While much of what Bernal accomplished was not crystallographic – even within his wide interpretation of the subject – I give below an example of his non-crystallographic war work, followed by one war project which certainly did relate to materials and their functionality. Finally I say something about the work that involved me – the polywater story.

(a) Statistical analysis of bombing raids

If there was one area of science above all others in which Bernal, as scientific advisor, had left his distinctive mark it would, I believe, be on his use of statistical analysis; one might call it Bernal's war-time legacy. Bernal took its implementation to a new level, in terms of both the surveying and the interpretation of the raw data. This was particularly so for assessing the effectiveness of bombing raids of both sides in causing both physical and psychological damage.

An assortment of volunteers and students, as well as professional statisticians, were used to establish relationships between damage, tonnage of bombs deployed, type of environment ascertained from observation, previous conflicts and accurate maps. Similar methodology was used to measure loss of morale using medicinal changes and increased alcohol consumption as indicators of population morale. Interestingly, the loss of morale in the affected population was generally less than expected. The rigour and neutrality behind the data from the statisticians were a breath of fresh air compared with the unverified and sometimes exaggerated claims made by those carrying out the ‘work’, such as flight crews. One issue of great irritation to the statisticians was the ignoring of their predictions, meaning the lives of pilots were sometimes being put in danger unnecessarily as they carried out bombing missions that statisticians had already shown to be of little value.

I close this section with an anti-intuitive statistical gem: more damage was inflicted on RAF aircraft parked on the ground than during aerial combat.

(b) Habakkuk

Habakkuk was an Old Testament prophet who invited the heathen to look upon his unbelievable works in fear (“I will work a work in your days which ye will not believe, though it be told you”); Brown (2005) also suggests the name might have been chosen after a character in Voltaire’s Candide “who was, like Mountbatten, capable of anything”. So, it was perhaps appropriate that Habakkuk was used as the name for one of the most imaginative and ambitious projects of World War II: to produce man-made icebergs to serve as gigantic aircraft carriers. The reasons for Habakkuk were not obvious, nor was it obviously feasible. However, consider the following remarkable attributes:

• The cost of the building materials is virtually zero.
• The breakup cost is virtually zero since the main component simply melts away.
• The project is ecologically 'green'. 
• The project is economically independent of how long the structure operates.
• The structure is mobile with a range of thousands of miles, and so can be positioned to match war requirements.
• The structure would need to last for only a few years, unlike a commercial equivalent.

With these points taken into account the project does appear attractive. The iceberg would be powered by a number of propelling units, and steered by a combination of those units and a gigantic rudder. Inside the structure would be refrigerator units supplementing the natural environmental chill, keeping it at sub-zero temperatures. To give some idea of the feasibility of staying cool, a scaled-down model was built that took two years to melt naturally.

The icebergs were to be built using pykrete – a mixture of 15% wood shavings and 85% ice. This material, named after its inventor Geoffrey Pyke, has remarkable strength, and melts very slowly. The plan was to make each of these icebergs the size of 25 conventional aircraft carriers.

If correctly constructed, pykrete blocks are stronger than concrete by weight and their thermal conductivity is lower than ordinary ice by an order of magnitude – the wood shavings behave as thermo-mechanical barriers. It was what we now know as a composite material; then, it was 40 years ahead of its time.

It was a controversial project. Some believed it was outlandish. Others including Churchill thought it had sufficient merits to be worth a try. The pykrete block was tested by Max Perutz (see article by Christine Slingsby above) – the same Perutz who went on to win with John Kendrew the Nobel Prize for his work on haemoglobin.

A somewhat amusing scene took place in a briefing session on board the Queen Mary (see Fig. 2). A block of pykrete and a block of ordinary ice were presented to an audience, which we believe included Churchill. One General Arnold attempted to break the ice block with an axe. He succeeded easily. When trying to break the pykrete block, not only did the axe fail to cause any damage, but also the resulting shock-wave travelling up Arnold's arm caused him to gasp in pain.

In typical dramatic style, Lord Mountbatten decided to repeat the experiment, but with a revolver. The ordinary ice shattered. In contrast, the pykrete block suffered little damage. Moreover, the bullet ricocheted off, narrowly missing one of the observing officers. It is said that pykrete was demonstrated by Churchill at dinner parties. He would present two apparently similar blocks of ice, but one of them would appear not to melt, or would melt only very slowly, presumably amazing his guests.

