Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Solar fuel generation by graphitic carbon nitride composites Lachance, Robert 2020

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
24-ubc_2020_november_lachance_robert.pdf [ 12.87MB ]
Metadata
JSON: 24-1.0392943.json
JSON-LD: 24-1.0392943-ld.json
RDF/XML (Pretty): 24-1.0392943-rdf.xml
RDF/JSON: 24-1.0392943-rdf.json
Turtle: 24-1.0392943-turtle.txt
N-Triples: 24-1.0392943-rdf-ntriples.txt
Original Record: 24-1.0392943-source.json
Full Text
24-1.0392943-fulltext.txt
Citation
24-1.0392943.ris

Full Text

SOLAR FUEL GENERATION BY GRAPHITIC CARBON NITRIDE COMPOSITES by  Robert Lachance  B.S., Columbia University, 2012  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF APPLIED SCIENCE in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Chemical and Biological Engineering)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  August 2020  © Robert Lachance, 2020  ii   The following individuals certify that they have read, and recommend to the Faculty of Graduate and Postdoctoral Studies for acceptance, a thesis entitled:  SOLAR FUEL GENERATION BY GRAPHITIC CARBON NITRIDE COMPOSITES  submitted by Robert Lachance in partial fulfillment of the requirements for the degree of Master of Applied Science in Chemical and Biological Engineering  Examining Committee: Fariborz Taghipour, Chemical and Biological Engineering Supervisor  Kevin Smith, Chemical and Biological Engineering Supervisory Committee Member  Xiaotao Bi, Chemical and Biological Engineering Supervisory Committee Member          iii  Abstract  Solar fuels are an attractive medium for long term storage of solar energy, but the low efficiency of photocatalysts in visible light prevents large scale adoption. Graphitic carbon nitride (g-CN) is a promising photocatalyst for solar fuel generation with a medium bandgap and demonstrated ability for photocatalyst engineering to improve charge recombination and charge mobility.   In this work the improvement of g-CN material is investigated via phosphorous doping and post synthesis techniques to stabilize the g-CN surface chemistry and prepare the material for heterojunction integration. Further, the improved visible light absorption and activity of a self-assembled 2D-2D precious metal free g-CN type II heterojunction z-scheme is presented, targeting photocatalytic CO2 reduction and hydrogen evolution for solar fuel applications. The synergistic combination of highly reducing electrons in the conduction band of g-CN and very oxidizing holes in the valence band of BiVO4 enables tunable, small bandgap, visible light active materials to perform artificial photosynthesis. Strong interfacial ohmic contact induced by the electrostatic attraction of z-scheme constituent materials to promote charge separation is investigated with the addition of reduced graphene oxide (rGO) acting as an electron transfer medium.   A realistic flat plate circulating batch reactor operating in the vapor phase, illuminated by a Xe arc solar simulator and coupled with online gas chromatography is employed for the evaluation of photocatalyst materials. This z-scheme employs facile synthesis techniques to achieve a visible light active, tunable band edge constituent material with similar reaction overpotentials to iv  single, UV active materials such as TiO2. Photoreduction of CO2 improves over 100% and water splitting improves nearly 40% for the g-CN/BiVO4 composite and is attributed to strong interfacial z-scheme charge transfer. Protonation and phosphorous doping are shown to influence the amount and oxidation state of the amine functional groups present on the g-CN surface, with protonation demonstrating nearly 4X increase in CO2 reduction compared to washed g-CN and attributed to increased CO2 adsorption. The work presented provides a pathway to electrostatic self-assembly of 2D-2D z-scheme heterojunctions for improved photocatalytic solar fuel generation.  v  Lay Summary  This project aims to improve the efficiency of storing solar energy in chemical resources that can be readily used in existing infrastructure. By artificially simulating the growth of plants, carbon dioxide, water, and sunlight can be transformed into synthesis gas that can later be utilized to release this stored solar energy. The materials that can facilitate this transformation of carbon dioxide into solar fuels with sunlight have typically been rare or expensive and utilized precious metals such as gold and platinum. In this project, the precursor materials of urea and graphite are transformed and combined into composites with improved conversion of carbon dioxide and water into energy commodities. These material composites synchronize to improve the overall conversion of light to chemical energy and therefore more effectively store solar energy to be used on medium to long term timescales, tackling the intermittency challenges associated with renewable energy production.  vi  Preface  The design, identification and conduct of this research program was wholly performed by the author under the supervision of Prof. Fariborz Taghipour. Shared instrumentation resources were used to characterize materials synthesized in the Photo-Reaction Engineering Laboratory in the Department of Chemical and Biological Engineering at UBC as per the following:  x XRD, Anita Lam, Chemistry, UBC x Diffuse Spectroscopy, Clean Energy Research Lab, UBC x PL Spectroscopy, Shared Instrument Lab, Chemistry, UBC x XPS/EDX/SEM/XRD, Shihong Xu, Nanofab Lab, University of Alberta x SEM, Owen Gethin, Dentistry, Center for High-Throughput Phenogenomics UBC  Throughout the research Prof. Fariborz Taghipour has guided the development of research objectives, plans and data analysis. Two manuscripts were prepared from the materials presented in chapter 1 and chapter 2 and will be submitted for publication in peer-reviewed journals.   vii  Table of Contents  Abstract ......................................................................................................................................... iii  Lay Summary .................................................................................................................................v  Preface ........................................................................................................................................... vi  Table of Contents ........................................................................................................................ vii  List of Tables ................................................................................................................................ xi  List of Figures .............................................................................................................................. xii  List of Symbols .............................................................................................................................xv  List of Abbreviations ................................................................................................................. xvi  Acknowledgements .................................................................................................................. xviii  Dedication ................................................................................................................................... xix  Chapter 1: Introduction ................................................................................................................1  1.1 Significance..................................................................................................................... 1 1.2 Artificial Photosynthesis and Water Splitting................................................................. 1  Existing Challenges .................................................................................................... 4   Photocatalyst Design Strategies .................................................................................. 5  Semiconductor Materials ............................................................................................ 8   Doping and Post Synthesis Treatment ...................................................................... 10  Exfoliation................................................................................................................. 12   g-CN Composites ...................................................................................................... 13   Type II Heterojunctions ............................................................................................ 14 1.3 Knowledge Gaps ........................................................................................................... 16  viii  1.4 Research Objectives ...................................................................................................... 17  1.5 Thesis Outline ............................................................................................................... 17  Chapter 2: The Impact of Doping and Post Synthesis Strategies on the Surface Amine Functionalization g-CN Materials ..............................................................................................19  2.1 Experimental ................................................................................................................. 19  Chemicals and Materials ........................................................................................... 19  Photocatalyst Preparation.......................................................................................... 20  Characterization Methods ......................................................................................... 20   Evaluation of Photocatalytic Activity ....................................................................... 21 2.2 Results and Discussion ................................................................................................. 25   X-ray Diffraction (XRD) .......................................................................................... 25  Scanning Electron Microscopy (SEM) and Energy Dispersive X-ray Spectroscopy (EDX) 26  X-Ray Photoelectron Spectroscopy (XPS) ............................................................... 27  UV-Visible Spectroscopy ......................................................................................... 32  Photoluminescence (PL) Spectroscopy..................................................................... 33  Photocatalyst Band Structure .................................................................................... 35  Photocatalytic CO2 Reduction .................................................................................. 36  Photocatalytic Mechanism ........................................................................................ 39 Chapter 3: 2D Electrostatically Self-Assembled g-CN Composite Photocatalysts ................43 3.1 Experimental ................................................................................................................. 43  Chemicals and Materials ........................................................................................... 43  Photocatalyst Preparation.......................................................................................... 44 ix  3.1.2.1 p-g-CN .............................................................................................................. 44 3.1.2.2 BiVO4 Nanosheets............................................................................................. 44 3.1.2.3 Photocatalyst Composite Preparation ............................................................... 45  Characterization Methods ......................................................................................... 45   Evaluation of Photocatalytic Activity ....................................................................... 46 3.2 Results and Discussion ................................................................................................. 46   X-ray Diffraction (XRD) .......................................................................................... 46  Scanning Electron Microscopy (SEM) and Energy Dispersive X-ray Spectroscopy (EDX) 48  X-ray Photoelectron Spectroscopy (XPS) ................................................................ 52  UV-Visible Diffuse Reflectance Spectroscopy (DRS) ............................................. 57  Photoluminescence (PL) Spectroscopy..................................................................... 59  Photocatalyst Band Structure .................................................................................... 60  Photocatalytic CO2 Reduction .................................................................................. 62  Proposed Charged Transfer Mechanism ................................................................... 65  Chapter 4: Conclusions and Recommendations .......................................................................70  4.1 Conclusions ................................................................................................................... 70  The Effect of Post Synthesis Preparation Techniques .............................................. 70  Phosphorous Doping in g-CN Catalysts ................................................................... 71   The Effect of g-CN/BiOV4 Z-Scheme Mass Ratios on Photocatalyst Activity........ 73  The Performance of rGO as a Z-scheme Mediator in g-CN/BiVO4 Z- schemes ...... 74 4.2 Recommendations and Future Directions ..................................................................... 75  Further Catalyst Characterization ............................................................................. 75 x  4.2.1.1 Ultraviolet Photoelectron Spectroscopy (UPS) ................................................ 75 4.2.1.2 Electrical Impedance Spectroscopy .................................................................. 76  Experimentation ........................................................................................................ 76 4.2.1.1 Materials and Synthesis .................................................................................... 76 4.2.1.2 Z-Scheme Mediators ......................................................................................... 77  Experimental Apparatuses ........................................................................................ 77 4.2.2.1 Preliminary Experimentation ............................................................................ 77 4.2.2.2 Reactor Sealing ................................................................................................. 78 4.2.2.3 Reactor Geometries ........................................................................................... 78 4.2.2.4 Reactor Materials .............................................................................................. 78 4.2.2.5 Temperature and Pressure ................................................................................. 79 4.2.2.6 Reaction Conditions .......................................................................................... 79 4.2.2.7 Light Source ...................................................................................................... 80 References .....................................................................................................................................81   xi  List of Tables  Table 2.1 XPS atomic ratio of each g-CN material ...................................................................... 29 Table 2.2 The bandgap of each photocatalyst estimated from DRS ............................................. 33  Table 2.3  The band structure of each g-CN photocatalyst species. ............................................. 35 Table 3.1 Atomic ratios of p-g-CN, BiVO4, and theoretic materials ............................................ 53 Table 3.2 The bandgap of each photocatalyst and composite ...................................................... 58 Table 3.3 The band edges of each photocatalyst and their composites. ....................................... 61 xii  List of Figures  Figure 1.1 - Photo excitation and charge recombination with relevant reduction and H2 evolution reactions .......................................................................................................................................... 2  Figure 1.2 – Photocatalyst (a) direct Z-scheme and  (b) type II heterojunction with relevant reaction potentials and band structure features ............................................................................... 6  Figure 1.3 – p-g-CN band structure with relevant reaction potentials, reactants, and products ... 10  Figure 2.1 Solar simulator incident light intensity distribution .................................................... 22  Figure 2.2 Experimental apparatus flow diagram ......................................................................... 23  Figure 2.3 Circulating batch reactor flow diagram during operation ........................................... 24 Figure 2.4 XRD spectra for g-CN, g-CN washed, P-g-CN, p-g-CN and p-P-g-CN photocatalysts....................................................................................................................................................... 25  Figure 2.5 SEM images of (a) g-CN, (b) p-g-CN (c) P-g-CN and (d) EDX elemental mapping of P-g-CN .......................................................................................................................................... 27  Figure 2.6 XPS survey spectra of g-CN photocatalysts ................................................................ 28  Figure 2.7 XPS P 2p peaks of P-g-CN and p-P-g-CN .................................................................. 29  Figure 2.8 XPS (a) C 1s and (b) N 1s peaks of g-CN photocatalysts ........................................... 30  Figure 2.9 (a) Absorbance and (b) diffuse reflectance spectra of g-CN photocatalysts ............... 32 Figure 2.10 PL spectra of p-g-CN, p-g-CN/BiVO4, p-g-CN/rGO/BiVO4, p-g-CN/rGO photocatalysts ................................................................................................................................ 34  Figure 2.11 A schematic of band structure shown in conjunction with the potential of pertinent reactions ........................................................................................................................................ 