Microstructure of polypropylene and active center in Ziegler–Natta catalyst: effect of novel salicylate internal donor

Abstract

Five salicylates with different sizes of hydrocarbon substituents were firstly synthesized and employed as ecofriendly internal donors of the Ziegler–Natta catalyst for propylene polymerization. The influences of these salicylates and traditional, industrial, internal donor diisobutyl phthalates on the microstructure of polypropylene and active center in a Ziegler–Natta catalyst were studied. It was found that the catalyst activities of the catalysts containing salicylate internal donors with a proper volume were higher than the catalysts containing diisobutyl phthalate internal donors. GPC results showed that the molecular weights of polypropylene prepared by salicylate internal donors were lower than those prepared by diisobutyl phthalate, which indicated that the polypropylene chains produced by salicylate internal donors were easier to transfer than those prepared by diisobutyl phthalate internal donors. Deconvolution of the GPC curves exhibited that as the volume of the salicylate internal donor increased some of the active centers for low molecular weight transferred into the active centers for high molecular weight. The results of ¹³C-NMR and SSA both suggested that a salicylate internal donor with an appropriate catalyst size volume was beneficial for increasing the isotactic sequence length, isotacticity index and regular triads “mm” of polypropylene. However, further increasing the volume of the salicylate internal donor in a catalyst would lead to the polypropylene chain containing more stereo-defects. Moreover, the active centers with different stereospecificity parameters, p_iso, in the catalyst could explain the trend of stereo-defects in polypropylene chains when different internal donors were used. In addition, it was found that the isotactic sequence length and isotacticity index of polypropylene prepared by isobutyl 2-benzyloxy-3,5-isopropyl benzoate were close to that produced by a diisobutyl phthalate internal donor. Moreover, the lamella thickness distribution of the polypropylene produced by a salicylate internal donor was broad, which might have potential application for expanded polypropylene materials.

---

Introduction

Driven by continuous developments in catalyst technology and polymerization processes, polypropylene has many excellent properties and has become one of the most widely used thermoplastic resins.^1–3 At present, the forth Ziegler–Natta catalyst is one of the most widely used catalyst systems for propylene polymerization,⁴ which always contains MgCl₂-supported TiCl₄, an internal donor, triethyl aluminium (TEA) and an external donor.⁵ Since internal donors are applied to the Ziegler–Natta catalyst system, it has become a key component for improving comprehensive properties of polypropylene.^6–8 Many kinds of compounds^9–14 are used as Ziegler–Natta catalyst internal donors for research. Now, phthalate³ is the most widely used internal donor for the industrial production of polypropylene. However, phthalates are harmful to the human body, and the European Union and the United States have both restricted the use of phthalates in certain plastics. Our group uses aliphatic diesters as ecofriendly internal donors¹⁵ and obtains catalysts with high catalytic activity and polypropylene with high isotacticity. However, the synthesis of aliphatic diesters is difficult on an industrial scale. Salicylic acid is known for its ability to ease aches and pains and is a component of aspirin.¹⁶ Moreover, some salicylates are often used in cosmetics as a fragrance additive and UV light absorber. However, there are no reports on salicylate as an internal donor for polypropylene polymerization. In addition, it was found that the internal donor could strongly impact the active center of the Ziegler–Natta catalyst and the microstructure of polypropylene,^15,17,18 which determines the properties and industrial applications of polypropylene. Successive self-nucleating and annealing (SSA) thermal fractionation^19,20 can give lamella thickness and lamella thickness distribution of polypropylene, and both properties are related to the isotactic sequence length and isotactic sequence distribution of polypropylene. In addition, ¹³C-NMR^21,22 can be used not only to measure the stereo-defect distribution of polypropylene, but also to analyze the stereoselectivity of the active center in the Ziegler–Natta catalyst. Based on the results of ¹³C-NMR and DSC, a DSC curve model²³ could be used to analyze the active center with a different stereoselectivity in the Ziegler–Natta catalyst. In our study, five salicylates with different size hydrocarbon substituents were synthesized and employed as ecofriendly internal donors of the Ziegler–Natta catalyst for propylene polymerization. It was found that the catalyst activity of catalysts containing SID internal donors with a proper volume was higher than that of catalysts containing DIBP internal donors. Deconvolution of the GPC curves indicated that as the volume of the SID internal donor increased, some of the active centers for low molecular weight transformed into active centers for high molecular weight. The results of ¹³C-NMR and SSA both suggested that the SID internal donor with an appropriate catalyst size volume was beneficial for increasing the isotactic sequence length, isotacticity index and regular triads “mm” of polypropylene, while further increasing the volume of SID internal donors in the catalyst would lead to the polypropylene chain containing more stereo-defects. The DSC curve model was used to study the stereospecificity parameter, p_iso, of active centers in the catalyst, which could be used to explain the trend of isotactic sequence length of polypropylene prepared by different internal donors. In addition, it was found that the isotactic sequence length and isotacticity index of polypropylene prepared by SID-4 were very close to that produced by DIBP internal donors.

Experimental

Materials

Salicylic acid, 3-methyl salicylic acid, 3,5-isopropyl salicylic acid, 3,5-tert-butylation salicylic acid, benzoyl chloride, trimethylacetyl chloride, sodium bicarbonate, chloroform, propylene, hydrogen, triethylaluminium (TEA), diisopropyldimethoxysilane (donor-P) and diisobutyl phthalate (DIBP) were purchased from Aladdin in reagent grade and were used without further purification.

