Noncatalytic Liquid Phase Air Oxidation of Ethylbenzene to 1-Phenyl Ethyl Hydroperoxide in Low Oxygen Volume Fraction

Article abstract of DOI:10.1021/acs.oprd.7b00284

The catalyst- and solvent-free liquid phase air oxidation reaction of ethylbenzene (EB) to 1-phenyl ethyl hydroperoxide (PEHP) was investigated at scheduled temperature and pressure with a separate deacidifying zone. The oxidation reaction was done in the

Full text of DOI:10.1021/acs.oprd.7b00284

Subscriber access provided by STEPHEN F AUSTIN STATE UNIV
Full Paper
Non-catalytic liquid phase air oxidation of ethylbenzene to 1-
phenyl ethyl hydroperoxide in low oxygen volume fraction
Hossein Roohi, and Mehrdad Rajabi
Org. Process Res. Dev., Just Accepted Manuscript • DOI: 10.1021/acs.oprd.7b00284 • Publication Date (Web): 22 Jan 2018
Downloaded from http://pubs.acs.org on January 23, 2018
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Organic Process Research & Development is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 36
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Non-catalytic liquid phase air oxidation of
ethylbenzene to 1-phenyl ethyl hydroperoxide
in low oxygen volume fraction
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
†
∗,‡,¶
Hossein Roohi and Mehrdad Rajabi
†
Department of Chemistry, Faculty of Science, University of Guilan, Rasht, Iran.
‡
Department of Chemistry, University Campus 2, University of Guilan, Rasht, Iran
¶
Research Department, National Petrochemical Co., Research & Technology, Arak, Iran
E-mail: mehrdadrajabi@hotmail.com
Phone: +98 91 2289 6761. Fax: +98 13 3336 7066
1
ACS Paragon Plus Environment

Organic Process Research & Development
Page 2 of 36
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Graphical Table of Content
Air+N2
EB
Condenser
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
EB Recycle
PEHP+ACP+PE
Alkali
Solution
2
ACS Paragon Plus Environment

Page 3 of 36
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Abstract
The catalyst and solvent free liquid phase air oxidation reaction of ethylbenzene
EB) to 1-phenyl ethyl hydroperoxide (PEHP) was investigated at scheduled tempera-
(
ture and pressure with a separate de-acidifying zone. The oxidation reaction was done
in the titanium fabricated reactor at an oxygen volume fraction in the outlet gas less
than 0.3%. The effects of several reaction parameters were studied in detail. In the
optimal reaction conditions and instructions given in this work, the high selectivity to
PEHP (above 95%) was achieved with EB conversion around 13%, and a clean feed
(water and acid-free) was provided for propene epoxidation reaction.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Keywords
Ethylbenzene air oxidation, Liquid-phase, 1-Phenyl ethyl hydroperoxide, Non-catalytic
3
ACS Paragon Plus Environment

Organic Process Research & Development
Page 4 of 36
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Introduction
The selective liquid phase air oxidation (autoxidation) of alkylbenzenes to valuable oxy-
genated products (ketones, aldehydes, and carboxylic acids) through their hydroperoxide
intermediates is a chemical reaction of great industrial importance.1–3 The organic hydroper-
oxides (ROOHs) are interesting products of hydrocarbon autoxidation which are used as
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
oxidizing agents for olefins. The important examples include cumene hydroperoxide pro-
6
duction (more than 5×10 annual tons) from the oxidation of the cumene (Hock process; for
5
phenol and acetone production), and the synthesis of terephthalic acid from Para-xylene
6
7
6
oxidation, or from methane, (50×10 annual tons) for the polyester industry.
Liquid phase ethylbenzene (EB) autoxidation to 1-phenyl ethyl hydroperoxide (PEHP)
6
with worldwide production greater than 5×10 annual tons is another interesting hydrocar-
bon autoxidation process and improving the efficiency and selectivity toward the PEHP is
the first objective for this reaction. PEHP is not the target, but an intermediate product
used as an oxygen carrier in propene epoxidation reaction. As shown in Scheme 1, the
PEHP production is a key step in the simultaneous production of styrene monomer (SM)
and propene oxide (PO) through the well-known SMPO process.8
Propene
Air
EB
Alkali Washing and
Concentration
Oxidation
Flashing
Recycled EB
PEHP in EB+ACP+1-PE
Recycled EB
Focus area
EB oxidation to PEHP and
Epoxidation feed preparation)
(
Epoxidation
Hydrogenation
1
-PE
Recycled
Propene
ACP
1
-PE
Separation
PO
Dehydration
SM
Scheme 1: SMPO Process
At present, the EB oxidation reaction is carried out industrially under conditions that
◦
are free of catalyst and solvent at temperatures between 141 and 148 C and a pressure of
4
ACS Paragon Plus Environment

Page 5 of 36
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
three atmospheres.9,10 Generally, during the oxidation reaction, EB conversion is maintained
at about 12 wt.%. Nevertheless, for commercial applications, PEHP selectivity should be
more than 95%. The main by-products are acetophenone (ACP) and 1-phenylethanol (1-
PE). The ACP is converted to 1-PE by the hydrogenation process and 1-PE is converted
into the styrene by the dehydration process.11 Particular amounts of carboxylic acids and
water are produced during the reaction, which is traditionally removed by alkali treatments.
The oxidation reaction mixture is concentrated by vacuum distillation up to 20 wt.% with
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
respect to PEHP, which is utilized as a feed (oxygen carrier) for the catalytic reaction of
propene epoxidation.8
The oxidation reaction of EB in the presence of homogeneous or heterogeneous metal
catalysts using oxidants such as H O , tert-butyl hydroperoxide (TBHP), and O has been
2
2
2
studied by others.12–15 However, when metal-containing compounds are used in the alky-
laromatics oxidation processes, the ROOHs selectivity is low, and the products are mainly
ketones and alcohols.16,17 Therefore, in order to achieve high selectivity of ROOHs, it is
necessary to use metal-free systems. In addition, since the epoxidation process is strongly
hindered by the presence of metallic catalysts and it is very difficult to remove them by
distillation,17 it was decided that the oxidation reaction of EB was carried out without any
catalyst, that provided the very best performance in PEHP production.
The oxidation of EB in the presence of metal-free organic catalysts such as
N-Hydroxyphthalimids (NHPI) has also been studied by others.18–22 In these processes, NHPI
in combination with some metal salts or aldehydes acts as a precursor to phthalimide-N-Oxyl
(
PINO) radicals. The PINO radicals are able to abstract hydrogen atoms from organic sub-
strates. The use of NHPI in the oxidation processes of alkylaromatics has many advantages,
including higher conversion, greater hydroperoxide selectivity, and milder reaction condi-
tions, compared to non-catalytic processes. However, there are serious drawbacks with the
use of NHPI, for example, product separation is difficult, and the NHPI catalyst is not easily
recovered and reused, leads to low synthetic efficiency.
5
ACS Paragon Plus Environment