36  xiii  Figure 2.12 Cyclic experiments of p-g-CN repeated for 5 hours each with error bars indicating 90% confidence intervals .............................................................................................................. 37  Figure 2.13 Photocatalytic production of each g-CN photocatalyst with error bars indicating 90% confidence intervals ...................................................................................................................... 38  Figure 3.1 XRD spectra of p-g-CN, BiVO4 and p-g-CN composites ........................................... 47 Figure 3.2 SEM imagery of (a) p-g-CN (b) BiVO4 (c) p-g-CN/rGO (d) p-g-CN/BiVO4 and (e) p-g-CN/rGO/BiVO4 ......................................................................................................................... 49 Figure 3.3 EDX elemental mapping of (a) p-g-CN, (b) BiVO4, (c) p-g-CN/rGO, (d) p-g-CN/BiVO4, (e) p-g-CN/rGO/BiVO4 ............................................................................................. 51 Figure 3.4 The XPS survey scan spectra of p-g-CN, BiVO4, and p-g-CN composites ................ 52 Figure 3.5 XPS spectra of (a) the C1s and (b) the N 1s orbitals for p-g-CN and p-g-CN composites..................................................................................................................................... 54  Figure 3.6 XPS spectra of the (a) V 2p and (b) Bi4f orbitals ....................................................... 55 Figure 3.7 XPS spectra of the O1s orbital .................................................................................... 56  Figure 3.8 (a) Absorbance and (b) diffuse reflectance spectra of each photocatalyst and composite ...................................................................................................................................... 57  Figure 3.9 PL spectra of p-g-CN, p-g-CN/BiVO4, p-g-CN/rGO/BiVO4, p-g-CN/rGO photocatalysts ................................................................................................................................ 59  Figure 3.10 The band structure of p-g-CN and BiVO4 shown graphically .................................. 61 Figure 3.11 The yield of p-g-CN and investigated composites with error bars indicating 90% confidence invervals. .................................................................................................................... 62  Figure 3.12 The yield of the weight ratios of p-g-CN/BiVO4 with error bars indicating 90% confidence intervals. ..................................................................................................................... 63  xiv  Figure 3.13 The band structure of the p-g-CN/BiVO4 composite with a proposed charge transfer mechanism .................................................................................................................................... 66  Figure 3.14 The band structure of p-g-CN/rGO/BiVO4 with a proposed charge transfer mechanism .................................................................................................................................... 68                    xv  List of Symbols  A  Absorbance d  Lattice spacing e-  Electron ECB  Conduction band potential [eV] Ee  Energy of an electron at the vacuum level [eV] Eg  Bandgap EVB  Valence band potential [eV] F(R)  Kubelka-Munk function gcat  Grams of catalyst h  Planch Constant [eV s]  h+  Hole hr  Hour n  Electronic transition coefficient R  Reflectance X  Mulliken Electronegativity α  Adsorption coefficient [nm-1] η  Reaction overpotential θ  Diffraction angle [degree] λ  Wavelength [nm] ν  frequency of light [s-1] π  Pi bond  xvi  List of Abbreviations  2D  Two dimensional BG  Bandgap CB  Conduction band DRS  Diffuse reflectance spectroscopy EDX  Energy dispersive x-ray spectroscopy GC  Gas chromatography  g-CN  Graphitic carbon nitride GO  Graphene oxide HER  Hydrogen evolution reaction HOMO Highest occupied molecular orbital  LUMO Lowest unoccupied molecular orbital NHE  Normal hydrogen electrode OER  Oxygen evolution reaction P-g-CN Phosphorous doped graphitic carbon nitride p-g-CN Protonated graphitic carbon nitride PL  Photoluminescence spectroscopy p-P-g-CN Protonated Phosphorous doped graphitic carbon nitride rGO  Reduced graphene oxide SEM  Scanning electron microscopy  UV  Ultra-violet VB  Valence band xvii  Vis  Visible XPS  X-ray photoelectron spectroscopy XRD  X-ray diffraction spectroscopy    xviii  Acknowledgements  Writing and completing the work associated with this thesis has been a journey for me both personally and professionally. I have learned more about myself in this time than I have about the intended research topic and I am certain that if I am able to apply these lessons, they will serve to me well. I would like to use this opportunity to acknowledge those who helped to facilitate this time of personal and professional growth.  Professor Fariborz Taghipour has been a guiding light and advocate of my work since my arrival at UBC, but particularly as I have concluded my work in the midst of the COVID-19 pandemic and family health issues. His flexibility and understanding of my personal challenges have allowed me to polish this work to a degree that I would not have been able to accomplish without his support.  I would like to acknowledge my colleagues in the Photo-reaction Engineering Lab who have provided a sounding board in tackling challenges in the implementation and interpretation of the work completed in 626. In particular, I would like to express my sincere appreciation to Shahriar Rouhani Anaraki for his optimism and tireless smile in the face of trials and Hengameh Delbari for her comradery in tackling the challenging topics of CO2 reduction and life.  Finally, I would like to thank all of those people who provided support to me in both my personal and professional life since beginning this endeavor. Each of you brought fulfillment and purpose to the time spent and I have learned so much from you in the lab and in the mountains. xix  Dedication          To my mother, whose dedication to raising her children and now grandchildren is unparalleled.1  Chapter 1: Introduction 1.1 Significance Anthropogenic carbon emissions and CO2 levels have exponentially increased since industrialization due to consumption of fossil fuels and these activities are widely accepted to be the primary contribution to climate change [1]. A promising method to reduce CO2 emissions from the burning of fossil fuels is the photocatalytic reduction of CO2 to carbon fuels and the splitting of water to yield carbon free hydrogen fuel. Photocatalytic CO2 reduction and water splitting permits long term solar energy storage in the form of chemical energy, solving the challenge of intermittent solar energy production limiting grid penetration and allowing the use of current energy distribution networks. However, current photocatalysts for CO2 reduction and water splitting exhibit low quantum efficiency (QE) and often use expensive noble metal materials, making the process economically infeasible. Therefore, an effective and inexpensive catalyst material is sought to carry out CO2 photoreduction to solar fuels at improved efficiencies.  1.2 Artificial Photosynthesis and Water Splitting Photocatalysis for CO2 reduction and water splitting are facilitated by utilizing the energy of the radiative spectrum and can be represented by the two overall equations [2]:  𝑛(𝐶𝑂ଶ) + 𝑚(𝐻ଶ𝑂) + ℎ𝜈 → 𝐶௡𝐻௠ + (𝑛 + 𝑚 2ൗ )𝑂ଶ   (1) 𝑛(𝐻ଶ𝑂) + ℎ𝜈 → 𝑛𝐻ଶ + (𝑛 2ൗ )𝑂ଶ     (2)  2  Although these equations are written with energetic input, photocatalysis mechanistically proceeds downhill after an initial photoexcitation of an electron in a semiconductor material from the lowest unoccupied molecular orbital (LUMO) or conduction band (CB)  to the highest occupied molecular orbital (HOMO) or valence band (VB). Photo excitation proceeds when a photon of sufficient energy excites an electron in the VB of a semiconductor to the CB and leaves behind a hole (h+) as seen in Figure 1.1. Both the electron and the remaining hole can then act as charge carriers, performing reduction and oxidation reactions at the potential of their orbitals. The potential difference between the CB and the VB is referred to as the bandgap (BG, Eg) and the VB and CB are collectively termed the band edges (BEs). Although the CB and VB do not overlap in energetic potential for semiconductors, the fermi level can be understood as the average energy of an electron. Semiconductors with a fermi level lower than the average between the CB and VB potentials vs. the vacumm level are termed p-type while semiconductors with higher fermi levels are n-type. Due to the quantum limitation of the band gap versus the   Figure 1.1 - Photo excitation and charge recombination with relevant reduction and H2 evolution reactions 3   continuous solar spectrum, there is a maximum theoretical efficiency for any semiconductor material according to the Shockley–Queisser relationship4. Since photons can only be adsorbed at potentials greater than or equal to Eg for the semiconductor, all photons at wavelengths below the BG are not absorbed [3]. Further, the energy of photons in excess of the Eg is emitted as light, resulting in higher energy photons not being fully captured. The fixed band gap of semiconductors vs the continuous solar spectrum leads to an optimal Eg of 1.34 eV for single junction semiconductor materials, corresponding to a wavelength of 925nm [3].  Reactions of interest for CO2 reduction and water splitting as well as the reaction standard potentials (Eo) are presented in Figure 1.1 with the half reactions below:  CO2 + 2H+ + 2e- Æ CH2O2    E[eV] = -0.61  (3) CO2 + 2H+ + 2e- Æ CO + H2O   E[eV] = -0.53  (4) 2H+ + 2e- Æ H2    E[eV] = -0.41  (5) 2CO2 + 12H+ + 12e- Æ C2H4 + 4H2O  E[eV] = -0.34  (6) CO2 + 8H+ + 8e- Æ CH4 + 2H2O   E[eV] = -0.24  (7) 2H2O + 4h+ Æ O2 + 4H+     E[eV] = +0.81  (8)  Although the difference between the oxygen evolution reaction (OER) and the reducing potentials required for artificial photosynthesis and water splitting above is quite small, a potential difference in excess of the standard potential for a reaction is typically required to drive the reaction, termed the reaction overpotential (η). In addition, the consumption of photo-generated electrons and holes typically proceeds via separate reduction and oxidation reactions. 4  This means that a semiconductor must have a CB with a negative enough potential to drive the reduction reaction of interest and a positive enough potential to drive an oxidation reaction on the VB.    Existing Challenges The mismatch between the required reaction potentials of a semiconductor and the optimal bandgap for adsorption of the solar spectrum is a major source of inefficiency in photocatalysis [4]. CO2 reduction in particular is challenging due to the large overpotential required to initiate reduction of the C=O bond [3]. As a result of the stability of the CO2 molecule, it has proven challenging to find semiconductor materials with the catalytic activity to decrease the required overpotential for CO2 reduction.    The catalytic activity of photocatalysts is often attributed to morphological defects such as vacancies, steps, corners and edges. However, in semiconductors these defects often create surface states at intermediary positions within the BG promoting the diffusion of electrons towards the holes from which they originated resulting in charge carrier recombination [5]. Charge carrier recombination is one of the primary reasons for low efficiency in photocatalytic CO2 reduction. Therefore an optimization problem exists between photocatalyst activity and recombination of charge carriers with respect to surface morphology. Finally, the slow kinetics of the multielectron reduction reactions are further hindered by the low solubility of CO2 in water, resulting in low reactant concentrations in solution [5].   5  In summation, the following factors must be taken into account in the design of semiconductor materials for CO2 reduction and water splitting applications: x Mismatch between semiconductor band gap and the solar spectrum x Charge recombination x Slow reaction kinetics x Low solubility of CO2 in water   Photocatalyst Design Strategies A material design strategy which targets charge recombination is heterojunction schemes [6].  These are formed when two photocatalysts or co-catalysts are brought into intimate interfacial contact. This contact  induces band bending, an alteration of band edge potential, while the band gap remains unchanged. Band bending induces the formation of an internal magnetic field which can act to increase charge separation.  Depending on the relative position of their band gaps, heterojunctions can behave differently. The most interesting applications for CO2 reduction are type II heterojunctions as can be seen in Figure 1.2. Type II heterojunctions are able to spatially separate charge carriers, limiting the amount of recombination due to electron diffussion, increasing charge carrier lifetime. A specific type II heterjunction of interest is the photocatalyst Z-scheme as seen in Figure 1.2(a). Z-schemes mimic the natural mechanism of photosynthesis by intentionally shuffling electrons to achieve a spatial separation of electons and holes as well as higher reduction and oxidation potential available for the overall reaction. Z-scheme type II heterojunction charge transfer mechanisms for g-CN allow photogenerated electrons in the conduction band of a second photocatalyst to recombine with the holes in the valence band of a primary catalyst, mitigating charge recombination. As a result, strongly reducing electrons on the 6  CB of one photocatalyst and strongly oxidizing holes on the VB of a secondary photocatalyst enhance the redox ability of the photocatalytic system to drive the CO2 reduction reaction. The efficacy of heterojunction photocatalyst systems can be tuned depending on composition, morphology, and the interfacial interaction between the two materials. Z-schemes can be carried out with or without a charge transfer intermediate material.   The primary benefit of using a z-scheme is the improved reduction and oxidation potential available for reaction while retaining moderate photocatalyst bandgaps of each material, allowing them to operate in the visible light range [6]. A single photocatalyst needs  Figure 1.2 – Photocatalyst (a) direct Z-scheme and  (b) type II heterojunction with relevant reaction potentials and band structure features only to perform either a reduction or oxidation reaction and the other photocatalyst in the pair can perform the other. As a result, the band edge potentials of each photocatalyst material can be more appropriate for the reaction of interest and the size of the bandgap can be smaller. 7  Solid state Z-schemes can also be employed in vapor phase reaction rather than in solution, removing the limitation of low reactant concentration typical in aqueous conditions. However, the small spatial separation required for z-scheme charge transfer also can result in back reactions that limit the overall production of the photocatalyst system. Despite this limitation, z-schemes have been of great interest in recent literature [3, 5, 6].   The qualities that constitute a favorable z-scheme are as follows: x Appropriate band edge potentials and catalytic activity for the reaction potential and required overpotential of interest x Capable of intimate Ohmic contact at semiconductor interfaces over a large portion of photocatalyst area x Stable under irradiation and in time  For direct z-schemes, intimate ohmic contact is typically achieved via complex deposition techniques in catalyst synthesis that allow alignment of the crystal lattices of each semiconductor [6]. This contact is difficult to achieve but is necessary since grain boundaries and defects could result in poor electron transfer and a poor type II heterojunction. Indirect z-schemes still require this contact, but typically achieve it via noble metal and other conductor interfaces acting as electron shuttles [5]. Therefore, strong performance of an indirect z-scheme typically requires expensive noble metals and semiconductor materials with good conductivity for electrons to reach the noble metal interface before recombination occurs. In addition, the contact area between noble metals and photocatalysts is small due to the point contact of metal particles on semiconductor planar surfaces. Therefore, z-schemes which can avoid the complex synthesis 8  issues of direct z-schemes without the use of expensive noble metals, have appropriate reaction potentials and catalytic activity for CO2 reduction are sought.    Semiconductor Materials Since the discovery of the photodecomposition of H2O in 1972 by Fujishima and Honda [7] the most widely studied semiconductor to date for water splitting and CO2 reduction has been TiO2. Although TiO2 has notched some of the highest generation rates of CO2 reduction and water splitting to date, the band gap of the material is too large to utilize the vast majority of the solar spectrum [8]. Despite various attempts at narrowing the bandgap of TiO2 via doping and utilizing different morphological phases, the pursuit of TiO2 based semiconductor systems for direct renewable energy generation continues to be an impractical exercise as is true for other large band gap materials. Various systems of noble metal doped semiconductors have also achieved relatively high reaction rates, including TiO2 materials. However, these precious metals are expensive for renewable energy applications at scale [8].  Various other metal oxides have also shown good photocatalytic activity, but often have band gaps too large to develop solar energy applications. Bismuth Vanadate (BiVO4), Bismuth Iodide Oxide (BiIO) and Copper (I) Oxide (Cu2O) have shown promising results with medium bandgaps that permit use of a portion of the solar spectrum [9, 10, 11]. In particular, BiVO4 and BiIO have demonstrated impressive oxidative potential, but lack the overpotential necessary to drive CO2 reduction. Conversely, Cu2O has a large overpotential for CO2 reduction, but does not have a large overpotential for oxygen evolution. In comparison to TiO2, BiVO4 and BiIO represent a compromise between redox reaction overpotential and solar spectrum utilization, but 9  could be used effectively in a z-scheme as the oxidizing photocatalyst [11]. Monoclinic BiVO4 has been reported as the most active phase of BiVO4 and is an interesting photocatalyst due to a z-scheme compatible band structure, narrow bandgap, and its strong photocatalytic activity for oxygen evolution [12, 13, 14]. These characteristics make BiVO4 integrated z-scheme systems capable of artificial photosynthesis with much more strongly oxidizing holes to perform oxygen evolution.  