A virtual factory for steel

Materials World magazine
,
20 Dec 2018

Professor Steve Brown, Professor Claire Davis and Professor Martin Brunnock, look into how £7m of funding for a new virtual factory will aid the future of UK steel.

Steel is the most widely used structural material in the world. It is at the heart of major manufacturing sectors including automotive, construction, packaging and defence, and is indispensable for national infrastructure, supporting transport, communications and energy. Too often seen as old-fashioned, steel is in fact essential for high-tech 21st Century manufacturing, from energy-positive buildings and plastic-free packaging, to wind turbines and electric vehicles.

Innovation has never been more important. The steel industry needs to move ever faster to keep pace with changing technologies and customer requirements. This, however, is where it has faced a major problem. Currently, developing new steel alloys can be a very slow process with lots of different stages, and it requires expensive trials on hundreds of tonnes of material.

Swansea University, Tata Steel and Warwick Manufacturing Group (WMG), at the University of Warwick, UK, which have a long history of collaboration on steel research, have won £7m in funding from the Engineering and Physical Sciences Research Council (EPSRC), through the Prosperity Partnership initiative, to tackle this problem.

Slow product development

Alloy development in the steel industry has traditionally been iterative in nature, reflecting the high cost for the pilot plant and large-scale trials required.

Intelligent and computer-based approaches are already in use to aid the alloy design process. However, the complex interactions between steel composition and processing, for example for strip grades, mean it is currently difficult to rely on computer predictive approaches for new grades. Therefore neither of these methods is suitable for rapid development of bulk production commodity steel materials. In steel, step changes in performance still often come about through serendipitous discovery rather than design, which is something we want to change with this new project.

Also, innovations come most often in response to customer needs. Customer demand for new steel products drives increased requirements for strength for reduced CO2 emissions in the auto sector, durability for construction and infrastructure, and even grid energy efficiency – electrical steels.

The steel industry does, of course, boast a strong track record in developing new differentiated steel products. For example, Tata Steel ultra-smooth, galvanised outer body auto panel solutions are hot-formed, high-precision structural hollow sections that reduce the weight of buildings and new metallic coatings.

However, development costs are high, time to market is long and production yields are low – up to 20% lower than less complex commodity products. And this has not always translated effectively into profits, particularly in the early ramp up phase of the process.

This is due to the current development process being incremental, not disruptive. It is also heavily reliant on full-scale production trials, which require up to 900t of material to be produced. Up to 98% of this does not reach the customer and is scrapped at a cost of up to £200/t to the business.

In short, this approach, with its constraints, means that products arrive late to market. In addition, more risky, radical ideas don’t see the light of day, as the cost of testing them is so high.

RAP in a virtual factory

With the funding, the partnership will be introducing a new process called rapid alloy prototyping (RAP). Effectively, this means much of the testing can be carried out in research labs and imaging suites – a virtual factory – rather than in an actual steel plant.

Physical testing is combined with computational modelling to rapidly assess hundreds of small-scale samples, covering areas such as strength, electrical and mechanical properties, processability, durability and resistance to corrosion.

Test data can be fed into computational models, further refining their accuracy and allowing for better predictions on the final material properties. Alloys that show promise can then be investigated on a larger scale and in more detail.

Initially, a three-month benchmarking activity will be undertaken against a range of product exemplar alloys that are representative of UK sectors supplied by Tata, including automotive steels, lifting and excavating steels, metallic coatings, material durability, functional coatings and construction grades, oxidation and surface properties and electrical steels.

This early benchmarking against existing steel alloys will allow refinement of modelling and experimental techniques and provide confidence in predictions. Modelling will range from detailed FE models of miniature test specimens, thermodynamic-kinetic modelling for surface characteristics, macroscale process models and exploratory data analysis techniques to investigate large laboratory and industrial data sets.

Specimens of different alloys will be produced in the weight range of tens of grams. These specimens will be cast and then may be heat-treated and machined for mechanical and other testing.

A major challenge will be to develop a way to confidently bridge the gap between measurements made on this laboratory-scale material and actual products. In particular, specific issues related to specific grades will need to be considered. For example, in dual phase (DP) grades, the martensite can be banded in the rolling direction, which arises from segregation and will be a challenge to reproduce using small-scale samples. In other grades different challenges will present.

