PhD defence Michiel Nijboer | Microporous Ceramic Nanofiltration Membranes using Atmospheric Pressure Atomic-and Molecular Layer Deposition

Microporous Ceramic Nanofiltration Membranes using Atmospheric Pressure Atomic-and Molecular Layer Deposition

Michiel Nijboer is a PhD student in the department Inorganic Membranes. (Co)Promotors are prof.dr.ir. A. Nijmeijer and dr.ing. M.W.J. Luiten-Olieman from the faculty Science & Technology, university of Twente.

The motivation for this work stems from the demand for robust membranes in the nanofiltration range for the purification of aqueous streams, which is key in enabling the reuse of water.

To that extent, the objectives of this thesis were split into two parts. First, the work aims to characterise and use a dedicated ALD reactor design that operates under atmospheric pressure for the modification of tight-ultrafiltration membranes. The second part of this thesis describes the development of a methodology that enables pore size tuning of nanoporous layers with a narrow pore size distribution.

Chapter 1 of this thesis introduces the concepts of membrane filtration as well as atomic layer deposition, the latter being an ultrathin-film deposition technique that originates from the field of semiconductor fabrication. The key parameters of the practical work and the motivation are outlined.

The available literature on the topic of atmospheric-pressure atomic layer deposition (AP-ALD) was reviewed in Chapter 2. The chapter highlights an opportunity to mature the application of the technique to more areas; dedicated reactor design can be used more in order to suit the deposition characteristics to the substrate.

A novel AP-ALD reactor with tubular design is presented in Chapter 3. The reactor is further characterised and improvements were suggested. Similar growth-per-cycle values were obtained when compared to state-of-the-art vacuum-based reactors. Furthermore, an in-line pressure-based method was developed for monitoring the closure of pores as a function of the number of layer deposition cycles. Lastly, the molecular weight cut-off of membrane samples was reduced to 300 Da, which is lower than what is commercially available at 450 Da for state-of-the-art ceramic NF membranes.

The following chapters present the adoption of a method making use of diversified reactants. This substitution of water with simple, bifunctional organic molecules as a co-reactant enables the deposition of hybrid inorganic/organic materials. The deposition strategy also changes, meaning that the aim now is to cover the pores with the hybrid layer, followed by a calcination step to remove the organic constituent of the layer. This removal of the organic molecules templates the formation of porosity, leading, in theory, to well-defined porous structures with a narrow pore size distribution. Chapters 4 and 5 show that the hydrophobicity and hydrophilicity of these membrane layers can be tuned between <10° and 102° (as measured by their water contact angle) by varying the temperature and atmosphere (air or nitrogen) during the heat treatment after deposition. These resulting properties were shown to originate from the molecular properties of the co-reactants and the temperature of the calcination step (i.e. thermal stability).

The same principal method is used in Chapter 6, where the material properties and morphology of the calcined layer are investigated in more detail, using a zeta potential measurement and scanning transmission electron microscopy (STEM), among others. The high-resolution micrographs provide new insights into the morphology of the hybrid layers after calcination, as well as the confinement of the deposited material to the uppermost region as a result of the deposition under atmospheric pressure.

Chapter 7 is a reflection on the results in the preceding chapters as well as some results not presented in this thesis. New directions of research are suggested on the basis of these observations in combination with related findings in the literature. For example, the design of the reactor was evaluated and key areas of reactor design are summarised. Furthermore, the confinement of the growth location of the hybrid layer is presented, along with suggestions for initial experiments about removal of the sacrificial layer by for example, washing with water or calcination.

Where the work in this thesis focuses on a limited number of variables for the sake of understanding, future work is suggested to rationally expand the used substrates, reactants and calcination conditions (like temperature range, heating- and cooling rates, duration, and atmosphere) in order to fulfil the potential of this promising method.