Decene formation in ethylene trimerization reaction catalyzed by Cr-pyrrole system

Article abstract of DOI:10.1016/j.apcata.2014.01.051

Decene formation in the ethylene trimerization reaction was studied using a chromium(III) 2-ethylhexanoate/2,5-dimethylpyrrole/triethylaluminum/ diethylaluminum chloride catalyst system. Kinetic investigations revealed that some decene formation reactions did not depend on 1-hexene concentration, because 1-hexene and catalyst may react with ethylene before dissociation of 1-hexene-catalyst complex after 1-hexene formation. The results demonstrated that decene formation is an intrinsic part of the trimerization reaction mechanism. It was also shown that a stepwise elimination mechanism for the decomposition of the chromacycloheptane intermediate cannot explain the observed product distribution. The dependencies found allow selection of appropriate conditions for low or high decene formation in the ethylene trimerization reaction.

Full text of DOI:10.1016/j.apcata.2014.01.051

Accepted Manuscript
Title: Decene formation in ethylene trimerization reaction
catalyzed by Cr-pyrrole system
Author: Timur M. Zilbershtein Vladislav A. Kardash
Vladlena V. Suvorova Anatoly K. Golovko
PII:
S0926-860X(14)00057-X
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.apcata.2014.01.051
APCATA 14689
To appear in:
Applied Catalysis A: General
Received date:
Revised date:
Accepted date:
30-9-2013
22-1-2014
29-1-2014
Please cite this article as:
Decene formation in ethylene trimerization
reaction catalyzed by Cr-pyrrole system, Applied Catalysis A, General (2014),
http://dx.doi.org/10.1016/j.apcata.2014.01.051
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Decene formation in ethylene trimerization reaction catalyzed by Cr-
pyrrole system
Timur M. Zilbershteina,b *, Vladislav A. Kardasha, Vladlena V. Suvorovaa,
Anatoly K. Golovkob
a NIOST LLC, Kuzovlevsky trakt 2 bld. 270, 634067, Tomsk, Russia, ztm@niost.ru
b Institute of Petroleum Chemistry, Siberian Branch of Russian Academy of Sciences,
Akademichesky. Ave. 3, 634021, Tomsk, Russia, golovko@ipc.tsc.ru
Abstract
Decene formation in the ethylene trimerization reaction was studied using a chromium(III) 2-ethylhexanoate/2,5-
dimethylpyrrole/triethylaluminum/diethylaluminum chloride catalyst system. Kinetic investigations revealed that some
decene formation reactions did not depend on 1-hexene concentration, because 1-hexene and catalyst may react with
ethylene before dissociation of 1-hexene – catalyst complex after 1-hexene formation. The results demonstrated that
decene formation is an intrinsic part of the trimerization reaction mechanism. It was also shown that a stepwise
elimination mechanism for the decomposition of the chromacycloheptane intermediate cannot explain the observed
product distribution. The dependencies found allow selection of appropriate conditions for low or high decene formation
in the ethylene trimerization reaction.
Keywords:Ethylene trimerization, Cr-pyrrole catalyst, metallacyclic mechanism, co-trimerization
1. Introduction
The ethylene trimerization reaction predominantly produces 1-hexene, which is chiefly used
as a comonomer in polyethylene production. The reaction, which has been applied industrially by
Chevron Phillips since 2004, is catalyzed by transition metal complexes which are typically
chromium-based and activated by organoaluminum compounds. Comprehensive reviews on
trimerization catalysts and the mechanism of the reaction have been published [1-3].
It is generally accepted that the reaction mechanism includes the formation of metallacyclic
intermediates. The basic mechanism proposed by Briggs [4] (Scheme 1) was later supported by
experimental [5,6] and theoretical [7,8] findings. It involves the coordination of two ethylene
*
Corresponding author at: NIOST LLC, Kuzovlevsky trakt 2 bld. 270, 634067, Tomsk, Russia,
Tel: +7(3822) 70-2222, Fax: +7 (3822) 70-2332, e-mail: ztm@niost.ru (T.M. Zilbershtein)
1
Page 1 of 21

