Article published in the Journal of Industrial & Systems Engineering

S F Bush

The academic purpose of the Centre is the development of a science of manufacture for which the experimental foundations are derived from direct engagement in process and product research, and from factory and business operations. The theoretical framework for this endeavour is the four-level system of molecular processes, mechanical equipment, factory control systems, and company management.

This academic purpose is motivated by the long-term problems in British manufacture expressed in familiar fashion by the following contemporary quotations:

“The good news is that we have a remarkable science base in the UK, but we are not very good at linking it up with industry to reap the (economic) rewards”^[1].

“In general, a productivity gap exists between UK and US manufacturing. Boosting productivity in the UK is vital to improving competitiveness and overall living standards”^[2].

These are two facets of an enduring problem, bearing directly on the country’s future^[3] which, by its complexity and subtlety both technical and human, have resisted so far all attempts to solve it generally. It is surely worthy of research in a university whose research range stretches from nanoscience to marketing and which is located in the largest concentration of manufacturing enterprises in Britain.

Of course the Centre for Manufacture is not alone in its efforts to address aspects of this linked problem, which in any case, like most serious diseases, is multifaceted and varied in its incidence.   But in our view CfM has settled on the most comprehensive approach yet tried, aimed mainly at the most intractable part of our manufacturing industries, namely the small and medium-sized enterprises (SMEs).

We are focused on the process industries (which represent about 60% of all manufacturing). That is to say on those industries – specifically polymers, food, chemicals, metals, fibres and plastics – where materials are changed both in chemical composition and physical form.   We also have a major focus on healthcare as an expanding market for our process industries, and on electronic systems as an enabling technology, particularly for embedded control as part of process and product designs^[4].

Manufacturing processes for the above industries have a substantially common research base – particularly in the fields of mixing, rheology, reaction kinetics, physio-chemical interfaces, energy transfer and systems technology. Polymer composite materials and processes exemplify all these elements, – are continually expanding in scope and application, – and represent an area where the Centre has an international research position^{[5] [6]} and a track record of successful process innovation* which is being added to each year**.

Most ideas for new products and processes come, in fact, from within manufacturing industry and its immediate customer base. However, whether originating from within or without – a university or independent inventor for instance – there will be a long knowledge-and-cash intensive path to traverse before a commercial product or functioning process is obtained. For this reason, an integral part of CfM’s programme has been to construct a unique techno-economic (TE) mathematical model, using the cell-balance principle^[3], which connects the research design, production and sales functions of a business in quantitative terms.

The TE model in effect provides the intellectual scaffolding for our four-level system approach to the basic manufacturing problem alluded to by the quotations above^{[1] [2]}. Field data for this model are progressively being obtained both from research-based new processes applied in FTSE-sized companies at home and abroad and from the 70+ regionally-based SME projects which the Centre has run under its Innovation, Strategy and Technology Assessment (ISTA) programmes in the last 5 years^[7]. These programmes are run in part with NEPPO Ltd which with CfM disposes of the resources needed to traverse the design and innovation pathway in the plastics and allied sectors, thus providing an established organisational model for other industrial sectors. NEPPCO Ltd has now spun off its own subsidiary SURGIPLAS to develop its own and CfM’s innovative ideas in the healthcare field.

There are currently three new innovation areas – one process, two product – which are increasingly engaging the Centre’s attention. New processes for the conversion of waste materials into saleable materials are a pressing environmental need. These new processes are likely to involve all of the common research base defined above, as well as the TE model. Our embryo ‘Rubicon’ process for combining waste tyre rubber with cement to give a concrete with some elastic properties is a case in point, based as it is on our CIRRAC rubber – calcium carbonate – polymer alloys.

