(148a) The ROLE of Physical Property Databases IN CH. E. EDUCATION

       Numerical problem solving in
Chemical Engineering typically requires a mathematical model of the problem
along with physical and thermodynamic data and correlations (equations) of the
chemical substances involved. The required pure compound properties are usually
divided into constant properties and temperature and/or pressure dependent
properties. In most ChE textbooks, the preparation of the mathematical model
associated with a particular problem is typically emphasized. The property data
required for solution of the problems are usually provided in tabular or
graphical form in the appendix of the textbook and/or in a CD associated with
the book (see for example, Felder and Rousseau, 2000 and Himmelblau and Riggs,
2004).  

       Recently, various databases that
contain extensive physical property data for large number of compounds have
become available. Typical examples are the DIPPR (Rowley et al., 2010) and the
NIST (http://webbook.nist.gov/chemistry/)
databases. The use of the data available in the databases for problem solving
has significant advantages over the use of the data provided in the textbooks.
Some of the more important advantages are:

1.    In engineering practice, the
databases are used as a principal source of property data and correlations. Thus it is important that students become
experienced with the application of these sources during their educational
programs.

2.    The databases provide consistent
sets of correlations for temperature-dependent properties enabling the use of
solution techniques independent of the format in which the property data is
provided. If
temperature-dependent data are provided in tabular or graphical forms, for
example, this prevents the use of standard numerical methods for problem
solution, and data format dependent ad hoc solution techniques have to be
used. 

3.    The data provided in databases are
continuously updated with the new data as they become available. Property data in textbooks are often taken "as
is" from references that may be as much as half a century old. The
textbook data may be incorrect or even contradictory. It is important that
students become accustomed to using reliable data.

4.    The data available in the databases
are evaluated and in cases when multiple values are available for the same
property, one recommended value is selected by the database professionals. There may be very substantial differences between
property values reported by different investigators. Brauner et al., 2005, for
example mention the case of the melting point of 4-methyloctane for which the
recommended value is 159.95 K while one reported experimental value is 219.62
K. The students must be made aware of the fact that several property values may
be available and learn how to find and utilize the value with the highest
confidence in its correctness.

5.    The databases usually provide
uncertainty values (upper limit on experimental error) which enable estimation
of the uncertainty of the problem solution using error propagation analysis. An important goal of chemical engineering education
is to impart to students that the numerical "solution" of a problem
almost always has some associated uncertainty.

       A convenient option for chemical engineering
educators to incorporate the use of databases into their teaching has been
integrated into the Polymath package (www.polymath-software.com ). This
widely-used computational tool now contains a "sample database"
subset of the DIPPR database.  The "sample database" contains 34
constant properties and 16 temperature dependent property correlations for ~120
compounds that are most frequently used in chemical engineering textbooks. A
special interface within Polymath now enables a convenient search of the sample
database for selected compounds. The desired properties can be selected, and
the data and correlations outputted in a format that can be copied and pasted
directly into a computer code with all the significant figures given in the
database. The formats that are currently supported are for Polymath, MATLABand Excel.

        In the extended abstract and the
presentation, several examples will be provided to demonstrate the
incorporation of physical property databases in numerical problem solving.  Our
experience with the in-class use of the Polymath and the associated DIPPR
sample database will also be described.  

References

1.      
Brauner,
N., Shacham, M., Cholakov, G. St. and Stateva, R. P., ?Property Prediction by
Similarity of Molecular Structures ? Practical Application and Consistency
Analysis?, Chem. Eng. Sci. 60, 5458 ? 5471 (2005)

2.      
Felder,
R. M. and Rousseau, R. W., Elementary Principles of Chemical Processes,3^rd
Ed, John Wiley & Sons, Inc, Hoboken, New-Jersey, 2000.

3.      
Himmelblau
D. M., and Riggs, J. B., Basic principles and Calculations in Chemical
Engineering, 7^th Ed., Prentice-Hall, Upper Saddle River, New-Jersey,
2004.

4.      
Rowley,
R. L.; Wilding, W. V.; Oscarson, J. L.; Yang, Y.; Zundel, N. A., DIPPR Data
Compilation of Pure Chemical Properties, Design Institute for Physical Properties,
(http//dippr.byu.edu),
Brigham Young University, Provo, Utah, 2010.