molecules to the catalyst center, followed by metallacyclopentane formation. A metallacycloheptane
is then formed by expansion of the ring with a third ethylene molecule. The metallacycloheptane
presumably undergoes β-elimination and reductive elimination, which results in 1-hexene formation
and regeneration of the catalyst. An alternative explanation suggests a concerted 3,7-hydrogen shift
in the metallacycloheptane leading to the same product.
By-products of the reaction, usually about 10% in total, include 1-butene, 1-octene, internal
hexenes, decenes, and higher olefins. 1-Butene may be formed due to non-selective oligomerization,
or from the metallacyclopentane. 1-Octene is believed to be formed due to further metallacycle
expansion to metallacyclononane, which also undergoes reactions analogous to the
metallacycloheptane. Decene isomers are formed in a co-trimerization reaction of 1-hexene and two
ethylene molecules [9,10]. Decenes are often the most significant by-products in the reaction,
regardless of the catalyst used. [1,9–11]. Heavier olefins can be formed through other co-
trimerization reactions [10].
Cr
2 C2H4
Cr
C2H4
"Cr"
Cr
Cr
Scheme 1. Ethylene trimerization mechanism.
Most kinetics studies of the ethylene trimerization reaction have considered the formation of
the main product, 1-hexene, or studied the total ethylene consumption or the catalyst productivity
[11–16]. Details of the decene formation kinetics in the ethylene trimerization reaction remain
unknown. We have not found studies devoted to the topic, although there are papers that consider the
formation of co-trimerization products using various conditions and catalysts [6,10,17]. This work
was intended to investigate the key features of decene formation kinetics in the ethylene
2
Page 2 of 21

trimerization reaction, as it affects the 1-hexene selectivity in the reaction. These data could also help
to increase the understanding of the reaction mechanism.
2. Experimental
2.1. Materials
Chromium(III) 2-ethylhexanoate (Cr(EH)3) was prepared by a known method from aqueous
chromium(III) chloride and sodium 2-ethylhexanoate [18], but the final drying with ethylhexanoic
acid was carried out at 200° C and 200 Pa, in order to maximize the associated water removal. A
25% solution of triethylaluminum (TEA) in toluene and 2,5-dimethylpyrrole (DMP) were purchased
from Aldrich. 1-Hexene, 1-octene and 1-decene for standards and diethylaluminum chloride (DEAC)
(1,0 M solution in hexanes) were purchased from Acros Organics. These reagents were used without
further purification. Hydrocarbon solvents were refluxed over Na and NaH, and freshly distilled
before use. The polymer-grade ethylene was further purified by passing it through a steel bottle filled
with activated molecular sieves 5Å.
2.2. Instruments
A stainless steel jacketed reactor (0.5 l) equipped with an overhead stirrer, thermostat,
pressure and temperature sensors, nitrogen line, hydrogen line, bottom needle valve, and ethylene
dosing line with a flow meter was used for the study. A computer-based control system was used to
control the stirrer speed, reactor and jacket temperatures, flow rate of ethylene, and other parameters.
Reaction products were analyzed using gas chromatography (Agilent 7890A with FID-
detector, Agilent 7890A with 5975C mass-selective detector) with HP-5 capillary column. NMR
spectra were recorded in CDCl3 using Bruker Avance III 400 (400 MHz) spectrometer.
2.3. Procedures
2.3.1. Catalyst preparation
The catalyst was prepared according to a previously described enhanced method [19] by using
Cr(EH)3/DMP/TEA/DEAC in the ratio 1 : 5 : 36 : 14. Cr(EH)3 (29 mg, 60 μmol) and DMP (29 mg,
305 μmol) were placed in a flask. Then 5 ml of ethylbenzene was added. The mixture of 1.15 ml of
3
Page 3 of 21

