Pd–phosphinesulfonate bravely battles the “vinyl halide insertion copolymerization” barricade



This overview provides insights into the current state-of-the-art solutions to insertion copolymerization of functional olefinic monomers. The challenges in insertion copolymerization of functional olefinic monomers, with a special emphasis on vinyl halides, are highlighted. The crucial design of the Pd–phosphinesulfonate [Pd(PO)] enables up to 3.6 mol % incorporation of vinyl fluoride (VF) in an ethylene–VF copolymerization reaction. In a significant development, insertion copolymerization of industrially relevant functional olefin, that is, vinyl chloride (VC), was unambiguously ascertained, and a detectable amount of VC (0.4 mol %) was incorporated (at the chain end). In a detailed investigation, the in situ existence of (PO)Pd[BOND]H species during the polymerization was revealed, and it was demonstrated that these are indeed responsible for VC incorporation. © 2013 Wiley Periodicals, Inc. J. Polym. Sci., Part A: Polym. Chem. 2014, 52, 1–6


Synthetic methodologies, which provide a predictable control over product properties, are indispensable in molecular and macromolecular synthesis. Insertion polymerization, a metal catalyzed polymerization method that affords polymers with a desired architecture, is extensively used for olefin polymerization.[1] Despite this progress, insertion (co)polymerization of electron-deficient functional olefins remained inaccessible until recently. This is in part due to the functional group intolerance of the highly acidic early transition metal catalysts employed in classical Ziegler–Natta systems. In 1996, Brookhart finally succeeded and reported the first insertion copolymerization of functional olefin; although only a few percent of acrylate could be incorporated.[2] His success was partly attributed to the clever use of late transition metals such as palladium and nickel as metal precursors. The seminal work of Brookhart laid the foundation of insertion copolymerization of functional olefins.[3] Built on this foundation, Drent et al. introduced a neutral phosphinesulfonate ligand system in 2002, and insertion copolymerization of a wide range of polar monomers was made amenable.[4] Although the current surge in the metal catalyzed insertion copolymerization of polar olefins has achieved limited success, clearly there is still a long way to go in order to demonstrate their commercial potential.[5]


Fundamental Challenges

There are at least three fundamental challenges that must be addressed, and the success or failure of insertion (co)polymerization of functional olefins will rely on how many of these can be precisely addressed. First, the foremost obstacle has been functional group coordination to the metal to form a σ-complex (Fig. 1, left), leading to a complete loss of activity. Second, in a coordination competition between the two olefins to the metal, the olefin (α-olefins) wins over the electron deficient functional olefin due to its poor π-donating ability. Third, the last bottleneck has been the formation of stable chelate after the insertion of the functional olefin, which halts the growing chain. A paradigm shift in the insertion copolymerization of functional olefins could be achieved if these fundamental challenges are effectively addressed. However, the current state-of-the-art, most successful catalytic system, palladium-o-phosphonebenzenesulfonate (Fig. 1), only partly satisfies these criteria by shifting the equilibrium between cis σ-complex and trans π-complex towards the right, and enables the incorporation of functional olefins to a certain extent. The functional olefin incorporation is facilitated due to special features associated with the Pd-Phosphonebenzenesulfonate system: a) an overall neutral charge on the complex, b) an enhanced stabilization of the trans π-complex, and c) a Pd-back donation into the π-orbital of the functional olefin (dπ-pπ).

Figure 1.

The decisive equilibrium between a nonproductive cis σ-complex and the productive trans π-complex.

