2019-2020
Iron Sulphide(s): A sustainable semi-conductor with a multitude of
applications
FeS
xProperties, Synthesis & applications
Student:
Floris Kersten
Student ID:
11850558
University: UvA
Supervisor:
Dr. S. Grecea
1
Table of Contents
List of abbreviations ...2 Abstract ...3 1. Introduction ...4 2. Theory ...5 2.1 A world of semi-conductors ...52.2 Pyrite Synthesis & Properties ...6
2.3 Applications ... 12
3 Discussion & Future Prospects ... 23
4 Conclusion ... 26
2
List of abbreviations
Hydrogen Evolution Reaction (HER) Oxygen Evolution Reaction (OER) FeS2/FeS (FeSx)
Bandgap/Energy Gap (Eg)
electrophoretic deposition (EDP) anodic aluminium oxide (AAO)
chemical vapor electrodeposition (CVD) polyvinylpyrrolidone (PVP)
Indium Tin Oxide (ITO) X-Ray Diffraction (XRD)
Di-tert-butyldisulphide or [(CH3)3C]2S2 (TBDS) Grazing-Incidence X-ray Diffraction (GIXRD) Scanning Electron Microscopy (SEM) Atomic Force Microscopy (AFM) Dye-sensitised solar cell (DSSC) Cyclic Voltammetry (CV) Rose Bengal (RB) Methylene blue (MB) Safranine T (ST) Methyl orange (MO) Rhodamine B (RHB) Pyronine B (PRB)
3
Abstract
Pyrite or FeS2 has a history of human applications such as the ‘’ore’’ used to obtain elemental sulphur11. In more recent years pyrite has grasped attention due to its semi-conduction properties, low bandgap (0,95 eV Eg) and its vast natural abundance. With a dawning era of green and sustainable chemistry the non-toxic properties of pyrite have proven to be quite interesting as well12,13. With that in mind, the main question this thesis tries to answer is: Can FeSx be used as a sustainable and multifunctional replacement for current semi-conducting materials?
Semiconductors are very important materials and are applied in a huge variety of technological and industrial applications. Certain properties of semiconductors lead to classifications such as N or P and intrinsic or extrinsic semi-conductors, which give insight in the conductive behaviour of a said material. Pyrite is an intrinsic semiconductor with either N or P type behaviour, depending on the crystal structure. The bandgap (which is the most important property of a semiconductor) lies close to that of silica (1,1 eV) and pyrite is thus interesting to study15,16,17.
To obtain pyrite, many different synthesis methods were developed21,22,23,24,25,26,29. The most suited method depends entirely on the application and thus material requirements. With a specific synthesis, specific material properties can be controlled such as: (I) phase purity, (II) stoichiometry, (III) surface morphology, (IV) bandgap, (V) crystallinity, (VI) surface area, (VII) Defects. These properties can vary heavily with each synthesis method and in turn translate to different semi conducting properties such as different conductivity and a varying bandgap or adsorption coefficient18,19.
Important applications of pyrite could be: Counter electrodes in Dye sensitised solar cell30, hydrogen Evolution Catalyst(s)33,34,9 and photocatalysts for the degradation of organic dyes13,35. Which are highlighted in this thesis, with increasing interest in pyrite and more research done in this area more promising applications might make an appearance in the near future.
One of the main discussion points regarding the research into pyrite is the amount of fragmented information. Even though research dates back to the 1970’s30 information regarding the synthesis and application of pyrite is fragmented. There is no overview regarding which synthetic methods result in specific properties. The properties I to VII have to be selected and controlled in order to reach the optimal performance of your material, this is something that hasn’t been properly done yet.
To conclude the main research question, FeSx has the potential to replace semi-conductors because it’s much cheaper, more abundant and nontoxic. However it seems that the areas in which it can be applied are limited and it will thus, not change the current industry as we know it.
4
1. Introduction
FeS2, also known as pyrite or Fool’s gold is a mineral composed of iron and sulphur (figure 1)14. The word stems from the Greek word of fire ‘’as sparks may be struck from it’’11. It’s the most commonly found sulphide mineral on earth11. Throughout history pyrite has been used for a variety of applications such as the use in the ignition system of fire arms and later on as source for obtaining pure elemental sulphur11.
Figure 1: Pyrite crystal as can be found in nature14
More recently pyrite has grasped the attention of scientists due to a multitude of reasons. First of all, FeS2 has semi-conducting properties, a high absorption coefficient and a low bandgap (around 0,95 eV)12. These properties make it an interesting choice for photovoltaics, photo-catalytic chemistry or simply catalysis. Consecutively, the different ways FeSx can be synthesised and the varying morphology and/or doping of the corresponding coatings, layers, crystals or nano-particles of FeSx opened the material up to many new possible applications12,13.
These physical properties are however not the only reason why FeS2 is an interesting material. There are many materials with semi-conducting properties and competition is stiff13. FeS
2 is a non-toxic, sustainable materials with a high natural abundance and thus is ideally suited as a sustainable resource for a variety of industrial applications13.
With increasingly tighter regulations surrounding the utilisation of natural resources, strong fluctuating prices and the increasing public demand for a more cyclic and sustainable economy companies are forced to look for greener alternatives3,4. Materials made of FeS
x can possibly follow green chemistry principles where others can’t and thus offer both economically, technologically and environmentally positive future prospects opposed to some of the current non-sustainable semi-conducting materials13.
Lastly, due to the O-life project2, which is a project focused on finding answers related to the origin of life, FeSx has gained more popularity as well. It is assumed that FeSx might have played a role in the catalysis of early life forms here on earth and that the composite material might have come from space in the form of meteorites7,8.
This thesis will focus on a main hypothesis which is divided into multiple sub sections. The main hypothesis states: Can FeSx be used as a sustainable and multifunctional replacement for current semi-conducting materials? Which can then further be split up into what spiked interest into the research of FeSx and why is FeSx a sustainable material? How does FeSx differ from current semi-conductors? How does the synthesis and morphology of FeSx influence its properties? And What properties does FeSx have and what applications could utilize these properties? The goal of this thesis is to convey the answers to these questions to the reader.