In the product field, the pending “End of Life (EOL)” legislation has the potential to bring about a paradigm shift in (consumer) product design, away from maximising short-term value with its throw-away corollary, towards maximising maintainability and component replaceability. To a considerable degree, new research under this heading will feed into the second area of product innovation opportunity, namely the Centre’s “product pipeline” concept^{[4] [8]}, where, as in the pharmaceutical industry, the market is anticipated and research and design are done in advance. There is potentially a wide gain from this approach: where ideas emanating from university research groups or specifically from the ‘pipeline’ objective meet the TE criteria, there will be a very high chance of public sector funding for the follow-on-research.

Overall then, CfM’s academic vision is to establish a new corpus of knowledge in the fields described above, using the standard Baconian methodology, and having a particular relevance to manufacturing enterprise^[8]. It will do this mainly in partnership with SMEs – and if opportunity presents – with other groups in the University***. It will continue to seek research contracts and grants-in-aid at levels around the four-year average (£190,000 per annum per academic excluding the £400,000 STRIX grant in 2001/02^[9].

To accelerate progress towards what is an ambitious goal, the Centre will seek to increase its academic establishment to at least the number (4) foreseen when it was set up^[4]. In the light of the accelerating flow of results, the Centre will further accelerate their publication. It will continue also to disseminate the basic concepts, and where appropriate the results, through the medium of its final year undergraduate courses^[4] [8] – principally Engineering Foresight, Product Design with Polymers & Composites, Process Manufacture – and its Technology for Business post-graduate short courses.

Footnotes

* Pre-2004: SMARTFORM, SAFIRE, GRANEX, MAP, CIRRAC, ROTOFOAM processes and materials.

** In 2004: RUBICON (a waste tyre rubber compound), ROLLET (for food distribution), BIOKAB (for healthcare)

*** At present: Textiles, Chemical Engineering, Mechanical Engineering and the Medical School.

References

[1] Professional Engineering 17 (14) 18 August 2004 p.35.

[2] “Manufacturing at the Crossroads” – Engineering Employers Federation report, December 2001.

[3] S F Bush “On the Importance of Manufacture to the Economy”, Trans Manchester Stat Soc 169 (1999) 1-46.

[4] Centre for Manufacture “Future Development”, CfM report, April 2004 (for Faculty Working Party.)

[5] E.g. (1) Bush, S F: References in “Scale, Order and Complexity” in Polymer Processing. Proc Inst Mech Eng 214E, pp217-232. Invited Paper for Millennium Edition 2000.

[6] E.g. (2) Methven, J M et al “Manufacture of Fibre-Reinforced Composites by Microwave Assisted Pultrusion (MAP”, Polymer Composites 21 (4) (2000) 586-594.

[7] G D Davidson “The competitiveness of UK manufacturing and the role of innovation” M.Phil thesis, CfM, UMIST 2004.

[8] “Where we are and where we are going”, CfM report (Jan 2003) (for official opening of STRIX Centre).

[9] Centre for Manufacture “Statistical Record” Rev IV, 22 June 2004.

Paper to Polymer Processing Society Americas Regional Meeting, Florianopolis, Brasil, 7th-10th November 2004

S F Bush

Introduction

It has long been appreciated that the addition of glass or other stiff fibres to a thermoplastic or thermoset in a suitable fashion usually brings increased stiffness and strength to the processed material. In the case of injection moulded thermoplastics, the glass fibres have until the 1980s been very short, usually in the range 0.3-1.0 mm.

In the case of thermoset compositions the fibres have generally either been 25-50 mm discrete fibres as in sheet moulding compounds (SMCs) or continuous woven structures. If 25-50 mm discrete fibres are used, they are usually in tows (bundles) of 30 or more individual filaments, either constructed into a loosely woven mat and then impregnated with thermoset resins or scattered in a random overlapping fashion on to a layer of resin with further resin poured on top. In either case, a form of semicoherent fibre structure is obtained within the polymer liquid, this structure being maintained after the composite sets to solid. This structure is one of the two main reasons why fibre-reinforced thermoset composites commonly show greater strength and stiffness than do the thermoplastic varieties based on short fibres, which do not usually form such structures, the other being the chemical cross-linked character of the thermoset.