A plethora of carbon based photocatalysts have also shown promising photocatalytic activity in recent literature. Specifically, graphitic carbon nitride (g-CN) has been shown to have strong reductive activity for CO2 reduction and water splitting with a medium bandgap and a high degree of catalyst tunability via synthesis precursors and methodology [15, 16, 17]. Bulk polymeric g-CN is composed of 2 dimensional layers of polymerized Tris-s-triazine. The primary method of synthesis of g-CN is via heating melamine, urea, thiourea, or a combination of these materials to between 500 and 600oC [18]. Depending on the pyrolysis treatment temperature, the crystalline structure present in the final product can be altered resulting in changes to the material band gap and surface chemistry.  The redox band edge potentials of g-CN can be seen in Figure 1.3. The tunable bandgap of g-CN has been demonstrated to be capable of CO2 reduction, hydrogen evolution (HER) and oxygen evolution (OER) making it a viable catalyst for artificial photosynthetic fuel production applications [18, 19, 20].  g-CN is of interest since it contains no metals, is abundant and easily synthesized via pyrolysis techniques, is tunable via morphological changes, is non-toxic and is highly stable [18]. 10   Figure 1.3 – p-g-CN band structure with relevant reaction potentials, reactants, and products    However, the challenges of this material include small surface areas in bulk samples, moderate visible light adsorption, fast charge recombination, moderate oxidation ability and low charge carrier mobility. Strategies for the improvement of g-CN studied to date have included metal nanoparticle decoration, non-metal doping, nano-structuring and exfoliation methods, heterojunctions with other photocatalysts, and protonation of the surface [18, 19].    Doping and Post Synthesis Treatment Doping has long been a common method to improve photocatalyst qualities in semiconductors and the literature shows it has also been quite effective for g-CN. Intrinsic doping with various 11  non-metals has been shown to be effective for H2 evolution, decreasing the bandgap, increasing photocatalyst surface area, and minimizing non-radiative recombination through the creation of midgap states [21, 22]. Specifically, doping with I [23, 24], P [25, 26, 27], B [28, 29], O [30] and S [31, 32, 33] have been shown to lead to improved performances for H2 evolution, but the topic has been only recently and far less extensively researched for CO2 reduction (O [34], P[30, 35, 36, 37] and S [38]). Since doping strategies typically create decrease the band gap and lower the reducing potential of the CB, CO2 reduction is limited by the reduced overpotential due to the stability of the C=O bond [20].   However, some promising results have been achieved in recent research showing O doped g-CN have better CO2 reducing photoactivity than bulk g-CN [34]. Improved results for CO2 reduction with S doped g-CN and enhanced capture of CO2 through an amine rich surface of P doped g-CN have been shown to lead to higher photocatalyst production and high selectivity [36]. P doping is of particular interest since it results in only slightly reducing the potential of the CB while increasing the VB, resulting in a shorter band gap with similar overpotential for CO2 reduction and water splitting [18, 36]. P doping has also been shown to create more amorphous material as a result of P integration into the g-CN heterocycles [36, 35]. Recent research has found that CO2 adsorption over 0.02 wt% P in P-g-CN resulted in a 3.14X increase, resulting in a 3.1 and 13.9X increase in CO production and CH4 evolution with H2 evolution also increasing 2.69 times [36]. However, Raziq et al. has shown that H2, CO, and CH4 generation increase only 2X in much higher P content doped material with the addition of Pt cocatalyst sites, indicating that lower doping concentrations may be beneficial [35]. The high nitrogen content of P doped g-CN and it’s resulting increased CO2 adsorption has been found to be more effective for increasing CO2 12  reduction than increases in surface area, suggesting CO2 reduction may be adsorption limited [39].  Exfoliation Due to the minimal effects of surface area, exfoliation techniques for g-CN have focused on decreasing layer thickness as a way to mitigate the effects of poor charge transfer. The layers of g-CN have been shown to be easily exfoliated via sonication techniques in various solvents into ultrathin 2D sheets [40]. Evidence suggests that layer separation increases surface area, allows for more facile charge transport to the material surface, increases the CB potential due to quantum confinement effects and provides a larger overpotential for CO2 reduction. However, increases in CB potential is accompanied by an increased BG, limiting light absorption. Methods to exfoliate thinner layers have included the use of acids, which has been documented to have the additional effect of protonating the g-CN surface, giving it a positive charge while retaining its activity [41, 42]. Specifically, H2SO4 [43] and HCL [16] have been shown to protonate the heterocycles of g-CN while maintaining activity despite increasing the bandgap.   Post synthesis treatment of g-CN materials is highly inconsistent when comparing g-CN materials altered to improve the photocatalyst and in attributing reasons for increased activity. Similarly, P doping has been attributed to improvements in g-CN activity via increased conductivity, introduction of mid gap states,  band structure changes, surface chemistry, morphology and crystallite size, improved optical adsorption, separation of charge carriers and charge transfer efficiency, but the attribution of the primary improvements are still debated. As a result, the cause for the effects of P doping and post-synthesis treatments on CO2 reduction and H2 evolution are still an area requiring further understanding. 13    g-CN Composites One of the primary advantages to positively charged g-CN material is that it permits the electrostatic self-assembly of composite structures with negatively surface charged materials to act as cocatalysts [16, 41]. Other methods such as hydrothermal techniques are energy and time intensive while others such as chemical reduction have the capacity to affect the activity of the photocatalyst [44].   A prominent example of this self-assembly technique found in literature is the composite of reduced graphene oxide (rGO) with g-CN. The g-CN/rGO composite has been shown to increase the mobility of surface charges as a result of rGO’s high conductivity while also increasing the light absorption of the composite as a result of a redshift [15, 45, 46]. The increased radiation adsorption at higher wavelengths has been attributed to 𝜋 bond conjugation between the two materials. Further, oxygen containing groups on the surface of rGO have been shown to act as charge localization centers, facilitating multi-electron transfer reactions such as CO2 reduction and water splitting by acting as a co-catalyst [47, 48]. The lamellar structure of p-g-CN and the 2D sheets of rGO result in an electrostatically self-assembled 2D composite with high interfacial contact area and improved photocatalytic production for CO2 reduction and H2 evolution.   Many authors have studied rGO and other carbon based cocatalysts and shown that electrostatic self-assembly can result in a 2.3-5.4X improvement in CO2 reduction [15, 17]. Further, more complex core-shell structures have also been shown to be capable of self-assembly, making this technique interesting for future applications [42, 49].  14    Type II Heterojunctions Direct and indirect Type II heterojunctions and z-schemes have also been demonstrated to be a successful strategy for the improvement of g-CN through material composites, but often require time consuming calcination, chemical reduction and hydrothermal reactions [20]. ‘One-pot’ pyrolysis reactions have been shown to be effective, but the poor control of synthesis conditions is limiting [50]. A method to bypass the challenges of establishing a direct z-scheme is to utilize an interfacial material to facilitate charge transfer to create an indirect z-scheme. Precious metal nanoparticles as intermediate materials for indirect z-schemes have been shown to perform well in the literature [6, 51]. For example, a g-C3N4/Ag/BiVO4 indirect z-scheme utilizing silver nanoparticles as an interfacial conductor has shown improved performance attributed to mitigating the charge recombination challenges associated with g-CN [10]. However, the point contacts between semiconductor materials limit the interfacial charge transfer area and may instead result in charge recombination if charge carriers are not able to reach these metal nanoparticles.   An alternative interfacial material that creates planar interfacial contact between the components of the indirect z-scheme without the use of noble metals is rGO [52]. rGO has been shown to perform well as an interfacial contact medium in z-schemes due to its moderate conductivity and very large interfacial contact area [47]. For example, rGO was able to increase the degradation of dyes 29X by acting as an intermediate between the active planes of BiVO4 crystals and g-CN through a hydrothermal synthesis route [2]. Similarly, a Bi2WO6/rGO/g-CN z-scheme was able to increase CO2 reduction over 10X compared to g-CN [53] and an indirect ZnV2O6 /rGO/g-CN 15  z-scheme was able to increase CO2 reduction over 8X compared to g-CN [54] among other notable successes [55, 56, 57].  Direct z-schemes are considered the next evolution in heterojunction synthesis, though implementation has been challenging due to the need of a very specific relative band structure, activity for the reaction of interest and strong interfacial contact [6]. To some degree, band edge engineering and morphology alterations can create favorable conditions for direct z-scheme creation, and recent research has shown some success for direct z scheme systems. However, this area is mostly explored for dye degradation and water splitting and less so for CO2 reduction [5, 28, 51, 58, 59, 60, 61, 62, 63]. For example, a CO3O4-BiVO4/g-CN z- scheme has shown an 31.5X increase in reaction rate for dye degradation [64], while a BiVO4/g-CN z-scheme has shown a  4.5X increase in degradation rate [60]. These results show promising applications for expansion of this research into solar fuels generation. Indeed, A recent publication found a 2D-2D z-scheme was hydrothermally synthesized between g-CN and BiVO4, resulting in a 4.5X improvement in CO2 reduction [59]. However, this synthesis relied upon time intensive reductions reactions that are difficult to scale for solar energy applications.   Electrostatic self-assembly to create scalable z-scheme heterojunctions for solar fuel creation may be able to solve these synthesis challenges but has been explored in only a limited manner for CO2 reduction. For example, glyphosate degradation was recently shown to increase over 12X through the use of a g-CN/BiVO4 z-scheme assembled electrostatically [44]. A recent electrostatic g-CN/ZnO z scheme was established for CO2 reduction, generating a 3X increase in CO2 reduction, but was inappropriate for solar fuel applications due to the large bandgap of ZnO 16  and operation under UV radiation [49]. These promising results indicate that electrostatic self-assembly schemes are an area of promising research for CO2 reduction applications.  Electrostatically self-assembled z-schemes require appropriate band structure, suitable morphology, opposing surface charges between materials, large surface area and strong interfacial contact between materials. Recent publications for protonated g-CN and BiVO4 show the possibility of an electrostatically self-assembled z-scheme [44]. Both g-CN and BiVO4 have been shown to be synthesized in 2D nanosheets, decreasing electron diffusion length, increasing the overall area for reaction and improving interfacial contact between the materials [17, 64, 65, 66, 67]. Therefore, a g-CN/BiVO4 direct z-scheme and a g-CN/rGO/BiVO4 indirect z-scheme are of key interest to advancing the field of solar fuels generation.  1.3 Knowledge Gaps The literature review indicates vapor phase visible light photocatalysis with g-CN/BiVO4 z-schemes for CO2 reduction and hydrogen evolution have not been studied in depth and that phosphorus doping, post synthesis treatments and photocatalyst preparation could have an effect when extended to z-scheme photocatalyst activity. Specifically, these areas are:  x Effect of photocatalyst post synthesis preparation techniques of washing and protonation on the morphology, surface chemistry, crystallinity, optical absorption, and ultimately photocatalytic performance of the obtained photocatalyst 17  x Effect of phosphorous doping which may affect the parameters above as well as providing centers of charge accumulation due to the introduction of mid-gap states, potentially increasing the density of surface charges. x Effect of g-CN/BiVO4 ratios on the g-CN/BiVO4 z-scheme activity for CO2 reduction and hydrogen evolution. x Effect of rGO introduction as an electron mediator in a g-CN/rGO/BiVO4 z-scheme  1.4 Research Objectives The overall goal of this project is to explore various techniques to tailor the optical, structural, and photochemical properties of a noble metal free g-CN/BiVO4 z-scheme for visible light CO2 reduction and hydrogen evolution to solar fuels. To address this goal, the following objectives were defined for this research: x To investigate the effect of photocatalyst washing and protonation techniques on g-CN activity and z-scheme integration. x To determine the impact of phosphorous doping on g-CN photocatalyst activity.  x To investigate the effect of g-CN:BiVO4 mass ratios in a 2D-2D photocatalyst z-scheme. x To determine the impact of rGO as an electron mediator in a 2D-2D-2D g-CN/rGO/BiVO4 z-scheme.  1.5 Thesis Outline This introductory chapter (Chapter 1) reviews the fundamentals of photocatalysis as they pertain to CO2 reduction, the photocatalytic characteristics of g-CN, BiVO4, and rGO. A review of recent literature is presented regarding the amelioration of g-CN photocatalysts through surface 18  chemistry modifications, doping, and z-scheme synthesis strategies with emerging 2D material morphologies and composite material self-assembly strategies.  Chapter 2 comprehensively characterizes and analyzes the impact of doping and post synthesis strategies on the surface amine functionalization of g-CN and the resulting changes in photocatalyst characteristics and performance in the context of CO2 reduction and H2 evolution solar fuel applications.  Chapter 3 discusses the effect of g-CN z-scheme integration with and without the presence rGO as an electron mediator in sonication assisted electrostatically self-assembled 2D g-CN/BiVO4 composite photocatalysts. Characterization of these photocatalyst composites and the mechanisms of CO2 reduction and H2 evolution improvement is discussed.  The final chapter (Chapter 4) of this thesis presents the overall conclusion of this study and recommendations for future work in light of the observations presented.   19  Chapter 2: The Impact of Doping and Post Synthesis Strategies on the Surface Amine Functionalization g-CN Materials  g-CN protonation and phosphorous doping strategies have been documented to yield promising visible light photocatalysts for solar fuels applications, promoting enhanced CO2 reduction and hydrogen evolution. However, the amine functionalized surface chemistry of these materials has gained little attention compared to their influence in CO2 adsorption and reduction activity. In this chapter, g-CN and phosphorus doped g-CN (P-g-CN) are synthesized in facile one-pot pyrolysis techniques and subjected to washing and protonation post-synthesis treatment to quantify the influence of common strategies on surface amine content and subsequent photocatalyst activity for hydrogen evolution and CO2 reduction.   2.1 Experimental  Chemicals and Materials Urea (CO(NH2)2, ≥99.0%), sodium phosphate monobasic (NaH2PO4, ≥99%) and 2-propanol ((CH3)2CHOH), >99.5%) were obtained from Fischer Scientific. hydrocholoric acid (HCl, 37%) and nitric acid (HNO3, 70%) were obtained from Sigma Aldrich. Chemically reduced graphene oxide powder (C7.7-8.7O2.2-1.3) was obtained from Graphenea. Carbon dioxide (CO2, 99.995%) and nitrogen (N2, 99.999%) were obtained from Praxair. All the reagents in this work were at the analytical reagent grade and used without further purification. Deionized water was used throughout this study. The water was purified with a Milli-Q water ion-exchange system.  20    Photocatalyst Preparation g-CN was synthesized by heating urea in an air atmosphere in a pyrolysis method. Urea was dried in an oven for 24 hours at 80 oC. 20 g of dry urea was then ground to a fine powder in a ball mill and placed in a 50mL crucible with a lid. The crucible was heated in an air atmosphere to 520oC at a ramp rate of 10 oC/min, held at 520 oC for 2 hours, then cooled to room temperature. The photocatalyst was used directly as g-CN and used further in other syntheses. Washed g-CN was washed in 0.1M nitric acid and centrifuged three times, followed by washing with distilled water until the product was neutral. The product was then placed in the oven to dry at 80 oC for 24 hours. Protonated graphitic carbon nitride, hereafter referred to as p-g-CN, was stirred for eight hours in 1M HCl and then washed with distilled water in a centrifuge until neutral. The resulting material was returned to the oven to dry for 24 hours at 80 oC. Phosphorus doped graphitic carbon nitride, P-g-CN, was prepared with the same methodology as g-CN except 25mg of NaH2PO4 was added to the ball mill prior to grinding and pyrolized directly after. Protonated phosphorus doped graphitic carbon nitride, p-P-g-CN, was prepared in the same manner as p-g-CN except P-g-CN was used as a precursor instead of g-CN. P-g-CN yield was 3.4 wt% while g-CN yield was 5.1 wt%, showing a limitation to polycondensation in P doped samples.   