The theory section of this thesis focuses on the synthesis, morphology, and applications of FeSx. After the required theory, an elaborate discussion will be presented which is focused on the bottlenecks regarding the aforementioned topics, including the future prospects. The thesis will be rounded up by a conclusion regarding the discussed matter.
5
2. Theory
2.1 A world of semi-conductors
Semiconductors are a very important part of modern days technology. Semiconductors find most of their applications in electric devices such as resistors, capacitors, diodes and transistors. These devices are consecutively found in nearly all electronic equipment we use ranging from mobile phones and computers to the satellites in space15.
A semiconductor, as name gives away is a material which has electric conductive properties in between an insulator (plastics) and a conductor (metals). Semi-conductors have, on the molecular level a crystalline structure which can be tuned by adding small impurities15,16. This so called ‘’doping’’ results in an N or P type semiconductor. N type semiconductors have excess free electrons and P type semiconductors have so called ‘’holes’’. In chemical terms, this relates to the elements used, e.g. a crystal lattice of silica can be N-doped with elements from the 15th row of the periodic table (namely N, P, As, Sb). These elements have one extra valence electron compared to silica and therefore create free electrons in the crystal lattice. For P doping of silica one uses elements from the 13th row (namely B, Al, Ga) which have one less valence electron than silica and therefore create an electron deficiency15,16.
Semiconductors can be divided into two groups: (I) Intrinsic semiconductors and (II) extrinsic semiconductors. Intrinsic semiconductors have an equal amount of free electrons and holes in their conduction band. Extrinsic semiconductors are doped and thus have a varying electron excess or deficiency15. The bandgap of a semiconductor is one of their most important features, it shows the energy gap between the valence and conduction band. The ‘’valence’’ and ‘’conduction’’ bands are basically all the HOMO’s (Valence) and LUMO’s (Conduction) added together in a semiconductor bulk material17.
Most commonly used intrinsic semiconductors are silica and germanium, with respective 1,1 eV and 0,66 eV bandgaps. These elements are used in for example photovoltaic cells in which photons excite electrons from the valence to the conduction band, when prevented to recombine these charges can create a potential difference which can be utilized to execute electric work17,18.
Pyrite(FeS2) has a bandgap(Eg) of 0,95 eV which is similar to the bandgap of silica. Pyrite also has a very high absorption coefficient of 5*105 cm-1 for wavelengths smaller than 700 nm18,19. This absorption coefficient shows how far light of a certain wavelength can penetrate into the material before it’s absorbed. It’s essential for photovoltaic materials to have a high absorption coefficient and suitable bandgap. These properties are partly what gave rise to the research done in applying pyrite to applications such as photovoltaic cells or photocatalysis19. Figure 220 shows a pyrite photovoltaic cell which produces electrons for a hydrogen evolution reaction.20
Figure 2: A photon strikes an FeS2 electron in the valance band and excites it to the conduction band. The electron is then
6
2.2 Pyrite Synthesis & Properties
The way FeS2 is synthesised is very important for its final properties such as surface area, crystallinity, possible defects, crystal structure, Fe/S content, particle size, impurities and pore sizes. These properties have a direct influence on how well the synthesised materials is suited for certain applications. Thus, depending on the application, a variety of methods have been developed to synthesise FeS2. A review of some of the synthetic routes with comments regarding the morphology is given here. These methods only summarize some of the most promising procedural parts to obtain FeSx suited for applications such as HER or Catalysis.
1. Sol-Gel Hydrothermal method with electrophoretic deposition (EPD)21
The sol-gel method with EDP is a multi-step process to prepare Pyrite films. The sol-gel method started with adding FeSO4*H2O together with a certain amount of polyvinylpyrrolidone (PVP). This was then dissolved into distilled water. The solution was heated to 90°C and stirred for one hour. After one hour pyridine and of thiourea were added. Consecutively the solution was refluxed at 95°C. The result was a grey colloid precursor which was added to a Teflon beaker together with a certain amount of sulphur. This mixture was then heated to 200°C in an autoclave for 40 hours. The result was a black powder which was further filtered and washed (multiple times) with distilled water, carbon disulphide and ethanol. This powder was then applied on a Indium Tin Oxide (ITO) glass substrate using EPD.21. Duan et al.21 have investigated the effects of the molar ratios of reactants in multiple steps of the synthesis. Starting with the formation of the amorphous precursor, which is composed of nanoparticles imbedded in PVP. Utilising X-Ray Diffraction (XRD) Duan et al.21 observed the mass ratio of PVP-to-FeSO4 to be 1:2. This resulted however in a mix of amorphous and not amorphous parts. When this ratio was increased to 1:1 the entire precursor was shown to be amorphous. This shows that PVP assists in dispersing the FeSO4homogeneously. Since the homogeneity is an important aspect for the next step, only the 1:1 ratio was used.
The 1:1 precursor was heated and resulting from this procedure a black powder was obtained. Utilising XRD again it was identified that samples at 100°C were not solely forming pyrite but also marcasite (polymorphic variant of pyrite) and FeFe2O4. However when heated to 200°C the amount of Pyrite formed was the major product. Duan et al.21 speculates 200°C is the right temperature for thermal-kinetic control to form FeS2 (cubic) crystals, however marcasite is still formed but in lower amounts. To increase the formation of Pyrite opposed to other structures or compounds further heating was executed at 200°C under varying quantities of sulphur. Duan et al.21 found the ‘’magic’’ molar ratio to be 3:1 Sulphur-to-FeSO4*H2O. This lead to the full conversion of marcasite to pyrite, ending up with pure and crystalline pyrite.
