Metallocenes and post-metallocenes immobilized on ionic liquid-modified silica as catalysts for polymerization of ethylene

Applied Catalysis A: General 484 (2014) 134–141

Contents lists available at ScienceDirect

Applied Catalysis A: General journal homepage: www.elsevier.com/locate/apcata

Metallocenes and post-metallocenes immobilized on ionic liquid-modified silica as catalysts for polymerization of ethylene ∗ ˛ Wioletta Ochedzan-Siodłak , Katarzyna Dziubek Faculty of Chemistry, University of Opole, Oleska 48, 45-052 Opole, Poland

a r t i c l e

i n f o

Article history: Received 4 March 2014 Received in revised form 10 June 2014 Accepted 10 July 2014 Available online 18 July 2014 Keywords: Organometallic catalyst Ionic liquid Silica Polyethylene

a b s t r a c t The supported ionic liquid (SIL) strategy was used for the first time to metallocene and post-metallocene heterogeneous catalysts for olefin polymerization. The metal complexes: Cp2 TiCl2 , Cp2 ZrCl2 , FI–Ti, and Sal–Ti were immobilized in the 1-(3-triethoxysilyl)propyl-3-methylimidazolium alkylchloroaluminate ionic liquid, anchored on the surface of the mesoporous amorphous silica. The SIL systems were characterized by FTIR, 29 Si NMR, N2 adsorption, EA, AAS, TG, and SEM techniques. The developed supported catalytic systems were found to be active in the ethylene polymerization and produce the polyethylene of various properties. © 2014 Elsevier B.V. All rights reserved.

1. Introduction Polyethylene (PE) is the most common and important polyolefin polymer. It is characterized by good mechanical properties, chemical inertness, and low cost of productions [1,2]. Therefore, there is a continuous search of new organometallic catalysts for ethylene polymerization, which will have better activity or selectivity. Metallocene and post-metallocene catalysts attract a special attention because variation in their structure not only provides a possibility to control polyreaction performance but also enable to tailor polymer properties [3–7]. These catalysts are usually investigated as homogeneous systems, which show high activities in olefin polymerizations [8–10]. However, use of large amounts of solvents, lack of ability to control polymer microstructure and morphology, and undesirable phenomenon of reactor fouling are main disadvantages of the homogeneous catalysts [11–14]. For industrial applications, supported metallocenes and post-metallocenes are more appropriate because they can be used in gas- and slurryphase processes. The polymer of uniform particles with narrow size distribution and high bulk density is produced and problems with the reactor fouling are prevented [15,16]. Amongst various supports of the organometallic catalysts, amorphous silica is the most widely used. This environmentally friendly support has good mechanical properties, stability and inertness under reaction and processing conditions, relatively large surface area and porosity.

∗ Corresponding author. Tel.: +48 77 452 7147; fax: +48 77 452 7101. ˛ E-mail address: [email protected] (W. Ochedzan-Siodłak). http://dx.doi.org/10.1016/j.apcata.2014.07.016 0926-860X/© 2014 Elsevier B.V. All rights reserved.

There is also possible to control the amount and distribution of hydroxyl groups on its surface [11,13,15,17–19]. Usually, the silica surface is modified with appropriate thermal treatment or chemical treatment by using dehydroxylating agent such as hexamethyldisilazine (HMDS) [20]. However, in ethylene polymerization using zirconocene catalyst, after HMDS-treatment, the catalyst shows lower activity and produces polymers of lower molecular weight and broader polydispersities, in comparison to the catalyst on non HMDS-treated silica. Another approach for immobilization of metallocene on silica support involves use of silane coupling agents, such as Cp(CH2 )3 Si(OCH2 CH3 )3 [21,22]. The zirconocene catalysts supported on silica, chemically modified by (CH3 )3 SiCl (TMCS), show a sufficiently high activity in ethylene polymerization (up to 1638 kgPE /molZr h bar) [23], but obtained polyethylene consists of shapeless aggregates similar to those obtained with homogeneous catalysts [24]. Alonso et al. prepared a supported metallocene catalyst (Cp2 ZrCl2 ) by chemical modification of silica by silicon ethers, EtOSiMe3 and (Me3 Si)2 O, or a silazane, (Me3 Si)2 NH. The activity of these systems in polymerization ethylene was 441 and 620 kgPE /molZ h, respectively for Cp2 ZrCl2 /[SiO2 -EtOSiMe3 ] and for Cp2 ZrCl2 /[SiO2 -(SiMe3 )2 O]. In the case of Cp2 ZrCl2 /[SiO2 (SiMe3 )2 NH] only traces of product were obtained [24]. The most common method to modify silica is to tread this support, first with alkylaluminium compound (mainly MAO or AlMe3 ), and then to absorb metallocene on it. Janiak and Rieger reported that activity of the zirconocene catalyst supported on the methylalumoxane-pretreated SiO2 was up to 4716 kgPE /molZr h, but without modification by using MAO it did not exceed 70 kgPE /molZr h [25]. Both nano- and micro-sized MAO-modified