Characterization Methods The crystal structure of photocatalyst materials were analyzed by a powder x-ray diffractometer (Rigaku Ultima IV) with Cu-kα radiation and a scan rate of 0.03/s with a 2θ range of 5o to 80o. The structural morphology, elemental composition and distribution was observed via field 21  emission scanning electron microscope with energy dispersive spectroscopy system (FE-SEM, Sigma with Oxford detector) as well as a Helios Nanolab 650 focused ion beam with scanning electron microscope (FIB/SEM). X-ray photo-electron spectrometer (Kratos Axis Ultra) with Al kα radiation. The optical absorption of photocatalysts were observed via diffuse reflectance spectrometer (Ocean Optics FLAME-S-XR1-ES). Photoluminescence spectroscopy was performed with a Cary Eclipse fluorescence spectrometer. A Sciencetech solar simulator (SS1kW) operating with a 1000W Xe arc lamp was used as a light source.   Evaluation of Photocatalytic Activity To compare the performance of the photocatalysts synthesized, experiments were conducted in a continuously flowing vapor phase batch reactor maintained at 70 oC, 2 atm absolute pressure with 13.62kPa partial pressure of H2O(g) via a LabVIEW online control and data acquisition module. A Sciencetech solar simulator model SS1kW powered by a 1000W Xe arc lamp was the light source for the experiment. The solar simulator maintained class B short term temporal instability (~1%), spatial non-uniformity of 12.42%, and Class B spectral non-uniformity compared to the solar spectrum [66] over the quartz window of the reactor. A spatial representation of light distribution is seen in Figure 2.1(a). The total average intensity entering the reactor after the placement of a 420nm cutoff filter and quartz window was 39.3mW/cm2.  The reactor design utilized here was a flat plate reactor utilizing a glass slide with drop-coated photocatalyst material. Symmetric design strategies implemented here created redundancy within the design implementation. For example redundant temperature and pressure detection ports sealed with swagelok fittings and gas flow diffusers at both the entrance and exit to the reactor 22  allowed the reactor to be utilized in a modular manner. Temperature was monitored with J type thermocouples in both temperature-controlled water column humidifier and in the reactor volume. The quartz window for the reactor entrance is 12.7 cm X 12.7 cm making use of the largest portion of the solar simulator output possible while maintaining fair non-uniformity of the incident light source. Total reactor volume was minimized by reducing the z-component of the reactor dimension while maintaining strong irradiance on the photocatalyst material surface in order to increase the overall concentration of detected product gases. Gas flow was maintained at 150mL per minute with a passively water cooled, epoxy potted electric pump controlled and calibrated via LabVIEW.  Figure 2.1 Solar simulator incident light intensity distribution   High purity CO2 and N2 were bubbled through the temperature controlled water collumn to achieve a consistent H2O:CO2 ratio. Prior to each experiment, the apparatus was purged with N2 23  gas to eliminate air and then the reactor was charged with the desired reactant mix. The reactant mix was measured at selected intervals with an online GC (7890A – Agilent). All experiments were carried out in an enclosure to prevent additional light partisipating in the reaction. A detailed schematic of the experimental apparatus is available in Figure 2.2 and Figure 2.3.  Figure 2.2 Experimental apparatus flow diagram  24    Figure 2.3 Circulating batch reactor flow diagram during operation   25  2.2 Results and Discussion  X-ray Diffraction (XRD) The XRD spectra for the synthesized catalysts can be seen in Figure 2.4. The spectra for all g-CN materials show two diffraction peaks at the (100) and (002) planes corresponding to g-C3N4 (JCPDS: 87-1526) [65, 67]. The g-CN and washed g-CN sample’s interlayer stacking of conjugated aromatic systems is evidenced by the (002) diffraction peak at 27.5o corresponds to a graphitic stacking spacing of 0.324nm, while the (100) peak at 13.1o reflects the in-plane repeating of tri-s-triazine of g-CN materials with a d spacing of 0.675nm [59, 67].   Figure 2.4 XRD spectra for g-CN, g-CN washed, P-g-CN, p-g-CN and p-P-g-CN photocatalysts For phosphorus doped g-CN samples the (002) peak broadens and shifts towards lower 2θ values showing that the interlayer spacing is less ordered and larger, respectively [36, 61, 62, 68]. 26  Protonation of g-CN also shifts the (002) peak to lower values demonstrating an increase in inter-layer spacing but retains peak intensity and therefore inter-layer order [16]. The (100) peak of P doped and protonated samples also broadens, suggesting a less ordered inter-later tri-s-triazine structure [61, 62, 68]    The more amorphous structure of P-g-CN suggests suppressed g-CN growth, resulting in smaller g-CN particle size and explaining the lower yield of P-g-CN syntheses [67].   Scanning Electron Microscopy (SEM) and Energy Dispersive X-ray Spectroscopy (EDX) The structure and composition of each sample was studied via scanning electron microscopy (SEM) and energy dispersive electron spectroscopy (EDX). As can been seen in Figure 2.5(a), g-CN shows a 2D sheet structure as has been reported for g-CN materials [41]. p-g-CN shows the sheet materials have been broken up during the protonation process in Figure 2.5(b), revealing more edge sites. This aligns with the XRD result that shows p-g-CN is less ordered and crystalline by virtue of the broader spectral peaks. Phosphorus doped g-CN material was revealed to consist of smaller particle size as shown in Figure 2.5(c) which is also supported by the more amorphous XRD spectra of P-g-CN, indicating that growth of the P-g-CN network was inhibited. The yield of phosphorus doped material during synthesis was 3.4% compared to a yield of 5.1% for g-CN further supporting growth of the P-g-CN network is inhibited. EDX data shows the presence of C, N, trace amounts of O and P in phosphorus doped samples. Figure 2.5(d) shows a uniform distribution of phosphorus in P-g-CN and p-P-g-CN, indicating incorporation of P into the g-CN structure and persistence of consistent levels of phosphorus after protonation. 27    Figure 2.5 SEM images of (a) g-CN, (b) p-g-CN (c) P-g-CN and (d) EDX elemental mapping of P-g-CN   X-Ray Photoelectron Spectroscopy (XPS) The XPS survey spectra in Figure 2.6 reveals that all g-CN materials synthesized are composed of C, N, a small amount of O and the presence of P for doped samples. Elemental ratios and stoichiometry of each catalyst is seen in Table2.1. The higher atomic % N in the untreated g-CN material has been suggested to be an abundance of surface functional amine groups and relatively less tertiary carbon [36, 48]. Interestingly, p-g-CN shows drastically less N after protonation, which may be due to a removal of these surface amine groups after acid treatment (a) (b) (c) (d) 28  while protonating of sp2 hybridized N [16, 43]. The amount of P doping has been determined to be 0.3% by XPS analysis and is maintained in protonated and doped samples as seen in Figure 2.7.  Figure 2.6 XPS survey spectra of g-CN photocatalysts 29   Figure 2.7 XPS P 2p peaks of P-g-CN and p-P-g-CN        Table 2.1 XPS atomic ratio of each g-CN material High resolution scans of the C1s have been deconvoluted into three peaks which for g-CN, which are at 284.8 eV, 288.1 eV, and 285.8 eV corresponding to adventitious carbon (C-C), hybridized carbon in N containing aromatic rings (N=C-N), and ammonia species (C-NH2), respectively [56, 67]. Washed g-CN, p-g-CN, P-g-CN, and p-P-g-CN samples shift the C-NH2 binding energy down 0.42eV, 0.27eV, 0.14eV, 0.35eV suggesting more protonated amine groups Sample description C:N atomic ratio Stoichiometric ratio Theoretic 1:1.33 C3N4 g-CN  1:1.36 C3N4.10 g-CN washed 1:1.34 C3N4.04 P-g-CN  1:1.33 C3N3.99P0.003 p-g-CN 1:1.19 C3N3.582 p-P-g-CN 1:1.36 C3N4.09P0.003 30  occupy these sites [67]. The C1s peak shifts down 0.1eV for N=C-N species in both the washed and P-doped samples suggesting the removal of surface amine groups and the effect of the incorporation of P into the heterocycles, respectively. The position of the P2p peak at 133.4eV corresponds with P-N coordination with P replacing C in the g-CN heterocycles [35, 36]. These spectra support the XRD result showing a more amorphous structure of P-g-CN since the relatively longer P-N bond length has been reported to yield a more amorphous structure with higher surface area arising from P doping [21, 69, 70].    Figure 2.8 XPS (a) C 1s and (b) N 1s peaks of g-CN photocatalysts (a) (b) 31  The N1s spectra seen in Figure 2.8 (b) can be deconvoluted into four peaks for each g-CN material. For g-CN the peaks are at 398.5eV, 399.7eV, 401.1eV and 404.0 eV which have been attributed to sp2 hybridized nitrogen (C-N=C) in tri-s-triazine rings, tertiary nitrogen (N-(C)3), free amine groups (C-N-H), and inter-layer π bond excitations. The position of sp2 hybridized nitrogen shifts down 0.1 eV in p-g-CN, which may be due to the π bonds becoming positively charged after protonation of the sp2 hybridized nitrogen [17, 35]. P-g-CN doesn’t observe the same shift after protonation, which may be due to addition of a valence electron when substituting P for C in P-g-CN heterocycles [36, 69]. The peak position of amine groups shifts down 0.1 eV, 0.1 eV, 0.4 eV, and 0.1eV for washed, P-g-CN, p-g-CN, and p-P-g-CN, respectively. The large peak shift observed upon protonation of g-CN is not observed when protonating P-g-CN, which may be attributed to the more amorphous structure of P-g-CN already containing a high degree of amine groups leaving fewer remaining sites for protonation. There is a shift in the peak attributed to π excitations of 0.3eV, 0.2eV, 0.1eV and 0.1 eV for washed, P-g-CN, p-g-CN, and p-P-g-CN, respectively. For p-g-CN, it is likely reflective of the degree of exfoliation as well as the net positive charge of π bonds resulting from the preferential protonation of sp2 hybridized N [43]. In the case of P-g-CN, the more amorphous, amine rich structure may similarly positively charge the heterocycles [40]. The peak associated with tertiary nitrogen groups also demonstrates peak position shift decreases of 0.1eV, 0.2 eV, and 0.4 eV for washed, P-g-CN and p-g-CN samples, respectively, whereas p-P-g-CN shifts back up 0.2 eV upon protonation. This suggests that N-(C)3 and C-N=C groups are both protonated by acid treatment [43].   32   UV-Visible Spectroscopy UV-vis diffuse reflectance spectroscopy (DRS) was used to characterize each photocatalyst across wavelength. Figure 2.9(a) shows the absorption each material and indicates that washed and protonated graphic carbon nitride photocatalysts absorb less light at higher wavelength than do other samples. However, p-P-g-CN shows the same level of light absorption after protonation, which is likely due to the relatively amorphous and structure of the material as evidenced by the XRD and SEM results presented [21].  Figure 2.9 (a) Absorbance and (b) diffuse reflectance spectra of g-CN photocatalysts  The blue shift associated with washing and protonation is most likely due to the removal of surface and edge amine groups as evidenced by the XPS results and as reported [70, 71, 72].  The relationship between light adsorption and bandgap can be written as: 𝛼ℎ𝑣 = 𝐴൫ℎ𝑣 − 𝐸௚൯௡ Where n is a coefficient associated with an electronic transition, ℎ is the Planch constant, and 𝑣 is the frequency of light. g-CN has been documented to demonstrate an indirect allowed (a) (b) 33  transition, where n=2 [73]. The bandgaps in Table2.2 have been calculated via the Kubelka-Munk method [74],  𝐹(𝑅) =(1 − 𝑅)ଶ2𝑅 Where F(R) is the Kulbelka-Munk function proportional to the extinction coefficient, α, and R is the reflectance. In order to estimate the bandgap of these materials, Figure 2.9(b) is plotted as (𝐹(𝑅) ∗ ℎ𝑣)ଵ/௡ versus ℎ𝑣. The bandgap of each photocatalyst is estimated as the x-intercept of the tangent of the slope and displayed in Table2.2. Across all samples, the harsher the post processing and therefore the smaller the particle size, the larger the bandgap of the photocatalyst, likely due to quantum confinement effects [72]. However, P-g-CN has a smaller shift in bandgap upon protonation, which is the result of the already reduced particle size of P-g-CN as revealed by SEM above.     Table 2.2 The bandgap of each photocatalyst estimated from DRS   Photoluminescence (PL) Spectroscopy  PL spectroscopy was utilized to characterize each photocatalyst across wavelength. The peaks in PL intensity seen in Figure 2.10 correspond well with the bandgaps resulting from the DRS analysis above. P-g-CN and p-P-g-CN show less intense photoluminescence than undoped samples by a substantial margin, likely due to the introduction of mid-gap states as a result of P Material Bandgap [eV] g-CN 2.64 g-CN nitric wash 2.69 P-g-CN 2.65 p-g-CN 2.72 p-P-g-CN 2.67 34  doping and reflecting a longer charge carrier lifetime. Protonation results in a slight increase in PL intensity for p-g-CN and p-P-g-CN samples and is believed to be the result of exfoliation and an increase in the number of surface defects. However, washing and protonation are shown to have opposing effects on PL intensity and by extension charge carrier lifetime. Washed g-CN samples demonstrate a decrease in PL intensity while p-g-CN shows an increase. This disparity demonstrates that the amino functional groups identified by XPS on p-g-CN and washed g-CN samples increase and decrease the amount of charge recombination occurring on the photocatalyst, respectively. This effect can be understood in the context of the previously observed shift in XPS N1s spectra, which indicated a more reduced surface chemistry present in the amino groups of p-g-CN.  Figure 2.10 PL spectra of p-g-CN, p-g-CN/BiVO4, p-g-CN/rGO/BiVO4, p-g-CN/rGO photocatalysts   35   Photocatalyst Band Structure The position of the CB and VB edges of g-CN have been estimated from Mulliken electronegativity theory [75, 76, 77, 78] and confirmed with literature values [40, 75, 79]. The position of the valence band can be estimated via: 𝐸௏஻ = 𝜒 − 𝐸௘ + 0.5𝐸௚  Where 𝐸௏஻ is the valence band edge potential, 𝜒 is the electronegativity of the semiconductor calculated as geometric mean of the constituent atoms’ Mulliken electronegativity,  𝐸௘ is the free energy of electrons (4.5 eV), and  𝐸௚  is the measured bandgap. The electronegativity of g-CN is 4.72 [80, 81, 82, 83, 84]. The position of the conduction band can be calculated given the valence band and the bandgap: 𝐸஼஻ =  𝐸௚ − 𝐸௏஻ The conduction band of g-CN is estimated to be -1.04eV. The position of the fermi level has also been estimated from previously collected data and given in Table2.3 in addition to the band positions for all photocatalysts. However, band structure has been shown to shift for doped, and protonated samples with P doped samples typically having lower conduction band edge and protonated samples a higher conduction band edge as suggested by the fermi levels measured [16, 36, 41].       Table 2.3  The band structure of each g-CN photocatalyst species  Species BG VB vs NHE CB vs NHE fermi level g-CN 2.64 1.60 -1.04 -0.64 g-CN washed 2.68 1.62 -1.06 -0.54 P-g-CN 2.65 1.61 -1.05 -0.69 p-P-g-CN 2.67 1.62 -1.06 -0.55 p-g-CN 2.72 1.64 -1.08 -0.52 36  Figure 2.11 shows the band positions of g-CN and the charge carrier potentials required to perform the reactions of interest. The conduction band of g-CN materials is capable of the hydrogen evolution reaction as well as CO2 reduction while the valance band can perform the oxygen evolution reaction. Shifts in the conduction and valence bands due to protonation and P doping influence the activity of a catalyst for each reaction.    Figure 2.11 A schematic of band structure shown in conjunction with the potential of pertinent reactions   Photocatalytic CO2 Reduction The activity of each catalyst synthesized was evaluated for CO2 reduction in the vapor phase with water as a hole scavenger. The two control experiments were conducted in a black box and 37  in the presence of only N2 did not lead to the formation of any reduced products, demonstrating that the products of further experiments were the result of photocatalytic reduction alone, and not a reflection of photocatalyst decomposition. CO and H2 were the products detected, which were found to evolve linearly in time. The stability of the catalysts was tested with repeated experiments as seen in Figure 2.12, demonstrating reproducibility of the experimental setup and stability of the catalyst across cyclic conditions with minor activity degradation for H2 evolution as widely demonstrated for g-CN materials [41, 42, 85].  Figure 2.12 Cyclic experiments of p-g-CN repeated for 5 hours each with error bars indicating 90% confidence intervals  Figure 2.13 shows the production of each photocatalyst under incident radiation from the solar simulator as described in section 2.1.4. As can be seen there is a large difference in reduction products for g-CN used directly after pyrolysis and g-CN washed with 0.1M nitric acid. Both of 38  these methods have been used as a point of comparison in various publications for both H2 and CO2 reduction [27, 36, 85, 86, 87]. However, the resulting photocatalyst activity for CO2 reduction reduces to just over 25% while H2 reduction increases, revealing a change in selectivity. The activity of g-CN is maintained and slightly improved after protonation into p-g-CN for CO2 reduction and H2, respectively. P-g-CN is has the strongest result for H2 evolution, increasing almost 4x vs g-CN and maintaining much of the CO2 reduction activity observed in g-CN as well. However, protonation of P-g-CN into p-P-gCN results in a 4X decrease in H2 evolution while slightly increasing CO2 reduction.  