Despite these positives, the project was turned down after the requirements expanded. The required travelling range grew to 7,000 miles, the required lifetime expanded to 3–5 years, and the weight to be carried grew from that of fighter planes only to bombers and fighters. These increases in requirements, and the increase in distances that aircraft could fly without refuelling, combined to result in the project's cancellation.

(c) Polywater

Polywater was the name given to an allegedly new structural form of water being worked on by the Russian group of Boris Deryaguin, and was being followed up by various international groups. I was aware of the work of Deryaguin mainly because of his classic publication on the DLVO theory of surface interactions. He had a high reputation in the scientific world, so when he made the surprising claims regarding the existence of a new from of water, he was given the benefit of the doubt. He claimed that water could be persuaded to condense onto confined silicate surfaces (e.g. glass or quartz capillaries) in a form that exhibited significant changes in its physical properties, such as boiling point, freezing point and density–temperature profile. Deryaguin's claims went further: he said that when ordinary and polywater droplets co-existed in proximity but with no physical contact, the polywater droplets could grow at the expense of the ‘ordinary’ water droplets. He was saying ordinary water could transform easily into polywater, implying polywater was a more stable form of H₂O.

When I was a postdoctoral researcher in Cambridge in 1968, I attended a lecture on polywater given by Deryaguin who was visiting the UK. I then wrote to Bernal at Birkbeck and asked if I could join his team which I had heard was to start research into polywater. Thus my four decade career at Birkbeck began. So in some ways Deryaguin was responsible for my association with Birkbeck. This is somewhat ironic, since he later tried to get me – and the rest of the team (including John Finney, current editor of Crystallography News) – sacked.

At the time I joined Birkbeck, the scientific community was full of polywater rumours and counter-rumours, and these extended into the wider academic world, industry and government. The issue started to feature in mainstream newspapers and TV. I was even interviewed about polywater, along with some of my Birkbeck colleagues, by the BBC.

Several ad hoc scientific meetings were arranged to address the issue. I remember one in particular in 1969, attended by about 40 physicists and chemists. The morning session consisted of everyone having an opportunity to describe their experiences making or using the alleged polywater. Reports fell into three categories:

1. I am trying hard but cannot make any polywater.
2. I can make polywater, and I am trying to measure some property (e.g. X-ray spectra, boiling point, freezing point).
3. Beware of the misleading effects of surface chemistry.

At the end of that meeting, various follow-up meetings were arranged to take place depending on the outcomes of further work. It may be that none ever took place...

Further communications on the subject became progressively quieter and less frequent when the Birkbeck group decided it was time we went public with the results of our work on the problem: that all such ‘sightings’ were nothing more than surface effects and contamination. And there was no real danger to our planet, as there would have been had ‘polywater’ indeed been a more stable form of H₂O as claimed by some polywater adherents. We said polywater did not exist; what others had found were just ‘polypollutants’ (Bernal et al., 1969; Barnes et al., 1971).

After publication we waited for the response which we expected would be an onslaught of invective against anything found carrying the name of Birkbeck. In the event it was the opposite: near-silence except for a letter from Deryaguin to Bernal urging that the Birkbeck team should be sacked for incompetence! Communications on polywater then appeared to vanish altogether, its champions seeming to have given up and moved onto other projects. Polywater was dead, killed by a deafening silence.

I think Deryaguin had been reporting what he believed to be true, based on information given to him by his experimental colleagues. He himself was more of a theoretician. A few years after our paper, Deryaguin, to his credit, retracted his previous claims in print.

A final comment

A question about Bernal: one long mission or multiple short-term solutions?

While waiting for a bus I sometimes play my mind game Categorize in which I try to place all members of a chosen group into the lowest possible number of categories. For example if the group comprised all famous crystallographers, dead or alive, known by me, I might try to sub-divide the members into two categories: those who are driven or inspired in their research as part of a life-long mission (e.g. seeking a cure for cancer; developing a new quantum theory) and those who are driven by the repeated thrill of solving many short-term problems (e.g. regularly solving the Times crossword). I can usually make an informed choice between the two but when the case of Bernal comes up, clarity disappears, as I uncover signs in his history that can apply to either of the two forms of inspiration. This ambiguity might in turn be related to another of Bernal’s renowned abilities, that of multi-tasking, evident long before the modern use of the term had been invented.

