- Email: [email protected]
Combinations of 2,6-bis(imino)pyridine iron with zirconocenes in ethylene polymerization: A cyclic and differential pulse voltammetry study
G Model
ARTICLE IN PRESS
MOLCAA-9740; No. of Pages 9
Journal of Molecular Catalysis A: Chemical xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Journal of Molecular Catalysis A: Chemical journal homepage: www.elsevier.com/locate/molcata
Combinations of 2,6-bis(imino)pyridine iron with zirconocenes in ethylene polymerization: A cyclic and differential pulse voltammetry study Emanoel S. Oliveira a , Diógenes R. Gramacho a , Carlos D. Silva a,b , Fernando Silveira c , João Z. Santos d , Zênis N. Rocha a,∗ a
Institute of Chemistry, Federal University of Bahia, Campus Ondina, 40170-290 Salvador, BA, Brazil Federal Institute of Bahia, Street Emídio dos Santos, n/n, Barbalho, 40301-015 Salvador, BA, Brazil c Braskem S.A.—Innovation and Technology Center—Pilot plants, PP1, III Petrochemical Polo, Via West, Lot 05, 95853-000 Triunfo, RS, Brazil d Institute of Chemistry, Federal University of Rio Grande do Sul, Avenue Bento Gonc¸alves, 9500, 91501-970 Porto Alegre, RS, Brazil b
a r t i c l e
i n f o
Article history: Received 1 October 2015 Received in revised form 6 January 2016 Accepted 7 January 2016 Available online xxx Keywords: 2,6-Bis(imino)pyridine iron Zirconocene Ethylene polymerization Cyclic voltammetry Differential pulse voltammetry
a b s t r a c t In the present study, binary systems based on 2,6-bis(imino)pyridine iron and zirconocene complexes were evaluated in the polymerization of ethylene using different catalyst molar ratios and methylaluminoxane as the cocatalyst. The catalyst activity and polydispersity were evaluated, supported by electrochemical data of isolated and binary systems. The results indicated that in binary systems, each complex solely forms an active species. The combination effect was shown to be dependent on the nature of the complexes and on the ratio between them. For an equal molar ratio, a synergistic effect occurs on the catalyst activity. A redox reaction between the complexes was detected, which led to a decrease in the concentration of active species for the other ratios. In addition, it was found that such combined systems can produce polyethylene with a broad polydispersity with bimodal patterns and higher cristallinity than isolated systems. © 2016 Elsevier B.V. All rights reserved.
1. Introduction Polyethylenes are widely employed materials in several segments of modern society. Using some active catalyst systems and polymerization processes, these polymers can be produced by catalytic polymerization using coordination compounds with high catalytic activity, which may produce high density polymers. In this system, the growth step occurs from a single active site (metal center) [1,2]. Among such single-site catalyst systems, metallocene compounds are active in this reaction and are already established in a few industrial plants, and they can provide, in addition to the features already mentioned, control over the polymer microstructure [3]. In the coordination sphere around the metal center in zirconocenes there is a bond between the aromatic ring and Zr(IV), an weak bond between carbon (aromatic ring) and zirconium makes such compounds kinetically more active [4]. Other compounds, known as non-metallocene or post-metallocene, have
∗ Corresponding author. E-mail addresses: [email protected], [email protected] (Z.N. Rocha).
shown promising results [5]. Among them, the complex of iron(II) coordinated to the ligand 2,6-bis(imino)pyridine (FeIpy) is less toxic, more stable under oxidizing and reducing atmospheres and is easier to obtain than metallocene catalysts [6–12]. Paulino and Schuchardt [6,7] reported high catalytic activity for the FeIpy complex with Al/Fe molar ratios of 1000 and 2000. Both compound classes have in common versatility in modulating the stereo-electronic features of the structure through various substituents in the rings and bridges between the rings, which allows the polymerization conditions and the characteristics of the resulting polyethylenes to be adjusted [13,14]. The active species in the polymerization is a cationic complex stabilized by methylaluminoxane (MAO), a compound that acts as the cocatalyst [15]. Among the polyethylene properties of interest, the molar mass and polydispersity can be considered to be the most important because they affect the mechanical properties and the processability of the resin. Monodisperse polymers tend to be very viscous, and thus, a certain degree of dispersion is desirable [16]. Among the strategies adopted for this purpose [16,17], the following can be highlighted: (i) anchoring catalytic systems on solid matrices—leads to the formation of surface active sites with different reactivities, which influences the degree of polymer-
http://dx.doi.org/10.1016/j.molcata.2016.01.008 1381-1169/© 2016 Elsevier B.V. All rights reserved.
Please cite this article in press as: E.S. Oliveira, et al., J. Mol. Catal. A: Chem. (2016), http://dx.doi.org/10.1016/j.molcata.2016.01.008
G Model MOLCAA-9740; No. of Pages 9
ARTICLE IN PRESS E.S. Oliveira et al. / Journal of Molecular Catalysis A: Chemical xxx (2016) xxx–xxx
2
ization and consequently the chain size; (ii) use of dual-site catalysts—combination of different compounds in a single catalyst system. The challenge is to combine compounds with similar activities but with different preferred chain transfer reactions to produce different chain sizes. The objective of this study was to evaluate the effect of binary combinations of metallocene and non-metallocene catalysts in the polymerization of ethylene using cyclic voltammetry (CV) and differential pulse voltammetry (DPV). 2. Experimental 2.1. Materials and methods 2.1.1. Reagents All the chemicals were manipulated in inert atmosphere using Schlenk techniques. MAO (Witco, 10.0 wt.% toluene solution, average molar mass of 900 g mol−1 ) and the metallocenes [ZrCl2 (Cp)2 , [ZrCl2 (i-BuCp)2 ], [ZrCl2 (n-BuCp)2 ] and [ZrCl2 Me2 Si(Ind)2 ] were purchased from Boulder or Aldrich and were purified by recrystallization from stock solutions. Ethylene and argon (White Martins) were passed through a 13 Å molecular sieve prior to use. Toluene was purified by refluxing with sodium, followed by distillation under nitrogen immediately prior to use. Acetonitrile (MeCN) was refluxed for 6 h followed by distillation, and it was stored over a molecular sieve. Tetrabutylammonium tetrafluoroborate (TBTF), which was obtained from Merck or Fluka, was purified through successive extractions with ethyl acetate at 78 ◦ C. The crystals were collected by filtration and were vacuum dried. The reagents 2,6diacetylpyridine, 2,6-dimethylaniline, formic acid and anhydrous iron chloride were purchased from Aldrich. 2.1.2. Synthesis of Ipy ligand The 2,6-bis-[1-(2,6-dimethylphenylimino)ethyl]pyridine (Ipy) was prepared as previously described [6,18] with slight modifications. 1.00 g (6.13 mmol) of 2,6-diacetylpyridine was dissolved in deaerated methanol (10 mL), and argon was bubbled through the mixture. After dissolution, it was added under stirring to a solution of 1.8 mL (14.7 mmol) of 2,6-dimethylaniline and 5 drops of formic acid. The mixture was heated to 65 ◦ C and maintained at this temperature under stirring and bubbling argon for 4 h. Thereafter, the mixture was stirred for 20 h. The resulting yellow solid was then collected by filtration and washed with cold methanol and vacuum dried. The average yield was 80 %. 2.1.3. Synthesis of FeIpy complex The 2,6-bis-[1-(2,6-dimethylphenylimino) ethyl]pyridinedichlorideiron(II)] (FeIpy) was prepared as described in the literature [6] with slight modifications. 0.05 g (0.354 mmol) of anhydrous iron chloride was dissolved in 10 mL of n-buthanol, and argon was bubbled through the solution. The mixture was stirred at 40 ◦ C, and then 144.5 mg (0.394 mmol) of Ipy was added and continuous
argon bubbling was maintained. After 2 h, the mixture was filtered, and the resulting blue power was collected and washed with cold n-buthanol. The average yield was 68.8 %. 2.1.4. Polymerization reactions Polymerizations were performed in toluene (150 mL) in a 300 mL Pyrex glass reactor connected to a constant temperature circulator equipped with a mechanical stirrer and inlets for argon and ethylene. For each experiment, a mass of catalytic system corresponding to 10−5 mol L−1 of metal center (M = Fe or Zr) was suspended in 10 mL of toluene and transferred into the reactor under argon. The polymerizations were performed under atmospheric pressure with ethylene at 60 ◦ C for 30 min at a molar ratio of Al/M = 1000, using MAO as the cocatalyst. Ethanol acidified by HCl was used to quench the processes, and the reaction products were separated by filtration, washed with distilled water, and finally dried under reduced pressure at 60 ◦ C. 2.2. Characterization 2.2.1. Electrochemical measurements The DPV and CV measurements were performed using a Parc model 273 potentiostat/galvanostat. All experiments were conducted using a conventional three-electrode cell. Glassy carbon was used as the working electrode. An Ag/AgCl electrode was used as the reference electrode, and a platinum wire was used as the auxiliary electrode. The Ag/AgCl electrode was calibrated with ferrocene, which was used as an internal reference. These measurements versus the ferrocene (Fc+ /Fc0 ) redox couple (+0.50 V in MeCN solutions) were reported versus an Ag/AgCl electrode. Electrochemical data were obtained using 0.1 mol L−1 solutions of TBTF in MeCN as the supporting electrolyte. In the DPV or CV, neither anodic nor cathodic peaks were observed in the absence of zirconocene or FeIpy in the investigated potential range. All solutions were deaerated by bubbling high purity argon. The cocatalyst solutions were prepared with different molar ratios of Al/M = 2.5, 5, 10 and 30. DPVs were recorded with a scan rate of 100 mV s−1 , and CVs were recorded with a scan rate of 200 mV s−1 . 2.2.2. Polydispersity measurements The molar mass distribution of polyethylene was measured at 150 ◦ C using a Waters CV Plus high temperature GPC instrument equipped with a viscosimetrical detector and three Styragel HTtype columns (HT3, HT4 and HT6) with an exclusion limit of 107 for polystyrene. 1,2,4-Trichlorobenzene was used as the solvent at a flow rate of 1 cm3 min−1 . The analysis were performed at 140 ◦ C. A universal calibration method with narrow polystyrene standards was used. 2.2.3. Calorimetric measurements DSC measurements were performed to determine the fusion temperature (Tm ), crystallization temperature (Tc ) and crystallinity
Fig. 1. Molecular structures of complexes: (a) FeIpy; (b) and (c) zirconocenes, where “R” can be H, n-butyl or i-butyl and “X”, the bridge, is dimethylsilane.