Figure 2.13 Photocatalytic production of each g-CN photocatalyst with error bars indicating 90% confidence intervals  39   Photocatalytic Mechanism A general photocatalytic mechanism is proposed for the CO and H2 results observed here and shown in Figure 2.11 for p-g-CN. Under simulated sunlight, electrons from the valence band of g-CN are photoexcited into the conduction band, creating electrons with high reduction potential and leaving behind holes with high oxidation potential. It is well documented that the valence band position of g-CN is capable of performing oxygen evolution as seen in equation 8, by splitting water into H+ and O2. These H+ can then combine with 2 electrons from the conduction band to produce hydrogen gas. It is well understood that CO2 can be adsorbed onto surfaces as a result of delocalized 𝜋 bonding, which destabilizes the bond. CO2 can then can be similarly reduced to CO and H2O with the addition of two conduction band electrons and 2 adsorbed H+.   The decreased CO2 photoreduction of nitric washed g-CN compared to unwashed material is understood in the context of this photocatalytic mechanism. Washed g-CN demonstates decreased light adsorption as observed in the spectroscopy results and the π bonding peak shift of the N1s XPS sprectra presented previously. It is likely that washing with weak nitric acid with the intention of cleaning the photocatalyst surface is inducing some exfoliation of g-CN, causing the observed shift in light adsorption due to quantum comfinement effects. The resulting decrease in light adsorption therefore limits the amount of photogenerated electrons and leads to the conduction band electron density necessary for CO2 reduction. Further, it is likely that the removal of amine groups on the surface of g-CN via washing with nitric acid alters the chemical environment of the g-CN surface, decreasing CO2 adsorption and limiting photoreduction [36, 88].   40  This explanation is further supported by the photoreduction activity and chracterization of p-g-CN material. p-g-CN demonstrates evidence of more extensive exfoliation and quantum confinement than nitric washed g-CN with spectroscopy results that measure a larger bandgap and similar shifts in the π bonding peak position while maintaining activity for CO2 reduction and increasing activity for H2 evolution.  However, the protonation of the g-CN heterocycles and edge sites into amine functionalized groups observed via XPS results are the likely explaination for the maintenance of photocatalytic activity despite decreased light absorption [36]. XRD results for both washed g-CN and p-g-CN show decreased peak height reflecting lower crystalinity of the materials compared to g-CN which is due to the exfoliation processes these materials are subject to [72]. However, p-g-CN edges have been shown to be protonated after this delamination occurs, replacing the amine content on the surface of g-CN material and increasing the activity for CO2 reduction through increased adsorption [42]. This allows p-g-CN to show increased overall photoreduction despite less light absorption by maintaining CO2 adsorption characteristics compared to g-CN.  The enhanced photoreduction demonstrated by P-g-CN can be understood in a similar mechanistic context. The XPS spectra shows that the substitution of P for N in the heterocycles of P-g-CN results in n-type doping [42]. The effect of P doping is an enhancment of  P-g-CN conductivity compared to g-CN and increased charge carrier lifetime while increasing charge transfer efficiency as evidenced by greatly reduced PL spectra for P-g-CN and p-P-g-CN [42, 36, 89]. These material improvements have been attributed to the creation of mid-band gap states that are able to attract and hold charges, stifling recombination [35]. The addition of P to the pyrolysis synthesis reaction of g-CN has also been shown to limit the polycondensation by virtue 41  of the longer P-N bond length as evidenced by reduced synthesis yield for P-g-CN[59]. This results in smaller particle size, more amorphous material with more specific suface area and amine rich edge sites [59, 21]. These characteristics lead to P-g-CN having a slightly smaller bandgap than g-CN with the majority of the bandgap shift coming from an increase in the valence band edge as previously descussed [18, 36]. The characteristics attributed to the doping of P in g-CN increase the number and duration of charge carriers on the surface of the material, supporting the enhanced P-g-CN result for CO2 reduction.   Upon the protonation of P-g-CN, the evolution of H2 plummets and CO2 reduction activity increases only marginally. Substantial positive peak shifts in the N1s XPS spectra attributed to the C-N-H and N-(C)3 peaks point to changes in the surface chemistry of p-P-gCN that may explain this dramatic loss in activity. The shift in peak position without a relative change in peak intensity indicates protonation of P-g-CN heterocycles and also the removal of surface amine groups in a similar fashion to p-g-CN as evidenced by their very similar N1s XPS spectra shifts. The protonation of P-g-CN appears to result in the negation of the positive effects of P doping as seen in the loss of photocatalyst activity despite retention of the P doping content as evidenced by XPS and EDX spectra. It is hypothesized that the additional charge mobility and enhancement of the band structure resulting from the additional valence electron of P when integrated into the heterocycles of P-g-CN is inhibited by the removal of surface protonation and the addition of N-H bonds in the heterocycles of p-P-g-CN [36]. The light absoption of p-P-g-CN is approximately the same before and after protonation, with only a small difference in band gap detected, indicating that the chemical environment created by intercalated amine groups after pyrolysis is more influential to activity than band structure[36, 88, 89]. 42   In summary, the amine content of g-CN materials is highly influential to the CO2 reduction and water splitting activity they can exhibit. Washed g-CN material CO2 reduction activity shows a 4x decrease when compared to g-CN, which is attributed to decrease CO2 adsorption driven by surface and intercalated amine groups as evidenced by XPS. Protonation restores water splitting and CO2 adsorption through the addition of N-H groups to the heterocycles of p-g-CN. Shifts in the N1s XPS peaks indicate that the surface amine groups removed, and the amine groups added to the heterocycles of p-g-CN result in a different chemical environment of the g-CN surfaces as evidenced by activity changes. Decreases in H2 evolution indicate that removal of surface amines in p-P-g-CN affects the charge transfer efficiency of P-g-CN. Control of surface amine content via doping and protonation has been shown to alter the selectivity, charge transfer efficiency and activity of g-CN materials. As a result, vapor phase CO2 reduction and water splitting offers the advantage of controlling surface and intercalated amine content and therefore selectivity compared to the liquid phase.  43  Chapter 3: 2D Electrostatically Self-Assembled g-CN Composite Photocatalysts  Here a self-assembled 2D-2D z-scheme photocatalyst composites between protonated amine functionalized p-g-CN and monoclinic BiVO4 nanosheets was synthesized via a facile sonication assisted method. The catalytic activity of various mass ratios of p-g-CN and BiVO4 were investigated via vapor phase CO2 reduction and hydrogen evolution under visible light (λ ≥ 420 nm). Based upon catalyst characterization methods, a z-scheme photocatalytic mechanism was proposed as the source of improved photosystem performance.  3.1 Experimental  Chemicals and Materials Ammonium vanadate (NH4VO3, ≥ 99.0%), bismuth nitrate pentahydrate (Bi(NO3)3·5H2O, ≥99.999%), urea (CO(NH2)2, ≥99.0%), 2-propanol ((CH3)2CHOH), >99.5%), 1-octadecene (C18H36, 90%), Oleic acid (C18H34O2, 90%), oleyl amine (C18H37N, 70%), hexane (C6H18, ≥98.5%), Ethanol (CH3CH2OH, ≥99.8%) were obtained from Fischer Scientific. Hydrochloric acid (HCl, 37%) and nitric acid (HNO3, 70%) were obtained from Sigma Aldrich. Chemically reduced graphene oxide powder (C7.7-8.7O2.2-1.3) was obtained from Graphenea. Carbon dioxide (CO2, 99.995%) and nitrogen (N2, 99.999%) was obtained from Praxair. All the reagents in this work were at the analytical reagent grade and used without further purification. Deionized water was used throughout this study. The water was purified with a Milli-Q water ion-exchange system.  44   Photocatalyst Preparation 3.1.2.1 p-g-CN p-g-CN was synthesized by the pyrolysis of urea. 20 g of urea was dried in an oven for 24 hours at 80 oC. It was then ground to a fine powder in a ball mill and placed in a 50mL crucible with a lid. The crucible was heated in an air atmosphere to 520oC at a ramp rate of 10 oC/min, held at 520 oC for 2 hours, then cooled to room temperature. The product was stirred for one hour in 1M HCl and then washed with distilled water in a centrifuge until neutral. The resulting material was returned to the oven to dry for 8 hours at 80 oC.  3.1.2.2 BiVO4 Nanosheets BiVO4 nanosheets were synthesized as published elsewhere [65]. Briefly, a flask with a thermocouple, N2 gas line, septum, stir rod, and a reflux condenser were prepared over a hot plate with magnetic stirring. 1.2g of Bi(NO3)3*5H2O, 5mL of 90% Oleic acid, 5mL of 70% Oley amine and 50mL of 90% octadecene were added to the flask. N2 was sparged to remove all O2 while stirring and then heating to 170 oC until all Bi(NO3)3 was dissolved. Temperature was then reduced to 130oC. In a separate flask, 0.6g of NH4VO3 and 10mL of HNO3 was added to 50mL of distilled water and stirred until dissolved. The NH4VO3 was injected through the septum of the flask containing the Bi(NO3)3 solution and then refluxed at 100 oC for 40 minutes. The solution was quickly cooled to room temperature and then stirring and sparging was stopped. A 2:1 mixture of hexane to ethanol was used to wash the resulting product via centrifuge and dried at 80oC in an oven for 8 hours.  45  3.1.2.3 Photocatalyst Composite Preparation The self-assembly of all photocatalyst composites was accomplished through an ultrasonication assisted electrostatic interaction between materials [16]. For the g-CN/rGO/BiVO4 composite, 150 mg of p-g-CN was ultrasonicated for 30 minutes to exfoliate thin sheets of material. 3.75 mg of rGO was then added under stirring for 30 minutes followed by sonication for an additional 30 minutes. BiVO4 in varying amounts was similarly added under stirring for 30 minutes, followed by sonication for 30 minutes. The stepwise addition of materials allowed the layering of individual materials through electrostatic self-assembly. For the p-g-CN/rGO and p-g-CN/BiVO4 composites, BiVO4 and rGO were omitted from the procedure, respectively.   Characterization Methods The crystal structure of photocatalyst materials were analyzed by a powder x-ray diffractometer (Rigaku Ultima IV) with Cu-kα radiation and a scan rate of 0.03/s with a 2θ range of 5o to 80o. The structural morphology, elemental composition and distribution was observed via field emission scanning electron microscope with energy dispersive spectroscopy system (FE-SEM, Sigma with Oxford detector). X-ray photo-electron spectrometer (Kratos Axis Ultra) with Al kα radiation. The optical absorption of photocatalysts were observed via diffuse reflectance spectrometer (Ocean Optics FLAME-S-XR1-ES). Photoluminescence spectroscopy was performed with a Cary Eclipse fluorescence spectrometer. A Sciencetech solar simulator (SS1kW) operating with a 1000W Xe arc lamp was used as a light source. . 46   Evaluation of Photocatalytic Activity To compare the performance of the photocatalysts synthesized, experiments were conducted in  the same experimental apparatus and conditions presented in chapter 2 with the only difference being the photocatalyst materials used.   3.2 Results and Discussion  X-ray Diffraction (XRD)  The XRD spectra for the synthesized catalysts and their composites are shown in Figure 3.1. The spectra for p-g-CN shows two diffraction peaks at the (100) and (002) planes and matching g-C3N4 (JCPDS: 87-1526) with a slight shift towards lower 2θ values for the (002) peak [65, 67]. The p-g-CN sample’s interlayer stacking of conjugated aromatic systems is evidenced by the (002) diffraction peak at 27.25o corresponding to a graphitic stacking spacing of 0.327nm, while the (100) peak at 13.1o reflects the in-plane repeating of tri-s-triazine of g-CN materials with a d spacing of 0.675nm [59, 65]. The spectra for monoclinic scheelite BiVO4 (JCPDS 14-0688) matches that of the synthesized BiVO4 without any additional peaks or impurities indicated [50] 47  . Figure 3.1 XRD spectra of p-g-CN, BiVO4 and p-g-CN composites  Monoclinic scheelite BiVO4 is differentiated from tetragonal scheelite BiVO4 by the presence of a minor peak where 2θ is 15.1o [13, 50].  The characteristic peaks at 2θ values of 28.8o, 30.6o, 42.4o and 53.2o correspond to the (112), (040), (051) and (161) planes, respectively [7, 50, 59]. Upon integration of the Z-scheme materials, all peaks of p-g-CN and BiVO4 are present, indicating that both materials are in the composite. The intensity of  the (040), (051) and (161) planes disproportionately decreases in the composite materials and both the (112) and (040) peaks shifts down from 2θ values of 28.9o to 28.8o and 30.6 to 30.5, respectively, which may indicate integration with g-CN [61]. The intensity of the (002) peak of p-g-CN also decreases markedly as has been previously observed and may also indicate composite integration [62]. The presence of rGO is likely not indicated because of its low wt% and low crystallinity [53, 54]. 48   Scanning Electron Microscopy (SEM) and Energy Dispersive X-ray Spectroscopy (EDX) The morphology of the photocatalysts and their composites were also analyzed by SEM and EDX shown in Figure 3.2a-e. As seen in Figure 3.2a, p-g-CN exhibits a lamellar sheet structure as expected of g-CN materials [18, 19]. As can be seen in Figure 3.2b, BiVO4 is predominantly nanosheets, shown here to have curled onto themselves to reduce surface charges as well as nanoplates [65]. Interestingly, when BiVO4 is added to composite catalysts, the nanosheets unfurl and conform to the p-g-CN material as seen in Figure 3.2d-e. A similar phenomenon is seen in rGO containing composites due to the same electrostatic attraction, resulting in large surface area interfacial contact between materials [16, 56]. This behavior is an expected prerequisite for heterojunction composites [44]. Figure 3.2e shows this interaction of p-g-CN/rGO/BiVO4 composites with BiVO4 and rGO sheet materials draped over p-g-CN.       49     Figure 3.2 SEM imagery of (a) p-g-CN (b) BiVO4 (c) p-g-CN/rGO (d) p-g-CN/BiVO4 and (e) p-g-CN/rGO/BiVO4 (a) (c) (d) (b) (e) 50  EDX data shows the presence of C, N, O, Bi, and V in Figure 3.3a-e. The results show that the g-CN material contains C, N, and trace amounts of O and the BiVO4 sample contains Bi, V, O as well as some C in Figure 3.3a and 3.3b, respectively. The p-g-CN/rGO sample is marked by the increase of the C/N content ratio in the EDX spectra due to the addition of rGO, but as a result of the low weight percentage, thin sheets and uniformity of distribution, rGO does not show clusters of material. In the p-g-CN/BiVO4 and p-g-CN/rGO/BiVO4 composites, the presence of BiVO4 nanoplates is demarcated strongly in the EDX spectra, but similarly to the rGO sheets the nanosheets of BiVO4 appear only in the relatively weak uniform distribution of Bi and V. This even distribution of rGO and BiVO4 nanosheets is evidence of the electrostatic interaction of the thin sheets on the surface of p-g-CN and the resulting strong interfacial contact.        51   Figure 3.3 EDX elemental mapping of (a) p-g-CN, (b) BiVO4, (c) p-g-CN/rGO, (d) p-g-CN/BiVO4, (e) p-g-CN/rGO/BiVO4 (c) (d) (e) (a) (b) 52    X-ray Photoelectron Spectroscopy (XPS) The XPS survey spectra in Figure 3.4 reveals that all g-CN materials synthesized are composed of C, N and a small amount of O, while BiVO4 samples show the presence of Bi, V, O and C. Elemental ratios and stoichiometry of each catalyst is seen in Table3.1. The theoretic ratio of C:N in an ideal g-CN material is 1:1.33.  Protonated g-CN shows less N after protonation, which may be due to a decreased number of edge N sites after acid treatment while protonating the sp2 hybridized N [16, 43]. The atomic ratio of BiVO4 nanosheets approaches stoichiometry as determined by XPS analysis.  Figure 3.4 The XPS survey scan spectra of p-g-CN, BiVO4, and p-g-CN composites    53       Table 3.1 Atomic ratios of p-g-CN, BiVO4, and theoretic materials   High resolution scans of the C1s have been deconvoluted into three peaks for p-g-CN which are at 284.8 eV, 288.1 eV, and 285.8 eV corresponding to adventitious carbon (C-C), hybridized carbon in N containing aromatic rings (N=C-N), and ammonia species (C-NH2), respectively [56, 67]. The large C1s peak for BiVO4 at 284.8eV can be solely attributed to adventitious carbon (C-C) and was likely introduced during sample preparation. In p-g-CN/rGO, p-g-CN/rGO/BiVO4, and p-g-CN/BiVO4 composites, the peak attributed to sp2 hybridized carbon shifts +0.4eV, +0.3eV and -0.1eV, respectively. The upward shift of composites containing rGO indicate a strong π-π stacking interaction but the effect of BiVO4 z-scheme heterojunction interaction is also present [46, 56]. The C-NH groups shift p-g-CN/rGO, p-g-CN/BiVO4 and p-g-CN/rGO/BiVO4 down 0.4 eV, 0.3 eV and 0.5eV, respectively, indicating increased electron density as a result of z-scheme and π-π bond stacking interactions [2, 68, 90].  The N1s spectra seen in Figure 3.5(b) can be deconvoluted into four peaks for each g-CN material. For p-g-CN the peaks are at 398.5eV, 399.3eV, 400.7eV and 404.1 eV which have been attributed to sp2 hybridized nitrogen (C-N=C) in tri-s-triazine rings, tertiary nitrogen (N-(C)3), free amine groups (C-N-H), and inter-layer π excitations, respectively [56, 67]. In all composite samples, the N1s peaks all shift upwards, which indicates a chemical environment shift due to Sample description C:N/BI:V:O atomic ratio Stoichiometric ratio Theoretic g-C3N4 1:1.33 C3N4 p-g-CN 1:1.19 C3N3.58 Theoretic BiVO4 1:1:4 BiVO4 BiVO4 1:1.03:3.74 Bi1V1.03O3.71 54  interaction with BiVO4 in a Z-scheme heterostructure [68, 90]. Specifically, in p-g-CN/rGO, p-g-CN/BiVO4 and p-g-CN/rGO/BiVO4 composites, the peak attributed to sp2 hybridized nitrogen shifts 0.5eV, 0.1eV and 0.5eV, respectively. This large shift is due to the π-π stacking interaction of p-g-CN and rGO as well as the z-scheme interaction [2]  Figure 3.5 XPS spectra of (a) the C 1s and (b) the N 1s orbitals for p-g-CN and p-g-CN composites (a) (b) 55   Figure 3.6 XPS spectra of the (a) V 2p and (b) Bi4f orbitals  The Bi4f and the V2p spectra of BiVO4 both reveal two peaks as seen in Figure 3.6. The Bi4f peaks at 159.0 eV and 164.3 eV correspond to Bi3+ spin states, Bi4f7/2 and the Bi4f5/2 and the V2p peaks at 524.1 eV and 516.6 eV are attributed to V2p1/2 and the V2p3/2 spin states of V5+, respectively [68, 90]. These peaks uniformly shift up in binding energy upon the creation of composite materials, indicating z-scheme composite interaction between BiVO4 and p-g-CN [68, 90].   (a) (b) 56   Figure 3.7 XPS spectra of the O1s orbital  All samples show peaks in the high resolution O1s spectra as seen in Figure 3.7. The peak at 532.1 eV corresponds to surface absorbed H2O and doesn’t shift in position among samples, whereas the peak at 529.7eV attributed to lattice O in the BiVO4 crystal structure shifts up in binding energy in composite samples [2]. The presence of surface adsorbed H2O evident here explains the trace O signal observed in the EDX spectra of p-g-CN above.  Both p-g-CN/rGO and p-g-CN/rGO/BiVO4 samples show peaks at 533.0 eV and 532.9 eV, respectively, which are attributed to C=O species of rGO [40].  57   UV-Visible Diffuse Reflectance Spectroscopy (DRS) Photocatalyst optical characterization was performed via UV-vis diffuse reflectance spectroscopy (DRS) to measure the absorption and utilization of light incident photocatalyst across wavelength. Figure 3.8(a) shows the absorption spectra of each photocatalyst and their composites, the absorbance of the composite catalysts compared to the individual catalysts is enhanced, with greatest absorbance occurring in rGO containing catalysts. Increased absorbance of rGO composites is a result of a red shift related to the π-π bonding and subsequent electron storage and localization on rGO [15, 53]. Heterojunction composites also show a redshift and a bimodal absorption due to the smaller bandgap of BiVO4 nanosheets which is shown to absorb most strongly over the visible range.   Figure 3.8 (a) Absorbance and (b) diffuse reflectance spectra of each photocatalyst and composite          (a) (b) 58    Material Bandgap [eV] BiVO4 2.30 p-g-CN 2.72 p-g-CN/rGO 2.65 p-g-CN/BiVO4 2.63 p-g-CN/rGO/BiVO4 2.56  Table 3.2 The bandgap of each photocatalyst and composite  The relationship between light adsorption and bandgap can be written as: 𝛼ℎ𝑣 = 𝐴൫ℎ𝑣 − 𝐸௚൯௡ Where n is a coefficient associated with an electronic transition, ℎ is the Planch constant, and 𝑣 is the frequency of light. BiVO4 and g-CN have been documented to demonstrate an indirect allowed transition, where n=2 [73, 76]. The bandgaps in Table3.2 have been calculated via the Kulbelka-Munk method [74],  𝐹(𝑅) =(1 − 𝑅)ଶ2𝑅 Where F(R) is the Kulbelka-Munk function proportional to the extinction coefficient, α, and R is the reflectance. To estimate the bandgap of these materials, Figure 3.8b is plotted as (𝐹(𝑅) ∗ ℎ𝑣)ଵ/௡ versus ℎ𝑣. The bandgap of p-g-CN and BiVO4 are estimated as the x-intercept of the tangent of the slope, 2.72 eV and 2.30 eV, respectively. All composite bandgaps are listed in Table3.2 were evaluated considering n=2 as g-CN is the primary photocatalyst material. Red shifts in BiVO4 spectra deviating from the widely accepted bandgap of 2.4 eV have been attributed to oxygen vacancies [65]. The XPS and EDX spectra show that the BiVO4 nanosheets 59  synthesized here are stoichiometrically deficient in lattice oxygen, which likely explains the narrowed bandgap.   Photoluminescence (PL) Spectroscopy   PL spectroscopy was utilized to characterize each photocatalyst composite across wavelength. As can be seen in Figure 3.9, p-g-CN has by far the most intense PL spectra and the peak of the p-g-CN spectra corresponds well with the bandgap results of the DRS previously performed. Compositing p-gCN and BiVO4 results in more than halving the PL intensity, suggesting that the combination results in a reduction in charge recombination and an increase in charge carrier lifetime. The greatest reduction in PL intensity is shown for the p-g-CN/rGO composite, demonstrating how efficiently rGO is able to reduce recombination as a result of the π-π stacking interaction between these 2D materials. The p-g-CN/rGO composite also shows a bimodal PL  Figure 3.9 PL spectra of p-g-CN, p-g-CN/BiVO4, p-g-CN/rGO/BiVO4, p-g-CN/rGO photocatalysts  60   spectra with an additional shoulder redshifted. This redshifted peak is likely recombination of charge carriers from the rGO surface to the VB of p-g-CN. Finally, the p-g-CN/rGO/BiVO4 composite is shown to demonstrate further PL reduction upon the addition of rGO as is expected from the drastic reduction demonstrated for the g-CN/rGO composite. However, the effect of rGO and BiVO4 compositing is not synergistic and does not result in the lowest intensity.    Photocatalyst Band Structure The position of the CB and VB edges of BiVO4 and g-CN have been estimated from Mulliken electronegativity theory [77, 78] and confirmed with literature values [39,79]. The position of the valence band can be estimated via: 𝐸௏஻ = 𝜒 − 𝐸௘ + 0.5𝐸௚  Where 𝐸௏஻ is the valence band edge potential, 𝜒 is the electronegativity of the semiconductor calculated as geometric mean of the constituent atoms’ Mulliken electronegativity,  𝐸௘ is the free energy of electrons (4.5 eV), and  𝐸௚  is the measured bandgap. The electronegativity of g-C3N4 and BiVO4 are 4.72 and 6.04, respectively, which when utilized here corresponds to a valence band of 1.64 eV for g-C3N4 and 2.76 eV for BiVO4 vs NHE. Given the valence band position, the conduction band can be deduced: 𝐸஼஻ =  𝐸௚ − 𝐸௏஻ The conduction band is estimated to be -1.08 eV and 0.44 eV for p-g-CN and BiVO4, respectively. The fermi level of each photocatalyst has also been estimated from the previously collected VB-XPS analysis and shown in graphic of Figure 3.10 and Table3.3. The conduction 61  band, valence band and fermi levels estimated here correlate well with existing literature sources for these materials [2, 65, 91].      Table 3.3 The band edges of each photocatalyst and their composites.   Figure 3.10 The band structure of p-g-CN and BiVO4 shown graphically   As can be seen in these figures, only the conduction band of p-g-CN has an appropriate band energy to facilitate CO2 reduction to CO and hydrogen evolution [90, 91] and therefore Species BG VB vs NHE CB vs NHE fermi level BiVO4 nanosheets 2.29 2.76 0.44 1.21 p-g-CN 2.72 1.64 -1.08 -0.52 p-g-CN/rGO/BiVO4 2.65 1.61 -1.05 -0.82 p-g-CN/BiVO4 2.63 1.60 -1.04 -0.82 p-g-CN/rGO 2.56 1.56 -1.00 -0.98 62  alterations to CO and H2 production must be a result of changes to the conduction band of p-g-CN or factors that affect the chemistry at the conduction band of p-g-CN.   Photocatalytic CO2 Reduction The photocatalytic performance of each composite catalyst in was investigated via vapor phase CO2 reduction with water vapor acting as a hole scavenger. Control experiments conducted in an atmosphere of pure N2 did not result in the detection of any reduced products and dark experiments conducted in a black box detected only trace amounts of reduced products. Together, these two results demonstrate that all products were the result of photocatalytic reaction and not photocatalyst decomposition. Detected products for each material composite included H2 and CO, which increased linearly in time. The production rates across various composite types is shown in Figure 3.11.   Figure 3.11 The yield of p-g-CN and investigated composites with error bars indicating 90% confidence intervals 63  Selectivity differences are observed between the different catalyst composites with p-g-CN producing 2.03 μmol/gcat/hr CO and 7.24 μmol/gcat/hr H2 while the addition of rGO in p-g-CN/rGO elevated the production to 2.64 μmol/gcat/hr CO and 13.23 μmol/gcat/hr H2. The 83% increase in H2 production and the 30% increase in CO production upon the addition of rGO is likely due to an increase in conductivity and mobility across the highly conductive rGO surface, which is in strong interfacial contact with p-g-CN [16]. It has also been shown that the COOH groups present on rGO can act as centers of charge carrier localization, inhibiting recombination and acting as a cocatalyst [46]. However, charge carriers become less reducing when transferred to the surface of rGO, decreasing the overpotential driving the CO2 reduction reaction to a greater extent than the HER and demonstrating a selectivity change in p-g-CN/rGO as a result [47].   Figure 3.12 The yield of the weight ratios of p-g-CN/BiVO4 with error bars indicating 90% confidence intervals 64  Various mass ratios of the p-g-CN/BiVO4 z-scheme composites were created and tested for photocatalytic CO2 reduction as seen in figure 3.12. An optimum ratio emphasizing the effect of the effect of the z-scheme heterojunction was found at a ratio of 150 mg p-g-CN to 25mg BiVO4. Increasing amounts of BiVO4 added after this point only briefly increased hydrogen evolution before inhibiting CO2 reduction and hydrogen evolution. Since BiVO4 does not have a conduction band potential high enough to reduce CO2 or participate in the HER (see Figure 3.10), all CO2 reduction and hydrogen evolution must occur on the surface of p-g-CN [59]. The 150/25 p-g-CN/BiVO4 z-scheme ratio increased the rate of CO2 reduction 107% to 4.23 μmol/gcat/hr CO and the rate of water splitting to 9.89 μmol/gcat/hr H2. The optimum ratio of photocatalyst z-scheme components is likely the result of increasing optical thickness of the deposited photocatalyst layer and shadowing by BiVO4, ultimately restricting the amount of light available to p-g-CN. This creates an apparent change in selectivity as BiVO4 absorbs a greater percentage of incident photons. This improvement in CO2 reduction and HER is evidence of an electrostatically self-assembled heterojunction interface between BiVO4 and p-g-CN.  The creation of an intermediate z-scheme via the addition of BiVO4 to the previously tested p-g-CN/rGO composite underperformed the p-g-CN/rGO composite, producing 2.40 μmol/gcat/hr and 5.84 μmol/gcat/hr  of CO and H2, respectively. The poor result of the p-g-CN/rGO/BiVO4 structure is likely due to the formation of a typical type II heterojunction facilitated by the increased charge mobility and intermediate charge states introduced by rGO as well as the combined shadowing effects from BiVO4 and rGO [40, 92, 93]. The layering of the rGO and BiVO4 sheets on the surface of p-g-CN may have also limited the mass transport at the surface.  65   Proposed Charged Transfer Mechanism As a result of the band relative band positions described above, the coupling of p-g-CN and BiVO4 materials must initiate a type II heterojunction in either the form of a classical type II heterojunction or a z-scheme where electrons either flow from the conduction band of p-g-CN to the conduction band of BiVO4 or from the conduction band of BiVO4 to the valence band of p-g-CN, respectively. The exact charge transfer mechanism is dependent on the interfacial contact between the materials and the position of their bands and fermi levels [94]. Figure 3.13 shows the band structures of p-g-CN and BiVO4 before contact. The increased CO production in the direct p-g-CN/BiVO4 composite suggests that the electron transfer mechanism taking place is most probably a z-scheme. Photogenerated elections in the BiVO4 conduction band would fill holes generated in the valance band of p-g-CN, inhibiting charge recombination and resulting in charge accumulation in the conduction band of p-g-CN to participate in multi-electron transfer CO2 reduction and hydrogen evolution reactions [19]. Extremely oxidative holes in the valence band of BiVO4 are then consumed by OER. This increase in electron density in the conduction band of p-g-CN positively correlates with an increase in CO and H2 production. The reduced PL intensity of the p-g-CN/BiVO4 composite reflecting reduced charge recombination further supports this charge transfer mechanism. 66   Figure 3.13 The band structure of the p-g-CN/BiVO4 composite with a proposed charge transfer mechanism    Typically, the relatively higher fermi level of p-g-CN would drive electrons from the conduction band of p-g-CN to the conduction band of BiVO4 until their fermi levels reached equilibrium at the interface. However, the resulting electron transfer mechanism would not support the increase in CO2 reduction and hydrogen evolution observed. Although some research has suggested that p- type BiVO4 would result in the creation of a type II heterojunction where electrons transition from the conduction band of BiVO4 to the conduction band of p-g-CN and show a similar increase in conduction band electron density, the conduction band potential of n-type monoclinic BiVO4 synthesized here is not expected to exceed that of the p-g-CN CB [40, 59, 60].  67  In the case of the p-g-CN/rGO composite, charge transfer is most likely occurring from the p-g-CN surface to rGO as a result of the higher work function and high conductivity of rGO [92]. Electrons can then accumulate in the oxygen containing localization sites to more rapidly drive multi-electron transfer CO2 reduction and hydrogen evolution reactions [92, 93]. The observed selectivity for water splitting is likely due to the decreased reduction potential for CO generation, but the still substantial overpotential for water splitting as observed in the band structure presented here. This hypothesis is supported by the greatly reduced, redshifted, bimodal PL spectra of the p-g-CN/rGO composite, suggesting that the addition of rGO stifles charge recombination, but results in less reduced photogenerated electrons on the rGO surface. In other words, the mechanism that enhances charge separation and mobility may also decrease the reduction potential of photogenerated electrons resulting in reduced CO2 reduction and enhanced H2 evolution.   The creation of a p-g-CN/rGO/BiVO4 composite resulted in only very small overall changes to the production of p-g-CN, suggesting that charge transfer in an indirect z-scheme mechanism is being partially halted in the transfer from rGO to BiVO4. The fermi level of rGO and the interfacial contact with p-g-CN and BiVO4 photocatalysts are instrumental in determining charge transfer in this intermediate z-scheme [94]. rGO may be providing intermediate charge states between the conduction band of p-g-CN and the conduction band of BiVO4, facilitating charge transfer in a type two heterojunction mode that then can further transfer in a z-scheme mode. This mixed type II heterojunction and z-scheme state may increase charge carrier lifetime in the conduction band of p-g-CN but result in relatively lower charge density compared to the direct p-g-CN/BiVO4 z-scheme. This hypothesis is supported by the decreased but not synergistically 68  reduced PL intensity of the three component p-g-CN/rGO/BiVO4 composite when compared to the p-g-CN/BiVO4 direct z-scheme. The proposed charge transfer mechanism for p-g-CN/rGO/BiVO4 is shown in Figure 3.14.  Figure 3.14 The band structure of p-g-CN/rGO/BiVO4 with a proposed charge transfer mechanism  In summation, experiments were conducted to investigate vapor phase CO2 reduction of direct and indirect p-g-CN/BiVO4 composite catalysts at various mass ratios. The optimal p-g-CN/BiVO4 ratio was found to exhibit more than twice the photocatalytic CO2 reduction and the addition of rGO resulted in  >100% improvement in CO2 reduction and a nearly 40% increase in H2 evolution compared to bare p-g-CN under visible light (λ>420 nm). (high stability during cycling tests?) The enhancement of the composite photocatalyst is attributed to improved charge 69  separation and increased charge carrier lifetime as a result of an electrostatically self-assembled type II z-scheme heterojunction between p-g-CN and BiVO4 photocatalysts. A z-scheme charge transfer mode was investigated via photocatalyst band structure and a potential mechanism is proposed and supported by PL spectroscopy results. In summary, this research elucidates the improvement possible as a result of direct z-scheme heterostructures with p-g-CN and BiVO4 photocatalysts and the role of the fermi level of rGO in indirect heterostructure composites. 70  Chapter 4: Conclusions and Recommendations 4.1 Conclusions The goal of this research was to study: x Effect of photocatalyst post synthesis preparation on the morphology, surface chemistry, crystallinity, optical absorption, and ultimately photocatalytic performance of the obtained photocatalyst x Effect of phosphorous doping which may affect the parameters above, as well as providing centers of charge accumulation due to the introduction of mid-gap states, potentially increasing the density of surface charges. x Effect of g-CN/BiVO4 ratios on the g-CN/BiVO4 z-scheme activity for CO2 reduction and hydrogen evolution. x Effect of rGO introduction as an electron mediator in a g-CN/rGO/BiVO4 z-scheme  This section discusses the contribution of this work to current research in the field, conclusions arising from the study’s findings, the strengths and limitations of the conducted research.   The Effect of Post Synthesis Preparation Techniques Protonation and washing techniques have been used in literature to prepare g-CN catalysts materials for further enhancement and as a basis for comparison despite the large differences in their yields for CO2 reduction and H2 evolution in the vapor phase. Further, inconsistent yet effective results from nontreated g-CN material have also been used as a basis of comparison for CO2 reduction. XRD and XPS characterization techniques showed that washing and protonation both exfoliated the photocatalyst as well as changed the amine functionalization of the surface 71  compared with untreated material. The differences in CO2 reduction among these catalysts have been primarily attributed to differences in CO2 adsorption as a result of amine functionalization whereas the differences in H2 evolution appear to be primarily driven by differences in photocatalyst surface area and particle size.  