beijing). Sodium dihydrogen phosphate dihydrate (99.5% in
mass), disodium hydrogen phosphate anhydrous (99.0% in mass), sodium hydroxide (NaOH) and sodium chloride (NaCl) were purchased
from Beijing Chemical Works. Epichlorohydrin (ECH, 99.5% in mass)
was purchased from Sinopharm Chemical Reagent Beijing Co., Ltd. All
the reagents were of analytical grade and used directly without further
purication. Deionized water was used throughout the experiments.
2.2. Experimental setup
The microchemical setup, as shown in Fig. 2, can be found elsewhere
[22,23]. Briey, DCP solution (Feed 1) and NaOH solution (Feed 2) were
delivered by two metering pumps (LB-80, Beijing Satellite Co. Ltd.) into
the feeding tubes, respectively. The two solutions were preheated to the
reaction temperature and then mixed in Microreactor 1 to start the reaction. The preheating zone, Microreactor 1 and most of the delay loop were
placed in Thermostat 1 (SC-15, Ningbo Xinzhi) to guarantee a thermostatic reaction process. At the end of the delay loop, the process was
extended to a low-temperature thermostat (Thermostat 2, SC-15, Ningbo

Xinzhi), in which phosphate buffer solution (pH = 7.0) delivered by another metering pump was introduced into the reaction system through
Microreactor 2. Due to the sudden decrease of temperature and concentration of hydroxyl ions, all the reactions were assumed to be quenched
during the sample processing and analysis. In this microchemical system,
the nominal reaction time was dependent on the length of the delay loop.
2.3. Analysis
The sample collected from the outlet of Microreator 2 was diluted
with ethanol, and then 0.5 l of sample was injected into a gas chromatograph (Shimadzu GC-2014). The GC analysis used AB-InoWax
(30 m 0.25 mm 0.25 m) for the column and polyethylene glycol
for the stationary phase. The GC conditions included: carrier gas, N2;
injection temperature, 220 C; column temperature, 70 C to 200 C
(30 Cmin1); and detector, ame ionization detector (FID) at
280 C. The quantitative analysis by GC was based on the internal
standard method. 1-Octanol was used as the internal standard. The
dichloropropanol conversion (X) and selectivity (S) are calculated
with the following equations.
X 1

W DCP  Mpro
F DCP

W ECH  M pro
F DCP W DCP  M pro

where WDCP and WECH are the mass fraction of DCP and ECH in the sample, respectively. Mpro (g) is the sampling mass, and FDCP (g) the feeding
mass of DCP during sampling. The determination errors of WDCP and
WECH were about 0.5% and 2.4%, respectively.
3. Results and Discussion
3.1. Reaction performance on the time scale of seconds
A microchemical system can provide a resolution of residence time
in the sub-second range. For a reaction which nears completion within
seconds, this is quite a valuable characteristic for process control. Our
previous research showed that the dehydrochlorination of 2,3-DCP,

Feed 3

Feed 2

Feed 1

Delay loop

Microreactor 1

Microreactor 2

Thermostat 1

Thermostat 2

Sampling

Fig. 2. Schematic of the microchemical system for ECH synthesis.

Y. Lu et al. / Chinese Journal of Chemical Engineering 23 (2015) 11231130

3.1.1. Stoichiometric reaction

As depicted in Fig. 1, NaOH can cause undesired ECH hydrolysis.
Thus, a stoichiometric reaction seems to be a reasonable option for the
desired reaction. Fig. 3 shows the time proles of DCP conversion in a
series of stoichiometric reactions. In the caption of gures, [DCP]10 refers to the initial DCP concentration of Feed 1 and [NaOH]20 the initial
NaOH concentration of Feed 2. Solid points correspond to the experiments using 3% DCP (in mass) for Feed 1. As seen, the conversion always
skips from zero to a considerable value (over 0.3) within the rst measurement interval (12 s), demonstrating that the conversion of 1,3DCP is almost an instantaneous process. However, the discrepancy of
the conversion curves at various temperatures is still remarkable with
time elapsed. The increasing of conversion is always slow and smooth
within the rst 20 s at 50 C, while occurs rapidly within the rst 5 s
at 85 C. This discrepancy is attributed to the temperature-sensitive
kinetics of 2,3-DCP conversion. Our previous work [22] indicated that
the activation energy of dehydrochlorination of 2,3-DCP was as high
as 150 ( 10) kJmol1. Fig. 3(b) illustrates the dependency of the
time prole of conversion on the initial DCP concentration. Apparently,
the increasing of [DCP]10 can increase the DCP conversion ratio at every
time intervals, indicating that the kinetic order of [DCP] is larger than 1
for the DCP dehydrochlorination reaction. A sufcient DCP conversion
(N95%) in a microchemical system is accessible under proper conditions
including DCP concentration of 5.0% (in mass), reaction temperature of
85 C, and residence time of no more than 15 s.
Since the necessary residence time in the microchemical system is
quite short for a sufcient DCP conversion, it is attempted to predict
the DCP conversion kinetics without considering side reactions. Assuming that 1,3-DCP converts to ECH completely instantaneously, the
overall conversion of DCP can be calculated by following equations
d2; 3 DCP
dt
dt
1; 3DCP0 2; 3DCP0
t