Table 1 summarizes the relative incorporation of functional olefins in the insertion copolymerization of ethylene and polar olefins. The copolymerization is generally carried out in a toluene solution or in a neat monomer at 60–100°C under an ethylene pressure of 0.1–3.0 MPa. Insertion copolymerization of various functional olefins affords copolymers with incorporations ranging between 2 and 52 mol % (Table 1).[6] However, the molecular weights are considerably lower than ethylene homopolymers, and the activity also decreases compared to ethylene homopolymerization.[17]

Table 1. Incorporation of Functional Olefins in Ethylene-Functional Olefin Copolymerizationa
FGIncor. (mol %)Mn (103 g/mol)FGIncor. (mol %)Mn (103 g/mol)
  1. a

    Analysis details (such as PDI, etc.) have been excluded.

  2. b


SO3Me13.9[11]1.8C([DOUBLE BOND]O)Me7.7[12]9.6b
C([DOUBLE BOND]O)NHiPr3.1[13]2.4COOH9.6[14]6.1
NC([DOUBLE BOND]O)(CH2)32.6[15]3.0F3.6[16]1.6

Insertion Copolymerization of Vinyl Halides

Incorporation of vinyl halides in linear polyolefins by insertion polymerization could provide an attractive alternative to the long sought challenge of halogenated polyethylene.[3] Traditionally, halogenated polyethylene is produced either by halogenation of polyethylene (PE)[18] (e.g., chlorinated PE) or dehalogenation of homopolymers of vinyl halides.[19] Polyvinylchloride (PVC) is commercially manufactured by radical polymerization of vinyl chloride (VC); however, this method provides a very limited control over the polymer architecture and leads to various “defects” in the PVC. Metal catalyzed coordination polymerization of VC may offer direct access to “low defect” PVC with enhanced control over the polymer microstructure and properties. Hence, insertion (co)polymerization of VC has long been on the radar of the scientific community.[20] This is not only because of scientific curiosity, but also because of its industrial significance. The two predominant inhibition modes in functional olefin polymerization, that is, coordination of the monomer functionality and the chelate formation after insertion, are believed to be less relevant in VC (co)polymerization due to relatively lower coordinating abilities of C[BOND]Cl species (Fig. 2).

Figure 2.

Two distinctly different modes of interaction of functional olefins with metal catalysts: inhibition through lone pair donation (σ-complex formation) (left), and insertion followed by β-halide elimination (right).

Initial reports using the classical Ziegler–Natta system claimed insertion copolymerization of VC; however, these were later shown to proceed via a more common radical pathway.[21] Jordan and colleagues evaluated the performance of metallocene catalysts (based on early transition metals) in insertion polymerization of VC.[22] It was concluded that VC inserts into the M[BOND]C bond, but the intermediate undergoes rapid β-Cl elimination to yield M[BOND]Cl and oligopropylene. The failure of this system can be attributed to the large differences in bond dissociation energies of the Zr[BOND]Cl (116 kcal/mol) relative to the Zr[BOND]Me (67 kcal/mol) bond, which most likely favors the β-Cl elimination process. In contrast, late transition metals are less prone to β-halide elimination, because of the smaller differences in bond dissociation energies of M[BOND]Cl relative to M[BOND]Me bond (Pt[BOND]Cl vs. Pt[BOND]Me: ca. 10 kcal/mol). Provoked by these findings, Jordan et al. investigated the reaction of VC with a range of late transition metal olefin polymerization catalysts.[23] It was demonstrated that VC coordinates to the metal (though relatively more weakly than ethylene and propylene); however, after net 1,2-insertion, it undergoes quick β-Cl elimination: a behavior that was anticipated to be less problematic with late transition metal catalysts. Recently, a bisphosphine monoxide palladium complex that incorporates various functional olefins was evaluated in the insertion copolymerization of VC with ethylene, but the incorporation was either undetectable or negligible.[24] However, it is interesting to note that insertion copolymerization of allyl chloride with ethylene resulted in a copolymer with 0.9% incorporated allyl-chloride.[25] Although VC could not be incorporated, insertion copolymerization of vinyl fluoride (VF) using a late transition metal catalysts was successfully demonstrated by Jordan et al. (Scheme 1).[16] The copolymerization experiments were carried out in toluene at 80 °C with 80–240 psi VF pressure, which resulted in a linear fluorinated PE with low levels (0.1–0.45 mol %) of VF incorporation (Scheme 1). The in-chain incorporation of VF was confirmed by a proton ([BOND]CH2CHFCH2) NMR signal at 4.49 ppm (2JH-F = 48 Hz), 19F resonance at −179.4 ppm, and 13C peaks at δ 94.6 (d, 1JC[BOND]F = 168 Hz), 35.6 (2JC[BOND]F = 36 Hz), and 25.5 (3JC[BOND]F = 5 Hz) corresponding to [BOND]CH2CHFCH2[BOND] in-chain fluoride unit. The neutral phosphinesulfonate–PdMe catalysts were further tuned, and under the best polymerization conditions a significantly high VF incorporation (up to 3.6 mol %) was reported.[26] Among the various chain ends, observation of [BOND]CH2CHFCH3 and [BOND]CH2CH[DOUBLE BOND]CHF chain ends indicates that α-F-alkyl-Pd species are formed in the copolymerization reaction, which supports the insertion mechanism. Apart from this, control experiments unambiguously rule out the radical mechanism and implicate the insertion mechanism. The success was partly attributed to the high C[BOND]F bond energy, which most likely supersedes the β-F elimination and facilitates VF incorporation.