To distribute the pyrite homogeneously onto a surface EPD was used, the powder (after the 3:1 sulphur treatment) was sonicated for 24h in ethanol to reduce particle size. Homogeneous pyrite films were formed during the EPD procedure. A pyrite suspension was used and a continuous current of 6,5mA cm-2 was applied. This resulted in an average pyrite grain size of 1 micrometre onto the ITO glass substrates(Figure 3)21.
7
Figure 3: Pyrite particles on ITO glass, picture taken by SEM21 Duan et al.21 observed the E
g to be between 1,4 and 1,19 eV depending on the film thickness, with thicker films with larger grains resulting in a lower Eg. This Eg is however not yet near the theoretical value of 0,95 eV. Duan et al.21 also argues the smaller the grain size the higher optical absorption edge is observed. Duan et al.21 found the synthesised films to be mostly of the N-type conduction. The Hall mobility was found to be in the range of 429-32,9 cm2 V-1 S-1 with a carrier density of 1019 cm-3. This is superior to film synthesised utilising a spray method(200-1 cm2 V-1 S-1 with a carrier density of 1016 -1020)27 rather than EPD.
2. Sulfurizing at high temperature utilizing anodic aluminium oxide (AAO)22
Wan et al22 first synthesised the AAO which is the blueprint for the nanowires. AAO was formed in a twostep anodization process. Firstly an Al plate was anodized in sulphuric acid, consecutively Fe was electrodeposited in the formed pores, this was also executed at 15V at a frequency of 200Hz. The plating bath solution used was FeSO4 and boric acid. After electrodeposition the AAO-Fe was etched in a phosphoric and chromic acid bath to further remove the AAO. The remaining highly porous material was annealed in a vacuum sealed quarts ampoule together with sulphur powder. This was heated for 8h at temperatures ranging from 300-450°C, this resulted in the formation of FeS2 nanowires.22
Wan et al22 Analysed both the iron nanowires (figure 4) 22 and FeS
2 nanowires (figure 5) 22 utilising SEM.
8
Figure 5: Pyrite nanowires on AAO
To obtain the best results the iron nanowires(figure 4) were heated to 450°C under 1:1 sulphur atmosphere, this resulted in a stochiometric S/Fe ratio of 1,97:1 and the formation of Pyrite nanowires (figure 5). Further XRD analysis was carried out, this further confirmed the ‘’magic’’ ratio was about 2:1 S:Fe which resulted in the sole formation of pyrite nanowires. The XRD values of the nanowires were similar to the XRD values one would get when simply taking an XRD of FeS2 powder.
It can be observed in both (figure 4 & 5) that the formed nanowires are of similar length and width also the spacing between them is roughly the same. These properties are mainly controlled by choosing the right AAO template, these templates range from having pores sizes between 10 to 200 nm and pore lengths from 1 to 100 micrometres. Changing these values has a great impact on the properties of said wires. It is however unclear from the paper which exact sizes they used to start with in order to obtain their presented results.
Looking back at pyrite films, the ideal annealing temperature is around 200-300°C and a magic ratio of 3:1 S/Fe is observed. According to Wan et al22 the lower required temperature used for pyrite films is due to the easy accessibility for sulphur to reach and react with the iron. In other words: Films are a homogeneous surface with evenly distributed grains and are thus evenly and easily accessible. The nanowires are formed in pores with very little space between them. This results in diffusion limitations of the sulphur reaching and reacting with each iron wire. To fully convert the nanowires to pyrite, elevated temperatures were needed in order to overcome the diffusion gradient.
3. Thin film creation through chemical vapor electrodeposition (CVD)23
Bessergenev et al23 have synthesised a variety of manganese/iron/cobalt oxide & sulphur films. The metal-sulphur film were synthesised using CVD and complex precursors MnPh/Co/Fe[(C2H5)2NCS2]3. The oxide films were produced by annealing the sulphide films in the presence of an Ar/O2 flow at low. For the specific synthesis of pyrite films Di-tert-butyldisulphide (TBDS) was used during the annealing process. Excess sulphur was required in order to obtain a pyrite structure. The annealing temperatures of all synthesised films ranged from 350-500°C.
Bessergenev et al23 utilised the Ar/TBDS flow to regulate the sulphur content in the pyrite films. By controlling the pressure, evaporator temperature (200-226°C), substrate temperature (340-550°C). The substrates used to deposit the films on were either glass or fused-silica based substrates. The grow rates of the films on said substrates varied between 7 and 62 nm/min. The minimum temperature required in order to have film formation was 320°C.
Structural analysis was performed utilising Grazing-Incidence X-ray Diffraction (GIXRD) at the DESY/HASYLAB synchrotron. This analysis showed that fused-silica substrates annealed at 340°C with
9 TBDS and Fe[(C2H5)2NCS2]3 yielded an FeS2-x structure (0<x<1). This means the final structure has many defects in it and is somewhere between the structures of FeS and FeS2. The Sulphur loss is due to the evaporation of sulphur at low pressure and high temperature. To prevent this from happening a large excess of TBDS was required, which lead to the formation of pure FeS2. From the results of this analysis can be concluded that both the temperature and the TBDS vapour pressure dictate the final crystal structure.
The further oxidation of the formed pyrite films under Ar/O2 atmosphere was very easy and delivered good results. This shows that pyrite in general might be sensitive to heating and/or oxidation under aerobic circumstances, which in turn requires certain precautions to be taken when utilising these materials in certain applications which perform under the aforementioned conditions.
4. FeSx coatings through plasma spraying24,29
Wang et al24 presented a way to create an FeS coating via plasma spraying, the sprayed material was AISI 1045 steel. FeS grains of approximately 40 micrometre were used for plasma spraying. These grains were first coated by chemical heating treatment with a nickel-base alloy to prevent burning while being plasma-sprayed. The total nickel content in the FeS grains was around 5%. After spraying the grains, the FeS film thickness was around 0,5mm.