Please cite this article in press as: E.S. Oliveira, et al., J. Mol. Catal. A: Chem. (2016), http://dx.doi.org/10.1016/j.molcata.2016.01.008
G Model
ARTICLE IN PRESS
MOLCAA-9740; No. of Pages 9
E.S. Oliveira et al. / Journal of Molecular Catalysis A: Chemical xxx (2016) xxx–xxx Table 1 Catalytic activity of binary systems. Zirconocene (Zr) [ZrCl2 Cp2 ] [ZrCl2 (n-BuCp2 )] [ZrCl2 (n-BuCp2 )] [ZrCl2 (n-BuCp2 )] [ZrCl2 (i-BuCp2 )] [ZrCl2 (i-BuCp2 )] [ZrCl2 (Me2 Si)(Ind)2 ] [ZrCl2 (Me2 Si)(Ind)2 ] [ZrCl2 (Me2 Si)(Ind)2 ]
Molar ratio FeIpy/Zr 1:1 1:1 1:2 2:1 1:2 2:1 1:1 1:2 2:1
3
Table 2 Redox potentials related to FeIpy and uncoordinated Ipy. Catalytic activity (kg(PE) mol−1 (Zr + Fe) h−1 ) 5816 5940 840 1264 2439 1522 2459 674 633
[Fe + Zr] = 10−5 mol L−1 ; molar ratio of Al/(Zr + Fe) = 1000; solvent = toluene; T = 60 ◦ C
(c ) of the resulting polymers. These analyses were conducted using a TA instruments DSC model 2920 calorimeter connected to a thermal analyzer integrator and calibrated with an indium pattern. The heating rate was 20 ◦ C min−1 within the temperature range of 30–150 ◦ C. The thermal cycle was performed twice, but only data from the second scan were employed to remove the thermal history of the polymers. 3. Results and discussion 3.1. Catalytic activity of the isolated and binary systems Combinations of FeIpy and zirconocene complexes (Fig. 1) were evaluated in the polymerization of ethylene with MAO. The catalytic activity for FeIpy is higher (on the order of 3000 kg(PE) mol−1 (Fe) h−1 ) than that exhibited by most of the investigated zirconocenes: from 700 for [ZrCl2 (Me2 Si)(Ind)2 ] and 2800 for [ZrCl2 (i-BuCp)2 ], except for [ZrCl2 (n-BuCp)2 ], which is approximately 4000 kg(PE) mol−1 (Fe) h−1 . To assess the behavior of mixtures of the different components in terms of catalytic activity, these properties for single systems were also determined for binary systems in different proportions, namely, 1:1, 1:2 and 2:1, as shown in Table 1. Based on the catalytic activity results, it appears that the binary systems are more active than the single catalytic systems. Thus, the catalytic activity was found to be dependent on the nature of the zirconocene and on the molar ratio between catalysts, suggesting a synergistic effect of the two systems. In a previous study [19], individual metallocenes were evaluated in the polymerization of ethylene. The catalytic activity was shown to be dependent on the nature of the ligand. This trend is also verified for the binary system of FeIpy/[ZrCl2 Cp2 ] in a 1:1 ratio in comparison with FeIpy/[ZrCl2 (nBuCp2 )] in a 1:1 ratio. The decrease in catalytic activity for binary mixtures in ratios of 1:2 and 2:1 can be attributed to electrochemical behavior that will be discussed later (see Section 3.4).
FeIpy 1a/1c 2a 3a/3c 4a/4c 5a/5c 6a/6c 8a/8c 9a/9c
Ipy +1.50 +0.80 −0.80 −1.20 −1.55 −1.80 +0.57 +0.15
1a/1c 2a 3a/3c 4a/4c 5a/5c 6a/6c 7a/7c
+1.30 +0.82 −0.75 −1.38 −1.70 −1.90 −2.10
of the change in the electronic density of Ipy promoted by the {FeCl2 } fragment. The potential values corresponding to the redox processes for the ligands and metal center are presented in Table 2. Redox processes associated with the metal center were allocated and compared to the DPV of the ligand. The redox couples 8a/8c and 9a/9c, which were absent in the DPV profile of Ipy, can be attributed to the reduction of FeIII/II (Eq. IV in Scheme 2) on the complex with the ligand in the neutral form (FeIpy0 ) (the Ipy in coordinated form is consistent with its reduction below +1.50 V (1a/1c)) and oxidized form (FeIpy+ ) (Eq. V in Scheme 2), respectively. Because measurements were performed in MeCN – a possible ligand for Fe(II) – and because the FeIpy complex is coordinatively unsaturated, the potential existence of redox processes centered on the possible product [FeCl2 (Ipy)(MeCN)] was investigated. The coordination of MeCN to Fe(II) results in a change in the electronic density around the Ipy ligand and metal center. Thus, with the coordination of solvent to Fe(II), it is possible that the signals related to both centered redox processes in the ligand and in the metal center appear positioned in potential to differ from that observed in the FeIpy complex, in which the metal center is coordinatively unsaturated. Based on the DPV profile of the uncoordinated ligand, it can be proposed that the 3a/3c pair in the complex peak around −0.8 V is related to the reduction of coordinated Ipy, as shown in Eq. VI (Scheme 3). These electrochemical results are interesting for the assignment of active catalyst species generated by the addition of MAO to the polymerization reaction. The active species have been reported to be zirconocenes such as [Cp2 ZrIV (Me)Cl]+ , a cationic species that
Scheme 1. Redox process centered on the Ipy ligand.
3.2. Electrochemical behavior of the isolated systems The DPV profiles of Ipy and FeIpy (Fig. 2) show that these compounds are electroactive on positive and negative potentials. The studies of controlled potential electrolysis indicated the involvement of three electrons, which supports the electrode process proposed in Scheme 1. Fig. 2 shows the DPV of FeIpy: one of the signals can be associated with Ipy. There are other signals that may be associated with the metal center. Conversely, for the metallocene catalytic system, studies [19,20] have shown that the same redox signals are associated with the reduction of Zr at negative potentials and the oxidation of ligands at positive potentials. The signals associated with Ipy present in the DPV of FeIpy are displaced in relation to the uncoordinated ligand. This is a result
Scheme 2. Redox process centered on metal center.
Scheme 3. Redox process centered on the coordinated Ipy ligand.
Please cite this article in press as: E.S. Oliveira, et al., J. Mol. Catal. A: Chem. (2016), http://dx.doi.org/10.1016/j.molcata.2016.01.008
G Model MOLCAA-9740; No. of Pages 9 4
ARTICLE IN PRESS E.S. Oliveira et al. / Journal of Molecular Catalysis A: Chemical xxx (2016) xxx–xxx
Fig. 2. DPVs of 1.61 × 10−3 mol L−1 Ipy (red line) and 2.66 × 10−3 mol L−1 FeIpy (black line). (a) cathodic scan and (b) anodic scan. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
is coordinatively unsaturated. Due to its electronic density, the ethylene molecule is attracted to the Zr(IV) of the active catalytic species. The charge/atomic radius ratio of Lewis acid contributes to the intensity of the electronic field, leading to an attractive interaction between them. Because ethylene is classified as a relatively soft base and Zr(IV) is a hard acid, the Zr(IV)-ethylene interaction is relatively weak. Similar to FeIpy, Fig. 3 illustrates that there is consumption of the original complex and a shift of signals in the −1.60 V to −2.10 V region, probably due to the coordination of the methyl group to Fe(II) forming cationic species.
