Accepted Manuscript

Title: Reaction mechanism of ionic liquid catalyzed

alkylation: alkylation of 2-butene with deuterated isobutene

Author: Ying Liu Lihong Wang Rui Li Ruisheng Hu

PII: S1381-1169(16)30176-5
DOI: http://dx.doi.org/doi:10.1016/j.molcata.2016.05.005
Reference: MOLCAA 9880

To appear in: Journal of Molecular Catalysis A: Chemical

Received date: 26-1-2016

Revised date: 29-4-2016
Accepted date: 9-5-2016

Please cite this article as: Ying Liu, Lihong Wang, Rui Li, Ruisheng Hu,
Reaction mechanism of ionic liquid catalyzed alkylation: alkylation of 2-
butene with deuterated isobutene, Journal of Molecular Catalysis A: Chemical
http://dx.doi.org/10.1016/j.molcata.2016.05.005

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.
Reaction mechanism of ionic liquid catalyzed alkylation:
alkylation of 2-butene with deuterated isobutene

Ying Liu, Lihong Wang, Rui Li, Ruisheng Hu

School of Chemistry and Chemical Engineering, University of Inner Mongolia, Hohhot, China

1/25
Graphical Abstract
(DMH)
+ Light
+ ends

- C
+ + -
C
CIL + +
- Cu C12+ TMP
C

+ - - +
C

+ i-C4H10
Heavy
+ +
ends
(TMP)

Highlights

 Deuterated isobutane was used for investigating the reaction mechanism.

 High solubility of C4 leads to the induction period of ionic liquid very short.
 The scission of C12+ plays an import role for forming trimethylpentanes.
 Dimethylhexanes are formed primary by the alkylation of sec-C4H9+ with 2-butene.
 The formation of light and heavy ends should be attributed to the C12+ cations.

2/25
Abstract
The ionic liquid catalyzed alkylation of 2-butene with deuterated isobutane was studied in a

continuous flow equipment. Product analyses with time and deuterated distribution determinations

were obtained. It is found that the induction period of ionic liquid alkylation is much shorter than

that of sulfuric acid. A considerable difference in isobutane solubility between ionic liquid and

sulfuric acid was observed with ionic liquid having a greater tendency to dissolve isobutane at the

start-up of alkylation. Deuterated product distributions indicate that trimethylpentane fractions

stemmed primarily from the self-alkylation of isobutane, the direct alkylation reaction of C4

hydrocarbons, and the scission of C12+ intermediates. Most dimethylhexanes should come from the

direct addition of sec-butyl carbonium ions to 2-butenes.

Keywords: Alkylation; Isobutane; Ionic liquid; Reaction mechanism.

3/25
1. Introduction
The acid catalyzed alkylation of butenes with isobutanes is an important process in the

petrochemical industries. The products from a sulfuric or hydrofluoric acid alkylation reactor

consist of a mixture of mono-methyl, di-methyl, and tri-methyl alkanes. Of these, the tri-branched

octanes are the most desired ingredients in gasoline because they have high octane numbers. In

order to improve the yield and selectivity of isooctanes in alkylate, the alkylation mechanism of

protonic acids (i.e., H2SO4 and HF) has been studied for many years. C4 alkylation reaction is

believed to proceed via a carbonium ion mechanism [1-3]. First, rapid protonation of 2-butene to

sec-butyl carbonium ion (sec-C4H9+) followed by the hydride transfer between isobutane and

sec-C4H9+ forms tert-butyl carbonium ion (tert-C4H9+) and n-butane. The resulting tert-C4H9+

reacts rapidly with 2-butene to produce trimethylpentane cation (TMP+), which is converted to

2,2,3-trimethylpentane (2,2,3-TMP) via hydride transfer from isobutane. A new tert-butyl

carbonium ion is generated during the hydride transfers process and another reaction cycle starts.

A general reaction scheme for isobutane/ 2-butene alkylation can be summarized as follows:

2-C4H8 + H+ → sec-C4H9+ (1)

sec-C4H9+ + i-C4H10 → tert-C4H9+ + n-C4H10 (2)

tert-C4H9+ + 2-C4H8 → C8H17+ (2,2,3-TMP+) (3)

2,2,3-TMP+ + i-C4H10 → 2,2,3-TMP + tert-C4H9+ (4)

Albright [4-6] has proposed a more complete reaction scheme when H2SO4 is used as a

catalyst. The reactions of butyl sulfates as well as the role of acid-soluble hydrocarbons are

considered in the global scheme. When hydrofluoric or trifluoromethanesulfonic acid was used to

catalyze isobutanes and isobutylenes, a detailed reaction pathway was described by Olah et al. [7,

8] Density functional theory and Ab initio method have also been employed to determine the

reaction mechanism of C4 alkylation [9, 10]. The results of theoretical calculation tended to

support the classic mechanism assumption that 2-butene is rapidly protonated to form sec-C4H9+,

and then reacts with isobutane to form tert-C4H9+. The resulting carbonium ion is deprotonated to

form isobutene that reacts rapidly with tert-C4H9+ to form TMP+.

