Pre-Conference Courses

Separate registration will be necessary:
Extra fee EUR 100 (students EUR 50). Participation only possible for conference ticket holders.

Tuesday, April 22, 2008

Morning Workshops 9:00-12:30
Multivariate analysis in spectroscopy: introduction to the UNSCRAMBLER
CAMO Software AS, Oslo/N

Multivariate data analysis, experiences from PAT implementation using the SIMCA software
Umetrics AB, Umea/S

PAT in crystallisation process development, scale-up and prodution
B. O'Sullivan, Mettler-Toledo, Dublin/IRL

Introduction to process control for chemists
J. Morris, University of Newcastle upon Tyne/UK

Chemical imaging
R. Kessler, Reutlingen University/D

Afternoon Workshops 13:30-17:00
Advanced multivariate data analysis: introduction to multivariate curve resolution
R. Tauler, IIQAB-CSIC, Barcelona/E; A. de Juan, University of Barcelona/E

Design of experiments for beginners
W. Kessler, Steinbeis Technology Transfer Centre for Process Control and Data Analysis, Reutlingen/D

Process RAMAN spectrometry
A. Nordon, University of Strathclyde/UK ; P. Dallin, J. Andrews, Clairet Scientific, Northampton/UK

Mass spectrometry in process analysis
InProcess Instruments GmbH, Bremen/D

Chemometrics for beginners using PLS_Toolbox
R. Bro, University of Copenhagen/DK

Detailed description of each workshop:

Multivariate Analysis in Spectroscopy: Introduction to The Unscrambler
With this course the participant is introduced into multivariate modeling of spectroscopic data. In the first part of the course data import from various instruments and other sources (like EXCEL, MyInstrument) are presented and demonstrated. Then follows the inspection of the data by statistics and PCA, in order to identify outliers. The theory of the main pretreatment methods (derivation, scatter correction, SNV etc) will be explained and applied in examples. The second part of the course will include the theory and practise of calibration and prediction using PLS algorithms, as well as model validation. Moreover wavelengths selection methods will be shown and applied. The basics of Multivariate Curve Resolution (MCR) are introduced and illustrated by case studies. Finally the possibilities of on-line predictions with OLUP, OLUC and Unscrambler On-line based on a regression model with automatic pretreatments will be presented.

Multivariate data analysis, experiences from PAT implementation using the SIMCA software
Umetrics will present the strategy behind successful multivariate PAT implementation. From initial feasibility study of off-line data to fully implemented on-line solutions for the entire process.
The seminar will also give a basic introduction to the multivariate technology including software demonstration. More information as pdf.

PAT in crystallisation process development, scale-up and prodution
The course is aimed at providing general concepts of crystallization processes as well as a variety of PAT application examples, monitoring both, liquid and solid phase.
Theoretical background is combined with experimental case studies, e.g. in optimizing a crystallization process

Introduction to process control for chemists
The aim of this short course is to provide an overview of process modelling and control for scientists and engineers that have no background in process control. Sufficient knowledge will be provided to allow delegates to appreciate the scope of process control techniques and the challenges that need to be considered when implementing process control strategies. The topics will include: classical feedback control with process examples; feedforward and inferential control; multi-loop control; analytical feedback control; and will finish with a brief overview an overview of more advanced control systems now being applied, and will finish with a brief review of control benefits analysis methods and practice.

Chemical imaging
„Chemical Imaging” or “Spectral Imaging” plays more and more an important role in process analysis. The technique allows analysing the lateral distribution of a spectral signature in a material but can also be used for in inline process control with multipoint spectral detection in a plant.

The course will focus on optical spectroscopy (UV/Vis-, Fluorescence-, NIR-, IR- and Raman- Spectroscopy) for Spectral Imaging. The different terms like line scanning, mapping and imaging will be explained and the state of the art technology in Whiskbroom Imaging, Staring Imaging and Pushbroom Imaging will be presented together with a complex multivariate data analysis.