3.3. Electrochemical behavior of the binary systems To understand the influence of each catalyst on the behavior of the combined system, the electrochemical behavior of the mixture of FeIpy and [ZrCl2 Cp2 ] (Zdc) was evaluated. Electrochemical studies of Zdc [19,20] show negative potential redox processes that are centered on the metal. By comparing these signals with the FeIpy signals, it was possible to perform an accurate assessment of the electrochemical behavior of the binary mixture (Fig. 4). By comparing the DPV profiles of the isolated complexes with that of the mixture, one can observe the presence of anodic and cathodic signals common to the mixture and to some of the com-
Please cite this article in press as: E.S. Oliveira, et al., J. Mol. Catal. A: Chem. (2016), http://dx.doi.org/10.1016/j.molcata.2016.01.008
G Model MOLCAA-9740; No. of Pages 9
ARTICLE IN PRESS E.S. Oliveira et al. / Journal of Molecular Catalysis A: Chemical xxx (2016) xxx–xxx
5
Fig. 3. (a) Successive CVs of 10−3 mol L−1 FeIpy in ethylene atmosphere at a molar ratio of Al/Fe = 10; (b) CV of 10−3 mol L−1 FeIpy for comparison.
plexes, i.e., these signals have already been assigned to cases involving Zr or Cp in the Zdc and Fe or Ipy in FeIpy. In addition, the DPV of binary mixture shows the existence of a cathodic signal around −0.42 V, which is absent in the profiles of Zdc and FeIpy alone. In the anodic scan for the binary mixture, an anodic signal around +0.1 V has a high current intensity compared to the isolated FeIpy complex. These changes in the profile suggest that chemical reactions between the complexes are occurring. To evaluate this hypothesis, the DPVs of the mixture were recorded for the freshly prepared solution as a function of time after contact (Fig. 5). The analysis of Fig. 5 shows that there are decreases in the peak current intensity of redox couples around + 0.55 V (8c), −0.80 V (3c) and −1.55 V (5c), which are assigned to isolated FeIpy associated with the reduction of FeIII/II of the starting complex (Eq. IV in Scheme 2), the reduction Ipy0/−I in complex with coordinated MeCN (Eq. VI in Scheme 3) and Ipy0/−1 in FeIpy− (Eq. VII in Scheme 3), respectively. Simultaneously, there is an increase in signal intensity 9c, which is assigned to the reduction of FeIII/II coordinated to the oxidized Ipy (Eq. V in Scheme 2). These data are coherent because the signal increase indicates that there are more species in which Ipy is oxidized and consequently fewer species containing Ipy0 and Ipy−I . This explains the decrease in the foregoing signals, which are associated with the processes described in Scheme 3. In Fig. 5, there is also the appearance of a signal at +1.0 V (10c), which has a low current intensity in the curve recorded immediately after mixing and is absent in the isolated complex. Two other signals appear in the region of +0.6 V to +0.8 V (11c), and redox processes focused on Zdc disappear. These data indicate that there is a reaction between Zdc and FeIpy, generating catalytically active species that are distinct from those generated when each catalyst acts alone. A similar behavior has been reported in binary systems based on metallocenes, in which there was the possibility of electron transfer reactions between the metallocenes that in turn provided longer stability to the active site [20] The DPV profile for the binary system composed of FeIpy and [ZrCl2 (n-BuCp)2 ] in a molar ratio of 1:1 was assessed (Fig. S1). In this binary solution, a change in current intensity centered in the redox processes equivalent to the two catalytic components occurs. The
cathodic signals of the curve recorded immediately after mixing the catalysts show that there is a profile change in the range of −1.50 V to −1.80 V, a fact that, combined with the decrease in the current intensity of the cathodic signal positioned at + 0.55 V and −0.80 V and increased signal intensity around + 0.15 V, suggests the existence of a chemical reaction between the complexes for a molar ratio of Zdc/FeIpy = 1. The proposal for the chemical reaction can also be supported by finding the change in current intensities of the anodic process. For the binary mixture composed of FeIpy and [ZrCl2 (CH3 )2 Si(Ind)2 ], comparison of the DPV profiles between individual complexes and those recorded immediately after mixing shows only signals involving the isolated systems. However, the curves recorded later show a change that is the result of a chemical reaction between the different complexes. In all cases, it is common that increasing the current intensity on the redox centered around +0.15 V – attributed to the redox process of the FeIII/II ligand coordinated to the oxidized Ipy – may lead to assigning a redox reaction between complexes, in which the zirconocene is reduced and the complex containing Ipy is oxidized. 3.4. Relationship between electrochemical behavior and catalyst activity The results presented here have shown the possible reaction between the components of the binary mixture. Based on the proposal of coordination from MeCN to Fe(II), which is evident in the binary mixture, it can be proposed that in the polymerization reaction whose experimental condition is performed using a noncoordinating solvent, the monomer ethylene may be coordinated to the metal ion of the FeIpy catalyst in its active form and the active species of the Zdc. This proposal justifies the fact that the binary mixture is formed when present at a molar ratio of Zdc/FeIpy = 1 with an increased catalytic activity compared to the isolated catalysts. Therefore, the electrochemical data suggest that there is a possibility that the binary mixture in the ethylene polymerization reaction results in a number of different catalytic species of unit systems. Note that in the polymerization conditions, the FeIpy complex interacts with ethylene, just as it occurs with zirconocene,
Please cite this article in press as: E.S. Oliveira, et al., J. Mol. Catal. A: Chem. (2016), http://dx.doi.org/10.1016/j.molcata.2016.01.008
G Model MOLCAA-9740; No. of Pages 9 6
ARTICLE IN PRESS E.S. Oliveira et al. / Journal of Molecular Catalysis A: Chemical xxx (2016) xxx–xxx
Fig. 4. DPVs of Zdc, FeIpy and binary mixture with a molar ratio of FeIpy/Zdc = 1; (a) cathodic scan and (b) anodic scan.
and this step is important for the formation of the polymer product. Based on the data obtained and evaluated, it can be concluded that the combination of catalysts in a 1:1 ratio generated more active systems than the FeIpy complex, which should result from the interaction observed by DPV (Fig. 4). However, for the binary mixtures in ratios of 1:2 or 2:1, the presented activities are lower compared to those of the isolated systems, which is due to the use of catalytic systems based on electron transfer between the systems. Thus, the amount of catalytically active species interacting with the monomer at a given time is minimized. For the molar ratio of Zdc/FeIpy = 2, the change in the DVP profile compared with the molar ratio of 1:1 was more complex (Fig. S2). Note that in this case, the signal of 8a/8c (Eq. IV in Scheme 2) has an increased current density. Thus, excess Zdc appears to induce redox processes at the electrode surface. Therefore, the electron transfer
reaction between the two complexes appears to be more favored, thereby changes in the catalytic activity. 3.5. Relationship between electrochemical behavior and polyethylene properties To verify the contribution effect of the FeIpy/Zr ratio on the molar mass of the polymers produced with the combination of FeIpy and [ZrCl2 (n-BuCp2 )], GPC analysis of the resulting polymers was performed. The chromatograms were also deconvoluted according to Flory [21] as shown in Fig. 6. In the deconvoluted chromatograms shown in Fig. 6, the bold lines represent the experimental data, and the other lines are the corresponding polymers with lower and higher molar masses. The polymers obtained using the FeIpy complex (Fig. 6a) resulted in
Please cite this article in press as: E.S. Oliveira, et al., J. Mol. Catal. A: Chem. (2016), http://dx.doi.org/10.1016/j.molcata.2016.01.008
G Model MOLCAA-9740; No. of Pages 9
ARTICLE IN PRESS E.S. Oliveira et al. / Journal of Molecular Catalysis A: Chemical xxx (2016) xxx–xxx
7
Fig. 5. DPVs of binary mixture with a molar ratio of FeIpy/Zdc = 1: (1) Immediately after mixing, (2) two minutes after mixing and (3) five minutes after mixing.
Fig. 6. GPC polymer chromatograms and signal deconvolution. Polymers produced by [ZrCl2 (n-BuCp)2 ] (Zr) (a), FeIpy (b) and FeIpy/Zr systems in molar ratios of 1:1 (c), 1:2 (d) and 2:1(e). Polymers produced by isolated systems are presented for comparison.