1; 3DCP0 0

d2; 3DCP
25 156000
2
1:61  10 e RT 2; 3DCP NaOH
dt

where all the concentrations correspond to the mixture of Feed 1 and

Feed 2, and the activation energy for 2,3-DCP dehydrochloration

reaction is taken as 156 kJmol-1. The lines in Fig. 3 represent the calculated values. They match with the experimental data well, indicating the
assumptions are reasonable and the chemical process within the
microchemical system can be well predicted or designed.
Besides of the conversion of DCP, the selectivity of DCP to ECH is another concern. Fig. 4 gives selectivities at relatively high temperatures
(77 C and 85 C) within 15 s. As seen, all the dots are around 1.0, almost
independent of the temperature, DCP initial concentration and residence time. This indicates that an operating window exists for achieving
high conversion and high selectivity simultaneously in a microchemical
system.

1.0
0.8

selectivity

although much slower than that of 1,3-DCP, was a fast reaction. Therefore, the reaction performance of mixed DCP isomers in a microchemical
system on the time scale of seconds was rst investigated. Typical ow
rates were determined to be 10 mlmin1 of Feed 1 and 3 mlmin1 of
Feed 2 in preliminary experiments, over which the time prole of
conversion was almost consistent since the mixing performance in the
microchemical system was high enough.

1125

0.6
o

77 C
o
85 C

0.4
0.2
0.0

10

15

t/s
Fig. 4. Selectivities of ECH from DCP under various reaction conditions. Feed 1: [DCP]10 =
3.0% (by mass) (solid points) or 5.0% (by mass) (hollow points), 10 mlmin1. Feed 2:
[NaOH]20 = 3.16% (by mass) (solid points) or 5.2% (by mass) (hollow points),
3 mlmin1.

3.1.2. Effects of the molar ratio of reactants

An exact stoichiometric reaction is not always practical for industrial
use since errors in feeding and controlling systems are inevitable.
Moreover, when the conversion of some specic reactant like DCP is
preferred, other reactants are commonly excessively introduced. Therefore, the effects of the molar ratio of NaOH to DCP in terms of conversion
and selectivity at a high temperature (85 C), high concentration (5% in
mass) and short residence time (up to 15 s) are investigated. As illustrated in Fig. 5(a), when the mole ratio of NaOH to DCP is 0.9, the
conversion ratio reaches 0.9, the theoretical limit, and when the mole
ratio of NaOH to DCP is 1.2, the conversion ratio reaches 0.98, 2%
lower than the theoretical limit. Apparently, the DCP concentration
has a much more crucial inuence on the reaction kinetics than the
NaOH concentration. The time proles of conversion ratio at various

1.0

1.0

conversoin

conversion

0.8
0.6
0.4

50 C
o
60 C
o
70 C

0.2
0.0

10

t/s

(a)

15

20

0.9

0.8

0.7

77 C
o
85 C
0

10

15

20

t/s

(b)

Fig. 3. Time proles of DCP conversion under various reaction conditions. Feed 1: [DCP]10 = 3.0% (by mass) (solid points) or 5.0% (by mass) (hollow points), 10 mlmin1. Feed 2:
[NaOH]20 = 3.16% (by mass) (solid points) or 5.2% (by mass) (hollow points), 3 mlmin1.