Guidotti et al29 First tried plasma spraying a mixture of Pyrite and elemental sulphur on a steel substrate in which the elemental sulphur functioned as an adhesive for the pyrite grains. Issues arose from this spraying method because the sprayed material was used as a cathode in a battery. The elemental sulphur would volatilize from the cathode and react exothermically with the anode, destroying the battery in the process.
To prevent this from happening Guidotti et al29 used different materials to co-plasma-spray with the pyrite grains. When using a LiCl-KCl electrolyte mixture together with Pyrite, better results were observed.
The plasma spraying of these experiments was executed at >9000°C, pyrite powder was fed to the plasma sprayer. The substrate used was a graphite paper named Grafoil®. The substrate was sprayed at roughly 5-10 cm distance. The spraying conditions were under normal atmospheric pressure and argon atmosphere to prevent sample oxidation. The formation of the spraying powder was synthesised by blending pyrite and the co-spray together at 600°C for 3 hours. The pyrite powder used had an average grain size of 44 micrometre and was purified by HCL before reacting and spraying.
Wang et al24 show their surface morphology after plasma spraying the FeS onto their substrate in figure 6. Opposed to other methods which left clear spikes or grains, the plasma-spraying method melts the particles down and create a surface with very broad spikes.
10 The grain size of the ‘’spikes’’ was roughly 1 micrometre, 1/40th of the starting size of the material. No further electronic or structural properties are known.
Guidotti et al29 have sprayed a mix of pyrite and LiCl-KCl onto a graphite surface (figure 7). The thickness of the coatings ranged from 50 to 150 micrometre. This method replaced the method of simply cold-pressing powders onto a substrate to form a cathode. Without going into the exact details of these processes the results of the plasma sprayed coatings were much better in terms of electric capacity (long low-voltage plateaus) but also adhesion of the coating. XRD analysis was executed and found the major phase of the coating to be Pyrite. However there is a difference in yield, the starting material contained pyrite (36-43% of the sprayed material) but the final material observed in figure 7 only contained about 20% pyrite. Guidotti et al29 argue this is due to pyrite particles bouncing off the substrate during the spraying procedure.
Figure 7: Plasma sprayed Pyrite & LiCl-KCl onto graphite surface
5. Hydro-solvo-thermal methods to synthesize FeS2 powder and nano-wires25,26
Wu et al25 used a Hydrothermal synthesis to form FeS
2 with very high purity in a 1:2 ratio. The reagents used were FeSO4*7H2O and Na2S2O3 these were added together with distilled water in a stainless steel autoclave. The autoclave ran for a range of times from 6 to 48 hours at temperatures ranging from 90 to 280°C. The products were cooled to room temperature and washed sequentially with distilled water, CS2 and dilute acid. The obtained slush was consecutively dried to a powder.
Wu et al25 have performed around 40 experimental runs at varying temperature and duration. XRD was used to analyse the crystallinity of the samples. A Rietveld refinement was executed to obtain insight in the % of marcasite vs. pyrite.
11
Figure 8: Pyrite crystals taken with an SEM25
It was found that at a reaction temperature of 200°C for a duration of 24h only FeS2 was synthesised(Figure 8), however, this material was only phase pure for 95%, meaning 95% of the material was pyrite while 5% yielded the orthorhombic version called marcasite. It seems though, compared to other preparation methods which in general did not achieve an exactly pure stoichiometry of Fe:S 1:2, the cost of achieving this ratio caused the unavoidable formation of a small amount of marcasite, leading to phase impurities opposed to crystal defects (which occur at a suboptimal stoichiometric ratio).
As can be observed in figure 8 one sees clear polycrystalline FeS2 grains of approximately 500 nm in size. The sample was put in a spectrometer to observe the absorption spectrum. It was found that this sample had an Edg of 1,38 eV according to qualitative determination utilising (ahv)2. It is not entirely clear to me if this is just the optical or actual electronic bandgap for their synthesised material.
12
2.3 Applications
Dye sensitised solar cell:
A dye sensitised solar cell also known as (DSSC) or Grätzel cell is a photovoltaic cell which converts light energy or photons into electrical energy. The cell is made up of different components and can be divided into three main sections: photoanode, photocathode and the electrolyte (figure 9). At the anode light falls through the conducting glass on the dye-coated TiO2. The photons excite electrons in the dye, these electrons are then transferred to the TiO2. The electrons can then flow through the conducting class into the external circuit (as a current which can be utilised for electric work) to the platina counter electrode. In the counter electrode the electrons are taken up by the redox couple which transfers them back to the dye, closing the cycle.30
Figure 9: schematic representation of a DSSC31
Shukla et al.30 found a way to use Pyrite as a replacement for the Pt (counter electrode) in figure 931. This is highly beneficial since pyrite is a very abundant, cheap and nontoxic material whereas platina is a very rare and expensive metal. The use of platina in many applications is not sustainable due to the cost and rarity of the material and thus, alternatives are sought for.
The redox potential of the redox mediator/couple is important for the maximum photovoltage of the cell. This directly relates to the used counter electrode. Where carbon based electrodes perform better with a cobalt redox mediator, platina counter electrodes perform best with a iodine redox mediator. There are alternatives available for counter electrodes which utilise the cobalt mediator but very few alternatives for counter electrodes are available in regard to the iodine mediator31.
Shukla et al.30 used spray pyrolysis by subsequent annealing to produce glass substrates with pyrite films. The have analysed the sample with many different techniques such as XRD, Raman, XPS, CV, SEM.
SEM images as shown in figure 10 (C) and (D) respectively are showing the morphologies of the pyrite film after spraying (C) and after annealing (D). These morphologies are similar to those of plasma-sprayed coatings however the grain size in figure 10 ranges from 30 to 50 nm30 opposed to around 1 micrometre for the plasma sprayed coating29.
13
Figure 10: Both SEM images: (C) showing surface morphology after spraying and (D) shows surface morphology after annealing30
Shukla et al.30 XRD and Raman (figure 11) analysis showed phase purity of FeS
2 to be very high, which can be seen from figure 11 A and B, both blue graphs only show strong peaks at the same degree(A) and wavenumber(B) for Pyrite formation and absence of marcasite formation.