W. Och˛edzan-Siodłak, K. Dziubek / Applied Catalysis A: General 484 (2014) 134–141

silica were used to support the Cp2 ZrCl2 catalyst for ethylene polymerization [26]. The nano-sized catalyst exhibited much better activity in ethylene polymerization than micro-sized catalyst (3.8 × 104 and 8730 kgPE /molZr h, respectively), which was attributed to the large specific external surface area, the absence of internal diffusion resistance, and the better active site dispersion for the nano-sized catalyst. Despite of many advantages, application of silica modified by MAO or TMA as a support of metallocene catalyst requires use of toluene as solvent and additional amount of MAO as catalyst activator. Moreover, it is observed unfavorable effect of leaching catalyst from silica surface [14,27–30]. Recently, there has been an interest in immobilization of ionic liquids (ILs), on the surface of a support, usually silica, so-called supported ionic liquid (SIL) concept [31–33]. Ionic liquids, due to their negligible vapour pressure, large liquid range, and high thermal stability gain increasing attention and they are considered as useful solvents with multiple applications in synthesis and catalysis [34–42]. In most cases the ionic liquids are applied in homogeneous or biphasic systems, which reveal a great potential in laboratory scale, but seem do not show a significant promise for large-scale industrial catalytic applications [43]. In these systems, large quantities of the ILs are used as solvents or catalysts, which inevitably generate lots of waste materials. Their disposal is extremely difficult and makes the overall process unacceptable from environmental and economic viewpoint. Application of the SIL systems enables to avoid such obstacles [31,32,44,45]. These systems offer a very efficient ionic liquid utilization, provide relatively short diffusion distances of reactants as compared to the biphasic catalyst systems organic liquid – ionic liquid, and thus solve problems of mass transport limitation [32,45–47]. Moreover, this type of the catalyst systems can be tuned by optimizing structures of the IL, transition-metal catalyst, and solid support. The ILs immobilized on various supports were successfully checked in many catalytic reactions, for example, in hydroformylation [48], alkylation [49], epoxidation [50], hydrogenation [51]. So far, there is no literature data regarding applications of the SIL strategy for development of a heterogeneous catalytic system for polymerization of olefin. In this work, we report application of the SIL system with metallocene and post-metallocene catalysts in ethylene polymerization. The influence of the kind of catalyst used, and applied reaction conditions on the performance of the polymerization as well as the on properties of the obtained polyethylene is also presented.

2. Experimental 2.1. Preparation of the SIL system All experimental steps have to be carried out under an inert atmosphere. The amorphous silica (3 g, 0.05 mol) was calcined for 4 h at 500 ◦ C and stored under argon. Calcined silica was transferred to a round bottom flask and 1-(3-triethoxysilyl)propyl3-methylimidazolium chloride (5.0 × 10−3 mol) in toluene (80 cm3 ) was added. The mixture was stirred at 80–90 ◦ C for 18 h. In the next step, solvent and ethanol created in the grafting step were distilled off. The remaining solid in form of powder was dried under vacuum, washed with hexane (5× 30 cm3 ) in a Schlenk apparatus, dried under vacuum, then extracted for 24 h with boiling dichloromethane in an Soxhlet apparatus, and dried under vacuum. Such obtained solid was then added to a solution of AlCl3 (2.0 × 10−3 mol) in toluene (50 cm3 ), and left stirring for 1.5 h at room temperature. After filtration in the Schlenk apparatus the support was washed hexane (4× 30 cm3 ), dried under vacuum, extracted for 20 h with boiling dichloromethane in the Soxhlet apparatus, and dried under reduced pressure. The solid was mixed

135

Table 1 Characteristics of the catalytic systems. Catalytic system

Support

Catalyst precursor

A B C D E

SIL SiO2 /AlEtCl2 SIL SIL SIL

Cp2 TiCl2 Cp2 TiCl2 Cp2 ZrCl2 FI–Ti Sal–Ti

SIL: SiO2 /IL-Cl/AlCl3 /AlEtCl2 ; Fi–Ti: bis[N-(salicylideno)anilinato]-titanium(IV) dichloride; Sal–Ti: N,N -ethylenebis[5-chloro-salicylideneiminato]titanium(IV) dichloride.

with AlEtCl2 (2 × 10−4 mol) in hexane (60 cm3 ) in a ball mill for 2 h. The obtained support was filtered in Schlenk apparatus, washed hexane (4× 30 cm3 ) and dried under reduced pressure.

2.2. Immobilization of catalysts on the ionic liquid-modified silica support All experimental steps were performed in inert atmosphere. The prepared support (1 g) was mixed with catalyst precursor (2.5 × 10−5 mol Mt) dissolved in deoxygenated toluene (5 cm3 ) and hexane (50 cm3 ) in a ball mill for 24 h. The obtained catalyst system was filtered in Schlenk apparatus, washed hexane (4× 20 cm3 ), and dried under nitrogen stream. The Cp2 TiCl2 precursor mixed with the support gave the catalytic systems denoted as A. Analogously, the precursors Cp2 ZrCl2 , FI–Ti, and Sal–Ti were mixed with the support in order to obtain respectively the catalytic systems C, D, and E (Table 1). The obtained catalytic systems were in the form of powder, slightly yellow for Ti systems and off-white for Zr system. For comparison, the catalyst system B was prepared (Table 1), in which the titanocene catalyst (Cp2 TiCl2 ) was immobilized without IL, directly on silica calcinated in 500 ◦ C, and modified only by AlEtCl2 (molar ratio SiO2 /AlEtCl2 = 10/1).

2.3. Polymerization of ethylene To a glass reactor (500 cm3 ) filled with inert nitrogen atmosphere, hexane (150 cm3 ), AlEt2 Cl or AlEtCl2 or MAO (1.0–6.0 × 10−3 mol) and the supported catalytic system (2.0 × 10−6 mol Ti or Zr) were added. The ethylene was introduced at 0.5 MPa and the polymerization reaction was carried out at 30 ◦ C or 50 ◦ C for 30–90 min. The reaction was quenched by closure of the ethylene feeding, reduction of the pressure to 0.1 MPa, and addition of acidified methanol to the reaction mixture. The polyethylene product was filtered, washed thoroughly with methanol, dried at room temperature, and then stored for subsequent characterization.

2.4. Leaching experiment for SIL catalytic system Hexane (150 cm3 ), AlEt2 Cl or AlEtCl2 or MAO (4.0 × 10−3 mol), and the supported catalytic system (2.0 × 10−6 mol Ti or Zr) were added in inert atmosphere to glass flask equipped with mechanical stirrer and heating jacket. The mixture was mixed vigorously at 30 ◦ C or 50 ◦ C for 30 min. Maintaining inert atmosphere, the solid was filtered and hexane phase was placed into reactor where it was taken into the ethylene flush at typical polymerization conditions (0.5 MPa, 30 ◦ C or 50 ◦ C, 30 min.). The solution was transparent and no polyethylene was obtained. The absence of Ti or Zr in the hexane phase was also confirmed by atomic absorption spectrometry (AAS).