1126

Y. Lu et al. / Chinese Journal of Chemical Engineering 23 (2015) 11231130

1.0

0.9

NaOH:DCP=0.9
NaOH:DCP=1.1
NaOH:DCP=1.2

0.8

0.7

selectivity

conversion

1.00

10

0.95

0.90

0.85

NaOH:DCP=0.9
NaOH:DCP=1.1
NaOH:DCP=1.2

0.80
0

15

10

t/s

t/s

(a)

(b)

15

Fig. 5. Effects of the molar ratio of NaOH to DCP on ECH synthesis at 85 C. (a) Conversion; (b) selectivity. Feed 1: [DCP]10 = 5.0% (by mass), 10 mlmin1; Feed 2: [NaOH]20 = 5.2%
(by mass); the owrate of Feed 2 is varied according to the molar ratio of NaOH to DCP; the lines represent the calculated values.

molar ratios of NaOH to DCP can also be predicted. The selectivity is

always near 1.0 when the mole ratio of NaOH to DCP is 1.0 or less
with a slight decreasing trend when the mole ratio of NaOH to DCP is
over 1.0. Therefore, excessive NaOH should be avoided for achieving
high selectivity in a microchemical system.
3.1.3. Effects of extra chloride ions
In industry, the ECH synthesis is usually joined directly with DCP
preparation. DCP can be prepared by allyl chloride chlorohydrination
or glycerol chlorination with hydrochloride (HCl) generation. Therefore,
in the ECH synthesis, NaOH also acts as the neutralization agent for HCl.
Considering the acidbase neutralization is generally accepted as an
instantaneous reaction, it is assumed that NaCl is pre-existed for a dehydrochlorination process of DCP. Therefore, some experiments were
conducted using NaOH-NaCl solution instead of NaOH solution as Feed
2 to investigate the effects of extra chloride ions on ECH synthesis. The
results are shown in Fig. 6. Compared with the reference experiments,
the inuence of chloride ions on the ECH synthesis is relatively insignificant in a microchemical system.
3.2. Side reactions on the time scale of minutes
Our aforementioned results conrmed that high conversion
and high selectivity could be achieved simultaneously within a sufciently short residence time (no more than 15 s in this work) in a
microchemical system. However, if ECH product continues to contact
with a hot aqueous solution containing plenty of electrolytes, side reactions may consistently take place. Therefore, the side reactions in the efuent of the microchemical system must still be considered before an
effective separation and chilling of ECH. Since the temperature of the efuent of the microchemical system is high, distillation appears to be an
alternative choice for separating and enriching ECH. Unlike the reactive

distillation process in conventional ECH synthesis, a simple distillation

process can be enhanced without the compromise of conversion. However, the effects of side reactions need to be investigated on the time
scale higher than that controlled in a microchemical system. Therefore,
the side reactions on the time scale of minutes are investigated, during
which most of conventional separation process can be completed.
3.2.1. Conversion ratio and selectivity
Fig. 7 shows comparisons of the conversion ratio and selectivity after
~10 s, 5 min, and 10 min residence. As seen, the DCP conversion ratio
has a variance with the time scale but it always corresponds to the lowest value after 10 min. The selectivity decreases substantially when the
reaction time extends to 5 min or more. Considering that 10 min may be
long enough for the exhaustion of NaOH by reactions shown Fig. 1, these
results indicate in the neutral aqueous solution containing Cl at 85 C
that the reversible conversion to DCP exists in the reaction system and
ECH decomposition cannot be neglected. Gaca et al. [24] have discussed
the mechanisms of ECH hydrolysis in acidic and neutral medium and in
the presence of chloride ions. Zong et al. [25] investigated the changing
of ECH concentration in the presence of NaCl under different conditions
and discussed the salt effect on the cyclization reaction. Both studies
pointed out that the presence of chloride ions may promote hydrolysis
side reaction, of which perhaps routes are proposed in Fig. 8. Plenty of
Cl ions with H+ generated from water dissociation can promote the
conversion of ECH to DCP (Step 1), meanwhile, DCP reacts with OH
to generate 3-chloro-1,2-propanediol (Step 2). The overall conversion
is a hydrolysis process from ECH to 3-chloro-1,2-propanediol and Cl
ions play a catalyst-like role. It is noticed that the above mentioned hydrolysis routes are not conceivable as alkaline (NaOH) still remains in
reaction system. Therefore, alkaline mediated side reaction should be
considered in priority, and characterizing its kinetic performance is
helpful for nding or understanding a way to guarantee the selectivity

1.0

1.0

selectivity

conversion

0.8
0.9

0.8

0.6
0.4
0.2

0.7

10

15

0.0

10

t/s

t/s

(a)

(b)

15

Fig. 6. Effects of the extra chloride ions on ECH synthesis from DCP at 85 C. (a) Conversion; (b) selectivity. Feed 1, [DCP]10 = 5.0% (by mass), 10 mlmin1; Feed 2, [NaOH]20 = 5.2%
(by mass), 3 mlmin1; hollow points in (a) are of experiments without NaCl pre-addition for reference.

