Synthesis of Polymers: New Structures and Methods

Polymers are huge macromolecules composed of repeating structural units. While polymer in popular usage suggests plastic, the term actually refers to a large class of natural and synthetic materials. Due to the extraordinary range of properties accessible, polymers have come to play an essential and ubiquitous role in everyday life - from plastics and elastomers on the one hand to natural biopolymers such as DNA and proteins on the other hand. The study of polymer science begins with understanding the methods in which these materials are synthesized. Polymer synthesis is a complex procedure and can take place in a variety of ways. This book brings together the "Who is who" of polymer science to give the readers an overview of the large field of polymer synthesis. It is a one-stop reference and a must-have for all Chemists, Polymer Chemists, Chemists in Industry, and Materials Scientists.

• Language: English
• Publisher: Wiley
• Release date: May 23, 2012
• ISBN: 9783527644087

Book preview

2

Polymer Synthesis: An Industrial Perspective

Sebastian Koltzenburg

2.1 About this Chapter

Despite polymer-based materials having been regarded as one of the major scientific success stories of the twentieth century, the polymer story just goes on and on! Starting with a global production volume of no more than 1.5 million tons in 1950 (which even today cannot be considered negligible on an absolute scale), the annual production of polymers rose to 245 million tons in 2008 – equivalent to an average growth rate of 9% per year for more than 50 years! Yet, this forms only part of the story, because polymer-based materials don't represent the only use of macromolecules in chemical products. Today, many applications are based on so-called functional polymers – in other words, polymers that are not used as a solid plastic material (e.g., a polyethylene bag) but rather can serve as invisible additives in complex systems, such as pharmaceutical products or detergents. Consequently, there is a very good chance that, for every chemist in the chemical industry, a polymer-related topic will one day appear in his or her project list. According to the statistics of the Verband der Chemischen Industrie (VCI; German chemical industry association), during 2009 one out of every four employees in the German chemical industry was, at some time, working in the area of polymers. But this figure is almost certainly a lower limit, based on the many activities in the field of functional polymers that are difficult to identify based on the information available.

But, despite there being such a continuous demand for educated polymer scientists, in many countries there is a significant lack of trained polymer chemists – which is good news for every expert in the field, and the main reason that books such as this one are being written today! So, perhaps the most important point that should be raised regarding this situation is, Why?

2.2 Why?

Within this context, the main questions center on:

Is there really a need for highly trained people in polymer chemistry to develop new products, processes, and applications – considering that most of the important polymers produced today have in fact been marketed for decades, and that no really new polymer chemistry has made it to the top of the charts for years?

Has industrial polymer chemistry really reached its limits? Is the problem now simply one of production optimization – to create, ultimately, a series of totally routine processes?

While there is certainly no final answer to these questions, I would like to present my very personal point of view. Shortly, I will introduce you to some general features of polymer synthesis that should be considered particularly relevant at this point. These will be followed by some examples (from both BASF and third parties) that most likely represent only three of a large number of fascinating stories to tell regarding polymer synthesis. I will also summarize how interaction between the individual players in this field – whether large industrial groups, small companies, universities, and other research institutes – can interact for their mutual benefit. Of course, it should be noted that this is a quite subjective selection – others might come to rather different conclusions!

This topic will be approached in the classical dialectic manner over thesis and antithesis, ending at – what else – synthesis!

2.3 Thesis: There Are No Limits to the Fantasy of a Synthetic Polymer Chemist

Many of the problems in polymer chemistry that some years ago appeared irresolvable are, today, state-of-the-art processes. Examples include the formation of block copolymers by controlled radical polymerization, or the increasingly broad application of transition metal-catalyzed polymerization techniques in aqueous environments. Clearly, polymer synthesis is a highly dynamic art form rather than a mature technological field.

Although, today, we are faced with an ever-increasing number of tools available to the synthetic polymer chemist, even with such new tools there remain many blank spaces on the map of polymer chemistry. The main reason for this is that the number of polymers that are accessible synthetically are absolutely comparable to – if not bigger than – the world of classical low-molar mass organic compounds, which are explored to a much better extent than the universe of polymers.

Basically, all that is required to begin the synthesis of a polymer is a compound that can form two (or more) bonds; by establishing bonds between individual molecules, the result is a polymer. As the principle can be applied generally, the number of potential building blocks is huge. Subsequently, by copolymerizing more than one different organic compound to serve as the building blocks for a polymeric material, it is possible to create a multitude of polymers that differ in their:

Chemical composition One fascinating point about polymer synthesis is that, apart from synthesizing polymers from only one chemical species, it is possible to copolymerize different monomers in deliberate ratios.¹ Simply by examining the case of statistical (e.g., radical) copolymerization, the monomer ratios can be varied almost deliberately, though still in a controlled fashion – unlike the stoichiometric reactions of low-molar mass organic compounds. This permits an enormous, practically unlimited degree of freedom to a synthetic polymer chemist in order to fine-tune the chemical properties of the polymer to as precise a state as needed.

Molecular architecture In addition, there are different ways in which the different monomers can be bonded chemically to each other, leading to the formation of block-like versus random comonomer sequences (Figure 2.1). Moreover, the polymer may be linear, branched, dendritic, or comb-like (see Figure 2.2).

Molar mass Today, by utilizing modern polymerization techniques, the molar mass of the resulting polymers can be controlled. Naturally, molar mass is an essential characteristic for controlling the properties of the final product. It must be admitted that a polymer with a molar mass of 1 000 000 g mol−1 will probably not differ very much from one with a molar mass of 1 000 100 g mol−1; however, the ability to control the order of size of a molecule can definitely help to tailor its properties.

Figure 2.1 Different types of monomer sequence in a binary copolymer.

2.1

Figure 2.2 Different architecture schemes for polymeric compounds.

2.2

In many cases, the resultant polymers do not have a uniform molar mass, but rather a molar mass distribution; this is often perceived as a flaw from the perspective of the purist organic chemist, who mostly longs for the preparation of a clean, well-defined compound with an as-well-defined molar mass as possible. However, in some cases the deliberate preparation of polymers with a very broad (sometimes polymodal) molar mass distribution will be exactly the route to follow in order to prepare a material that outperforms its clean, narrowly distributed analogs (more on this point later).