Figure 11: (A) XRD pattern of the sample compared to database XRD patterns of pyrite (red) and marcasite (green). (B) Raman spectrum in blue, also set of to the database Raman patterns of pyrite (red) and marcasite (green).30
Since the pyrite is used as an electrode, the electronic and catalytic properties are very important, a Hall analysis was performed to analyse the electronic properties, which are shown in figure 12. literature predicts30 that pyrite crystals generally show N type behaviour while pyrite films show P type behaviour, the reason for this change in behaviour isn’t clear yet. The films created by Shukla et al.30 show highly conductive P type behaviour.
14 Beside the Hall analysis Shukla et al.30 also conducted a CV analysis (figure 13) for both the standard Pt electrode (red) and the FeS2 electrode (Blue). The top, positive half of the graph shows the oxidation reaction while the bottom, negative half shows the reduction reaction. As can be seen in figure 13, the CV diagrams overlap fairly well which shows that similar redox-behaviour is observed for the pyrite electrode compared to platina electrode.
Figure 13:Overlapping CV diagrams of Pt and FeS2 electrodes showing the oxidation and reduction reactions of the iodine
redox couple.
Figure 14 shows further analysis into the current density (A) and quantum efficiency (B) also yielded good results for pyrite. The current density for the pyrite film was found to be higher than the conventional Pt electrode, the quantum efficiency between roughly 425 nm and 700 nm is also higher for pyrite than for platina.
Figure 14: (A) shows a current density vs. potential plot of pyrite in blue and platina in red. (B) shows the quantum efficiency of the photovoltaic system at wavelengths comparable to the emission of the sun, again pyrite in blue and platina in red.
The overall efficiency of the pyrite electrode based DSSC was reported30 to be around 8%. For the platina electrode based DSSC this was found to be 7,5%. Shukla et al.30 largely attributes the positive pyrite efficiency to the good catalytic activity towards the iodine redox mediator.
15 Hydrogen Evolution Reaction (HER):
Hydrogen is an abundant element on earth, it is however usually not found as molecular hydrogen (H2) but rather bound in the form of H2O or hydrocarbons34,9. The natural abundance of H2 is very low because, due to the low molecular weight H2 is able to escape the earth’s gravitational field and is lost in space.34,9
Hydrogen is an important reagent for many chemical reactions such as for example the Haber-Bosch process which produces fertilisers34. Beside being a chemical reagent, H
2 is also an energy carrier and can be used as energy storage (basically store it as a fuel) or used as a fuel to burn and generate electricity or heat. If produced sustainably it has no carbon or nitrogen footprint and the only emission of the process will be water vapour, all these things make it into a very green and sustainable alternative for the replacement of fossil or biobased-fuels34.
The hydrogen evolution reaction is the electrochemical process of splitting water into hydrogen and oxygen utilising electricity also known as electrolysis. The process consists of two main redox reactions: (OER) or oxygen evolution reaction and (HER) hydrogen evolution reaction. The OER is the oxidation of water while HER is the reduction of hydrogen34,9.
This process requires electricity, the minimal theoretical voltage required to split H2O into H2 and 1/2O2 is 1,23V.33,34 However in practise this voltage lies higher due to the activation barriers of chemical reactions, resistance of the solution etc. This extra potential needed is called the overpotential. To split water as efficiently as possible one wants to keep the overpotential as low as possible. This can be achieved by finding good catalysts which can facilitate the HER or OER, lower the energy barrier of the reaction and therefore bring the practical potential closer to the theoretical value of 1,23V.
Miao et al33 have looked at pyrite as an electrocatalyst for the HER. Mio et al33 have synthesised Fe 2O3 utilising a sol-gel method followed by annealing under H2S atmosphere and elemental sulphur. This delivered meso-porous pyrite which, when used for HER only required a low overpotential of roughly 96 mV33 at a current density of 10mA/cm-2. It was also found to have very good stability during a 24h test in alkaline solution.
Figure 1533 shows various results of the (morphological) analysis of pyrite, executed by Mio et al33. Figure 1533 (A) shows the XRD powder pattern of the normal FeS
2 standard and synthesised mesoporous FeS2 (in red). Even thought the intensity is low and there is some noise, the characteristic FeS2 peaks are still quite visible and it is shown to be a crystalline material. The crystallinity of the mesoporous pyrite is less than that of the standard. Mio et al33 deems this to be because of the meso-porous nature, which might have destroyed some of the crystalline structure and thus results in being less crystalline.
Figure 1533 (B-C-D-E) are all SEM pictures of the meso-FeS
2 at various magnifications showing the grain size and distribution. Figure 1533 (F-G-H-I) are all EDX measurements and show the elemental composition of the pyrite particles, figure 1533 (G) shows the amount of iron, (H) shows sulphur and (I) shows both iron and sulphur.
16
Figure 15: (A) XRD of pyrite, (B-C-D-E) SEM images of the pyrite, (F-G-H-I) EDX elemental map of pyrite33
Figure 1633 (A&B) shows the HER behaviour of mesoporous pyrite in red, compared to different other materials. The system used was a three electrode setup at pH 13. The overpotential required at 10mA/cm-2 is found to be 96 mV. This is much better than the Ni based reference and the commercial pyrite reference. However the platina electrode with roughly 50 mV overpotential at a 10mA/cm-2 current density is still nearly twice as efficient.
Figure 16: (A) Current density vs potential plot of different catalysts, (B) required overpotential for HER for various catalysts33 According to Miao et al33 the results in Figure 1633 (B) show the presence of a fast Volmer-Heyrovsky mechanism. This means the hydrogen discharge reaction followed by the desorption of OH- is fast. This concludes to the desorption of H2 being the rate determining step in the reaction.
