Article

pubs.acs.org/IECR

Measurement and Modeling for Isobaric Vapor−Liquid Equilibrium

of Binary 1,3,5-Trichlorobenzene + 3,5-Dichloroaniline and
3‑Chloroaniline + 3,5-Dichloroaniline Systems
Shuo Han,† Gan-Bing Yao,† Yun-Hao Zhong,‡ Cun-Bin Du,† Long Meng,† and Hong-Kun Zhao*,†
†
College of Chemistry & Chemical Engineering, Yang Zhou University, Yang Zhou, Jiangsu 225002, People’s Republic of China
‡
Overseas Education College, Nanjing Tech University, Nanjing, Jiangsu 225002, People’s Republic of China
*
S Supporting Information

ABSTRACT: The isobaric vapor−liquid equilibrium data for binary systems of 1,3,5-trichlorobenzene (1) + 3,5-dichloroaniline
(2) and 3-chloroaniline (1) + 3,5-dichloroaniline (2) were measured experimentally under pressures of 20.00, 60.00 and 101.20
kPa using an inclined ebulliometer. Thermodynamic consistency of the vapor−liquid equilibrium data was tested by means of the
Herington semiempirical method, and the isobaric vapor−liquid equilibrium data were correlated with three activity coefficient
models, which were Wilson, NRTL and UNIQUAC. The energy interaction parameters were acquired by a nonlinear least-
squares regression method for the three thermodynamic models. Results indicate that all the three activity coefficient models can
describe the measured isobaric vapor−liquid data acceptably. The isobaric vapor−liquid equilibrium data determined in the
present work are important in the design and operation for purification of 3,5-dichloroaniline by distillation.

■ INTRODUCTION
3,5-Dichloroaniline (CAS: 626-43-7) is a specific chlorinated
Herington Test (Area Test) method and (3) fit the isobaric
vapor−liquid equilibrium data by Wilson, NRTL and UNI-
QUAC models.
aniline derivative. It is an important substance for production of
agricultural fungicides, herbicides, insecticides and plant growth
adjusting agents. Furthermore, in medicine industry, it is also
applied to synthesis of quinoline derivatives.1,2 In recent years,
■ EXPERIMENTAL SECTION
Materials. 3,5-Dichloroaniline was provided by Jiangsu Kuai
some new applications have been found for 3,5-dichloroaniline. Da chemical Co., LTD, China, and had a purity of 0.978 with
It is of importance to study 3,5-dichloroaniline for improvement mass fraction. It was further refined by mean of recrystallization
of ring phthalimide agricultural fungicide.3−10 in (methanol + water) mixture with the volume ratio of 1:1. The
Many methods have been put forward to produce 3,5- final purity of 3,5-dichloroaniline used in experiment was
dichloroaniline. The earliest way reported in the literature was confirmed by gas chromatography equipped with a flame
the reduction of 3,5-dichlorobenzene,11 which is still an ionization detector (Agilent 7890A Infinity GC, Agilent
important method in producing 3,5-dichloroaniline at present. Technologies), the mass fraction of which was 0.9972. 1,3,5-
However, some reactants and byproducts, e.g., 1,3,5-trichlor- Trichlorobenzene and 3-chloroaniline were provided from
obenzene and 3-chloroaniline, exist in the aim product during the Shanghai Ziyi Reagent Factory, which had a purity of 0.9964
preparation process. The crude 3,5-dichloroaniline should be and 0.9989 in mass fraction, respectively. No further purification
purified from its mixture before use. Distillation is an effective was made for the two substances. More details about the 3,5-
operation process to purify the crude mixtures, and the saturated dichloroaniline, 1,3,5-trichlorobenzene and 3-chloroaniline are
vapor pressure data and vapor−liquid equilibrium for the system summarized and presented in Table 1S of the Supporting
of 3,5-dichloroaniline + 1,3,5-trichlorobenzene + 3-chloroaniline Information.
must be known in advance. To the best of the authors’ present Apparatus and Procedure. The apparatus employed to
knowledge, the saturated vapor pressure data of pure 3,5- measure the vapor−liquid equilibrium data was described
dichloroaniline were determined by Li and co-workers.12 elsewhere.13,14 The apparatus consisted of an inclined ebulli-
However, the vapor−liquid equilibrium data of the 3,5- ometer, a set of digital vacuum pressure controller with high
dichloroaniline with 1,3,5-trichlorobenzene and 3-chloroaniline, accuracy and a vacuum pump (U4.20 model, Becker, Germany).
which are essential for the design of a suitable distillation tower, The pressure was adjusted by a needle valve and recorded by
have not been reported so far. To enrich the detailed means of a pressure sensor (type: RL-P-Y). The precision of the
fundamental data of distillation for engineering fields, the object pressure controller was about 0.02 kPa for the range from 0 to
of the present paper is to (1) determine the isobaric vapor−liquid 200 kPa. A Hewlett-Packard quartz thermometer (model
equilibrium data for binary systems of 1,3,5-trichlorobenzene (1)
+ 3,5-dichloroaniline (2) and 3-chloroaniline (1) + 3,5- Received: January 4, 2015
dichloroaniline (2) under 20.00, 60.00 and 101.20 kPa by Revised: March 19, 2015
using an inclined ebulliometer, (2) check thermodynamic Accepted: March 27, 2015
consistency of the vapor−liquid equilibrium data by the Published: April 6, 2015

