Photoelectrochemical Hydrogen Generation Using QDs Sensitized Photocathode
Mahilet T. Meidenbauer
University of Rochester, Material Science Department
1. Introduction and Motivation
The technical and environmental challenges associated with fossil fuel extraction and usage pose risks globally due to factors such as price volatility, environmental pollution, and distribution challenges.  Therefore, developing and employing clean and renewable energy sources is crucial for the future of our planet. H2 is a promising renewable fuel as it has the highest specific heat energy of any chemical and it combusts clean with water as the only byproduct. Today 95% of the U.S. H2 is produced by reforming natural gas.  The two main disadvantages of natural gas reforming are that the starting material, methane, is itself a valuable energy source and methane reforming requires a high temperature process.
Motivated to develop and study a sustainable H2 production system, this study investigates quantum dot (QDs) synthesized electrode for photoelectrochemical water splitting. The photoelectrode can be divided into two layers, p-type NiO layer on ITO coated substrate that functions as a support to the layer of n-type semiconducting quantum dots. Upon illumination, an electron hole pair are generated with in the quantum dot. The electrons are transferred to a hydrogen evolving catalyst and are used to reduce a proton into H2. The p-type NiO layer fills the hole left in the QDs.  The rate of hole diffusion through the NiO layer to the QDs plays an important role in the overall efficiency of the system.
The full water splitting reaction requires both water oxidation to oxygen (O2) and proton reduction to H2. This study will be focused on the reductive side of water splitting (equation 1) by using a quantum dot synthesized electrode.
2 H+ + 2e- → H2
QDs have been shown to have a great potential for photoelectrochemical reduction of proton to H2; however, the underlying QDs, film, and catalyst interaction and the charge transfer at the surface of QDs is poorly studied.  Motivated to close this intellectual gap and develop a scalable and deployable H2 production technology; this study investigates physical and chemical properties of the NiO layer as well as current production of the system.
2. Design and Sample Preparation
The photocathode was prepared by doctor blading a NiO film on a conductive ITO coated glass. The film was the calcined at 450o. CdSe quantum dots with first exitonic peak at 525nm was then affixed on the NiO layer via spin coating. Chronoamperometry using CHI potentiostat was used to measure current produced by the system.
3. Results and discussion
Scanning electron microscope imaging is done using a beam of electrons in raster scanning pattern. The scanning electron beam has high kinetic energy and upon impact with sample atoms, the kinetic energy is dissipated due to electron-sample interactions. The main signals generated by primary beam and specimen atom interactions is secondary electrons. Secondary electrons are most valuable for showing morphology and topography of samples. The SEM micrograph collected using the SE2 detector is given by image 1. This image shows that the ITO electrode is completely coated with an NiO and the surface is textured. SEM image taken at 54o, image 2, provides a better insight to the surface roughness. The image taken at an angle shows more surface detail due to the SE2 detector location relative to the sample.
Image 1 NiO coated electrode at (left image at 0o
and right at 54o
Lithography is a patterning method that can be used to mill or deposit material on a sample in the micro and nano scale. Focused ion beam lithography (FIB) is one of the three lithography methods employed in the SEM. In FIB, sample material is ablated away using focused ion beam. In this study FIB was used to gain insight into the cross-section of the electrode. Figures 2A and 2B show the overall FIB pattern that was used. Figure 2C shows a higher magnification of the FIB well. The FIB well reviles that the NiO layer is irregular and has voids within. Since the charge transfer is reliant on the NiO layer, further studies on optimizing the NiO layer will be needed.
3.3 X-ray Spectroscopy
X-ray signal is generated when primary beam of electrons interaction with electrons in discrete orbitals of the specimen atoms and cause the ejection of inner orbital electron from the sample. As an outer orbital electron relaxes to inner orbital, it gives of X-rays that is of a fixed wavelength that is related to the difference in energy levels of electrons in different shells for a given element.  As a result, there is a characteristic X-rays for each element making elemental analysis using SEM possible. The peaks in an EDS spectrum are labeled using characteristics of K, L, and M shell X-ray signal of an element.