Synthesis of the salicylate internal donor

The synthesis method for the five salicylate internal donors was based on the literature.²⁴ The purity of salicylate used for propylene bulk polymerization was above 98%. The synthetic route for salicylate and the structures of the five salicylates are shown in Schemes 1 and 2, respectively. As a typical example, the synthesis process of isobutyl 2-benzyloxybenzoate (SID-1) was carried out as follows:

Scheme 1 Synthetic route of salicylate internal donors.

Scheme 2 Structures of salicylate internal donors with different substituent groups.

A 100 ml flask contained 40 ml isobutanol, 0.3 mol salicylic acid and 0.5 ml concentrated sulfuric acid. The mixture was stirred at 120 °C for 5 h. When water no longer appeared in the water separator, the heater was turned off. After the formation of isobutyl salicylate was confirmed by gas chromatography, isobutyl salicylate was purified by decompressing distillation. Isobutyl salicylate (0.05 mol), 0.08 mol benzoyl chloride and 0.005 mol bismuth(III) oxychloride were stirred at 40 °C for 6 h. After isobutyl 2-benzyloxybenzoate was confirmed by gas chromatography, the mixture was stirred with aqueous sodium hydrogen carbonate. The organic phase was extracted from the water phase, and then a transparent and oily liquid 2-benzyloxybenzoate was obtained by decompressing distillation.

The synthesis processes for the other salicylates, including isobutyl 2-benzyloxy-3-methyl benzoate (SID-2), isobutyl 2-trimethyloxy-3-methyl benzoate (SID-3), isobutyl 2-benzyloxy-3,5-isopropyl benzoate (SID-4) and isobutyl 2-benzyloxy-3,5-tert-butyl benzoate (SID-5), were similar to that of SID-1. ¹H NMR (DMSO) δ (ppm): SID-1: 0.81 (d, 6H), 1.73 (m, 1H), 3.92 (d, 2H), 7.25 (t, 2H), 7.60 (m, 3H), 7.72 (d, 1H), 8.07 (d, 1H), 8.19 (d, 2H); SID-2: 0.82 (d, 6H), 1.21 (s, 9H), 1.92 (m, 1H), 2.13 (s, 3H), 4.01 (d, 2H), 7.08 (t, 1H), 7.38 (d, 1H), 7.80 (d, 1H); SID-3: 0.85 (d, 6H), 1.80 (m, 1H), 3.95 (d, 2H), 7.42 (d, 2H), 7.60 (t, 3H), 7.75 (d, 1H), 8.25 (d, 2H); SID-4: 0.85 (d, 6H), 1.14 (d, 12H), 1.91 (m, 1H), 2.81 (m, 1H), 2.99 (m, 1H), 3.96 (d, 2H), 7.47 (s, 1H), 7.53 (t, 2H), 7.66 (m, 2H), 8.16 (d, 2H); SID-5: 0.88 (d, 6H), 1.39 (s, 18H), 1.78 (m, 1H), 3.90 (d, 2H), 7.54 (t, 2H), 7.63 (t, 1H), 7.93 (s, 2H), 8.25 (d, 2H).

Preparation of catalyst

The Ziegler–Natta catalysts with salicylate as the internal donor were prepared in a reactor under dry nitrogen conditions. Magnesium ethoxide (5 g) and 150 ml dry toluene were added to a glass reactor under nitrogen at room temperature. Then, 50 ml TiCl₄ was added to reactor by a funnel over half an hour at 5 °C. Two millilitres of salicylate internal donor was injected into the reactor after dripping the TiCl₄. The mixture was stirred at 110 °C for 2 h. After reaction, the supernatant was removed, and the residue was washed by toluene. The precipitated solid was treated again with a mixture of 150 ml toluene and 50 ml titanium tetrachloride at 110 °C for 2 h. After discarding the supernatant, the residue was washed with toluene and hexane, respectively. A solid catalyst was obtained.

Propylene bulk polymerization

Firstly, 800 g of propylene gas and 0.3 MPa of hydrogen gas were charged with a reactor. The C(H₂)/C(propylene) ratio was 2.59 mmol mol⁻¹. Then, 20 mg of the Ziegler–Natta catalyst, 12 mmol TEA in n-hexane solution and 0.39 mmol donor-P were added to the system. The reactor was heated to 70 °C and stirred at 300 rpm for an hour. Then, the propylene was exhausted, and polypropylene was obtained as a white powder.

UV-Vis spectra

The content of Ti in the catalyst was measured by UV-Vis spectra (UV-1800). Firstly, the catalyst was dissolved in a dilute sulfuric acid solution. Then, titanium was converted to [TiO(H₂O₂)]SO₄ by the addition of H₂O₂. A spectrophotometer was used to record UV-Vis spectra of the resultant solutions of catalyst complexes.

Gas chromatography

The ester content of the catalysts was determined on a gas chromatograph (clarus 580) manufactured by PE in the US.

Differential scanning calorimetry

A Mettler 822e differential scanning calorimeter was used to measure the thermal properties of polypropylene. Polypropylene (2–4 mg) was put in an aluminium pan. To erase the thermal history, the pan with polypropylene was heated from 50 °C to 200 °C at a heating rate of 50 °C min⁻¹. It was held at 200 °C for 5 min in a nitrogen atmosphere. Then, polypropylene in the pan was cooled to 50 °C and held at 50 °C for 5 min. Finally, the polypropylene was heated to 200 °C at 10 °C min⁻¹.