136

W. Och˛edzan-Siodłak, K. Dziubek / Applied Catalysis A: General 484 (2014) 134–141

2.5. Analysis FTIR analyses were accomplished using a Nicole Nexus 2002 FTIR spectrometer from 4000 to 400 cm−1 with a 2 cm−1 resolution. The FTIR analyses of the SIL system samples each step of the synthesis were investigated in Nujol KBr/KBr. The polymer samples were prepared in the form of tablets made of a polymer powder and KBr. The numbers of methyl ( CH3 ) groups were estimated from the ratio of the absorption band with the maximum at 1379.3 cm−1 to the absorption of the band at 1368.7 cm−1 for the methylene ( CH2 ) groups. 1 H NMR and 13 C NMR spectra were recorded on a Bruker Ultrashield spectrometer (400 MHz). The NMR spectra of the ionic liquid were recorded in CDCl3 . The linearity of the polyethylene was analyzed using the NMR spectra recorded in o-dichlorobenzene-d4 . The number of branches was found from the ratio of integrals of signals of the methyl groups (0.9 ppm) to the methylene groups (1.3 ppm). 29 Si NMR spectra were recorded on Varian Mercury 300 VT spectrometer in benzene-d6 . Elemental analysis (EA) was performed using apparatus EA 1108 (Fisons Instruments). Thermogravimetric analysis (TGA) data were obtained in platinum pan (capacity 100 ␮L) in a dry box. Using the TGA 2050 (TA-Instruments) the sample was pierced while being inserted into the instruments dry N2 atmosphere. The TGA data were collected at 10 ◦ C/min under N2 . Nitrogen adsorption–desorption isotherms were obtained on a ASAP 2020 (Micromeritics) instrument after out gassing the samples in vacuum at 150 ◦ C for 12 h prior to the measurements. The total surface areas were calculated according to the BET method and the average pore diameter was calculated following the BJH method. Atomic absorption spectrometry (AAS) using an ICE 3500 model (Thermo Electron Corporation) was used to determine amount of Ti and Zr. The crystallinity and the melting temperature of the polyethylene were estimated with a DSC 2010 TA Instruments. The polymer crystallinity degree was calculated using the equation: C = (Hf /Ht,c ) × 100% where; Hf – heat of fusion of the polyethylene sample, Ht,c – heat of fusion of standard = 290 J/g, C – crystallinity degree, %. Gel permeation chromatography was used to determine the molecular weight and molecular weight distribution of each polymer sample on Waters Alliance GPCV 2000 apparatus using 1,2,4-trichlorobenzene as the solvent at 142 ◦ C. The data was analyzed using polystyrene calibration curves. The bulk densities of polyethylene were measured according to ASTM Standard D 1895. Scanning electron microscope (SEM) experiment of the polyethylene samples was carried out on a Hitachi model TM 3000 electron microscope. The samples were fixed on an aluminum sample stub and coated with gold by conventional sputtering techniques. The employed accelerating voltage was 5–15 kV for SEM.

3. Results and discussion 3.1. Characterization of SIL catalytic systems In the first step the surface calcined mesoporous amorphous silica was treated with 1-(3-triethoxysilyl)propyl3-methylimidazolium chloride to prepare the SIL system in which the ionic liquid was covalently anchored on the support. In the second step, the AlCl3 was added to the ionic liquid modified support, in order to obtain chloroaluminate anions. In the third step, AlEtCl2 was introduced to the prepared system, to ensure the slightly acidic conditions through the creation of the alkylchloroaluminate anions. Finally, the metallocene or post-metallocene catalyst was anchored on such prepared support (step 4). The successive steps of the preparation of the SIL system are presented in Fig. 1. The studied SIL systems differ only by type of catalyst precursors, which were anchored on their surface at the last step of the

calcined silica support Si

OH

Si

OH

+

(EtO)3Si

N

+ N

Cl

(step 1) silica support Si

O

Si

O

Si

N

+ N

OEt (step 2)

Cl

AlCl3 eq.

silica support Si

O

Si

O

Si

N

+ N

OEt (step 3)

AlCl4

AlEt2Cl

silica support Si

O

Si

O

Si

N

+ N

OEt (step 4)

catalyst precursor

AlEtxCl4-x

Cp2TiCl2 Cp2ZrCl2 FI-Ti Sal-Ti

catalytic SIL-system

ethylene

polyethylene (PE)

Fig. 1. Schematic presentation of the SIL systems synthesis.

synthesis. Detailed analysis of the prepared SIL systems is shown on example of the system A. The grafting of the ionic liquid on the silica support via a covalent bond manner was evidenced by FTIR spectroscopy. Fig. 2 presents spectra of the pure calcined silica, silica modified with the ionic liquid (step 1–3), and finally prepared SIL catalytic system A with Cp2 TiCl2 complex (step 4). The FTIR analysis of the samples of the SIL system from each synthesis step reveal specific IR adsorption characteristics for the imidazolium ionic liquid, reflected by the bands at 3156, 3117, 1573, and 621 cm−1 ascribed, respectively, to aromatic CH, CC, CN stretching vibrations of the imidazole ring and CH bending vibrations of the methyl group [51–56]. Those bands are absent for non-modified silica calcined at 500 ◦ C, which shows in contrast a characteristic band at 3700 cm−1 ascribed to (Si) OH stretching vibrations and bands at 1105 cm−1 and 470 cm−1 for Si O stretching vibrations. The band at 3700 cm−1 disappears in the SIL system what shows a graft of the ionic liquid on the silica support. Presence of individual components of the SIL system was analyzed on the basis of thermogravimetric analysis (TGA), which was

W. Och˛edzan-Siodłak, K. Dziubek / Applied Catalysis A: General 484 (2014) 134–141

137

809

Weight (%)

3700

100

Si-OH

Si-O-Si 470

1105 3600

step 1 step 2

80

step 3

SiO2 4000

90

step 4

70 0

3200

2800

2400

2000

1600

1200

800

400

200

400

600

800

1000

Temperature (°C)

621

3156

1573

Fig. 3. Thermogravimetric analysis (TGA) of the samples of the SIL system for each step of the synthesis. Derivative weights (%/◦ C) in supporting material.