Based on these principles, it is very easy to synthesize, for example, polymers with identical chemical compositions and equal molar masses, but which are still significantly different with respect to their molecular architecture – and hence their physical properties. Whilst one structure might be completely useless, the other might be a blockbuster; an example of this can be seen in the different stereoisomers of polypropylene. Stereoirregular polypropylene is an amorphous, sticky substance with a very limited application profile, whereas its big brother – stereoregular polypropylene – is a highly crystalline, solid material that has become one the world's leading polymers. Up until the 1950s, controlling the stereochemistry of polymers was a privilege of Nature in highly complex enzymatic reactions, leading to polymers such as DNA. Subsequently, the ability of Giulio Natta to control the stereochemistry of the single polymerization steps of propene, using Karl Ziegler's catalysts, represented some of the most groundbreaking and exciting revolutions in polymer science, from both academic and industrial standpoints.

Thesis: A synthetic polymer is master over an infinite playground of novel materials with new, potentially useful properties. Go and look at what is out there!

2.4 Antithesis: We May Be Able to Synthesize Millions of New Polymers – But Why Should We Do So?

Today, the manufacture of plastics can be characterized only as a mass production. Indeed, on examining the production volumes of today's industrial polymer manufacturers, the ranking of top-selling polymers worldwide has not changed at all over the past years (see Figure 2.3).

Figure 2.3 Five of the most important (by volume) polymer materials produced today.

2.3

About one-third of the industrial production volume is covered by polyethylene, a polymer which has been number one for decades, (which clearly exceeds the successes of modern rock stars). So, apparently, this is a very boring game; is there any room left for creative polymer chemists?

It is not that no attempts have been made, but the success of relatively young polymeric materials (e.g., liquid crystalline polyesters) was found to be limited to the order of some tens of thousands of metric tons, some orders of magnitude less than the polyolefins. But, there is also a significant technical hurdle that must be overcome, in that those polymers seeking to become the world's next polymer idol must not only deliver an excellent performance, but also demonstrate a good production cost profile.

Today, most of the polymerization plants that are used to create mass products are highly specialized. Therefore, to move from the laboratory, with a new polymer in hand that is to be introduced to the market as being superior to polyethylene, it is simply not possible to take an existing plant, exchange the ethylene tank for a monomer of choice, and restart production. Rather, it is essential to build a new plant. Yet that is exactly what will ultimately cause many projects to be unprofitable! So, when deciding on the production capacity of a new plant, there will be two basic options, both of which have the potential to spoil a business plan:

Start with a world-scale capacity plant to keep the specific investment² low As a rule of thumb, the specific investment cost for a chemical factory scales with the square root of the production capacity. In other words, for a product from a plant that is fourfold larger as a reference, the investment cost burdened on each ton of the product is cut by 50%. This may be a wide margin, given the fact that most mass plastics are low-cost materials with a very low margin. In other words, in order to make money in such a market, it is necessary to have one of the biggest production sites, worldwide. However, if a new material is being developed to substitute an old, well-established material, there is always a risk that the product will not be accepted on the market. Hence, the risk associated with a world-scale production for the market introduction of a new polymer is immensely high and, in most cases, the process will fail.

Start with a small-scale production site in order to penetrate the market step by step This option certainly takes into account the above-described risks of lacking market acceptance. However, as it will be impossible to produce at a competitive cost, a more expensive product will most likely be offered to customers, compared to the old materials. Clearly, in order to succeed, very convincing arguments are needed regarding product performance, and the probability of this is low (at least in the sector of conventional bulk materials).

So, whichever decision is taken, the market entry barriers for new mass products are very high, even in markets where the regulatory hurdles are relatively low. In applications where various regulatory demands must be met (e.g., for polymers in life sciences, such as pharmaceuticals), these additional hurdles will add significantly to making the polymer chemist's life even harder. So, the question arises: Do we need new polymers at all? Or, is polymer synthesis today simply a tool that is highly developed and works fine, but does not require any further optimization?

Finally, it is important to address the topic that is not only scary, but keeps many industrial chemists awake at night: cost (see Figure 2.4). The highly optimized production processes for polyolefins, combined with the low costs of the raw materials, lead to bulk prices as low as 1 € per kilogram, and sometimes even lower. This, in turn, leads to the realization that almost any other chemistry is more expensive than that produced from steam cracker products; there is no room for anything else – the low-hanging fruits are gone!

Antithesis: Industrial synthetic polymer chemistry is at its end. Almost all materials-related problems can be solved with existing polymer chemistry. There is no need for new monomers and/or new polymers.

Figure 2.4 The industrial chemist's nightmare – in particular for polymer chemists.

2.4

2.5 Synthesis

So far, a clash of two completely diverging opinions has been witnessed, although each of these – from their own perspective – is correct. So, how can such a situation be acceptable and these two statements be combined to provide a consistent picture? At this point, it may be best to diversify from the dry sphere of business plans and to examine three case studies on the real-life requirements of an industrial polymer chemist.

2.5.1 Polymer Chemistry in Two Dimensions: Coatings

In our everyday life, many artificial surfaces are coated. For example, houses are painted with architectural coatings, eye glasses are coated to reduce mirror effects and lens robustness, and motor cars have thin layers of polymers on top of their paintwork. Indeed, although most of the materials used to build motor car bodies are still metal-based, the outer surfaces of our modern toys on wheels are in fact plastic.

Typically, a modern car coating will consist of four different layers (see Figure 2.5). The innermost layer prevents corrosion, the next layer is applied to smooth the roughness of the first layer and to provide an even surface, while the third layer – referred to as the base coat – contains the actual color of the car. Finally, the so-called clear coat is applied to the surface. This fourth – and perhaps most important – layer is responsible for the surface properties of the final product, such as gloss and hardness/resistance to scratches. Taken together, the mass of coatings applied to a single modern car is approximately 1 kg, depending on the vehicle size. Clearly, with more than 50 million cars being produced each year this represents a market that is definitely worth seeking!

Figure 2.5 Set-up of a modern automotive four-layer coating.

Photograph courtesy of Daimler AG.

2.5

The uppermost layer (clear coat) is about 40 µm thick, and consists of a polymer network that is formed in situ by a chemical reaction (the reaction is not carried out in a conventional vessel but on the surface of a car body). Consequently, the main challenge here is to run a polymer build-up reaction on a surface – a process that is far from being trivial:

The primary problem is that the surface is not necessarily flat and horizontal. Rather, as cars are three-dimensional, a coating must be prepared that can be applied to flat or curved, horizontal, or vertical surfaces. Clearly, the material cannot be applied by direct contact with a brush or a blade; it must be sprayed onto the surface, to which it must first adhere and then form a uniform layer of defined thickness. Unfortunately, this immediately creates a challenge to the rheology³ of the coating. Initially, the viscosity must be low enough to enable atomization of the coating in the spray nozzle; however, when the coating touches the surface which, in a (frequent) worst case is vertical, it must be viscous enough to prevent sagging (down-flow), as this would lead to an uneven surface and ruin the appearance of the car.