Organic Process Research & Development
Page 6 of 36
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
In recent years, the heterogenization of homogeneous catalysts, such as immobilization of
homogeneous catalysts on solid supports, has been an important field of research because it
allows researchers to design and prepare more effective catalysts.23–25 However, immobilized
NHPI is rarely reported.26–28 Besides, due to the low solubility of NHPI in hydrocarbons,
the use of a polar solvent is necessary.29 Additionally, solvents such as acetonitrile or acetic
acid are hydrogen-bond acceptors and form complexes with NHPI, which reduces catalytic
activity.30,31Also, due to the polarity of PEHP, the solubility of NHPI increases in higher
concentrations of PEHP and its recovery by cooling the reaction mixture is more difficult.
These problems limit the use of NHPI in industrial applications.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Although the oxidation of EB to PEHP is important from an industrial point of view, the
works published in the literature for non-catalytic EB air oxidation to PEHP are extremely
scant.32–34 It has been substantially studied kinetically.35–37 However, in existing industrial
process for EB oxidation, relatively large amounts of ACP and 1-PE are produced as side
products and the processes necessary to convert these products to styrene is tedious. Al-
though this oxidation process has been invented as early as 1969,32 it is still necessary to
be considered from the scientific and industrial community. Among them, a few kinds of
literature are concentrated on the parameters of the oxidation reaction of EB, which are
essential for reaction optimization.
In this present work, we reported investigations on the liquid phase autoxidation reaction
of EB without the need for any metal or organic catalysts and with a particular attention
to PEHP synthesis. This study led to a better selectivity for PEHP and lower production
of secondary products (ACP and 1-PE). Also, a feed free of metal, water, and acid was
prepared for propene epoxidation reaction, which can improve the life of the epoxidation
catalyst. Since the distribution of the reaction products depends strongly on the reaction
conditions, the effect of several important reaction parameters was studied in detail.
6
ACS Paragon Plus Environment

Page 7 of 36
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Experimental
Ethylbenzene was oxidized in a stirred reactor of titanium with air as oxidant at the boiling
point of water at the selected pressure and without using any catalysts. The pressure was
adjusted so that the reaction mixture remained in the liquid phase while the water and light
acids produced during the reaction evaporated along with some EB and entered the vapor
phase. The evaporated mixture, then condensed and de-acidified in a separate zone by an
alkaline solution. In this separate zone, organic and aqueous phases were easily separated
from each other and a two-phase system was formed. The upper organic phase was pure
and de-acidified EB, which was continuously returned to the oxidation reactor to continue
the oxidation reaction.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Chemicals and reagents
EB was provided by the Pars Petrochemical Company in Iran, which contained 0.35% by
weight of PEHP and was employed without further purification. Synthetic zeroes air (21%
O with 79% N , dry and free of CO ) and nitrogen (99.999% of purity) were supplied by
2
2
2
Roham Gas Company in Iran. ACP and 1-PE were purchased from Aldrich and used without
further purification. Potassium Iodide, glacial acetic acid, sodium thiosulphate and isopropyl
alcohol all with an analytical grade were purchased from the Merck Company.
Triphenylphosphine (TPP) purchased from Fluka was crystallized by n-hexane and dried
◦
in a nitrogen atmosphere. Its purity was checked by its melting point at 80 C and also by
thin layer chromatography.
Reaction equipment and setup
The semi-continuous laboratory reactor utilized in this work for the oxidation of EB is shown
in Fig. 1. EB containing a small amount of PEHP is introduced as a feed into the reactor,
and the air diluted with nitrogen is continuously introduced into the reactor.
7
ACS Paragon Plus Environment

Organic Process Research & Development
Page 8 of 36
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Impeller Driver
Press.
Sensor
Temp.
Sensor
Gas Inlet
Outlet
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Heating
agent Out
Side
Glass
Side
Glass
Heating
agent in
Drain
Figure 1: Ethylbenzene oxidation reactor
The reaction vessel was a 1.2 liter double jacket continuous stirred-tank reactor (CSTR)
connected to a titanium turbine type impeller powered by a variable speed (0-3000 rpm)
engine, a dual jacket reflux condenser, a thermometer, a pressure sensor, two gas inlet
systems with porous titanium disk mounted on the bottom; one for air and the other for
N . For optimum performance, the baffles were installed inside the reactor. The gas and
2
vapor mixtures that passed through the reactor were transferred through a reflux condenser
◦
◦
at a temperature of 80 to 120 C and then cooled down through a condenser at −5 C. In
the condenser, the vapors turned into liquid, and the remaining gas was heated to room
temperature and then transferred to an oxygen-analyzer (Servomex 2223 oxygen analyzer)
to measure the Oxygen Volume Fraction (OVF) in the exhaust gas. The OVF in the exhaust
gas was adjusted by changing the flux of the inlet air. The condensed vapors containing EB,
water, and light acids were transferred to a neutralization/separation vessel containing 5
wt.% sodium hydroxide aqueous solution. In this separate vessel, the neutralization of acids
occurred and the neutralized EB (upper phase) was continuously directed to the oxidation
reactor.
Since hydroperoxides are relatively stable against titanium, the reactors, vessels, fittings,
8
ACS Paragon Plus Environment