Ar(C=C)

Ar(C-H)

CH3

step 1 2800

2400

2000

3156 3117

3200

1600

1200

800

400

1573

3600

T (%)

4000

Ar(C=C) 621

Ar(C-H) CH3

step 2 4000

3600

3200

2800

2400

2000

1600

1200

800

400

4000

3600

3200

2800

2400

2000

1600

1200

800

400

3200

2800

2400

2000

performed for samples taken from each step of the synthesis (Fig. 3, supplementary data). The smallest weight losses (14.5%) and the highest thermal stability was evident for the SIL system, in which the silica surface was modified by 1-(3-triethoxysilyl)propyl-3methylimidazolium chloride (step 1). In further synthetic steps, the weight losses were higher (20.1, 23.8, and 26.2% after steps 2–4, respectively). Three different regions of the weight losses were visible on all thermograms with clearly smoothing from step to step. The first weight loss (2.5–10.3%) from 50 to 150 ◦ C corresponds to desorption and removal of water and/or water physisorbed on the external surface. The second weight loss (4.7–7.2%), from 150 to 350 ◦ C is attributed to removal organic components of the system. Finally, the third weight loss (5.6–10.5%) from 350 to 550 ◦ C is related to water loss from condensation of adjacent silanol groups to form siloxane bond [57]. The successful immobilization of the imidazolium ionic liquid on the silica support was confirmed by the analysis of the solidstate 29 Si MAS-NMR spectra (Fig. 4). The spectrum of the silica calcinated at 500 ◦ C shows characteristic single signal at −101 ppm [58] which can be assigned to silanol (SiO)3 SiOH [54,59,60]. In the spectrum of the silica with the immobilized ionic liquid, the signal is shifted in the lower magnetic field, which indicate the presence of siloxane bridges (SiO)4 Si. Moreover, there are two more signals in the higher magnetic fields, at −54 and −61 ppm, that are assigned to organosiloxane: Si O Si R (OEt)2 and (Si O)2 Si R OEt, respectively [49,54,60]. This proves that the imidazolium cations are bonded to the silica surface via one or two Si O Si bonds. Elemental analysis (EA) data apparently revealed amount of the ionic liquid on the silica support, as it is also reported in literature [49,61]. The amount of the ionic liquid was determined from percentage of elemental nitrogen and was 0.58 mmol of the ionic liquid per gram of the final catalytic system.

step 4 4000

3600

Wavenumbers

1600

1200

800

400

(cm-1)

Fig. 2. FTIR spectra of the pure calcined silica and successive synthetic steps leading to the SIL catalytic system A.

Fig. 4. 29 Si NMR spectra of the pure silica calcined at 500 ◦ C (bottom) and finally prepared SIL catalytic system A (top).

138

W. Och˛edzan-Siodłak, K. Dziubek / Applied Catalysis A: General 484 (2014) 134–141 Table 2 Influence of the polymerization conditions on the activity of the studied SIL catalytic systems.

700 SiO2 system A

V N2 (cm 3/g STP)

600 500 400 300 200 100 0 0

0.2

0.4

0.6

0.8

1

Relative pressure (p/po) Fig. 5. N2 adsorption–desorption isotherms of the calcined silica and finally prepared SIL catalytic system A.

The measured BET surface areas and pore characteristics of the SIL system were performed. The N2 adsorption–desorption isotherms of the pure calcined silica and SIL system show their mesoporous structure (Fig. 5). The results show that the ionic liquid-modified silica support is characterized by lower surface area, smaller pore volume and smaller average pore diameter, in comparison to pure silica, as expected [47,51,62]. The surface area and the total pore volume decrease significantly, respectively from 340 to 160 m2 /g and from 0.73 to 0.58 cm3 /g, while the average pore diameter decreases only moderately from about 8.6 to 4.8 nm. The grafting of the IL on the silica carrier greatly influences its surface morphology [60]. The SEM images presented in Fig. 6a and b show a smoothing of the carrier surface with the grafted ionic liquid in comparison to the surface of the calcined non-modified silica. The amount of the catalyst immobilized in the SIL systems was determined by AAS on the basis of transition metal content. For the SIL catalytic system A with the Cp2 TiCl2 metal complex the catalyst amount was determined as 0.021 mmol Ti per gram catalytic system. For the catalytic system C with Cp2 ZrCl2 , D with FI–Ti, E with Sal–Ti, and system B with Cp2 TiCl2 but without the ionic liquid, the catalyst amount was determined as 0.015 mmol Zr, 0.010 mmol Ti, 0.010 mmol Ti, and 0.037 mmol Ti per gram catalytic system, respectively.

3.2. Activity test of the SIL catalytic systems in ethylene polymerization In the presented SIL systems, the ionic liquid modifies the support surface, and thus changes its morphological properties. Simultaneously, as the medium of the catalyst, IL prevents direct interaction between the surface of the support and the homogeneous catalyst. Thus, the SIL system combines the advantages both heterogeneous and homogeneous catalysts – the final material is solid, but the active species, which are immobilized in the IL phase, act as a homogenous catalyst [46,48]. Additionally, the alkylchloroaluminate ionic liquid ensures the proper – slightly acidic reaction medium, which is necessary to perform polymerization with the use of organometallic catalysts [63–66]. The developed SIL catalytic systems with the metal complexes: Cp2 TiCl2 , Cp2 ZrCl2 , FI–Ti, and Sal–Ti, were used for the low pressure ethylene polymerization (Table 2). Traditional alkylaluminum

Item

Catalytic system

Activator

Al/Mt molar ratio

Activity (kgPE /molMt)

1 2 3 4a 5b 6c 7 8 9 10 11

A

AlEt2 Cl AlEt2 Cl AlEt2 Cl AlEt2 Cl AlEt2 Cl AlEt2 Cl AlEt2 Cl AlEt2 Cl AlEtCl2 AlEtCl2 MAO