The second challenge is to run a chemical reaction on a surface and not in a reactor. A coating is basically a complex and highly optimized polymer system that comprises functional resins and crosslinker molecules as the key components (see Figure 2.6). These components must be polymerized to form a solid, high-performance polymer shell around a valuable product. Unfortunately, many of the tools employed by polymer chemists to conduct a polymerization reaction cannot be used, including stirring, the addition of compounds during the reaction to keep it alive, or the feeding of additional reactants. It is also necessary to control the polymerization enthalpy. Finally, the time available for the process is minimal; automotive coating lines are expensive to build, and they may also form a bottleneck that will limit daily production if their capacity is insufficient. The production line cannot afford a whole day for the coating on each car to cure. In fact, some people claim that it was for this reason that the world's first car to be built in a serial, high-volume production – the legendary Ford Model T – was available only in black, as this color was simply the fastest to dry.⁴

Figure 2.6 Example of components in a modern automotive clear coat.

2.6

One of the most important challenges for the coatings expert is that, unlike all other synthetic chemists, they need to create a product that is perfect at the first shot. There is no way that they can purify their product, as would the organic chemist on a routine basis. Neither is there any means to remove excess monomers that have not been integrated chemically into the polymer network (the latter process step is common in conventional technical polymerizations). Finally, the surface must be perfectly even and glossy, as excessive refinishing is not accepted by the car manufacturers. Clearly, for a product to be perfect at the first shot presents a major challenge to the polymer chemist!

In addition, the final physical properties of the product are expected to be at least as good as, or even superior to, the performance of each and every competitor's product worldwide. However, don't get frustrated – it can be done!

To summarize, coatings chemistry forms a part of synthetic polymer chemistry that is widely admired, and is an economically highly relevant field of polymer synthesis that even today is far from the end of its learning curve. Indeed, it is an area of science that continues to require highly trained, skilled, and motivated scientists.

2.5.2 Polymer Chemistry Going Broad: Effects of Molar Mass Distribution

As mentioned above, most polymers are characterized by a distribution of molar masses of the individual polymer chains; that is, almost every polymer sample is a mixture of polymers with different molar masses, an effect which is referred to as polydispersity. In the past, significant attempts have been made to produce polymers with a narrow molar mass distribution, and to prepare polymers with precisely identical molar masses. This is a consequence of the inherent desire of the synthetic chemist to produce a compound that is as well defined as possible – in just the way that Nature teaches us. Yet, only natural polymers such as DNA are really 100% monodisperse. In the following case study, it should be noted that even the absolute counterpoint to these longlasting attempts can open the way to a successful polymer in a highly competitive market. The subject here is probably the most competitive landscape in polymer chemistry over all, the polyolefins.

As outlined in Section 2.4 (The Antithesis), polyolefins are mass products that are produced under severe cost pressures. Nonetheless, skilled and creative scientists have been able to optimize these polymers, and to squeeze out the last percent of performance required for them to outperform in such a tough market segment.

A large proportion of the polyethylene produced worldwide is created by means of transition metal catalysts,⁵ while the molar mass of the product is controlled by a suitable choice of initiator and the reaction conditions. Conventional, heterogeneous Ziegler catalysts usually can be used to provide polyethylene with a relatively broad distribution of molar masses. Yet, while many scientists celebrated the advent of homogeneous, well-defined metallocene catalysts giving access to polyolefins with a relatively narrow molar mass distribution, such catalysts have remained niche products from an industrial point of view. Against all intuition, with respect to the materials' properties, polyolefins with a broad molar mass distribution are often superior to their better defined colleagues. Why is this?

The situation can be best understood by having a better understanding of materials science, and of the way in which these polymers crystallize. At this point, a brief introduction to the subject will be most useful.

Polyethylene is recognized as a semicrystalline polymer, where the prefix semi indicates that the material is not perfectly crystalline (such as low molar mass compounds, e.g. dimethoxy biphenyl, which forms amazing crystals from supersaturated solutions) but rather is a nanophase-separated material that consists of a crystalline (mostly lamellar) phase embedded in an amorphous phase. For high-molar-mass polymers, most polymer chains form part of several crystal lamellae that meander through the amorphous phases between them. Such a structure explains why polymeric materials have unique material properties: on the one hand, the crystalline part provides strength and hardness to the material, while on the other, the amorphous part is rubbery, so as to reduce the brittle nature of the final product. It is for these reasons that polymer chemists are able to produce materials that are hard, but not brittle.

Based on the explanation above it becomes clear that, in particular, high-molar-mass polymers where many chains form part of many different crystallites (causing them to link together) should have superior material properties. The crystallization of polymers is, however, a highly complex process that requires reorganization of the polymer chains within the material. It is easy to understand that, in particular, the very long chains of high-molar-mass polymers are strongly entangled, and require a long time to acquire a sufficient degree of crystallinity. On the other hand, their counterparts with a lower molar mass will crystallize more quickly, due to a higher degree of molecular mobility that leads in turn to a higher degree of crystallinity. Consequently, the most surprising finding was that polymer materials with bimodally distributed molar masses – that is, a molar mass distribution with more than one main fraction, such as a low-molecular and a high-molar mass portion – will outperform polymers with a clean, narrow, molar mass distribution.

Then how does this concern polymer synthesis? If the desired product is a bimodally distributed material, why not simply take two different batches and mix them?

The answer to this question is shown in Figure 2.4. Due to problems of viscosity, the mixing of polymers is less trivial than mixing water with acetone. The production of an intimate mixture requires much energy, and is economically unfavorable. This issue can be resolved by controlling the polymerization process in such a way that it provides the desired, bimodal molar mass distribution in a single reaction step. This process, which is referred to (among other names) as the Borstar process (as marketed by the Borealis Group) is shown schematically in Figure 2.7, together with a photograph of the actual production plant.

Figure 2.7 (a) Flow chart of a process for the production of bi- or multimodal polyolefins; (b) The industrial realization of the process.

