Topic of the Month April 2012
The Twenties are generally idealised as the “golden age”. Contrary to the cliché, they were in fact a decade with both ups and downs – for Hermann Staudinger too. The chemist, who had been working in Zurich since 1912, started it spectacularly: in 1920, he published his “Macromolecular Manifesto”, which gave plastics chemistry its foundations but was rejected resoundingly by the organic chemistry establishment. The opposition that Staudinger faced as a result threatened to isolate him, but he defended his theory stubbornly and continued his attempts to prove experimentally the existence of the “giant molecules” he had postulated in theory. This was a project with an uncertain outcome at first and Staudinger suffered setbacks in his private life too: his father died in 1921 and he was divorced from his wife Dora, née Förster (1886-1964), who bore him four children in the 20 years of their marriage, in 1926. 1926 marked the start of a new stage in his career as well – and one that was to prove successful: Staudinger left Zurich and returned to Germany and a position at Freiburg University. He enjoyed recognition and fame here in the Breisgau region – and finally retired from his academic career there too. It was also a happy time for Staudinger again in his private life, once he married the biologist Magda Woit (1902-1997) in 1928, who was his companion in his scientific endeavours as well up to his death in 1965.
Germany at the beginning of the 1920s: the war was over and the monarchy was a thing of the past. Hitherto unknown Republican freedom quickly helped people to forget the authoritarian state. “Anything goes” was the message spread by intellectuals; cities became the stage for experimenting with “liberté” and “libertinage”. A great deal was changing in plastics chemistry too. New empirical findings demanded a theoretical basis, but rigid, outdated thinking could not simply be abandoned as long as the explanatory concepts needed were still nebulous. A paradigm shift was in the air, but the “experimental stage” had not been passed yet:
“The term ‘plastic’ very gradually started to establish itself via a magazine of the same name that was started in 1911 by the (German, editor’s note) chemist Richard Escales (1863-1924). Nothing at all was, however, known about how these plastics were in actual fact structured and by what principles they could be synthesised in a laboratory until late in the 20s. The progress that was nevertheless evident [...] was not based on systematic research but was instead attributable to an explosive cocktail mixed together from such ingredients as experience, speculation, acquired know-how and plenty of sheer luck.” (Heimlich 1998, 79)
Basic research was vital in this uncertain situation. Hermann Staudinger did pioneering work in this field at his chair in Zurich. He was interested in “determining the composition” (Staudinger 1938, 15 and 1961, 77) of polymers, i.e. of the fascinating class of substances that included such natural substances as rubber in addition to the innovative new synthetic ones – “proper” plastics – like celluloid (1869), Galalith (1897) or Bakelite (1908). Biopolymers include, in addition, proteins, enzymes, polysaccharides (e.g. cellulose, glycogen and pectin) as well as nucleic acids, the basic components of our genetic structure, as research in subsequent decades was to show.
Fascinating class of substances with exceptional properties
The polymers produced by mankind (“synthetic”) and the polymers that are already available without mankind doing anything (“natural”) have exceptional properties and behaviour in common that no other class of substances can boast:
• In contrast to, for example, a saline solution, which cannot be distinguished visually from clear water, polymers form colloidal, i.e. glue-like, solutions, which already move between liquid and solid states at relatively low concentrations and are sometimes viscous and sometimes jelly-like (cf. Krüll 1978a, 45).
• Other properties that should be emphasised are a marked ability to swell and form fibres, high elasticity, tremendous strength and “above all the unique combination of very high stability with multiple reactivity” (Staudinger 1961, 302; cf. ibid., 95 and Staudinger 1938, 14).
It was not, however, clear at the time what gave polymers all these physical characteristics, why a polymer, as it were, has no alternative but to display such properties. Staudinger was convinced that chemists had to find the answers to these questions: “The great variety of the individual phenomena is based [...] on the fact that the ( ) atoms are joined together in very different ways.” (Staudinger 1938, 5) In order to “obtain an understanding” of the properties of the polymers, it was therefore necessary “[...] to determine the structure of their molecules; the nature of the bonds and the arrangement of the atoms in the molecule therefore need to be specified” (Staudinger 1938, 9). Understanding the specific chemical reaction that led to the creation of polymers also promised to shed light on this matter. The aim was to have this process, which was known as polymerisation, take place in a controlled fashion and to discover suitable auxiliary materials that initiated, maintained and ended the process – not least of all in order to be able to develop versatile new plastics and manufacture them on an industrial scale.
The four basic elements of organic chemistry
Staudinger’s primary interest was therefore to decipher the “structural principle” of the polymers (Staudinger 1938, 11; cf. ibid., 5). Anyone who set out to determine their composition could not restrict himself “merely to analysing the substance” (Staudinger 1938, 9). The composition of the polymers was “basically very simple, because just a few types of atom are involved in their structure; mainly carbon, hydrogen, oxygen and nitrogen, the four basic elements of organic chemistry.” (Staudinger 1938, 6; cf. Staudinger 1961, 311) What was in the final analysis involved was “the chemistry of a single element – carbon. The outstanding feature of its atom is, incidentally, that it has an exceptional ability to bond with others of its own kind as well as with the few above-mentioned other types of atom [...]. This distinctive feature of carbon leads to an enormous number of compounds.” (Staudinger 1938, 6; cf. Staudinger 1961, 85) The crucial statement Staudinger adds is: “Knowing about the composition of an organic compound does not, however, in itself involve any understanding of its formation and properties” (Staudinger 1938, 6).