Please cite this article in press as: E.S. Oliveira, et al., J. Mol. Catal. A: Chem. (2016), http://dx.doi.org/10.1016/j.molcata.2016.01.008
G Model MOLCAA-9740; No. of Pages 9 8
ARTICLE IN PRESS E.S. Oliveira et al. / Journal of Molecular Catalysis A: Chemical xxx (2016) xxx–xxx
Fig. 7. DSC profiles of the resulting polymers with the catalytic system FeIpy/Zr in 1:1 ratio: left (Tc ) and right (Tm ).
a polymer with a molecular weight of 16,376 Da. The polymers obtained using zirconocene species, [ZrCl2 (n-BuCp)2 ], (Fig. 6b) had a molecular weight of 208,035 Da. Comparing the chromatographic profiles generated by the isolated species, it is possible, using the deconvolution methodology, to determine which species are responsible for the production of low and high molar mass polymers. The combined system with the 1:1 FeIpy/Zr ratio (Fig. 6c) showed a higher catalytic activity compared to the other combined species, as previously shown in Table 1. The resulting polymers have a molecular weight of 76,682 Da and a broad polydispersion: the Mw/Mn (weight average molecular weight/number average molecular weight) is on the order of 20 compared to the isolated species. The low molar mass fraction appears to be consistent with the displacements and molar mass of the components that compose the polymer produced by the FeIpy species. The high molar mass fraction also appears coherent, according to the molar mass and displacements of the zirconocene catalyst. The ratio of the low and high mass polymers are 45% for FeIpy and 40% for the zirconocene, and the remaining 15% of species appear to correspond to the low mass fraction of the zirconocene species. An analysis that considers the relative fraction of each species shows that the resulting molar mass is not a simple combination of polymers produced independently by the two species; rather, it suggests some interaction between them, as indicated by the combined species voltammograms shown in Fig. 4, if compared to the isolated species. Equally, the broadening of the polydispersity may be associated with this interaction and, of course, the bimodality observed in the chromatogram. In the case of the 1:2 (Fig. 6d) and 2:1 ratio (Fig. 6e) systems, there is a decrease in the catalytic activity in relation to the isolated species. In the 1:2 ratio system, the molecular weight reaches 91,000 Da with a polydispersity of 11.7. In terms of fraction, the 1:2 ratio system shows one less fraction (4), which explains the lowest polydispersity among the three combined systems. The 2:1 ratio resulted in the highest molecular weight, 101,210 Da, and a polydispersity of 27.
3.6. Relationship between DSC measurements and polyethylene properties The contribution of each species on the crystallization temperature (Tc ), fusion temperature (Tm ) and crystallinity (c ) can be evaluated through DSC analysis. The Tm and c are in agreement with PEAD, confirming the production of this material. The resulting polymers from the binary systems exhibit properties that are
Table 3 Crystallization temperature (Tc ), fusion temperature (Tm ) and crystallinity (c ) of the polymers produced using isolated and binary systems. Polymer sample
Tc (◦ C)
Tm (◦ C)
c (%)
FeIpy [ZrCl2 (n-BuCp2 )] 1:1 FeIpy/Zr 1:2 FeIpy/Zr 2:1 FeIpy/Zr
108.8 109.0 100.1 107.5 108.4
135.1 141.1 144.0 138.2 137.3
83.2 59.3 78.8 76.0 85.3
Zr = [ZrCl2 (n-BuCp2 )].
compatible with those corresponding to the isolated systems, as shown in Table 3. The combined species in the 1:2 and 2:1 fractions present similarities in their crystallization and fusion temperatures, differing from each other by the sample resulting from the combination of the 2:1 fraction having a higher crystallinity, obeying the trend shown by the isolated species, although with a higher crystallinity value compared to that of the isolated species FeIpy. For the sample resulting from the 1:1 combination, one can observe lower Tc values and higher Tm values (see Fig. 7). The crystallization and fusion temperatures, associated with a high polydispersity and the several components that compose the molar mass distribution (Fig. 6c), confirm that the catalytic system combined in the 1:1 ratio possesses a higher perturbation resulting from the higher electronic interaction between the species, as shown by the DPV of this combination (Fig. 4).
4. Conclusions The electrochemical analysis enabled a detailed and accurate assessment of the behavior of binary catalytic systems formed by FeIpy and zirconocenes. When combined in the presence of MAO, each complex separately forms the catalytically active species for the polymerization of ethylene. It can be observed that the effect of the mixture depends on the nature of the zirconocene used and its proportion in the combined systems. Changes in the electrochemical signals of the binary mixtures, compared to those of the isolated complex, suggest that there is interaction between them. An equal ratio has an increased number of catalytically active species, which leads to increased activity. The occurrence of redox reactions between different complexes was observed, in which the metallocene is reduced and the complex containing Ipy is oxidized. This fact may justify the lower activity observed for the other ratios. Thus, an increase in the concentration of oxidized Ipy necessarily means a decrease in the concentration
Please cite this article in press as: E.S. Oliveira, et al., J. Mol. Catal. A: Chem. (2016), http://dx.doi.org/10.1016/j.molcata.2016.01.008
G Model MOLCAA-9740; No. of Pages 9
ARTICLE IN PRESS E.S. Oliveira et al. / Journal of Molecular Catalysis A: Chemical xxx (2016) xxx–xxx
of the reduced form of Ipy species that are part of the catalytically active species. GPC analysis showed that the polymer bearing bimodality varied between 4 and 27, suggesting comparable performance of both catalytic sites. Higher crystallinity values for the polymers obtained with binary systems probably reflect better mechanical properties. Binary mixtures of FeIpy and zirconocene 1:1 were more promising active systems than the isolated catalytic systems because it has a significant increase in catalytic activity and extension of the molar mass distribution in the polymeric chains, which facilitates the processing of the resulting polyethylene. Acknowledgements We thank the Brazilian agencie CNPq. The authors thank Teacher Ulf Schuchardt for his great contribution and help in catalysis olefin polymerization. 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.molcata.2016.01. 008.
9
References [1] F.M.B. Coutinho, I.L. Mello, L.C. Santa Maria, Polímeros 13 (2003) 1–13. [2] E.I.S. Oliveira, F. Silveira, A.M.V. Viveiros, N.R.S. Basso, Z.N. Rocha, J.H.Z. Santos, Quim. Nova 38 (2015) 97–105. [3] G.G. Hlatky, Coord. Chem. Rev. 181 (1999) 243–296. [4] H.G. Alt, A. Köppel, Chem. Rev. 100 (2000) 1205–1222. [5] V.C. Gibson, S.K. Spitzmesser, Chem. Rev. 103 (2003) 283–315. [6] I.S. Paulino, U. Schuchardt, J. Mol. Catal. A: Chem. 2011 (2004) 55–58. [7] I.S. Paulino, U. Schuchardt, Catal. Commun. 5 (2004) 5–7. [8] Z. Zhang, S. Chen, X. Zhang, H. Li, Y. Ke, Y. Lu, Y. Hu, J. Mol. Catal. A: Chem. 230 (2005) 1–8. [9] A.S. Abu-Surrah, K. Lappalainen, U. Piironen, P. Lehmus, T. Repo, M. Leskela, J. Organomet. Chem. 648 (2002) 55–61. [10] M.E. Bluhm, C. Folli, M. Doring, J. Mol. Catal. A: Chem. 212 (2004) 13–18. [11] N.V. Semikolenova, V.A. Zakharov, L.G. Echevckaja, M.A. Matsko, K.P. Bryliakov, E.P. Talsi, Catal. Today 144 (2009) 334–340. [12] S.S. Ivanchev, A.V. Yakimansky, D.G. Rogozin, Polymer 45 (2004) 6453–6459. [13] P. Mohring, N.J.G. Coville, Coord. Chem. Rev. 250 (2006) 18–35. [14] B. Wang, Coord. Chem. 250 (2006) 242–258. [15] H. Sinn, W. Kaminsky, Adv. Organomet. Chem. 18 (1980) 99. [16] A.G. Fisch, N.S.M. Cardozo, A.R. Secchi, J.H.Z. Santos, Quim. Nova 34 (2011) 646–657. [17] J.H.Z. Santos, A.G. Fisch, N.S.M. Cardozo, A.R. Secchi, Quim. Nova 31 (2008) 1199–1207. [18] R. Souane, F. Isel, F. Peruch, P.J. Lutz, Coord. Rev. Chim. 5 (2002) 43–48. [19] F. Silveira, L.M.T. Simplício, Z.N. Rocha, J.H.Z. Santos, Appl. Catal. A 344 (2008) 98–106. [20] F. Silveira, L.M.T. Simplício, Z.N. Rocha, J.H.Z. Santos, Macromol. React. Eng. 2 (2008) 253–264. [21] P.J. Flory, Principles of Polymer Chemistry, Cornell University, New York, 1953.
