Investigating introductory and advanced students' difficulties with entropy and the second law of thermodynamics using a validated instrument

TL;DR

This study uses a validated survey to analyze how introductory and advanced students understand entropy and the second law, revealing common misconceptions and difficulties across different education levels.

Contribution

It provides comprehensive data on student misconceptions about entropy and the second law, highlighting specific errors and misunderstandings at various levels of physics education.

Findings

Introductory students often confuse entropy with energy properties.

Advanced students expect entropy increases even in processes where it does not change.

Widespread incorrect application of the entropy change formula.

Abstract

We use the Survey of Thermodynamic Processes and First and Second Laws-Long (STPFaSL-Long), a research-based survey instrument with 78 items at the level of introductory physics, to investigate introductory and advanced students' difficulties with entropy and the second law of thermodynamics. We present analysis of data from 12 different introductory and advanced physics classes at four different higher education public institutions in the US in which the survey was administered in-person to more than 1000 students. We find that a widespread unproductive tendency for introductory students to associate the properties of entropy with those of energy leads to many errors based on an idea of "conservation of entropy," in which entropy increases are always balanced by equal entropy decreases. For many of the more advanced students (calculus-based and upper level), we detect a tendency to expect entropy increases even in processes in which the entropy does not change. We observed a widespread failure to correctly apply the relationship {\Delta}S = {\delta}Qreversible/T, either by using it for irreversible processes to which it does not apply, or by applying it incorrectly or completely neglecting it in reversible processes to which it does apply. We also noted that many introductory students are simply not aware that total entropy must increase in any "spontaneous" heat transfer process. Students at all levels were very frequently found to be confused that while net entropy (system + reservoir) in reversible isothermal processes does not change, the entropy of the working substance itself does indeed increase or decrease depending on whether the process is an expansion or compression. Our findings are broadly consistent with prior research findings in this area, expanding upon them and revealing previously unreported aspects of students' thinking.