1.9 M TEA solution in toluene (2.2 mmol) and 0.84 ml of 1 M heptane solution of DEAC (0.84
mmol) was microwave-irradiated in a polypropylene syringe using CEM MARS 5 microwave oven
at rated power 400W for 6 minutes (with little heating). Then, within 30 seconds after the end of the
irradiation, the TEA/DEAC solution was added to the mixture of Cr(EH)3 and DMP in ethylbenzene.
After 15 min since TEA/DEAC addition, solvents were removed in vacuo at 40 °C. The residue in
the flask was diluted with 20 ml of cyclohexane to obtain the catalyst solution (0.003 М Cr).
2.3.2. Ethylene trimerization
The reactor was dried at 120 °С under nitrogen flow, and then evacuated and filled with
hydrogen. Cyclohexane (200 ml) was added with a dosing pump under a hydrogen atmosphere. The
solvent was saturated with hydrogen at atmospheric pressure and the desired reaction temperature.
The catalyst solution in cyclohexane (2 ml, 6 μmol Cr) was injected into the reactor via a syringe
under hydrogen counter flow. Ethylene was swiftly added to build up the desired pressure, and then
ethylene was dosed to maintain the pressure constant. Reactor temperature was maintained at the set
value by an automated system. Sampling was performed from the bottom valve during the reaction.
The pressure was reduced and the reaction mixture was unloaded after 60 min.
During each run, seven samples were collected for GC analysis at 2, 5, 10, 15, 30, 45 and 60
min after ethylene addition (sample volume 1-1.5 ml).
2.3.3. Distillation of decene fraction
The combined reaction mixtures were exposed to air for 2 h, which resulted in precipitation
of small amount of brown solids. These were filtered off, and the filtrate was distilled using a
laboratory distillation column to obtain a bottom residue containing about 50% cyclohexane and
hexenes, 40% decenes, and 10% heavy products. The residue was additionally distilled using a
laboratory column in vacuo, and a fraction with the boiling point 108–112°C/18 kPa was collected.
GC/MS analysis confirmed a content of 94% C10 olefins. The sample also contained about 2% of
cyclohexane, 1.8% of ethylbenzene and 1.3% of 1-octene.
3. Results and discussion
3.1. The scope of the experiments
4
Page 4 of 21

We performed a series of trimerization runs, varying the ethylene pressure and reaction
temperature. Our measurements allowed us to determine the average reaction rate between sampling
points for every product, including the decene isomers. The average reaction rates for 1-hexene and
the decenes at 15–30 min intervals were calculated from GC data for all runs. The results are listed in
Table 1.
Table 1
Reaction rates at 15–30 min intervals.a
Average formation rate
P(C2H4), Activity at 30 min,
MPa
over 15–30 min interval,
TOF, mol·(mol Cr·s)-1
Run T, °C
kg·(g Cr·h)-1
b
1-C6
10.9
15.8
C10
0.32
0.32
0.42
0.51
0.93
0.76
1.35
0.13
0.33
0.50
1.18
1.11
1.00
1
2
25
40
0.8
0.8
72.0
132.7
3
4
5
6
7
8
9
10
11
12
13
50
50
50
50
50
65
65
65
65
65
65
0.4
0.8
1.2
1.5
2.0
0.4
0.6
0.8
1.0
1.2
1.45
79.4
131.8
170.9
216.2
277.7
25.7
56.2
84.4
166.7
156.0
156.0
11.1
20.0
25.9
33.0
40.1
3.73
8.65
16.2
28.2
25.9
25.3
a General conditions: 6 μmol Cr (0.03 mmol/L), cyclohexane,
Cr(EH)3:DMP:TEA:DEAC 1:5:36:14.
b Total decenes formation rate
The typical time course of product formation during the ethylene trimerization reaction is shown in
Fig. 1.
5
Page 5 of 21

m, g
0,8
0,7
0,6
0,5
0,4
0,3
0,2
0,1
0
Butenes
1-Hexene·0.01
Hexenes
1-Octene
Decenes·0.1
Heavy olefins·0.1
t, min.
0
20
40
60
Fig. 1. Product formation versus time during the ethylene trimerization reaction, run 6. In order to
present a concise diagram, 1-hexene and decene/heavy olefin amounts were scaled down by a factor
of 100 and 10, respectively.
3.2. Identification of the decenes
Scheme 2 illustrates pathways for decene formation in accordance with the metallacycle
mechanism. Seven main decenes were formed, including 1-decene (6), 4-decene (7) and 5-
decene (5). GC analysis could not distinguish 5 and 7, so only six major decene peaks were discerned
by GC/MS.
6
Page 6 of 21

1
Bu
Bu
2
Cr
Cr
Route I
Cr
Bu
Bu
3
4
5
Bu
Bu
Bu
Bu
Route II
Cr
"Cr"
Cr
Cr
Cr
Cr
Bu
Route III
Cr
Bu
Bu
Bu
Bu
Bu
6
7
Scheme 2. Mechanism of decenes formation.
The decenes were distilled from the reaction products for identification purposes. The
fraction contained about 94% decenes, according to GC/MS analysis. The mixture also contained
small amounts of 1-octene (1.3%) and ethylbenzene (1.8%) which had been used in the catalyst
preparation. The remainder was mostly cyclohexane solvent. Decenes were identified by 13C NMR
analysis of the decenes mixture. The Distortionless Enhancement by Polarization Transfer (DEPT)
method was used to establish the signal positions of the vinyl CH2 and CH carbons, as well as the
aliphatic CH (branching points). 5-Methyl-1-nonene (1) and cis-4-decene (7) were synthesized by
Bercaw et al [10], and 13C NMR data were given. It was found that cis-4-decene (7) was the major
isomer among linear olefins 5, 6 and 7, and only small amount of 5 was found (Fig. 2). The trans-
7
Page 7 of 21