500 1000 1500 1500 1500 1500 2500 3000 1500 2500 1500

105 1917 3602 4898 5336 908 2375 1706 144 124 1007

12 13 14

B

AlEt2 Cl AlEt2 Cl AlEtCl2

1000 1500 1000

557 516 69

15

C

MAO

1500

545

16c 17c

D

AlEt2 Cl AlEt2 Cl

2500 3000

70 90

18 19c 20 21

E

AlEt2 Cl AlEt2 Cl AlEtCl2 MAO

2500 2500 2500 1500

205 130 150 291

Polymerization conditions: catalyst (2.0 × 10−6 mol Mt), (0.5 MPa), temperature (30 ◦ C), polym. time (30 min). a Polym. time (60 min). b Polym. time (90 min). c Temperature 50 ◦ C.

ethylene

pressure

compounds (AlEtCl2 , AlEt2 Cl) as well as methylaluminoxane (MAO) were applied as activators. The presented SIL concept, where the catalyst precursor is immobilized in the ionic liquid on the silica surface, assumes that the catalyst leaching does not take place. Lack of such phenomenon was confirmed by the experiment (see Section 2) and AAS methods. The activity of the studied system A with the Cp2 TiCl2 catalyst increases in the following order: AlEt2 Cl > MAO > AlEtCl2 (Table 2, items 3, 9 and 11). This indicates that in the case of the Cp2 TiCl2 system for the ethylene polymerization, traditional alkylaluminium compound can be successfully applied instead of the expensive MAO. Also the activator/catalyst molar ratio (Al/Ti) influences the activity, as it was shown for AlEt2 Cl activator (Table 2, items 1–3, 7 and 8). In the applied activator concentration, increase of the activity with increase the activator/catalyst molar ratio was observed. It is explained by increase of the active site concentration [23,67]. It should be noted that no optimum activator/catalyst molar ratio was observed. The system A turned out to be very active and provides very high activity (3600 kgPE /molTi 0.5 h) at relatively low activator concentration (Al/Ti = 1500 mol/mol). For comparison, the activity of the system B with Cp2 TiCl2 catalyst anchored directly on the AlEtCl2 modified silica, but without the ionic liquid is presented (Table 2, items 12–14). As can be seen, the activity of the system B is considerably lower, regardless of the type of the activator used. Lower activity of the catalyst system with the silica support modified only by AlEtCl2 is caused probably by too strong Lewis acidity, which can lead to partial deactivation of the catalyst. This phenomenon can be limited by presence of chloroaluminate ionic liquid, which acts as buffer, and thus causes decrease of Lewis acidity [68,69]. It should be noted that the activity of the studied SIL systems in the ethylene polymerization were much higher than that for the previously investigated biphasic ionic liquid/hexane systems

W. Och˛edzan-Siodłak, K. Dziubek / Applied Catalysis A: General 484 (2014) 134–141

139

Fig. 6. SEM images of the pure calcined silica (a), the ionic liquid-modified silica (b), the SIL-metallocene catalytic system (c) and the obtained polyethylene (d).

[63,64,70,71]. The activity was also higher than that for the traditional heterogeneous catalyst, where the Cp2 TiCl2 catalyst was anchored directly on a modified or non-modified silica support [72,73]. Longer polymerization time increases the activity (up to 5300 kgPE /molTi), which is characteristic for supported heterogeneous catalysts. It also indicates the high stability of the studied SIL catalytic systems (Table 2, items 4 and 5). Conversely, the increase of the reaction temperature up to 50 ◦ C, results in a decrease of the catalytic activity (Table 2, item 6). This is because of deactivation of the titanocene catalyst [67]. Successful results obtained using the SIL system A with the Cp2 TiCl2 precursor led us to investigate the selected metal complexes (Cp2 ZrCl2 , FI–Ti, and Sal–Ti) (Table 2, items 15–21). It was found that the catalytic system C with Cp2 ZrCl2 shows the best activity with the activator MAO, whereas it is not active in conjunction with AlEtCl2 or AlEt2 Cl (Table 2, item 15). The activity of the system C in the ethylene polymerization, which was performed at 30 ◦ C, reaches 545 kgPE /molTi 0.5 h. This is comparable activity or even higher than those for the zirconocene catalysts, in which the Cp2 ZrCl2 precursor was immobilized directly on silica supports, non-modified or modified by MAO, AlMe3 or organosilane [23,24,73–75]. The catalytic system D with the FI–Ti complex was found to be inactive in the ethylene polymerization, which was performed at 30 ◦ C. The increase of the reaction temperature up to 50 ◦ C as well as the increase of the amount of the activator AlEt2 Cl gives barely 80 kgPE /molTi 0.5 h (Table 2, items 16 and 17). Unfortunately, that there is lack of literature reports concerning investigations of the FI–Ti complex supported on silica for ethylene polymerization. The activity of the homogeneous system reaches 300 kgPE /molTi 0.5 h at Al/Ti = 800 mol/mol, but it does not exceed