© 2015 American Chemical Society 3706 DOI: 10.1021/acs.iecr.5b00031

Ind. Eng. Chem. Res. 2015, 54, 3706−3710
Industrial & Engineering Chemistry Research Article

Figure 1. Isobaric vapor−liquid equilibrium data for the 1,3,5-trichlorobenzene (1) + 3,5-dichloroaniline (2) binary system. (a) 20.00 kPa; (b) 60.00
kPa; (c) 101.20 kPa; (■, ●) experimental data in this work; dotted line represents correlated data with Wilson model; dashed line represents correlated
data with NRTL model; solid line represents correlated data with UNIQUAC model.

2810A) with a 2850A sensing probe was employed to measure ρ = (τ 2 − B)/A (1)
the boiling point temperature, which was calibrated with a
standard platinum resistance thermometer (model 8163-B) where A and B represent equipment constants, which are
beforehand. The total uncertainty in the boiling point measure- calibrated in advance with the reference densities of pure water
ment was evaluated to be no more than 0.02 K, which was within and dry air at 360.15 K. The uncertainty of the calculated density
the limits of investigated temperature. The good performance of was estimated to be ±1 kg·m−3.
the equipment was confirmed in the previous experiment. Analysis. The contents of the equilibrium vapor phase and
In this study, about 75 g of sample (1,3,5-trichlorobenzene (1) liquid phase were analyzed by gas chromatography (GC, Model
+ 3,5-dichloroaniline (2) or 3-chloroaniline (1) + 3,5- 7890A, Agilent Technologies), which was equipped with a flame
dichloroaniline (2) with different ratios) was put into the ionization detector (FID). The type of capillary column used in
inclined ebulliometer. The pressure was maintained to a desired the experiment was DB-1701 (30m × 0.32 mm × 0.25 μm). The
value by the vacuum pump. The vapor phase and the liquid phase carrier gas was high purity of nitrogen, the speed of which was 10
were continuously circulating to ensure sufficient contact of the mL·min−1. The temperature of the injection chamber was
phases. The vapor phase was condensed in the condenser. When maintained at 573 K. At first the detector temperature was set to
the constant temperature was kept at least 35 min, the 410 K, and then it raised to 510 K with a heating rate of 10 K·
equilibrium was assumed to be achieved. In order to confirm min−1. The three compounds can be separated well under such
the equilibrium condition, the condensate and the liquid phase conditions. The GC analysis was calibrated with a series of
were examined with an interval of 15 min until the two phase standard solutions prepared by an electronic balance (precision:
composition did not vary. When equilibrium was reached, the ± 0.0001 g). The injected sample volume of each phase taken for
equilibrium temperature was recorded. The samples of the analysis was 0.2 μL. The absolute error was no greater than 0.001
equilibrium vapor phase and liquid phase were taken out in mole fraction. The analysis was carried out three times for each
simultaneously and then were analyzed by GC. sample, and the average value of the three measurements was
Density Measurement. The density of 3,5-dichloroaniline considered as the final value.
and 1,3,5-trichlorobenzene in liquid state are not found in the
previous publications. They were measured by a digital
densimeter (Anton Paar, model DMA 45) at 360.15 K in the
■ RESULTS AND DISCUSSION
Densities. The measured densities of 3,5-dichloroaniline and
present work. The temperature was measured by a glass 1,3,5-trichlorobenzene were 1580.8 and 1220.4 kg·m−3 at 360.15
thermometer with a precision of ±0.02 K, and controlled by a K, which are shown in Table 1S of the Supporting Information.
constant temperature water bath. The density of the sample in The density value of 3-chloroaniline was 1215.9 kg·m−3 at 293.15
the vibrating U-tube can be calculated according to the K, which showed good agreement with the literature value
oscillation period, τ, by using eq 1 (1216.1 kg·m−3).
3707 DOI: 10.1021/acs.iecr.5b00031
Ind. Eng. Chem. Res. 2015, 54, 3706−3710
Industrial & Engineering Chemistry Research Article