isomer of 7 was not detected. 1-Decene (6) and 5-methylene-nonane (2) were also identified using
the NIST mass spectral database. 1-Decene was additionally confirmed by comparison of the
retention time with a standard solution. The vinyl signals of 1-octene were indistinguishable from the
corresponding 1-decene signals, which was responsible for about 50% increase in the signal intensity
of the vinyl carbons of 6. The NMR-based identification was in agreement with the GC/MS data.
The results are given in the Table 2. All expected isomers were found; there was no indication of
other isomers in the NMR spectra. Only traces of other C10 olefin isomers were detected by GC/MS.
Table 2
Identification of the decenes using 13C NMR and GC/MS data.
Content
Decene in C10, %
(GC)
Signal 1
δ, ppm rel. int.
Signal 2
δ, ppm
Signal 3
Group δ, ppm rel. int.
Group
CH2=CH-
Group
CH2=CH-
rel. int.
3.8
1
2
3
4
33
11b
23
114.1a
5.1
1.4
3.1
2.7
139.5a
150.2c
143.8
137.8
32.5a
5.7
CH2CH(CH3)CH2
-
108.6c
114.0
115.6
0.6
CH2=C(C4H9)2
CH2=CH-CH
CH2=CH-CH2CH
CH2=C(C4H9)2
CH2=CH-CH
CH2=CHCH2CH
2.7
44.2
39.0
2.1
2.8
CH2=CHCH
CH2=CHCH2CH
18
1.8
5
6
-
130.4d
-
-
-
-
CH2CH=CHCH2
CH2=CH-
3b
114.2d
1.0
139.3d
130.0a
130.5a
0.5
1.1
1.1
CH2=CH-
CH2CH=CHCH2
7
12e
-
a
Confirmed by comparison with data for the pure compound in [10]
b Confirmed by comparison with NIST mass spectra database
c
Confirmed by comparison with data from [20]
d Confirmed by comparison with SDBS NMR database
e Together with 5; less than 1% of 5 in the mixture according to NMR data
8
Page 8 of 21

Fig. 2. Fragment of the 13C NMR spectra of the decenes mixture.
3.3. 1-Hexene concentration dependence of decene formation rate
The GC data were obtained for every sample taken during the reaction, so content of every
decene isomer was measured. The average formation rate between sampling in the form of TOF for
every isomer was calculated as a change in the decene isomer molar content per molar amount of the
catalyst and the time interval. It was previously shown that addition of 1-hexene to the reaction
mixture results in elevated amount of decenes formed [9]. It was also demonstrated by Do and
coworkers that adding 1-heptene leads to olefins C11 [10]. As decenes are formed from 1-hexene, it
was assumed that the decene formation rate would be dependent on 1-hexene concentration. Indeed,
during the reaction, it was found that the decene formation rate increased proportionally to the 1-
hexene concentration, calculated as an average value for the same time intervals. Good linearity was
observed in most cases after the initial period (5–10 min of the reaction). Examples are shown in Fig.
3. However, at low 1-hexene concentrations, the decene formation rate did not stay close to zero, but
rather remained near a constant value. This was true for all decenes 1–7 under different reaction
conditions. The constant component of the formation rate for each decene could be determined using
a trend line for the dependence of the formation rate on 1-hexene concentration. The trend lines
intercepted the vertical axis above zero.
9
Page 9 of 21