Scheme 1.

A general representation of ethylene–VF copolymerization.

Insertion of another vinyl halide, that is, vinyl bromide (VB), in a Ta-complex was investigated by Wolczanski, and, surprisingly, VB insertion was found to be faster than the VC insertion rate.[27] This behavior was rationalized by assuming that the halides operate according to their inductive capacities [Cl (σI = 0.47) and Br (σI = 0.45)]. Although the insertion of VB was investigated,[28] unfortunately insertion (co)polymerization of VB remains attended. Thus repeated attempts of Jordan[22-26] and Nozaki[24] could not yield the desired insertion (co)polymerization of VC, but key barriers were identified: (1) VC coordinates weakly compared to analogous ethylene/propylene coordination, resulting in lower incorporation; (2) VC insertion most likely follows a net 1,2-insertion pathway and subsequent β-Cl elimination, leading to irreversible catalyst poisoning; and (3) It is believed that the VC insertion rate is slower than the β-Cl elimination rate. Apart from identifying these challenges, it was hypothesized that the novel catalyst design that favors 2,1-insertion and has a higher insertion rate than β-Cl elimination will most likely enable the VC (co)polymerization. In spite of all these efforts, the insertion copolymerization of VC remained elusive until recently.[29]

Insertion Copolymerization of VC

In their pursuit to realize the insertion copolymerization of VC, Mecking et al. recently disclosed a method that finally incorporates a detectable amount of VC as a chain-initiating group.[30] Inspired by the performance of phosphinesulfonate ligands (Scheme 2, L1) in functional olefin copolymerization (in general) and VF copolymerization (in particular), they set out to evaluate catalysts 1–5 (Scheme 2) in ethylene–vinyl chloride copolymerization. The unique reactivity of these complexes is associated with the net neutral charge on the metal and enhanced back donation from metal into the monomer acceptor π-orbitals. By the virtue of their design, these catalysts destabilize σ-complex formation and enhance π-complex formation with the incoming functional olefinic monomer. These features make phosphinesulfonates the best ligand system for functional olefin polymerization.

Scheme 2.

Representative o-phosphonebenzenesulfonate ligand L1 and the best performing Pd-catalysts.

