School of Chemical Engineering and Advanced Materials

Staff Profile

Dr Mark Willis

Senior Lecturer

Background

Roles and Responsibilities

Degree Programme Director. Msc. Applied Process Control.

This course, which is supported through the CTA, brings together the disciplines of process control, process engineering, optimisation and industrial statistics alongside management. I designed this course and I am responsible for the overall management of the course. I gained accreditation for the course (InstMC) in 2005. Management of the course has given me experience in the complete range of course administration and support roles.

I have been (or am currently) a member of the following, Executive board, Marketing and recruitment committee (chair), School Liaison, Undergraduate teaching and learning committee, Postgraduate teaching and learning committee, Staff student committee, UG Board of studies, Board of studies MSc Applied Process control, PACT management committee, Molecular Engineering Industrial Advisory Board.

Qualifications

Chartered Engineer (Institute of Measurement and Control), 1997

PhD in the modelling and control of chemical reactors.
School of Chemical Engineering and Advanced Materials, University of Newcastle upon Tyne, 1991

Masters Degree (Hons) in Chemical Engineering.
School of Chemical Engineering, University of Leeds, UK, 1986

Honours and Awards

Best Poster Award, European Congress of Chemical Engineering, Copenhagen, 2007.

ICI Prize for best journal paper, Trans. InstMC., 2000

Kelvin Premium, Inst. Of Electrical Engineers for the journal paper ‘Artificial neural networks in process engineering’, 1992

Informal Interests

Skiing, Scuba Diving, Travel.

Research

Research Interests

My research is aimed at creating enabling technologies for process development. The area focuses on process development using High Throughput Technologies. I work closely with a colleague Dr Allen Wright. The work is intrinsically multi-disciplinary and systems engineering orientated, spanning chemical engineering, computing science, synthetic and physical chemistry, modelling, simulation, mathematics, statistics and kinetics.

Major achievements include:

(1)Establishment of a £750K SRIF II-funded Process Development Laboratory. These facilities include an HTT platform (a Chemspeed SLT106 synthesizer), an HEL SIMULARTM reaction calorimeter and a GC-MS (Varian Saturn 2200 with an auto-sampler) and a HPLC with an auto-sampler.

(2)The development of mathematical procedures for the determination of chemical reaction networks from batch process data which has resulted in 1) The first use in the literature of a simple non-linear polynomial model structure to reverse engineer plausible kinetic models of unknown chemical reaction networks from experimental batch and fed-batch composition data 2) The first use in the literature of the biochemical systems theoretic (BST) method of “S-systems” to demonstrate that the topology of unknown chemical reaction networks can be reverse-engineered from experimental composition data, acquired from batch, fed-batch or continuous flow chemical reactors.

(3)To date, the study of chemical systems with nonlinear dynamics has been a relatively pure science with few applications that are of practical importance. Along with co-workers we have reported an experimental study which achieves reproducible oscillations in both pH and heat output (Qr) during a palladium-catalysed phenylacetylene oxidative carbonylation reaction in an homogeneous catalytic system (PdI2–KI–Air–NaOAc in methanol solution). This reaction system is of interest for two reasons, firstly it provides a novel pH oscillator operating in a stirred batch reactor, and secondly because of the potential for the chemo-selective synthesis of several commercially valuable products. Despite the significance of this system, oscillatory behaviour has not been achieved by other workers.

(4)The development of approaches to integrating fundamental process knowledge into empirical approaches. In the area of genetic programming this has resulted in 1) The first description in the literature of co-operative co-evolutionary GP for building structured non-linear chemical process models (this won the ICI prize for the best paper in the journal Trans. InstMC in 2000. This technology has been widely adopted by Dow (USA) and has largely supplanted their use of neural networks for inferential estimation.

(5)Recognition within the area of GP as a significant contributor to published works (see www.cs.bham.ac.uk/~wbl/biblio/gp-coauthors/index.html)

Mark's research profile can be viewed at Google Scholar

Other Expertise

We provide consulting services through LyraChem (www.lyrachem.com) a joint spin out with Durham University Chemistry. Work ranges across the R&D lifecycle, including chemical route development, mechanism and kinetics determination, scale up and optimisation, thermal safety studies, incident investigation etc.
We also offer a service providing fast reliable and highly effective batch to continuous assessment with the opportunity for pilot trials when appropriate.

Current Work

We are currently involved in a large EPSRC funded research project in collaboration with the Department of Chemistry, University of Durham and the Department of Chemistry, University of Leeds. We are developing the use of robotic workstations to carry out parallel chemical synthesis alongside novel algorithms and software tools to automatically translate the reaction data into a chemical reaction network (reaction mechanism). A reliable reaction mechanism is a prerequisite for the regression of kinetic parameters and the use of process modelling and simulation tools.

