How to use this list
Membrane gas separation is used to remove carbon dioxide from natural gas and flue gas, to purify hydrogen and to enrich oxygen or nitrogen from air. Graphene-based membranes attract doctoral work because a single layer of graphene is only one atom thick, so pores cut into it can in principle let small molecules through quickly while blocking larger ones. In practice, "graphene membrane" covers three different materials: atomically thin graphene with engineered nanopores, stacked laminates of graphene oxide flakes, and polymer membranes with graphene or graphene oxide added as a filler. Each needs different equipment, skills and tests.
The rebuilt titles below keep every idea but say which material class, which gas pair and which test conditions each project will use. Each table gives a possible research question, a likely design and why the topic matters now. For related reading see the critical review of advanced functional materials, the review of AI-driven nanomaterial design, green hydrogen production titles and the topics hub.

Theme 1: Nanopores and rim chemistry in atomically thin graphene
| Dissertation title | Possible research question | Likely design and data | Why it matters now |
|---|---|---|---|
| Rim functional groups and their thermal motion in graphene nanopores: molecular simulation and single-layer membrane tests for O2/N2 and CO2/N2 separation | How do the type and density of functional groups on pore rims, and their motion with temperature, change permeance and selectivity for each gas pair? | Molecular dynamics with validated force fields; pore creation in CVD graphene; single- and mixed-gas permeation over a temperature range | Rim motion has been shown to govern selective transport through graphene nanowindows (Vallejos-Burgos et al., 2018; Cheng et al., 2025) |
Evidence that the gap is real. Vallejos-Burgos et al. (2018) combined simulation and experiment to show that the concerted motion of functional groups on the rims of graphene nanowindows mediates the separation of oxygen, nitrogen and argon, with very high permeability. Cheng et al. (2025) reported that thermal fluctuations of edge groups change the effective pore size in graphene nanopores, giving hydrogen/hydrocarbon selectivity above 70 at 220 °C and stable membranes after more than a year of ageing. What is still under-studied is how different rim chemistries compare across gas pairs, and how well simulations predict measured performance when pore sizes vary across a real membrane.
Theme 2: Graphene oxide laminates for hydrogen and carbon dioxide
| Dissertation title | Possible research question | Likely design and data | Why it matters now |
|---|---|---|---|
| Graphene oxide laminate membranes for hydrogen purification: H2/CO2 and H2/CH4 separation under realistic temperature and humidity, benchmarked against the upper bound | How do flake size, stacking and interlayer spacing affect hydrogen permeance and selectivity, and how far beyond the polymer upper bound do the laminates sit at operating temperature? | Laminates by vacuum filtration or coating on porous supports; XRD of interlayer spacing; mixed-gas permeation with and without humidity | Hydrogen purification needs membranes that stay selective at elevated temperature (Cheng et al., 2025; Rowe et al., 2010) |
| Crosslinking strategies for durable graphene oxide nanofilm membranes: polydopamine, amine and melamine crosslinkers compared for long-term CO2/N2 separation in humid flue gas | Which crosslinking route best limits swelling and loss of selectivity over months of exposure to humid CO2/N2 mixtures? | Laboratory fabrication; accelerated and real-time ageing; permeation at set intervals; spectroscopy of the crosslinks | Graphene oxide laminates swell in water, which changes interlayer spacing and performance (Abraham et al., 2017; Tiwary et al., 2024) |
Evidence that the gap is real. In graphene oxide laminates, molecules travel through the channels between stacked flakes, so the spacing between layers controls what passes. Abraham et al. (2017) showed that this spacing can be tuned by physical confinement, and that unconfined laminates swell in water. Tiwary et al. (2024) review fabrication routes for graphene oxide membranes and identify stability and scale-up as continuing problems. For gas separation, humidity in real feeds is the main threat to stability, so a durability study under humid gas would add evidence that short laboratory tests cannot.