Please cite this article in press as: E.S. Oliveira, et al., J. Mol. Catal. A: Chem. (2016), http://dx.doi.org/10.1016/j.molcata.2016.01.008
ARTICLE IN PRESS
MOLCAA-9740; No. of Pages 9
Journal of Molecular Catalysis A: Chemical xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Journal of Molecular Catalysis A: Chemical journal homepage: www.elsevier.com/locate/molcata
Combinations of 2,6-bis(imino)pyridine iron with zirconocenes in ethylene polymerization: A cyclic and differential pulse voltammetry study Emanoel S. Oliveira a , Diógenes R. Gramacho a , Carlos D. Silva a,b , Fernando Silveira c , João Z. Santos d , Zênis N. Rocha a,∗ a
Institute of Chemistry, Federal University of Bahia, Campus Ondina, 40170-290 Salvador, BA, Brazil Federal Institute of Bahia, Street Emídio dos Santos, n/n, Barbalho, 40301-015 Salvador, BA, Brazil c Braskem S.A.—Innovation and Technology Center—Pilot plants, PP1, III Petrochemical Polo, Via West, Lot 05, 95853-000 Triunfo, RS, Brazil d Institute of Chemistry, Federal University of Rio Grande do Sul, Avenue Bento Gonc¸alves, 9500, 91501-970 Porto Alegre, RS, Brazil b
a r t i c l e
i n f o
Article history: Received 1 October 2015 Received in revised form 6 January 2016 Accepted 7 January 2016 Available online xxx Keywords: 2,6-Bis(imino)pyridine iron Zirconocene Ethylene polymerization Cyclic voltammetry Differential pulse voltammetry
a b s t r a c t In the present study, binary systems based on 2,6-bis(imino)pyridine iron and zirconocene complexes were evaluated in the polymerization of ethylene using different catalyst molar ratios and methylaluminoxane as the cocatalyst. The catalyst activity and polydispersity were evaluated, supported by electrochemical data of isolated and binary systems. The results indicated that in binary systems, each complex solely forms an active species. The combination effect was shown to be dependent on the nature of the complexes and on the ratio between them. For an equal molar ratio, a synergistic effect occurs on the catalyst activity. A redox reaction between the complexes was detected, which led to a decrease in the concentration of active species for the other ratios. In addition, it was found that such combined systems can produce polyethylene with a broad polydispersity with bimodal patterns and higher cristallinity than isolated systems. © 2016 Elsevier B.V. All rights reserved.
1. Introduction Polyethylenes are widely employed materials in several segments of modern society. Using some active catalyst systems and polymerization processes, these polymers can be produced by catalytic polymerization using coordination compounds with high catalytic activity, which may produce high density polymers. In this system, the growth step occurs from a single active site (metal center) [1,2]. Among such single-site catalyst systems, metallocene compounds are active in this reaction and are already established in a few industrial plants, and they can provide, in addition to the features already mentioned, control over the polymer microstructure [3]. In the coordination sphere around the metal center in zirconocenes there is a bond between the aromatic ring and Zr(IV), an weak bond between carbon (aromatic ring) and zirconium makes such compounds kinetically more active [4]. Other compounds, known as non-metallocene or post-metallocene, have
∗ Corresponding author. E-mail addresses: [email protected], [email protected] (Z.N. Rocha).
shown promising results [5]. Among them, the complex of iron(II) coordinated to the ligand 2,6-bis(imino)pyridine (FeIpy) is less toxic, more stable under oxidizing and reducing atmospheres and is easier to obtain than metallocene catalysts [6–12]. Paulino and Schuchardt [6,7] reported high catalytic activity for the FeIpy complex with Al/Fe molar ratios of 1000 and 2000. Both compound classes have in common versatility in modulating the stereo-electronic features of the structure through various substituents in the rings and bridges between the rings, which allows the polymerization conditions and the characteristics of the resulting polyethylenes to be adjusted [13,14]. The active species in the polymerization is a cationic complex stabilized by methylaluminoxane (MAO), a compound that acts as the cocatalyst [15]. Among the polyethylene properties of interest, the molar mass and polydispersity can be considered to be the most important because they affect the mechanical properties and the processability of the resin. Monodisperse polymers tend to be very viscous, and thus, a certain degree of dispersion is desirable [16]. Among the strategies adopted for this purpose [16,17], the following can be highlighted: (i) anchoring catalytic systems on solid matrices—leads to the formation of surface active sites with different reactivities, which influences the degree of polymer-
http://dx.doi.org/10.1016/j.molcata.2016.01.008 1381-1169/© 2016 Elsevier B.V. All rights reserved.
Please cite this article in press as: E.S. Oliveira, et al., J. Mol. Catal. A: Chem. (2016), http://dx.doi.org/10.1016/j.molcata.2016.01.008
G Model MOLCAA-9740; No. of Pages 9
ARTICLE IN PRESS E.S. Oliveira et al. / Journal of Molecular Catalysis A: Chemical xxx (2016) xxx–xxx
2
ization and consequently the chain size; (ii) use of dual-site catalysts—combination of different compounds in a single catalyst system. The challenge is to combine compounds with similar activities but with different preferred chain transfer reactions to produce different chain sizes. The objective of this study was to evaluate the effect of binary combinations of metallocene and non-metallocene catalysts in the polymerization of ethylene using cyclic voltammetry (CV) and differential pulse voltammetry (DPV). 2. Experimental 2.1. Materials and methods 2.1.1. Reagents All the chemicals were manipulated in inert atmosphere using Schlenk techniques. MAO (Witco, 10.0 wt.% toluene solution, average molar mass of 900 g mol−1 ) and the metallocenes [ZrCl2 (Cp)2 , [ZrCl2 (i-BuCp)2 ], [ZrCl2 (n-BuCp)2 ] and [ZrCl2 Me2 Si(Ind)2 ] were purchased from Boulder or Aldrich and were purified by recrystallization from stock solutions. Ethylene and argon (White Martins) were passed through a 13 Å molecular sieve prior to use. Toluene was purified by refluxing with sodium, followed by distillation under nitrogen immediately prior to use. Acetonitrile (MeCN) was refluxed for 6 h followed by distillation, and it was stored over a molecular sieve. Tetrabutylammonium tetrafluoroborate (TBTF), which was obtained from Merck or Fluka, was purified through successive extractions with ethyl acetate at 78 ◦ C. The crystals were collected by filtration and were vacuum dried. The reagents 2,6diacetylpyridine, 2,6-dimethylaniline, formic acid and anhydrous iron chloride were purchased from Aldrich. 2.1.2. Synthesis of Ipy ligand The 2,6-bis-[1-(2,6-dimethylphenylimino)ethyl]pyridine (Ipy) was prepared as previously described [6,18] with slight modifications. 1.00 g (6.13 mmol) of 2,6-diacetylpyridine was dissolved in deaerated methanol (10 mL), and argon was bubbled through the mixture. After dissolution, it was added under stirring to a solution of 1.8 mL (14.7 mmol) of 2,6-dimethylaniline and 5 drops of formic acid. The mixture was heated to 65 ◦ C and maintained at this temperature under stirring and bubbling argon for 4 h. Thereafter, the mixture was stirred for 20 h. The resulting yellow solid was then collected by filtration and washed with cold methanol and vacuum dried. The average yield was 80 %. 2.1.3. Synthesis of FeIpy complex The 2,6-bis-[1-(2,6-dimethylphenylimino) ethyl]pyridinedichlorideiron(II)] (FeIpy) was prepared as described in the literature [6] with slight modifications. 0.05 g (0.354 mmol) of anhydrous iron chloride was dissolved in 10 mL of n-buthanol, and argon was bubbled through the solution. The mixture was stirred at 40 ◦ C, and then 144.5 mg (0.394 mmol) of Ipy was added and continuous
argon bubbling was maintained. After 2 h, the mixture was filtered, and the resulting blue power was collected and washed with cold n-buthanol. The average yield was 68.8 %. 2.1.4. Polymerization reactions Polymerizations were performed in toluene (150 mL) in a 300 mL Pyrex glass reactor connected to a constant temperature circulator equipped with a mechanical stirrer and inlets for argon and ethylene. For each experiment, a mass of catalytic system corresponding to 10−5 mol L−1 of metal center (M = Fe or Zr) was suspended in 10 mL of toluene and transferred into the reactor under argon. The polymerizations were performed under atmospheric pressure with ethylene at 60 ◦ C for 30 min at a molar ratio of Al/M = 1000, using MAO as the cocatalyst. Ethanol acidified by HCl was used to quench the processes, and the reaction products were separated by filtration, washed with distilled water, and finally dried under reduced pressure at 60 ◦ C. 2.2. Characterization 2.2.1. Electrochemical measurements The DPV and CV measurements were performed using a Parc model 273 potentiostat/galvanostat. All experiments were conducted using a conventional three-electrode cell. Glassy carbon was used as the working electrode. An Ag/AgCl electrode was used as the reference electrode, and a platinum wire was used as the auxiliary electrode. The Ag/AgCl electrode was calibrated with ferrocene, which was used as an internal reference. These measurements versus the ferrocene (Fc+ /Fc0 ) redox couple (+0.50 V in MeCN solutions) were reported versus an Ag/AgCl electrode. Electrochemical data were obtained using 0.1 mol L−1 solutions of TBTF in MeCN as the supporting electrolyte. In the DPV or CV, neither anodic nor cathodic peaks were observed in the absence of zirconocene or FeIpy in the investigated potential range. All solutions were deaerated by bubbling high purity argon. The cocatalyst solutions were prepared with different molar ratios of Al/M = 2.5, 5, 10 and 30. DPVs were recorded with a scan rate of 100 mV s−1 , and CVs were recorded with a scan rate of 200 mV s−1 . 2.2.2. Polydispersity measurements The molar mass distribution of polyethylene was measured at 150 ◦ C using a Waters CV Plus high temperature GPC instrument equipped with a viscosimetrical detector and three Styragel HTtype columns (HT3, HT4 and HT6) with an exclusion limit of 107 for polystyrene. 1,2,4-Trichlorobenzene was used as the solvent at a flow rate of 1 cm3 min−1 . The analysis were performed at 140 ◦ C. A universal calibration method with narrow polystyrene standards was used. 2.2.3. Calorimetric measurements DSC measurements were performed to determine the fusion temperature (Tm ), crystallization temperature (Tc ) and crystallinity
Fig. 1. Molecular structures of complexes: (a) FeIpy; (b) and (c) zirconocenes, where “R” can be H, n-butyl or i-butyl and “X”, the bridge, is dimethylsilane.