Gel permeation chromatography

PL-GPC 220 high-temperature gel permeation chromatography (Polymer Laboratories Ltd.) was used to measure the molecular weights (M_n and M_w) and the polydispersity index (PI) of polypropylene. The measuring temperature was 413.15 K. 1,2,4-Trichlorobenzene was the solvent for polypropylene. The injection volume was 100 ml, and the flow rate was 1.0 ml min⁻¹. The calibration was made by used linear polystyrene as the standard sample.

Successive self-nucleation and annealing

Successive self-nucleation and annealing (SSA) technology has been widely used to analyze the lamellar thickness and isotactic sequence length of polymers.²⁵ The method includes several heating and cooling steps: (1) DSC heating of polypropylene to 200 °C at 50 °C min⁻¹ and then holding for 5 min; this step is for the erasure of the crystalline thermal history of polypropylene. (2) Cooling polypropylene to 50 °C at 20 °C min⁻¹ and holding for 5 min. (3) Heating polypropylene from 50 °C to a partial melting temperature denoted T_s at 20 °C min⁻¹. (4) Keeping polypropylene at T_s for 10 min. (5) Cooling polypropylene from T_s to 50 °C at 20 °C min⁻¹. (6) Repeat steps ‘c’ to ‘e’ at increasingly a new lower T_s, which were varied from 164 to 144 °C, at 5 °C intervals for a total five self-nucleation/annealing steps. (7) Finally, heating polypropylene from 50 °C to 200 °C at 10 °C min⁻¹. Then, multiple melting curves (SSA curve) were obtained. Pijpers gave the method for looking at the optimum T_s temperatures of polypropylene in the literature.²⁶

Nuclear magnetic resonance spectroscopy

DMX 300M (Bruker) was used to test the ¹³C-NMR spectra of polypropylene. Polypropylene (80–100 mg) was dissolved in 0.5 ml of deuterated o-dichlorobenzene at 383 K. The o-dichlorobenzene solvent was used to provide the internal lock signal with its highest peak at 132.700 ppm. The number of pulses was more than 5000; pulse angle was 30; spectrum width was 25 000 Hz; and relaxation delay was 7 s. All spectra were completely proton decoupled.

Results and discussion

Components and catalytic activity of the Ziegler–Natta catalyst

Five salicylates with different hydrocarbyl substituents on the phenyl group were firstly prepared, as Scheme 2 shows. Then, the five salicylates (SID) and di-butyl phthalate (DIBP) were used as internal donors to prepare six Ziegler–Natta catalysts, as Table 1 lists. Table 1 also displays the Ti content and ester content in the Ziegler–Natta catalyst. As shown in Table 1, the Ti contents and ester contents in SID catalysts were both lower than that of the DIBP catalyst, indicating that the Ti and ester contents in the catalysts could be affected by the types of internal donors. Besides, comparing Cat-1, Cat-3 and Cat-4, the ester content of the catalysts increased from Cat-1 to Cat-4, while their Ti content decreased from Cat-1 to Cat-4. It was reported that the ester can coordinate with Mg atoms on the (110) facets of MgCl₂ crystallites,²⁷ which could block surface sites on the MgCl₂ support and lead to Ti hardly coordinating with the support. Therefore, the trend of relative Ti and ester content was contrary. Previous research found that an increase in the ester content in a catalyst was beneficial for improving the catalyst activity of the Ziegler–Natta catalyst.²⁸ Table 1 shows that the catalyst activity increased from 6.6 × 10⁵ g PP per g Ti to 24.1 × 10⁵ g PP per g Ti as the ester content of catalyst increased from 1.1% to 2.2%. The trend of catalyst activity and ester content in the catalyst also matched previous conclusions. In addition, it was found that catalyst activities of the SID catalyst (except for SID-1) were much higher than that of the DIBP catalyst, which indicated that SID internal donors with a proper volume were more beneficial for the catalyst activity than DIBP internal donors in catalysts.

Table 1 Components and catalytic activity of Ziegler–Natta catalysts containing different internal donors

Samples	Donor	Catalytic activity (×10^5 g PP per g Ti)	Ti^a (wt%)	Ester^b (wt%)
a The weight percent of Ti in the catalyst.b The weight percent of ester in the catalyst.
Cat-1	SID-1	6.6	2.5	1.1
Cat-2	SID-2	13.7	1.7	1.9
Cat-3	SID-3	10.3	1.8	1.7
Cat-4	SID-4	24.1	0.9	2.2
Cat-5	SID-5	23.7	0.7	1.9
Cat-D	DIBP	8.3	2.7	6.8

---

Molecular weight and polydispersity index of polypropylene

Table 2 lists the molecular weight and polydispersity index of polypropylenes prepared by different internal donors. It was shown that the polydispersity index of the polypropylene prepared by the SID internal donors ranged from 5.15 to 7.19, which was similar to that prepared by the DIBP internal donors. Table 2 also shows that the molecular weights of the polypropylene prepared by SID internal donors were all lower than that prepared by DIBP, which indicated that the polypropylene chain produced by the SID internal donors was easier to transfer than that prepared by the DIBP internal donors. Furthermore, it was also found that the average molecular weight, M_w, of polypropylene successively increased from SID-1 to SID-5. In order to deeply analyze the trend of the molecular weight of polypropylene prepared by different internal donors, GPC curves were deconvoluted by Schulz–Flory distribution functions.

