Progress in the thermal sciences: AIChE Institute Lecture

Abstract

Rather than merely sur®eying progress in the prediction of heat transfer o®er the past half century, attention is focused on the factors that ha®e led to significant ad®ances in understanding and practice. An almost one-to-one correspondence is demonstrated between ad®ances in heat transfer and those in computer hardware and software. Howe®er, the de®elopment of specialized algorithms for flow and heat transfer by indi®idual in®estigators appears in most cases to be an essential ingredient. Direct contributions of experimentation are relati®ely minor, but the synergy of combined experimentation and numerical analysis is most often the key to disco®ery and inno®ation. A quiet re®olution has occurred in the form of correlati®e equations, and the primary and almost sole contribution of closed-form analysis has been the deri®ation of asymptotes for that purpose. Finally, progress in heat transfer has occurred primarily in discrete steps, and most often as a consequence of exploratory research andror the obser®ation and pursuit of the explanation for an anomaly. IntroductionThe first Institute Lecture was given at the Annual Meeting in Pittsburgh in 1949 by William H. McAdams and was on the subject of Heat Transfer. I was present at his talk, and the intervening period of 49 years corresponds closely to that of my own career in research and teaching in this very field. In that half of a century, the experimental and analytical advances in heat transfer, per se, have been rather modest but those associated with machine computation have been overwhelming, particularly in heat-exchanger design and the numerical solution of differential and integral models. The net result is a total change in the character of the thermal sciences and their role in our profession. Heat transfer has virtually disappeared as a distinct field of industrial practice and academic research in chemical engineering, but remains a subject of critical importance in traditional chemical and materials processing, as well as in every current new endeavor of chemical engineers from biomedical technology to the manufacture of computer chips. As an example of a traditional application, an endothermic homogeneous reactor may be considered to be essentially a pure heat exchanger in that the rate of reaction is proportional to the rate of heat transfer and nearly independent of the chemical kinetic mechanism.As an example at the modern end of this spectrum of applications, the speed and mechanical design of the largest computers are now limited by the rate of heat removal. For these reasons, heat transfer has rightfully retained a prominent place in the curriculum of chemical engineering.This article describes a few important advances in the past half century that have led to the present state of the thermal sciences, but it is not intended to be a review in the conventional sense. The future is of greater interest than the past, and, as a guideline for the achievement of future advances, I will try to identify what prompted some of those of the recent past....

On temperature prediction at low Re turbulent flows using the Churchill turbulent heat flux correlation

“…In Refs. [11,12,14,20,21], Churchill and coauthors have also discussed an effect that has been neglected in previous research: the deviation of the heat flux density distribution due to the shear stress distribution across the channel. They have suggested the use of the correction term c, which can be included in the calculation of the heat flux ratio, as shown in the following equations for equal and uniform heating from one plate, Eq.…”

“…In analogy to momentum transfer, turbulent heat transfer based on the fraction of the heat flux that is due to turbulence has been studied more deeply in recent papers by Churchill and coauthors [12,14,20,21]. These papers were focused on the temperature predictions in turbulent flow fields, in round tubes [20] and between parallel plates [21].…”

“…Results from direct numerical simulations have shown that scaling with the wall parameters for low Reynolds numbers does not provide universal behavior for the fluctuating thermal field [8][9][10]. However, an effort to explore the scaling of heat transfer similar to that for momentum has not been vigorously pursued, with the notable exception of Churchill and coworkers [11][12][13][14] who suggested the use of the local fraction of the heat flux density due to turbulent fluctuations to predict the mean temperature, introducing, thus, the use of a scale different than the conventional friction temperature.…”

Lagrangian Modeling of Turbulent Dispersion from Instantaneous Point Sources at the Center of a Turbulent Flow Channel

“…While a much broader research effort has been devoted to the modeling of turbulent flow [4,5], the effort to simulate and develop models for turbulent heat transfer has been more limited [10][11][12][13][14][15][16][17]. The simulations are mainly divided into two categories: those that are based on the Eulerian approach to describing transport phenomena (and where the system of reference is not moving), and those that…”

Understanding Macroscopic Heat/Mass Transfer Using Meso- and Macro-Scale Simulations

“…However, experimental data have demonstrated scatter around this correlation, implying that there is another functional relationship between the dimensionless numbers, Nu = f (Re, P r), (see Churchill's insightful discussion about this issue [83]). Regarding P r dependence, there is a controversy in the literature among investigators who argue for a heat transfer coefficient that goes as h ∼ P r −3/4 and those who argue for h ∼ P r −2/3 .…”