Page 9 of 36
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
condensers, thermocouples, pressure sensors, impeller and sampling facilities were all selected
from titanium to prevent the unwanted hydroperoxide decomposition. The reactor vessel was
purchased from Buchiglasuster Company/Switzerland/(www.buchiglas.ch).
Using USH400 Lauda heating circulator, both the reactor vessel and the reflux condenser
were heated. The task of this circulator was to heat the reaction mixture at the beginning of
the oxidation reaction, as well as cool the reaction mixture after the reaction was sufficiently
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
progressed. The synthetic zeroes-air (dry and free of CO ) and nitrogen gas, through two
2
separate pressure cylinders and after passing through the mass flow controllers, Brooks model
5
850TR, were introduced into the oxidation reactor.
Ethylbenzene autoxidation: general procedure
The oxidation experiments of EB were carried out in batch and semi-continuous mode with
gas injection continuously. In a typical experiment, 900 mL (780.30 g) of EB containing
approximately 0.35% by weight of PEHP was charged to the reactor. The reactor was
flushed with N gas to replace existing air and was pressurized to around three atmospheres.
2
◦
Then the reactor content was heated to 155 C and stirred vigorously at 1400 rpm. As soon
as the temperature reached a stable state, a mixture of air and N was sent to the reactor
2
at a rate of 200-1700 mL/min and the reaction began. In order to prevent the explosion,
it is necessary to control the OVF in the reactor outlet gas in less than 5%. This can be
accomplished by adjusting the flow of air into the reactor.
◦
In the next step, the temperature program started, so that it remained at 155 C for 30
◦
minutes and then dropped to 140 C for 240 minutes. By decreasing the reactor pressure,
water and light acids along with a small amount of EB were continuously released from the
reaction mixture. The pressure program was adjusted so that the reaction mixture remained
at three atmospheres for 30 minutes and then dropped to 1.2 atmospheres for two hours and
remained at this pressure before the end of the reaction. The reflux condenser temperature
◦
was set at 80 − 120 C so that almost all reaction products, including PEHP, ACP, 1-PE,
9
ACS Paragon Plus Environment

Organic Process Research & Development
Page 10 of 36
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
and more than 75% by weight of EB, were condensed into the reactor.
By choosing suitable temperatures for the reactor and the reflux condenser, as well as the
reactor pressure and the flow rate of the outlet gas, a mixture of gas and vapor containing
N , O , EB, water, and light acids were passed through the reflux condenser and were trans-
2
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
◦
ferred to a cold condenser. In this cold condenser with a temperature of −5 C, the vapors
condensed into a separate zone of neutralization/phase separation. To eliminate remaining
vapor that can be condensed, the exhaust gas was then passed through an additional cold
◦
trap with a temperature of −20 C, was warmed to room temperature and then directed to
an oxygen analyzer to measure the OVF.
The condensed liquid part was consisted mainly of EB and small quantities of water and
light acids produced during the oxidation reaction, which after neutralization, by contacting
with 5 wt.% NaOH aqueous solution, and the phase separation, the organic phase, which
was pure and de-acidified EB, was continuously transferred to the oxidation reactor using a
pump or gravity.
According to the data reported by Alcantara et al.,17 we selected 1400 rpm as a suitable
stirring speed. For the experiment at 1400 rpm, a sharp decrease in PEHP selectivity
was observed when the EB conversion reached 13.45%. That’s why in all experiments EB
conversion was kept at less than 11%.
Safety
Autoxidation reactions of hydrocarbons are usually exothermic and proceed through a free
radical chain mechanism and the reaction heat from non-isothermal conditions in the re-
actor decreases the selectivity of the reaction.38 For safety reasons, the concentration of
PEHP should be less than 20% by weight relative to the total weight of the reaction mix-
ture. According to the results of the experiments performed in this work, in the selected
◦
temperature range from 140 to 155 C, the PEHP was slowly decomposed and there was
always a competition between the rate of production and the rate of PEHP decomposition.
10
ACS Paragon Plus Environment

Page 11 of 36
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
During the oxidation reaction, the PEHP decomposition rate was increased with increasing
concentrations. However, after reaching a concentration of 11%, the decomposition rate was
much higher than that of PEHP production. Generally, the best way to avoid the thermal
decomposition of PEHP is to scrupulously adhere to the recommended temperature, time,
and concentrations proposed in this work.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
In order to have a reaction mixture in the liquid phase, it is necessary to perform the
reaction under pressure. The nature of the corrosion of the reaction mixture at high tempera-
tures forces the usage of equipment made of titanium. The reactor surface effect is especially
big for a small reactor in which the volume-to-wall surface ratio is low. The temperature of
the outlet gas from the reactor is usually as high as the reactor temperature. To avoid the
risk of explosion, usually after cooling to ambient temperature, the OVF in the outlet gas
is kept at less than 5%. Fortunately, since in this work the condenser temperature was less
◦
than −5 C, without the risk of explosion, the outlet gas from the reactor can have up to
2
1% oxygen.
Sampling and analytical methods
The sampling of the reaction mixture was performed at the specified time intervals. The
samples were cooled down before the gas chromatography (GC) analysis to avoid evaporation
of the hydrocarbons in the samples and the loss of the materials.
Quantitative measurement of PEHP was done by iodometric titration, utilized by Alcan-
tara et al.17 method or by a modified method, which has been described in the next section.
The quantitative measurements of EB, ACP, and 1-PE were performed by GC analysis. Af-
ter measuring the amount of PEHP and before GC analysis, toluene was added to the cooled
samples as an internal standard. The identification of oxidation products was performed by
comparing retention times for pure 1-PE and ACP in the chromatograph. In addition, the
PEHP in the chromatograph did not completely decompose into the 1-PE and ACP.
On the other hand, PEHP is easily and quantitatively reduced by Triphenylphosphine
11
ACS Paragon Plus Environment

Organic Process Research & Development
Page 12 of 36
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
TPP) to yield 1-PE,39 (Eq. 1). This reaction previously used by Toribio et al.18 for GC
(
analysis of EB oxidation reaction mixtures.
PEHP + TPP −−→ 1-PE + TPPO
(1)
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Indeed, if the chromatographic analysis of the sample is performed without hydroperoxide
reduction with TPP, the actual concentrations of the components of the sample are not
obtained correctly because only a small amount of hydroperoxide can provide a separate
peak on the chromatogram. Also, a small amount of hydroperoxide in the chromatogram
is decomposed into various derivatives, which did not actually exist in the prototype. In
addition, at high temperatures of the gas chromatography system, the reaction between
PEHP and EB can be continued and lead to the production of products that were not
actually in the reaction mixture. Therefore, PEHP cannot be analyzed by GC. So, before
analyzing the samples, it is necessary to reduce the PEHP with TPP to 1-PE.
The EB oxidation products, except PEHP, were analyzed periodically after adding a
stoichiometric amount of solid TPP and also adding a certain amount of toluene as the in-
ternal standard. They were analyzed on an Agilent Technologies 6890N gas chromatography
equipped with Cp-Wax 52CB capillary column (30 m length, 0.32 mm internal diameter
◦
and 0.5 µm film thickness) and flame ionization detector (FID), 240 C, He as a carrier gas
◦
◦
and injection temperature 210 C. The oven temperature was adjusted at 100 C for just
◦
◦
one minute and was increased by 13 C/min up to 230 C, where it was kept for just two
minutes. This GC was calibrated for EB, ACP, and 1-PE.
If 15 to 30 minutes before the GC analysis, the stoichiometric amount of solid TPP is
added to each sample and the resulting chromatogram differs significantly from the chro-
matogram of the sample without the addition of TPP, then it can be concluded that there
was PEHP in the list of oxidation products. Actually, after reduction with TPP, the 1-PE
peak raised markedly whilst the ACP peak decreased. Usually, the sum of 1-PE and ACP
concentrations in the reduced sample was not equal to the sum of 1-PE and ACP concentra-
12
ACS Paragon Plus Environment