Whatever the specific objectives laid down for the composite, two factors in particular will determine how it meets these objectives. These are (i) fibre-polymer contact, and (ii) fibre management. The over-arching requirement is of course that of minimum cost of the product as defined by its required shape and load-bearing characterisation.

Paper to the 19th Annual Meeting of the Polymer Processing Society, Melbourne, Australia, 7th-10th July 2003

S F Bush with O K Ademosu

Abstract

Rotomoulding is an established process for forming relatively large hollow structures by rotating a mould containing polymer powder (typically polyethylene) either about two perpendicular axes or about one axis combined with a rocking back and forth motion along the second, perpendicular axis. Up to now, if such a rotomoulded hollow form needed to be foam-filled, the hollow form has had to be made first in one operation, demoulded, and then, as a comparatively costly second operation, taken to another station where it is filled with polyurethane foam. The UMIST Rotofam process allows the foaming step to proceed at the same time as the moulding step, giving a solid outer skin of one material and a foamed interior made of another. The paper describes experiments on the Rotofoam process at both laboratory scale and full-scale as rotation speeds, feed materials and temperature-time profiles are varied. Large bore steam pipe insulators, damage resisting post covers, cold store doors, harbour fenders and pallets, all made on our industrial collaborators’ plants will illustrate the results obtained in practice from this new industrial proces. A further variant – Rotofil – in which long glass fibre filaments are distributed into the polyethylene skin will also be briefly described.

Paper to 19th Annual Meeting of Polymer Processing Society, Melbourne, Australia, 7th-10th July 2003

S F Bush with J D Tonkin and F G Torres

Abstract

Ref 1 (in the Society’s Carl Klason [1999] memorial issue of International Polymer Processing) summarised the main experimental and theoretical results from a major long-term programme of research to produce and apply long glass fibre compounds to the extrusion of pipes, and the injection moulding of relatively complex shapes.

This work has been commercialised over the last 10 years under the trade name SAFIRE – the acronym for Self Assembling Fibre Reinforcement – which records the fact that a major part of the technology is concerned with the use of fibre management devices which cause fibres, above a certain length dependent on concentration, to form themselves into coherent mat structures within the melt as it flows towards the shaping die or moulds. These fibre management devices have been protected by international patents during the on-going commercial exploitation phase. The formation of these all-important mat structures is dependent on the number N of virtual touches experienced by one fibre in the presence of the others. N is give as A.c(l/d) where A depends on the flow field. This paper records new results obtained with this technology for blowmoulding and for thermoforming of extruded sheet.

References

[1] S F Bush, Long Glass-Fibre Reinforcement of Thermoplastics, International Polymer Processing 14 (1999) 282-90.

[2] S F Bush, Fibre reinforced polymer compositions and process and apparatus for production thereof, US Patent 5 264 261 (1993)

[3] S F Bush, Filament Separation in Liquids, US patent 5 035 848 (1991)

[4] S F Bush, F Yilmaz, P F Zhang, Impact strengths of injection moulded polypropylene long-glass fibre composites, Plast Rubber Composites (1995) 24 (3), 139-147.

[5] S F Bush, M Dreiza, J D Tonkin, Blow moulding of long-glass fibre composites, Plast Rubber Composites (1999) 28, 379-384.

[6] F G Torres and S F Bush, Sheet extrusion and thermoforming of discrete long-glass fibre reinforced polypropylene, Composites Part A. 31 (2000), 1289-94.

[7] D R Blackburn and O K Ademosu, Investigation of the production of rotationally moulded composites, Proc 9th Intl Conf Fibre Reinforced Composites, Ed A G Gibson, Conf Design Consultants publ (2002), 402-07.

Report II on the design of the Rollet

S F Bush

Introduction

The basic design given in our first report (3 August 2001) is unchanged, but we have decided to alter the initial manufacturing method for the superstructure (the side panels and shelves).