Ionic liquids (ILs) as new catalysts have shown many interesting properties [11, 12]. For

instance, the most important properties of C4 alkylation may be the acid strength and the solubility

of isobutane in the acid [13]. But for ILs, these properties can be readily tuned by using different
4/25
organic cations or inorganic anions [14-17]. Additionally, many ILs can maintain high catalytic

activity during the whole alkylation reaction without carbon deposition problems as compared to

solid acid alkylation. Some ILs have been reported for the pilot-scale alkylation test [11]. However,

few studies have been devoted to the reaction mechanism of IL catalyzed alkylation reaction. One

possible reason may be that the alkylation reaction process is complex because of the

simultaneous reactions including olefin polymerization, protonation, and cracking. The difficulty

in detecting the intermediates under the rigorous reaction conditions also leads to the alkylation

mechanism of ILs has not been validated [18]. On the other hand, almost all of previous research

works on both the alkylation reaction and the role of ILs had been carried out in batch reactors, in

which the acid was contacted with the isoparaffin-olefin and the entire mixture analyzed at the end

of the reaction. It is almost impossible to detect the intermediates and to study the reaction

mechanism by the product distribution over time [19].

There are some experimental results of IL alkylation indicating that the classical reaction

mechanism cannot provide a full explanation. For example, when chloroaluminate ILs are used as

catalysts as compared to H2SO4 alkylation, a short contact time between the acid phase and feed is

favored [20, 21]. If a composite IL [22, 23] or a Brønsted-Lewis acidic IL [16] is employed under

the optimal operating conditions, the alkylate will contain about 5-10 wt% C5-C7 compounds

(light ends) and 5-10 wt% compounds of higher molecular weight (heavy ends). Particularly, the

most valuable trimethylpentanes are often greater than 80 wt%. These results from the point of

view of product distributions are far superior to those of traditional acids.

The aim of this paper is to study the reaction mechanism of IL catalyzed C4 alkylation

through experimental measurements of reaction. Two ionic liquids, [BMIm]Cl-AlCl3 and

[BMIm]-AlCl3-CuCl, were used as catalysts. The approach adopted in this investigation was to

utilize flow equipment, in which the reaction could be followed as a function of time, and the

product distributions with time could be obtained. In addition, deuterated isobutane (i-C4D10) was

used as an isoparaffin feed and a tracer to react with 2-butenes. By means of analyzing the

deuterated product distributions, a relatively easy way was found to investigate the reaction

mechanism. A better understanding of the chemistry of C4 alkylation might help explain the

formations of trimethylpentanes, light ends, and heavy ends in different ionic liquids.

5/25
2. Experimental
The chloride salt of the 1-butyl-3-methyl- imidazolium cations ([BMIm]Cl) was obtained from

Sigma-Aldrich Chemical Company and dried under vacuum at 100 oC. [BMIm]Cl-AlCl3 ionic

liquid (BIL) and [BMIm]Cl-AlCl3-CuCl (composite ionic liquid, CIL) were prepared and

characterized using methods as early described [17]. In this work, the mole ratios of AlCl3 and

CuCl to organic salt are 1.6:1 and 0.4:1, respectively. Isobutane and 2-butene were commercial

products from China National Petroleum Corporation (CNPC) with 99% purity and used without

further purification. Deuterated isobutane (i-C4D10, 98%) was obtained from Cambridge Isotope

Laboratories, Inc.

Experimental work was carried out in a small continuous flow reactor, as shown in Fig. 1. A

typical experiment is performed in the following manner. After purging the entire system with

nitrogen, the ionic liquid is charged to the reaction tube (over-all length 12 cm, inside diameter 1.5

cm) to a level about 5 cm above the bottom, so that it completely fills the reaction zone. The feed is

a premixed 2-butene/isobutane (or deuterated isobutane) blend with a mole ratio of i-C4H10/C4H8 of

approximately 7/1. The feed of hydrocarbon flows from the reactant charge cylinder and is

catalyzed by IL in the reaction tube. The hydrocarbon-acid emulsion is disengaged in the top of the

reaction tube and the hydrocarbon is depressurized and then vaporized to analyze. The total

hydrocarbon sample is periodically analyzed on a gas chromatograph (Hewlett-Packard, 6890

Series II) equipped with a mass spectrometer (Hewlett-Packard, 5972 Series II column). The initial

ionic liquid charge remains in the reactor for the duration of the experiment.

Studies on the induction period were carried out in the same reactor using method as literature

[24] described. Isobutane and 2-butene were all gas-phase feeds. In a typical experiment, the

IL/isobutane mixture was put into the reactor at first and the reactor was initially at atmospheric

pressure. After mixing to assure saturation of the IL with isobutane, 2-butene was pumped into the

reactor at a constant rate (e.g. 10 ml/min). Meanwhile, the consumption of isobutane with time

was measured by observing the input of isobutane required to maintain a constant pressure (e.g.