17
Figure 17: (A) shows a Nyquist plot which shows the charge transfer resistance. (B) shows the current density over a 24h period.33
Figure 1733 (A) shows a Nyquist plot, this plot reveals a huge difference in charge transfer resistance of the standard pyrite(238 Ohm) compared to the mesoporous synthesised pyrite (7 Ohm). This is directly related to the kinetics of the HER, which is much more favourable for the mesoporous pyrite. This is probably because the mass and charge transfer is better facilitated in the mesoporous material than the standard pyrite crystal structure/pores. Figure 1733 (B) shows the current density of the material over a 24h period of time under alkaline (13 pH) conditions at a 100 mV overpotential. There is no remarkable decrease, which indicates good catalytic stability under the aforementioned conditions. Miao et al33 have also carried out DTF calculations to get insight into the reaction mechanism of the HER. Miao et al33 have used their XRD results for the mesoporous and standard pyrite to compute part of the crystal structure which was used in the DFT calculation.
18 Figure 1833 shows the reaction pathway of both the standard pyrite (blue) and mesoporous pyrite (red). Computer simulations indicate the reaction starts with the adsorption of H2O on an Fe site forming an Fe-O bond, this bond is heavily polarised towards oxygen. In the next step we can see deformation of the water molecule and elongation of an O-H bond which finally results into the breaking of the elongated bond, splitting the H2O into a bound H+ and OH-.33 My personal assumption is (not shown in figure 18) that after multiple water atoms are split into H and OH the bound protons can, with appropriate current, form H2 gas which will desorb from the surface, forming your final product. In turn it’s possible for OH- to split off another hydrogen which will undergo the same process. Eventually the remaining oxygen atoms can desorb and form O2.
19 Pyrite as a photocatalyst:
Looking back at the DSSC30 or HER33,34,9 it’s clear that pyrite has catalytic activity and is suited to be used as a catalyst. In DSSC the pyrite acts as a counter electrode and catalyses the redox reaction of an iodine redox mediator while in the HER the pyrite acts as a hydrogen evolution catalyst. This part will be focused on the use of pyrite as a photocatalyst and shows how versatile the catalytic properties of pyrite are.
Photocatalysis is similar to other fields of catalysis but instead of for example adding energy in the form of heat to a reaction, photons (generally with a specific wavelength) are introduced to initiate the reaction or speed the reaction up. Some reactions require excited states in order to be reactive towards specific substrates or to undergo specific reactions. A good example of this are some Diels-alder reactions in which the diene and dienophile’s electronic states are mismatched to undergo a cyclo-addition. In this case photons can be used to excite electrons and make the reaction feasible36 without the use of a photocatalyst I would not be possible to execute some of these reactions due to forbidden transitions.
Looking at FeS2, some studies have been performed by Liu et al.35 and Bhar et al.13 which both look at the degradation of organic dyes, with the use of FeS2 and (UV) light.
Bhar et al.13 used a single source precursor to create a nanocrystalline iron sulphide thin film on a glass substrate. Their synthesis method is spread out over more than one paper and is a bit vague however their analysis holds good results.
The created films were polycrystalline and the crystal phase according to their XRD was orthorhombic FeS2, which means the structure isn’t pyrite but marcasite. Figure 19 shows the grainsize was roughly 25 nm with unevenly distributed pores.13
Figure 19: SEM image of marcasite phase on glass substrate, average grainsize of 25 nm.13
Figure 2013 shows the UV-vis absorption spectrum of the marcasite film. A sharp peak can be seen at 460 nm which corresponds to the optical bandgap energy of the material. The optical bandgap was calculated to be 2,7 eV13. This bandgap is higher than the bulk bandgap. According to Bhar et al.13 this has to do with the quantum confinement effect due to having relatively small crystals. This effect causes the bandgap to be larger for small crystals compared to bandgap of the material in homogeneous bulk.
20
Figure 20:UV-vis spectrum of the marcasite phase13
Nitrogen sorption studies were performed on some scraped off marcasite. The measured surface-area of the synthesised marcasite was 49,53 m2/g13. The surface area is important because it allows many substrates to attach to the surface at once, which in general leads to better catalytic properties. To test the catalytic properties of the marcasite, multiple experiments were performed. Rose Bengal (RB) dye was used and the absorption peak maxima decrease was studied at 546 nm light at 100W power. To compare the catalytic event, both a beaker with only dye & light was used (no catalyst present) and a beaker with both the dye & catalyst present (no light).
Figure 2113 shows a graph with the irradiation time compared to the relative presence of the dye, this graph was made by taking out small amounts of dye solution and measuring them in a spectrophotometer throughout the time the reaction ran for. Throughout a 300 minute experiment approximately 84% of the dye was degraded by the catalyst in combination with light, the other two experiments showed no degradation whatsoever. This shows clear photocatalytic activity for this specific reaction.13
21 Liu et al.35 used the solvothermal method which is comparable to the one mentioned in paragraph 2.2 (5.). This method yielded cubic FeS2 crystals also known as pyrite. According to Liu et al.35 their measurements yielded good crystalline pyrite without phase impurities (as confirmed by their XRD (A) & SEM (B) shown in figure 22). The average grainsize is not uniform and ranges from 0,6 to 1,5 micrometre (figure 22 (B)).35
Figure 22: (A) shows the XRD pattern set off to the internal standard for pyrite. (B) shows a SEM image of the synthesised pyrite35
For the catalysis part, Liu et al.35 have done something very similar to what Bhar et al.13 have done. Organic dyes were added in solution to the synthesised pyrite microcrystals and irradiated with light for a certain amount of time, the degradation of the dyes was studied and put into graphs shown in figure 23(A-E)13
22 The graphs shown in figure 2313 (A-E) all show a clear difference between a non-catalysed and catalysed solution. However the degradation ratios after reacting for 140 minutes are only between 56-88% for all the dyes. To improve the degradation the experiments were, after reacting for 140 minutes kept in the dark for 24h and degradation ratios of 97-99% were observed for all the dyes.13
The reason for the lower degradation after 140 minutes is due to the high surface area in combination with high adsorption of the dyes on the catalyst surface. This oversaturates the catalyst and blocks some of the light, required for the reaction. Excessive adsorption also creates a diffusion barrier which blocks products from leaving and reactants from entering. This is why according to Liu et al.35, when given 24 hours, nearly all the dye was degraded opposed to only a lower percentage after 140 minutes.13
23
3 Discussion & Future Prospects
Discussion:
In general research about pyrite as a semiconductor is very fragmented. Even though research into pyrite as a semiconducting material dates back30 to the 1970’s, it seems few articles really combine knowledge of different papers and research areas but rather focus on their own ‘’idea or application’’. Some articles took the time to very briefly review a variety of synthesis methods22,24,25,26 and based upon their findings and application, chose a suitable method. A variety of reasons were mentioned for choosing a synthesis method, such as the ease to perform the method21,25,26, because it was specifically suited for their mentioned application22,24,29,the phase purity of the formed product26 or because not a lot of research has been conducted in this method13,21,22. I personally find the latter a very bad reason to conduct research in an area, science is based on a well formulated hypothesis with experimental work to confirm this hypothesis. Simply fishing for good results to be able to publish something does not seem fruitful. However only some papers used this reason, others24,25 have carefully reviewed previously conducted research and based on that chosen a method.