Figure 2. Isobaric vapor−liquid equilibrium data for the 3-chloroaniline (1) + 3,5-dichloroaniline (2) binary system. (a) 20.00 kPa; (b) 60.00 kPa; (c)
101.20 kPa; (■, ●) experimental data in this work; dotted line represents correlated data with Wilson model; dashed line represents correlated data with
NRTL model; solid line represents correlated data with UNIQUAC model.

Vapor−Liquid Equilibrium Data. The measured isobaric minimum boiling points of the studied binary systems,
vapor−liquid equilibrium data for the binary 1,3,5-trichloroben- respectively. The activity coefficient, γi, can be evaluated
zene (1) + 3,5-dichloroaniline (2) and 3-chloroaniline (1) + 3,5- according to the experimental vapor−liquid equilibrium data.
dichloroaniline (2) systems at 20.00, 60.00 and 101.20 kPa are The calculated values of (D − J) for the two binary systems of
reported in Tables 2S and 3S of the Supporting Information. 1,3,5-trichlorobenzene (1) + 3,5-dichloroaniline (2) and 3-
Furthermore, the plots of T−x−y and y−x at the three pressures chloroaniline (1) + 3,5-dichloroaniline (2) under 20.00, 60.00
are plotted in Figures 1 and 2. It can be seen from these figures and 101.20 kPa are presented in Table 4S of the Supporting
that the two binary systems are not azeotropic. Information, which illustrates that the isobaric vapor−liquid
Thermodynamic Consistency Test. Some experimental equilibrium data measured in the present work can meet the
errors could be introduced from deviation of the activity criteria and are confirmed to be thermodynamically consistent.
coefficients derived from the Gibbs−Duhem equation.15 To Data Correlations. When a vapor−liquid system reaches
confirm the reliability of the data for isobaric vapor−liquid equilibrium, the chemical potential or fugacity of each
equilibrium, the thermodynamic consistency of the experimental component are equal in two phases. The thermodynamic
data for the binary systems under three pressures was checked by relationships for the vapor−liquid phase equilibrium can be
the Herington semiempirical method.16,17 According to the test expressed as18
of the Herington area, if the (D − J) value is less than 10, the
⎧ L S ⎫
⎪ Vi (p − p ) ⎪
vapor−liquid equilibrium data are regarded as to be
yi φi V p = xiγipiS φiSexp⎨ i
⎬
thermodynamically consistent. D and J can be expressed by eqs ⎪
RT ⎪
⎩ ⎭ (4)
2 and 3, respectively.
x =1 where T is equilibrium temperature under pressure p, ViL
∫x =1 0 ln(γ1/γ2)dx1 represents the molar volume of constituent i, φiV and φiS
1
D = 100 × x1= 1 represent the fugacity coefficients of constituent i in the vapor
∫x = 0 |ln(γ1/γ2)|dx1 (2) phase and in the pure state, γi represent the activity coefficient of
1
constituent i, pSi represents the saturated vapor pressure of pure
Tmax − Tmin constituent, yi and xi represent the content of component i in
J = 150 × molar fraction in the vapor phase and liquid phase, respectively,
Tmin (3) and R is the universal gas constant. As the experimental pressure
studied in the present work is low, the exponential term in eq 4 is
where ∫ xx11==01|ln(γ1/γ2)|dx1 is total area of ln(γ1/γ2) − x1 above
1. The vapor phase can be regarded as ideal gas; as a result, the
the x-axis and below the x-axis, and ∫ x1x1==01ln(γ1/γ2)dx1, the values of φVi and φSi are also equal to 1. Equation 5 can be
difference between the area of ln(γ1/γ2) − x1 above the x-axis and deduced from eq 4, which is usually used to calculate the activity
below the x-axis. Tmax and Tmin stand for the maximum and coefficients for equilibrium liquid phase and vapor phase.
3708 DOI: 10.1021/acs.iecr.5b00031
Ind. Eng. Chem. Res. 2015, 54, 3706−3710
Industrial & Engineering Chemistry Research Article