80 kgPE /molTi 0.5 h at Al/Ti = 100 mol/mol [76]. The studied system E with the Sal–Ti complex was found to be active in conjunction with all activators used (Table 2, items 18–21). The highest activity (290 kgPE /molTi 0.5 h) was obtained, when MAO was used at Al/Ti = 1500 mol/mol. When AlEt2 Cl and AlEtCl2 were applied, the activities about 150 kgPE /molTi 0.5 h and 205 kgPE /molTi 0.5 h were obtained, respectively. The obtained results are much higher than those known from literature where the silica supported Sal–Ti catalyst gives, respectively, 97 kgPE /molTi 0.5 h and 22 kgPE /molTi 0.5 h, using the MAO and AlEt2 Cl activators [77]. 3.3. Polyethylene properties Linear polyethylene is obtained, regardless of the type of the metal complex, type and amount of the activator, the reaction time and temperature. This is indicated by the singlet of methylene groups at ı = 29.9 ppm (13 C NMR). The number of branches in the polymer chain is less than one CH3 group per 1000 CH2 , as determined by 1 H NMR and FTIR. Melting temperature is relatively high (Tm = 138–141 ◦ C) (Table 3), which is typical for high density polyethylene [78]. Crystallinity degree varies in the range of 62–82%. The metal complex has a considerable influence on molecular weight. The highest value was obtained using the Cp2 TiCl2 and Cp2 ZrCl2 metallocenes (Table 3, items 2–11, and 15). The polyethylene obtained using the studied SIL-metallocene catalytic systems is characterized by monomodal GPC elution peak and narrow distribution of molecular weight. It should be noted that homogenous metallocene systems, with MAO or other suitable activator exhibit single-site behavior and they yield product with narrow molecular weight distribution (Mw /Mn < 2) [11,12,14]. However, the molecular weight distribution of polyethylene may broaden upon catalyst

140

W. Och˛edzan-Siodłak, K. Dziubek / Applied Catalysis A: General 484 (2014) 134–141

Table 3 Selected properties of the PE obtained on the studied SIL catalyst systems. Itemd

Mw (kg/mol)

2 3 4a 5b 6c 7 11 12 13 14 15 16c 17c 18 19c 20 21

592.8 477.6 534.9 601 339.1 305.6 334.3 235.5 230.8 119.7 307.3 149.2 140.3 160.1 308.3 142.3 154.1

a b c d

Mw /Mn 3.4 3.1 2.9 2.5 12.6 3 2.3 6.7 7.2 11.3 2.4 11.3 14.2 15.7 13.2 11.8 19.1

Tm (◦ C) 141 141 140 140 140 141 141 141 140 140 140 140 140 139 141 140 138

Crystallinity (DSC) (%) 76 77 78 75 71 79 62 72 75 70 74 63 64 82 71 72 80

Bulk density (g/dm3 ) 209 298 340 378 231 316 214 nd 202 246 nd nd nd nd nd nd nd

Polym. time (60 min). Polym. time (90 min). Temperature 50 ◦ C. Numeration according to Table 2.

supporting on a solid carrier [15] as a result from interactions between the metallocene and the silica [16,25,79] or nonuniform monomer concentration within the particle [12]. In the case of the system A with Cp2 TiCl2 catalyst, the molecular weight depends on the amount of the activator (Table 3, items 2, 3, and 7). Increase of the Al/Ti molar ratio leads to a decrease of the molecular weight (Table 3, items 3). This effect is caused by the chain transfer reaction to activator [25,80,81]. Longer polymerization time also leads to considerable increase of the molecular weight, what is known for supported organometallic catalysts (Table 3, items 4, 5). In contrast, the increase of the reaction temperature up to 50 ◦ C decreases the molecular weight, but increases the molecular weight distribution (Mw /Mn = 12.6) (Table 3, item 6), as compared to the polyethylene obtained at 30 ◦ C (Mw /Mn = 3.0–3.4). It should be noted that the Cp2 TiCl2 catalyst anchored on the AlEtCl2 modified silica, but without the ionic liquid produces the polyethylene of much lower molecular weight in comparison to the analogous SIL system, regardless of the type of the applied activator (Table 3, items 12–14). The SIL systems C and D with the FI–Ti and Sal–Ti catalysts produce the PE, which is characterized by lower molecular weight, monomodal GPC elution peak, but broad distribution of molecular weight (Mw /Mn = 11.3–19.1). Interestingly, the increase of the reaction temperature leads to an increase of the molecular weight and a decrease of Mw /Mn (Table 3, items 16–21). The value of bulk density reaches up to 380 g/dm3 , what is comparable to the values for the polyethylene produced using the metallocene catalyst anchored on the solid support [14,16,27,82–84]. Moreover, longer reaction time leads to an increase of the bulk density. The presented results indicate that the properties of the polyethylene produced can be controlled both by the kind of organometallic catalyst and polyreaction conditions. Scanning electron microscope (SEM) images of the resulting polyethylene are shown in Fig. 6c and d. The polyethylene samples have the shape of porous granules, similar to those produced using the supported organometallic catalysts [85]. There is a similarity of the morphology of the PE particles to the particles of the final SIL catalytic system. This means that, as in the case of supporting catalysts, the control of the product morphology is mainly achieved by the shape replication of the support particles [86].