140 kPa, absolute pressure) in the reactor.

3. Results and discussion

3.1. Induction period and the alkylate components with time

Stewart and Calkins have measured the induction period of the sulfuric acid alkylation by
6/25
observing the gaseous isobutane consumption [24]. We determined the induction period of the IL

alkylation using the similar method. Fig. 2 shows the results made during a typical experiment.

Both IL alkylation processes all have an induction period, during which little or no isobutane is

used. However, the induction period of IL alkylation is less than 4 min, which is far shorter than

that of H2SO4 alkylation. Table 1 lists the product distributions at the induction period, the steady

state period, and the terminal reaction period, respectively. The BIL and CIL product distributions

have a marked difference at the induction period. The CIL alkylate contains more C8 compositions,

however, the content of C12 is also higher than that of BIL. During the steady state and the

terminal reaction periods, C12 fractions of the CIL alkylate have a tendency to decrease while TMP

increasing with time.

The change of alkylate components with time was also investigated in the small flow unit

using liquid-phase i-C4H10/2-C4H8 feed, and the data are illustrated in Fig. 3. An important

difference observed when comparing BIL and CIL product compositions is that a marked decrease

of dodecene (C12=) and dodecane (C12o, saturation of C12= by protonation and hydride transfer)

occurred during the CIL alkylation. A higher amount of dodecene is formed at the initial 30 min of

continuous reaction. Meanwhile, the amount of octenes (C8=) was less than 3 wt%. With the time

increasing, the amounts of dodecenes decreased but trimethylpentanes and octenes (C8=) increased.

Finally, the amounts of dodecenes, trimethylpentanes, and octenes nearly remained constant. By

contrast, the octenes and octanes were the main products before 30 min for the BIL systems, but

octenes decreased and dodecenes increased with time increasing. Generally, the rate of 2-butene

oligomerization is faster than that of alkylation at low isobutane concentration. It is easy to

understand C8= could be transferred to C12= over time in the BIL system. But for CIL system, it is

proposed that 2-butene around Cu ions were firstly polymerized to C12=, and then C12o cleaved to

C8= and isobutane. The direct evidence is illustrated in Fig. 6.

3.2. C12+ intermediates

Table 2 lists the product distributions obtained when deuterated isobutane is alkylated with

2-butene in the presence of ILs.

7/25
 Most researchers have considered that scission of the C12+ cations will lead to the precursors

of pentanes, hexanes, and heptanes. In addition, C16 fractions are also a result of the

polymerization reaction of C12+ and butenes: C12+ + C4=→ C16+ + i-C4o →C16 + i-C4+.

However, the deuterated product distribution shows that C12 fractions may be formed through

two reaction pathways. Some dodecene and dodecane are found containing less deuterium element,

which molecular weight is 168 (C12H24) and 171 (C12H25D from Route 1), respectively. But many

dodecanes with more deuterium were also obtained, and they have higher molecular weight

(C12D10H16-180 from Route 2). These results suggest that the alkylation between C8+ and 2-butene,

as well as the oligomerization of 2-butene produce C12+ intermediates.

3 C4H8 → C12H24 (168) + H+ → C12H25+ + i-C4D10 → C12H25D (171) + i-C4D9+

(Route 1)

i-C4D9+ + C4H8 → i-C8D9H8+

i-C8D9H8+ + C4H8 → C12D9H16+ + i-C4D10 → C12D10H16 (180) + i-C4D9+

(Route 2)

It is noteworthy that the formation of C12+ intermediates in BIL and CIL has obvious

differences. A relatively high quantity of C12D10H16 (180) is detected, indicating the secondary

alkylation between isooctyl ion and 2-butene in BIL should be the main pathway of the formation

of C12+ intermediates (Route 2). In CIL system, most of dodecanes are found with lower molecular

weight (C12H25D-171). The only reasonable explanation is that the C12 intermediates are formed

by the oligomerization of non-deuterated 2-butene when CIL is used as the catalyst (Route 1).

3.3. The formation of light ends

As Table 2 listed, the deuterium elements were detected in the fragments of C5-C7, and the

saturated C5-C7 hydrocarbons were the primary products of light ends. Deuterated C5-C7 products

would be attributed to the scission of deuterated C12+ intermediates. The precursors of pentanes,

hexanes, and heptanes were generated by the following reaction:

C12+ → C5+ + C7= (5)

C12+ → C5= + C7+ (6)

C12+ → C6+ + C6= (7)

Saturation of the ionic fragments by hydride transfer and of the olefins by protonation would

increase these low boiling saturated hydrocarbons observed in the continuous experiments. If a
8/25
C12+ fragment containing a uniform distribution of deuterium were to split into C5+ and C7+

fragments, the amount of deuterated pentanes would be the same as that of deuterated hexanes.