In situ NMR experiments devised to track the reactivity of VC with palladium Complex 1 revealed net 1,2-insertion of VC, which generated propylene after quick β-Cl elimination. The in situ generated propylene reacts faster than VC with 1 and releases thermodynamically stable products. However, no [BOND]CHCl unit could be spectroscopically detected in these products. As anticipated, the net 1,2-insertion of VC and β-Cl elimination constitutes the major roadblock in VC (co)polymerization. The higher tendency for propylene insertion rather than VC (as indicated by NMR experiments) prompted the authors to plan the polymerization experiment with substantially lower ethylene pressure. Hence, catalyst 2–5 were screened for insertion copolymerization of VC under moderate ethylene pressure (3–4 bars), and were found to be active. The resultant polymers were analyzed by NMR spectroscopy, which ruled out the existence of midchain [BOND]CHCl units. But the same analysis revealed proton and carbon signals that are very characteristic for 2-chloroalkane compounds. The proton NMR of the polymers displayed a sextet at 4.08 ppm, which is a very characteristic signal pattern for the CH3CHClCH2-R type of unit. The existence of [BOND]CHCl units at the polymer chain end was unambiguously ascertained by recording various 1D and 2D NMR spectra, and by comparing it with various 2-chloroalkanes. Detailed polymer analysis revealed that catalysts 2, 4, and 5 incorporate about 0.03–0.1 mol % of the VC (Table 2). Application of an independently prepared and isolated Pd–D Complex 3 as the catalyst resulted in a much-enhanced incorporation of VC (0.4 mol %). The VC unit can be incorporated either by 1,2-insertion of VC into the Pd[BOND]Me bond and ethylene insertion, or via 2,1-insertion of VC into the Pd[BOND]H/D bond and subsequent ethylene insertion. The former mode of insertion would generate polymers of type (X), and the later of type (Y) (Scheme 3). Among the two possible insertion pathways, 13C labeling investigations ruled out the existence of the 1,2-insertion mode, as the anticipated product (X) could not be spectroscopically detected. More than 95% of the 13C labeled methyl groups were incorporated in nonfunctionalized polymer chain ends. On the basis of these leads, the authors postulated that the observed CHCl incorporation might originate from a 2,1-insertion of VC in a Pd[BOND]H species, followed by ethylene insertion (Y). The first confirmation of this hypothesis came from a NMR tube deuterium labeling experiment. Treatment of VC with an independently prepared palladium deuteride complex resulted in deuterium incorporation/scrambling at the [BOND]CH2 end of the VC (Scheme 3, Z). These results were further corroborated by the fact that the application of an isolated palladium–deuteride Complex 3 resulted in the highest incorporation of VC (Table 2, Run 2).

Table 2. Palladium-Catalyzed Ethylene–Vinyl Chloride Copolymerizationa
Serial No.Cat. (μmol)Inc. VC (mol %)n (PE)/n (mCPE)Mn (NMR) (g/mol)
  1. a

    Polymerization details have been ignored for simplification purpose.

12 (47)0.1065.0:14,500
23 (19)0.3991.8:13,900
34 (45)0.0299.8:18,400
45 (45)0.0483.4:113,400
Scheme 3.

Decoding insertion modes: (X) 1,2-insertion of VC, (Y) 2,1-insertion of VC, and (Z) deuterium scrambling in VC.

However, if Pd[BOND]H species are responsible for 2,1-insertion of VC and subsequent incorporation, one might wonder why even 0.1 mol % VC is incorporated with Pd[BOND]Me catalysts (Table 2, Run 1). As the authors indirectly point out, a fair assumption would be that the ethylene inserts into the Pd[BOND]Me bond first, and after multiple ethylene insertions followed by β-hydride elimination releases PE, and generates the “desired” Pd[BOND]H species. A simplistic model based on this assumption is depicted in Scheme 4. The in situ generated Pd[BOND]H species can catalyze 2,1-insertion, and incorporates VC in the polymer chain. Alternatively, part of the Pd[BOND]H species can undergo ethylene insertion and the subsequent cycle to regenerate the Pd[BOND]H species (Scheme 4). In direct validation of this assumption, ethylene homopolymers with vinyl chain ends have been found in the resultant polymers. The authors even report the ratio between PE and mono-chlorinated PE (mCPE) (Table 2, Column 4 and Scheme 4). The best-performing Catalyst 3 produces PE:mCPE in the ratio of 1.8:1. This observation suggests that ∼36% of the Pd[BOND]H species initiate the formation of mCPE, which means roughly every third polymer chain that is initiated by Pd[BOND]H species produces VC containing polymers (mCPE).