Photograph courtesy of Borealis Polyolefine GmbH.

2.7

The Borstar process involves the use of two cascaded reactors. In the first stage, ethylene is polymerized in supercritical propane by the addition of a transition metal catalyst in a loop reactor, which leads to low-molecular-weight polyethylene. The reaction mixture is then transferred into a gas-phase reactor in which high-molecular-weight polymers are formed. The direct result of this two-stage process is an intimate mixing of the two polymer fractions, which differ in their molar masses.

This is an excellent example of how close the cooperation between the different scientific disciplines must be if polymer synthesis is to be moved forward in an industrial environment. Among the different competencies (beyond polymer synthesis) that are required for such a process to function effectively, the following ones need to be included:

The catalytic nature of the Borstar process, whereby polymerization proceeds at a transition metal center. A profound knowledge of metal-organic chemistry is also required to achieve a targeted design of exactly the catalyst geometry required to create the desired molar masses.

The crystallization kinetics and thermodynamics must be fully understood, in order to control the solid-state morphology that ultimately provides the desired material properties. This is the domain of the rheologists, polymer physicists, and polymer-processing engineers.

Finally, the concept of an in-reactor blending of two polymer fractions with different molar masses must be transferred to the industrial scale – a procedure that involves intensive interaction with plant-construction engineers.

Based on experience, this situation is absolutely typical of the way in which synthetic polymer chemists operate in an industrial setting, being integrated constantly in interdisciplinary expert networks, all parts of which have their own competencies, and which are all inter-dependent. Clearly, without a team, a single polymer chemist cannot do anything – but the others won't be able to work without him or her, either!

Although, in this particular case, a quite technical and physical network is evident, the interacting partners can have totally different backgrounds, as shown in the following example taken from the fields of biology and medicine.

2.5.3 Polymer Chemistry Meets The Life Sciences: Polymeric Drug-Delivery Systems

The design of innovative polymer-based medication systems driven by the ever-increasing need for new and more efficient therapies of disease serves as a constant driving force for polymer chemists at this interface of chemistry, physics, biology, and medicine. Evidently, polymer synthesis alone cannot help to optimize the complex routes that a drug can take within a physiological system before it eventually reaches its molecular target. Intensive interaction with experts from fields other than chemistry is required to understand how an optimal polymeric delivery system should be designed. Only an ultimate realization of the controlled synthesis of exactly the polymer needed will then be the domain of the synthetic polymer chemist.

This situation can be clarified with an example of both academic and technical relevance, namely carriers for targeted and controlled drug delivery. The dream of developing highly selective medication systems – especially in the field of cancer medicine – is to deliver cytostatic compounds (e.g., doxorubicin or paclitaxel) more or less exclusively to the desired site of action – that is, the tumor tissue. At present, because a significant part of the drug will also reach healthy parts of the body, this will lead to the well-known and often extremely adverse side effects of chemotherapy. The use of a polymeric drug-delivery system could, however, allow the dream of targeted delivery to become reality.

Nature teaches us that amphiphilic systems, such as phospholipids, assemble in water to form so-called liposomes (see Figure 2.8). Liposomes are hollow spheres on the nanometer scale, that consist of a bilayer of phospholipids arranged in a similar fashion to the cell membrane; thus, they consist of a hydrophilic (aqueous) core, around which is sited a hydrophobic shell. Drugs can be incorporated into both the core and the shell, depending on their solubility in water and nonaqueous phases; as a result, liposomes can serve as carriers for cosmetic as well as medicinal active ingredients. An example of the practical use of liposomes includes a carrier role for doxorubicin. Unfortunately, phospholipids are relatively well-defined chemical species that do not allow any extensive chemical modification. Nonetheless, considering the huge arena is available to synthetic polymer chemists (see Section 2.2), the replacement of a lipid with a biocompatible, synthetic polymer would offer extensive possibilities to tailor the properties of the resultant delivery systems to exact needs.

Figure 2.8 Schematic representation of a liposome.

2.8

As with lipids, amphiphilic block copolymers can be divided into a polar moiety (the "head) and an unpolar moiety, referred to as tail"; consequently, they are also able to form supramolecular associates in aqueous systems. Depending on the exact structure of the block copolymer, the resultant supramolecular structure can vary, and is controlled by the polymer architecture; examples include the lengths of the individual hydrophilic building blocks and the overall hydrophilic/hydrophobic balance. Just like phospholipids, block copolymers (notably triblock copolymers) can form supramolecular hollow spheres with both hydrophilic and hydrophobic compartments, that allow for the encapsulation of hydrophilic and hydrophobic substances. In analogy to liposomes, these associated structures are referred to as polymersomes.

Due to the variability of synthetic polymers, the polymersomes can be tuned to afford a wide range of functionalities, including:

Enhanced stability Due to the larger size of polymers, the associative forces between the hydrophobic moieties are more pronounced than in the comparatively small lipids; this leads to an increase in supramolecular stability. Many pharmaceutically active compounds are proteins, and all of these suffer from very short lifetimes within the blood (perhaps of only a few minutes), due to their biodegradation. Hiding such materials in the interior of a polymersome can significantly increase the time for which they can act in the body.

Drug targeting Polymersomes have relatively well-defined surfaces that can be used as anchor groups for binding to biological receptors. This property can be used to provide the selective accumulation of a drug in a specific region within the body; this is referred to as targeted delivery. The well-defined size and shape of polymersomes can also be used to passively enrich nanoparticles in tumor tissues. Tumor tissues have relatively large openings toward blood vessels, and this enables the selective accumulation of nanoparticles with a certain, well-defined particle size. This effect, which is known as enhanced penetration and retention (EPR), is currently undergoing intensive investigation for application to cancer treatments.

Triggered release In a similar manner, the physiological conditions within a target tissue can be employed for selective destruction of the polymersomes, leading to the release of their payload (see Figure 2.9). Such release can be triggered by pH, temperature, and redox potential. As an example, in normal tissues the pH of the blood is relatively constant (ca. 7.4), but in tumor and inflammatory tissues it is lower. The mechanism used to release the active compound is relatively simple: the formation of a supramolecular associate relies mainly on the hydrophobicity of the hydrophobic block. If this contains basic functions that can be protonated at low pH, the block will become hydrophilic, which in turn will cause the whole assembly to be destabilized, releasing the active compound from the core.