Table 2 Molecular weight and polydispersity index of polypropylenes prepared by different internal donors

Donor	Mn × 10^−4 (g mol^−1)	Mw × 10^−4 (g mol^−1)	Mw/Mn
SID-1	4.53	26.39	5.82
SID-2	5.13	26.45	5.15
SID-3	4.25	27.24	6.40
SID-4	4.99	27.38	5.48
SID-5	3.83	27.55	7.19
DIBP	7.54	43.26	5.74

---

Active centers based on the molecular weight of polypropylene

The M_w/M_n ratio of polypropylene prepared by a single active center catalyst was narrow (2–3), and the wider distribution of molecular weight indicated that the catalyst contained multiple active centers. Judging by the polydispersity index of polypropylene, catalysts with SID and DIBP internal donors both contained several types of active centers. Fig. 1 showed that GPC curves of polypropylene prepared by SID and DIBP internal donors were both deconvoluted into five kinds of Flory compounds with small polydispersity indices, which could correspond to different active centers in the catalyst as it did in the literature.²⁹ Comparing the active centers in the catalyst with different SID internal donors, it was found that each type of active center in different SID catalysts produced polypropylene with a similar molecular weight; however, the relative content of each kind of active center in the five SID catalysts changed regularly. The relative contents of and increased from Cat-1 to Cat-5, while the relative contents of , and all decreased from Cat-1 to Cat-5 (Table 3). It was indicated that as the volume of the SID internal donor in the catalyst increased some of , and transferred into and . In addition, it was found that the molecular weight of the polypropylene prepared by each kind of active center in the DIBP catalyst was obviously different with that in the SID catalyst, which indicated that the active centers in the SID catalyst were different with that in the DIBP catalyst. The molecular weight of polypropylene mainly depended on the hydrogen response, as the chain transfer to hydrogen was more sensitive than the chain transfer to a monomer or TEA.²⁷ The sensitivity of the chain transfer to hydrogen was primarily dependent on the regioselectivity of the active center, which had a great relationship with the proportion of secondary (2,1-) monomer insertion.³⁰ Taking into account the same addition amount of hydrogen for SID and DIBP systems, the different molecular weights of polypropylene prepared by active centers in SID and DIBP catalysts were ascribed to the different internal donors in the catalyst. Although it was reported that some of the ester internal donors could be removed by TEA, internal donors that strongly coordinated with the support could not be removed,³¹ which could influence the active center in the catalyst. It was suggested that the different types of internal donors remaining in the catalyst might have different impacts on the regioselectivity of the active center in the Ziegler–Natta catalyst, which led to SID and DIBP catalysts producing polypropylene with different molecular weights.

Fig. 1 Resolution of GPC curves into five Flory components for polypropylene.

Table 3 Results of GPC deconvolution by Flory components

Samples
	Fr^a	Mw^b	Fr	Mw	Fr	Mw	Fr	Mw	Fr	Mw
a Fr was the weight percentage of the fraction produced by a certain active center in catalyst.b Weight average molecular weight, in 10^4 g mol^−1.
Cat-1	16.3	56.4	33.9	19.7	32.3	8.2	12.9	2.9	4.4	0.9
Cat-2	18.1	56.5	35.0	19.8	31.1	8.0	11.6	3.0	3.9	1.2
Cat-3	18.3	57.9	35.7	19.8	30.7	7.7	11.4	2.9	3.6	1.1
Cat-4	19.0	57.6	35.9	19.2	30.1	8.1	11.3	3.3	3.5	1.2
Cat-5	19.8	58.4	37.9	18.5	29.3	7.3	10.2	2.5	2.7	0.7
Cat-D	8.2	138.9	30.7	51.2	38.8	17.9	18.0	6.7	4.3	2.1

---

The stereo-defect in polypropylene chain

The stereo-defect of the molecular chain could significantly impact the thermal and mechanical properties of polypropylene; therefore, detailed characterization of the stereo-defect in the molecular chain was important to obtain polypropylene with better properties. The fraction of heptane-insoluble crystalline polypropylene was defined as the isotactic index (I.I.) of polypropylene, which could be used to roughly analyze the stereo-defect in the polypropylene chain. Table 4 shows the isotactic index of the polypropylene prepared by different internal donors. The isotactic index of polypropylene increased from 96.3% to 98.6% as the internal donor successively changed from SID-1 to SID-4, which indicated that the volume of the SID internal donor could impact the stereoregularity of polypropylene. In addition, the isotactic index of polypropylene produced by SID-4 was close to that prepared by the DIBP internal donor. In order to obtain more information on the stereo-defect in the molecular chain, successive self-nucleation and annealing (SSA) and nuclear magnetic resonance spectroscopy (¹³C-NMR) were both used to study the microstructure of polypropylene.^32,33

Table 4 SSA results of polypropylene prepared by different internal donors

Donor	ΔHm^a (J g^−1)	I.I.^b (%)	Tm1^c (°C)	Tm2 (°C)	Tm3 (°C)	Tm4 (°C)	Tm5 (°C)
a The isotactic index (I.I.) was tested by extraction with boiling n-heptane for 6 h.b The value of the endothermic enthalpy was determined from the SSA curve.c The melting temperature was determined from the peak value in the SSA curves.
SID-1	104.2	96.3	175.9	170.2	165.8	160.6	154.8
SID-2	99.1	96.9	175.9	169.4	165.1	160.3	154.7
SID-3	109.9	98.0	176.0	169.9	165.6	160.5	155.0
SID-4	118.1	98.6	176.3	169.9	165.5	160.3	153.5
SID-5	106.2	97.7	176.1	169.8	165.2	160.0	154.0
DIBP	121.0	98.5	171.4	166.0	160.2	—	—