Y. Lu et al. / Chinese Journal of Chemical Engineering 23 (2015) 11231130

~10 s

5 min

~10 s

10 min

5 min

10 min

1.0

selectivity

conversion

1.0

1127

0.9

0.8

0.8
0.6
0.4

0.7

1.0

1.1

0.2

1.2

1.0

1.1

1.2

mole ratio

mole ratio

(a)

(b)

Fig. 7. Changing of conversion (a) and selectivity (b) on the time scale of minutes. T = 85 C, [DCP]10 = 5.0% (by mass), [NaOH]20 = 5.2% (by mass); NaOH and NaCl are approximately
equimolar.

Fig. 9 shows the hydrolysis reaction conversion ratio (based on

NaOH) versus time. Considering the species in the sodium hydroxide
mediated reaction, the reaction rate can be expressed as

r kECH NaOH

1.0

0.8

XNaOH

Fig. 8. Perhaps hydrolysis routes of ECH in neutral aqueous solution containing Cl.

from DCP to ECH. To this end, a well-established method should be

employed to eliminate the effect from other hydrolysis reaction
pathways.

0.6

[ECH] 0 =5.83%,by mass

0.4

[ECH] 0 =5.00%,by mass

0.2

[ECH] =4.17%,by mass

0

[ECH] =3.33%,by mass

0

0.0
0

10

15

20

25

t/min

(a)
0.0

-0.5

ln(1-XNaOH )

3.2.2. Sodium hydroxide mediated hydrolysis reaction

Since sodium hydroxide mediated hydrolysis is not a fast reaction, a
setup as shown in Fig. 9 is used to investigate its reaction kinetics. The
proper stirring speed was explored to eliminate the effect of mass transfer in advance. In experiments, ECH and NaOH solutions were prepared
and preheated to a preset temperature. Next, the solutions were added
to a container with a magnetic stir-bar and heat jacket and the reaction
was timed accordingly. So as to inhibit the hydrolysis routes shown in
Fig. 8, the initial molar ratio of NaOH to ECH was kept at or less than
0.1 to decrease the Cl ion concentration. The sodium hydroxide mediated hydrolysis process was traced by determining the NaOH concentration with a pH electrode (with temperature calibration).

-1.0

-1.5

[ECH] =3.33%,by mass

0

[ECH] =4.17%,by mass

0

-2.0

[ECH]0=5.00%,by mass
[ECH]0=5.83%,by mass

-2.5
0

12

16

t/min

(b)
0.15

PH
6.98

C
B

k[ECH]0

0.12
0.09
0.06
0.03

0.00
0.0

0.1

0.2

0.3

0.4

[ECH]0 /molL

0.5

0.6

0.7

-1

(c)
Fig. 9. Set up for hydrolysis reaction kinetics determination. A heat carrier inlet; B heat
carrier outlet; C pH electrode; D magnetic stirred rod; E pH meter.

Fig. 10. Time prole of the sodium hydroxide mediated hydrolysis reaction at 50 C
[NaOH]0 = 1.2% (by mass); ECH is absolutely excessive.

1128

Y. Lu et al. / Chinese Journal of Chemical Engineering 23 (2015) 11231130

where [ECH] and [NaOH] represent the concentration of ECH and NaOH
after mixing, respectively. Since ECH is in extreme excess during the
reaction, [ECH] is taken as [ECH]0 approximately.
If the order of NaOH is rst, we derive

ln 1X NaOH  kECH0 t

where XNaOH represents the conversion of NaOH.