Figure 2.9 Schematic representation of the formation of a polymersome from block copolymers, encapsulation of drugs in its core, and its release.

Reproduced with permission from Onaca, O., Enea, R., Hughes, D.W., and Meier, W. (2009) Macromol. Biosci., 9, 129–139.

2.9

Alternatively, based on the fact that the local temperature in solid tumors is slightly higher than the regular body temperature, polymer chemists have designed hydrophobic blocks that are able to react to an increase in temperature. It is well known that polymers can undergo relatively well-defined transitions, from a water-soluble to a water-insoluble material, and vice versa. Hence, if the transition temperature of the hydrophobic block is tuned in a suitable way, then a thermally triggered release of the drug can be achieved.

In a similar way, block copolymers can react to reductive or oxidative media, a situation that often involves the use of sulfur-containing polymers (due mainly to the ease of performing redox chemistry with sulfur atoms). In most physiological systems, the environment outside the cell tends to be more oxidative than inside, and this differential can be used as a trigger for drug release.

The use of polymers in medicinal applications is, naturally, not limited to drug-delivery systems. The covalent binding of drugs to polymers (so-called conjugates) and the design of complex nanodevices for medical applications are only two among numerous, promising fields with a high potential impact in the treatment of disease.

2.6 Conclusions

Returning to the dialectic arguments proposed in Sections 2.3 and 2.4, it must be concluded that, to a certain extent, both positions are correct. In the case of high-volume plastics, for many years no new chemistry has been introduced among the very top selling materials. Yet, many problems of the future – that are issues of both economical relevance and scientific fascination – will not be resolved without help from the often unusual properties of new, innovative polymer molecules. This is particularly true for the so-called niche markets, which are served initially by a few kilograms of material but then slowly develop the to dimensions that industrial polymer chemists are used to. It goes without saying that, from an industrial perspective, such small-scale materials need to add high value to the overall system in order for them to be developed, produced, and marketed in an economically reasonable fashion.

It is difficult to assess what the future growth areas for polymers will be. Luckily, the traditional applications of typical plastics with their main applications (motor vehicle industry, construction, packaging) will presumably keep growing during the course of general economic growth, especially in Asia. However, global megatrends – such as the increasing need for energy, as well as problems related to the growing and ageing population – will create new challenges, many of which will not be tackled without the use of the often unique properties of polymeric materials. Thus, polymer synthesis is – and will always be – an essential part of this development.

Some of these developments can be realized by industrial companies alone, as demonstrated by recent BASF product launches in the field of polymers for pharmaceutical applications, or in the field of electronic polymers. However, the point must be made clear that the discovery of new polymers and new polymerization techniques is an area where nonprofit research organizations – not only universities but also institutes such as those operated by the Max-Planck or the Fraunhofer Society – can successfully follow up on new technologies and identify new trends in polymer science, without considering economical boundary conditions. In this way, such institutions can continue to investigate areas that do not necessarily pay out on an industrial ten-year business plan – and this is an extremely important role!

In my eyes, this fundamental research represents an essential contribution to the success of polymer science. A significant number of new polymer types and polymerization methods, which have been successfully introduced to the market by different companies in the past years, were initiated by discoveries and investigations conducted at universities. One case of such relatively young, new materials is the new class of polymers produced by controlled radical polymerization. An example of such a polymerization process, which appears attractive from an industrial viewpoint, is the so-called nitroxide-mediated polymerization, which was discovered during studies conducted by D.H. Solomon at the Commonwealth Scientific and Industrial Research Organization in Australia. The fundamental technology for these materials was essentially developed in a non-industrial laboratory, in cooperation with industry, and has made its way into technical reality. Also not to be forgotten is the huge class of polyolefins, which was started on the gram scale in the laboratories of Karl Ziegler at the Max-Planck Institute in Mülheim an der Ruhr. Clearly, universities and research institutes – as well as companies specialized in the customized manufacture of smaller-scale quantities⁶ – represent an essential part of the discovery chain in the search of new materials. Moreover, they can effectively nucleate the development of new polymers, even in the third millennium.

All of these cases show that – besides a strong competence in polymer synthesis – cooperation is the key to success. It is for this reason why large, multinational companies operate successful multiple external R&D cooperations: in 2009, BASF alone had almost 1900 external cooperations, of which 660 were with universities and scientific institutions. Only multidisciplinary teams of scientists from different fields and with different training can move forward in such highly complex environments. Each team member, as well as needing to be an expert in his or her field, must be able to talk to scientists from other fields – which, in many cases, is not as trivial as one might imagine! In such interdisciplinary teams, polymer chemists with both a profound knowledge of polymer science and excellent communication skills will always be invaluable and indispensable members, facing the challenges of tomorrow.

Acknowledgments

The author thanks his colleagues and friends, Thomas Grösser, Andreas Mühlebach, Reinhold Schwalm, Joseph Lupia, Melanie Steigelmann, Anja Feldmann, Erich Beck (all with BASF), Henrik Meincke (German chemical industry association), and Oskar Nuyken (Technical University of Munich) for their support during the writing of this chapter, and for their manifold fruitful and constructive comments.

¹ Naturally, for some polymerization reactions, for example, polyester formation, it is required that the number of acid groups essentially matches the number of alcohol functions; however, as long as you keep to this condition, you can use basically any mixture of different acids and different alcohols that you like.

² The specific investment is the investment cost divided by the production capacity of the plant under consideration.

³ Rheology is the science of describing the flow of liquids. Polymer solutions often follow complex, non-Newtonian flow patterns; hence, polymer rheology is rather an art than a science.

⁴ During the 1920s, the daily production of the Model T reached 9000 cars at peak times. Such an output was way beyond that of any other production system of the time!

⁵ Alternatively, polymerization can be carried out by radical polymerization in supercritical ethene under extremely harsh reaction conditions such as 2000 bar and 300 °C. Clearly, when handling such a dangerous (highly explosive) gas in very large quantities under such conditions, you must really know what you are doing!

⁶ Examples are the Kaufbeuren (Germany)-based company Polymaterials, or HOS-Technik GmbH in St Stefan, Austria, as well as numerous other companies.