Scheme 4.

Plausible in situ generation of Pd[BOND]H species from a Pd[BOND]Me catalyst and the subsequent production of mCPE.


In conclusion, the induction of novel Pd–phosphinesulfonate system enable insertion (co)polymerization of ethylene with various polar vinyl monomers with incorporations ranging between 2 and 52 mol %. Insertion copolymerization of VF and VC has been investigated on a few occasions using the phosphinesulfonate ligand system. By the virtue of their special design, Pd–phosphinesulfonate catalysts incorporate a significantly higher amount of VF (3.6 mol %) in ethylene–VF copolymerization. Among the vinyl halides, the most curious case is of VC insertion copolymerization. This is not only because of the scientific challenge, but also because of its industrial significance. Initial efforts using the cationic (α-diimine)PdMe+ catalyst led to the isolation of (α-diimine)Pd(Cl)(propene)+ species and no VC incorporation.

In a significant development, incorporation of VC in insertion copolymerization was unambiguously ascertained for the first time. In spite of the limited VC incorporation, the most significant contribution by Mecking et al. is in revealing the in situ existence of Pd[BOND]H species during the polymerization and demonstrating that these are indeed responsible for 2,1-insertion and VC incorporation. Contemporary strategies that would (a) suppress the problematic chain-walking; (b) enhance VC insertion over β-Cl elimination; and (c) offer access to net 2,1-insertion by tailoring the design of catalysts might facilitate higher VC incorporation. Now that a VC insertion to the Pd[BOND]H species has been demonstrated to initiate the ethylene polymerization, the remaining challenge is VC insertion to Pd[BOND]C species that can be followed by subsequent ethylene insertion. It is reasonable to assume that future investigations will focus on VC insertion to Pd[BOND]C species and higher VC incorporation.


The authors thank the DST-Ramanujan Grant (SR/S2/RJN-11/2012) for its support. DCPC India (CoE SPIRIT) and CSIR-National Chemical Laboratory (MLP026326) are gratefully acknowledged for financial support.


  • Image of creator

    Shahaji Gaikwad was born in 1986 in Maharashtra, India. He received his M.Sc. with a specialization in organic chemistry from the University of Pune, India. He was qualified for the National Eligibility Test conducted by the University Grant Commission in 2011. Subsequently, he joined the group of Samir Chikkali at the National Chemical Laboratory, Pune, India as Junior Research Fellow (JRF). Currently he is working on insertion (co)polymerization of functional olefins.

  • Image of creator

    Satej Deshmukh was born in Amravati, Maharashtra, India in 1986. He received his master's in organic chemistry from the Swami Ramanand Teerth Marathwada University (S.R.T.M.U.), Nanded, India. After qualifying for the National Eligibility Test, he joined the group of Samir Chikkali at the National Chemical Laboratory, Pune, India as Junior Research Fellow (JRF). He is working on the synthesis of Pd-phosphine sulfonate catalysts for insertion (co)polymerization of functional olefins.

  • Image of creator

    Samir Chikkali (1978) earned his Ph.D. under the supervision of Dietrich Gudat (University of Stuttgart, Germany). Subsequently, he did postdoctoral research with Joost Reek (University of Amsterdam, Netherlands) and Stefan Mecking (University of Konstanz, Germany), the latter as an AvH Fellow. In 2012, he joined the National Chemical Laboratory, Pune, India, to begin his independent research career. His scientific interests are functional olefin polymerization, renewables to polymers, P-stereogenic supramolecular phosphines, and cooperative catalysis.