4. Conclusions In this work, the ionic liquid-modified silica (supported ionic liquid – SIL strategy) was used for the first time to heterogenization of metallocene and post-metallocene catalysts for polymerization of ethylene. The developed SIL systems varied in the kind of the metal complex type: Cp2 TiCl2 , Cp2 ZrCl2 , FI–Ti, or Sal–Ti. All prepared SIL systems were active in the ethylene polymerization, regardless of the kind of organometallic catalyst. The highest activity was obtained using the SIL system with the Cp2 TiCl2 complex and AlEt2 Cl activator. The produced polyethylene was linear, which was characterized by the monomodal molecular weight distribution, high melting points, crystallinity degree, and bulk density. The presented results clearly demonstrated the potential utility of the SIL systems for the ethylene polymerization, and thus, possibly other olefins. Advantages resulting from a combination of the ILs and solid support show that the SIL technology can be a promising method to obtain heterogeneous catalyst as well as to apply the metallocene and post-metallocene catalysts on industrial scale. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.apcata. 2014.07.016. References [1] B. Heurtefeu, C. Bouilhac, É. Cloutet, D. Taton, A. Deffieux, H. Cramail, Prog. Polym. Sci. 36 (2011) 89–126. [2] W. Kaminsky, Catal. Today 62 (2000) 23–34. [3] W. Kaminsky, Chemosphere 43 (2001) 33–38. [4] H. Makio, H. Terao, A. Iwashita, T. Fujita, Chem. Rev. 111 (2011) 2363–2449. [5] W. Kaminsky, A. Laban, Appl. Catal. A 222 (2000) 47–61. [6] J.A. Ewen, in: J. Scheirs, W. Kaminsky (Eds.), Metallocene-based Polyolefins, John Wiley & Sons, Chichester, 2000, pp. 3–31. [7] A.O. Patil, G.G. Hlatky, in: A.O. Patil, G.G. Hlatky (Eds.), Beyond Metallocenes: Next-Generation Polymerization Catalysts, American Chemical Society, Washington, 2003, pp. 1–11. [8] W. Kaminsky, Macromol. Chem. Phys. 197 (1996) 3907–3945. [9] J. Saito, M. Mitani, S. Matsui, Y. Tohi, H. Makio, T. Nakano, H. Tanaka, N. Kashiwa, T. Fujita, Macromol. Chem. Phys. 203 (2002) 59–65. [10] H. Terao, S. Ishii, J. Saito, S. Matsukura, M. Mitani, N. Nagai, H. Tanaka, T. Fujita, Macromolecules 39 (2006) 8584–8593. [11] J.C.W. Chien, Top. Catal. 7 (1999) 23–36. [12] J.R. Severn, J.C. Chadwick, R. Duchateau, N. Friederichs, Chem. Rev. 105 (2005) 4073–4147. [13] G. Fink, B. Steinmetz, J. Zechlin, Ch. Przybyla, B. Tesche, Chem. Rev. 100 (2000) 1377–1390. [14] G.G. Hlatky, Chem. Rev. 100 (2000) 1347–1376. [15] Y. Choi, J.B.P. Soares, Can. J. Chem. Eng. 90 (2012) 646–671. [16] V.F. Tisse, Ch. Boisson, F. Prades, T.F.L. McKenna, Chem. Eng. J. 157 (2010) 194–203. [17] V.F. Tisse, F. Prades, R. Briquel, Ch. Boisson, T.F.L. McKenna, Macromol. Chem. Phys. 211 (2010) 91–102. [18] C. Covarrubias, R. Quijada, Catal. Commun. 10 (2009) 995–1001. [19] A.J.D. de Freitas, J.H.Z. dos Santos, S.M.P. Meneghetti, M.R. Meneghetti, J. Appl. Polym. Sci. 119 (2011) 3051–3057. [20] H. Schneider, G.T. Puchta, F.A.R. Kaul, G. Raudaschl-Sieber, F. Lefebvre, G. Saggio, D. Mihalios, W.A. Herrmann, J.M. Basset, J. Mol. Catal. A: Chem. 170 (2001) 127–141. [21] H. Juvaste, E.I. Iiskola, T.T. Pakkanen, J. Organomet. Chem. 587 (1999) 38–45. [22] E.I. Iiskola, S. Timonen, T.T. Pakkanen, O. Härkki, P. Lehmus, J.V. Seppälä, Macromolecules 30 (1997) 2853–2859. [23] B.L. Moroz, N.V. Semikolenova, A.V. Nosov, V.A. Zakharov, S. Nagy, N.J. O’Reilly, J. Mol. Catal. A: Chem. 130 (1998) 121–129. ˜ [24] C. Alonso, A. Antinolo, F. Carrillo-Hermosilla, P. Carrión, A. Otero, J. Sancho, E. ˜ J. Mol. Catal. A: Chem. 220 (2004) 285–295. Villasenor, [25] Ch. Janiak, B. Rieger, Angew. Makromol. Chem. 215 (1994) 47–57. [26] K.-T. Li, C.-L. Dai, C.-W. Kuo, Catal. Commun. 8 (2007) 1209–1213. [27] V.N. Panchenko, L.G. Echevkaya, V.A. Zakharov, M.A. Matsko, Appl. Catal. A 404 (2011) 47–53. [28] A.-M. Uusitalo, T.T. Pakkanen, M. Kröger-Laukkanen, L. Niinistö, K. Hakala, S. Paavola, B. Löfgren, J. Mol. Catal. A: Chem. 160 (2000) 343–356. [29] X. Cheng, O.W. Lofthus, P.A. Deck, J. Mol. Catal. A: Chem. 212 (2004) 121–126. [30] F. Silveira, M. do Carmo Martins Alves, F.C. Stedile, S.B. Pergher, A. Rigacci, J.H. Zimnoch dos Santos, J. Mol. Catal. A: Chem. 298 (2009) 40–50.

W. Och˛edzan-Siodłak, K. Dziubek / Applied Catalysis A: General 484 (2014) 134–141 [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] [47] [48] [49] [50] [51] [52] [53] [54] [55] [56] [57] [58] [59]