The product of BIL alkylation provides a better chance to observe this process for its C12

components (C12D10H16) containing more deuterium elements. As data listed in Table 2, the total

amount of deuterated C5 fragments is 4.95 wt%, which is almost equal to the value of deuterated

C7 components 4.92 wt% (2,4-dimethylpentane + 2,3-dimethylpentane). The agreement between

theoretical and experimental values for BIL system again indicates that the C5- and C7- fractions

come from scission of a C12-intermediate.

However, most C5-C7 fractions of the CIL alkylate do not contain deuterium element. These

hydrocarbons would be a result from the protonation of dodecene (C12H25+). It is an indication that

2-butenes in the CIL system prefer to form dodecenes by oligomerization reaction even at high

isobutane concentration (I/O=7:1).

3.4. Heavy ends

Most heavy ends (i.e., C9+ isoparaffins) are produced by alkylation-type reactions of butenes

in the acid phase. For instance, the addition of a 2-butene molecule to a C5+ carbocation can form

a trimethylhexane (2,2,4- or 2,2,5-TMH). Decane and undecane would come from the alkylation

of 2-butene with C6+ and C7+ fragments, respectively. A general reaction scheme for 2-butene

alkylation can be summarized as Route 3.

i-Cn+ + 2-C4H8 → i-Cn+4H2n+9+ + i-C4H10 → i-Cn+4 H2n+10 + i-C4H9+ (n=5-7)

(Route 3)

Dodecanes undoubtedly arise from saturation of the C12+ intermediate by hydride transfer

with isobutane. In the presence of CIL systems, dodecenes and dodecanes would also be formed

by the oligomerization reaction of 2-butenes as Route 1, for most of them with less deuterium

element (Table 2). Insufficient material with molecular weight higher than dodecane was present

in the IL alkylates, so that it was impossible to obtain analyses on these fractions.

3.5. The formation of trimethylpentanes

Trimethylpentanes (TMP) are the most desired isooctane product because of their high

research octane numbers. Although the TMP selectivity of CIL is much greater than that of BIL,

there are few profound discussion and research on this difference. Table 2 also lists the product

distribution of trimethylpentanes for both ILs with deuterated isobutane as the reactant, which
9/25
would provide insight into the formation mechanism of TMP using different ILs. According to the

carbonium ion mechanism, four reaction pathways lead to forming trimethylpentanes:

(1) Self-alkylation of isobutanes, the reaction involves the transfer of hydrogen from

isobutane to 2-butene and then alkylation of the resulting isobutylene with isobutane.

i-C4H10 + C4H8 → i-C4H8 + n-C4H10 (8)

i-C4H8 + i-C4H10 → 2,2,4-TMP (9)

(2) Isobutane is directly alkylated with 2-butene.

C4H8 + tert-C4H9+ → 2,2,3-TMP+

or

2-C4H8 + tert-C4D9+ →2,2,3-TMP+ + H- → 2,2,3-TMP (C8H9D9-123) (10)

(3) 2-butene is isomerized to isobutylene followed by the alkylation reaction with isobutane.

2-C4H8 → i-C4H8 (11)

i-C4H8 + i-C4H10 → 2,2,4-TMP

or

i-C4H8 + i-C4D10 → 2,2,4-TMP (C8H8D10-124) (12)

(4) Isomerization of dimethylhexanes.

Deuterated TMP distribution data in Table 2 indicate that self-alkylation reactions are

important in the IL systems, although self-alkylation is not so in H2SO4 [25]. Over 8 wt%

TMP-132 (molecular weight=132) are observed in the two IL alkylates, suggesting that these

trimethylpentanes should come from the deuterated isobutane through self-alkylation (reactions 8

and 9): i-C4D8 + i-C4D10 → 2,2,4-TMP (C8D18-132). This may be due to the higher solubility of

isobutane with respect to butenes in IL than in H2SO4. Reactions 10-12 might predominate for the

formation of trimethylpentanes in the BIL system, because a large proportion of trimethylpentanes

in the alkylate is TMP-123 (from reaction 10) and TMP-124 (from reaction 12).

Additionally, if more 2-butene transferred to isobutylene, more TMP-124 would be obtained

through reactions 11 and 12. Thus, the amounts of TMP-123 and TMP-124 also indicate the extent

of isomerization of 2-butene to isobutylene. In both IL systems, the amount of TMP-123 is

significantly lower than that of TMP-124, indicating that 2-butene is more likely to form

isobutylene before the alkylation reaction.

The CIL catalyst yielded significantly less TMP-123 and TMP-124 as compared to BIL.
10/25
However, TMP-116 fraction accounts for about 30% of total trimethylpentanes. The most

plausible explanation for these data involves rapid oligomerization of the non-deuterated 2-butene

followed by cracking and hydrogen transfer reactions that produce trimethylpentanes with less

deuterium. A mechanistic scheme for this is presented in reactions 13 through 16.

