вход по аккаунту


Experimental Design for Combinatorial and High Throughput Materials Development. Edited by James N. Cawse

код для вставкиСкачать
Experimental Design for
Combinatorial and High
Throughput Materials
Edited by James N.
Cawse. John Wiley
& Sons, New York
2003. 317 pp.,
£ 38.40.—ISBN
Combinatorial and high-throughput
approaches are currently of great interest in materials and polymer research.
Triggered by small start-up companies
or by large multinationals, combinatorial and high-throughput paradigms are
becoming increasingly common in both
industry and academia. The book
edited by James N. Cawse is a very
timely contribution to this area. Cawse
is a leading industrial researcher, working in one of the top industrial research
laboratories active in combinatorial
materials research (General Electric
has the second largest number of patents
in this field). The book is suitable both
for beginners and for advanced readers,
and towards the end it introduces some
very innovative concepts into the discussion. The contents cover the entire field
of combinatorial materials research,
including inorganics, catalysts, and polymers. In contrast to most other books
currently appearing on the market, this
contribution by Cawse deals with the
key intellectual step associated with
(combinatorial) exploratory research,
namely the planning of experiments.
Angew. Chem. Int. Ed. 2004, 43, 4123 – 4124
The book (13 chapters with over 200
charts and illustrations) is structured in
such a way as to first introduce a
description of standard methods, and
subsequently progresses to a discussion
of cutting-edge mathematical developments.
Chapter 1 introduces some of the
background, by comparing the current
state of combinatorial materials
research with the situation in pharmaceutical combinatorial research, and discussing historical examples and the
explosion of interest in combinatorial
methods. In particular, the discussion
regarding the number of compounds
that are theoretically accessible using
the entire Periodic Table is highly
impressive. This part serves as an introduction to the rest of the book. The
motivation concerning “why to plan
experiments” is convincingly discussed,
in particular with a view to experiments
involving automation (“A poorly
designed experiment will give bad information with unprecedented speed and
in outstanding quantity”, and “If anything, planning must be done even
more carefully, since we now have the
possibility of going in the wrong direction faster than ever”). In my opinion
this chapter is outstanding and, although
written from an industrial perspective, it
is also very suitable for teaching purposes. In Chapter 2, the editor presents
his personal view of the field, and introduces the topics that will be covered by
the individual chapters (contributed by
different authors who are all leading
international experts, for example,
van Dover, Amis, Baerns). Important
keywords in this context are “optimization strategies”, “non-intuitive experiments”, “experimental space” and
Chapters 3 and 4 do not really fit
into the general concept of the book.
They discuss masking methods, preparation approaches, strategies for the exploration of defined phase spaces within the
overall inorganic phase space, and a
wide range of characterization techniques. Unfortunately, the design and
planning of experiments is not an integral part of these chapters. The same is
true of Chapter 5, where the wellknown NIST approach to the preparation of thin-film libraries by continuous
gradient methods is described. Here
the reader finds only a few statistical
The following chapters (on the optimal-coverage approach and on overlapping gradient arrays) deal with the
design and evaluation of combinatorial
experiments, specifically with gradientbased design in which the gradient is
sampled intermittently by robot-mixed
formulations. The advantages and disadvantages of split-plot designs are discussed in Chapter 8. These approaches
are used in particular for the design of
experiments in complex mixture/process
situations. The following two chapters
discuss more sophisticated forms of
genetic algorithms and artificial neural
networks. The latter two methods were
developed for highly complex and nonintuitive experimental spaces. Chapter
11 discusses the very important topic of
descriptor generation, as well as the
development of quantitative structure–
property (QSPR) models. Finally, Chapters 12 and 13 present two very different
approaches to the design of libraries and
experiments, namely Monte Carlo and
Spatial Sampling Design schemes. The
book is completed by a detailed index.
Overall, there are very few points for
criticism in this book. Apart from the
absence of aspects concerning the
design of experiments in Chapters 3
and 4 (see above), only the repeated
use of some figures in several places is
somewhat inappropriate (first in black/
white and, later on, collected in color).
However, these are only minor points,
which do not diminish the very positive
overall impression. This book represents
a highly important contribution to the
field of combinatorial materials
research, and will be of interest to beginners in the field, students, and practicing
researchers, as well as experimental
design specialists.
Ulrich S. Schubert
Eindhoven University of Technology and
Dutch Polymer Institute (DPI)
Eindhoven (The Netherlands)
DOI: 10.1002/anie.200385086
3 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Без категории
Размер файла
59 Кб
development, experimentov, design, throughput, high, cawse, edited, material, combinatorics, james
Пожаловаться на содержимое документа