Theme 3: Graphene fillers in polymer membranes
| Dissertation title | Possible research question | Likely design and data | Why it matters now |
|---|---|---|---|
| Machine-learning-guided screening of graphene oxide functional groups for CO2/N2 selectivity in graphene oxide-polymer mixed-matrix membranes, with experimental validation | Can a model trained on simulated and published data rank functional groups by their effect on membrane selectivity, and do the top-ranked groups perform as predicted in permeation tests? | Dataset from simulation and the literature; interpretable models; synthesis of a small number of predicted candidates; single- and mixed-gas tests | Machine learning is increasingly used to choose materials, but predictions need laboratory checks |
| Interfacial engineering of polysulfone-graphene oxide mixed-matrix membranes for CO2/CH4 separation: filler loading, surface treatment and interfacial voids | How do filler loading and surface treatment affect interfacial voids, and how do these defects change CO2/CH4 selectivity under mixed gas and rising pressure? | Membrane casting at several loadings; electron microscopy of the interface; mixed-gas permeation at several pressures; comparison with the upper bound | Natural gas and biogas upgrading need membranes that keep selectivity at high CO2 pressure (Rowe et al., 2010) |
Evidence that the gap is real. Polymer membranes are judged against an empirical upper bound: as permeability rises, selectivity tends to fall. Rowe et al. (2010) showed that this bound shifts with temperature, so a comparison is fair only when it uses the same conditions. Adding graphene or graphene oxide to a polymer can raise performance, but poor contact between filler and polymer leaves voids that let gas bypass the selective path. A study that links measured interfacial defects to mixed-gas results, rather than reporting single-gas data alone, answers a question that matters for scale-up.
How to test a title before you commit
Use the general method in how to choose a PhD topic and how to select a PhD topic, or our topic selection support, and screen each idea for feasibility, interest, novelty, ethics and relevance (Hulley et al., 2013). For membrane projects, also check the following.
| Check | What to confirm | Why it matters |
|---|---|---|
| Material class | Single-layer graphene, graphene oxide laminate or mixed-matrix membrane | Each needs different fabrication and characterisation |
| Gas pair and feed | Which gases, what composition, humidity and pressure | Mixed and humid feeds often differ from single-gas results |
| Test rig | Single- and mixed-gas permeation, temperature control, gas analysis | Without a reliable rig, selectivity cannot be measured |
| Benchmark | Upper bound at the test temperature and a standard polymer | Shows whether the result is a real advance (Rowe et al., 2010) |
| Durability | Test duration and ageing conditions | Short tests can hide swelling and loss of selectivity |
Worked example (illustrative). A student likes the live title on "hierarchical graphene nanolaminate architectures for high-performance hydrogen purification". Her laboratory can make graphene oxide laminates by vacuum filtration and has a mixed-gas rig that runs up to 150 °C. She narrows the title to H2/CO2 separation, plans tests at three temperatures with dry and humid feeds, and decides to plot every result against the upper bound at the matching temperature. The title now promises a question her equipment can answer rather than an outcome she cannot guarantee.
Common mistakes
- Using "graphene membrane" without saying which of the three material classes the project studies.
- Treating higher CO2 adsorption as proof of higher membrane selectivity.
- Reporting only single-gas, dry, room-temperature results for an industrial application.
- Comparing new results with published data measured at different temperatures or pressures.
- Promising to "eliminate" the permeability-selectivity trade-off instead of measuring how far a membrane moves beyond it.
Key takeaways
- Rim chemistry and its thermal motion control selective transport through graphene nanopores (Vallejos-Burgos et al., 2018; Cheng et al., 2025).
- Graphene oxide laminates depend on interlayer spacing, which swelling can change (Abraham et al., 2017; Tiwary et al., 2024).
- Benchmark every result against the upper bound at the same temperature (Rowe et al., 2010).
- Name the material class, gas pair and test conditions in the title itself.
- Screen each shortlisted title for feasibility, novelty and relevance before writing a proposal (Hulley et al., 2013).