We now propose to form the superstructure side panels by rotomoulding a polythene powder instead of thermoforming (TF) and extruded sheet of either ABS or polypropylene SAFIRE sheet. The reason for the change is that on the initial quantities we are working on (10,000 per annum) thermoforming of extruded sheet would take the manufactured cost above £60 which we all agree is too high. (Note that the price of SAFIRE or ABS sheet is volume sensitive, so this manufacturing route will still be a candidate once volume has built up.)

Besides the cost consideration, rotomoulded panels have distinct advantages of their own:

1. They will be essentially doubled-walled of low temperature impact resistant polyethylene. As a variant the cavity could be filled using our proprietary ROTOFOAM technology, although the present design doesn’t require this.
2. Without foam, the side panels will be very abuse resistant: with foam, they will be super-abuse resistant. This manufacturing method can therefore be retained long-term for niche markets requiring this performance.
3. A second variant is a SAFIRE reinforced roto-moulded skin where higher stiffnesses are required for certain applications. Again, this will NOT be necessary for our main target application – food distribution trolleys. However we are pursuing this variation as a CfM research project outside the Rollet project itself.
4. We are aiming to clip the side panels together, so eliminating the cost of two corner posts for a three-sided TF superstructure.

 
The price we have been quoted for rotomoulded panels brings the overall manufactured cost of the Rollet down to around £54. Moulds are around £4,000 each so this is a very suitable approach for pilot full-scale Rollets – as we agreed we should aim for (rather than a scaled-down version).

We have a rotomoulder who is very keen to advance the project. He has provided some useful detailed design ideas. Since he has already quoted a competitive price to make the bases, he will have the strongest possible motive to ensure the whole thing fits together.

Invited paper published in the Proceedings of the Institution of Mechanical Engineers: Process Mechanical Engineering, volume 214, Part E, 2000, Special Millenium issue, ISSN 0954-4089.

S F Bush

Abstract

From slow beginnings in the 1860s, the evolution of the polymer industry has been marked in the second half of the twentieth century by rapid increases in the scales of production, by increasing power to control order at the molecular level, and by the variety and complexity γ of the resultant processes and products. The paper reviews some of the key developments over the last 100 years or so with a view to identifying themes likely to be of continuing importance in the new century.

A general model for the cost of a processing technology is proposed in terms of the factors Q and γ involved in producing a given artefact. Particular technologies are discussed in terms of the order in which basic processing functions are carried out. A major trend likely to continue into the twenty-first century is the way in which the supramolecular organization of the polymer chains is increasingly being brought under control, either directly by processing or indirectly by self-ordering properties of the polymers themselves. Self-organization of reinforcing fibres during processing to produce optimal performance of polymer composites is a parallel trend also likely to develop further into the next century. To illustrate these ideas the paper draws on examples from major polymer processes: extrusion, injection moulding, film blowing, reaction moulding, thermoforming, fibre making and coating.

Paper to FRC 8th International Conference 13th-15th September, “Composites for the Millennium”

Published as ISBN 85573 5504

S F Bush with D R Blackburn, A J Neuendorf and J M Methven

Abstract

While much of fibre reinforcement of polymers has rightly concentrated on solid forms, there is a significant demand also for lightweight open structures of the wire-cage type. The paper will report results obtained from a variety of polymer-fibre compositions in wire form.

These wire-cage results draw on the laboratory’s extensively reported work on long-glass fibre reinforcement of thermoplastics and the pultrusion of both thermoplastics and thermosets. However, for the new wire-cage technology, the behaviour of the synthetic fibre and natural fibres in place of glass fibres has also been investigated. The results obtained show that for a number of significant applications these soft fibres are better than glass fibres in terms not only of their formability into wire structures, but also in terms of their elastic recovery from imposed stress or strain.

The development opens up a significant new field for polymer-fibre composites both as an alternative to existing metal wire structures in the food distribution and textile industries and as an alternative to certain solid structures more generally.