pyi no more than 0.11 K and 0.0013 for the Wilson equation, 0.14 K
γi =
piS xi and 0.0016 for the NRTL equation and 0.33 K and 0.0045 for the
(5)
UNIQUAC equation. Therefore, the three models, Wilson
The isobaric vapor−liquid equilibrium data determined in the model, NRTL model and UNIQUAC model, can satisfactorily
present work at three different pressures (20.00, 60.00 and describe the vapor−liquid equilibrium behavior of the binary
101.20 kPa) were correlated by three activity coefficient models, systems of 1,3,5-trichlorobenzene (1) + 3,5-dichloroaniline (2)
Wilson, NRTL and UNIQUAC, respectively (for details, see the and 3-chloroaniline (1) + 3,5-dichloroaniline (2) at the different
Three Activity Coefficient Models of the Supporting Informa- pressures. In general, the Wilson model gives the best correlated
tion). results for the two binary systems.
The two parameters’ values of UNIQUAC model for 3,5- To illustrate the effect of pressure on the isolation of 3,5-
dichloroaniline, 1,3,5-trichlorobenzene and 3-chloroaniline are dichloroaniline via distillation, the relative volatilities (αi) of the
listed in Table 5S of the Supporting Information. Based on the two binary system at different pressures were evaluated by means
vapor−liquid equilibrium data, the values of interaction of the equation expressed as follows
parameters Λji in the Wilson equation, Δgij in the NRTL x 2iy1i
equation and uji − uii in the UNIQUAC equation were obtained αi =
by a method of nonlinear least-squares regression. The objective x1iy2i (9)
function (OF) is expressed as where i refers to the equilibrium temperature. The evaluated
⎡⎛ ⎛ Δp ⎞2 ⎤ relative volatilities are presented in Tables 2S and 3S of the
ΔTi ⎞
N 2

OF = ∑ ⎜⎜⎢ ⎟ + ⎜⎜ i ⎟⎟ ⎥ Supporting Information. It shows that the relative volatilities

⎢⎝ σT , i ⎟⎠ ⎝ σp , i ⎠ ⎥⎦ increase with decreasing in pressure for 1,3,5-trichlorobenzene
k=1 ⎣ (6) (1) + 3,5-dichloroaniline (2) system; however, for the system of
Where N stands for the number of experimental point. T is the 3-chloroaniline (1) + 3,5-dichloroaniline (2), the effect of
boiling point temperature. σT,i and σp,i are estimated standard pressure on the relative volatility was not obvious.
deviations in measured temperature and pressure for the ith Furthermore, to understand the nonideality of the binary
experiment, respectively. For all calculations with the original systems studied, the excess Gibbs free energies GE were evaluated
data presented in this work, they were assigned values σT,i = 0.01 from the following equation:
K and σp,i = 0.0005pexpt,i for all observations as a priori estimated GE(J·mol−1) = RT (x1ln γ1 + x 2 ln γ2) (10)
from the properties of the experimental setup.
The Antoine constants for 3,5-dichloroaniline, 1,3,5-trichlor- where γi represents the activity coefficient, which can be
obenzene and 3-chloroaniline are collected and shown in Table evaluated from the corresponding activity coefficient model. In
6S of the Supporting Information. The regressed binary Figures 3 and 4, the variation of GE with the composition of 1,3,5-
interaction parameters for the Wilson, NRTL and UNIQUAC
models are presented in Table 7S of the Supporting Information.
The root-mean-squared deviations (σT and σy1) between the
evaluated and experimental values are defined as
N
∑i = 1 (Tical − Tiexp)2
σT =
N (7)