Another general discussion point is the fact that papers only seem to apply some analytical technologies. To clarify, they only perform analysis which they deem useful, which leaves the reader with fragmented data. It is understandable that when one synthesises an oxidation catalyst, cyclic voltammetry is performed to review the oxidation and reduction reaction. On the other hand it’s also logical such an experiment is not conducted when investigating the influence of synthesis methods on the morphology, XRD and SEM is much more useful in this case. However, some analytical methods should be considered to be mandatory, no matter the application.
For pyrite a few parameters seem to stand out in importance, being: (I) phase purity, (II) stoichiometry, (III) surface morphology, (IV) bandgap, (V) crystallinity, (VI) surface area. Respectively these parameters can be analysed by: (I) XRD, (II) Stoichiometry of the synthesis or ICP-MS, (III) SEM/TEM, (IV) spectrometry, (VI) gas sorption experiments. There is not a single article which has collected all analytical results of the aforementioned parameters but they rather have a selection of some of them. This makes the comparison of different synthesis methods difficult.
The next part of the discussion will focus on the thesis in the order it is written. Looking at the semiconductor industry many materials have been synthesised over time and many more will be created in the future, although pyrite is a cheap, abundant and nontoxic material it will most likely not take over the market because it simply can’t match the properties of widely used silica semiconductors15,16. However this doesn’t mean pyrite is useless, it can find applications in specific areas which can utilise the properties of pyrite. In those areas it could replace currently used materials30,33,35.
As mentioned before both the synthesis and analytical research related to the morphology as well as the performance of pyrite in a variety of applications is fragmented. Almost all the papers21,22,26 do show that the morphology is key for the properties of the synthesised pyrite. The most important parameters are (I) phase purity, (II) stoichiometry, (III) surface morphology, (IV) bandgap, (V) crystallinity, (VI) surface area, (VII) Defects. I will shortly review why these parameters are so important.
Phase purity is important because the elemental composition of a crystal structure can be the same while the organisation of these elements differ, this is what’s called polymorphism. FeS2 can have multiple polymorphic forms, however most of the time it resides in either a cubic form known as pyrite
24 or orthorhombic form known as marcasite. Because the unit cell metrices differ for the different polymorphic forms, so do the macroscopic properties and thus having a phase pure material is important.
Stoichiometry and defects are closely related, where the stoichiometric amount differs, defects emerge. FeS2 has a 1:2 Fe:S ratio, when this is broken, the material won’t have a perfect crystal structure. Having a slight sulphur deficient material and thus a broken stoichiometric order does not necessarily have to mean the material is of bad quality. It depends entirely on the application if the defects and stoichiometric deficiency are wanted or not. It has to be mentioned that, in heterogeneous catalysis, often in a material the most catalytically active sites are the sites which have defects imbedded in them37.
Surface morphology is important for catalysis since it will heavily influence the diffusion of reactants/products and for photoactive materials since the morphology will decide how easily light gets absorbed or reflected. It’s also important to know the surface area and pore size and debt, this gives again information about the possible order of magnitude for the TON/TOF of your catalytic material and thus in turn how well your material might be suited for a catalytic application. The shape of the surface is also important for the aforementioned properties since shapes such as grains, needles, spikes, rods or balls all might have different photo, electronic or catalytic properties.
The bandgap gives information about the electronic properties of the material and is required for a multitude of applications such as HER, DSSC etc. Since these applications consist of electrochemistry (redox reactions in combination with an electric current) the Eg gives insight in how well the material is suited for the aforementioned concepts.
Crystallinity is also related to the surface morphology and defects but encapsulates the entire morphology of the material, through and through. When looking at a material it’s important to know the composition on the inside as well since materials might be milled, processed, crushed, damaged and reveal material that previously wasn’t on the surface. Also properties such as the bandgap and conductivity relate to the entire material in bulk and not just to the surface where catalytic processes take place.
A large surface area makes a material suited as a catalyst due to the large accessibility of catalytic active sites. It also increases the amount of defects (in absolute values) which in turn often increases catalytic performance.
All the aforementioned properties come forth out of a different synthesis methods such as: sol-gel method, plasma spraying, hydrothermal method, annealing etc. However there is not one single ‘’magical’’ synthesis method which works for everything. From the mentioned papers and their wide variety of synthesis methods it seems that depending on the characteristics you want your material to have, to best suit the application you design, you will need to search for a specific synthesis method. This will delivers these best combination of properties. In practise this will most likely come down to a compromise between the different parameters, selecting the ones which are deemed most important for your selected application. I personally think, this is also the reason why the research is somewhat chaotic and fragmented, each research group goes through the aforementioned process and makes a selection, they should however give more insight in the ‘’how’’ and ‘’why’’ because in the end, sharing of information in a logical an reproducible way is what science is about.