3 C4H8 → C12H24 + H+ → C12H25+ (13)

C12H25+ → i-C4H9+ + C8H16 (14)

i-C4D10 + C8H16 → i-C8H16D2 (TMP-116) + i-C4D8 (15)

i-C4D10 + i-C4D8 → i-C8D18 (TMP-132) (16)

As compared to the BIL alkylate, more TMP-132 was obtained when CIL was employed as

catalyst. These extra TMP-132 compositions should be attributed to reaction 16. Thus, reactions

13-16 could be considered as another pathway of isobutane self-alkylation to form TMP.

3.6. The formation of dimethylhexanes

When 2-butene is alkylated with sec-butyl carbonium ions, the expected primary products

would be 3,4-dimethylhexane (3,4-DMH, reaction 17). Clearly, other paths such as the

rearrangement of a trimethylpentyl cation (reaction 18) and cracking of C12+ (reaction 19) must

also contribute to dimethylhexanes’ formation.

2- C4H8 + sec-C4+ → 3,4-DMH+

or

2-C4H8 + sec-C4D9+ →3,4-DMH+ + H- → DMH-114 (C8H18) (17)

TMP+ → DMH+ (18)

C12+ → DMH + i-C4+ (19)

The formation of dimethylhexanes may in fact proceed via several routes at once. In the

previous studies, it is not clear which pathway is the most important for IL systems. DMH

distribution data in Table 2 would help study more details of the formation of DMH fractions.

Because C12+ intermediates in BIL systems were mainly generated from the deuterated

isobutane through Route 2. If the scission of C12+ were the main source of DMH, the

dimethylhexane would contain more deuterium or has a higher molecular weight. When BIL was

employed, however, the amount of DMH-114 (C8H18, from reaction 17) in Table 2 is found much

higher than that of DMH-124 (C8H8D10, from TMP-124 → DMH-124), 10.7 wt% vs. 1.1 wt%.

Similarly, in the CIL system only less than 0.3 wt% of the product are DMH with deuterium
11/25
(DMH-124), in contrast with 3.1 wt% DMH without deuterium (DMH-114). Obviously, Route 4

would not form most BIL dimethylhexanes after saturation of the iso-olefins produced in the

cracking reaction or after hydride transfer to the i-C8+ released in the fragmentation.

i-C12+ → i-C8= + i-C4+ or i- C12+ → i-C4= + i-C8+ (Route 4)

As Fig. 4 shown, no obvious fluctuations of the amounts of DMH were observed in both IL

systems after 10 min of the continuous reactions. Dimethylhexanes should be formed at the initial

period of alkylation. It is possible to conclude that reaction 17 is responsible for the formation of

most dimethylhexanes. In the IL systems, sec-butyl carbonium ions (sec-C4H9+) are believed to be

the primary intermediates to form dimethylhexanes.

3.7. The protonation of 2-butene

According to the aforementioned mechanism, sec-butyl carbonium ions would be the main

source of DMH for the IL alkylation process. It is necessary to learn more information about the

formation of sec-butyl cations. In fact, tert-butyl cations are the principal cations during the whole

C4 alkylation reaction, because tert-butyl cations are easily formed through hydride transfer

reaction of isobutanes. But the product distribution with time suggests that most DMH are formed

in the initial step of the alkylation reaction. DMH mainly comes from the alkylation of butenes

with sec-butyl carbonium ions in the ILs. Few studies have discussed the formation of sec-butyl

ions in Lewis acids. If deuterated isobutanes are used as isoparaffin feed, the following reactions

seem suggest that n-butane with molecular weight 59 could indicate the formation of sec-C4H9+ in

the Lewis acidic IL.

2-C4H8 + H+ → sec-C4H9+ (20)

sec-C4H9+ + i-C4D10 → n-C4H9D (59) + i-C4D9+ (21)

Data in Table 2 shows that about 0.2 wt% n-C4H9D was detected during the first 30 min of

continuous reaction. This result could be considered as the evidence that the protonation of

butenes did occur in the Lewis acids.

3.8. The participation of isobutane and reaction mechanism

A higher concentration of isobutane on the acid might increase the selectivity to TMP and

decrease the length of the induction period [1]. Thus, a test on isobutane solubility was carried out

in the reactor to investigate the participation of isobutane at the start-up of alkylation reaction. The

mixture of i-C4D10, i-C4H10, and 2-butene (molar ratio = 1:10:1) was used as a liquid-phase feed
12/25
for the alkylation reaction. The total hydrocarbon sample, the product with an excess of isobutane

feed, is periodically analyzed on a mass spectrometer. The mass spectrum of initial

i-C4D10/i-C4H10 mixture is shown in Fig. 5 (a), and the peak at m/z= 41.0 shows a 97% relative

abundance due to the presence of i-C4D10. When i-C4D10 was dissolved in the acid phase over time

and reacted with 2-butene, the isobutane sample from the reactor outlet would contain less i-C4D10.