In order to dig deeper here, Staudinger put his concept of macromolecules (“giant molecules”) to scientific discussion and publicised it on an ongoing basis. Staudinger made a start on this in the essay “About polymerisation” that appeared in the “Reports from the German Chemical Society” on 12. June 1920, in which he postulated a “structure of long chain molecules” for polymers – mention being made, among others, of polystyrenes, polyvinyl chlorides and rubber (Staudinger 1961, 77). In this context, Staudinger coined the term “high polymers”, which was to be replaced by the term “macromolekel” (Staudinger/Fritschi 1922) and, finally, “macromolecule” (Staudinger 1924) in subsequent years. In Staudinger’s first essay about the chemistry of high polymers, which Priesner 1980, 351 calls the “macromolecular manifesto”, the central definition is: “Polymerisation processes [...] are all the processes in which two or more molecules combine to form a product with the same composition but a higher molecular weight” (Staudinger 1920; quoted in Priesner 1980, 35-36). A chemical molecule could “reach practically any size” (Staudinger 1961, 7) and therefore grow into a giant molecule in this way: “Identical or similar small groups of atoms join together in constant repetition to form a pattern, as a result of which macromolecules of enormous size are, finally, produced.” (Minssen/Walgenbach 1985/I, 16)
Simply defining terminology does not, however, by any means settle adequately what exactly happens in polymerisation and what enables this process to take place. This is therefore explained in further detail step by step below, based on statements made by Staudinger. An appropriate place to start is the phenomenon level, because it can be described and because it presents the mysteries that electrify both naive observers and passionate chemists. Looking back on the early days of macromolecular chemistry, Staudinger writes in 1961, 169: “It had already been known for a long time that some unsaturated compounds turn into products with the same composition but completely different physical properties when left standing for a long time, when exposed to light or when heated.” Styrene, for example, “[...] gradually becomes a highly viscous substance [...], finally forming glassy polystyrene” (Staudinger 1961, 170). Such processes, that could be described as spontaneous polymerisation, correspond to polymerisation that is triggered actively with human involvement, e.g. by heating or the exertion of pressure.
What is known as the vulcanisation of rubber is an excellent example of this: in 1839, the American chemist Charles Nelson Goodyear (1800-1860) succeeded in transforming the rubber that occurs naturally into the polymer product that we now call rubber by adding sulphur and applying heat. The undesirable tendency of the rubber to become sticky when heated and crumbly when cooled was overcome as a result.
Charles Nelson Goodyear (1800-1860)
Strictly speaking, vulcanisation is a type of polymerisation that is comparable to what is called addition polymerisation. The definition of this is that two different raw materials – rather than one and the same raw material – combine in chains to form macromolecules, as is the case – for example – with polyurethanes (see Staudinger 1961, 316). If there are by-products, water in particular, as is the case with Galalith (from casein and formaldehyde) or nylon (from hexamethylene diamine and adipic acid), for instance, this is called condensation polymerisation instead (cf. Staudinger 1961, 315-316; cf. ibid., 175 and Staudinger 1938, 15). (Chemistry of plastics)
Friedrich Carl Duisberg (1861-1935)
About monomers and car tyres
Vulcanisation is not a completely accurate example, because rubber itself is already a polymer when it combines with sulphur. The British chemist Charles Greville Williams (1929-1910) was the first to propose this hypothesis. Rubber therefore has to be considered the product of natural polymerisation that is attributable to basic components called monomers that have joined together repetitively and continuously, i.e. have combined to form a polymer. The conclusion from this is that it ought to be possible to create synthetic rubber by polymerising the isolated rubber monomer – the hydrocarbon isoprene. Experiments to do this started at Farbenfabriken Bayer in 1906 under the direction of Carl Duisburg and Staudinger already tackled this research assignment during his time in Karlsruhe (1907-1912) too (see Staudinger 1961, 5; cf. Westermann 2007, 67 and Krüll 1978b, 229) “(at this time, editor’s note) there was great demand for synthetic rubber due to the rapid growth of the car industry and the rising prices for plantation rubber on the world market associated with this – particularly in the German empire, which depended on raw material supplies from the English and French colonies. This economic situation of his country was another particularly strong incentive [...] for Staudinger to focus on polymerisation reactions like those occurring with isoprene very early on.” (Krüll 1978b, 230)
In order to make it easier to understand what follows, let us recap here: the term “monomer” is used for basic molecules that form macromolecules via standard polymerisation, addition polymerisation or condensation polymerisation. “So macromolecules represent chains of one and the same basic molecule. The number of the latter in the macromolecule is called its degree of polymerisation.” (Staudinger 1938, 11) Staudinger also characterised polymerisation as a “peculiar chain reaction” (Staudinger 1961, 315; cf. ibid., 179: “chain polymerisation”) and drew a comparison with a box of matches that has been set on fire: “just one match has to be ignited to set all the matches on fire” (Staudinger 1947, 95).
Carbon double bonds of critical importance
It is not, however, the case that all monomers are capable of forming macromolecules. (Chemically unsaturated) hydrocarbons are what primarily have the ability to create a polymer chain. In them, the carbon atoms have multiple bonds and the number of hydrogen atoms is reduced accordingly:
• Single carbon bond, e.g. ethane: each of the two carbon atoms has bonds to the other carbon atom as well as to three hydrogen atoms. As long as no atom is removed, no bonds are available to join a polymer chain (saturated state).
• Double carbon bond, e.g. ethylene: there are two bonds between the carbon atoms. One is easy to break (unsaturated state), so that the molecule can join a polymer chain. Rubber, for example, has numerous ethylene bonds (see Staudinger/Fritschi 1922, 785, quoted in Minssen/Walgenbach 1985-I, 55; cf. Krüll 1978b, 240, footnote 42).
• Triple carbon bond, e.g. acetylene: the triple bond of acetylene is so easy to break that the molecule falls apart explosively; for this reason, it is only suitable as the component of a polymer chain to a very limited extent.
Unsaturated raw materials with at least one carbon double bond are therefore the primary candidates for the production of macromolecules. This bond can be opened (“activated”) under the influence of heat, high pressure or auxiliary agents known as “initiators” (Initiators and catalyts); it then tries to find other molecules that are capable of forming a bond. This initial step is known as the “start reaction”. The chain formation process (polymerisation) that then follows leads to polymers / plastics with very different properties, depending on when the process is terminated. The termination reaction can be initiated in a controlled fashion, e.g. by adding water, atmospheric oxygen (cf. Staudinger 1961, 176) or solvents. In this context, a hydrogen atom changes its position and a saturated giant molecule is created. Polymerisability and polymerisation speed do not therefore depend solely on the structure of the molecules; they are also influenced to a large extent by agents that are added to initiate (start reaction), maintain (growth reaction) or end (termination reaction) polymerisation. Staudinger 1961, 171 says that substituents “can both increase [...] and decrease polymerisability (cf. Staudinger 1938, 7) thanks to their impact on the carbon double bond. Oxygen, for example, turns “soluble rubber with unlimited swelling properties [...] into rubber that is insoluble and only swells to a limited extent [...]. The soluble rubber remains unchanged in nitrogen atmospheres, on the other hand.” (Staudinger 1938, 26; cf. Staudinger 1961, 330 about polystyrene) What is particularly spectacular in this context is that even “minute amounts of substances can lead to exceptionally large changes in the physical properties (of macromolecular substances, editor’s note)” (Staudinger 1961, 329). “In certain circumstances, it is sufficient for the reactive substance to react with a single specific group of the macromolecule that only accounts for a small fraction of its mass; the behaviour of the entire macromolecule can be changed as a result” (Staudinger 1938, 27). Chemists have unexpected creative powers as a result – as if they were modern alchemists.