Page 13 of 36
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
tions in the perfect sample that is untreated with TPP. This indicates that the sample has
PEHP as the main component that is not decomposed completely during the GC analysis.
In summary, to determine the actual values of the components of the EB oxidation
mixture, the samples were cooled at first, and the PEHP value was determined using an
iodometric method. Then, the stoichiometric amount of TPP was added to reduce the
PEHP to the 1-PE, and after 15 to 30 minutes, a certain amount of toluene was added to
the samples as the internal standard. The actual concentrations of EB, ACP, and 1-PE,
were quantified by integrating the corresponding chromatogram peaks and simultaneously
solving equations; (2), (3), and (4).
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
[EB]real−[EB]TPP
(2)
[
ACP]real−[ACP]TPP
(3)
(4)
[
1-PE]TPP−[1-PE]real + [1-PE]PEHP
where, [EB]real, [ACP]real, and [1-PE]real are the actual concentrations of EB, ACP, and
-PE contained in the reaction mixture at the certain moment, respectively. The [EB]TPP
ACP]TPP, and [1-PE]TPP are the observed concentrations of EB, ACP, and 1-PE determined
1
,
[
by integration of the GC peak related to EB, ACP, and 1-PE after treating the reaction
mixture with TPP, respectively. The [1-PE]PEHP is the stoichiometric quantity of 1-PE that
is produced from the reaction of PEHP with TPP (Eq. 1).
It should be noted that the ACP concentration obtained from the chromatogram be-
fore reduction with TPP (direct injection of reaction mixture) is not actually the [ACP]real
;
because PEHP is decomposed to ACP and 1-PE in the GC injector. Subsequently, after
determining the actual concentrations of the components of the EB oxidation reaction mix-
ture, the EB conversion (%) and the PEHP selectivity (%) were calculated using the most
13
ACS Paragon Plus Environment

Organic Process Research & Development
Page 14 of 36
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
4
common calculation procedure, often defined as follows:
[
PEHP] + [ACP] + [PE] + [Other]
Con.EB
=
× 100
× 100
(5)
(6)
[
EB]0
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
[
PEHP]
Sel.PEHP
=
[
PEHP] + [ACP] + [PE] + [Other]
where, [PEHP], [ACP] and [PE] are the weight percent concentrations of 1-phenyl ethyl
hydroperoxide, acetophenone and 1-phenylethanol, respectively, in the sample and [EB] is
0
the initial weight percent of EB.
Hydroperoxide determination
The standard iodometric titration for the determination of hydroperoxides has been used for
many years. According to our knowledge, only procedure in which the focus had been on the
determination of PEHP has been introduced by Alcantara et al.17 The Alcantara method
was modified in the present study, using absolute ethanol (Abs.EtOH) and solid potassium
iodide (KI) instead of 99% isopropyl alcohol (IPA) and a saturated solution of KI in IPA.
The procedure was as follows.
Sufficient Abs.EtOH (usually 20-25 mL) was added to the 0.5 g of KI in 100 mL volumetric
flask, followed by 0.5 mL of acetic acid-sodium acetate buffer. The mixture was heated almost
◦
to boiling point (65 − 70 C), kept at this temperature range for three to five minutes with
slow magnetic stirring. Then, without cooling, 0.1-0.3 g of hydroperoxide containing sample
was added and the solution was heated for further five to ten minutes. Depending on the
amount of hydroperoxide present in the sample, the reaction mixture turns yellow to red.
The liberated iodine was titrated with 0.1 N sodium thiosulfate solution until the yellow
color of the iodine disappeared. A parallel analysis was done for the blank. All results
indicated an average of at least three determinations. Because of the non-aqueous solution,
starch is not a satisfactory indicator in this solvent system. The PEHP concentration (wt.%)
in the samples was calculated according to Alcantara method.17 For the same sample, both
14
ACS Paragon Plus Environment

Page 15 of 36
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
the Alcantara method and the modified method, presented in this work, provided almost
1
identical results which were confirmed by HNMR analysis.
Actually, the selected sample for confirmation of the amount of PEHP was the crude
product obtained from the oxidation reaction of EB with 10% by weight PEHP content,
calculated by Alcantara method.17 1HNMR spectrum was obtained in deuterochloroform
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
CDCl ) with a 250 MHz Bruker instrument. Tetramethylsilane (TMS) was used as an
3
internal standard and chemical shift of the proton that is presented as δ(TMS=0). A typical
1HNMR spectrum is given in Fig. 2.
1
Figure 2: HNMR spectra in CDCl of the crude product obtained upon oxidation of EB
3
The hydroperoxy proton appeared as a broad singlet downfield from the TMS. The
chemical shift was concentration dependent, as Fujiwara et al. showed;40 at approximately
1
M in CDCl the hydroperoxy proton gives a signal at about 8 ppm downfield. Interpretation
3
and assignment of these peaks are summarized in Table 1.
Although aromatic protons appear in the same downfield region as ROOHs do, there is
no problem in identifying the O−O−H signal as the hydroperoxy proton readily undergoes
deuterium exchange. Methyl protons, β to the hydroperoxy, are unaffected by it and show
a typical sharp singlet at nearly the same position as that of the non-peroxidic analog. The
O−O−H signal was shifted up-field from 8.15 at 1.0 M to 7.75 ppm at 0.01 M; at infinite
dilution, the signal is at 7.3-7.5 ppm (extrapolation from 0.01 M). It is presumed that
the concentration dependence results from a decrease in intermolecular hydrogen bonding
15
ACS Paragon Plus Environment