This process could decrease the relative abundance of isobutane isotope at m/z = 41.0. The

relativity abundance of peak at 41.0 might indicate the changes of isobutane solubility in the acid

phase. As Fig. 5 (b) shown, under the identical operation conditions, the relative abundance of

isobutane isotope at m/z = 41.0 in both ILs decreased quickly. To the contrary, in sulfuric acid the

deuterated isobutane were consumed completely needing more time. It suggests that the solubility

of isobutane in acid is higher for IL than for sulfuric acid. This in turn decreases the length of the

induction period.

In the alkylation reaction of 2-butene with isobutane, the participation of isobutane is the

most important factor as butene is much more active. In order to investigate the participation of

isobutane and 2-butene in the continuous alkylation reaction, the changes of deuterated C8

fractions with time were illustrated in Fig. 6 by using i-C4D10/2-C4H8 as feed.

Deuterated C8 products would indicate the participation of isobutane or 2-butene in the

alkylation reaction. As mentioned before, C8= fractions mainly come from the dimerization of

2-C4H8. TMP-124 is the product of primary reaction i-C4D10 + C8H16 → i-C8H16D2 (reaction 15),

and TMP-132 is generated by the self-alkylation i-C4D10 + i-C4D8 → i-C8D18 (reaction 16). As Fig.

6 shown, the alkylation of 2-butene with deuterated isobutane catalyzed by the CIL gave about 27

wt% C12 products at the first 5 min. Nevertheless, these products could be significantly decreased

with time through the scission of C12+ (reactions 13-16). The increase of TMP-116 with time

indicates that the alkylation in CIL has such way to form TMP as compared to the BIL (reactions

1-4). 2-butene is readily to be adsorbed by IL in the presence of Cu ions or CuAlCl4, which would

lead to a relatively high amount of C12 fractions in alkylate at the start-up of alkylation. However,

this disadvantage could be weakened by increasing the mole ratio of isobutane to 2-butene as the

previous study discussed [23]. Moreover, many C12 fractions could be cracked to form TMP with

time, as can be seen in Figs. 3 and 6.

In addition, the amount of DMH in the CIL alkylate is lower than that of BIL, which is also a
13/25
result of the π-bond complexation of Cu-olefin. If a large number of 2-butenes are adsorbed by the
CIL at the start-up of the alkylation process, the possibility of 2-butene reacting with sec-C4+
(reaction 17, 2-butene + sec-C4+ → 3,4-DMH+) is small. Thus, less DMH be produced in CIL
systems.
Due to the higher solubility of isobutane in the IL, hydrogen transfer reactions between C8+
ions and isobutane are relatively enhanced with respect to H2SO4. It increases the selectivity to
TMP and allows shorter induction period and shorter contact time to be used in the alkylation
reactor. The complexation of Cu-olefin enhances the hydrogen transfer reactions between TMP+
ions and isobutane, promotes the transformation from C12 to TMP, and inhibits the formation of
DMH. These effects finally improve the quality of CIL alkylate. A general reaction scheme for the
CIL alkylation with time is illustrated in Fig. 7.

4. Conclusion
The following conclusions can be drawn from our experimental results:
(1) By means of the relative abundance of deuterated isobutane at m/z = 41.0, the induction
periods of isobutane/2-butene alkylation are obtained. Due to a high solubility of isobutane in the
acid phase, the induction period of IL alkylation is shorter than that of H2SO4.
(2) The product distribution of i-C4D10/ 2-C4H8 alkylation is obtained in the continuous flow
equipment. The formation of C12+ intermediates in BIL and CIL has obvious differences. For the
CIL catalyst, C12 fractions are primarily generated by olefin polymerization reactions, while the
secondary alkylation of 2-butene with C8+ cations is the main source of C12+ intermediates for BIL
system.
(3) Most light ends should be attributed to the β-scission of C12+ cations, while heavy ends
are coming from the reaction of 2-butene with C5-C7 alkanes.
(4) In IL alkylation, the self-alkylation of isobutanes and the scission of C12+ all can produce
trimethylpentane fractions. C12+ intermediates play an import role for forming trimethylpentanes
when CIL is used.
(5) 2-butenes are easily protonated in the presence of ILs. The signal of forming sec-C4H9+,
n-C4H9D, is observed during the protonation of 2-butenes. Dimethylhexanes are mainly formed by
the alkylation of sec- C4H9+ with 2-butene.

Acknowledgements
We thank the National Natural Science Foundation of China for financial support (No.
21266015, No. 20806036).

14/25
References
[1] F.C. Whitmore, J. Am. Chem. Soc. 54 (1932) 3274-3283.

[2] L. Schmerling, J. Am. Chem. Soc. 67 (1945) 1778-1783.

[3] L. Schmerling, J. Am. Chem. Soc. 68 (1946) 275-281.

[4] L.F. Albright, Ind. Eng. Chem. Res. 42 (2003) 4283-4289.

[5] L.F. Albright, Industrial and Laboratory Alkylations, (ACS Symposium Series, Vol. 55), ACS,

Washington DC, 1977.

[6] L.F. Albright, K.W. Li, Ind. Eng. Chem. Proc. Des. Dev. 9 (1970) 447-454.