We can conclude that there are two groups of reactions that lead to decene formation: one that
is first-order dependent on 1-hexene concentration, and a second group that is independent of 1-
hexene concentration.
The decenes formation rate dependence values shown in Fig. 3 allowed the determination of
conditional constants for the hexene-dependent and hexene-independent formation reactions of every
isomer. The hexene-dependent conditional constants were determined from the slopes of the trend
lines. The hexene-independent constants were established as the vertical axis intercepts by the trend
lines.
The following equations were used for hexene-dependent (1) and hexene-independent
reactions (2):
d[C10]/dt = k′[1-C6][Cr]
d[C10]/dt = k″[Cr]
(1)
(2)
where [C10] is the concentration of decenes 1–7 (mol L-1); [Cr] is the concentration of
chromium (mol L-1); and [1-C6] is the 1-hexene concentration, (mol L-1). The conditional constants k′
(L mol-1 s-1) and k″ (s-1) include the ethylene pressure, which was kept constant during the runs.
a
c
b
TOF, s-1
0,15
TOF, s-1
0,50
TOF, s-1
0,40
1
1
1
0,45
0,40
0,35
0,30
0,25
0,20
0,15
0,10
0,05
0,00
0,35
0,30
0,25
0,20
0,15
0,10
0,05
0,00
2
2
2
0,10
0,05
0,00
3
3
3
4
4
4
5+7
6
5+7
6
5+7
6
0,0
1,0
2,0
0,0
0,5
1,0
1,5
0,0
0,5
1,0
1-hexene, mol L-1
1-hexene, mol L-1
1-hexene, mol L-1
Fig. 3. Decenes 1–7 formation rates under various conditions: a) run 1, 25 °C, 0.8 MPa; b) run 6,
50 °C, 1.5 MPa; c) run 10, 65 °C, 0.8 MPa.
3.4. Ethylene pressure dependence of decenes formation rates
The conditional constants k′ and k″ were found to be dependent on the ethylene pressure.
Plots of the ethylene pressure dependence at 50 and 65°C are shown in Figs. 4 and 5.
10
Page 10 of 21

Fig. 4. Dependence of the conditional constants for decenes formation on ethylene pressure at 50°C:
a) hexene-dependent reactions and b) hexene-independent reactions.
Fig. 5. Dependence of the conditional constants of decenes formation on ethylene pressure at 65°C:
a) hexene-dependent reactions and b) hexene-independent reactions.
At 50°C, decenes formation in hexene-dependent reactions was negatively dependent on the
ethylene pressure, which indicated inhibition by ethylene (Fig. 4, a). At 65°C, a similar dependence
was observed (Fig.5, a) for ethylene pressure above 1.0 MPa (Table 1, entries 11–13), as 1-hexene
formation occurred at almost the same rate. That is in accordance with the observations that 1-butene
addition at the initial stage of the reaction results not only in the increased octenes formation due to
1-butene co-trimerization, but also causes decreased decenes formation [9]. At lower pressure, the 1-
hexene formation rate showed approximately second-order dependence on ethylene pressure (Table
1, runs 8-11), and the hexene-dependent decene formation rate revealed nearly first-order
dependence with respect to ethylene. We believe that the difference between Fig. 5, a and Fig. 4, a is
11
Page 11 of 21

caused by different kinetics of trimerization reaction at 50 °C and 65 °C, because at 50 °C 1-hexene
formation is nearly linear-dependent on ethylene pressure (Table 1, runs 3-7). The detailed
investigation of the 1-hexene and 1-octene formation at different temperatures is under way.
Hexene-independent reactions were approximately first-order dependent with respect to
ethylene at both studied temperatures (Fig. 4, b and Fig. 5, b). This indicated the absence of
concurrent inhibition by ethylene for this group of reactions.
3.5. Mechanism of chromacycloheptane decomposition
Generally, the decene ratios under various reaction conditions for the hexene-dependent and
the hexene-independent reactions were similar (Table 3). All six decenes were present in both cases,
which meant that at least two of the routes I, II, and III (Scheme 2) were implemented in both groups
of reactions.
Table 3
Decenes 1–7 formation ratios in the hexene-dependent and -independent reactions.
Decene
Share of total decenes,
Share of total decenes,
hexene-dependent reactions, %
hexene-independent reactions, %
1
2
38.2–57.8
8.2–12.5
12.7–20.4
12.3–16.9
4.6–11.4
3.4–7.2
23.6–34.3
9.3–12.9
24.5–30.2
17.8–23.5
9.0–11.7
0.6–2.3
3
4
5+7
6
The formation of 1 and 2 was highly preferred for the diphosphinoamine (PNP) catalyst,
which resulted in 95% total selectivity for these decenes [17]. With our catalyst, compound 1 also
prevailed in both the hexene-dependent and hexene-independent reactions. However, its share did
not exceed 60% under any of the conditions used.
The route III should have little impact on the distribution of the products. Otherwise, the
subsequent insertion of ethylene would have resulted in higher share of olefins 5-7, because in the
route III they are formed after ethylene insertion from the less hindered unsubstituted side. The route
II is unlikely to be hexene-independent, as 1-hexene insertion here should compete with ethylene
insertion. Therefore, we may conclude that the route I is the major pathway, especially for the
hexene-independent reactions, as the olefins 1-4 make up more than 85% of the products in that case.
The isomers 5-7 are formed only via the route III in the case of hexene-independent reactions. The
12
Page 12 of 21