Organic Process Research & Development
Page 16 of 36
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
Table 1: Assignments of peaks of HNMR spectra of the crude EB oxidation products
Chemical shift (ppm) Peak kind Iintegration Assignments
1
1
.2 ∼ 1.6
Triplet
Doublet
Singlet
4.35
2.18
1.00
3.03
0.05
0.32
10.42
CH protons of EB
3
.6 ∼ 1.8
CH protons of 1-PE
3
2.6 ∼ 2.8
CH protons of ACP
3
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
5
5
7
.8 ∼ 3.0
.0 ∼ 5.1
.2 ∼ 5.4
.3 ∼ 7.7
Tetralet
Tetralet
Tetralet
Multiplet
Doublet
α-H on EB
α-H on 1-PE
α-H on PEHP
aromatic protons
a
b
8.1 ∼ 8.2
0.89
CH protons of PEHP
3
a
b
This peak is covvered by O−H proton of 1-PE. This peak is covered by O−O−H proton of PEHP.
due to dilution.41,42 Compared with the hydroxyl proton of alcohols, the large downfield
shift of the signal of the hydroperoxy proton cannot, therefore, be attributed to an unusual
hydrogen bonding effect but is a reflection of the large de-shielding effect of the peroxy
1
group. As can be seen in table 2, for the same sample, the results of HNMR and those
of the standard iodometric method of Alcantara,17 as well as our modified method for the
PEHP determination, were in good agreement.
Table 2: Comparision of three methods for hydroperoxide determination
Method
Concentration (wt.%)
PEHP
ACP
1-PE
EB
Alcantara17
Modified
10.00
10.20
10.57
11.20
11.53
11.88
25.20
24.80
25.89
47.20
53.47
51.66
1
a
HNMR
a1
HNMR spectra was obtained in CDCl3
Results and Discussion
The non-catalytic reaction of EB autoxidation was carried out in a liquid phase at a temper-
◦
ature range of 140 to 155 C and a pressure range of 1 to 3 atm. The three main products,
including PEHP, ACP, and 1-PE were produced in this reaction (Scheme 2). Phenol, ben-
zaldehyde, benzoic acid, and water were also obtained in small quantities.
16
ACS Paragon Plus Environment

Page 17 of 36
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
H
C
H
OOH
C
O
C
OH
C
CH3
CH3
CH3
CH3
+
O −−→
+
+
2
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 2 Ethylbenzene oxidation main products
◦
In fact, the autoxidation of EB at a temperature of more than 130 C is exactly an
example of an auto-catalytic free radical chain reaction that prevents selectivity control.
The mechanism of this reaction has been extensively studied by others,12,36,37 which the
most accepted one is shown in Scheme 3;
Initiation by heat or light:
·
·
ROOH −−→ RO + HO
(7)
(8)
(9)
·
·
RO + RH −−→ ROH + R
·
·
HO + RH −−→ H O + R
2
Propagation:
Termination:
·
·
R + O −−→ ROO
(10)
(11)
2
·
·
ROO + RH −−→ ROOH + R
·
·
ROO + ROO −−→ ROH + Q−O + O
(12)
2
Scheme 3 Classical free radical chain reaction mechanism of hydrocarbon autoxidation.
Q=O denotes the respective carbonyl compounds.18
Since oxygen is too inactive for hydrogen absorption, the autoxidation of a hydrocarbon
bond (C−H) to the hydroperoxide (C−O−O−H) group with oxygen is very slow. But when
a small amount of free radicals is created by some of the initiation processes, it reacts with
·
oxygen and produces peroxy radicals (ROO ). These peroxy radicals attract hydrogen from
a hydrocarbon and promote the radical chain reaction.
17
ACS Paragon Plus Environment

Organic Process Research & Development
Page 18 of 36
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Although the oxidation of EB is favorable thermodynamically, it is kinetically hindered;
so it is often necessary to carry out oxidation at fairly high temperatures. Therefore, the
◦
temperatures of 140 to 155 C) are not unusual for the autoxidation of EB. In this temper-
ature range, the decomposition of PEHP in order to produce free radicals (reaction 7) is a
key step in the oxidation reaction. For this reason, before starting the oxidation reaction
in the liquid phase, a small amount of hydroperoxide is added as a radical initiator to the
hydrocarbon feed.37 That is why in all experiments a small amount of PEHP (0.35 wt.%)
was added to the reaction mixture.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The rate of such reactions is greatly increased by adding small amounts of transition
metal ions, which catalyze the formation of radicals upon homolytic bond cleavage of the
hydroperoxide.43,44 Nevertheless, in the absence of active metal ions, a thermal oxidation re-
action occurs. Although it was often assumed that radicals are generated upon O-O scission
in hydroperoxide (reaction 7), Turra et al.45 believed that unimolecular cleavage is too slow
and inefficient to play an important role. Alternatively, they assumed a bimolecular initia-
tion reaction in which the hydroxyl radical produced from the peroxide homolytic thermal
decomposition, abstracts a hydrogen atom from an assisting molecule, leading to the forma-
tion of water (reaction 13). They also emphasized that the water acts as a ’spin-insulator’
between the two radical products, hindering the immediate recombination to ROR. The
α-hydrogen in EB molecule, making reaction (13) energetically more favorable.
·
·
ROOH + R−H −−→ RO + H O + R
(13)
2
·
·
The RO and R radicals will subsequently react with the substrate or with O , respec-
2
·
tively, producing peroxy radical (reactions 10). The peroxy radicals (ROO ) can propagate
the radical chain (reaction 11), or react together and lead to the radical termination (reac-
tion 12). As mentioned by Herman et al.45 the lifetime of the peroxy radicals ROO (≈ 1-10
s) is significantly shorter than the changing time of peroxy radicals to other radicals (≈ 1-30
·
18
ACS Paragon Plus Environment