In the experimental programme, a series of decision gates will progressively eliminate all but the most promising alloy chemistries. Progressively more detailed analyses, such as microscopy or mechanical testing, will be deployed the further a specimen passes through the decision points.

For a suitably promising alloy that passes all the decision points, steps may then take place on the pilot plant facility at Swansea, which houses a VIM furnace, Lewis rolling mill and high-pressure water de-scaler. At this point a transition from grams to kilograms takes place. Material weighing 30–60kg will be produced in the VIM furnace, rolled and tested.

It will take time to generate the large data sets of different alloys but it is envisaged that around 2,000 specimens will be produced every six months when the project is at full speed.

The final step is a scale-up from kilograms to tonnes. It is anticipated that the first full production run of a new alloy will take place at the Port Talbot Tata Steelworks sometime in the third year. At this point, the virtual factory should be running at full capacity and subsequent full production runs will become increasingly more frequent during the second half of the project.

Professor Steve Brown of Swansea University College of Engineering explains, ‘The partnership will create a new methodology for product and process development, based on smart designed alloys and processes, which will mean the steel industry can respond far more quickly to customer needs.

‘The challenge is to be able to predict properties from rapid tests and modelling. For example, two steels with the exact same chemistry can have vastly different mechanical properties just by altering the cooling rate or rolling temperature.’

Brown said the new  approach will allow for the synthesise of over 100 different alloy samples per week, which can be rapidly characterised in terms of their thermo-physical properties, such as thermal conductivity and hardness, and their magnetic and electric properties. Combined with small-scale deformation/heating experiments, this approach will enable the team to establish phase transformation behaviour and generate thermodynamic and kinetic data.

Dr Cameron Pleydell-Pearce, Associate Professor at Swansea University College of Engineering, said, ‘The approach will involve fewer stages in the testing process. This can be done because it will draw on much better data. As one example, it will allow for the use of properties such as hardness to screen samples, without needing to conduct full-scale mechanical tests on every sample. High throughput screening is a tried and tested approach in the pharmaceutical industry. It is being adapted here for steel.’

A steel revolution

The difference this new approach will make to the steel industry can be significant:

  • 100 samples can be tested in the time it currently takes to test one
  • It produces datasets big enough to make predictions about properties
  • In overall terms, it means newer and better steel products can be made ready for customers far more quickly – from development to market entry can sometimes take the steel industry years using the current approach.

One area where the new approach could be particularly important is the elevation of steel’s environmental credentials. In principle, it is infinitely recyclable with no loss of performance. Rapid alloy prototyping is likely to lead to new products and processes, which contribute towards a greener low-carbon future.

Professor Claire Davis from WMG explains, ‘This project provides an exciting opportunity to accelerate the translation of innovative steel chemistry and process improvements into the steel industry.

‘For example, we will be able to explore the opportunities for increasing steel scrap levels in new steel production by understanding the influence residual elements such as copper and tin have on the properties and processability. This will help inform decisions on scrap sorting, selection and the levels that can be tolerated in different products, which will all contribute to the circular economy in the UK.

‘Rapid alloy processing facilities will allow us to trial new chemistries and process routes quickly to make recommendations for industrial take up.’

Supporting partners

This new approach is the work of all three organisations, with support from the EPSRC through their Prosperity Partnership initiative.

Prosperity Partnerships are EPSRC’s flagship approach to co-investing with business in long-term, use-inspired, basic research. They are five-year, multimillion pound research collaborations on topics of national and global importance, which have been co-created by leading UK universities and businesses with a strong research presence in the UK.

Professor Martin Brunnock, Tata Steel’s UK Technical Director, said, ‘This project will help us to accelerate the process of developing exciting new steels to give them a competitive edge. Steel is playing an essential role in helping to solve major societal challenges such as the transition to sustainable energy and mobility, and it’s vital to keep pace through the faster development of innovative steel products.’

Brown added, ‘Innovation is at the heart of the 21st Century steel industry. This project is a huge boost for innovation as it massively speeds up the development of new alloys. It means steel producers can deliver new and better products to their customers far quicker.’