Postgraduate Supervision

PhD Students Supervised:

(1) Heather Todd, A Rational Approach to Predicting Catalyst Performance in Reactive Distillation Systems, 2009
(2) Ankur Mukherjee,Oscillations and Chaos in chemical reactions, 2009
(3) Robert Worth, Engineering Scalability: Extending high throughput technologies to chemical process development, 2009
(4) Simon Horne, Using S-systems to model chemical reactions, 2009
(5) Sam Burnham, Towards the automated determination of chemical reaction networks, 2008
(6) Adam Errington, Hybrid modelling of a polyethylene process to predict polymer end-use properties, 2008
(7) Katie Kipling, Knowledge engineering in bioprocesses, 2004
(8) Mark Hinchliffe, Dynamic modelling using genetic programming, 2003
(9) Mark Porter, Modelling using symbolic annealing, 2004
(10) Dominic Searson, Nonlinear PLS using Genetic Programming, 2002
(11) Kevin Brodie, Inferential predictive control, 1999
(12) Hugo Hiden, Data based modelling using genetic programming,1998
(13) Andrew Mitchell, Inferential estimation and control of viscosity index on a lubricant production plant, 1998
(14) Simon Harris, A graph based framework for dynamic process modelling, 1998

Post-Doctoral Research Assistants:

(1) Dominic Searson, 2004 - 2008
(2) Katarina Novakovic 2004 - 2008
(3) Mark Hinchliffe 2002 - 2003
(4) Dominic Searson 2002 - 1999
(5) Ben McKay 2001 - 1998
(6) Hugo Hiden 1998 – 1996

Funding

Enhancing manufacturing by linking models derived from ultra-scale down and early process data via an intelligent agent. EPSRC.
Co-investigators: Dr. J. Glassey, Prof. G. Montague (£456,024).2008 - 2005

Extending high throughput technologies to synthetic route selection EPSRC.
Co-investigators: Prof. A. Wright. (£18,650) 2005 - 2004

Essential mathematics for the process industries. EPSRC.
Co-investigators: Prof. E. Martin, Prof. G. Montague (£40,196), 2005 - 2004

Extending high throughput technologies to chemical process development. The automation of mechanism determination. EPSRC.
Co-investigators: Prof. J. Morris, Prof. A. Wright, Prof. E. Martin, Dr. A. Whiting (Durham), Prof. S. Scott (Leeds) (£1,078,263), 2008 - 2004

Advanced Control of fermentation processes.
Co-investigator: Prof. G. Montague (£25,288), 2005 - 2004

Industrial Relevance

Consultancy

Evolutionary Computation:
Genetic programming,
Co-evolutionary component modeling,
Differential evolution,
Multi-objective optimisation.

Inferential models - soft sensors:
Symbolic regression,
Neural networks,
Hybrid models.

Multivariate Statistics:
Non-linear PLS,
Non-linear PCA,
Multi-dimensional clustering.

Software Development:
Matlab,
Java,
Enterprise applications (ConneXience),
Object orientated analysis.

We provide consulting services through LyraChem (www.lyrachem.com) a joint spin out with Durham University Chemistry. Work ranges across the R&D lifecycle, including chemical route development, mechanism and kinetics determination, scale up and optimisation, thermal safety studies, incident investigation etc.

We also offer a service providing fast reliable and highly effective batch to continuous assessment with the opportunity for pilot trials when appropriate.

Specialist Equipment
The High Throughput Technologies Research Group has its own purpose built research lab with state of the art experimental facilities including:
Chemspeed SLT 106 Synthesiser, a highly flexible, modular, and scalable robotic platform used to perform automated unattended parallel multi-step synthesis including reagent preparation, solids dosing, reaction, filtration, extraction, evaporation, work-up, and analysis (via an HPLC interface).
HEL Reaction calorimeter equipped with precision controlled solids, liquid and gas dosing, reflux operation, turbidity and pH monitoring and control.
GCMS and on line HPLC.

Teaching

Undergraduate Teaching

I have taught at B.Eng., M.Eng. and MSc levels with class sizes ranging from 5 to 50 to students with mixed ability with a diverse range of student backgrounds. I have lectured / taught the following subjects while at Newcastle, Material and Energy Balances, Heat Transfer, Process Control, Optimisation, Computing, Analytical techniques, Modelling and Simulation, Neural Networks, Design Project, Research project, Process Control Laboratory.

Publications