Part I: Introduction to Spectral Imaging

Definition: Whiskbroom-, Staring- and Pushbroom Imaging
Advantages and disadvantages with practical examples
Lateral resolution, penetration and information depth in scattering media
Imaging beyond diffraction limit
Spin offs: e.g. Multidimensional Fluorescence, Micro Reactor Plant Imaging, Multipoint Spectral Detection, Reactor Tomography, Diffuse Optical Imaging,

Part II: Leading edge technology and industrial applications

Bruker Optics: Whiskbroom Imaging IR/NIR and Raman
J&M Whiskbroom Imaging UV/VIS/NIR and Fluorescence:
Malvern Instruments: Staring Imaging:
Specim: Pushbroom Imaging:
Burgermetrics: Multivariate Data Analysis in Chemical Imaging

Advanced Multivariate Data Analysis: Introduction to Multivariate Curve Resolution
Multivariate Curve Resolution (MCR) has been shown to be a very powerful tool for advanced multivariate data analysis. It may be used in the analysis of any data set ordered in a data table or data matrix or in multiple data tables or matrices. The basic assumption of these methods is that measured data variance may be decomposed in the sum of individual contributions (mixture) coming from different sources or components, each one defined by a set of profiles describing the changes in the different orders o modes of
measurement, for instance in their concentration profiles and spectra profiles.
This short course will cover the basic aspects of MCR and will give a fast overview of the implementationm of the MCR method using Alternating Least Squares (ALS). Examples of different complexity will be given.

Topics:
Session 1 (90 minutes): Decomposition of a data matrix using a Bilinear Model. Multivariate Curve Resolution of a single data matrix. Number of Components. Initial estimates. Data structure, local rank and Evolving Factor Analysis. Natural Constraints. Alternating Least Squares. Example of analysis and resolution of a chemical reaction process. Example of
resolution of a hiphenated cromatographic-spectroscopic system. Software

Session 2 (90 minutes). Extension of the Bilinear model for the decomposition of multiway and multiset data structures. Matrix augmentation. Multivariate Curve resolution of augmented data matrices. Extension of constraints to multiple data matrices. Alternating Least Squares. Example of resolution of an enviromental multiset system. Example of resolution of a
spectroscopic image. Software.

Design of experiments for beginners
Searching for new products or improving existing processes involves changing large numbers of controlled variables to find optimal process conditions which meet all specifications at minimal cost. This can be accomplished efficiently by doing Experimental Design. The workshop teaches you how to find the most important factors you need to focus on and how to set up the experiments to discover previously unknown interactions, which too often misled you by ignoring them. Learn how to use statistical methods to give you confidence in your findings. Statistical Design of Experiments is a key to success with a minimum number of experiments.

How and when design of experiments should be used
The basics of other designs: Full Factorial designs
Exploit factorial designs
Discover hidden interactions
Interpret the results with analysis of variance (ANOVA)
Robustness testing
What to do after screening?
Optimisation designs

Process Raman Spectrometry
Process Raman spectrometry is becoming a more accessible technique that extends the range of procedures available in process analytical chemistry. As yet, it is not as widely used as e.g. near infrared spectrometry, which may partially be due to a lack of appreciation of the attributes of Raman spectrometry for process monitoring and control. This course will provide an introduction to Raman spectrometry and explore the range of process applications in a number of industries. The course will include sessions on the following:

Theory of Raman spectrometry
Instrumentation
Process applications
Comparisons with other process analysis techniques
Discussion and questions

Mass Spectrometry in Process Analysis
Process mass spectrometry is used for on-line monitoring, process control and quality assurance in various branches of industry and research for more than ten years.
This short course will give an overview about quadrupole mass spectrometers as a powerful analytical tool for process analytics.
Fundamentals of mass spectrometry and the specifics of process mass spectrometers will be explained. Application examples will show the strengths and opportunities of this method. System design, flexibility, ease of use and integration into the industrial environment will be discussed.

13:30 - 15:00
- Basics of process mass spectrometry for on-line gas analysis
InProcess Instruments GmbH, Bremen, Germany
- On-line monitoring of contaminations in pure hydrogen using a mass spectrometer
- On-line monitoring of a vinyl acetate monomer reactor in an industrial environment with a mass spectrometer
- Optimization of a ketene plant using on-line mass spectrometry
Thomas List, Wacker Chemie AG, Burghausen, Germany

15:00 - 15:30 Coffee Break

15:30 - 17:00
- Application examples for on-line monitoring, process control and quality assurance
- Instrument demonstration
InProcess Instruments GmbH, Bremen, Germany

Chemometrics for Beginners using PLS Toolbox
In this course, the basic principles in chemometrics and multivariate analysis are explained in simple terms. Many examples are given on the use of NIR, fluorescence and chromatographic data for process analytical applications. It is shown how multivariate methods can help understand, monitor and control a process with examples from food, pharma and related areas.

1. Why chemometrics is needed?
2. Principal Component Analysis (PCA) - how it works
3. PCA for fault detection and raw material identification
4. Predicting product quality
5. Understanding the process using advanced spectroscopic sensors
6. New frontiers in chemometrics