---

As an effective method for the microstructure features of polypropylene, SSA can give lamellar thicknesses and thickness distribution of the polypropylene, which could be connected with the isotactic sequence length of the polymer. The melting curves of polypropylene prepared by different internal donors after SSA thermal fractionation are shown in Fig. 2, and the melting enthalpy ΔH_m of each polypropylene is listed in Table 4. Comparing the melting enthalpies ΔH_m of polypropylene prepared by SID-1, SID-3 and SID-4, it was found that the melting enthalpy ΔH_m of polypropylene increased from SID-1 to SID-4, while the melting enthalpy ΔH_m of polypropylene prepared by SID-5 was lower than those prepared by SID-4, which reflected a trend in the crystallinity degree of polypropylene.

Fig. 2 SSA melting curves of polypropylene samples prepared by different internal donor.

Each melting peak of polypropylene in the SSA melting curves was related to a different thickness of the lamellar crystallites, which formed and annealed at each self-nucleation temperature.³⁴ The Thomson–Gibbs equation³⁵ can give the information about the lamellar thickness of polypropylene:

T_m = T⁰_m(1 − 2σ/ΔH₀L_i)

where the equilibrium melting temperature is T⁰_m = 460 K,³⁶ ΔH₀ = 184 × 10⁶ J m⁻³, the surface energy σ = 0.0496 J m⁻² and L_i is the lamellar thickness.³⁷

The lamellar thicknesses of polypropylene after SSA thermal fractionation were calculated and are listed in Table 5. It was shown that the polypropylene prepared by SID internal donors had five kinds of lamellar thicknesses but polypropylene prepared by DIBP internal donors had only three kinds of lamellar thicknesses, which indicated that the distribution of the lamellar thickness in polypropylene produced by SID internal donors might be different than that prepared by DIBP internal donors. It was also found that all of the SID internal donors produced polypropylene with similar lamellar thicknesses, which differed from those prepared by DIBP internal donors, revealing that the type of internal donor could affect the thickness of each kind of lamellae in polypropylene.

Table 5 The lamellar thicknesses of polypropylene prepared by different internal donors

Donor	L1	L2	L3	L4	L5
SID-1	22.65	13.81	11.78	9.45	7.74
SID-2	22.65	13.57	11.52	9.34	7.71
SID-3	22.68	13.66	11.71	9.42	7.82
SID-4	22.79	13.66	11.67	9.34	7.44
SID-5	22.71	13.62	11.56	9.23	7.55
DIBP	16.04	11.89	9.30	—	—

---

Moreover, it was found that the isotactic sequence length of polypropylene was closely related with lamellar thickness. The following equations³⁸ were used to calculated the average lamellar thickness, thickness distribution (arithmetic average L_n, weighted average L_w and broadness index I) and average isotactic sequence length (arithmetic average MSL_n, weighted average MSL_w). n_i is the content of each fraction on the SSA curve, and L_i is the thickness of lamella for each fraction, L_helix = 0.65 nm.

I = L_w/L_n MSL = 3L/L_helix

Comparing the average isotactic sequence length MSL_n of polypropylene prepared by different internal donors in Table 6, the isotactic sequence length MSL_n of polypropylene prepared by SID internal donors ranged from 57.04 to 59.82, which was close to that produced by DIBP internal donors. Meanwhile, the distribution of the lamella thickness I of polypropylene prepared by SID internal donors ranged from 1.056 to 1.074, which was broader than that prepared by DIBP internal donors. It was suggested that SID internal donors in the catalyst were more conducive to producing polypropylene with a broad distribution of lamella thickness than DIBP. In addition, comparing SID-4 and DIBP, it was found that the average lamella thickness of polypropylene prepared by SID-4 was similar to that produced by DIBP, but the thickness and thickness distribution of each lamella in polypropylene for SID-4 and DIBP internal donors were different.

Table 6 Lamellar thickness statistical parameters and isotactic sequence length of polypropylene prepared by different internal donors

Donor	Ln (nm)	Lw (nm)	MSLn	MSLw	I
SID-1	12.36	13.27	57.04	61.25	1.074
SID-2	12.77	13.37	58.93	61.71	1.057
SID-3	12.62	13.39	58.24	61.80	1.061
SID-4	12.96	13.52	59.82	62.40	1.056
SID-5	12.66	13.48	58.43	62.22	1.065
DIBP	13.11	13.76	60.50	63.09	1.043

---

Nuclear magnetic resonance spectroscopy (¹³C-NMR) was always used to study the macrostructure of polypropylene.³⁹ The methyl regions' results from the 300 MHz ¹³C-NMR spectrum of the polypropylene are listed in Table 7. It was found that the content of the regular triads “mm” of polypropylene prepared by SID-1–SID-4 internal donors increased gradually from 85.46% to 90.98%, while the amount of “mr” decreased from 7.81% to 3.74% and the amount of “rr” also decreased from 6.66% to 4.47%. In addition, comparing polypropylene prepared by SID-4 and DIBP internal donors, the content of the regular triads' “mm” of polypropylene prepared by an SID-4 internal donor was slightly lower than that prepared by a DIBP internal donor. It suggested that both type and volume of the internal donors could impact the regular triads' “mm” of polypropylene. ¹³C-NMR results could also be used to analyze the average meso sequence length of polypropylene. Average isotactic sequence length calculated from ¹³C-NMR results was defined as MSL, which could be calculated by the following equation:⁴⁰