Fig. 10(b) shows the time proles of ln(1-XNaOH) at various [ECH]0.
The experimental results agree with Eq. (6) well, supporting the
common assumption that the sodium hydroxide mediated hydrolysis
reaction is rst order for NaOH. Fig. 10(c) is a plot of k[ECH]0 versus
[ECH]0, where an approximate linearity across the base point can be
found. Evidently, is equal to unity, indicating that the sodium hydroxide mediated hydrolysis reaction is also rst order for ECH. In addition,
the slope of the line in Fig. 10(c) is the value of k at 50 C.
The experiments of hydrolysis reaction were further carried out at
varied temperature. Results are shown in Fig. 11(a). Suppose that k is
dependent on temperature according to the Arrhenius equation,
Eq. (7) is obtained as
E
RT

k k0 e

where k is the rate constant, k0 is pre-exponential factor, E is the activation energy, R is the universal gas constant (8.314 Jmol1K1) and T
is the absolute temperature. k0 and E can be obtained by linearity calibration of lnk vs. 1/T, as shown in Fig. 11(c). Finally, the following kinetic
equation for the sodium hydroxide mediated hydrolysis reaction is
obtained
11 87305
RT

r 3:42  10 e

ECHNaOH:

3.3. Requirements for enhancing ECH synthesis

Since the desired reaction could be carried out in the microchemical
system with high efciency, the focus of ECH synthesis enhancement
was how to inhibit side reactions. Dependent on whether reactants of
the desired reaction were involved or not, the sodium hydroxide
mediated hydrolysis reaction and other side reactions were discussed
separately.
NaOH is the reactant of either the sodium hydroxide mediated
hydrolysis reaction or the desired reaction. Therefore, the sodium
hydroxide mediated hydrolysis reaction will be inhibited by the fast
desired reaction as long as the molar ratio of NaOH to DCP is no more
than 1:1. In such cases, assuming that only a small proportion of ECH
is hydrolyzed, the rate ratio of the desired reaction and the sodium hydroxide mediated hydrolysis reaction at any point of DCP conversion
ratio can be approximately predicted, as shown in Fig. 12. An overall
optimization strategy involves: (1) increasing the initial DCP concentration [DCP]0 [Fig. 12(a)] and temperature [Fig. 12(b)], and/or (2) using
excessive DCP [Fig. 12(c)]. Fig. 12(d) indicates that the optimization
on reaction conditions can strongly inhibit the effect of sodium hydroxide mediated hydrolysis on the desired reaction. Moreover, as long as
most of the NaOH is consumed in the microchemical system, it may
be able to ignore the effect of sodium hydroxide mediated hydrolysis.
This requirement can be met within a residence time of 1020 s under
optimized reaction conditions.
Unlike the sodium hydroxide mediated hydrolysis reaction, other
hydrolysis reaction pathways are not terminated even if NaOH is
exhausted. Thus, it is exactly necessary to separate ECH from the reaction system before considerable hydrolysis reactions occur as well. Synthetically viewing the time dependency of selectivity on various time
scales, nishing ECH separation within 1 min may achieve acceptable
selectivity (N 95%) at a temperature around 85 C. Fig. 13 gives an illustration on the whole process including a microchemical system and a
separation unit for ECH synthesis. However, more details on other

1.0

0.0

0.8

ln(1-XNaOH)

-0.5

XNaOH

0.6
0

40 C

0.4

50 C
0

0.2

16

24

32

50 C
o

60 C
o
70 C

-2.5

0.0
8

40 C

-1.5

-2.0

60 C
0
70 C
0

-1.0

40

t/min

t/min

(a)

(b)

10

-3
-4

lnk

-5
-6
-7
-8

0.0029

0.0030

0.0031

0.0032

1/T

(c)
Fig. 11. Effects of temperature on the sodium hydroxide mediated hydrolysis reaction. [ECH]0 = 6% (by mass), [NaOH]0 = 1.2% (by mass); ECH is absolutely excessive.

Y. Lu et al. / Chinese Journal of Chemical Engineering 23 (2015) 11231130

-1

1 0 00 0

(R e f 1 )

1129

10000

-1

85 C (Ref 1)
o
90 C
o
95 C

1 00 0

1000

10 0

100

r1/r2

r1 /r 2

-1

10

10

0 .1
0 .0

0.1

0 .2

0 .9 2

0 .96

0 .9 8

0.1
0.0

1.0 0

10000

Ref 1
Ref 2

1000

r1/r2

100

100

10

10

0.2

0.92 0.94 0.96 0.98 1.00

(b)

Ref1
1 % excess
2 % excess

0.1

0.2

(a)

1000

0.1
0.0

0.1

10000

r1/r2

0.9 4

0.1
0.0

0.92 0.94 0.96 0.98 1.00

0.1

0.2

0.92 0.94 0.96 0.98 1.00

(c)

(d)

Fig. 12. Prediction on the rate ratio of desired reaction to sodium hydroxide mediated hydrolysis reaction. Ref 1: T = 85 C, [DCP]0 = [NaOH]0 = 0.2 molL1; Ref 2: T = 95 C, [DCP]0 =
0.51 molL1, [NaOH]0 = 0.5 molL1; the curves in each subgure are calculated by using the same parameters with the exception of (a) initial concentrations, (b) temperature, or
(c) excessive amount of DCP.