Page 19 of 36
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
min.). Therefore, a quasi-steady state of peroxy radical concentration will be established.
By increasing the PEHP concentration during the oxidation reaction, its thermal decompo-
sition is increased by reaction (7) and according to the reaction (10), the concentrations of
peroxy radicals, increases and the possibility of a termination reaction (12) increases. For
this reason, the concentrations of the side products, ACP and 1-PE will increase over time.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
◦
In order to prevent this behavior, the temperature program was selected from 155 to 140 C,
and the PEHP concentration was kept below 10% by weight.
·
In a stationary state, the initiation rate (of the R radicals) must be equal to the termi-
·
nation rate (of the ROO radicals):
.
2
r = k [ROO ]
(14)
7
12
.
.
The reaction rate: (Rate = −d[O ]/dt = −d[RH]/dt = k [R ][O ] = k [ROO ][RH])
2
10
2
11
is:
√
k .[RH] r
7
1
1
Rate =
√
(15)
k12
The hydroperoxide might be decomposed according to a first or even a second order reaction.
·
·
Since there are different radicals (RO and HO ), the reaction rate constant for the initiation
is k + k . However, the overall equation of the initiation velocity can be written as:
8
9
r = k [ROOH]
(16)
7
7
The equation for the reaction rate is:
√
p
k . k .[RH]. [ROOH]
11
7
Rate =
√
(17)
k1
2
·
The addition reaction of alkyl free radical (R ) to oxygen results in the synthesis of peroxy
free radical ROO (reaction 10). Based on the previous similar work,46 it is believed that
this reaction step is instantaneous because the conditions for high Oxygen Partial Pressure
·
19
ACS Paragon Plus Environment

Organic Process Research & Development
Page 20 of 36
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
OPP) are achieved. Anyway, in the present work, as previously assumed by Shang et al.,47
(
this step of the reaction was recognized as the rate-determining step.
As concluded by Gutmann and co-workers,16 the aerobic oxidation of EB to produce
ACP through the intermediate of PEHP is generally accomplished by the decomposition
of PEHP into 1-PE as an intermediate. The 1-PE is subsequently oxidized to ACP, and
then the ACP is gradually more oxidized and converted to benzoic acid. This pathway is
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
44
verified by Romeo and co-workers, using hydrogen peroxide as an initiator for air oxidation
of 1-PE to ACP with good yield. In order to avoid the subsequent oxidation of the target
hydroperoxide products, the oxidation of hydrocarbons is usually carried out in less than
1
0% conversion. In any case, at high OPP for the oxidation reaction of EB, the subsequent
oxidation of the ACP to benzaldehyde and then to benzoic acid is performed.
The concentration of hydroperoxides is also affected by the presence of acids and water.
Indeed, the decomposition of hydroperoxides in the presence of acids is accelerated, for
example, the decomposition of cumyl hydroperoxide with sulfuric acid in phenol production
5
process, can be expressed. Water has a negative effect on PEHP selectivity and product
quality. The reaction rates of EB oxidation to 1-PE, as well as the oxidation of 1-PE to
ACP, is the same and are faster compared to the subsequent oxidation of the ACP to benzoic
acid. As a result, the amount of benzoic acid produced during the EB oxidation reaction
was relatively small. In order to optimize the reaction parameters, the reaction conditions
including reaction temperature and pressure, gas flow rate, reaction time, and OVF in outlet
gas were investigated on the reaction performance.
Reaction time effect
In EB autoxidation reaction, a brief induction time was observed. Usually, if a small amount
of PEHP (0.25 to 1% by weight) is added from the beginning to the reaction mixture, this
step can be bypassed.37 In this induction time, the original radicals are formed by thermal
decomposition of the PEHP present in the reaction feed.
20
ACS Paragon Plus Environment

Page 21 of 36
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
The effect of the reaction time on the conversion of EB at various temperatures is shown in
Fig. 3. As can be seen in the figure, the EB conversion increased over time. The experiments
◦
showed that the initial reaction rate at temperatures below 140 C was much slower than the
other ones, for which an induction time of about 30 minutes was observed. After 30 minutes,
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
◦
the EB conversion rapidly increased over time. At the temperature of 155 C, the induction
time of the oxidation reaction was not apparent.
3
3
2
2
5
0
5
0
1
1
1
1
55 ⁰C
48 ⁰C
40 ⁰C
20 ⁰C
15
1
0
5
0
0
2
4
6
8
10
Time (hr.)
Figure 3: EB conversion at various temperatures for EB autoxidation reaction. OVF in outlet gas= 0.1%,
◦
gas flow rate= 1700 mL/min, pressure=1.2 atm., Temperature=148 C.
In optimum reaction conditions, the effect of reaction time on products selectivity is
shown in Fig. 4. As seen in the figure, after four hours from the induction time, the
selectivity of PEHP decreased while selectivity for ACP, 1-PE, and other side products
increased. However, with the accumulation of PEHP, its decomposition accelerated and
reduced the selectivity of PEHP. These results indicated that four hours was a good time to
achieve the 91% selectivity of PEHP. At higher temperatures, the PEHP accumulation took
place relatively less, because the thermal decomposition of PEHP was dominant and more
side products were produced.
Fig. 5 shows the variations in PEHP concentration (wt.%) relative to the reaction time
(hr.) at different temperatures. It can be seen that, after five to six hours, the concen-
◦
◦
◦
◦
tration of PEHP begins to decrease at 148 C and 155 C, while at 120 C and 140 C, the
concentration of PEHP increases linearly for 10 hours.
21
ACS Paragon Plus Environment

Organic Process Research & Development
Page 22 of 36
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
00
80
60
40
20
0
PEHP
ACP
PE
other
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
0
2
4
6
8
10
Time (hr.)
◦
Figure 4: Selectivity of EB autoxidation products at 148 C. OVF in outletl gas= 0.1%, gas flow rate=
700 mL/min, pressure=1.2 atm.
1
1
4
2
155 ⁰C
148 ⁰C
40 ⁰C
120 ⁰C
1
1
10
8
6
4
2
0
0
2
4
6
8
10
Time (hr.)
Figure 5: PEHP concentration at various temperatures for EB autoxidation reaction. OVF in outlet gas=
.1%, gas flow rate= 1700 mL/min.
0
As illustrated in Fig. 6, there is a relationship between the conversion of EB and the
selectivity of PEHP. In fact, when the EB conversion increased by about 12%, the selectivity
of PEHP was dramatically reduced while selectivity for ACP, 1-PE and other by-products
increased. So we kept the concentration of PEHP below 12% by weight. It can be concluded
that by increasing the conversion of EB, the selectivity of PEHP increases. However, prolong-
ing the reaction time up to 10 hours led to a substantial decline in the selectivity of PEHP
and an increase in the selectivity of other by-products. It means that the reaction products
are formed as a result of a competition between PEHP production and decomposition.
Indeed, over time, the concentration of PEHP increased from 0.35% to about 10% by
22
ACS Paragon Plus Environment

