Laboratory of Engineering Thermodynamics (LTD)

Student Works

As part of your studies at RPTU Kaiserslautern, we are happy to offer you topics in thermodynamics and process development for your project work, student research project, diploma thesis, bachelor’s thesis, or master’s thesis. Your work at the LTD is always embedded in one of our ongoing research projects. You will be supervised by one of our research associates throughout.

Our research associates are always looking for capable, motivated, and committed students. Those who generally offer student theses are listed below, together with a brief outline of their work. If you are interested in a particular topic area, please contact the respective research associate directly. You can also find further information on their individual pages.

No special prior knowledge is required for most theses; we will make sure you get up to speed quickly. That said, prior knowledge is welcome. For many theses, it is an advantage if you have attended the lecture course “Thermodynamics of Mixtures.” Depending on the focus of the thesis, it is advisable to have taken the corresponding advanced courses, such as “Process Thermodynamics,” “Electrolyte Thermodynamics,” or “Machine Learning in Process Engineering.”

In addition to scientific supervision, you will receive support with project management and with presenting your results. This includes not only the write-up but also a talk on your work, which you will give in the LTD seminar.

We place great value on committed collaboration between student and supervisor. At the start of your thesis, we will discuss its goals and how the work will proceed. You will then receive a written task description that clearly defines the scope and objectives of the thesis. The scope is determined by the study and examination regulations that apply to you. We set up the conditions for your work with us so that you can achieve the objectives within the agreed time frame.

Below you will find descriptions of the work of the research associates who are currently offering student theses. If you are interested, please contact them directly.

If you have any general questions about student theses, please contact our student thesis coordinator:

Mery Hach (mery.hach(at)rptu.de).

Urgent openings in the following projects

Further development of hybrid group contribution methods for predicting thermodynamic properties

 Hannah Mennecke

(Download PDF hier)

Automation of vapor-liquid equilibrium measurements

Tom Schanne

(Download PDF hier)

Characterization of unknown mixtures using NMR spectroscopy and machine learning

Jens Wagner

(Download PDF hier)

Modeling of diffusion coefficients using entropy scaling and machine learning

Sebastian Schmitt und Jens Wagner

(Download PDF hier)

Anomaly detection in batch distillation processes

Justus Arweiler und Indra Jungjohann

(Download PDF hier)

Research associates currently offering student theses

Mery Hach

Research Areas
  • Nuclear magnetic resonance (NMR) spectroscopy 
  • Prediction of mixture properties 
  • Application of machine learning (ML) methods 
 
Description
Data on mixture properties are essential for chemical engineering processes. However, they are often scarce because measuring them requires considerable experimental effort. My work investigates how comparatively simple NMR measurements can be used and how the resulting data can be integrated into ML-based predictive models to gain new insights into mixture behavior and improve the accuracy of these models.

Rébecca Loubet

Research Areas
  • Evaluation of the performance of large language models (LLMs) on thermodynamic tasks
  • Application of LLMs in "human-in-the-loop" workflows
  • Literature search, data extraction, and data curation using LLMs
 
Description
Experimental thermodynamic data form the foundation for the development of thermodynamic models and are essential for process design. Although large databases of thermodynamic properties exist, many are not freely accessible. Consequently, researchers often need to compile problem-specific databases from scratch, a time-consuming and labor-intensive task. My research focuses on investigating the potential of large language models (LLMs) to support the construction of thermodynamic databases, from literature search and data extraction to structured storage and data quality assessment.

Hannah Mennecke

Research Areas
  • Prediction of mixture properties
  • Combination of group contribution methods with an Machine Learning (ML) approach for matrix completion
 
Description
Hybrid ML models combine the flexibility and broad applicability of data-driven ML approaches with the physical consistency and interpretability of physics-based models. In my research, I investigate such a hybrid model that combines a physical group contribution method with ML, enabling reliable predictions of mixture properties even when experimental data are scarce. In particular, I am interested in understanding how different types of data and the underlying group classification influence the model's predictive accuracy and applicability.

Jens Wagner

Research Areas
  • NMR Fingerprinting
  • Rational Definition of Pseudocomponents
  • Thermodynamic Modeling of Poorly Specified Mixtures
 
Description
Poorly specified mixtures, such as fermentation broths or the products of complex reactions, are widespread in process engineering. Since a complete analysis is often too time-consuming, these mixtures cannot be described using classical thermodynamic models. Therefore, I develop methods based on nuclear magnetic resonance (NMR) spectroscopy and machine learning algorithms to enable the thermodynamic modeling of poorly specified mixtures.