Please cite this article in press as: E.S. Oliveira, et al., J. Mol. Catal. A: Chem. (2016), http://dx.doi.org/10.1016/j.molcata.2016.01.008
G Model
ARTICLE IN PRESS
MOLCAA-9740; No. of Pages 9
E.S. Oliveira et al. / Journal of Molecular Catalysis A: Chemical xxx (2016) xxx–xxx Table 1 Catalytic activity of binary systems. Zirconocene (Zr) [ZrCl2 Cp2 ] [ZrCl2 (n-BuCp2 )] [ZrCl2 (n-BuCp2 )] [ZrCl2 (n-BuCp2 )] [ZrCl2 (i-BuCp2 )] [ZrCl2 (i-BuCp2 )] [ZrCl2 (Me2 Si)(Ind)2 ] [ZrCl2 (Me2 Si)(Ind)2 ] [ZrCl2 (Me2 Si)(Ind)2 ]
Molar ratio FeIpy/Zr 1:1 1:1 1:2 2:1 1:2 2:1 1:1 1:2 2:1
3
Table 2 Redox potentials related to FeIpy and uncoordinated Ipy. Catalytic activity (kg(PE) mol−1 (Zr + Fe) h−1 ) 5816 5940 840 1264 2439 1522 2459 674 633
[Fe + Zr] = 10−5 mol L−1 ; molar ratio of Al/(Zr + Fe) = 1000; solvent = toluene; T = 60 ◦ C
(c ) of the resulting polymers. These analyses were conducted using a TA instruments DSC model 2920 calorimeter connected to a thermal analyzer integrator and calibrated with an indium pattern. The heating rate was 20 ◦ C min−1 within the temperature range of 30–150 ◦ C. The thermal cycle was performed twice, but only data from the second scan were employed to remove the thermal history of the polymers. 3. Results and discussion 3.1. Catalytic activity of the isolated and binary systems Combinations of FeIpy and zirconocene complexes (Fig. 1) were evaluated in the polymerization of ethylene with MAO. The catalytic activity for FeIpy is higher (on the order of 3000 kg(PE) mol−1 (Fe) h−1 ) than that exhibited by most of the investigated zirconocenes: from 700 for [ZrCl2 (Me2 Si)(Ind)2 ] and 2800 for [ZrCl2 (i-BuCp)2 ], except for [ZrCl2 (n-BuCp)2 ], which is approximately 4000 kg(PE) mol−1 (Fe) h−1 . To assess the behavior of mixtures of the different components in terms of catalytic activity, these properties for single systems were also determined for binary systems in different proportions, namely, 1:1, 1:2 and 2:1, as shown in Table 1. Based on the catalytic activity results, it appears that the binary systems are more active than the single catalytic systems. Thus, the catalytic activity was found to be dependent on the nature of the zirconocene and on the molar ratio between catalysts, suggesting a synergistic effect of the two systems. In a previous study [19], individual metallocenes were evaluated in the polymerization of ethylene. The catalytic activity was shown to be dependent on the nature of the ligand. This trend is also verified for the binary system of FeIpy/[ZrCl2 Cp2 ] in a 1:1 ratio in comparison with FeIpy/[ZrCl2 (nBuCp2 )] in a 1:1 ratio. The decrease in catalytic activity for binary mixtures in ratios of 1:2 and 2:1 can be attributed to electrochemical behavior that will be discussed later (see Section 3.4).
FeIpy 1a/1c 2a 3a/3c 4a/4c 5a/5c 6a/6c 8a/8c 9a/9c
Ipy +1.50 +0.80 −0.80 −1.20 −1.55 −1.80 +0.57 +0.15
1a/1c 2a 3a/3c 4a/4c 5a/5c 6a/6c 7a/7c
+1.30 +0.82 −0.75 −1.38 −1.70 −1.90 −2.10
of the change in the electronic density of Ipy promoted by the {FeCl2 } fragment. The potential values corresponding to the redox processes for the ligands and metal center are presented in Table 2. Redox processes associated with the metal center were allocated and compared to the DPV of the ligand. The redox couples 8a/8c and 9a/9c, which were absent in the DPV profile of Ipy, can be attributed to the reduction of FeIII/II (Eq. IV in Scheme 2) on the complex with the ligand in the neutral form (FeIpy0 ) (the Ipy in coordinated form is consistent with its reduction below +1.50 V (1a/1c)) and oxidized form (FeIpy+ ) (Eq. V in Scheme 2), respectively. Because measurements were performed in MeCN – a possible ligand for Fe(II) – and because the FeIpy complex is coordinatively unsaturated, the potential existence of redox processes centered on the possible product [FeCl2 (Ipy)(MeCN)] was investigated. The coordination of MeCN to Fe(II) results in a change in the electronic density around the Ipy ligand and metal center. Thus, with the coordination of solvent to Fe(II), it is possible that the signals related to both centered redox processes in the ligand and in the metal center appear positioned in potential to differ from that observed in the FeIpy complex, in which the metal center is coordinatively unsaturated. Based on the DPV profile of the uncoordinated ligand, it can be proposed that the 3a/3c pair in the complex peak around −0.8 V is related to the reduction of coordinated Ipy, as shown in Eq. VI (Scheme 3). These electrochemical results are interesting for the assignment of active catalyst species generated by the addition of MAO to the polymerization reaction. The active species have been reported to be zirconocenes such as [Cp2 ZrIV (Me)Cl]+ , a cationic species that
Scheme 1. Redox process centered on the Ipy ligand.
3.2. Electrochemical behavior of the isolated systems The DPV profiles of Ipy and FeIpy (Fig. 2) show that these compounds are electroactive on positive and negative potentials. The studies of controlled potential electrolysis indicated the involvement of three electrons, which supports the electrode process proposed in Scheme 1. Fig. 2 shows the DPV of FeIpy: one of the signals can be associated with Ipy. There are other signals that may be associated with the metal center. Conversely, for the metallocene catalytic system, studies [19,20] have shown that the same redox signals are associated with the reduction of Zr at negative potentials and the oxidation of ligands at positive potentials. The signals associated with Ipy present in the DPV of FeIpy are displaced in relation to the uncoordinated ligand. This is a result
Scheme 2. Redox process centered on metal center.
Scheme 3. Redox process centered on the coordinated Ipy ligand.
Please cite this article in press as: E.S. Oliveira, et al., J. Mol. Catal. A: Chem. (2016), http://dx.doi.org/10.1016/j.molcata.2016.01.008
G Model MOLCAA-9740; No. of Pages 9 4
ARTICLE IN PRESS E.S. Oliveira et al. / Journal of Molecular Catalysis A: Chemical xxx (2016) xxx–xxx
Fig. 2. DPVs of 1.61 × 10−3 mol L−1 Ipy (red line) and 2.66 × 10−3 mol L−1 FeIpy (black line). (a) cathodic scan and (b) anodic scan. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
is coordinatively unsaturated. Due to its electronic density, the ethylene molecule is attracted to the Zr(IV) of the active catalytic species. The charge/atomic radius ratio of Lewis acid contributes to the intensity of the electronic field, leading to an attractive interaction between them. Because ethylene is classified as a relatively soft base and Zr(IV) is a hard acid, the Zr(IV)-ethylene interaction is relatively weak. Similar to FeIpy, Fig. 3 illustrates that there is consumption of the original complex and a shift of signals in the −1.60 V to −2.10 V region, probably due to the coordination of the methyl group to Fe(II) forming cationic species.