DCP

NaOH

Microreactor

ECH-riched stream

Delay loop

Flash Flash
separation
separation

ECH-leaned stream
Time line
0s

15s

30s

45s

60s

Fig. 13. Schedule of ECH synthesis process using a microchemical system.

hydrolysis reactions of ECH need further exploration for a more accurate

process design. The process could also benet from technologies involving ash or highly efcient separation for the ECH synthesis system as
well.
4. Conclusions
In this work, the reaction between mixed DCP (1,3-DCP/2,3-DCP:
1/2) and NaOH in aqueous solution in a microreactor was investigated. It was observed that N95% conversion of DCP can be achieved
using 5.0% (by mass) of DCP at 85 C with a residence time of 15 s
in a microchemical system. On the time scale of seconds, extra chloride ions introduced by DCP feed did not show a signicant effect on
the reaction performance. The effects of side reactions were limited
and the time prole of DCP conversion ratio could be accurately
predicted.
By extending the time scale to minutes, the effects of side reactions
become obvious at high temperature like 85 C. Besides a sodium
hydroxide mediated pathway, plenty of Cl ions, independent of
sodium hydroxide, led to hydrolysis reactions with a high impact on

the main reaction. Through specically designed experiments, the

kinetics of the sodium hydroxide mediated hydrolysis of ECH was determined. Both calculated and experimental results indicated: (1) a practical direction to inhibit the effect of the sodium hydroxide mediated
hydrolysis of ECH was to increase the initial DCP concentration and
temperature as well as introduce a slight excess of DCP; and (2) as
long as most of NaOH was consumed in the microchemical system,
the side reactions caused by the sodium hydroxide mediated hydrolysis
were insignicant in following.
Systematically viewing the time dependency of selectivity on
various time scales, nishing ECH separation within 1 min could
achieve acceptable selectivity (N 95%) at a temperature around
85 C. Compared with the traditional reactive distillation process,
this new strategy for ECH synthesis exploited microchemical system
and decoupled the reaction and separation with potentials of higher
productivity and better reliability in scaling up. However, both the
hydrolysis reactions of ECH without alkaline and the technologies eligible for ash and high efcient separation of ECH synthesis system
need further exploration for more accurate process design and more
effective production.

1130

Y. Lu et al. / Chinese Journal of Chemical Engineering 23 (2015) 11231130

References
[1] J.W. Bijsterbosch, A. Das, F.P.J.M. Kerkhof, Clean technology in the production of
epichlorohydrin, J. Clean. Prod. 2 (1994) 181184.
[2] K. Weissermel, H.J. Arpe, Industrial Organic Chemistry, Wiley-VCH, Weinheim and
New York, 1997.
[3] M. Wang, Y.B. She, X.T. Zhou, H.B. Ji, Efcient solvent-free synthesis of
chloropropene carbonate from the coupling reaction of CO2 and epichlorohydrin
catalyzed by magnesium porphyrins as chlorophyll-like catalysts, Chin. J. Chem.
Eng. 19 (2011) 446451.
[4] N., Nagato, H., Mori, K., Maki, R., Ishioka, Process for production of epichlorohydrin,
US Patent 4,634,784 (1987).
[5] E. Santacesaria, R. Tesser, M. Di Serio, L. Casale, D. Verde, New process for producing
epichlorohydrin via glycerol chlorination, Ind. Eng. Chem. Res. 49 (2010) 964970.
[6] S.M. Danov, A.V. Sulimov, A.V. Ovcharova, Modeling of the vaporliquid equilibrium
in the system formed by the epichlorohydrin synthesis products, Theor. Found.
Chem. Eng. 46 (2012) 3143.
[7] L.L. Wang, Y.M. Liu, W. Xie, H.J. Zhang, H.H. Wu, Y.W. Jiang, M.Y. He, P. Wu, Highly
efcient and selective production of epichlorohydrin through epoxidation of allyl
chloride with hydrogen peroxide over Ti-MWW catalysts, J. Catal. 246 (2007)
205214.
[8] S. Carr, E. Santacesaria, M. Morbidelli, P. Schwarz, C. Divo, Synthesis of epichlorohydrin by elimination of hydrogen chloride from chlorohydrins. 1. Kinetic aspects of
the process, Ind. Eng. Chem. Process. Des. Dev. 18 (1979) 424427.
[9] S. Carr, E. Santacesaria, M. Morbidelli, P. Schwarz, C. Divo, Synthesis of epichlorohydrin by elimination of hydrogen chloride from chlorohydrins. 2. Simulation of the
reaction unit, Ind. Eng. Chem. Process. Des. Dev. 18 (1979) 428433.
[10] L. Ma, J. Zhu, X. Yuan, Q. Yue, Synthesis of epichlorohydrin from dichloropropanols:
Kinetic aspects of the process, Chem. Eng. Res. Des. 85 (2007) 15801585.
[11] F. Bezzo, A. Bertucco, A. Forlin, M. Barolo, Steady-state analysis of an industrial
reactive distillation column, Sep. Purif. Technol. 16 (1999) 251260.
[12] R. Taylor, R. Krishna, Modelling reactive distillation, Chem. Eng. Sci. 55 (2000)
51835229.