Table 7 Meso sequence length calculated from ¹³C-NMR and the results of ¹³C-NMR

Donor	mm	mr	rr	MSL
SID-1	85.46	7.81	6.66	15.56
SID-2	88.08	6.11	5.81	22.97
SID-3	89.61	5.25	5.15	19.16
SID-4	90.98	3.74	4.47	28.58
SID-5	88.88	5.50	5.63	22.76
DIBP	91.67	3.82	4.51	30.66

---

Table 7 showed the average isotactic sequence length MSL calculated from the results of ¹³C-NMR. Comparing polypropylene prepared by SID-1, SID-3, and SID-4, it was found that the average meso sequence length MSL calculated from ¹³C-NMR data increased from SID-1 to SID-4; however, SID-5 gives polypropylene with a lower meso sequence length than SID-4, which was in good agreement with the trend of MSL_n calculated from SSA. Moreover, the trend of the average isotactic sequence length was also similar for the isotacticity index and regular triads' “mm” of polypropylene prepared by different SID internal donors, which suggested that the appropriate volume size for SID internal donors in the catalyst would benefit a decrease in the content of stereo-defect in the polypropylene chain. However, further increasing the volume of SID internal donors in the catalyst might lead to a polypropylene chain containing more stereo-defect. It was supposed that SID internal donors with a different volume might have a different effect on the stereoselectivity of the catalyst, which led to the production of polypropylene with different stereo-defects.

The thermal properties of polypropylene

Fig. 2 shows that the SSA curves of polypropylene prepared by SID internal donors were all broader and had more peaks than that prepared by DIBP internal donors, which represented more types and a broader distribution of lamellar thickness for polypropylene prepared by SID internal donors. Therefore, it was assumed that SID internal donors might produce polypropylene with broad DSC curves. Fig. 3 displays that the DSC curves prepared by SID internal donors were indeed broader than DIBP. Many researchers found that the broad melting curve of the polymer was beneficial for expanding the crystallization temperature range for expanded polypropylene, which was conducive to control the foaming of polypropylene.⁴¹ Hence, that polypropylene produced by SID internal donors might possibly be used as the material for expanded polypropylene.

Fig. 3 DSC curves of polypropylene prepared by different internal donors.

Active centers with different stereoselectivity

It was reported that there were multiple active centers with different stereoselectivity in heterogeneous Ziegler–Natta catalysts⁴² that could produce polypropylene chains with varying distributions of stereo-defects. The results of ¹³C-NMR and SSA confirmed that internal donors could significantly impact the relative content and distribution of stereo-defects in the molecular chain of polypropylene. Different internal donors in the catalyst might have different influences on the stereoselectivity of active centers in the Ziegler–Natta catalyst. Combining the results of ¹³C-NMR and DSC, a modeling of the DSC melting curve²³ could be used to analyze the stereoselectivity of active centers in the Ziegler–Natta catalyst. If all of the active centers in the catalyst were perfectly isospecific, which means that the probability of meso linking of adjacent propylene monomer units was highest (stereospecificity parameter p_iso was the highest and defined as 1), polypropylene prepared by the active centers would have very narrow DSC curves and the T_m of the DSC curves would be highest (T⁰_m); the “mmmm” value of the polymer would be close to 1. There are certain relationships between T_m, “mmmm” value and stereospecificity parameter p_iso that have been reported by Kissin.⁴² Fig. 4 is the straight fitting of T_m from the DSC curve and “mmmm” value from ¹³C-NMR. It was found that the highest melting temperature T⁰_m was 171.2 °C, which was obtained by extrapolating the “mmmm” value to 1. The highest melting temperature T⁰_m can be used to calculate the stereospecificity parameter p_iso of the active center.⁴³ Fig. 5 is the modeling of the DSC curve of polypropylene prepared by SID catalyst. It shows that there are four fitting curves to fit the DSC melting curve of polypropylene, which corresponds to four active centers with different stereospecificity parameters, p_iso. Combining the melting temperature of the fitting curve from Fig. 5 with T⁰_m from Fig. 4, the stereospecificity parameter p_iso for each active center could be obtained.⁴³ The stereospecificity parameter p_iso and the relative content of the active centers I_stereo and II_stereo for different catalysts are listed in Table 8, and correspond to T_m1 and T_m2 in the DSC curve model and had high stereoselectivity in the catalyst. Comparing active centers I_stereo and II_stereo for Cat-1, Cat-3 and Cat-4, the relative content of active I_stereo and II_stereo both increased from Cat-1 to Cat-4, while Cat-5 had a lower content of active centers I_stereo and II_stereo than Cat-4. Meanwhile, the trends for stereospecificity parameters p_iso1 and p_iso2 were similar with the rule of the relative content of the active centers I_stereo and II_stereo. It was indicated that the stereoselectivity of the catalyst increased as the volume of the SID internal donor in the catalyst increased, but a SID internal donor with an excessively large volume could decrease the stereoselectivity of the catalyst. It was reported that Ti active centers attach to the surface of the (110) facets of MgCl₂ crystallites by coordination of Cl⁻ on TiCl₄ with Mg on MgCl₂.⁴⁴ Meanwhile, ester internal donors can also coordinate with Mg, which would lead to dispersed Ti on the surface of MgCl₂ crystallites and decrease the space of the Ti active centers.²⁷ It was also confirmed that although some of the ester internal donors could be exchanged by external donors during polymerization, the ester internal donors remaining in the catalyst could significantly impact the active centers of the catalyst.³¹ Therefore, it was speculated that as the SID internal donors with increased volume remained in the catalyst, the space of the active centers around the internal donors became smaller, which favors improving the stereoselectivity of the catalyst. However, an excessively large volume of the SID internal donors would result in an excessively small space for active centers, which might be harmful to meso-insertion of propylene units in the polymer chain. Therefore, the stereoselectivity of the catalyst decreased. In addition, the structures of SID and DIBP were different, which might lead to the different stereoselectivities of the SID and DIBP catalysts.