3.3. Electrochemical behavior of the binary systems To understand the influence of each catalyst on the behavior of the combined system, the electrochemical behavior of the mixture of FeIpy and [ZrCl2 Cp2 ] (Zdc) was evaluated. Electrochemical studies of Zdc [19,20] show negative potential redox processes that are centered on the metal. By comparing these signals with the FeIpy signals, it was possible to perform an accurate assessment of the electrochemical behavior of the binary mixture (Fig. 4). By comparing the DPV profiles of the isolated complexes with that of the mixture, one can observe the presence of anodic and cathodic signals common to the mixture and to some of the com-
Please cite this article in press as: E.S. Oliveira, et al., J. Mol. Catal. A: Chem. (2016), http://dx.doi.org/10.1016/j.molcata.2016.01.008
G Model MOLCAA-9740; No. of Pages 9
ARTICLE IN PRESS E.S. Oliveira et al. / Journal of Molecular Catalysis A: Chemical xxx (2016) xxx–xxx
5
Fig. 3. (a) Successive CVs of 10−3 mol L−1 FeIpy in ethylene atmosphere at a molar ratio of Al/Fe = 10; (b) CV of 10−3 mol L−1 FeIpy for comparison.
plexes, i.e., these signals have already been assigned to cases involving Zr or Cp in the Zdc and Fe or Ipy in FeIpy. In addition, the DPV of binary mixture shows the existence of a cathodic signal around −0.42 V, which is absent in the profiles of Zdc and FeIpy alone. In the anodic scan for the binary mixture, an anodic signal around +0.1 V has a high current intensity compared to the isolated FeIpy complex. These changes in the profile suggest that chemical reactions between the complexes are occurring. To evaluate this hypothesis, the DPVs of the mixture were recorded for the freshly prepared solution as a function of time after contact (Fig. 5). The analysis of Fig. 5 shows that there are decreases in the peak current intensity of redox couples around + 0.55 V (8c), −0.80 V (3c) and −1.55 V (5c), which are assigned to isolated FeIpy associated with the reduction of FeIII/II of the starting complex (Eq. IV in Scheme 2), the reduction Ipy0/−I in complex with coordinated MeCN (Eq. VI in Scheme 3) and Ipy0/−1 in FeIpy− (Eq. VII in Scheme 3), respectively. Simultaneously, there is an increase in signal intensity 9c, which is assigned to the reduction of FeIII/II coordinated to the oxidized Ipy (Eq. V in Scheme 2). These data are coherent because the signal increase indicates that there are more species in which Ipy is oxidized and consequently fewer species containing Ipy0 and Ipy−I . This explains the decrease in the foregoing signals, which are associated with the processes described in Scheme 3. In Fig. 5, there is also the appearance of a signal at +1.0 V (10c), which has a low current intensity in the curve recorded immediately after mixing and is absent in the isolated complex. Two other signals appear in the region of +0.6 V to +0.8 V (11c), and redox processes focused on Zdc disappear. These data indicate that there is a reaction between Zdc and FeIpy, generating catalytically active species that are distinct from those generated when each catalyst acts alone. A similar behavior has been reported in binary systems based on metallocenes, in which there was the possibility of electron transfer reactions between the metallocenes that in turn provided longer stability to the active site [20] The DPV profile for the binary system composed of FeIpy and [ZrCl2 (n-BuCp)2 ] in a molar ratio of 1:1 was assessed (Fig. S1). In this binary solution, a change in current intensity centered in the redox processes equivalent to the two catalytic components occurs. The
cathodic signals of the curve recorded immediately after mixing the catalysts show that there is a profile change in the range of −1.50 V to −1.80 V, a fact that, combined with the decrease in the current intensity of the cathodic signal positioned at + 0.55 V and −0.80 V and increased signal intensity around + 0.15 V, suggests the existence of a chemical reaction between the complexes for a molar ratio of Zdc/FeIpy = 1. The proposal for the chemical reaction can also be supported by finding the change in current intensities of the anodic process. For the binary mixture composed of FeIpy and [ZrCl2 (CH3 )2 Si(Ind)2 ], comparison of the DPV profiles between individual complexes and those recorded immediately after mixing shows only signals involving the isolated systems. However, the curves recorded later show a change that is the result of a chemical reaction between the different complexes. In all cases, it is common that increasing the current intensity on the redox centered around +0.15 V – attributed to the redox process of the FeIII/II ligand coordinated to the oxidized Ipy – may lead to assigning a redox reaction between complexes, in which the zirconocene is reduced and the complex containing Ipy is oxidized. 3.4. Relationship between electrochemical behavior and catalyst activity The results presented here have shown the possible reaction between the components of the binary mixture. Based on the proposal of coordination from MeCN to Fe(II), which is evident in the binary mixture, it can be proposed that in the polymerization reaction whose experimental condition is performed using a noncoordinating solvent, the monomer ethylene may be coordinated to the metal ion of the FeIpy catalyst in its active form and the active species of the Zdc. This proposal justifies the fact that the binary mixture is formed when present at a molar ratio of Zdc/FeIpy = 1 with an increased catalytic activity compared to the isolated catalysts. Therefore, the electrochemical data suggest that there is a possibility that the binary mixture in the ethylene polymerization reaction results in a number of different catalytic species of unit systems. Note that in the polymerization conditions, the FeIpy complex interacts with ethylene, just as it occurs with zirconocene,
Please cite this article in press as: E.S. Oliveira, et al., J. Mol. Catal. A: Chem. (2016), http://dx.doi.org/10.1016/j.molcata.2016.01.008
G Model MOLCAA-9740; No. of Pages 9 6
ARTICLE IN PRESS E.S. Oliveira et al. / Journal of Molecular Catalysis A: Chemical xxx (2016) xxx–xxx
Fig. 4. DPVs of Zdc, FeIpy and binary mixture with a molar ratio of FeIpy/Zdc = 1; (a) cathodic scan and (b) anodic scan.
and this step is important for the formation of the polymer product. Based on the data obtained and evaluated, it can be concluded that the combination of catalysts in a 1:1 ratio generated more active systems than the FeIpy complex, which should result from the interaction observed by DPV (Fig. 4). However, for the binary mixtures in ratios of 1:2 or 2:1, the presented activities are lower compared to those of the isolated systems, which is due to the use of catalytic systems based on electron transfer between the systems. Thus, the amount of catalytically active species interacting with the monomer at a given time is minimized. For the molar ratio of Zdc/FeIpy = 2, the change in the DVP profile compared with the molar ratio of 1:1 was more complex (Fig. S2). Note that in this case, the signal of 8a/8c (Eq. IV in Scheme 2) has an increased current density. Thus, excess Zdc appears to induce redox processes at the electrode surface. Therefore, the electron transfer
reaction between the two complexes appears to be more favored, thereby changes in the catalytic activity. 3.5. Relationship between electrochemical behavior and polyethylene properties To verify the contribution effect of the FeIpy/Zr ratio on the molar mass of the polymers produced with the combination of FeIpy and [ZrCl2 (n-BuCp2 )], GPC analysis of the resulting polymers was performed. The chromatograms were also deconvoluted according to Flory [21] as shown in Fig. 6. In the deconvoluted chromatograms shown in Fig. 6, the bold lines represent the experimental data, and the other lines are the corresponding polymers with lower and higher molar masses. The polymers obtained using the FeIpy complex (Fig. 6a) resulted in
Please cite this article in press as: E.S. Oliveira, et al., J. Mol. Catal. A: Chem. (2016), http://dx.doi.org/10.1016/j.molcata.2016.01.008
G Model MOLCAA-9740; No. of Pages 9
ARTICLE IN PRESS E.S. Oliveira et al. / Journal of Molecular Catalysis A: Chemical xxx (2016) xxx–xxx
7
Fig. 5. DPVs of binary mixture with a molar ratio of FeIpy/Zdc = 1: (1) Immediately after mixing, (2) two minutes after mixing and (3) five minutes after mixing.
Fig. 6. GPC polymer chromatograms and signal deconvolution. Polymers produced by [ZrCl2 (n-BuCp)2 ] (Zr) (a), FeIpy (b) and FeIpy/Zr systems in molar ratios of 1:1 (c), 1:2 (d) and 2:1(e). Polymers produced by isolated systems are presented for comparison.