[13] E. Milchert, A. Krzyzanowska, A. Wolosiak-Hnat, W. Pazdzioch, The inuence of

technological parameters on dehydrochlorination of dichloropropanols, Ind. Eng.
Chem. Res. 51 (2012) 35753579.
[14] K. Jahnisch, V. Hessel, H. Lowe, M. Baerns, Chemistry in microstructured reactors,
Angew. Chem. Int. Ed. 43 (2004) 406446.
[15] T. Illg, P. Lb, V. Hessel, Flow chemistry using milli-and microstructured
reactorsFrom conventional to novel process windows, Bioorg. Med. Chem. 18
(2010) 37073719.
[16] C.B. Ye, G.W. Chen, Q. Yuan, Process characteristics of CO2 adsorption by aqueous
monoethanolamine in a microchannel reactor, Chin. J. Chem. Eng. 20 (2012) 111119.
[17] K. Wang, Y.J. Wang, G.G. Chen, G.S. Luo, J.D. Wang, Enhancement of mixing and mass
transfer performance with a microstructure minireactor for controllable preparation
of CaCO3 nanoparticles, Ind. Eng. Chem. Res. 46 (2007) 60926098.
[18] T.B. Zhang, D.W. Xin, Y.C. Lu, G.S. Luo, Direct precipitation for a continuous synthesis
of nanoiron phosphate with high purity, Ind. Eng. Chem. Res. 53 (2014) 67236729.
[19] K. Wang, Y.C. Lu, Y. Xia, H.W. Shao, G.S. Luo, Kinetics research on fast exothermic
reaction between cyclohexanecarboxylic acid and oleum in microreactor, Chem.
Eng. J. 169 (2011) 290298.
[20] N. Kockmann, M. Gottsponer, B. Zimmermann, D.M. Roberge, Enabling continuousow chemistry in microstructured devices for pharmaceutical and ne-chemical
production, Chem. Eur. J. 14 (2008) 74707477.
[21] J.S. Zhang, Y.C. Lu, Q.R. Jin, K. Wang, G.S. Luo, Determination of kinetic parameters of
dehydrochlorination of dichloropropanol in a microreactor, Chem. Eng. J. 203 (2012)
142147.
[22] J.S. Zhang, K. Wang, X.Y. Lin, Y.C. Lu, G.S. Luo, Intensication of fast exothermic reaction by gas agitation in a microchemical system, AIChE J 60 (2014) 27242730.
[23] J.S. Zhang, K. Wang, Y.C. Lu, G.S. Luo, Characterization and modeling of micromixing
performance in micropore dispersion reactors, Chem. Eng. Process. 49 (2010)
740747.
[24] J. Gaca, G. Wejnerowska, P. Cysewski, Mechanism of the acidic hydrolysis of epichlorohydrin, J. Phys. Org. Chem. 24 (2011) 10451050.
[25] M. Zong, H. Jiang, Y. Fang, G. Ren, Synthesis of epichlorohydrin, Spec. Petrochem. 24
(2007) 3235.