N
∑i = 1 (y1cal
i
− y1exp
i
)2
σy1 =
N (8)
The calculated root-mean-squared deviations are also given in
Table 7S of the Supporting Information. The calculated
temperature and vapor mole fraction by using the Wilson, Figure 3. Calculated excess energy from Wilson equation against liquid
NRTL and UNIQUAC models are presented in Tables 2S and molar composition for the binary system of 1,3,5-trichlorobenzene (1) +
3S of the Supporting Information, and plotted in Figures 1 and 2. 3,5-dichloroaniline (2): ---, 20.00 kPa; -·-, 60.00 kPa; , 101.20 kPa.
Furthermore, the absolute deviation of the vapor fraction (Δy)
and equilibrium temperature (ΔT) between the experimental
and calculated data of the vapor composition and the trichlorobenzene and 3-chloroaniline is plotted. Figures 3 and 4
temperature for the systems of 1,3,5-trichlorobenzene (1) + illustrate that positive deviations behavior from the Raoult’s law
3,5-dichloroaniline (2) and 3-chloroaniline (1) + 3,5-dichlor- is observed for the binary systems of 1,3,5-trichlorobenzene (1) +
oaniline (2) are also listed in Tables 2S and 3S of the Supporting 3,5-dichloroaniline (2) and 3-chloroaniline (1) + 3,5-dichlor-
Information, respectively. The results show that the calculated oaniline (2) at the different pressures over the whole
composition range.

■
data have a good agreement with the experimental ones. For the
1,3,5-trichlorobenzene (1) + 3,5-dichloroaniline (2) system, the
average absolute deviations (AD) of the boiling point and the CONCLUSIONS
vapor mole fraction fitted by activity coefficient models are no The isobaric vapor−liquid equilibrium data for the binary
more than 0.60 K and 0.0055 for the Wilson equation, 0.42 K and systems of 1,3,5-trichlorobenzene (1) + 3,5-dichloroaniline (2)
0.0055 for the NRTL equation and 0.57 K and 0.0078 for the and 3-chloroaniline (1) + 3,5-dichloroaniline (2) were
UNIQUAC equation. For the system of 3-chloroaniline (1) + determined at 20.00, 60.00 and 101.20 kPa, respectively. The
3,5-dichloroaniline (2), the average absolute deviations (AD) are two binary systems exhibited no azeotropic behavior. The
3709 DOI: 10.1021/acs.iecr.5b00031
Ind. Eng. Chem. Res. 2015, 54, 3706−3710
Industrial & Engineering Chemistry Research Article

(4) Gu, X.; Li, D.; Zhang, Q. H.; Jiang, J. L.; Liu, X. Q. Synthesis of 3,5-
Dichloroaniline by Catalytic Hydrogenation-Catalytic Hydrogenation
of Pentachloronitrobenzene. Agrochemicals 2010, 49, 723 (China).
(5) Jersak, U. D. C. D.; Scheuermann, H. D. C. D. Verfahren Zur
Herstellung Von 3,5-Dihalogenanilinen. DE Patent 2,720,316,
November 16, 1978.
(6) Li, X. N.; Lu, C. S. Method for Synthesizing Halogen Aromatic
Amines by High-Selectivity Liquid Phase Hydrogenation under the
Condition of no Solvent. CN Patent 101,811,973, August, 25, 2010.
(7) Qiu, Z. G. Method for Preparing Halogenated Nitrobenzene
Selective Hydrogenation High-Activity Nano-Ruthenium Catalyst. CN
Patent 101,569,859, November 4, 2009.
(8) Li, J. X. 3,5-Dichloroaniline Preparing Process. CN Patent
1,690,040, November 2, 2005.
Figure 4. Calculated excess energy from Wilson equation against liquid (9) Smith, J. O.; Petruska, M. A.; Longlet, J. J. Process for the
molar composition for the binary system of 3-chloroaniline (1) + 3,5- Preparation of Substituted Halogenated Aniline. WO Patent
dichloroaniline (2): ---, 20.00 kPa; -·-, 60.00 kPa; , 101.20 kPa. 2,004,054,961, July 1, 2004.
(10) Wang, S.; Gu, X. Three Steps of para-Chloronitrobenzene
Synthesis Method of 3,5-Dichloroaniline Process Optimization and
thermodynamic consistency of the experimental data was Research. Guangxi J. Light Ind. 2008, 24, 17 (China).
checked by using the Herington semiempirical method, and all (11) Valentovic, M. A.; Rogers, B. A.; Meadows, M. K.; Conner, J. T.;
the vapor−liquid equilibrium data were confirmed to be Williams, E.; Hong, S. K.; Rankin, G. O. Characterization of
thermodynamically consistent. The isobaric vapor−liquid Methemoglobin Formation Induced by 3,5-Dichloroaniline, 4-Amino-
2,6-dichlorophenol and 3,5-Dichlorophenylhydroxylamine. Toxicology
equilibrium data were correlated by three thermodynamic 1997, 118, 23.
models, which corresponded to Wilson, NRTL and UNIQUAC. (12) Li, Y. X.; Zhao, H. K.; Mao, M. L.; Yang, Z. P. Measurement and
The average absolute deviations (AD) of the boiling point and Correlation of the Vapor Pressure of 3,5-Dichloroaniline. J. Chem. Eng.
the vapor mole fraction calculated by three thermodynamic Data 2013, 58, 1629.
models are all least than 0.60 K and 0.0079 for the two systems. (13) Yao, G. B.; Yang, Z. P.; Zhang, B.; Xu, H.; Zhao, H. K. Vapor
The calculated results using the three models agree well with Pressure and Isobaric Vapor−Liquid Equilibrium for Dichloronitro-
experimental values for the binary systems of 1,3,5-trichlor- benzene Isomers. Fluid Phase Equilib. 2014, 367, 103.
obenzene (1) + 3,5-dichloroaniline (2) and 3-chloroaniline (1) + (14) Dong, H.; Wu, C.; Yang, X. F.; Qiu, H. Y. Measurement and
3,5-dichloroaniline (2) at three different pressures. Correlation of Saturated Vapor Pressure of 2,4,6,8,10-Pentamethylcy-