Please cite this article in press as: E.S. Oliveira, et al., J. Mol. Catal. A: Chem. (2016), http://dx.doi.org/10.1016/j.molcata.2016.01.008
G Model MOLCAA-9740; No. of Pages 9 8
ARTICLE IN PRESS E.S. Oliveira et al. / Journal of Molecular Catalysis A: Chemical xxx (2016) xxx–xxx
Fig. 7. DSC profiles of the resulting polymers with the catalytic system FeIpy/Zr in 1:1 ratio: left (Tc ) and right (Tm ).
a polymer with a molecular weight of 16,376 Da. The polymers obtained using zirconocene species, [ZrCl2 (n-BuCp)2 ], (Fig. 6b) had a molecular weight of 208,035 Da. Comparing the chromatographic profiles generated by the isolated species, it is possible, using the deconvolution methodology, to determine which species are responsible for the production of low and high molar mass polymers. The combined system with the 1:1 FeIpy/Zr ratio (Fig. 6c) showed a higher catalytic activity compared to the other combined species, as previously shown in Table 1. The resulting polymers have a molecular weight of 76,682 Da and a broad polydispersion: the Mw/Mn (weight average molecular weight/number average molecular weight) is on the order of 20 compared to the isolated species. The low molar mass fraction appears to be consistent with the displacements and molar mass of the components that compose the polymer produced by the FeIpy species. The high molar mass fraction also appears coherent, according to the molar mass and displacements of the zirconocene catalyst. The ratio of the low and high mass polymers are 45% for FeIpy and 40% for the zirconocene, and the remaining 15% of species appear to correspond to the low mass fraction of the zirconocene species. An analysis that considers the relative fraction of each species shows that the resulting molar mass is not a simple combination of polymers produced independently by the two species; rather, it suggests some interaction between them, as indicated by the combined species voltammograms shown in Fig. 4, if compared to the isolated species. Equally, the broadening of the polydispersity may be associated with this interaction and, of course, the bimodality observed in the chromatogram. In the case of the 1:2 (Fig. 6d) and 2:1 ratio (Fig. 6e) systems, there is a decrease in the catalytic activity in relation to the isolated species. In the 1:2 ratio system, the molecular weight reaches 91,000 Da with a polydispersity of 11.7. In terms of fraction, the 1:2 ratio system shows one less fraction (4), which explains the lowest polydispersity among the three combined systems. The 2:1 ratio resulted in the highest molecular weight, 101,210 Da, and a polydispersity of 27.
3.6. Relationship between DSC measurements and polyethylene properties The contribution of each species on the crystallization temperature (Tc ), fusion temperature (Tm ) and crystallinity (c ) can be evaluated through DSC analysis. The Tm and c are in agreement with PEAD, confirming the production of this material. The resulting polymers from the binary systems exhibit properties that are
Table 3 Crystallization temperature (Tc ), fusion temperature (Tm ) and crystallinity (c ) of the polymers produced using isolated and binary systems. Polymer sample
Tc (◦ C)
Tm (◦ C)
c (%)
FeIpy [ZrCl2 (n-BuCp2 )] 1:1 FeIpy/Zr 1:2 FeIpy/Zr 2:1 FeIpy/Zr
108.8 109.0 100.1 107.5 108.4
135.1 141.1 144.0 138.2 137.3
83.2 59.3 78.8 76.0 85.3
Zr = [ZrCl2 (n-BuCp2 )].
compatible with those corresponding to the isolated systems, as shown in Table 3. The combined species in the 1:2 and 2:1 fractions present similarities in their crystallization and fusion temperatures, differing from each other by the sample resulting from the combination of the 2:1 fraction having a higher crystallinity, obeying the trend shown by the isolated species, although with a higher crystallinity value compared to that of the isolated species FeIpy. For the sample resulting from the 1:1 combination, one can observe lower Tc values and higher Tm values (see Fig. 7). The crystallization and fusion temperatures, associated with a high polydispersity and the several components that compose the molar mass distribution (Fig. 6c), confirm that the catalytic system combined in the 1:1 ratio possesses a higher perturbation resulting from the higher electronic interaction between the species, as shown by the DPV of this combination (Fig. 4).
4. Conclusions The electrochemical analysis enabled a detailed and accurate assessment of the behavior of binary catalytic systems formed by FeIpy and zirconocenes. When combined in the presence of MAO, each complex separately forms the catalytically active species for the polymerization of ethylene. It can be observed that the effect of the mixture depends on the nature of the zirconocene used and its proportion in the combined systems. Changes in the electrochemical signals of the binary mixtures, compared to those of the isolated complex, suggest that there is interaction between them. An equal ratio has an increased number of catalytically active species, which leads to increased activity. The occurrence of redox reactions between different complexes was observed, in which the metallocene is reduced and the complex containing Ipy is oxidized. This fact may justify the lower activity observed for the other ratios. Thus, an increase in the concentration of oxidized Ipy necessarily means a decrease in the concentration
Please cite this article in press as: E.S. Oliveira, et al., J. Mol. Catal. A: Chem. (2016), http://dx.doi.org/10.1016/j.molcata.2016.01.008
G Model MOLCAA-9740; No. of Pages 9
ARTICLE IN PRESS E.S. Oliveira et al. / Journal of Molecular Catalysis A: Chemical xxx (2016) xxx–xxx
of the reduced form of Ipy species that are part of the catalytically active species. GPC analysis showed that the polymer bearing bimodality varied between 4 and 27, suggesting comparable performance of both catalytic sites. Higher crystallinity values for the polymers obtained with binary systems probably reflect better mechanical properties. Binary mixtures of FeIpy and zirconocene 1:1 were more promising active systems than the isolated catalytic systems because it has a significant increase in catalytic activity and extension of the molar mass distribution in the polymeric chains, which facilitates the processing of the resulting polyethylene. Acknowledgements We thank the Brazilian agencie CNPq. The authors thank Teacher Ulf Schuchardt for his great contribution and help in catalysis olefin polymerization. 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.molcata.2016.01. 008.
9
References [1] F.M.B. Coutinho, I.L. Mello, L.C. Santa Maria, Polímeros 13 (2003) 1–13. [2] E.I.S. Oliveira, F. Silveira, A.M.V. Viveiros, N.R.S. Basso, Z.N. Rocha, J.H.Z. Santos, Quim. Nova 38 (2015) 97–105. [3] G.G. Hlatky, Coord. Chem. Rev. 181 (1999) 243–296. [4] H.G. Alt, A. Köppel, Chem. Rev. 100 (2000) 1205–1222. [5] V.C. Gibson, S.K. Spitzmesser, Chem. Rev. 103 (2003) 283–315. [6] I.S. Paulino, U. Schuchardt, J. Mol. Catal. A: Chem. 2011 (2004) 55–58. [7] I.S. Paulino, U. Schuchardt, Catal. Commun. 5 (2004) 5–7. [8] Z. Zhang, S. Chen, X. Zhang, H. Li, Y. Ke, Y. Lu, Y. Hu, J. Mol. Catal. A: Chem. 230 (2005) 1–8. [9] A.S. Abu-Surrah, K. Lappalainen, U. Piironen, P. Lehmus, T. Repo, M. Leskela, J. Organomet. Chem. 648 (2002) 55–61. [10] M.E. Bluhm, C. Folli, M. Doring, J. Mol. Catal. A: Chem. 212 (2004) 13–18. [11] N.V. Semikolenova, V.A. Zakharov, L.G. Echevckaja, M.A. Matsko, K.P. Bryliakov, E.P. Talsi, Catal. Today 144 (2009) 334–340. [12] S.S. Ivanchev, A.V. Yakimansky, D.G. Rogozin, Polymer 45 (2004) 6453–6459. [13] P. Mohring, N.J.G. Coville, Coord. Chem. Rev. 250 (2006) 18–35. [14] B. Wang, Coord. Chem. 250 (2006) 242–258. [15] H. Sinn, W. Kaminsky, Adv. Organomet. Chem. 18 (1980) 99. [16] A.G. Fisch, N.S.M. Cardozo, A.R. Secchi, J.H.Z. Santos, Quim. Nova 34 (2011) 646–657. [17] J.H.Z. Santos, A.G. Fisch, N.S.M. Cardozo, A.R. Secchi, Quim. Nova 31 (2008) 1199–1207. [18] R. Souane, F. Isel, F. Peruch, P.J. Lutz, Coord. Rev. Chim. 5 (2002) 43–48. [19] F. Silveira, L.M.T. Simplício, Z.N. Rocha, J.H.Z. Santos, Appl. Catal. A 344 (2008) 98–106. [20] F. Silveira, L.M.T. Simplício, Z.N. Rocha, J.H.Z. Santos, Macromol. React. Eng. 2 (2008) 253–264. [21] P.J. Flory, Principles of Polymer Chemistry, Cornell University, New York, 1953.
Please cite this article in press as: E.S. Oliveira, et al., J. Mol. Catal. A: Chem. (2016), http://dx.doi.org/10.1016/j.molcata.2016.01.008