[7] G.A. Olah, J. Welch, J. Am. Chem. Soc. 97 (1975) 208-210.

[8] G.A. Olah, T. Mathew, A. Goeppert, B. Torok, I. Bucsi, X.Y. Li, Q. Wang, E.R. Marinez, P.

Batamack, R. Aniszfeld, G.K. Prakash, J. Am. Chem. Soc. 127 (2005) 5964-5969.

[9] P. Wang, D. Wang, C. Xu, J. Liu, J. Gao, Catal. Today 125 (2007) 263-269.

[10] P. Wang, D. Wang, C. Xu, J. Gao, Appl. Catal. A: Gen. 332 (2007) 22-26.

[11] H. Olivier-Bourbigou, L. Magna, D. Morvan, Appl. Catal. A: Gen. 373 (2010) 1-56.

[12] Z.C. Zhang, Adv. Catal. 49 (2006) 153-237.

[13] A. Corma, A. Martinez, Catal. Rev. - Sci. Eng. 35 (1993) 483-570.

[14] P. Cui, G.Y. Zhao, H.L. Ren, J. Huang, S.J. Zhang, Catal. Today 200 (2013) 30-35.

[15] X.Q. Xing, G.Y. Zhao, J.Z. Cui, S.J. Zhang, Catal. Commun. 26 (2012) 68-71.

[16] S.W. Liu, C.G. Chen, F.L. Yu, L. Li, Z.G. Liu, S.T. Yu, C.X. Xie, F.S. Liu, Fuel 159 (2015)

803-809.

[17] K. Yoo, V.V. Naboodiri, R.S. Varma, J. Catal. 222 (2004) 511-519.

[18] J.E. Hofmann, A. Schriescheim, J. Am. Chem. Soc. 84 (1962) 957-961.

[19] J.E. Hofmann, A. Schriescheim, J. Am. Chem. Soc. 84 (1962) 953-957.

[20] Z.C. Liu, X.H. Meng, R. Zhang, C.M. Xu, H. Dong, Y.F. Hu, AIChE J. 60 (2014) 2244-2253.

[21] Y. Chauvin, A. Hirschauer, H. Olivier, J. Mol. Catal. 92 (1994) 155-165.

[22] Y. Liu, R.S. Hu, C.M. Xu, H.Q. Su, Appl. Catal. A: Gen. 346 (2008) 189-193.

[23] Y. Liu, R. Li, H. Sun, R. Hu, J. Mol. Catal. A: Chem. 398 (2015) 133-139.

[24] T.D. Stewart, W.H. Calkins, J. Am. Chem. Soc. 70 (1948) 1006-1009.

[25] L.F. Albright, M.A. Spalding, J. Faunce, R.E. Eckert, Ind. Eng. Chem. Res. 27 (1988) 391-397.

15/25
 Back pressure regulator

Sample
analyzed by GC-MS
Reactant charge Reaction tube
cylinder

i-C4H10 or Reciprocating
i-C4D10 / 2-C4H8 magnet for mixing
Ionic
Pump liquid
2-butene feed

Fig. 1 Continuous flow reactor for C4 alkylation

16/25
 B

Terminal reaction period

300 S 300 S

250 250
Steady state period
Volume delivered, mL

200 B 200 B

150 150

100 100
[BMIm]Cl-AlCl3-CuCl [BMIm]Cl-AlCl3

50 50
Induction period

0 0
0 2 4 6 8 10 12 14 16 18 0 2 4 6 8 10 12 14 16 18
Time, min Time, min

Fig. 2 Rate of consumption of isobutane (S) gas produced by the addition of 2-butene (B) 2-butene

gas at a constant rate, in ILs

17/25
 50

40
=
30
Hydrocarbon in Alkylate, wt%

TMP C8 C12
20 [BMIm]Cl-AlCl3-CuCl

10

30

20

=
10 TMP C8 C
12
[BMIm]Cl-AlCl3
0
0 20 40 60 80 100 120 140 160 180
Time, min

Fig. 3 TMP, C8= and C12 components vs. continuous reaction time

18/25
 12
10
8 DMH-114
Dimethylhexane in Alkylate, wt%

6 DMH-124
4 [BMIm]Cl-AlCl3
2
0
3.5
3.0
2.5
DMH-114
2.0
DMH-124
1.5
1.0 [BMIm]Cl-AlCl3-CuCl
0.5
0.0
0 20 40 60 80 100 120 140 160 180
Time, min

Fig. 4 The amounts of dimethylhexanes with time

19/25
 100

Relative abundance of m/z = 41.0, %

43.1 100 [BMIm]Cl-AlCl3-CuCl
80
i-C4H10 [BMIm]Cl-AlCl3
90
Relative Abundance, %

60
41.0 H2SO4
40
80
20 27.1 39.2
15.1 57.1 72.4 85.1 105.0
0 70
100
43.1
80 41.0
60
60 i-C4H10 / i-C4D10
40
27.1 39.2 50
20 57.1
15.1 77.1 94.9
0 40
20 30 40 50 60 70 80 90 100 110 0 20 40 60 80 100 120 140 160 180 200
(a) m/z (b) Continous reaction time, min

Fig. 5 Solubility of i-C4D10 / i-C4H10 mixture: (a) mass spectra of i-C4D10 / i-C4H10 mixture, (b)

relative abundance at m/z = 41.0 with time.