Staudinger’s first encounter with polymers
It is worth remembering that Staudinger’s interest in the structure of high-polymer compounds was aroused in direct connection with his research in the low-molecular field. After he synthesised a new class of substances – ketenes – when he was qualifying to teach at a university in Strasbourg (see Krüll 1978b, 229 for details), he carried out autoxidation experiments on them during his time in Karlsruhe, which “in addition to a number of interesting and analysable products occasionally led to undefinable, resin-like substances as well that are practically impossible to dissolve and have an unclear composition and structure. This was his first, unedifying encounter with polymer substances.” (Krüll 1978b, 229) “In connection with his initial work on isoprene, Staudinger found out that the synthetic rubber he produced was not completely identical to natural rubber – an observation that was bound to arouse curiosity and chemical interest. He therefore began to produce and make closer examinations of other unsaturated hydrocarbons like polyoxymethylene too. This means that the connection to high-polymer chemistry was established as early as 1911. When he moved to Zurich (a year later, editor’s note), he was forced to shelve this work to a large extent for the time being due to greater demands made on this time by teaching commitments, administrative assignments of all kinds and other research projects.” (Krüll 1978b, 230) He worked systematically on making a gradual shift in the focus of his research, however: “I myself have concentrated on macromolecular chemistry since 1920 [...], starting at the Swiss Federal Institute of Technology in Zurich.” (Staudinger 1961, 312) What Staudinger is referring to here is the essay “About polymerisation” that he published in 1920 and that has already been mentioned before, in which he summarises and thinks through his experiences with polyoxymethylenes, polystyrenes, synthetic rubber etc. and then proposes the thesis in question, that high polymers consist of long chain molecules: “This molecular structure in particular is often of crucial importance for the properties of macromolecular substances – both natural macromolecular substances and plastics.” (Staudinger 1961, 95) An apt example: “The lower links in the polystyrenes with molecular weights between 2,000 and 10,000 [...] are powdery and dissolve without swelling, whereas the highest-molecular representatives with a molecular weight of 100,000 and more [...] are tough glass materials that acquire elastic properties when heated to more than 120°C.” (Staudinger 1961, 95)
About primary valences and secondary valences
Heimlich 1998, 79 summarises the situation in rather direct fashion: Staudinger “was brutal in his destruction of the legend of small molecules and replaced it by his convictions about giant molecules.” Ibid., 83 says: “While molecules with what is called a molecular weight of 300 were classified as huge [...] in classic organic chemistry, Staudinger downgraded them to dwarfs in relation to the macromolecules he proposed that had molecular weights of 10,000 or more.” Looked at from our current perspective, this was a scientific revolution and a paradigm shift, with which Staudinger laid the foundations for plastics chemistry. Most of his contemporaries failed to realise the significance, however: “The response to Staudinger’s article was minimal [...]. At this time, Staudinger was still unable to provide any proof of the existence of long-chain molecules.” (Deichmann 2001, 251) Doubts about the accuracy of Staudinger’s theory dominated; there was opposition primarily to his theories about the bonding forces that existed in high polymers. The predominant view in organic chemistry at the time was that the basic molecules in polymers did not lose their independence, i.e. they were only bonded to form a unit by low electromagnetic attraction. In other words: the existence of high polymers was no reason to give up the concept of low molecules and to postulate macromolecules, which many chemists claimed were nothing more than a figment of the imagination.
Detailed information about this controversy and the people involved will be provided later on. Before this is done, here is an outline of Staudinger’s antithesis and the necessary preconditions. The basic rule is: electromagnetic attraction takes place between all the atoms of a piece of material, but the degree of attraction varies. The strongest interaction between the atoms is within the individual molecules. These inter-atomic and/or intra-molecular forces are called primary or covalences (primary bonds). In contrast to them, weaker bonding forces known as secondary or partial valences (secondary bonds) are responsible for inter-molecular cohesion (cf. Priesner 1980, 17). For his macromolecular model, Staudinger now excluded the “assumption of secondary valences” from the outset as being “not necessary” (Minssen/Walgenbach 1985-II, 13). This was a logical conclusion, because the claim was that a macromolecule was an independent entity of a size that had not been considered possible before and not just a loose collection of familiar small molecular units. With respect to the existing bonding relationships in the macromolecule, Staudinger therefore worked on the assumption of primary valences in the same way as with any other molecule. Secondary valences would only be a subject requiring examination when the discussion moved on to inter-(macro)molecular attraction.
At the latest from 1920 onwards, Staudinger was certain “that standard valence formulae explain the wide range of different polymerisation products sufficiently” (Deichmann 2001, 251; cf. Priesner 1980, 35). In other words: the “thousand to one million atoms” that macromolecules consist of are “’bonded via primary valences” (Staudinger 1961, 93; cf. ibid., 77). Since this was the case, the chemist had a stable building material that made him the architect of buildings of a variety that exceeded everything ever known in the past – an analogy that Staudinger liked to use:
“Not only molecules but also [...] macromolecules can be compared to buildings that are made essentially from just a few kinds of building materials – carbon, hydrogen, oxygen and nitrogen atoms. If there are only a few dozen or hundred of them, all that can be made with them are small molecules and, therefore, relatively primitive buildings. However, when 10,000 or 100,000 are available, buildings of endless variety can be produced: residential buildings, factory halls, skyscrapers, palaces etc. Structures can also be produced then that are unimaginable when only a small amount of building material is available. The same is true of macromolecules. It is obvious that new properties are of course observed here too that are not possible with small molecules of low-molecular substances. The number of possible macromolecular compounds is infinitely large. The size of the macromolecules also means that they can be designed in no end of different ways, again in the same way as is the case with buildings” (Staudinger 1961, 94-95; cf. ibid., 330-331; see also Heimlich 1998, 84 and Kunststoff-Museums-Verein e.V. Düsseldorf 2004, 26).