So the answer to the above question is that Bernal was in both categories.

Concluding remarks

In my opinion, Bernal was one of the greats, both for who he was and for what he did. His abilities included an amazing memory, a fertile imagination and a determination to make science work for the benefit of mankind.

I rate him as one of the four founding pillars of X-ray crystallography. Röntgen discovered X-rays; von Laue confirmed X-ray diffraction; the Braggs used X-rays to reveal crystal structure. For me, the fourth pillar is Bernal, since he linked the above discoveries and systematised X-ray crystallography.

A case could be made that Bernal should have been awarded whole or part of four Nobel Prizes: a share in the one awarded to Dorothy Hodgkin, a share in the one awarded to Max Perutz and John Kendrew, and two prizes that were never awarded; one for his founding work linking X-ray diffraction with the structure, properties and function of crystalline material, and another for his founding work on liquids, water and hydrogen bonding. As to why Bernal was never a Nobel laureate, the reasons offered range from political forces to being active in too many different areas of science. These reasons are still debated now, 50 years after his death.

One final remark. Over the last 18 months we have seen scientists around the world work together for the benefit of mankind as they delivered vaccines effective against COVID-19. The level of cooperation and the speed with which life-saving results have been delivered has been unprecedented and remarkable. This is exactly the role that Bernal wanted for science. If had lived to see this, he would have been very happy.

Acknowledgements

This article is derived from my own experience plus a host of discussions and written sources too numerous to mention, with the exception of Andrew Brown’s excellent biography (Brown, 2005), which has been my main written source. Special thanks are due to Michael O’Callaghan, friend and colleague, who has been a huge help in the research and writing of this article.

References

Barnes, P., Cherry, I. A., Finney J. L. & Peterson. S. (1971). Nature, 230, 351.

Bernal, J. D., Barnes, P., Cherry, I. A. & Finney, J. L. (1969). Nature, 224, 393–394.

Brown, A. (2005). J. D. Bernal: The Sage of Science. Oxford University Press.

 

Bernal, water and liquids

John Finney, University College London, UK

Though Bernal threw out ideas for others to develop, he kept the development of his approach to liquids for himself. Initially, his interest was sparked by the biological relevance of water, saying “My interest in the subject... came about... through my biochemical interests, in that all living structures are mostly composed of water” (Bernal, 1964). This interest was to lead to a landmark paper in 1933 published in the first number of the Journal of Chemical Physics (Bernal & Fowler, 1933). Bearing in mind that this was only 20 years since X-rays were first diffracted from a crystal, and barely a decade since the first methods of working back from a diffraction pattern to a crystal structure were successful, to tackle the much more difficult problem of liquids, where there was inadequate understanding of how to interpret the broad diffraction patterns that liquids gave, was ambitious to say the least.

Arguing that the charge distribution on a water molecule was near-tetrahedral, Bernal proposed that as opposite charges would attract, the local molecular structure should be essentially tetrahedral (see Fig. 1).

Noting that silica (SiO₂) also forms similar local structures, he developed ‘disordered’ versions of two silica structures (quartz and tridymite) to fit the X-ray data. With this model, he was able to explain a wide range of the properties of water and ionic solutions, as listed in the short summary heading that paper shown in Fig. 2.

It is also worth noting that the not-quite tetrahedral charge distribution he used is essentially the father of the effective pair potentials used today in simulations of water and aqueous systems.

In Birkbeck, he returned to the water problem in the 1950s, when he recognised that his 1933 approach “was, frankly, one of crystal structure, trying to picture water structure as that of a mixture of the analogous four co-ordinated structures of… quartz and tridymite”, and that “This was ultimately to prove rather a delusive approach, postulating a greater degree of order… in the liquid than actually exists there” (Bernal, 1964).

However, rather than return to the specific problem of water, he recognised that he first needed to understand the structures of simpler liquids. Theoretical approaches to the liquid state at the time treated liquids either as disordered (crystalline) solids (as Bernal had done in the 1933 water paper) or as dense gases. Though the disordered crystal approach was mathematically tractable and could yield correct densities, it assigned too much order to the liquid – the predicted entropy was too low. On the other hand, treating liquids as dense gases required unphysical mathematical approximations: though the entropy could come out OK, the densities that could be handled were too low to be representative of real condensed-phase liquids.