Fig. 4 The straight fitting of T_m from DSC curve and “mmmm” value from ¹³C-NMR.

Fig. 5 Resolution and fitting of DSC curves for four components.

Table 8 Stereospecificity parameter p_iso and activation energy ΔE_act(si/re) of active centers calculated from a DSC model

Samples	Istereo			IIstereo
	piso1 (°C)	ΔEact(si/re)1	Yield (%)	piso2 (°C)	ΔEact2(si/re)2	Yield (%)
Cat-1	99.09	−3.14	30.2	98.62	−2.86	56.2
Cat-2	99.21	−3.23	31.8	98.66	−2.88	58.0
Cat-3	99.26	−3.28	32.2	98.77	−2.93	58.2
Cat-4	99.27	−3.29	33.9	98.81	−2.95	60.5
Cat-5	99.23	−3.25	32.5	98.76	−2.93	59.6
Cat-D	99.49	−3.45	25.6	99.06	−3.12	64.8

---

In addition, stereospecificity parameter p_iso, defined as k_si/(k_si + k_re) in kinetics (Scheme 3), can be used to analyze the activation energy ΔE_act(si/re), as the following equation: ΔE_act(si/re) = RT ln(k_si/k_re) = RT ln[(1 − p_iso)/p_iso].²³ Table 8 shows that the law of ΔE_act(si/re)₁ and ΔE_act(si/re)₂ values for active centers in the SID catalyst was contrary to the trend of the isotactic sequence length (from SSA and ¹³C-NMR), regular triads “mm” (from ¹³C-NMR) and isotacticity index of polypropylene. This trend further confirmed that the internal donor in the catalyst can impact the stereoselectivity of active centers in the Ziegler–Natta catalyst, which led to changing of the activation energy ΔE_act(si/re) and further influencing the distribution and content of stereo-defects of propylene during polymerization. Finally, it showed the general trend of isotactic sequence lengths and thermal properties of polypropylene for different internal donors.

Scheme 3 Active centers (C_si) and active centers (C_re).

Conclusions

Five salicylates with different size hydrocarbon substituents on phenyl groups were firstly synthesized and employed as ecofriendly internal donors of the Ziegler–Natta catalyst for propylene polymerization. The effect of these salicylates and traditional industrial internal donor DIBP on the microstructure of polypropylenes was studied by GPC, SSA and NMR. Meanwhile, the active centers in the catalyst were studied by deconvolution of GPC curves and DSC model. GPC results showed that the molecular weights of polypropylene prepared by SID internal donors were all lower than that prepared by DIBP, which indicated that the polypropylene chains produced by SID internal donors were easier to transfer than those prepared by DIBP internal donors. Deconvolution of GPC curves showed that as the volume of the SID internal donors increased, the relative contents of the active centers for high molecular weights increased, while the relative contents of the active centers for low molecular weights decreased. The isotactic sequence lengths from ¹³C-NMR and SSA both gradually increased from SID-1 to SID-4, while SID-5 gave polypropylene with a lower meso sequence length than SID-4, which was similar with the trend of the isotacticity index and regular triads “mm” of polypropylene. The general trend of stereo-defects in the polypropylene chain was explained by the stereoselectivity of the active centers in the catalyst for different internal donors, indicating that appropriately increasing the volume of SID internal donors could improve the stereoselectivity of the catalyst. However, excessively large volumes of SID internal donors would reduce the stereoselectivity of the catalyst. In addition, it was found that the catalyst activities of catalysts containing SID internal donors with proper volume were higher than that of catalysts containing DIBP internal donors. The isotactic sequence length and isotacticity index of polypropylene prepared by isobutyl 2-benzyloxy-3,5-isopropyl benzoate were close to that produced by DIBP internal donors.

Acknowledgements

The authors express thanks for the supports of the National Natural Science Foundation of China (NO. 51073170) and Innovation Program of CAS Combination of Molecular Science and Education.

