Ian Bell (Fed)

Brief:

Dr.Bell’s research interests include computational routines and tools used for the development and application of thermodynamic mixture models.  These mixture models are used in software tools like NIST REFPROP, a modeling tool widely used in industry. 

Google Scholar Citation Page

NIST REFPROP – The industry standard in thermophysical properties

Dr. Bell is part of the team involved in the development of NIST REFPROP.  This software tool is the industry standard for thermophysical properties of pure fluids and mixtures, and is one of NIST’s flagship products.  His interest is especially focused on development of interfaces to facilitate the interfacing of REFPROP with other high-level environments like python, C++, Microsoft Excel, etc.  A key theme of Dr. Bell's work is to develop new numerical tools to be used in the new version of NIST REFPROP.  This includes:

• Superancillary equations reproducing orthobaric densities to within numerical precision (10.1021/acs.iecr.1c00847 and 10.1021/acs.iecr.2c02916)
• teqp, a library for evaluating EOS with advanced numerical tools (10.1021/acs.iecr.2c00237)
• entropy scaling transport property models, described in the next section

Entropy Scaling

Isomorph theory forms the basis for the observation that in many cases, there is a monovariate relationship between macroscopically scaled transport properties and the residual (or excess) entropy.  A number of publications have investigated this link, and highlighted how the residual entropy is a very important thermodynamic property that is connected to many other thermophysical properties. Some highlights:

• Paper: Probing the link between residual entropy and viscosity of molecular fluids and model potentials
• Paper: Modified Entropy Scaling of the Transport Properties of the Lennard-Jones Fluid
• Paper: Entropy Scaling of Viscosity for Molecular Models of Molten Salts
• Paper: Entropy Scaling of Viscosity - III: Application to Refrigerants and Their Mixtures
• Paper: Dynamic Crossover in Fluids: From Hard Spheres to Molecules

Refrigerants

Dr. Bell developed a number of open-source tools based in C++ and python for carrying out parameter optimization for a wide range of mixtures.  In addition, a number of recent efforts have focused on refrigerant (pure and mixture) screening.

• Fitting of refrigerant models for HFO-containing blends: 10.1063/5.0086060 and 10.1063/5.0135368
• Paper: The hunt for nonflammable refrigerant blends to replace R-134a
• Paper: Survey of Data and Models for Refrigerant Mixtures Containing Halogenated Olefins 
• Paper: Automatic Fitting of Binary Interaction Parameters for Multi-fluid Helmholtz-Energy-Explicit Mixture Models

Algorithms for thermodynamic calculations of mixtures

Dr. Bell is actively involved the development of algorithms for mixture calculations.  In this category are algorithms for the calculation of all the critical points for mixtures, phase envelope cross-sections, robust and efficient algorithms for calculating density values from multi-parameter equations of state, etc.  In most cases, the code for these algorithms are provided as supplemental information along with the associated publication.

• Algorithm to Identify Vapor-Liquid-Liquid Equilibria of Binary Mixtures from Vapor-Liquid Equilibria
• Calculation of critical points from Helmholtz-energy-explicit mixture models

Development of open-source software tools:

In the last few years Dr. Bell has been involved in a number of open-source software projects;  open-source software forms the cornerstone of replicable science.

Those projects are publicly available on NIST’s github page:

• teqp (website) : An open-source library for efficiently carrying out derivatives of an equation of state in C++ with automatic differentiation
• multicomplex (website) : An open-source C++ and Python library for doing derivatives with multicomplex algebra
• potter (website) : An open-source C++ library for calculating virial coefficients from pair potentials
• ChebTools (website) : C++ tools for constructing and working with Chebyshev expansions of functions in one dimension.  Can also be used to obtain roots of nonlinear functions in one dimension over fixed interval.

2022: Awarded the I&EC Research 2022 Excellence in Review Award

2018: Recognized as Emerging Investigator by the Journal of Chemical and Engineering Data