Bernal found both these approaches unsatisfactory. So instead he tried to find an approach that:

• was a concrete picture of the structure (Bernal was a crystallographer and so naturally would want to visualise the atomic arrangement);
• was consistent with Ockham’s razor (‘entities should not be multiplied beyond necessity’);
• was homologous to that of the crystalline solid as well as radically different in kind;
• had a general quality of homogeneity without the assumption of any special groups.

The most general hypothesis he came up with was to treat the liquid “as homogeneous, coherent and essentially irregular assemblages of molecules containing no crystalline regions” (Bernal, 1964). This concept he realised in the laboratory with assemblies of steel ball bearings, contrasting liquids as irregular heaps of molecules as against crystals as regular piles. Fig. 3 illustrates the differences!

 

This structural approach was indeed radically different from that of other workers on the liquid problem – and indeed Bernal apologised to “the modern theoretical physicist for introducing such a simple way of looking at things, but I believe on the whole that it is better to start with a model that has some resemblance to reality” (Bernal, 1959).

And indeed the model was successful for simple liquids such as those of the inert gases. It gave correct densities, explained density changes on melting, had the right degree of disorder and essentially predicted the observed X-ray scattering. In the late 1960s, it was also successful in explaining structures of amorphous metal alloys. The coordinates of a large laboratory model on which much of the later work was based continues to be requested and used for a variety of theoretical and practical purposes (Finney, 1970). And John Ziman, a key theoretician of liquids in the second half of the 20th century, commented that “This simple idea… is now seen to be the key to any qualitative or quantitative understanding of the physics of liquids” (Ziman, 1979). Similar comments were made in 1970 by John Rowlinson of Imperial College, one of the foremost theoretical chemists who has spent a lifetime working on liquids:

"It has therefore been hard to admit that the form or even the existence of the attractive forces has little direct effect on the structure of a liquid, as described, for example, by the pair distribution function g(r). The recent realization of this truth has followed the extensive studies… of the properties of assemblies of hard spheres without attractive forces” (Rowlinson, 1970).

Recent work is also suggesting that Bernal’s model can explain the behaviour of liquids above the critical point, where the liquid/vapour coexistence line that vanishes at the critical point continues into the supercritical region as a line of maximum heat capacity (Finney & Woodcock, 2014).

So how did Bernal move from this ‘irregular heap’ model to the more complex water problem? Simply by recognising that a disordered non-crystalline arrangement could also be built up of molecules interacting in the essentially tetrahedral fashion of Fig. 1 to produce a random network of molecules, as against the random packing of the spherical molecules of simple liquids. Fig. 4 shows a ball and stick visualisation of a random network compared to the ordered crystalline arrangement of hexagonal ice.

So Bernal’s final view of water (Finney, 2007, 2015) was that:

• Water is essentially a ‘random network’ of water molecules.

• Each molecule interacts with its neighbours in an approximately tetrahedral geometry.

• Local coordination is ideally fourfold, but with some variation.

And it compared well with experimental results. It explains the main properties of water such as expansion on freezing, the temperature of maximum density and other so-called anomalies, the mobility of hydrogen, and structural changes with temperature and pressure. It is consistent with current state-of-the-art experimental and computational work, which demonstrates that Bernal’s random network concept is essentially correct. And it has indeed helped us to understand water’s biological role – the reason that Bernal started working on the problem in the first place.

References

Bernal, J. D. (1959). Proc. Roy. Inst. 37, 355.

Bernal, J. D. (1964). Phil. Trans. Roy. Soc. A, 280, 299.

Bernal, J. D. & Fowler, R. H. (1933). J. Chem. Phys. 1, 515–548.

Finney, J. L. (1970). Phil. Trans. Roy. Soc. A, 319, 479. Data available at https://www.digitalrocksportal.org/projects/47

Finney, J. L. (2007). J. Phys. Conf. Ser. 57, 40–52.*

Finney, J. L. (2015). Water. A Very Short Introduction. Oxford University Press.**

Finney, J. L. & Woodcock, L. V. (2014). J. Phys. Condens. Matter, 26, 463102.

Rowlinson, J. S. (1970). Disc. Faraday. Soc. 49, 30.

Ziman, J. M. (1979). Models of disorder, p. 78. Cambridge University Press. 

*For a fuller account of the development of Bernal’s ideas on the structure of water.