T. Selvam, A. Machoke, W. Schwieger, Appl. Catal. A 445–446 (2012) 92–101. Ch.P. Mehnert, Chem. Eur. J. 11 (2005) 50–56. M.H. Valkenberg, C. de Castro, W.F. Hölderich, Green Chem. 4 (2002) 88–93. H. Olivier-Bourbigou, L. Magna, J. Mol. Catal. A: Chem. 182–183 (2002) 419–437. H. Olivier-Bourbigou, L. Magna, D. Morvan, Appl. Catal. A 373 (2010) 1–56. T. Welton, Coord. Chem. Rev. 248 (2004) 2459–2477. K. Binnemans, Chem. Rev. 107 (2007) 2592–2614. S. Chowdhury, R.S. Mohan, J.L. Scott, Tetrahedron 63 (2007) 2363–2389. J.P. Hallett, T. Welton, Chem. Rev. 111 (2011) 3508–3576. J. Dupont, F.R. de Souza, P. Suarez, Chem. Rev. 102 (2002) 3667–3692. V.I. Parvulescu, Ch. Hardacre, Chem. Rev. 107 (2007) 2615–2665. F. Van Rantwijk, R.A. Sheldon, Chem. Rev. 107 (2007) 2757–2785. H. Li, P.S. Bhadury, B. Song, S. Yang, RSC Adv. 2 (2012) 12525–12551. H.N.T. Ha, D.T. Duc, T.V. Dao, M.T. Le, A. Riisager, R. Fehrmann, Catal. Commun. 25 (2012) 136–141. A. Riisager, R. Fehrmann, M. Haumann, B.S.K. Gorle, P. Wasserscheid, Ind. Eng. Chem. Res. 44 (2005) 9853–9859. Ch. Van Doorslaer, J. Wahlen, P. Mertens, K. Binnemans, D. De Vos, Dalton Trans. 39 (2010) 8377–8390. J. Lemus, J. Palomar, M.A. Gilarranz, J.J. Rodriguez, Adsorption 17 (2011) 561–571. Ch.P. Mehnert, R.A. Cook, N.C. Dispenziere, M. Afeworki, J. Am. Chem. Soc. 124 (2002) 12932–12933. M.H. Valkenberg, C. deCastro, W.F. Hölderich, Top. Catal. 14 (2001) 139–144. K. Yamaguchi, C. Yoshida, S. Uchida, N. Mizuno, J. Am. Chem. Soc. 127 (2005) 530–531. L.-L. Lou, X. Peng, K. Yu, S. Liu, Catal. Commun. 9 (2008) 1891–1893. L.-L. Lou, K. Yu, F. Ding, W. Zhou, X. Peng, S. Liu, Tetrahedron Lett. 47 (2006) 6513–6516. J. Gao, Q.-W. Song, L.-N. He, C. Liu, Z.-Z. Yang, X. Han, X.-D. Li, Q.-C. Song, Tetrahedron 68 (2012) 3835–3842. P. Han, H. Zhang, X. Qiu, X. Ji, L. Gao, J. Mol. Catal. A: Chem. 295 (2008) 57–67. J. Sun, W. Cheng, W. Fan, Y. Wang, Z. Meng, S. Zhang, Catal. Today 148 (2009) 361–367. K.P. Boroujeni, M. Jafarinasab, Chin. Chem. Lett. 23 (2012) 1067–1070. M. Selvaraj, P.K. Sinha, K.S. Seshadri, A. Pandurangan, Appl. Catal. A 265 (2004) 75–84. S.B. Oepen, H. Gunter, Magn. Reson. Imaging 14 (1996) 993–994. M. Luhmer, J.B. d’Espinose, H. Hommel, A.P. Legrand, Magn. Reson. Imaging 14 (1996) 911–913.

141

[60] F. Wang, Z. Zhang, J. Yang, L. Wang, Y. Lin, Y. Wei, Fuel 107 (2013) 394–399. [61] Eur. Pat., 230 023 (2000). [62] A. Riisager, R. Fehrmann, M. Haumann, P. Wasserscheid, Top. Catal. 40 (2006) 91–102. ˛ [63] W. Ochedzan-Siodłak, K. Dziubek, K. Czaja, Polym. Bull. 70 (2013) 1–21. ˛ [64] W. Ochedzan-Siodłak, K. Dziubek, Chem. Eng. Process. 72 (2013) 74–81. [65] Y. Chauvin, S. Einloft, H. Olivier, Ind. Eng. Chem. Res. 34 (1995) 1149–1155. [66] D. Thiele, R.F. de Souza, J. Mol. Catal. A: Chem. 264 (2007) 293–298. [67] S.S. Sarma, S. Sivaram, Macromol. Chem. Phys. 198 (1997) 495–503. [68] J.P. Blitz, R.E. Diebel, C.A. Deakyne, J.M. Christensen, V.M. Gun’ko, J. Phys. Chem. B 109 (2005) 5667–5677. [69] M. Dötterl, H.G. Alt, ChemCatChem 4 (2012) 660–667. ˛ [70] W. Ochedzan-Siodłak, B. Sacher-Majewska, Eur. Polym. J. 43 (2007) 3688–3694. ˛ [71] W. Ochedzan-Siodłak, K. Dziubek, D. Siodłak, Eur. Polym. J. 44 (2008) 3608–3614. [72] S. Sensarma, S. Sivaram, Polym. Int. 51 (2002) 417–423. [73] S.-K. Ihm, K.-J. Chu, J.-H. Yim, in: K. Soga, M. Terano (Eds.), Catalyst Design for Tailor-Made Polyolefins, Kodansha Ltd., Tokio, 1994, pp. 299–306. [74] N. Semikolenova, V. Zakharov, Macromol. Chem. Phys. 198 (1997) 2889–2897. [75] L.M. Tosta Simplício, F. Gomes Costa, J. Soares Boaventura, E. Andrade Sales, S. Teixeira Brandão, J. Mol. Catal. A: Chem. 216 (2004) 45–50. ´ M. Białek, K. Czaja, Polimery (Wars.) 54 (2009) 712–718. [76] A. Glen, [77] M. Białek, A. Garłowska, O. Liboska, J. Polym. Sci. A: Polym. Chem. 47 (2009) 4811–4821. [78] C. Vasile, M. Pascu, in: C. Vasile, M. Pascu (Eds.), Practical Guide to Polyethylene, Rapra Technology Limited, Shrewsbury, 2005, pp. 31–88. [79] J. Moreno, R. van Grieken, A. Carrero, B. Paredes, Macromol. Symp. 302 (2011) 198–207. [80] L. D’Agnillo, J.B.P. Soares, A. Penlidis, Macromol. Chem. Phys. 199 (1998) 955–962. [81] J.-N. Pédeutour, K. Radhakrishnan, H. Cramail, A. Deffieux, Macromol. Rapid Commun. 22 (2001) 1095–1123. [82] H. Hammawa, S.E. Wanke, Polym. Int. 55 (2006) 426–434. [83] K.-J. Chu, J.B.P. Soares, A. Penlidis, J. Polym. Sci. A: Polym. Chem. 38 (2000) 1803–1810. [84] Ch.-H. Lin, Catal. Lett. 68 (2000) 63–65. [85] K.-J. Chu, J.B.P. Soares, A. Penlidis, J. Polym. Sci. A: Polym. Chem. 38 (2000) 462–468. [86] T. Diesing, G. Rojas, M. Klapper, G. Fink, K. Müllen, Angew. Chem. Int. Ed. 48 (2009) 6472–6475.