3

From Heterogeneous Ziegler – Natta to Homogeneous Single-Center Group 4 Organometallic Catalysts: A Primer on the Coordination Polymerization of Olefins

Lawrence R. Sita

3.1 Introduction

The successful commercialization of polyolefins that are derived from the transition metal-mediated coordination polymerization and copolymerization of ethylene, propylene and, to a smaller extent, longer-chain α-olefins such as 1-butene, 1-hexene, and 1-octene, has forever altered the course and progress of humankind's anthropological evolution [1–4]. Indeed, with a combined global production of over 140 million metric tons per year in 2007 for just polyethylene and polypropylene materials alone – which by one estimate [5] is equivalent in volume to 44 pyramids the size of Kufu's Great Pyramid at Giza being manufactured each year (!) – the sheer magnitude of worldwide production, transportation, and end-use manufacturing of finished goods virtually guarantees that civilization, as it is known today, will be critically dependent on polyolefins for the foreseeable future. Furthermore, this dependency will undoubtedly continue to grow even stronger as new polyolefin materials are invented and commercialized to fill existing and newly created technological voids, or as replacements for legacy plastics that are plagued with real or perceived environmental- and health-related issues, such as plasticized polyvinylchloride [6].

The quest for new structural forms and subtle variations, or grades, of polyolefins has been pursued in earnest for the past 60 years, ever since Karl Ziegler and coworkers [7] at the Max Planck Institute for Coal Research at Mülheim-an-der-Ruhr first introduced a heterogeneous catalyst obtained from a mixture of TiCl4 and triethylaluminum, AlEt3, that provided highly crystalline, high-molecular-weight polyethylene, also known as high-density polyethylene (HDPE; I in Figure 3.1), through the controlled polymerization of ethylene at low pressure and temperature. Prior to this discovery, only low-density polyethylene (LDPE; II in Figure 3.1) was available as an amorphous, highly branched polymeric material obtained from the high-pressure radical polymerization of ethylene [1]. With a melting temperature of 135 °C, Ziegler quickly grasped the commercial significance and potential value of HDPE as a remoldable thermoplastic, and actively pursued the licensing and commercialization of this material within a global chemical industry that was in the early stages of embracing readily available and inexpensive ethylene (H2C = CH2) over acetylene (HC ≡ CH) as the preferred C2 commodity feedstock chemical [8]. Finally, a later development for polyethylene-based materials, was the introduction of linear low-density polyethylene (LLDPE; III in Figure 3.1) that is produced through the copolymerization of ethylene with different 1-alkenes [3, 4]. By varying the level of comonomer incorporation, a broad range of different LLDPE grades can be generated and their physical properties tailored to suit different specific applications.

Figure 3.1 Fundamental structural forms of various polyethylene and polypropylene materials.

3.1

Following close on the heels of this seminal discovery of the coordination polymerization of ethylene, Giulio Natta and coworkers [2, 9] at the Milan Polytechnic Institute in Italy employed Ziegler's compositional class of heterogeneous catalysts for the polymerization of propylene in a series of investigations that ultimately yielded the first samples of highly crystalline, highly stereoregular isotactic polypropylene (iPP; IV in Figure 3.1), in which all of the chiral centers along the polymer backbone have the same relative configuration (isotactic is derived from the Greek iso for equal, and taktikos for relating to arrangement or order). With a melting temperature higher than that of HDPE (cf., 160–166 °C), finished goods manufactured from iPP could withstand standard autoclaving conditions that are required for sterilization, in contrast to those made from HDPE. This provided one of the many reasons for the ensuing commercial acceptance and popularity of this new polyolefin thermoplastic material.

During an energetic period of fundamental discovery, the Natta group reported that several different stereochemical microstructures (i.e., different tacticities) could actually exist for polypropylene, although initially samples of these alternative forms were available in only low yield and with low regioregularity and stereoregularity. An extensive structural characterization of these materials led to a further codification of general terms to describe different stereochemical microstructures for polymers, and in the present case, to the identification of syndiotactic polypropylene (sPP) and atactic polypropylene (aPP) as uniquely different microstructural forms (V and VI in Figure 3.1, respectively) [10, 11]. As an interesting historical side note, it was apparently Natta's wife – who was an accomplished linguist – who first suggested the use of the appropriate Greek terms, isotactic, syndiotactic, and atactic, to describe the different stereochemical relationships of the relative configurations of the chiral centers along the polymer backbone of IV, V, and VI, respectively [12].

Natta also serendipitously isolated polypropylene fractions that exhibited novel elastomeric behavior that he proposed were a manifestation of properties linked to an unique isotactic–atactic stereoblock polypropylene (sbPP) microstructure (VII in Figure 3.1) [13]. In this model, the elastic properties of sbPP were hypothesized to originate with interchain associations of hard, crystalline (isotactic) domains that function as nonbonded physical crosslinks within an amorphous (atactic) matrix, with the former serving to dimensionally restore the material upon the removal of a deforming strain. Unfortunately, this sbPP material was not the principal product of a controlled polymerization for which a sound mechanism could be established to account for chain growth that, in this case, must proceed in alternating stereoselective and nonselective fashion. Indeed, both sPP and sbPP were obtained by the Natta group in only small quantities through extensive and laborious fractionation and solution-phase chromatographic separation of complex mixtures that represented the true as-synthesized polypropylene (PP) material produced using the now-renamed Ziegler–Natta (ZN) catalysts.

On looking back from the present vantage point – which is now 60 years into the future – the pace and breadth of the fundamental discoveries made by both the ZN groups within such a brief period of time still generates amazement and respect for these efforts that rapidly established the polyolefins as a class of academically interesting and commercially important materials [1–4, 12–16]. The collective results of these investigations, which probed the structures and operative mechanisms of catalyst active sites and elucidated the key concepts relating polyolefin structure with physical properties, laid the critical foundation that still supports and guides polyolefin research today. As a matter of fact, the industrial production of HDPE and iPP using ZN catalysts was already in full swing by 1957 – only a short time after the initial discovery of these materials [12]. Accordingly, with the Plastic Age well ensconced by 1963, it was a timely and fitting tribute to the enormous contributions made to society by both Ziegler and Natta that they were co-awarded the Nobel Prize in Chemistry in that year [17].

3.2 Chapter Prospectus

Given the heterogeneous nature of ZN catalysts, the mechanistic details for the key steps involved in metal-mediated chain-growth propagation and the origin of stereocontrol that is responsible for generation of all the different stereochemical microstructural forms of PP remain experimentally inaccessible for systematic investigation. On the other hand, major improvements in the activity, regioregularity, and stereoregularity for polyolefin production by later generations of ZN catalysts continue to be made through an iterative, empirically driven optimization that, during recent years, has involved the high-throughput screening of thousands of additives that help to control the number and nature of multiple active sites that are an intrinsic feature of these heterogeneous systems [1–4, 18]. In fact, due to favorable physical and processing properties that arise with subtle differences in, for instance, molecular weight, molecular weight distributions (MWDs), and long-chain branching (LCB), to name just a few, the vast majority of commercialized poly(ethylene) (PE)- and PP-based materials are still produced globally on the commodity-scale through the use of heterogeneous ZN catalysts.