presence of route II in the hexene-dependent reactions leads to higher amount of 6, and much more
pronounced increase of isomer 1 formation. This indicates that unsubstituted chromacyclopentane is
preferably expanded by 1,2-insertion of 1-hexene rather 2,1-insertion. Both 1-hexene insertions are
much slower than the competing ethylene insertion leading to 1-hexene formation. The same
preference was observed in the case of Cr-PNP catalyst [10,17].
Two possible mechanisms for 1-hexene release from the chromacycloheptane intermediate
have been suggested: a two-step process involving β-elimination followed by reductive elimination,
and a one-step concerted 3,7-hydrogen shift [2] (Scheme 3). The computational studies [7,21]
indicate that the concerted shift is more energetically favorable for chromium trimerization catalysts.
However, no experimental evidence in favor of one of these mechanisms has been published yet.
Both of these proposed reaction pathways may be stereoselective, and therefore, the structure
of the substituted chromacycloheptane can determine the structure of the product.
H
Cr
Cr
[Cr]
H
Cr
Scheme 3. Possible mechanisms of 1-hexene formation from the chromacycloheptane.
13
Page 13 of 21

Scheme 4. Possible stepwise elimination from 3-substituted chromacycloheptane.
Conclusion: 5 ≤ 6 < 7, mix of cis-7 and trans-7; experiment: 5 < 6 < 7, only cis-7.
Our analysis of the decenes mixture allowed us identify and measure the ratio of the linear
decenes 5, 6 and 7, which contribute about 15% of the decenes composition. In the previous work
[10] only 3% of the linear decenes was found in the mixture, which made their identification and
quantification difficult. The difference should be attributed to the different catalyst systems used in
this work and by authors of [10]. We can use the known composition of these decenes in our
mechanistic investigations. Scheme 4 illustrates the formation of decenes 5, 6 and 7 according to the
stepwise mechanism of β-elimination followed by reductive elimination. Structures a and b are
possible isomers of chromacycloheptane, with yet unknown substituents on the chromium centre.
β-Elimination can be either stereo- and regioselective (E2 mechanism), or regioselective only
(E1 mechanism). In the latter case, the most substituted olefins should be formed. This was
disproven by the experimental data, as α-olefins prevailed in the C10 mixture.
14
Page 14 of 21

During β-elimination according to the E2 mechanism, Cr should be in the same plane with
the hydrogen in position 3 or 6 and carbons C(2) and C(3), or C(6) and C(7), respectively. In the case
of the exocyclic elimination, Cr should be coplanar with hydrogen from the alpha-methylene group.
For the 2-substituted chromacycloheptane (Scheme 4) β-elimination from position 6 is preferable or
at least comparable with elimination from position 3, because there is no hindrance for elimination
from position 6. Consequently, the proportion of 6 should be equal to or higher than that of 5. This
was in accordance with the experimental data, because 13C NMR showed very small signal for 5
(Fig. 2). The isomer 7 was formed due to the exocyclic hydrogen elimination (or the hydrogen
transfer). The exocyclic reaction should have less structural limitations, so the product 7 should
prevail among linear decenes. This was also in accordance with the experimental data. The exocyclic
β-elimination should lead to a mixture of cis and trans isomers of 7, with excess of the trans-olefin,
which is typical for E2 reaction. However, we found that the only product of the exocyclic reaction
was cis-4-decene. The corresponding trans isomer would have been easily distinguished in the 13C
NMR spectra, but no such signals were found.
Such a high selectivity cannot be explained using β-elimination mechanism, especially for the
structure a with pseudo-equatorial orientation of the butyl substituent. That is an experimental proof
that the stepwise elimination is unlikely to be the mechanism for substituted chromacycloheptane
decomposition. Other mechanisms, such as the concerted hydrogen shift, may better explain the
observed selectivity, which is in accordance with the computational data [7,21]. However, the
mechanism of chromacycloheptane decomposition requires further investigation.
The concerted hydrogen shift mechanism is an intramolecular reaction, unlike the stepwise
elimination. One may suppose that if the reaction indeed proceeds via concerted hydrogen shift, then
the nature of the catalyst center should affect this stage more than the chemical environment. It is not
likely that this stage can be influenced by any additives, provided they do not alter the catalyst
species.
3.6. Temperature dependence and decene formation pathways
The hexene-independent decene formation reactions, which were not inhibited by ethylene,
should have a non-limiting stage of 1-hexene coordination to the catalyst. The experimental data
demonstrated that this stage remained non-limiting even at very low 1-hexene concentration in the
reaction mixture (see 3.3, Fig. 3). It is reasonable to consider that there is no difference in the
following stages between the hexene-dependent and the hexene-independent reactions after the
15
Page 15 of 21