References

1. J. Qiao, M. Guo, L. Wang, D. Liu, X. Zhang, L. Yu, W. Song and Y. Liu, Polym. Chem., 2011, 2, 1611–1623.
2. T. Taniike and M. Terano, in Polyolefins: 50 Years after Ziegler and Natta I: Polyethylene and Polypropylene, ed. W. Kaminsky, 2013, vol. 257, pp. 81–97.
3. J. Brandrup and E. H. Immergut, Polymer Handbook, Wiley, New York, 3rd edn, 1989.
4. H. Youliang, C. Hefei, L. Huayi and Z. Liaoyun, Petrochem. Technol., 2013, 11, 1189–1196.
5. V. Busico, MRS Bull., 2013, 38, 224–228.
6. K. Soga, T. Shiono and Y. Doi, Macromol. Chem. Phys., 1988, 189, 1531–1541.
7. M. L. Ferreira and D. E. Damiani, J. Mol. Catal. A: Chem., 1999, 150, 53–69.
8. D. M. Xu, Z. Y. Liu, J. Zhao, S. M. Han and Y. L. Hu, Macromol. Rapid Commun., 2000, 21, 1046–1049.
9. Z. L. Ma, L. Wang, W. Q. Wang, L. F. Feng and X. P. Gu, J. Appl. Polym. Sci., 2005, 95, 738–742.
10. S. Tanase, K. Katayama, N. Yabunouchi, T. Sadashima, N. Tomotsu and N. Ishihara, J. Mol. Catal. A: Chem., 2007, 273, 211–217.
11. Y. Yang, X. F. Dang, J. Y. Yao, H. Y. Li and Y. L. Hu, Acta Polym. Sin., 2013, 511–517,  DOI:10.3724/sp.j.1105.2013.12268.
12. N. Cui, Y. Ke, H. Li, Z. Zhang, C. Guo, Z. Lv and Y. Hu, J. Appl. Polym. Sci., 2005, 99, 1399–1404.
13. K. W. Choi, K. J. Chu, J. H. Yim and S. K. Ihm, Eur. Polym. J., 1998, 34, 739–745.
14. D. R. Burfield and P. J. T. Tait, Polymer, 1974, 15, 87–96.
15. X. F. Dang, Q. Li, H. Y. Li, Y. Yang, L. Y. Zhang and Y. L. Hu, J. Polym. Res., 2014, 21(619), 1–8.
16. R. K. Madan and J. Levitt, J. Am. Acad. Dermatol., 2014, 70, 788–792.
17. M. Chang, US Pat., 8003558, 2011.
18. A. G. Potapov, V. A. Zakharov and G. D. Bukatov, Kinet. Catal., 2007, 48, 403–408.
19. A. J. Muller, Z. H. Hernandez, M. L. Arnal and J. J. Sanchez, Polym. Bull., 1997, 39, 465–472.
20. M. L. Arnal, Z. H. Hernandez, M. Matos, J. J. Sanchez, G. Mendez, A. Sanchez and A. Muller, Use of the SSA technique for polyolefin characterization, Soc Plastics Engineers, Brookfield Center, 1998.
21. G. Morini, E. Albizzati, G. Balbontin, I. Mingozzi, M. C. Sacchi, F. Forlini and I. Tritto, Macromolecules, 1996, 29, 5770–5776.
22. V. Busico, R. Cipullo, G. Monaco, G. Talarico, M. Vacatello, J. C. Chadwick, A. L. Segre and O. Sudmeijer, Macromolecules, 1999, 32, 4173–4182.
23. Y. V. Kissin, Q. Zhou, H. Li and L. Zhang, J. Catal., 2015, 332, 156–163.
24. R. Ghosh, S. Maiti and A. Chakraborty, Tetrahedron Lett., 2004, 45, 6775–6778.
25. A. J. Müller, A. T. Lorenzo and M. L. Arnal, Macromol. Symp., 2009, 277, 207–214.
26. T. F. J. Pijpers, V. B. F. Mathot, B. Goderis, R. L. Scherrenberg and E. W. van der Vegte, Macromolecules, 2002, 35, 3601–3613.
27. J. C. Chadwick, G. M. M. Van Kessel and O. Sudmeijer, Macromol. Chem. Phys., 1995, 196, 1431–1437.
28. Z. Chen, Master Degree of Science, Tianjin University, 2008.
29. P. Y. Li, S. T. Tu, T. Xu, Z. S. Fu and Z. Q. Fan, J. Appl. Polym. Sci., 2015, 132(41689), 1–10.
30. J. C. Chadwick, F. van der Burgt, S. Rastogi, V. Busico, R. Cipullo, G. Talarico and J. J. R. Heere, Macromolecules, 2004, 37, 9722–9727.
31. A. G. Potapov, G. D. Bukatov and V. A. Zakharov, J. Mol. Catal. A: Chem., 2010, 316, 95–99.
32. J. Chau and J. Teh, J. Therm. Anal. Calorim., 2005, 81, 217–223.
33. M. L. Arnal, V. Balsamo, G. Ronca, A. Sanchez, A. J. Muller, E. Canizales and C. U. de Navarro, J. Therm. Anal. Calorim., 2000, 59, 451–470.
34. V. Virkkunen, P. Laari, P. Pitkänen and F. Sundholm, Polymer, 2004, 45, 4623–4631.
35. U. W. Gedde, Polymer physics, Chapman & Hall, London, 1995.
36. M. Iijima and G. Strobl, Macromolecules, 2000, 33, 5204–5214.
37. A. Wlochowicz and M. Eder, Polymer, 1984, 25, 1268–1270.
38. M. Keating, I. H. Lee and C. S. Wong, Thermochim. Acta, 1996, 284, 47–56.
39. N. Tan, L. Q. Yu, Z. Tan and B. Q. Mao, J. Appl. Polym. Sci., 2015, 132, 42487.
40. R. Paukkeri, E. Iiskola, A. Lehtinen and H. Salminen, Polymer, 1994, 35, 2636–2643.
41. Y. Zou, Master Degree of Science, Harbin Institute of Technology, 2015.
42. Y. V. Kissin, J. C. Chadwick, I. Mingozzi and G. Morini, Macromol. Chem. Phys., 2006, 207, 1344–1350.
43. Y. V. Kissin, J. Res. Updates Polym. Sci., 2013, 2, 8.
44. U. Giannini, Die Makromolekulare Chemie, 1981, 5, 216–229.

---