'Fracture Mechanics and Elastomers seminar: 50 not out'

Fifty years ago a scientific paper was published in the Journal of Polymer Science that initiated fracture mechanics work on elastomers. The authors were R.S. Rivlin and A.G. Thomas - young researchers working at the time at the laboratories of the British Rubber Producers' Research Association (BRPRA) in Welwyn Garden City. In September this year, at the Hertford home of the Tun Abdul Razak Research Centre (the successor organisation to BRPRA), a one-day seminar took place to mark this important anniversary.


Rivlin and Thomas at the seminar

'Fracture Mechanics: 50 not out', sponsored by TARRC and its consultancy unit Rubber Consultants, attracted a good audience of industrialists and academics who were welcomed by Director of Research, Dr Andrew Tinker. There were more than fifty participants from all over the world, and distinguished speakers included the authors of the paper, another ex-BRPRA physicist Professor Alan Gent, and others who have contributed to the development of the fracture mechanics approach. In fact, the first four speakers had all received the prestigious Goodyear Medal from the Rubber Division of the American Chemical Society.

First to talk was Professor Rivlin. After graduating from Cambridge University he spent seven years researching various aspects of electrical communications, before joining BRPRA in 1944, where he almost immediately became interested in the mechanics of materials. He embarked on the development of a phenomenological theory of the mechanics of vulcanised rubber that would be valid for large elastic deformations. The theory he developed, known as finite elasticity theory, has had a great influence on research on the mechanics of materials, and established Professor Rivlin as one of the principal architects of non-linear continuum mechanics. Another subject of interest in his early days at BRPRA was the tack of adhesive tape. He saw that the tack could be characterised by the work required to peel the tape - the work of adhesion. This concept, later termed the peel energy, formed the basis for studies of the adhesion of elastomers. Such an energy balance approach to the interpretation of a phenomenon was exploited in the work, with Professor Thomas, on the rupture of vulcanised rubber - the subject of the paper celebrated at the Seminar. They established the application of the Griffith energy-based criterion to the fracture of vulcanised rubber by a series of simple, but ingenious experiments coupled with elegant analysis drawing on Professor Rivlin's earlier studies on finite elasticity theory. Their seminal paper led to the later application of the approach to a wide range of failure processes in elastomers, such as fatigue, wear and cutting. Professor Thomas followed with an account of his early work demonstrating that the relationship between tearing energy (or energy release rate) and rate of crack growth is a characteristic of the material and is dependent of geometrical considerations specific to a particular specimen. Both Professors Rivlin and Thomas gave lively and interesting reflections on the ideas and research they undertook at that time, and provoked many questions from the floor during the discussion period.

The morning session was completed by a brief overview, presented by the programme organiser, Professor Graham Lake of the University of East London, of the application of fracture mechanics to cutting phenomena and failure in cord-reinforced elastomers.

The afternoon session began with a talk by Professor Gent, who set out to explain why cracks propagating in a rubber sheet tend to turn sideways. He was able to show that the fall of energy release rate with the change of propagation direction can be counterbalanced by the development of a weak direction parallel to the applied force as the rubber is stretched. Professor Lawrence presented intriguing results on the bursting of cylindrical balloons. High-speed photography showed that cracks in the highly strained membrane propagated at a speed many times faster than the normal stress wave velocity for elastomers.

Will Mars from Cooper Tire & Rubber Company, USA spoke of the need to have experimental procedures for the measurement of the crack-growth characteristic that could be reliably and consistently used by industry. He reported on an inter-laboratory comparison of measurements on standard compounds utilising a specified test procedure. The results demonstrated a lack of consistency particularly for filled compounds. The feeling on the floor was that it was very important to involve industry in further research to develop sufficiently defined procedures, and that the possibility of a standard was of great interest with many companies keen to participate. Dr Mars expressed appreciation for the care that Prof. Thomas must have exercised in his original experiments to produce such high quality data.

Dr Alan Muhr from TARRC gave an interesting review of experiments carried out at the laboratories at Brickendonbury on the fatigue of laminated bearings including recent work on cyclic tilting. He described how, with simple analytical formulae, the results could be interpreted in terms of fracture mechanics. The data showed that cracking can occur quite rapidly, a surprising observation in view of the excellent performance of such bearings in the field. Dr Muhr felt that bearing standards should address this type of failure more clearly.

In recent years, engineers have combined finite element techniques and the fracture mechanics approach to try to predict performance and service life of rubber components. Two presentations described several examples of such practical exploitation. John Harris from the Materials Engineering Research Laboratory in Hertford reported good agreement between predicted and observed crack growth for three small mounts. He concluded that, though FEA is a powerful tool for predicting tearing energy and fatigue life of components, software developers need to improve ease of use. The formal presentations were rounded off with a lively talk by James Busfield, from the Materials Department at Queen Mary, University of London, on predicting fatigue cracking in a gearbox mount. He found good agreement with test data for cracks over 4mm in length.

Before the Chairman for the afternoon session Dr Keith Fuller, Head of the Engineering Products and Design Division at TARRC closed the day's events, James Busfield noted with admiration the relative youth of both Professors Rivlin and Thomas at the time of the publication of their paper. He pointed out that the former was his own age and the latter was no older than his PhD students present in the audience. He remarked again on the importance of the work they had published, a point with which the audience unanimously agreed.