**For an updated account of our understanding of water.

 

Bernal’s Irish connections and his enduring legacy at the Bernal Institute

Vincent Casey, University of Limerick, Ireland

John Desmond Bernal was born at the dawn of the 20th century (1901) and subsequently became one of the pioneering scientists associated with that century and the renaissance in science which it heralded. He addressed the big questions of his day. For instance: what is life and how did it originate? (Bernal, 1967). He was convinced that the answer lay within the structure of matter and the molecules where form and function would be closely entangled. Those he inspired were subsequently involved in the nascence of structural biology (see article by Christine Slingsby above) and were duly acknowledged for their pioneering work by the Nobel Committee. It is fair to assert that Bernal laid the groundwork that ultimately led to the extinction of the ‘vital force’ postulate in much the same way that Einstein’s relativity theories banished the luminiferous ether. Bernal was very much aware of the important role played by science in history, as is clear from his four-volume book set of that title (Bernal, 1969), and of its critical role in shaping society and so was active in pioneering the use of scientific methods to solve major social (Cross & Price, 1988) and political (Bernal, 1958) problems of his time. For a comprehensive account of his life and work, Andrew Brown’s biography is a most interesting and engaging read (Brown, 2005).

Bernal was born in Nenagh, Co. Tipperary, before Ireland won its independence. He was therefore a British citizen. Nevertheless, he was deeply ‘imprinted’ by his Irish childhood. He traced his roots to a mid-nineteenth century Sephardic Jewish migration. His grandfather, John Bernal, ran a successful auctioneering business in Limerick city and was also active in Limerick city politics. Around 1900, John Bernal’s son, Samuel George Bernal (J. D. Bernal’s father) bought a 147 acre country estate on the outskirts of Nenagh, complete with estate house, Brookwatson (National Inventory of Architectural Heritage, 2004). Considering those class-conscious times, this is likely to have placed the Bernals amongst the Anglo Irish landed gentry ruling class. The 1911 Irish census lists those staying at Brookwatson to include his parents, siblings, two relatives and four servants (Casey, 2021).

Bernal commemoration

Until recently, Bernal was largely unknown in Ireland although better established, if neglected, among our colleagues in the UK. In recent times, Bernal’s contributions to science and to society have been acknowledged and he has been commemorated in various ways both in the UK and in Ireland. On the occasion of the unveiling of a plaque (on 20 July 2005) in honour of J. D. Bernal by the National Committee for Science and Engineering Commemorative Plaques in his home town of Nenagh (see Fig. 1), Martin Bernal (J. D. Bernal's son) wrote: “Now, 35 years after his death, John Desmond Bernal has received this ultimate accolade”. In 2006, the Munster branch of the Institute of Physics in Ireland organised a conference entitled ‘John Desmond Bernal: Science and Society’ with keynote speakers who knew and worked with Bernal (Finney, 2007; Mackay, 2007). Earlier, in 2001, an English Heritage blue plaque was erected at 44 Albert Street, London, where Bernal lived. And in 2013, after some years of planning, the magnificent Bernal Institute was opened by the Taoiseach (Irish Prime Minister), Mr Enda Kenny, on the University of Limerick riverside campus.

Figure 2. Official launch of the Bernal Institute: (left to right) Professor Edmond Magner, Professor Don Barry (President UL), Professor Bartek Glowacki (Bernal Chair in Energy), Professor Gavin Walker (Bernal Chair in Pharma Powder Engineering), Professor Ursel Bangert (Bernal Chair in Microscopy and Imaging), An Taoiseach Enda Kenny TD, Professor Jacques Huyghe (Bernal Chair), Jon O’Halloran (BI Operations Manager), Professor Michael Zaworotko (Bernal Chair in Crystal Engineering), Professor Harry Van den Akker (Bernal Chair in Fluid Mechanics), Mary Shire (Vice President Research, UL). Photograph courtesy of UL.

Figure 3. The sun about to rise over the Penrose tiling patterns that decorate ‘The Cube’, the large lecture hall in the Analog Devices Building of the Bernal Institute. Photograph courtesy of M. J. Zaworotko.

Figure 4. ‘Skyscrapers’ – an image of crystals of coordination polymers taken by Bernal Institute post-graduate student Shi-Qiang Wang. This image was awarded a prize in the Crystal in Art Virtual Competition 2020 of the British Association of Crystal Growth (BACG). Photograph courtesy of M. J. Zaworotko.