20/25
 =
C8 -112
25 TMP-116
Deuterated C8 and C12, wt%

TMP-124
TMP-132
20
C12-171

15

10

0 10 20 30 40 50 60
Time, min
Fig. 6 Deuterated TMP and C12 components vs. continuous reaction time

21/25
 sec-C4+

-H+ +H+ Alkylation 3,4-, 2,5-,

3,4-; 2,5-; 2,4-DMH
and 2,4-DMH+
i-C4H10
2-C4H8
Alkylation 2,2,3- and H-transfer 2,2,3- and
tert-C4+ 2,3,4-TMP+ (from i-C4H10) 2,3,4-TMP

-H+ +H+ Self-alkylation

2,2,4-TMP+ 2,2,4-TMP and 2,5-DMH

i-C4H8 + i-C4H10
+ tert-C4+ C8= β- scission
i-C5, i-C6
β- scission
Oligomerization
2-C4H8 C12+

+ tert-C4+
C16+

Fig. 7 Scheme of reaction mechanism for the CIL alkylation

22/25
Table 1 Comparison of alkylates produced at different periods a

[BMIm]Cl-AlCl3 [BMIm]Cl-AlCl3-CuCl

Reaction time, min 1 2 6 13 1 2 6 13

Isoparaffins, wt%

C5-C7 15.4 15.2 14.6 9.8 13.5 10.5 7.2 4.5

C8= 16.3 14.5 12.9 6.6 4.0 3.9 4.1 5.2

Trimethylpentanes 10.4 12.8 15.7 23.9 23.1 26.6 37.3 47.5

Dimethylhexanes 10.7 10.8 9.5 8.7 6.4 7.3 6.9 6.4

Other C8o 4.6 5.3 5.9 4.1 3.5 4.5 4.3 5.0

2,2,5- trimethylhexanes 5.6 5.1 4.6 4.9 3.8 3.6 3.9 5.7

C12= and C12o 3.8 4.9 9.1 13.4 22.1 20.4 12.7 5.1

Other C9+ 33.2 31.4 27.7 28.6 23.6 23.2 23.6 20.6
a Reaction conditions: temperature =25 oC, IL=5 mL, rate of 2-butene input= 20 mL/ min.

23/25
Table 2 GC-Mass spectrometric component analyses of alkylates a
Component Formula MW [BMIm]Cl-AlCl3 [BMIm]Cl-AlCl3-CuCl

C4
Isobutane i-C4D10 68 0.54 0.48
n-Butane C4H9D 59 0.23 0.19
C5
2-Methylbutane C5H12 72 0.01 0.83
2-Methylbutane C5H8D4 76 4.95 0.21
C6
2-Methylpentane C6H6D8 94 2.61 0.56

3-Methylpentane C6H14 86 8.77 5.85

C7

2,4-Dimethylpentane C7H10D6 106 4.59 0.89

2,4-Dimethylpentane C7H16 100 0.07 3.77

2,3-Dimethylpentane C7H10D6 106 0.32 0.3

2,3-Dimethylpentane C7H16 100 0.01 0.37

2-Methylhexane C7H16 100 1.97 1.28

C8= (Octenes) C8H16 5.66 4.89

C8 (Trimethylpentanes)

2,2,4-TMP C8H18 114 1.03 0.99

2,2,3-TMP C8H18 114 0.48 0.33

2,3,4-TMP C8H18 114 0.31 0.19

C8 (Deuterated TMP)

C8H16D2 116 3.17 14.2

C8H9D9 123 6.3 5.1

C8H8D10 124 13.43 10.8

C8H5D13 127 0.2 0.3

C8D18 132 8.21 14.4

C8 (Dimethylhexanes)

2,3-DMH C8H18 114 2.71 0.65

2,4-DMH C8H18 114 3.28 1.03

2,5-DMH C8H18 114 4.71 1.42

C8 (Deuterated DMH)

C8H8D10 124 1.1 0.23

C9 (Trimethylhexanes)

2,2,4-TMH C9H20 128 0.13 0.32

24/25
 2,2,5-TMH C9H20 128 0.71 2.47

C9 (Deuterated TMH) C9H10D10 138 2.73 3.12

Deuterated Decane C10H10D12 154 3.96 3.98

Deuterated Undecane C11H8D16 172 3.84 5.44

Dodecenes C12H24 168 0.82 0.76

Deuterated dodecanes
C12H25D 171 1.51 13.37
C12D10H16 180 11.64 1.28
a Alkylate sample of continuous reaction at 30 min, i-C4D10 and 2-butene feed, I/O =7:1, temperature =25 oC

25/25