The amines present on untreated g-CN photocatalyst surfaces are unstable, yielding inconsistent results depending on catalyst preparation techniques and solvents utilized. The amine functionalized surface of p-g-CN is much more stable but consist of different amine functional groups in the heterocycles of g-CN as evidenced by XPS peak shifts of the N1s spectra. These observations were more obvious due to experimentation in the vapor phase, where the transient amine groups of untreated g-CN were able to enhance CO2 adsorption. Further, the vapor phase allows increases in CO2 reactant partial pressure without resorting to the basic carbonate chemistry of liquid phase experimentation that may stifle the acidic surface amine sites on g-CN. This research demonstrates that amine functionalization of the g-CN surface is a pathway towards improved CO2 reduction for g-CN materials by virtue of enhanced CO2 adsorption. Future works should consider catalyst treatments’ effect on surface amine groups when attributing improvements of various photocatalyst enhancement techniques. Amine functionalization is discussed further in the recommendations section.    Phosphorous Doping in g-CN Catalysts The creation of centers of charge accumulation has been of general interest for photocatalysts participating in multielectron transfer reactions, but particularly for g-CN materials due to their high tendency toward charge recombination. The phosphorous doping of g-CN has been shown 72  to create midgap states that accumulate charge carriers and facilitate multi-electron transfer reactions. Further, the incorporation of P inhibits condensation of g-CN as evidenced by lower synthesis yields and more amorphous material as seen in XRD spectra. Amorphous phosphorous doped g-CN material is shown to have a high number of surface amine groups, as evidenced by the XPS results, that have been shown by this research to be instrumental in the adsorption of CO2 prior to CO2 reduction. Other works have also noted large increases in the conductivity and upward band edge shifts of P-g-CN as a result of doping, making P-g-CN a very interesting starting material to build improved g-CN catalysts. However, the transient amine groups on the P-g-CN surface result in inconsistent results for CO2 reduction experimentation with P-g-CN, showing high selectivity for H2 evolution.  As in earlier work, protonation of P-g-CN aimed at stabilizing the surface amine groups to retain and enhance CO2 adsorption characteristics and shift selectivity toward CO2 reduction. XPS results indicate that protonation occurs in the heterocycles of p-P-g-CN at both N and P substitutions sites as evidenced by the XPS spectra shifts of p-P-g-CN. However, the protonation of P-g-CN leads to only slightly enhanced CO2 and a collapse of H2 production. It is hypothesized that the protonation of P inhibits the formation of charge accumulation centers and stifles charge transport of p-P-g-CN as the extra valence electron of P is engaged in covalent bonding. As a result, the self-assembly applications for p-P-g-CN in CO2 reduction photocatalysts is not as promising as originally theorized. Future works should consider alternative amine functionalization methods of P-g-CN as detailed in the recommendations section.  73   The Effect of g-CN/BiOV4 Z-Scheme Mass Ratios on Photocatalyst Activity p-g-CN has been shown in literature to be a viable photocatalyst for electrostatic self-assembly mechanisms with stable surface chemistry and production, but still suffers from high levels of charge recombination. SEM and EDX imagery show that a self-assembled 2D-2D direct z-scheme between g-CN and BiVO4 is a promising method to inhibit charge recombination in a more rapid and facile approach than other reported calcination and hydrothermal methods. The addition of very small amount of BiVO4 as a result of the 2D nanosheet structure leads to the doubling of p-g-CN production for enhanced organic semiconductor applications without the costs associated with precious metal use. The demonstrated improvement shows that the intimate interfacial contact necessary for direct z-scheme charge transfer is possible through electrostatic interaction alone.  The removal of holes in the g-CN band structure by filling them with photogenerated electrons from BiVO4 offers improved holes to facilitate oxygen evolution and longer charge carrier lifetimes in the CB of p-g-CN, but this strategy is limited in generating higher charge densities in on the p-g-CN surface due to shadowing effects. Literature shows that in composite photocatalyst schemes there is often an optimal ratio between materials as a compromise between the improvement characteristics and shadowing effects on the photoactive materials and this has also been shown for this 2D-2D p-g-CN/BiVO4 z-scheme.   Difficulty in establishing CB and VB edge positions due to the hydrophilicity and low conductivity of p-g-CN, inhibiting Mott-Schottky experiments and UPS spectra, limits our 74  understanding of the charge transfer mechanism of the as synthesized material presented here. However, theoretical calculations backed by the characterizations presented in the literature indicate that the most likely charge transfer mechanism responsible for enhancement is a direct z-scheme. Further research is suggested in the recommendations section.   The Performance of rGO as a Z-scheme Mediator in g-CN/BiVO4 Z- schemes Literature widely debates the effectiveness of mediated z-schemes vs direct z-schemes in their overall effectiveness for visible light photocatalysis. Shadowing effects must be balanced with the ease of charge transfer and the tunability of a finite number of photocatalyst band structures capable of visible light photocatalysis in achieving effective photoreduction. Here, the introduction of rGO into a p-g-CN/BiVO4 z-scheme to achieve a 2D-2D-2D composite did not result in an overall increase in photoreduction. The photocatalyst characterization methods provided herein, calculations, and literature values for band structure suggest that the fermi level of rGO facilitated a type II heterojunction charge transfer mechanism rather than the z-scheme observed in the direct material composite. However, Mulliken band structure calculations have limited ability to explore the effects of protonation and 2D nanosheet morphologies of photocatalyst materials on photocatalyst band structures and do not yield a full understanding of charge transfer mechanism possibilities as compared to other effects such as multicomponent shadowing.  75  4.2 Recommendations and Future Directions In order to more fully understand the phenomena contributing to production changes of the various catalysts utilized above, further characterization and experimentation is recommended in the following section. In conjunction with some methodological changes in the experimental setup, improved yields are viable.   Further Catalyst Characterization The band structure of the photocatalysts presented in this study have not been exhaustively examined. Challenges associated with the characteristics of g-CN as opposed to traditional metal oxide semiconductors have resulted in technical obstacles in pursuing typical analytical measurements.   4.2.1.1 Ultraviolet Photoelectron Spectroscopy (UPS) Chiefly among them are UPS spectra, which struggles due to the low conductivity of the g-CN material coupled with the low penetration of this characterization technique, preventing the application of a precious metal coating to reduce samples surface charging as is common in XPS applications. This characterization technique would be valuable to pursue via the creation of very thinly Ir coated (<3nm) photocatalyst thin films on doped silica supports as it would allow an understanding of the valence band maxima, work function, and fermi level of the photocatalyst materials. Coupled with band gap information previously collected, the conduction band minima could also be implied. These measurements would constitute a more exhaustive study of the band structures of the photocatalyst examined herein and potentially reveal effects of band structure, but not necessarily bandgap on photocatalyst production.  76  4.2.1.2 Electrical Impedance Spectroscopy Similarly, the hydrophilic qualities of g-CN rapidly result in thin film delamination when attempting to conduct electrochemical techniques such as electrical impedance spectroscopy, yielding a Mott-Schottky analysis. This technique could imply the flat band potential and therefore the conduction band minima. In conjunction with bandgap information this would then yield a full understanding of band structure.    Experimentation 4.2.1.1 Materials and Synthesis Specific suggestions for the improvement of photocatalysis in this specific setup include: a) serial experimentation to determine the optimal weight percentage of rGO to g-CN and BiVO4, particularly in a 3-component indirect z-scheme whereas the present study only utilizes approximate findings from literature. b) Changing the order of composite self-assembly in the aim of facilitating better interfacial interaction between semiconductors. c) altering the doping concentration of P-g-CN to find optimal content rather that utilizing approximate content from the literature. d) Doping rGO and adjusting the degree of rGO surface reduction (C:O ratio) to adjust its fermi level and charge localization abilities and improve it as an intermediate charge transfer material for this research e) Utilizing Mulliken electronegativity values to achieve sensical doping strategies for new catalyst materials f) Exploring new methods to achieve surface amine functionalization of g-CN to increase the CO2 adsorption characteristics. This particularly applies to P-g-CN as it remains an interesting method to achieve improved production particularly through stabilized enhanced amine functionalization. 77  4.2.1.2 Z-Scheme Mediators The z-scheme mediator utilized here was not successful in improving results, despite the fermi level being energetically between the fermi level of each photocatalyst. A more comprehensive understanding of band bending scale should be used in the selection of future materials as well as accounting for shadowing effects of the semiconductor surfaces, if applicable. In the future, selection of a z-scheme photocatalyst of which the fermi level can be tuned is desirable in order to facilitate the development of better z-scheme mediators. The doping of materials like rGO and particle size in metal nanoparticles are two levers to utilize in establishing tunable conductive z-scheme materials [55, 92, 93, 94].     Experimental Apparatuses 4.2.2.1 Preliminary Experimentation Preliminary catalyst analysis can be conducted in a rapid manner with dye degradation experiments over the course of minutes compared to multi-hour experiments in CO2 reduction photoreactors. These experiments would also aid in allowing experimentation and grooming of analytical techniques to take place during reactor manufacturing and assembly. Although less applicable to results from the vapor phase, photocatalyst performance and ranking can be performed to hasten photocatalyst analysis. Similarly, photocatalyst absorbance via UV-vis spectroscopy, diffuse reflectance spectroscopy and PL spectroscopy should be utilized to more fully understand the mechanisms of synthesized photocatalyst improvements before investing in more expensive and time intensive analytical techniques. 78  4.2.2.2 Reactor Sealing  For almost all applications, the use of O-ring sealing is much preferred to the use of a gasket as is the case here. It is recommended that any further photoreactor development utilize O-rings which can be custom fabricated from the desired material in circular cross section as opposed to gaskets.  4.2.2.3 Reactor Geometries The use of a top or side radiated reactor simplifies quantum yield calculations and light attenuation uncertainties and is recommended in future use. Similarly, small reactor volumes achieved by small dimensions in the direction parallel to incident light can help to achieve smaller volumes, particularly for catalysts mounted on supports in the solid phase, which for similar amounts of catalyst should lead to higher reactor concentrations with less error as a result peak integration. Finally, smaller reactors coupled with GC online applications utilize only a very small amount of reactor contents in order to measure concentration and as such large reactor volumes are not needed to maintain reaction pressure conditions as would be the case when withdrawing larger samples in gas-tight syringes for GC injection. However, in this case in-line drying of gases sent to the GC as well as re-humidification upon return to the experimental setup is required, adding significant complexity to reactor implementation and design.  4.2.2.4 Reactor Materials Although initial concerns were present regarding the corrosion of Al alloys for CO2 reductions due to material compatibility charts published by Cole-Palmer [95], the experience of other researchers utilizing surface ground Al alloys did not show significant signs of corrosion after 79  experimentation over the course of the year. Use of Al alloys would increase the manufacturability of designs and decrease the material cost and amount of machine shop time required for fabrication compared to stainless steel utilized here. However, use of Al alloys limits the ability to incorporate welding into reactor design and decreases the strength of threaded connections, which would have to be considered in further reactor design applications.   The use of quartz windows of appropriate thickness for the pressures expected in experimental designs as well as options for pressure regulation and relief should be considered required applications for design creation. A lab incident related to reactor pressure buildup and the breaking of a quartz window and Hg lamp, contaminating the lab could have been avoided if design criteria were used in specifying thickness of quartz windows required for the desired geometries, spans, and pressures with the appropriate safety factors [96].  4.2.2.5 Temperature and Pressure As previously mentioned, reactor pressure and temperature monitoring are crucial in humidified gas phase reactions to prevent reactor failure due to over-pressure conditions. Temperature and pressure should be monitored to be consistent across experiments as these govern overall catalyst performance parameters. Smaller reactors are easier to implement these controls for.  4.2.2.6 Reaction Conditions High H2O vapor partial pressures require elevated reactor temperatures in order to prevent condensation. However, catalyst temperature also affects adsorption desorption equilibria and 80  residence time of reactants on the catalyst. Reactor temperature could be studied as a variable to achieve improvements in catalytic performance.  4.2.2.7 Light Source Light intensity utilized in CO2 reduction applications is highly irregular. Trends towards intensities corresponding to those comparable to concentrated sunlight applications have become standard in order to increase the yield reported in μmol per gram of catalyst. Further, the use of light sources without light collimation results in non-uniformly irradiated catalysts, making quantum yield calculations less meaningful. Conversely, the intensities used in this study appear to yield results that do not approach the best reported studies who use much more concentrated light than the reported light intensities here, nor does it lead to as reduced of products [41]. As a result, greatest importance is the relative improvement of photocatalyst materials given the same setup. It appears that for CO2 reduction, surface charge density may result in selectivity differences and light intensity is an important consideration in this application. Future research should consider more concentrated light sources as a mode to generate higher evolution rates on a μmol/gcat basis, but also weigh this against the advantages of light collimation. 81  References [1] “Special Report on Climate Change and Land — IPCC site.” [Online]. Available: https://www.ipcc.ch/srccl/. [Accessed: 01-Aug-2020]. [2]  Q. Zhang, M. Liu, S. Liu, L. Qiao, X. Hu, and H. Tian, “Z-scheme g-C3N4/BiVO4 photocatalysts with RGO as electron transport accelerator,” J. Mater. Sci. Mater. Electron., vol. 31, no. 1, pp. 667–676, Jan. 2020. [3]  Z. Zhu, H. Pan, M. Murugananthan, J. Gong, and Y. Zhang, “Visible light-driven photocatalytically active g-C3N4 material for enhanced generation of H2O2,” Appl. Catal. B Environ., vol. 232, pp. 19–25, Sep. 2018. [4]  D. A. Neaman, Semiconductor Physics and Devices, 2012. [5] K. Li, X. An, K. Hyeon Park, M. Khraisheh, and J. Tang, “A critical review of CO2 photoconversion: Catalysts and reactors,” Catal. Today, vol. 224, pp. 3–12, 2014. [6] J. Low, C. Jiang, B. Cheng, S. Wageh, A. A. Al-Ghamdi, and J. Yu, “A Review of Direct Z-Scheme Photocatalysts,” Small Methods, vol. 1, no. 5, p. 1700080, May 2017. [7] A. FUJISHIMA and K. HONDA, “Electrochemical Photolysis of Water at a Semiconductor Electrode,” Nature, vol. 238, no. 5358, pp. 37–38, Jul. 1972. [8] M. Tahir and N. S. Amin, “Advances in visible light responsive titanium oxide-based photocatalysts for CO2 conversion to hydrocarbon fuels,” Energy Convers. Manag., vol. 76, pp. 194–214, Dec. 2013. [9] J.-C. Wang et al., “Indirect Z-Scheme BiOI/g-C 3 N 4 Photocatalysts with Enhanced Photoreduction CO 2 Activity under Visible Light Irradiation.” [10] F. Chen et al., “Novel ternary heterojunction photcocatalyst of Ag nanoparticles and g-C3N4 nanosheets co-modified BiVO4 for wider spectrum visible-light photocatalytic degradation of refractory pollutant,” Appl. Catal. B Environ., vol. 205, pp. 133–147, May 2017. [11] W. J. Jo et al., “Phase transition-induced band edge engineering of BiVO 4 to split pure water under visible light.” [12] J. Yu and A. Kudo, “Effects of structural variation on the photocatalytic performance of hydrothermally synthesized BiVO4,” Adv. Funct. Mater., vol. 16, no. 16, pp. 2163–2169, Oct. 2006. 82  [13] Y. Zhang, Y. Guo, H. Duan, H. Li, C. Sun, and H. Liu, “Facile synthesis of V4+ self-doped, [010] oriented BiVO4 nanorods with highly efficient visible light-induced photocatalytic activity,” Phys. Chem. Chem. Phys., vol. 16, no. 44, pp. 24519–24526, Oct. 2014. [14] M. L. Guan, D. K. Ma, S. W. Hu, Y. J. Chen, and S. M. Huang, “From hollow olive-shaped BiVO4 to n-p core-shell BiVO 4@Bi2O3 microspheres: Controlled synthesis and enhanced visible-light-responsive photocatalytic properties,” Inorg. Chem., vol. 50, no. 3, pp. 800–805, Feb. 2011. [15] W.-J. Ong, L.-L. Tan, S.-P. Chai, and S.-T. Yong, “Graphene oxide as a structure-directing agent for the two-dimensional interface engineering of sandwich-like graphene–g-C 3 N 4 hybrid nanostructures with enhanced visible-light photoreduction of CO 2 to methane,” Chem. Commun. Chem. Commun, vol. 51, no. 51, pp. 858–861, 2015. [16] W.