■
clopentasiloxane by means of an Inclined Ebulliometer. Thermochim.
ASSOCIATED CONTENT Acta 2009, 483, 66.
(15) Jackson, P. L.; Wilsak, R. A. Thermodynamic Consistency Tests
*
S Supporting Information
Based on the Gibbs-Duhem Equation Applied to Isothermal, Binary
Equations of Three Activity Coefficient Models listing equations Vapor-Liquid Equilibrium Data: Data Evaluation and Model Testing.
of the three activity coefficient models, Wilson, NRTL and Fluid Phase Equilib. 1995, 103, 155.
UNIQUAC, tables showing provenance and purity of the (16) Herington, E. F. G. Test of Experimental Isobaric Vapor-Liquid
materials, experimental vapor−liquid equilibrium data, thermo- Equilibrium Data. J. Inst. Petrol. 1951, 37, 457.
dynamics consistency test for two systems, values of r and q, the (17) Chen, S. H.; Bao, Z. B.; Lü, Z. Z.; Yang, Y. W.; Xu, W. G.; Chen, Z.
Antoine constants of the three compounds used in the present M.; Ren, Q. L.; Su, B. G.; Xing, H. B. Vapor−Liquid Equilibrium for the
work, and correlation interaction parameters. This material is 1,1,1-Trifluorotrichloroethane + Sulfuryl Chloride System at 101.3 kPa.
J. Chem. Eng. Data 2014, 59, 16.
available free of charge via the Internet at http://pubs.acs.org.

■
(18) Ness, H. V.; Smith, J.; Abbott, M. Introduction to Chemical
Engineering Thermodynamics, 6th ed.; McGraw-Hill: New York, 2001.
AUTHOR INFORMATION
Corresponding Author
*H.-K. Zhao. E-mail: [email protected]. Tel.: + 86 514
87975568. Fax: + 86 514 87975244.
Notes
The authors declare no competing financial interest.

■ ACKNOWLEDGMENTS
We thank the National Natural Science Foundation of China for
their support (Project number: 21406192). Furthermore, the
Yangzhou City Science and Technology Bureau, China (Project
numbers 2012038-3 and YZ2011139) are also appreciated.

■ REFERENCES
(1) Valentovic, M. A.; Lo, H. H.; Brown, P. I.; Rankin, G. O. 3,5-
Dichloroaniline Toxicity in Fischer 344 Rats Pretreated with Inhibitors
and Inducers of Cytochrome P450. Toxicol. Lett. 1995, 78, 207.
(2) Yan, S. J.; Yang, J. Comment on the Preparation of 3,5-
Dichloroaniline. Shenyang Chem. Ind. 1998, 27, 22 (China).
(3) Cao, G. H. 3,5-Dichloroaniline Synthesis Review. Hubei Chem.
1993, 16, 24 (China).

3710 DOI: 10.1021/acs.iecr.5b00031

Ind. Eng. Chem. Res. 2015, 54, 3706−3710