Basic research triggers industrial boom
Staudinger himself was certain right from the start that his macromolecule concept was significant not only at the theoretical level or did not just help progress to be made in the laboratory. It was a milestone in basic research that pointed the way to new approaches in the industrial production of polymers too. Staudinger expected the “in-depth understanding of the inescapable connections between the structure of the [...] plastics, i.e. the size and shape of their macromolecules, and their physical properties to lead to new ways to improve the properties of these substances [...]. It will be possible to manufacture products that are adapted to their respective use more effectively than the products supplied by nature by deliberately changing the structure” – this quotation is taken from the introduction to the first German plastics manual entitled “Fortschritte der Chemie, Physik und Technik der makromolekularen Stoffe” of which he was one of the publishers (Staudinger/Rohrs/Vieweg 1939; quoted in Westermann 2007, 169, footnote 224). “Synthetic rubber is, for example, [...] tougher than natural rubber [...] and it is more suitable for car tyres.” (Staudinger 1938, 15)
Staudinger’s self-confident predictions proved to be correct; the macromolecular concept stimulated material research and really did lead to an industrial boom soon afterwards:
• “Thanks to the co-operation with Hermann Staudinger, the second half of the 1920s and the 30s were trailblazing years for industrial research [...], since Staudinger’s macromolecular model represented a very viable theoretical resource. It was possible to tackle specific development problems and create new experimental conditions with it.” (Westermann 2007, 60)
• “During the period between 1929, when the research team at I. G. Farbenindustrie produced the first (marketable, editor’s note) polystyrene, and 1932, the group developed synthetic polymers at a speed of about one new product per day. It goes without saying that not all of them were viable, but some were of tremendous economic significance. The latter included the first polyacrylic compounds, some of which were used later on to manufacture excellent materials – such as Orlon and Acrilan – and strong, transparent plastics – such as Plexiglas. These products alone were enough to form the basis for an extensive and large plastics industry.” (Mark 1970, 104; cf. Staudinger 1938, 15)
• “Global production of high-molecular materials (plastics, synthetic resins, chemical fibres etc.) amounted to 100,000 tonnes in 1933, one million tonnes in 1950 and more than two million tonnes in 1953.” (Source: www.benzolring.de)
From purely empirical optimisation of materials to molecular material design – this, in a nutshell, is the most tangible progress that has been made thanks to Staudinger’s macromolecular concept and that is highlighted when tribute is paid to Staudinger’s historical achievements: his “concept [...] that was revolutionary at the time paved the way for the molecular design of functional and decorative polymer materials, the property profiles of which are customised for specific applications via the molecular architectures.” (Mülhaupt 2004, 1072)
Richard Willstätter (1872-1942)
Rejection in Düsseldorf
All of this was of course still a long way off at the beginning of the Twenties, when the macromolecule concept was still in its infancy. Irrefutable experimental proof of the existence of macromolecular substances had not yet been obtained; some 20 dissertations (cf. Westermann 2007, 65) were compiled at Staudinger’s Institute of Organic Chemistry at the Swiss Federal Institute of Technology in Zurich between 1920 and 1926 for this purpose, the results of which Staudinger presented to the Society of German Natural Science Researchers and Doctors when it met in Düsseldorf on 23. September 1926. Instead of the triumphal reception he hoped for, Staudinger found himself almost completely isolated, however: “Everyone [...] rejected Staudinger’s theory as being thoroughly untenable. Only Richard Willstätter (1872-1942, editor’s note), the winner of the Nobel Prize (in 1915, editor’s note) declared to his astonished colleagues at the end of the meeting that he was now of the opinion that Staudinger had provided experimental proof of the existence of long chain molecules.” (Krüll 1978b, 232; cf. Krüll 1978a, 48). The physical chemist Hermann Mark (1895-1992), who was another of the speakers in Düsseldorf, put it more cautiously: “Willstätter [...], the Chairman, indicated in reticent form during his final remarks that he supported the macromolecular concept.” (Mark 1980, 482; quoted in Minssen/Walgenbach 1985-I, 82)
Heinrich Wieland (1887-1957)
Staudinger faced further resistance from colleagues the same year when he left the Swiss Federal Institute of Technology in Zurich after fourteen years of successful work to take up a position at Albert-Ludwigs-Universität Freiburg as successor to Heinrich Wieland (1887-1957), who in turn followed Richard Willstätter at Munich Technical University. Staudinger was to stay committed to Freiburg until he retired in the spring of 1951 at the age of 70, remaining the highly respected director of the chemical laboratory at the university for a quarter of a century, even though the conditions were anything but favourable at the start. Because “serious misgivings and even open protest were expressed by the (Freiburg, editor’s note) professors” against Staudinger before his appointment (Krüll 1978b, 228) – but for political reasons and not because of his provocative macromolecule hypothesis: since the dedicated appeals he made in 1917 to decide the outcome of the First World War by negotiation rather than by fighting, because Germany was certain to suffer a military defeat due to “material inferiority” [see part I of this series], nationalistic circles had branded Staudinger a “traitor to his country”. The Freiburg “Dean Friedrich Oltmanns (1860-1945, editor’s note) travelled to Zurich specifically to meet Staudinger and take him to task personally and it took the latter a great deal of effort to make it clear to Oltmanns and the other Freiburg colleagues that he was not by any means the detractor of Germany which he was to a large extent considered to be. Staudinger became a professor at Freiburg University in 1926 and was even Dean of the natural sciences faculty for a time, although not all of his colleagues succeeded in overcoming their animosity against him.” (Krüll 1978b, 228-229)
Reservations about “gunk chemistry”
Both his personal and professional reputations remained tarnished at first: “The rejection of the concept of macromolecules by most organic chemists turned into disdain at the end of the 1920s.” (Deichmann 2001, 253) The opponents included the already mentioned Heinrich Wieland, former holder of the Freiburg chair, a specialist for organic nitrogen compounds and winner of the 1927 Nobel Prize in Chemistry. It is reported that Wieland gave Staudinger the following piece of advice at the end of the 1920s: “My dear colleague, abandon the idea of giant molecules, organic molecules with a molecular weight of more than 5,000 do not exist. Purify your products, like rubber, and then they will crystallise and prove to be low-molecular substances.” (Quoted in Deichmann 2001, 253; cf. Staudinger 1961, 79 and Krüll 1978a, 47-48)
This criticism of Staudinger was based on two associated presuppositions that were themselves questionable:
• Premise A: substances or substance blends in a non-crystalline state, such as rubber and other resins, were not chemically pure. Such “gunk” was not something that deserved investigation from the outset, chemists were only supposed to focus on pure, crystalline compounds – following possible extraction from the sticky resins. Looked at from this point of view (“chemistry of pure substances”), not only the alleged giant molecules but also their supposed alternative, i.e. clusters (aggregates) of small molecules, disappeared into thin air – because both of them could only be found if the “gunk” in question was inadequately purified, so that they were, strictly speaking, only pseudomaterials (cf. Minssen/Walgenbach 1985-II, 12).