Page 23 of 36
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
1
00
80
60
40
PEHP
ACP
PE
other
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
20
0
0
5
10
15
20
Conversion(%)
◦
Figure 6: Product selectivity versus conversion for EB autoxidation reaction at 148 C. OVF in outlet
gas= 0.1%, gas flow rate= 1700 mL/min.
weight, and after about 5 hours, the thermal decomposition of PEHP increased faster than its
production rate, and the concentration of PEHP began to decrease. Under optimal reaction
conditions, EB conversion was 11.1% and selectivity of PEHP was 95.02%. In the industrial
unit, the EB conversion is controlled in less than 13% to ensure that the PEHP selectivity
is greater than 90%.4
Temperature effect
In order to evaluate the effect of temperature on EB conversion and selectivity to PEHP for
◦
EB autoxidation reaction, four temperatures of 120, 140, 148, and 155 C were selected. The
results are shown in Figures 3, 5 and 7. The pressure was adjusted so that, at the selected
temperature, the water and light acids produced during the reaction entered the vapor phase
and evolved from the reaction medium through the gas stream.
◦
By increasing the reaction temperature from 120 to 155 C, the EB conversion was in-
creased significantly (Fig. 3) and the selectivity of PEHP was decreased (Fig. 7). Based on
the data presented in Fig. 5, due to the higher value of the rate constant at higher temper-
atures, although the concentration of dissolved oxygen was reduced, when the temperature
was increased, the formation rate of hydroperoxide was high. The PEHP selectivity did not
improve with increasing temperature (Fig. 7) because the PEHP decomposition was accel-
23
ACS Paragon Plus Environment

Organic Process Research & Development
Page 24 of 36
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
00
80
60
40
20
0
120 ⁰C
140 ⁰C
148 ⁰C
155 ⁰C
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
0
2
4
6
8
10
Time (hr.)
Figure 7: PEHP selectivity (%) versus time for EB autoxidation reaction at various temperatures. OVF
in outlet gas= 0.1%, gas flow rate= 1700 mL/min.
erated at higher temperatures. However, this behavior for other side products (ACP, 1-PE,
and acids) was inverted, and as the temperature rises, the selectivity for ACP and 1-PE were
increased. Further increase in temperature facilitates the synthesis of other oxidation prod-
ucts (benzaldehyde, benzoic acid, and phenol). Although the stationary concentration for
PEHP is generally higher at lower temperatures, the reaction rate for industrial applications
is too low.
When the concentration of PEHP was more than 10 wt.%, it was observed that the
PEHP selectivity was decreased and the selectivity of the ACP and 1-PE was increased,
which can be due to the increase in decomposition rate relative to the rate of production of
PEHP. Thus, the reaction temperature is an essential factor in maintaining high selectivity
for PEHP and having the highest EB conversion, as well as minimal selectivity for ACP and
1
-PE.
◦
Based on the results of the experiments, the temperature range from 155 to 140 C was
◦
determined as the temperature range for the experiments. Temperatures below 140 C re-
◦
duced EB conversion, while temperatures above 155 C caused further oxidation to benzoic
acid. After extensive experiments, the most effective temperature program for the EB au-
◦
◦
toxidation reaction was 30 minutes at 155 C and then reduced to 140 C for 240 minutes and
kept at this temperature until the end of the reaction time. The reflux condenser tempera-
24
ACS Paragon Plus Environment

Page 25 of 36
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
ture was so adjusted that the water and light acids produced during the oxidation reaction
could be steadily released from the reactor, as soon as they were produced. This temperature
◦
range was from 80 to 120 C.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Pressure effect
Almost all previous authors believe that the pressure applied to the EB autoxidation reaction
is to maintain a reaction in the liquid phase. In any case, we found that the pressure should
be so adjusted that at the selected reaction temperature, the water and light acids produced
during the reaction can evaporate and leave the reaction mixture together with the gas
stream.
Korthals et al. believed that water and ACP are essentially produced by PEHP decom-
position under reaction conditions.36 Water and acids restrict the oxidation reaction and
reduce the PEHP production rate and product quality. Therefore, in the present work, the
water generated was condensed in a cold trap and acids produced from the ACP oxidation
reaction was controlled by controlling the OVF in the outlet gas. These acids were neutral-
ized in a separate zone containing dilute sodium hydroxide solution. At high temperatures,
the concentration of dissolved oxygen decreases and therefore, a higher pressure is required
to have at least the oxygen concentration in the reaction mixture. At a higher temperature
than the boiling point of EB and below the boiling point of water, at a selective pressure,
water and light acids, along with a small amount of EB, entered the gas phase and left the
reactor.
The vapors and gases released from the reactor were cooled down in a condenser at a
◦
temperature of −5 C and were treated with an alkaline solution and de-acidified. After
separating the phases, the neutralized EB was transferred to the reactor to continue the
oxidation reaction. By examining the results of the experiments, we found that, according
to the temperature schedule, the pressure should be reduced from 3.0 to 1.2 atmospheres.
The selected pressure program, used in this work, included 30 minutes at 3 atm. and then
25
ACS Paragon Plus Environment

