Synthesis of energy efficient distillation configurations

Anirudh Arun Shenvi, Purdue University

Abstract

Industrially, distillation is the method of choice accounting for around 95% of the separations. In spite of having evolved into a "mature" technology, distillation is an energy-intensive process with applications such as petroleum crude distillation alone consuming roughly 1.6 million barrels of oil/day worldwide. Even small energy efficiency improvements with development of better and cheaper separation process solutions can have a huge impact on the overall process economy. For instance, for the fractionation of the natural gas liquids (NGLs) that is found in the newly discovered shale gas reserves in the US, a potential energy savings of 10% can save in excess of 300 Quads of energy! To put this savings in context, the annual US energy consumption is a little less than 100 Quads. The overarching goal of this thesis is to create user-friendly and robust methods for identification and design of highly energy efficient industrial distillation configurations. In this work, we have developed a novel method based on temperature-profile analysis to identify the most energy efficient heat integration arrangement in a distillation column for any given separation. Our method dispels the conventional notion of direct proportionality between number of heat integration points and energy savings. For distillation of a multicomponent mixture into desired product streams, multiple distillation columns can be used in several different arrangements that can differ substantially in cost and energy efficiency. A long-standing challenge in multicomponent distillation has been the inability to systematically generate the search space of all distillation configurations. Conventional distillation configurations use n - 1 distillation columns for separating a multicomponent feed mixture into n product streams. With capital cost being an important parameter for choosing an optimal solution, distillation configurations that use "less" than n - 1 distillation columns are of significant interest. We have developed, for the first time, an easy-to-use mathematical framework to synthesize the complete search space of distillation configurations having less than n - 1 columns. With this framework, we systematically generated a large number of hitherto unknown distillation configurations. Some of these novel configurations demonstrate a feature that we refer to as simultaneous heat and mass integration. These novel configurations deserve special attention because they can be easily exploited to improve energy efficiency by withdrawing side-stream products. Another important challenge for a process engineer is to quickly prune the search space of hundreds and thousands of multicomponent distillation configurations and identify a handful of attractive energy efficient candidates. To this effect, we developed a quick screening optimization tool that identifies configurations with low heat duties and high thermodynamic efficiencies. Finally, we lay the foundation for equations analogous to the classic Underwood equations to estimate the minimum heat duty requirements of a distillation column with multiple feeds and products.

Degree

Ph.D.

Advisors

Agrawal, Purdue University.

Subject Area

Chemical engineering|Energy

COinS