• Premise B: the smallest atomic components of a crystal that could be determined with the help of X-rays were called basic elements or elementary cells, These three-dimensional structures in the low-molecular organic range were all larger than the molecules of the substance in question. With respect to high-polymer materials, X-ray structural analysis showed that crystalline cellulose, for example, only had an elementary cell consisting of a few glucose units. In view of past experience in the low-molecular range, it was concluded that cellulose was not a macromolecule candidate – after all, the cellulose molecule had to be even smaller than the elementary cell, which was small anyway (cf. Krüll 1978b, 232 and Priesner 1980, 30). It was of course unscientific to generalise this finding, i.e. to apply it to all supposedly macromolecular substances and solutions of them without carrying out appropriate experiments, but this did not inhibit the mainstream traditionalists in the low-molecular field much at all.
Other natural scientists apart from chemists were also dogmatic in their criticism of Staudinger’s giant molecules, such as the Swiss mineralogist Paul Niggli (1888-1953): “When Staudinger gave a lengthy lecture at a scientific conference in 1925 in which he presented his latest evidence demonstrating the existence of macromolecules, Niggli exploded right in the middle of it. He stood up and shouted across the room. ‘Such things do not exist!’” (Krüll 1978a, 232; cf. Staudinger 1961, 86) Later on, Niggli was to admit his error openly and laugh about his premature conclusion, in contrast to “colleagues, who chose to keep quiet about their misinterpretation and took over Staudinger’s macromolecular concept that they had fought so fiercely at first – as if it was a matter of course” (Krüll 1978b, 240, footnote 44).
The low-molecular dogma started to be questioned more and more and the anti-Staudinger front was far less unified than it appeared on the surface to be. The physical chemist Kurt Hans Meyer (1883-1952), for example, criticised the widespread inaccurate evaluation and/or interpretation of X-ray spectroscopic results. The head of the IG Farbenindustrie plant in Ludwigshafen (see Priesner 1980, 77 for extensive information about Meyer’s life) made it unmistakably clear that the size of the elementary cell did not dictate maximum molecular size: “It is [...] completely wrong to look for the limitations on organic molecules, i.e. on the atomic complex held together by primary valences, in the basic element.” (Meyer 1928; quoted in Minssen/Walgenbach 1985-I, 88). Hermann Mark also conceded that “an organic molecule could under certain circumstances be larger than the crystallographic basic element” (Mark 1980, 482; quoted in Minssen/Walgenbach 1985-I, 82). In his summary of the – for Staudinger frustrating – conference in Düsseldorf, annoyance is expressed too: “The situation for the representatives of X-ray structural analysis was somewhat disappointing. Before the conference, it seemed as if the small basic elements were a crucial objection to macromolecules; now, after settling their role, they were compatible with both small components and long chains.” (ibid.)
Amazingly enough, Staudinger was anything but enthusiastic about receiving “support from representatives of physical chemistry and X-ray structural analysis” (Deichmann 2001, 253). He did in fact maintain a long-running feud with Meyer and Mark. We will be looking into his reasons for this later on.
Staudinger succeeded in providing proof that “individual ( ) molecules can encompass a large number of elementary cells” (Krüll 1978b, 232) in 1927. His X-rays of polyoxymethylene showed “an elementary cell with only four methylene oxide groups [...], whereas it was, on the other hand, an undisputed fact that this substance definitely had to consist of far more such basic units” (ibid.). In spite of this – the evidence in favour of the macromolecule concept was still too tenuous to change the minds of opponents and notorious sceptics. Staudinger had to come up with proof that focussed on the core of his theory and made it watertight, i.e. that there were primary valence bonds between all the links in the postulated chain molecule with respect to electromagnetic attraction. Because only they were able to weld atoms and molecules together to form a stable unit irrespective of size (cf. Staudinger 1938, 6 and Staudinger 1961, 317) and substantiate the difference between an individual molecule and a molecular complex, between – if you like – a genuine macromolecule in the form of an integrated whole and a pseudo-macromolecule (in the sense of a combination of several molecules to form a compound that is only held together by weaker secondary valence bonds). But how was the difference between macromolecules and clusters of low-molecular particles, also known as micelles (Micelle or molecular colloid?), to be demonstrated specifically in a thoroughly convincing way?
Micelle or molecular colloid?