Organic Process Research & Development
Page 26 of 36
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
reduced to 1.2 atm. for 2 hours and then remained at this pressure until the end of the
reaction.
Oxidant concentration and gas flow rate
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Five semi-continuous experiments were carried out to investigate the effect of OVF in the
exhaust gas on the conversion of EB and selectivity of hydroperoxide. By changing the ratio
of air to nitrogen, the percentage of OVF in the outlet gas was adjusted for five different
OVF of (0.1, 0.3, 0.5, 3.0, and 5.0)%. The results are listed in Table 3.
a
Table 3: The effect of OVF in the exhaust gas on EB autoxidation reaction
OVF Conversion
(%)
Selectivity (%)
ACP
∗
PEHP
PE
Others
0
0
0
.1
.3
.5
11.05
13.85
17.85
20.25
22.20
95.02
80.29
66.39
65.28
63.42
02.90
12.64
19.61
19.75
20.27
01.36
06.28
09.52
09.88
10.36
0.72
0.79
4.48
5.09
5.95
3.0
5
.0
a
◦
Reaction conditions: Temperature=148 C, Pressure=3 atm., Time=5 hours, gas flow rate=1700
∗
mL/min, stirrer speed=1400 rpm. Others=mainly benzoic acid.
According to Table 3, EB conversion increased with increasing the OVF in the outlet gas
while the PEHP selectivity was reduced. However, the selectivity of ACP, 1-PE, and other
acidic side products have increased with the increase of EB conversion. This might be due
to the thermal decomposition of PEHP to water and ACP, and to the oxidation of the ACP
to benzaldehyde and then to benzoic acid. By increasing the gas flow rate of oxygen, the EB
conversion slightly increased, which was neglected for the desired reaction. Also, the light
acids and water produced during the reaction were removed from the oxidation reactor by
increasing the gas flow.
In most of the previous work, a simplified kinetics for the oxidation of EB in the liquid
phase was predicted based on the free radical mechanism,11,47,48 in which all the elementary
reaction steps were assumed to be of the first-order with respect to liquid reactants and zero-
26
ACS Paragon Plus Environment

Page 27 of 36
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
order relative to gaseous reactants i.e. oxygen. In none of this kinetics explicitly have referred
to the effect of the gaseous reactant. In fact, just as it was tested in the laboratory, only
experiments performed at low oxygen concentrations showed such a relationship. However,
as some previous authors have pointed out,49,50 the effect of oxygen on the autoxidation of
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
hydrocarbons in the liquid phase can only be ruled out if the oxygen partial pressure (OPP)
is greater than a minimum value. This minimum value is between 0.066 and 0.132 atm.51
Our experimental observations confirmed that the oxidation reaction rate of EB was clearly
reduced when the OVF in the outlet gas was less than 0.30%. This suggested that, when the
OVF was reduced from a certain value, the oxidation of EB was no longer zero-order with
respect to a gaseous reactant (oxygen).
In this work, in agreement with Sun and co-workers,47 for p-Xylene oxidation, we found
that there is a threshold for the OVF in the exhaust gas, which is between 0.10 and 0.30%.
When the OVF was less than 0.30%, it had a substantial effect on the oxidation kinetics
of EB. As long as the OVF was greater than 0.30%, the oxidation kinetics of EB was
almost independent of the OVF. Under the operating conditions of the industrial process
of oxidation of EB, the OVF in the reactor falls within this range. Therefore, the amount
of EB converted is equal to the amount of oxygen absorbed and only at very low OPP the
oxidation rate decreased. On the other hand, additional oxygen resulted in the formation of
more acidic by-products. For this reason, under optimized reaction conditions, the OVF in
the outlet gases was set to less than 0.3%. The air used as a feed gas was diluted with pure
N and gas from the oxygen analyzer to provide the required oxygen content in the feed.
2
The EB vapors, after neutralization in a separate zone, were condensed and returned
to the oxidation reactor;. The selectivity of PEHP was reduced by increasing the OVF,
which could be due to the formation of more by-products resulted from further oxidation of
EB with additional oxygen. Excellent selectivity for PEHP (95%) along with moderate EB
conversion (11%) was observed with OVF less than 0.1% in the outlet reactor gas. Increasing
the OVF means increasing the concentration of oxygen in the gas feed and also increasing
27
ACS Paragon Plus Environment

Organic Process Research & Development
Page 28 of 36
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
the PEHP yield. At high concentrations, and particularly for oxidation with OVF=5.0%,
PEHP tended to decompose. In fact, the reduction of PEHP yield at higher concentrations
has previously described.52
Increasing the concentration of oxygen in the gaseous feed was accompanied by an increase
in the conversion of EB, which was higher in high concentrations of oxygen. In parallel with
EB conversion, the formation of side products (ACP, 1-PE, and acids) was also observed.
ACP was the second major product with a higher concentration than the 1-PE. ACP is
formed directly from the termination reaction step,18 or from the oxidation of 1-PE, or
through the PEHP decomposition.36 Apparently, the selectivity toward PEHP was lowest
for the highest concentration of oxygen at the end of the test (5 hours).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figs. 8, 9 and 10 show the effect of the flow rate of the exhaust gas on the EB conversion
and the selectivity of PEHP, ACP, 1-PE, and other by-products.
14
13.5
13
1
2.5
12
0
300
600
900
1200 1500 1800
Gas flow rate (mL/min)
◦
Figure 8: The effect of gas flow rate on EB conversion (%) at 148 C, during 5 hours, and OVF in outlet
gas=0.1%.
As the gas flow rate rises from 200 to 1700 mL/min, the EB conversion increased slightly
(Fig. 8), while PEHP selectivity (Fig. 9) and the selectivity of ACP and 1-PE (Fig. 10)
remain at approximately the same level. Also, as shown in Fig. 10, by increasing the gas
flow rate, the selectivity for other acidic products is reduced. However, a further increase
in gas flow rate led to a nearly identical PEHP selectivity. Hence, the gas flow rate of 1700
mL/min was selected.
28
ACS Paragon Plus Environment

Page 29 of 36
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
1
00
80
60
4
0
0
0
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
0
300
600
900 1200 1500 1800
Gas flow rate (mL/min)
◦
Figure 9: The effect of gas flow rate on PEHP selectivity (%) at 148 C, during 5 hours, and OVF in outlet
gas=0.1%.
6
5
4
3
ACP
PE
Others
2
1
0
0
300 600 900 1200 1500 1800
Gas flow rate (mL/min)
◦
Figure 10: The effect of gas flow rate on ACP, 1-PE, and other by-products selectivity (%) at 148 C,
during 5 hours, and OVF in outlet gas=0.1%, others=mainly benzoic acid.
When air and nitrogen were fed into the reactor in such a way that the OVF was 0.1%, the
PEHP selectivity was 95%, and when the OVF in outlet gas was 5.0%, the PEHP selectivity
decreased to 63%. Nevertheless, careful scrutiny of the PEHP selectivity versus EB con-
version revealed that, by increasing the EB conversion, the PEHP selectivity was decreased
and at the same time selectivity to ACP, 1-PE, and other by-products were increased. The
differences observed in the five concentrations of oxygen studied were the result of differences
in different levels of EB conversions. This kind of behavior is inherent in the autoxidation
mechanism of hydrocarbons in liquid phase.53
29
ACS Paragon Plus Environment