formation of the butyl-substituted chromacyclopentane (see 3.2, Scheme 2). Consequently, the
difference should be connected with the previous stage of 1-hexene and ethylene coordination to the
catalyst.
The coordination of the 1-hexene double bond to the chromium center was predicted as the
final stage of the catalytic cycle by the computational study of the chromium-pyrrole system [7].
Then formed 1-hexene should be then released from the catalyst, which concludes the catalytic cycle.
We suggest that there is another way for the transformation of the catalyst – 1-hexene complex,
where ethylene molecule coordinates to it before 1-hexene is released. This results in formation of
substituted chromacyclopentanes, which in turn transform into decene isomers.
Taking into account the fact that 1-hexene release from the catalyst is very fast and usually
non-limiting stage in the catalytic cycle, we can conclude that the catalyst – 1-hexene complex is
quite unstable. Its decomposition should accelerate with an increase in temperature. Therefore, the
shortened lifetime of the complex at a higher temperature will result in lower probability of ethylene
coordination leading to decenes, and the hexene-independent reactions will be suppressed.
Indeed, we found that the hexene-independent reactions slowed when the temperature was
increased (Fig. 6). In contrast, the hexene-dependent reactions accelerated as the temperature
increased (Fig. 7). We can conclude that the pathways of the hexene-independent and hexene-
dependent reactions are different. The competition between ethylene and hexene in the hexene-
dependent decene formation reactions is in accordance with substitution of one of the ethylene
molecules with 1-hexene in the metallacyclic mechanism (Schemes 1 and 2).
Fig. 6. Temperature dependence of the conditional constants for hexene-independent decenes
formation at 0.8 MPa ethylene.
16
Page 16 of 21

Fig. 7. Temperature dependence of the conditional constants for hexene-dependent decenes
formation at 0.8 MPa ethylene.
The results discussed above allow us to suggest new details of the mechanism of by-product
formation in the ethylene trimerization reaction (Scheme 5). It should be noted that hexene-
independent decene formation reactions cannot be fully avoided, because they presumably occur due
to ethylene coordination to the complex of the newly formed 1-hexene and the catalyst. Therefore,
these reactions are an intrinsic part of the ethylene trimerization mechanism.
Scheme 5. Combined mechanism for 1-hexene and decene formation. The hexene-dependent decene
formation pathways are in the upper part, and the hexene-independent decene formation pathway is
in the bottom part of the scheme.
17
Page 17 of 21

Decenes formation during the ethylene trimerization reaction is a complex process. It is
affected by the ethylene pressure, 1-hexene concentration, temperature, and other factors. Decenes
formation can be inhibited by keeping the ethylene pressure high and minimizing the 1-hexene
concentration. However, this will affect only the hexene-dependent reactions. The hexene-
independent formation of decenes accelerates with increasing ethylene pressure, but slows at higher
temperatures. In order to achieve higher 1-hexene selectivity, one should keep temperature high in
order to minimize hexene-independent by-product formation, while maintaining low 1-hexene
concentration and high ethylene pressure to suppress hexene-dependent decene formation. In
contrast, if increased decenes formation is desired, one should provide high 1-hexene concentration
and use low ethylene pressure during the reaction. In that case, the temperature should be kept high
but not excessively to avoid suppression of hexene-independent decene formation reactions.
4. Conclusions
We investigated decenes formation in the ethylene trimerization reaction using a chromium-
pyrrole catalyst. The reaction products were identified, and it was shown that all major decene
isomers were formed in accordance with the metallacycle mechanism. Two groups of reactions
leading to decenes were found: those with first-order dependence on 1-hexene concentration, and
those which were independent of 1-hexene concentration. The hexene-dependent reactions
accelerated with an increase in temperature, whereas the hexene-independent reactions slowed. The
hexene-independent formation of decenes was explained as arising from coordination of ethylene to
the catalyst still in complex with just formed 1-hexene. It was shown that the stepwise elimination
mechanism of chromacycloheptane decomposition cannot explain the observed product distribution,
so an alternative hydrogen shift mechanism should be considered. The results help identify the
optimal conditions for low or high decene formation in ethylene trimerization reaction.
5. Acknowledgement
The authors thank Mr. Stanislav Kiselev for his help with NMR spectra, and Dr. Denis Lenev
for his comments on the paper. The authors also thank SIBUR Holding, Ltd., for financing the
research and approving the publication of this article.
References:
18
Page 18 of 21