From both academic and industrial perspectives, the critical deficiency of ZN catalysts remains an inability to produce fundamentally different polyolefin materials of unique structure or composition, at will, by applying a set of experimentally derived and theoretically validated first principles for the de novo design of new generations of ZN catalysts. As is often the case with heterogeneous catalysts, such advances have first required the development of homogeneous, solution-phase catalytic systems which, in the present case, are based on molecularly well-characterized transition-metal complexes that can function as discrete initiators and active site propagators for the coordination polymerization of olefins [19–27]. Here, the term homogeneous is further used to signify the uniform, or single-center, nature of the active propagating species. Indeed, it is this homogeneity of active centers that greatly facilitates the establishment of a clear mechanistic picture that can account for virtually all of the primary pathways involved in productive chain growth, as well as the elucidation of key structure–property relationships that can be used to guide additional improvements in catalyst design. To be sure, it is a fair appraisal that the current state of the art that has now been accomplished within the past 25 years for the homogeneous coordination polymerization of olefins represents one of the most successful and productive periods of academic–industrial collaborative research.

Perhaps due in large part to its success, until just a few years ago there had been growing sentiment – either expressed or implied – that polyolefin research was a mature field, and that any further innovations would be only incremental as the result of additional small refinements being made, rather than transformational in nature as the result of an infusion of new fundamental paradigms. After much initial hope and enthusiasm, the establishment of new commercial polyolefin grades produced using single-site homogeneous catalysts has also been a much more difficult path to blaze than originally envisioned, vis-à-vis the well-entrenched commodity-scale production of polyolefins using the latest embodiments of ZN catalysts. Furthermore, it has become increasingly obvious within the academic community that, perhaps in response to these influences and pressures, the next generation of young investigators has been discouraged from continuing the hunt for new advances involving either the coordination polymerization of olefins or polyolefin materials in general. Unfortunately, this decreasing interest and declining number of investigators is of great concern as they coincide with an ever-increasing need for new classes of recyclable structural materials that can support the development and commercialization of a host of advanced technologies for society that are sustainable in terms of reduced energy-, environmental-, and health-related impacts – and in all respects, polyolefins can meet these needs.

Strongly countering the rather bleak assessment for the future of polyolefins research just presented is the emergence within the past few years of several exciting new discoveries, from both the academic and industrial quarters, that are truly transformational in nature and which represent a critical evolutionary leap forward for polyolefin research and, in particular, with respect to the tailored-design of new polyolefin materials [28–34]. More to the point, while the present understanding of the mechanistic pathways that are operative during the homogeneous coordination polymerization of olefins is largely complete, this collection of key steps represents only the barest essentials that are required for achieving productive chain growth activity and control over polyolefin stereochemical microstructure or copolymer composition. The strategic layering-in of additional reaction pathways and, most notably, of dynamic fast and reversible processes that are competitive with propagation, not only dramatically increases the complexity of the overall polymerization process, but now also provides the opportunity to introduce new mechanistic control points that can be brought under external control for the purpose of greatly expanding the range of polyolefin structures and physical properties that were previously either inaccessible or simply inconceivable [29–31]. Some of these advances have even led to the successful commercial introduction of several new classes of polyolefin products of unprecedented structure and physical properties, within only a few years of the initial discovery of the new polymerization mechanisms [30]. Indeed, the pace at which these recent achievements have been made call to mind the original period of discovery enjoyed soon after introduction of heterogeneous ZN catalysts, and they serve to substantiate the belief that the beginnings of a Renaissance period for polyolefin research is being witnessed that will extend far into the twenty-first century.

The aim of the present chapter is to provide those interested in the subject with a basic working knowledge of the fundamental concepts and key discoveries that support the current state of the art that has now been established for the heterogeneous and homogeneous coordination polymerization of ethylene, propylene, and longer-chain α-olefins. As the pace of new innovations with polyolefins has rapidly accelerated during recent years, the intent is to provide a primer of the subject matter that includes an overview of the historical roots and nature of the ground-breaking advances that have been achieved to make these most recent innovations possible. Ultimately, the goal of this treatise is not to corner the reader into simply reflecting on what has been achieved in the past, but rather, to prepare them to undertake an evaluation of the current state of the art of the field for the purpose of asking the much more exhilarating question of what remains to be discovered within the scope of the coordination polymerization of olefins, and within the field of polyolefin research in general. It should be noted from the outset, however, that the aim of the chapter is not to provide a comprehensive review of all the scientific and technological advances that have been made regarding the continued evolution of heterogeneous ZN catalysts, nor of the enormous body of work that has now amassed for olefin polymerization that is mediated by homogeneous single-center catalysts. Several excellent reviews covering these aspects of polyolefin research are already available, and the reader is encouraged to explore these in order to obtain the broadest perspective of the accomplishments made and the challenges that still exist for the field [1–5, 18–34].

3.3 Fundamentals of Coordination Polymerization

3.3.1 Ziegler–Natta Catalysts

Since the earliest days after the first reports from the ZN groups, questions arose – and have persisted – regarding the nature of the active site(s) and the specific mechanism(s) by which chain-growth propagation is mediated by the heterogeneous ZN catalysts that originally were derived from a simple mixing of TiCl4 with AlEt3. In order to address these questions, a vast range of different recipes for ZN catalyst compositions have been extensively explored over the past 50 years, through investigations guided largely by empirical observations in Edisonian fashion. The result of this evolutionary process has been the emergence of several different generations of ZN catalysts, the latest of which is commonly referred to as the fifth generation. The primary distinguishing features that separate each of these generations are principally related to the productivity (activity) for ethylene and propylene polymerizations and, with the latter monomer, also to stereoselectivity. As such, it is informative to follow the course of improvements that have been made to the original TiCl4/AlEt3 recipe for the industrial production of iPP (IV in Figure 3.1).