Paper published in International Polymer Processing XIV (1999) 3, 282-290, ISSN-0930-777X, 1999.

S F Bush

Abstract: Experimental and Theoretical Results for Injection and Blow moulding, Sheet and Pipe extrusion.

The paper summarises the main experimental and theoretical results from a long-term programme of research (SAFIRE) to produce and apply long glass fibre compounds to the extrusion of pipes, sheets and profiles and to injection, blow and roto moulding. The overall objective is to obtain the processing speeds associated with short fibre reinforced thermoplastics with the reinforcement efficiencies associated with prepositioned or prepreg thermoset composites. Extrusion and injection moulding are now in the commercial domain, with industrial scale trials underway in the other technologies.

Long glass fibres are defined by their ability to form lace-like mat structures within the polymer melt which persist into the solid state. Such structures, which greatly increase both melt strength and solid state thermo mechanical properties, can be formed with fibre volume concentrations (c) as low as .0l. The formation of mat structures depends on the number N of virtual touches per filament. A minimum of around five touches is generally needed. From earlier work N is given as A.c l/d. A varies with mean fibre orientation in the mat: for the random in-plane case it is approximately 8/π², so that in contrast with typical fibre suspensions (c <d/l) extremely strong particle-particle interactions are involved in the melt state.

In the solid state, tensile strength is measured and modelled in terms of number average fibre length (l) and diameter (d), polymer yield strength, fibre distribution efficiency, interfacial shear strength and a specially defined matrix stress magnification factor M. The role of patented fibre management devices in optimising these variables as they appear in the solid state is defined and described.

Paper to the 5th International Conference on Manufacturing, Processing Composite Materials, Plymouth University, 12th-14th July 1999.

Published in the journal: Composites Part A: Applied Science and Manufacturing (incorporating Composites and Composites Manufacturing) ISSN 1359-835X, Volume 31, Issue 12, December 2000.

S F Bush with F G Torres

Abstract

The present paper summarises the main aspects and the developments in sheet extrusion and thermoforming of discrete long glass fibre (LGF) composites using the SAFIRE (Self Assembling Fibre Reinforcement) technology. During extrusion the long glass fibres are organised into coherent fibre mats which persist into the solid state, and are able to withstand the deformation process that takes place during thermoforming. A process analysis has been performed for extrusion and thermoforming indicating the main individual operations. Both processes have been studied with regard to their performance with the materials used in the studies, namely polypropylene homo and copolymer, with and without LGF reinforcement. Significant improvements in mechanical properties relative to the unreinforced materials have been found for the extruded sheets and the thermoformed products. Major improvements in processability relative to unreinforced PP have been found for the LGF materials. These are discussed in terms of the coherent fibre mat concept.

Paper to the International Conference on Manufacturing, Processing Composite Materials, Plymouth University, Professor of Polymer Engineering, UMIST, 12th-14th July 1999.

Published in the journal: Composites Part A: Applied Science and Manufacturing (incorporating Composites and Composites Manufacturing) ISSN 1359-835X, volume 31, issue 12, December 2000, 1421-1431.
S F Bush with F G Torres and J M Methven

Abstract

Three experimental techniques have been employed to assess the rheological behaviour of discrete long glass fibre reinforced polypropylene and propylene/ethylene copolymers. A Carri Med cone and plate rheogoniometer has been used to determine shear viscosity as a function of strain rate and time at temperatures relevant to the extrusion and injection moulding processes. A bubble inflation test (BIT) has been designed and used to characterise the behaviour of these composites under the extensional flow fields typical of blow moulding and thermoforming. Finally a squeeze load test (SLT), similar to those developed for sheet moulding compounds (SMC) and glass mat thermoplastics (GMT), has been used to explore the rheological behaviour of the long glass fibre (LGF) materials under compression moulding conditions, in particular to assess the relative importance of shear and extensional flow.