[1]
J.T. Dixon, M.J. Green, F.M. Hess, D.H. Morgan, J. Organomet. Chem. 689 (2004)
3641-3668.
[2]
[3]
[4]
[5]
T. Agapie, Coord. Chem. Rev. 255 (2011) 861-880.
D.S. McGuinness, Chem. Rev. 111 (2011) 2321-2341.
J.R. Briggs, Chem. Commun. (1989) 674-675.
R. Emrich, O. Heinemann, P.W. Jolly, C. Krüger, G.P.J. Verhovnik, Organometallics 16
(1997) 1511-1513.
[6]
[7]
T. Agapie, J.A. Labinger, J.E. Bercaw, J. Am. Chem. Soc. 129 (2007) 14281-14295.
J.W. van Rensburg, C. Grové, J.P. Steynberg, K.B. Stark, J.J. Huyser, P.J. Steynberg,
Organometallics 23 (2004) 1207-1222.
[8]
[9]
Bhaduri S., Mukhopadhyay S., Kulkarni S.A., J. Organomet. Chem. 694 (2009) 1297-1307.
Y. Yang, H. Kim, J. Lee, H. Paik, H.G. Jang, Appl. Catal. A 193 (2000) 29-38.
[10] L.H. Do, J. A. Labinger, J.E. Bercaw, Organometallics 31 (2012) 5143-5149.
[11] A. Carter, S.A. Cohen, N.A. Cooley, A. Murphy, J. Scutt, D.F. Wass, Chem. Commun.
8 (2002) 858-859.
[12] H. Hagen, Ind. Eng. Chem. Res. 45 (2006), 3544 –3551.
[13] R.M. Manyik, W.E. Walker, T.P. Wilson., J. Catal. 47 (1977) 197-209.
[14] R. Walsh, D. H. Morgan, A. Bollmann, J. T. Dixon, Appl. Catal. A 306 (2006) 184-191.
[15] A. Wöhl, W. Müller, S. Peitz, N. Peulecke, B.R. Aluri, B.H. Müller, D. Heller, U. Rosental,
M.H. Al-Hazmi, F.M. Mosa, Chem. Eur. J. 16 (2010) 7833-7842.
[16] S. Kuhlmann, C. Paetz, C. Hagele, K. Blann, R. Walsh, J.T. Dixon, J. Scholz, M. Haumann,
P. Wasserscheid, J. Catal. 262 (2009) 83-91.
[17] L.E. Bowen, D.F. Wass, Organometallics 25 (2006) 555-557.
[18] R.B. Sheele, A. Katzakian, J.J. Scigliano, E.E. Hamel, US Patent 3,962,182 (1976).
[19] T.M. Zilbershtein, A.A. Nosikov, A.I. Kochnev, M.V. Lipskikh, A.K. Golovko, Petr. Chem.
52 (2012) 253-260.
[20] Suzuki Y., Kinoshita S., Shibara A., Ishii S., Kawamura K., Inoue Y., Fujita T.,
Organometallics 29 (2010) 2394-2396.
[21] P.H.M. Budzelaar, Can. J. Chem. 87 (2009) 832–837.
19
Page 19 of 21

Highlights
• Isomeric decenes were formed in accordance with metallacyclic mechanism
• There are hexene-dependent and hexene-independent decene formation pathways
• Hexene-independent pathways seem to start before 1-hexene release from the catalyst
• The stepwise elimination mechanism is inconsistent with the product distribution
[22]
20
Page 20 of 21

*Graphical Abstract (for review)
Page 21 of 21