Staudinger 1961, 108 says: “The procedure adopted in explaining composition issues in macromolecular chemistry is exactly the same as in low-molecular chemistry, i.e. the substance is dissolved and the size and composition of its dissolved particles are investigated.” (cf. Staudinger 1938, 15) The premise: “In view of the size of the molecules, macromolecular substances can [...] only dissolve colloidally.” (Staudinger 1961, 119) If dissolved substances do in fact take on this glue-like consistency, less is, however, achieved than hoped, because it cannot be concluded that the dissolved substance is macromolecular in structure on the basis of the formation of a colloid alone; this can be a characteristic of micelles too (cf. Minssen/Walgenbach 1985-I, 10). In other words, it would only be definite that the substance consisted of macromolecules if it could be proved that “the colloidal nature [...] was due to the special composition of the substance” (Staudinger 1961, 111). Staudinger coined the term “molecular colloid” to describe this finding: “In micelle colloids, the colloid particles are loose collections of small molecules, whereas the colloid particles in molecular colloids are the macromolecules themselves.” (Staudinger 1961, 320)
Micelle colloids form, for example, in aqueous solutions of soaps and dyes (see Staudinger 1938, 8-9, Staudinger 1961, 80-81 and Deichmann 2001, 250). Soaps dissolve “normally” (Staudinger 1961, 81), i.e. without micelle formation, in alcohol, on the other hand (cf. Staudinger 1938, 8-9). This is, however, true of the high-polymer material rubber too, if menthol is used as the solvent (Staudinger 1961, 81). The crucial role played by the solvent (cf. Priesner 1980, 208) therefore makes it difficult to determine correctly whether a low- or high-molecular substance is involved. Depending on the nature, concentration and temperature of the solvent, it is evidently the case that primary valence bonds can break too, while secondary valence bonds remain stable. Even if a colloid proves to be resistant to many different solvents, there is still some uncertainty about whether the dissolved substance can be identified definitely as macromolecular. The process is not therefore conclusive enough. Staudinger himself also felt that resistance was merely “a valuable indication but not definite proof that the colloid particles are macromolecular in structure” (Staudinger 1961, 119). “( ) Determination of the size [...] does not reveal the inner structure of the particles. This question is answered via chemical experiments that are carried out here at the same time ( ), like when investigating the structure of particles of low-molecular organic compounds [...], in order to demonstrate that the atoms in a particle of a certain size are bonded by primary valences, i.e. that this particle represents a chemical molecule.” (Staudinger 1938, 15-16)
How Staudinger proved the existence of macromolecules
But how could the necessary proof be provided? This is exactly what the Japanese Emperor also wanted to know from Staudinger when he granted an audience to the man who was later to win the Nobel Prize: “Professor, are macromolecules merely concepts that enable many different phenomena to be explained or is there strictly scientific proof of their existence too and, if so, what methods are used to supply the proof?” (Staudinger, 1961, 115) The answer was: experimental proof of the existence of macromolecules has been provided when a substance “is transformed into derivatives without changing (or reducing) its degree of polymerisation” (ibid.): “Transformation of this kind [...] into derivatives with the same degree of polymerisation is known as polymer-analogue conversion.” (Staudinger 1938, 17)
This reasoning is based on the assumption “that a secondary valence bond [...] does not survive chemical conversion unchanged. [...] The secondary valences must disappear at least in the transition state of the reaction.” (Priesner 1980, 342) If the colloids prove to be resistant even so, i.e. their degree of polymerisation does “not” change even “in such profound chemical conversion processes as esterification or saponification”, it is definite that “all the basic molecules [...] are bonded to each other via primary valences” (Staudinger 1938, 17) and not by secondary valences, which “are definitely destroyed [...] by such chemical intervention” (Mark 1980, 482). In a nutshell: in this case, macromolecules and not micelles must be involved. “Such proof [...] has been provided for cellulose, starch, glycogen, rubber and various plastics, including polyvinyl acetates” (Deichmann 2001, 411).
Staudinger produced the first experimental results as early as 1922 together with his doctoral student Jakob Fritschi with the hydrogenation of rubber, i.e. the saturation of its carbon atoms with hydrogen. The hydrorubber created proved to be “just as tough as the original substance and produced only colloid solutions as well” (Krüll 1978a, 47), “which prompted the research scientists to work on the assumption that macromolecules were involved rather than micelles or relatively low-polymerised molecules” (Westermann 2007, 67). The original publication states: ”Rubber is [...] a very high-molecular hydrocarbon with numerous ethylene bonds [...]. The ethylene bonds can be saturated partially or completely by adding halogen, hydrogen halide or sulphur chloride in vulcanisation, without the colloidal properties changing, i.e. without the ‘macromolecule’ disintegrating.” (Staudinger/Fritschi 1922; quoted by Krüll 1978b, 240, footnote 42) “These conclusions about the macromolecular structure of rubber and hydrorubber were confirmed by experiments conducted on polystyrene between 1923 and 1926.” (Staudinger 1961, 84)
“Polymer-analogue conversion is a method that is based exclusively on the application of organic chemical principles, is intrinsically logical and is very convincing.” (Priesner 1980, 342) An excellent example of “how scientific progress [...] can be achieved using a modified concept with the help of established methods” (Deichmann 2001, 249). The dispute about Staudinger’s macromolecules was not over, however: although their existence had been confirmed in principle and the plastics industry took advantage of Staudinger’s model, there were still a number of controversial details and unsettled issues “relating in particular to explanation of the physical properties of the high polymers” (Priesner 1980, 208).
Staudinger’s dispute with Meyer and Mark
The physical chemists Hermann Mark and Kurt Hans Meyer, who have already been mentioned briefly, were particularly critical observers of Staudinger’s research. Both of them worked at the central laboratory of I.G. Farben in Ludwigshafen. Mark was appointed Professor of Physical Chemistry at Vienna University in 1932 and established a polymer chemistry teaching and research programme there. Meyer, who used to be on the staff of Fritz Haber and Richard Willstätter, left I.G. the same year and took up an appointment as professor at Geneva University. Both of these research scientists had “already acknowledged the existence of macromolecules in 1928 – following initial rejection of them – but had modified Staudinger’s concept in this context” (Deichmann 2001, 404). Mark and Meyer agreed with the assumption of “primary valence chains”, but considered that “micellar forces”, i.e. secondary valences, acted between them (cf. Priesner 1980, 337). A concept that Staudinger called the new or “second micellar theory” (Staudinger 1960, 90) and thus rejected as outdated. A close look reveals that Mark and Meyer were in fact firmly in the Staudinger camp, except that they tried by fluctuating outwardly between macromolecules and micelles to make more distinctions in the theory of high-polymer materials and, if necessary, to correct Staudinger. Meyer and Mark insisted in particular that the significance of secondary valences should not be underestimated:
• Meyer 1929a writes: “Staudinger assumes that association to form molecular groups or micelles has only been determined with soaps, that hold a special position because of their salt character. We would like to draw attention to the fact that they can be detected in all higher-molecular compounds [...].” (quoted by Priesner 1980, 96)
• Meyer 1929b writes: “In contrast to Staudinger, [...] we observe the structure of the [...] high polymers in solution, when Staudinger says [...] that they have no micellar character. We, however, are convinced that cluster or micelle formation plays a key role in the high-polymer materials in solution too.” (quoted in Priesner 1980, 108)
• Priesner 1980, 337 comments: “Whereas to Staudinger there was a clear distinction between primary and secondary valences and no attempt was made to obtain information about the nature of the individual types of bond, the physical chemistry approach demanded stronger distinction. [...] The strength of both primary and secondary valences was not observed to be constant; instead of this, it varied according to the structure of the molecules. As far as size was concerned, a strong secondary valence could therefore very definitely correspond to a weak primary valence.”