EDS and elemental mapping was used to prove the presence of CdSe as well as verify the chemical composition of the NiO layer. The region shown by image 2C was the area used for this analysis. Its important to note because of the interaction volume the Cd and Se signals are week. This effect is discussed further in 3.6
Elemental mapping of the area given by image 5 is shown below. The elements Ni, Si, In, and Sn were mapped along a fracture line. The images shows the regions where these elements are concentrated. 6
Image 3 EDS (Left image - Bare NiO coated electrode, Right Image - QDs (CdSe)sensitized Photocathode
SEM micrographs are usually recorded normal to the incident beam. While top view of a sample gives a detailed study of the topology, it does not provide a direct qualitative depth measurement. To gain depth measurements, samples can be fractured and mounted at a 90o. Image 5 shows that the NiO layer is approximately 8µm. This method also reveilles that the NiO layer does not make a smooth contact with the ITO layer and the hight of the layers is not regular.
Image 5 SEM image of a fracture line mounted at 90o
Transmission electron microscopy (TEM) is a type of microscopy where a transmitted electron beam through a sample specimen is used to form an image. TEM can be used to give morphological, compositional, and crystallographic information of a sample.  When transmitted beam goes through a sample specimen, some of the electrons will be absorbed, some will be scattered, while other go through the sample to form a micrograph. The lighter areas of TEM images represent the places where a greater number of electrons were able to pass through the sample and the darker areas reflect the dense areas of the specimen.  the quantum dots used to synthesized the working photocathode were investigated using TEM and the micrographs are given by image 6.1.
Image 6.1 TEM image of 2.6nm CdSe QDs
Image 6.2 TEM image of 4ML CdSe NPLs
3.6 Interaction Modeling
Micrographs produced by the SEM are dependent on the quality and quantity of signal collected. When primary beam interacts with a sample material, the beam penetrates into the sample and creates an excitation zone shaped like a tear drop. This region is called the interaction volume. The interaction volume will give insight into the imaging limits. Image 7 gives the approximate interaction shape. The NiO layers is infinitely thicker than the QDs used, it is expected that detecting the quantum dots on the surface will be a challenge. Under normal operating condition of 5kv to 20kv accelerating voltage the primary beam will drive through the surface dots. Some Cd and Se signal was detected by increasing the area over which EDS was performed and by increasing the takeoff angle and integration time. Similarly, because the NiO layers depth was larger than the penetration depth, the underling ITO coating was difficult to detect. This issue was solved by imaging and mapping the sample along a fracture line at 900 and through FIB.
Image 7 Interaction volume model (Left image - 3000 trajectories, right image - 200 trajectories)
4. Conclusion and Future Studies
This study has shed some light on to the physical and chemical properties of the NiO layer used in the preparation of QDs sensitized photocathode. Preliminary and minimally optimized current generated using QDs sensitized photocathode show promising and exciting results. Some of the result is given by image 8.1.
Image 8.2 shows false color SEM images of a used Photocathode while image 8.3 shows EDS elemental analysis. The used photocathod show Cl and K deposition from the electrolyte used. The effect of this deposition on charge transport will be studied in the future. The Cl and K deposition was further studied using EDS and is given by image 8.3.
Image 8.1 Preliminary current generated using QDs sensitized photocathode (Chronoamperometry current measurements)
8.2 Colonized image of used QDs sensitized photocathode
Using the information gathered through SEM, TEM, EDS, and FIB I hope to make continuous advanced on my project and work towards a sustainable hydrogen production system in the coming years. Future work includes optimization of the NiO layer and catalyst studies.
Image 8.3 EDS elemental analysis of used QDs sensitized photocathode
I sincerely would like to thank Brian McIntyre for hosting the most enjoyable class at UofR and for the endless help and encouragement throughout the this project and the course. I also would like to thank Ralph Wiegandt for training me on FIB and sample mounting at an angle. Last but not least I would like to thank my family for supporting me through my graduate career, Xenon for helping me through every deliverable, and my husband, Gary, for always being there for me and making me a better person each day.
2. Bren, K. L. et al. (2016) Inorg. Chem
3. Krauss, T. D. et al. (2017) PNAS
4. Eisenberg, R. (2017) PNAS
4. "Scanning Electron Microscopy (SEM)." Examples, 26 May 2017, serc.carleton.edu/research_education/geochemsheets/techniques/SEM.html.