3.3.1.1 First-Generation ZN Catalysts

Following their initial success with the coordination polymerization of propylene that yielded isolated fractions of iPP, the Natta group noted that, during preparation of the ZN catalyst, the addition of AlEt3 to a hydrocarbon solution of TiCl4 promoted a metal-centered reduction to generate a heterogeneous suspension of finely divided TiCl3 that existed in four crystalline modifications, the α-, β-, δ-, and γ-forms. In the first refinement, the reaction of TiCl4 with chlorodiethylaluminum, AlEt2Cl, at low temperatures largely provided the β-TiCl3 form that cocrystallized with AlCl3. Subsequent heating of this precursor to 160–200 °C led to its conversion to the more stereoselective γ-form. However, as originally produced, these ZN catalysts displayed rather poor activities (ca. 1 kg PP g−1 catalyst) and low stereoselectivities, such that removal of the catalyst residues (de-ashing) and separation of the coproduced atactic polymer fraction were required [35, 36]. A final improvement of this first generation of ZN catalyst involved using a mixture of TiCl3 and AlCl3, in combination with AlEt2Cl, to provide a catalyst that was much more active than pure TiCl3 – this presumably was due to an increase in the surface area for TiCl3 crystallites. Unfortunately, the stereoselectivity was still low, with the fraction of iPP present being on the order of only 90%. As a result, the manufacturing process for iPP using first-generation ZN catalysts was both complicated and expensive.

3.3.1.2 Second-Generation ZN Catalysts

During the 1970s, the Solvay company introduced a TiCl3 catalyst with improved activity by chemically treating the solid phase obtained from the TiCl3/AlCl3/AlEtCl2 mixture with diisoamyl ether to extract the cocrystallized AlCl3 and leave behind a pure β-TiCl3 phase with a greater porosity [37]. Lowering the temperature of the β- to γ-phase transition to 100 °C also limited the growth of the catalyst particles to provide a higher surface area of the active Ti sites. These low-Al catalysts demonstrated a fivefold increase in productivity (ca. 5–25 kg g−1 catalyst), and could achieve a stereoselectivity for iPP of ∼95%; this latter value was now sufficiently high to spare the need for removal of the atactic fraction [38, 39]. Finally, the ethers used in the preparation of these second-generation ZN catalysts served to herald the start of electron donor technology that was of importance to the success of subsequent generations.

3.3.1.3 Third-Generation ZN Catalysts

In an effort to further increase activity, a new generation of supported ZN catalysts was developed that employed a range of high-surface-area carriers, including silica, alumina, and Mg hydroxides. However, while this strategy proved to be effective for PE production, the higher activities associated with the new generation of supported ZN catalysts did not translate over to iPP productivity. In this regard, a major breakthrough came during the late 1960s, when the companies, Montecatini and Mitsui, independently introduced activated MgCl2 as the support for a ZN catalyst that was highly active for both PE and PP – albeit, with stereoselectivity in the latter case being only <50% for iPP [40–42]. Fortunately, it was shortly discovered that the addition of a Lewis base, such as ethyl benzoate (IX in Figure 3.2) during preparation of the activated MgCl2 yielded a ZN catalyst that was both highly active and stereoselective for iPP production.

Figure 3.2 Structures of various electron donors used in the preparation of different generations of Ziegler–Natta catalysts.

3.2

In a typical recipe, a third-generation ZN catalyst is prepared through the ball-milling of a mixture of MgCl2, TiCl4, and a Lewis base, referred to as an "internal donor (Di), to provide a mixture that is then combined with a trialkylaluminum, AlR3, as cocatalyst and a second Lewis base, usually referred to as an external donor" (De). The ball-milling process leads to the formation of very small (≤3 nm-thick) crystallites of MgCl2 in which the crystal structure is severely distorted. The adsorption of TiCl4 onto the crystalline faces of this activated MgCl2 support produces an octahedrally coordinated titanium center that serves as the precursor to the active and stereoselective sites. The choices for Di and De are usually identified through empirical optimization, but ethyl benzoate is typically employed as Di, and an aromatic ester, such as methyl-p-toluate (X in Figure 3.2), as De. The dominant active species is still believed to be associated with Ti(III) metal centers; indeed, it is conjectured that the role of the internal donor is to stabilize the Ti(IV) precursor from being over-reduced to Ti(II) upon contact with the AlR3 cocatalyst [43]. It is further believed that the internal donor plays an additional part in increasing stereoselectivity for propylene polymerization, while the role of the external donor De is to replace the amount of internal donor that is lost through reaction with the aluminum alkyl cocatalyst. This latter hypothesis is supported by observation that the most active and stereoselective catalysts are also those that allow the highest incorporation of De [44]. Finally, a major advantage of these third-generation ZN catalysts is that they are active enough so that de-ashing of the spent catalyst is not required, although, between 6% and 10% of aPP must still be physically separated from the desired iPP product.

3.3.1.4 Fourth-Generation ZN Catalysts

During the early 1980s, new combinations of Lewis base donors were discovered that led to highly active and highly stereoselective ZN catalysts, to the extent that removal of the atactic polymer was no longer required [45, 46]. In particular, the use of bidentate phthalates as Di, and alkoxysilanes (or silyl ethers) as De (such as XI and XII in Figure 3.2, respectively), in combination with a chemical process for MgCl2 activation that provides a spherical catalyst morphology with controlled particle size and porosity, are responsible for providing highly desirable catalyst properties. This has led to the tremendous commercial success of these fourth-generation of ZN catalyst systems that are, at present, the most widely employed for industrial iPP production.

At this point, a few additional comments should be made regarding the optimization of the general classes of bidentate phthalates and alkoxysilanes XI and XII, respectively, that have been employed as internal and external donors for fourth-generation ZN catalysts. Initially, the steric bulk of the hydrocarbyl group bound to silicon in XII is believed to protect the silane from its removal through interaction with the aluminum alkyl cocatalyst [47]. In addition to increasing the degree of stereocontrol at the isoselective propagation centers, the structure of these alkoxysilane donors may also have an important impact on the MWD of the polyolefin product [48, 49]. For instance, with dicyclopentyldimethoxysilane (XIII in Figure 3.2), a very high stereoselectivity is accompanied with a fairly broad MWD for iPP; together, these characteristics are ideal for applications that require a high melt strength, such as in the extrusion of pipes and thick sheets. In contrast, a narrow MWD and a low molecular weight are advantageous for fiber-spinning applications. The significant advantage of fourth-generation ZN catalysts then is that, through empirical optimization,