Hermann Mark (1895-1992)
On the basis of what we know now, Mark and Meyer were in actual fact “not completely wrong” (Krüll 1978a, 48), because it is true that macromolecules can “definitely in suitable conditions form micelles in their solutions too (ibid.; cf. Krüll 1978b, 233). “More or less highly aggregated groups of molecules are also solvated in colloidal solutions alongside individual molecules, depending on the solvent concentration. Micelles are just as real as individual macromolecules” (Priesner 1980, 115), although the term is reserved exclusively for “aggregates of small molecules” nowadays (Priesner 1980, 82). Minssen/Walgenbach 1985-II, 99 go even further: “The concept of chemical primary valence with its defined bonding relationships does not explain all the characteristics of a substance.” Denaturation of enzymes could, for example, be described best by saying that the primary valence bonds were maintained, whereas the secondary valence bonds were broken. Ibid., P. 60-61 goes on: “In the case of what are known as biological macromolecules, e.g. nucleic acids and ‘proteins’, particularly enzyme proteins, the sensitivity to heat [...] cannot be explained any more via a molecular structure involving primary valences. [...] Staudinger is wrong when he says that the reason for the instability when exposed to heat is because the molecules ‘disintegrate’ due to the elimination of primary valences (1926). The introduction of secondary valences accordingly allows [...] the description of more complicated structures and behavioural patterns than is the case when the theory is reduced to standard valences.” Staudinger’s concept needed “to be abandoned as too limiting. To this extent, his opponents are celebrating a belated triumph.”
Differences and deficits
Another point of contention was Staudinger’s insistence on the stick model in macromolecular theory that he propagated vehemently into the 1940s; he thought that the chain polymers were “always rigid, stretched structures. He liked to use long Mikado sticks to illustrate his ideas.” (Priesner 1980, 208; cf. Deichmann 2001, 254 and Krüll 1978b, 233) Meyer, on the other hand, already emphasised in 1928 “the elasticity of rubber with the tendency of the isoprene chains to form curves and to get tangled up, an interpretation that was new and correct at the time, as we now know” (Deichmann 2001, 254).
What was known as Staudinger’s viscosity formula, which assumed a correlation that was determined by the laws of nature between the degree of polymerisation and/or the molecular weight of macromolecular substances on the one hand and the viscosity level of their solutions on the other hand, was a source of further dissent. Staudinger’s widow remembers: “This formula occurred to Hermann Staudinger on a beautiful autumn day in 1929 while we were on one of our rambles in the Black Forest and we then used it in the laboratory on numerous occasions to determine molecular sizes – while leading to just as many attacks from the scientific community!” (Magda Staudinger 1976, 42) Hermann Mark considered the proportionality assumed by Staudinger to be too vague and started viscosity experiments of his own in the 1920s. His “goal [...] was to find a relationship that was based on precise mathematical principles.” (Priesner 1980, 111; cf. ibid., 348) “There was an additional complication for Staudinger in the form of the causal link between his (narrow, editor’s note) idea of the ‘form’ of the macromolecules and the accuracy of his viscosity law.” (Priesner 1980, 190)
Staudinger’s critics proved to be mistaken about the core issue – molecule size – however: “Whereas Mark and Meyer were right in assuming that there were strong inter-molecular forces, they continued to underestimate the length of the primary valence chain (of the macromolecule) for many years.” (Deichmann 2001, 254; cf. Priesner 1980, 82 and 208). It should be pointed out that neither of them claimed to be able to determine molecular sizes on the basis of their domain (X-ray structural analysis) (cf. Priesner 1980, 347) and that they said they had no particular ambitions in this area either: “In all our work, [...] we have considered it much less important to determine that chains exist and have given much higher priority to finding out exactly the location and shape of the chains, the bonds between the links in the chain, the micellar forces etc.” (Meyer 1929b; quoted in Priesner 1980, 108)
Feud between colleagues instead of coalition
Opposing positions here that were not irreconcilable in principle, that definitely had more in common than separated them. And, although “in a sense both sides were right” (Krüll 1978b, 233), the controversy refused to end, becoming increasingly fierce and polemic as the years went by. Priesner 1980, 211 concludes that Staudinger and Meyer/Mark had “no reason” at all “to compete with each other, because the former was at home in the preparative organic chemistry field, while the latter focussed on physical chemistry”. Both sides were committed to different angles and issues, which complemented each other rather than ruling each other out – in spite of “two different starting points”, the results were “very similar conclusions” (Priesner 1980, 58). Priesner therefore wonders what might have prompted the rivals to fight each other ruthlessly, instead of forming a coalition to combat established low-molecular thinking – the real opponent: “The opportunity of benefitting mutually from the skills of the other via close co-operation and of helping the macromolecular theory to make a breakthrough against the resistance of the strong group of the proponents of the low-molecular ‘aggregation theory’ [...] was squandered.” (Priesner 1980, 58; cf. ibid., 349)
Priesner found out the reason for the feud specifically once he analysed the correspondence between Staudinger and Mark/Meyer, which forms part of the Staudinger estate that is kept at the Deutsches Museum in Munich: “What this controversy involved was not [...] a theoretical dispute [...], but the question of to whom priority was due with respect to a position that was maintained by both parties in a similar way.” (Priesner 1980, 349; cf. ibid., 351) There never was a quarrel between Staudinger and Mark/Meyer in the sense of a dispute of fundamental significance about scientific theory, because Staudinger’s attacks to all intents and purposes ignored Mark’s and Meyer’s “actual Achilles’ heel”, the relatively small size of the primary valence chains” (Priesner 1980, 93). Psychology and not logic was therefore required to understand what fuelled the controversy (cf. Priesner 1980, 350).
