150 BOOK REVIEWS

[3] R. L. Burden and J. D. Faires, Numer- ing and a refresher on continuum mechanics.
ical Analysis, 6th ed., Prindle, Weber This is necessarily given at a brisk pace.
& Schmidt, Boston, MA, 1997. The governing equations in each area
[4] E. Isaacson and H. B. Keller, Analy- of continuum mechanics (referred to as
sis of Numerical Methods, John Wiley,
“domains” by MEMSfolk) are more than
New York, 1966.
just stated, however. Generous references
[5] J. Stoer and R. Bulirsch, Introduction to
Numerical Analysis, Springer-Verlag,
are given when more background may be
Berlin, Heidelberg, New York, 1980. needed, and specifics of the small scale are
emphasized. The section on the governing
EUGENE L. ALLGOWER equations of fluid mechanics, for example,
Colorado State University includes introduction of the Knudsen num-
ber (i.e., the ratio of mean free path length
to system size) and a form for a “slip”
Modeling MEMS and NEMS. By John A. boundary condition relating the difference
Pelesko and David H. Bernstein. Chapman & in tangential velocity across a fluid–solid in-
Hall/CRC, Boca Raton, FL, 2003. $69.95. terface to the fluid’s local velocity gradient.
xxiv+357 pp., hardcover. ISBN 1-58488-306-5. Numerical values of the Knudsen number
at which slip may begin to occur in practice
As the title indicates, this is a book about are given, followed by directions to recent
the modeling of MEMS and NEMS (micro- texts and the research literature for more
electromechanical and nanoelectromechan- information.
ical) devices. Its focus is on the modeling of A practiced applied mathematician or
systems and phenomena that is very much an engineer with a theoretical slant could
within the tradition of “physical” applied skip the first two chapters, but I would
mathematics but is directed specifically recommend that non-MEMSfolk jump in at
toward systems of small spatial scale— the third chapter “Small Is Different” and
sufficiently small in some current applica- begin to take note of the sections “Scaling”
tions, from sizes of microns to nanometers, and “Systems.” The term scaling is per-
that the applicability of continuum me- haps overused in general and may acquire
chanics can reasonably be questioned. an additional layer of meaning here, but
MEMS and NEMS is a burgeoning area its use in the book is given a fairly precise
of research and development in science and definition on page 65 (paraphrased): “how
engineering, with a market size that is being the characteristics of physical phenomena
estimated in terms of billions of dollars in change in magnitude and proportion as
a few years’ time. Applications range from length scales decrease from meter to mi-
developments in manufacturing to mechan- cron sizes.” Examples of what is meant are
ical and electrical engineering to material given and used later in the section on “Sys-
science and biology. But whether or not the tems,” in a discussion of scaling applied
next small thing turns out to be the next to the design of an electrostatic versus an
big thing, the topic offers interesting science electromagnetic actuator. Results of sim-
to be explored, and here is a good book to ple canonical problems—and here it is the
guide you through it. force between two infinitely long, charged,
The subject matter is of very broad or current carrying cylinders in the elec-
scope. Heat transfer, mechanics, elasticity, tromagnetic domain—are combined with
electromagnetism, and fluid mechanics, as results from a distinct domain, such as the
branches of continuum mechanics, all have critical compressive load for buckling of
a part to play in understanding MEMS and charged cylinders or the critical current for
NEMS, and some aspect of at least two will thermal overload of current carrying cylin-
usually be important in any one device. ders, to determine the relative merits of the
After carefully considered remarks about two designs and how the force derived from
preparation and prerequisites and a short, them depends on system size as system size
well-researched historical sketch, the first tends to zero.
two chapters of the book are given to dis- This example is presented as no more
cussion of nondimensionalization and scal- than back-of-the-envelope reasoning, not as
 BOOK REVIEWS 151

the result of a developed model. Nonethe- interest, however, and to date there is no
less, it is a good example of how dimensional other book on MEMS and NEMS that is fo-
analysis and scaling can be used to show the cused on the modeling process to the same
way from a simple model toward a more de- extent.
tailed model and a successful device. Graduate students of MEMS and NEMS
The first three chapters establish the will find the book to be between useful and
spirit of the text, while the next six de- indispensable. The exposure to modeling
velop the subject of MEMS and NEMS de- needed as preparation for the text is ap-
vices in greater detail. These are divided by proximately that of a good undergraduate
the domain or combination of domains of physics or engineering major, and the re-
continuum mechanics that pertain to them. marks in the preface suggest that it should
For example, Chapter 6 presents coupled be used after a first, graduate-level course
thermal-elastic systems. In each of these in modeling and continuum mechanics. The
six chapters, a number, from four to eight book would clearly be well suited to a grad-
or so, of existing specific devices and their uate course devoted to MEMS and NEMS,
purposes are described, accompanied by ref- and there is much that could be placed in
erences to their appearance in research jour- a second course on continuum mechanics.
nals. One or more core or prototype prob- Each chapter closes with a good selection
lem is identified that is common to each of well-thought-out examples, chapter high-
device and is then formulated mathemati- lights, and notes on the text.
cally and solved. The book is not without its misprints or
Nondimensionalization and scaling fea- typos, but I did not find any that really mat-
ture throughout, and solution methods in- ter, and I would not expect to see a book
clude phase plane analysis, separable so- in its first edition with fewer. The writing
lutions, stability, and perturbation meth- style is very clear and lively, producing a
ods. Elasticity and heat transfer are per- readable—even vibrant—text. But above
haps the domains that are most prominent, all, it is a thought-provoking book for those
although there is plenty of insight and in- with interests in modeling and is likely to
terest in other areas, and the choice of core have a longevity well beyond the usual half-
problems is excellent. Knowledge of math- life of a textbook.
ematical methods is developed in the text MICHAEL BOOTY
rather than assumed, and ample references New Jersey Institute of Technology
are given to more thorough or specialized
sources. The solution of core problems is
followed by exposition of how the model
might be developed further or adapted to High-Order Methods for Incompressible
specific devices that were described at the Fluid Flow. By M. O. Deville, P. F. Fischer, and
chapter’s beginning. E. H. Mund. Cambridge University Press, Cam-
The last chapter, “Beyond Continuum bridge, UK, 2000. $80.00. xxviii+499 pp., hard-
Theory,” contains a cautionary note on the cover. ISBN 0-521-45309-7.
relevance of continuum modeling for mi-
croscale systems, and outlines areas such This new book by Deville, Fischer, and
as nanostructured materials where another Mund is a fine work with a rather mis-
approach is required. leading title. First, it is not about “high-
The book is more didactic or instruc- order” methods in general, but rather is
tive in approach than a research monograph a comprehensive discussion of spectral and
would be. Indeed, the breadth and pace of spectral elements algorithms, also known
developments in MEMS and NEMS devices as “p-type finite elements.” And although
would probably preclude a research mono- the book certainly has a strong bias toward
graph on more than a small part of the fluid mechanics, most of the book is about
overall subject. Researchers in MEMS and the more general task of solving elliptic,
NEMS, where modeling and simulation are parabolic, and hyperbolic differential equa-
increasingly recognized as important first tions and would profit readers far removed
steps in design, are likely to find much of from fluids.