-J. Ong, L.-L. Tan, S.-P. Chai, S.-T. Yong, and A. R. Mohamed, “Surface charge modification via protonation of graphitic carbon nitride (g-C3N4) for electrostatic self-assembly construction of 2D/2D reduced graphene oxide (rGO)/g-C3N4 nanostructures toward enhanced photocatalytic reduction of carbon dioxide to methane,” Nano Energy, vol. 13, pp. 757–770, Apr. 2015. [17]  D. Jiang, L. Chen, J. Zhu, M. Chen, W. Shi, and J. Xie, “Novel p–n heterojunction photocatalyst constructed by porous graphite-like C 3 N 4 and nanostructured BiOI: facile synthesis and enhanced photocatalytic activity,” vol. 42, 2013. [18] J. Wen, J. Xie, X. Chen, and X. Li, “A review on g-C 3 N 4 -based photocatalysts,” Appl. Surf. Sci., vol. 391, pp. 72–123, 2017. [19] J. Fu, J. Yu, C. Jiang, and B. Cheng, “g-C3N4-Based Heterostructured Photocatalysts,” Advanced Energy Materials, vol. 8, no. 3. Wiley-VCH Verlag, Jan. 2018. [20] Z. Sun, H. Wang, Z. Wu, and L. Wang, “g-C3N4 based composite photocatalysts for photocatalytic CO2 reduction,” Catal. Today, vol. 300, pp. 160–172, Feb. 2018. [21] S. Wang et al., “Phosphorous doped carbon nitride nanobelts for photodegradation of emerging contaminants and hydrogen evolution,” Appl. Catal. B Environ., vol. 257, p. 117931, Nov. 2019. [22] J. K. Stolarczyk, S. Bhattacharyya, L. Polavarapu, and J. Feldmann, “Challenges and Prospects in Solar Water Splitting and CO2 Reduction with Inorganic and Hybrid Nanostructures,” ACS Catal., vol. 8, no. 4, pp. 3602–3635, Apr. 2018. [23] G. Zhang, M. Zhang, X. Ye, X. Qiu, S. Lin, and X. Wang, “Iodine modified carbon nitride semiconductors as visible light photocatalysts for hydrogen evolution,” Adv. Mater., vol. 26, no. 5, pp. 805–809, Feb. 2014. 83  [24] Q. Han, C. Hu, F. Zhao, Z. Zhang, N. Chen, and L. Qu, “One-step preparation of iodine-doped graphitic carbon nitride nanosheets as efficient photocatalysts for visible light water splitting,” J. Mater. Chem. A, vol. 3, no. 8, pp. 4612–4619, Feb. 2015. [25] T. Y. Ma, J. Ran, S. Dai, M. Jaroniec, and S. Z. Qiao, “Phosphorus-Doped Graphitic Carbon Nitrides Grown in Situ on Carbon-Fiber Paper: Flexible and Reversible Oxygen Electrodes,” Angew. Chemie - Int. Ed., vol. 54, no. 15, pp. 4646–4650, Apr. 2015. [26]  Y. P. Zhu, T. Z. Ren, and Z. Y. Yuan, “Mesoporous Phosphorus-Doped g-C3N4 Nanostructured Flowers with Superior Photocatalytic Hydrogen Evolution Performance,” ACS Appl. Mater. Interfaces, vol. 7, no. 30, pp. 16850–16856, Aug. 2015. [27] S. Guo et al., “Phosphorus-Doped Carbon Nitride Tubes with a Layered Micro-nanostructure for Enhanced Visible-Light Photocatalytic Hydrogen Evolution,” Angew. Chemie Int. Ed., vol. 55, no. 5, pp. 1830–1834, Jan. 2016. [28] Y. Wang, H. Li, J. Yao, X. Wang, and M. Antonietti, “Synthesis of boron doped polymeric carbon nitride solids and their use as metal-free catalysts for aliphatic C-H bond oxidation,” Chem. Sci., vol. 2, no. 3, pp. 446–450, 2011. [29] S. C. Yan, Z. S. Li, and Z. G. Zou, “Photodegradation of rhodamine B and methyl orange over boron-doped g-C 3N4 under visible light irradiation,” Langmuir, vol. 26, no. 6, pp. 3894–3901, Mar. 2010. [30] J. Li, B. Shen, Z. Hong, B. Lin, B. Gao, and Y. Chen, “A facile approach to synthesize novel oxygen-doped g-C3N4 with superior visible-light photoreactivity,” Chem. Commun., vol. 48, no. 98, pp. 12017–12019, Nov. 2012. [31] G. Liu et al., “Unique electronic structure induced high photoreactivity of sulfur-doped graphitic C3N4,” J. Am. Chem. Soc., vol. 132, no. 33, pp. 11642–11648, Aug. 2010. [32] C. Xu, Q. Han, Y. Zhao, L. Wang, Y. Li, and L. Qu, “Sulfur-doped graphitic carbon nitride decorated with graphene quantum dots for an efficient metal-free electrocatalyst,” J. Mater. Chem. A, vol. 3, no. 5, pp. 1841–1846, Feb. 2015. [33] J. Zhang, M. Zhang, R.-Q. Sun, and X. Wang, “A Facile Band Alignment of Polymeric Carbon Nitride Semiconductors to Construct Isotype Heterojunctions,” Angew. Chemie, vol. 124, no. 40, pp. 10292–10296, Oct. 2012.  [34] J. Fu, B. Zhu, C. Jiang, B. Cheng, W. You, and J. Yu, “Hierarchical Porous O-Doped g-C3N4 with Enhanced Photocatalytic CO2 Reduction Activity,” Small, vol. 13, no. 15, Apr. 2017. 84  [35] F. Raziq et al., “Photocatalytic solar fuel production and environmental remediation through experimental and DFT based research on CdSe-QDs-coupled P-doped-g-C3N4 composites,” Appl. Catal. B Environ., vol. 270, p. 118867, Aug. 2020. [36] B. Liu et al., “Phosphorus-Doped Graphitic Carbon Nitride Nanotubes with Amino-rich Surface for Efficient CO2 Capture, Enhanced Photocatalytic Activity, and Product Selectivity,” ACS Appl. Mater. Interfaces, vol. 10, no. 4, pp. 4001–4009, Jan. 2018. [37] E. M. Dias, K. C. Christoforidis, L. Francàs, and C. Petit, “Tuning Thermally Treated Graphitic Carbon Nitride for H2 Evolution and CO2 Photoreduction: The Effects of Material Properties and Mid-Gap States,” ACS Appl. Energy Mater., vol. 1, no. 11, pp. 6524–6534, Nov. 2018. [38] K. Wang, Q. Li, B. Liu, B. Cheng, W. Ho, and J. Yu, “Sulfur-doped g-C3N4 with enhanced photocatalytic CO2-reduction performance,” Appl. Catal. B Environ., vol. 176–177, pp. 44–52, Oct. 2015. [39]  K. S. Lakhi et al., “Mesoporous carbon nitrides: Synthesis, functionalization, and applications,” Chemical Society Reviews, vol. 46, no. 1. Royal Society of Chemistry, pp. 72–101, 01-Jan-2017. [40] H. Huo et al., “Synergy of Photocatalysis and Adsorption for Simultaneous Removal of Hexavalent Chromium and Methylene Blue by g-C3N4/BiFeO3/Carbon Nanotubes Ternary Composites,” Int. J. Environ. Res. Public Health, vol. 16, no. 17, p. 3219, Sep. 2019. [41]  W. J. Ong et al., “Unravelling charge carrier dynamics in protonated g-C3N4 interfaced with carbon nanodots as co-catalysts toward enhanced photocatalytic CO2 reduction: A combined experimental and first-principles DFT study,” Nano Res., vol. 10, no. 5, pp. 1673–1696, May 2017. [42] L. Ma et al., “Protonation of Graphitic Carbon Nitride (g-C3N4) for an Electrostatically Self-Assembling Carbon@g-C3N4 Core-Shell Nanostructure toward High Hydrogen Evolution,” ACS Sustain. Chem. Eng., vol. 5, no. 8, pp. 7093–7103, Aug. 2017. [43]  K. Wang, S. Song, Q. Zhang, Y. Jin, and Q. Zhang, “Fabrication of protonated g-C3N4 nanosheets as promising proton conductive materials,” Chem. Commun., vol. 55, no. 51, pp. 7414–7417, Jun. 2019. [44] R. Huo, X. L. Yang, J. Y. Yang, S. Y. Yang, and Y. H. Xu, “Self-assembly synthesis of BiVO4/Polydopamine/g-C3N4 with enhanced visible light photocatalytic performance,” Mater. Res. Bull., vol. 98, pp. 225–230, Feb. 2018. [45] K. Akaike, K. Aoyama, S. Dekubo, A. Onishi, and K. Kanai, “Characterizing Electronic Structure near the Energy Gap of Graphitic Carbon Nitride Based on Rational 85  Interpretation of Chemical Analysis,” Chem. Mater., vol. 30, no. 7, pp. 2341–2352, Apr. 2018. [46] K. Yu, X. Hu, K. Yao, P. Luo, X. Wang, and H. Wang, “Preparation of an ultrathin 2D/2D rGO/g-C3N4 nanocomposite with enhanced visible-light-driven photocatalytic performance,” RSC Adv., vol. 7, no. 58, pp. 36793–36799, Jul. 2017. [47] M. Rakibuddin and H. Kim, “Reduced graphene oxide supported C 3 N 4 nanoflakes and quantum dots as metal-free catalysts for visible light assisted CO 2 reduction,” Beilstein J. Nanotechnol., vol. 10, no. 1, pp. 448–458, 2019. [48] F. Raziq et al., “Promoting visible-light photocatalytic activities for carbon nitride based 0D/2D/2D hybrid system: Beyond the conventional 4-electron mechanism,” Appl. Catal. B Environ., vol. 270, p. 118870, Aug. 2020. [49] N. Nie, L. Zhang, J. Fu, B. Cheng, and J. Yu, “Self-assembled hierarchical direct Z-scheme g-C 3 N 4 /ZnO microspheres with enhanced photocatalytic CO 2 reduction performance,” Appl. Surf. Sci., vol. 441, pp. 12–22, May 2018. [50] R. Akbarzadeh, C. S. L. Fung, R. A. Rather, and I. M. C. Lo, “One-pot hydrothermal synthesis of g-C3N4/Ag/AgCl/BiVO4 micro-flower composite for the visible light degradation of ibuprofen,” Chem. Eng. J., vol. 341, pp. 248–261, Jun. 2018. [51] Y. Si et al., “g-C3N4/Pt/BiVO4 nanocomposites for highly efficient visible-light photocatalytic removal of contaminants and hydrogen generation,” Nanotechnology, vol. 31, no. 12, p. 125706, Jan. 2020. [52] D. Jiang, P. Xiao, L. Shao, D. Li, and M. Chen, “RGO-Promoted All-Solid-State g-C3N4/BiVO4 Z-Scheme Heterostructure with Enhanced Photocatalytic Activity toward the Degradation of Antibiotics,” Ind. Eng. Chem. Res., vol. 56, no. 31, pp. 8823–8832, Aug. 2017. [53] W. K. Jo, S. Kumar, S. Eslava, and S. Tonda, “Construction of Bi2WO6/RGO/g-C3N4 2D/2D/2D hybrid Z-scheme heterojunctions with large interfacial contact area for efficient charge separation and high-performance photoreduction of CO2 and H2O into solar fuels,” Appl. Catal. B Environ., vol. 239, pp. 586–598, Dec. 2018. [54] Y. Yang et al., “Urchin-like hierarchical CoZnAl-LDH/RGO/g-C3N4 hybrid as a Z-scheme photocatalyst for efficient and selective CO2 reduction,” Appl. Catal. B Environ., vol. 255, p. 117771, Oct. 2019. [55] Y. Bao and K. Chen, “Novel Z-scheme BiOBr/reduced graphene oxide/protonated g-C 3 N 4 photocatalyst: Synthesis, characterization, visible light photocatalytic activity and mechanism,” Appl. Surf. Sci., vol. 437, pp. 51–61, Apr. 2018. 86  [56] A. Bafaqeer, M. Tahir, A. A. Khan, and N. A. S. Amin, “Indirect Z-scheme assembly of 2D ZnV2O6/RGO/g-C3N4 nanosheets with RGO/PCN as solid-state electron mediators toward visible-light-enhanced CO2 reduction,” Ind. Eng. Chem. Res., vol. 58, no. 20, pp. 8612–8624, May 2019. [57] X. Hu, J. Hu, Q. Peng, X. Ma, S. Dong, and H. Wang, “Construction of 2D all-solid-state Z-scheme g-C3N4/BiOI/RGO hybrid structure immobilized on Ni foam for CO2 reduction and pollutant degradation,” Mater. Res. Bull., vol. 122, p. 110682, Feb. 2020. [58] W. Chen, M. Liu, X. Li, and L. Mao, “Synthesis of 3D mesoporous g-C3N4 for efficient overall water splitting under a Z-scheme photocatalytic system,” Appl. Surf. Sci., vol. 512, p. 145782, May 2020. [59] M. Lu et al., “Remarkable photocatalytic activity enhancement of CO2 conversion over 2D/2D g-C3N4/BiVO4 Z-scheme heterojunction promoted by efficient interfacial charge transfer,” Carbon N. Y., vol. 160, pp. 342–352, Apr. 2020. [60] G. J. Lee, X. Y. Lee, C. Lyu, N. Liu, S. Andandan, and J. J. Wu, “Sonochemical synthesis of copper-doped BiVO4/g-C3N4 nanocomposite materials for photocatalytic degradation of bisphenol a under simulated sunlight irradiation,” Nanomaterials, vol. 10, no. 3, Mar. 2020. [61] L. Li et al., “Direct Z-scheme photocatalyst for efficient water pollutant degradation: A case study of 2D g-C3N4/BiVO4,” Mater. Chem. Phys., vol. 241, p. 122308, Feb. 2020. [62] T. Ahmed, M. Ammar, A. Saleem, H. L. Zhang, and H. Bin Xu, “Z -scheme 2D-m-BiVO4 networks decorated by a g-CN nanosheet heterostructured photocatalyst with an excellent response to visible light,” RSC Adv., vol. 10, no. 6, pp. 3192–3202, Jan. 2020. [63] P. Raizada, A. Sudhaik, P. Singh, A. Hosseini-Bandegharaei, and P. Thakur, “Converting type II AgBr/VO into ternary Z scheme photocatalyst via coupling with phosphorus doped g-C3N4 for enhanced photocatalytic activity,” Sep. Purif. Technol., vol. 227, p. 115692, Nov. 2019. [64] Y. Wang et al., “Synthesizing Co3O4-BiVO4/g-C3N4 heterojunction composites for superior photocatalytic redox activity,” Sep. Purif. Technol., vol. 239, p. 116562, May 2020. [65] C. Dong et al., “Colloidal Synthesis of Ultrathin Monoclinic BiVO 4 Nanosheets for Z-Scheme Overall Water Splitting under Visible Light,” 2018. [66] “(SS1.0kW) 1000W Fully Reflective Solar Simulator - Fully Reflective Solar Simulators - Steady State Solar Simulators - Solar Simulation and PV Testing Equipment - Products | Sciencetech Inc.” [Online]. Available: http://www.sciencetech-inc.com/all-87  products/solarsimulators/continuous-solar/fully-reflective-solar-simulators/ss1-0kw-1000w-fully-reflective-solar-simulator.html. [Accessed: 01-Aug-2018]. [67] Y. Luo et al., “Nonmetal element doped g-C3N4 with enhanced H2 evolution under visible light irradiation,” J. Mater. Res., vol. 33, no. 09, pp. 1268–1278, May 2018. [68] Z. Li et al., “Novel rugby-like g-C3N4/BiVO4 core/shell Z-scheme composites prepared via low-temperature hydrothermal method for enhanced photocatalytic performance,” Sep. Purif. Technol., vol. 232, p. 115937, Feb. 2020. [69] L. Zhang et al., “Phosphorous doped graphitic-C3N4 hierarchical architecture for hydrogen production from water under visible light,” Mater. Today Energy, vol. 5, pp. 91–98, Sep. 2017. [70] Y. Zhang, Q. Zhang, Q. Shi, Z. Cai, and Z. Yang, “Acid-treated g-C3N4 with improved photocatalytic performance in the reduction of aqueous Cr(VI) under visible-light,” Sep. Purif. Technol., vol. 142, pp. 251–257, Mar. 2015. [71] P. Xia, B. Cheng, J. Jiang, and H. Tang, “Localized π-conjugated structure and EPR investigation of g-C3N4 photocatalyst,” Appl. Surf. Sci., vol. 487, pp. 335–342, Sep. 2019. [72] Y. Feng, J. Shen, Q. Cai, H. Yang, and Q. Shen, “The preparation and properties of a g-C3N4/AgBr nanocomposite photocatalyst based on protonation pretreatment,” New J. Chem., vol. 39, no. 2, pp. 1132–1138, Feb. 2015. [73] M. Caux, F. Fina, J. T. S. Irvine, H. Idriss, and R. Howe, “Impact of the annealing temperature on Pt/g-C3N4 structure, activity and selectivity between photodegradation and water splitting,” Catal. Today, vol. 287, pp. 182–188, Jun. 2017. [74] R. López and R. Gómez, “Band-gap energy estimation from diffuse reflectance measurements on sol-gel and commercial TiO 2: A comparative study,” J. Sol-Gel Sci. Technol., vol. 61, no. 1, pp. 1–7, Jan. 2012. [75] M. A. Butler, “Prediction of Flatband Potentials at Semiconductor-Electrolyte Interfaces from Atomic Electronegativities,” J. Electrochem. Soc., vol. 125, no. 2, p. 228, 1978. [76] J. K. Cooper et al., “Indirect bandgap and optical properties of monoclinic bismuth vanadate,” J. Phys. Chem. C, vol. 119, no. 6, pp. 2969–2974, Feb. 2015. [77] H. Shi, Z. Li, J. Kou, J. Ye, and Z. Zou, “Facile synthesis of single-crystalline Ag2V4O 11 nanotube material as a novel visible-light-sensitive photocatalyst,” J. Phys. Chem. C, vol. 115, no. 1, pp. 145–151, Jan. 2011. [78] J. Fu, B. Chang, Y. Tian, F. Xi, and X. Dong, “Novel C3N4-CdS composite photocatalysts with organic-inorganic heterojunctions: In situ synthesis, exceptional 88  activity, high stability and photocatalytic mechanism,” J. Mater. Chem. A, vol. 1, no. 9, pp. 3083–3090, Mar. 2013. [79] J. Zhang, F. Ren, M. Deng, and Y. Wang, “Enhanced visible-light photocatalytic activity of a g-C3N4/BiVO4 nanocomposite: A first-principles study,” Phys. Chem. Chem. Phys., vol. 17, no. 15, pp. 10218–10226, Apr. 2015. [80] R. Cheng, L. Shen, J. Yu, S. Xiang, and X. Zheng, “Photocatalytic Inactivation of Bacteriophage f2 with Ag3PO4/g-C3N4 Composite under Visible Light Irradiation: Performance and Mechanism,” Catalysts, vol. 8, no. 10, p. 406, Sep. 2018. [81] S. Ghattavi and A. Nezamzadeh-Ejhieh, “A visible light driven AgBr/g-C3N4 photocatalyst composite in methyl orange photodegradation: Focus on photoluminescence, mole ratio, synthesis method of g-C3N4 and scavengers,” Compos. Part B Eng., vol. 183, p. 107712, Feb. 2020. [82] Y. Che, Q. Liu, B. Lu, J. Zhai, K. Wang, and Z. Liu, “Plasmonic ternary hybrid photocatalyst based on polymeric g-C3N4 towards visible light hydrogen generation,” Sci. Rep., vol. 10, no. 1, pp. 1–12, Dec. 2020. [83] H. Pang et al., “Facile synthesis of few-layer g-C3N4 nanosheets anchored with cubic-phase CdS nanocrystals for high photocatalytic hydrogen generation activity,” J. Alloys Compd., vol. 839, p. 155684, Oct. 2020. [84] Y. Su et al., “Decoration of TiO2/g-C3N4 Z-scheme by carbon dots as a novel photocatalyst with improved visible-light photocatalytic performance for the degradation of enrofloxacin,” RSC Adv., vol. 7, no. 54, pp. 34096–34103, Jul. 2017. [85] Y. Zhou et al., “Brand new P-doped g-C3N4: Enhanced photocatalytic activity for H2 evolution and Rhodamine B degradation under visible light,” J. Mater. Chem. A, vol. 3, no. 7, pp. 3862–3867, Feb. 2015. [86] Z. Sun et al., “g-C3N4 foam/Cu2O QDs with excellent CO2 adsorption and synergistic catalytic effect for photocatalytic CO2 reduction,” Environ. Int., vol. 130, p. 104898, Sep. 2019. [87] N. Ojha, A. Bajpai, and S. Kumar, “Visible light-driven enhanced CO2 reduction by water over Cu modified S-doped g-C3N4,” Catal. Sci. Technol., vol. 9, no. 17, pp. 4598–4613, Aug. 2019. [88] Q. Huang, J. Yu, S. Cao, C. Cui, and B. Cheng, “Efficient photocatalytic reduction of CO 2 by amine-functionalized g-C 3 N 4,” in Applied Surface Science, 2015, vol. 358, pp. 350–355. 89  [89] Y. Zhang, T. Mori, J. Ye, and M. Antonietti, “Phosphorus-doped carbon nitride solid: Enhanced electrical conductivity and photocurrent generation,” J. Am. Chem. Soc., vol. 132, no. 18, pp. 6294–6295, May 2010. [90]  R. A. Rather, M. Khan, and I. M. C. Lo, “High charge transfer response of g-C3N4/Ag/AgCl/BiVO4 microstructure for the selective photocatalytic reduction of CO2 to CH4 under alkali activation,” J. Catal., vol. 366, pp. 28–36, Oct. 2018. [91] Y. Huang, M. Fu, and T. He, “Synthesis of g-C3N4/BiVo4 nanocomposite photocatalyst and its application in photocatalytic reduction of CO2,” Wuli Huaxue Xuebao/ Acta Phys. - Chim. Sin., vol. 31, no. 6, pp. 1145–1152, Jun. 2015. [92] Abid, P. Sehrawat, S. S. Islam, P. Mishra, and S. Ahmad, “Reduced graphene oxide (rGO) based wideband optical sensor and the role of Temperature, Defect States and Quantum Efficiency,” Sci. Rep., vol. 8, no. 1, pp. 1–13, Dec. 2018. [93] H. Zeng, X. Zhu, Y. Liang, and X. Guo, “Interfacial Layer Engineering for Performance Enhancement in Polymer Solar Cells,” Polymers (Basel)., vol. 7, no. 2, pp. 333–372, Feb. 2015. [94]  H. Li, H. Yu, X. Quan, S. Chen, and Y. Zhang, “Uncovering the Key Role of the Fermi Level of the Electron Mediator in a Z-Scheme Photocatalyst by Detecting the Charge Transfer Process of WO3-metal-gC3N4 (Metal = Cu, Ag, Au),” ACS Appl. Mater. Interfaces, vol. 8, no. 3, pp. 2111–2119, Jan. 2016. [95] “Chemical Compatibility Database from Cole-Parmer.” [Online]. Available: https://www.coleparmer.com/chemical-resistance. [Accessed: 01-Aug-2020]. [96]  “Technical Glass Products: Pressure Calculations.” [Online]. Available: https://technicalglass.com/pressure_calculations/. [Accessed: 01-Aug-2020].               

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0392943/manifest

Comment

Related Items