Since Staudinger considered himself to be the “intellectual father of macromolecular chemistry” (Priesner 1980, 250) and had the necessary self-confidence to claim that he alone was “responsible for determining the composition of high polymers” (ibid., 184), he understood “any assessment of his work that was not unreservedly positive to be an attack” (ibid., 240). For this reason, it could be said that he suffered from over-sensitivity (Priesner 1980, 240) or even a “kind of academic claustrophobia” (ibid., 330). And that is not all: in his determination to smother any perceived attempt to dispute his claim to priority at the earliest possible stage, Staudinger opted to go on the offensive before he needed to defend himself at all: “Staudinger initiated the controversy at the start and as it went on, there are no examples of Mark or Meyer attacking Staudinger themselves either.” (Priesner 1980, 351)
It was Mark in particular who tried repeatedly to calm things down. He explicitly took sides with Staudinger, because “we essentially think the same, i.e. that the high-molecular substances consist of long chains that are held together by primary valences, and are only unclear about the most appropriate term for this” (letter to Staudinger of 11. December 1928; quoted in Priesner 1980, 99). On another occasion, Mark pointed out: “I think that we [...] should proceed together and should not emphasise differences between our personal views that are in my opinion minor; if we did, the high-polymer community could easily make the mistake that is only too familiar from politics; that a major issue was not given close enough attention and was not presented clearly enough because of minor differences between opinions that were not far apart.” (Letter to Staudinger of 2. November 1928; quoted in Priesner 1980, 94)
Staudinger, however, dug his heels in even more and contradicted himself into the bargain. Priesner 1980, 350 reveals the paradox “that Staudinger claimed he was being copied by Mark and Meyer, while stating at the same time that their theory was wrong. The only way for anyone to get into such a situation was if he thought that any activities by other people in the high-polymer chemistry field were [...] a violation of scientific rights he claimed for himself, if the person in question also advocated the existence of large molecules.” Staudinger’s contemporary, Wallace Hume Carothers (1896-1937), the inventor of nylon, drew conclusions about him that were just as embarrassing. In a discussion held in 1932, Carothers started by paying tribute to Staudinger’s tremendous importance as a scientist, before outlining his personal weaknesses: “Opinions abandoned by former opponents are presented and refuted again; apart from this, the contributions made by other research scientists are not acknowledged to a sufficient extent.” (Carothers, 1932; quoted in Deichmann 2001, 255) As late as 1936, Meyer still criticised “Staudinger’s practice of repeatedly misquoting other research scientists and accusing them of holding the opposite of their true views” (quoted in Priesner 1980, 197).
The macromolecule has several fathers
In view of this, the final question that the scientific historian still has to answer is the extent to which Staudinger’s uncompromisingly formulated claim to priority is justified, not only with respect to Meyer and Mark (synchronic angle) but also with respect to possible predecessors from earlier days (diachronic angle). In other words: was the macromolecular concept the major discovery of a pioneer or did Staudinger benefit from the work of others before him and more or less tacitly uncover something that had already been discovered but then forgotten? The answer is complex:
• Priesner 1980, 350-351 writes: Staudinger’s first announcement about high-polymer compounds (= Staudinger 1920) “indisputably contained all the fundamental principles of macromolecular chemistry, but these principles were not exclusively original creations by Staudinger”, because “a large proportion of them had already been thought and expressed before him. It goes without saying that Staudinger was more than a compiler, but he was that as well.” (cf. Priesner 1980, 336)
• Meyer 1929b says: “It is not correct that ‘the publications by K.H. Meyer essentially reproduce opinions that Staudinger has been expressing in numerous publications and lectures for years’”. Together with Mark, he, Meyer, built “not on Staudinger but on general teaching in the past, which is outlined very soundly in Emil Fischer’s work about polypeptides and proteins in particular” (quoted in Priesner 1980, 107; cf. ibid., 95-96).
• Deichmann 2001, 249-250 uses rubber as an example to talk about different opinions, traditions and fashions that determined the concept of macromolecules and alternated up to 1930: “In 1860, the British chemist Charles G. Williams (1829-1892, editor’s note) expressed the suspicion that rubber could consist of numerous individual components, while the work done by other research scientists supported the theory that a large molecule was involved. The idea that the naturally occurring substances rubber, cellulose, starch and protein had a high-polymer structure was a widespread view at the end of the 19th century. Thinking then started to go in the other direction, represented most significantly by Carl Harries (1866-1923, editor’s note), who was one of the most well-known rubber chemists of his time in Germany and was convinced that rubber had a low-molecular structure.” Priesner 1980, 9 qualifies: “However, Harries too initially expressed the opinion that ‘rubber’ was ‘a hydrocarbon of very large, unknown molecular size’”. (see also Krüll 1978a, 45)
Staudinger’s achievements cannot be overstated in spite of all this: even if the macromolecule has several “fathers”, it is justifiably identified primarily with Staudinger. What is certain is that Staudinger is “the first chemist who confirmed the existence of macromolecules experimentally” (Deichmann 2001, 254). Krüll 1978b, 233 stresses: “It remains a fact that they (= macromolecules, editor’s note) have dimensions unsuspected in the past that are the reason for their specific properties and behavioural patterns which differ completely from molecules of ‘normal’ size. Credit is due to Staudinger for being the first to have claimed and proved this. Indirectly, however, we owe the basic theoretical concept behind macromolecular chemistry – and thus modern plastics chemistry – to Staudinger’s numerous scientific opponents in particular too. Because their constant doubts and counterarguments are what forced Staudinger to keep on looking for new ways and means to prove his theories.” Priesner, to whom Staudinger is “indisputably one of the most important polymer chemists ever”, delivers a balanced verdict from a historical distance: “All in all, the macromolecular concept is not the work of a single person. Like almost always in scientific history (and not just there), it becomes clear when a closer look is taken that the development of human insight is to a large extent the result of the achievements of many different people, co-operation between whom is the source of but also precondition for scientific development and human society.” (Priesner 1980, 359-360)
Staudinger’s position in Germany was already being considered in a similar way at the beginning of the Thirties: more and more chemists sheepishly joined the macromolecular camp, while the number of sceptics and adversaries shrank. Although this was gratifying for Staudinger, a new challenge was already lying in store for him in 1933, when the Nazis came into power: would the scientist, who faced political hostility, be allowed to continue his research unhampered or would he be unable to enjoy the results of his work?
Read more in part 3 of the series, which is our "Topic of the Month June 2012.The life and work of Hermann Staudinger. Part 3: 1933-1945
Copyright (for the whole series of the life and work of Hermann Staudingers):
Guido Deußing/Dr. Markus Weber