25 The application of pyrite as the counter electrode of DSSC’s seems like a very feasible alternative based on the results of Shukla et al.30. The current use of platina is not sustainable due to it being a very expensive and limited resource. Shukla et al.30 results show pyrite to exhibit similar redox behaviour, better current density, better quantum efficiency and better overall cell efficiency. However pyrite cannot fully replace all materials in a DSSC, as photovoltaic crystal that material underperforms compared to many conventional materials32. It is nonetheless a good option to replace the platina-electrode.
As for the HER pyrite doesn’t offer the same efficiency as platina but it does seem like a feasible alternative compared to other carbon based materials33. The cost & performance of the equipment with platina required for a HER should be set off versus the cost & performance of the equipment with pyrite. Depending on the performance requirements a choice could be made for one or the other. Also in light of the current switch to sustainable chemistry and green chemistry, utilising a non-toxic, cheap and abundant material such as pyrite to produce a renewable energy resource (H2) seems certainly appealing.
As for the photocatalytic part, I personally don’t see how the degradation of dyes is a major catalytic application. Being able to degrade biological compounds could be of use in for example wastewater treatment plants (WWTP) but no clear examples are made in which a pyrite catalyst is capable of photo catalytically converting wastewater to cleaner water13,35. The papers are also a bit vague, Bhar et al.13 describes the use of marcasite for the photocatalytic degradation of RB dye. While Liu et al.35 looks at the photocatalytic degradation of a variety of dyes using pyrite. There is some discussion related to the diffusion which causes a performance lag but it’s not very extensively discussed nor compared. In general more research would be needed in this area to be able to draw a concise conclusion.
Future Prospects:
Regarding the morphology it’s very much needed to have a review article which combines around 10 different commonly used synthesis methods and reviews the material properties (e.g. I-VIII in the discussion) with extensive analytical results. Having an overview of properties offset vs. the synthesis makes it possible for scientists to more accurately choose a method suited for their application rather than making guesses based on incomplete information spread out over many papers.
DSSC’s pyrite electrode performance is exceptional which begs the question if electrodes in other applications could be replaced by pyrite as well. This is a complicated issue since this specific electrode is used in mediating an iodine redox couple, for any other reaction this superior performance might not hold. Nonetheless it would be interesting to see a variety of reactions (other than HER) and compare performances of pyrite and conventional electrodes.
HER performance of pyrite is good but not exceptional, this begs the question if further changes in material properties (e.g. I-VII in the discussion) could bring this material closer to the values observed for platina. Further research will have to be conducted in order to finetune the material to perfection. The photocatalysis part leaves a lot of questions, which applications can utilise the photocatalytic capability of pyrite? If there is activity towards the degradation of a variety of organic molecules it might be possible to utilise this performance to degrade pathogens in the waste water of WWTP’s. It would certainly be interesting to see if something like this is possible. Also the influence of material properties (e.g. I-VII in the discussion) are not very well investigated and might be able to seriously improve the photocatalytic behaviour of pyrite or marcasite. Further research in this area will improve the knowledge and answer these burning questions.
26
4 Conclusion
Overall research papers have fragmented analytical results and synthesis methods, this is partly because to publish something papers want something ‘’new’’ and partly because for each application, other parameters are important. Thus a compromise between parameters is sought for in order to come up with the best results. However this means it’s hard to draw conclusions based on the comparison of methods since not all information is available to be compared.
Research into pyrite as a semiconducting material dates back to the 1970’s but more recently it gained increased interest due to the uprising of green chemistry and environmental friendly chemistry. Pyrite is an abundant, nontoxic and cheap material with semiconductor properties and thus could be used as replacement for expensive and/or toxic materials. The O-life project has also contributed to the intensified research into pyrite.
Pyrite(FeS2) has a theoretical bandgap(Eg) of 0,95 eV which is similar to the bandgap of silica. Pyrite also has a very high absorption coefficient of 5*105 cm-1 for wavelengths smaller than 700 nm. The way pyrite is synthesised directly relates to the final properties of the material. In summary the most important (synthetic) properties of pyrite are: (I) phase purity, (II) stoichiometry, (III) surface morphology, (IV) bandgap, (V) crystallinity, (VI) surface area, (VII) Defects. These properties change depending on the reaction conditions and synthesis methods used.
There isn’t one universal method which gives the ‘’best’’ universal properties but rather a set of methods such as for example sol-gel method, hydrothermal method or plasma spraying which give properties suited for specific application such as the use of pyrite for HER, in the electrode of a DSSC or as photocatalyst for the degradation of organic dyes.
FeSx has the capability to change certain areas of the semiconductor industry, the possibility to replace platina based electrodes in certain scenarios is exciting. The use of pyrite for water splitting or degradation of organic compounds is still a novel concept but would certainly be a valuable option to further dive into.
Since the material properties of FeSx are sensitive to a large amount of parameters it will take more research in order to synthesise and finetune FeSx to be perfectly suited for a specific application. This in turn means that pyrite does not yield the power to be widely applicable for a large area of semiconductors but It does show that FeSx can certainly be a powerful and sustainable replacement for targeted industrial semiconductor applications.
27
5 References
(1) Origin of Life: Iron-sulfur world. (2019). Retrieved 29 September 2019, from https://biopoiesis.blogspot.com/2006/10/iron-sulfur-world.html
(2) Rationale of oLife’s research programme. (2019). Retrieved 29 September 2019, from
http://www.olife-programme.eu/rationale-of-olifes-research-programme/
(3) Anastas, P., & Crabtree, R. (2009). Handbook of green chemistry. Weinheim: Wiley-VCH. (4) Marteel-Parrish, A. (2014). Green chemistry and engineering. New York: John Wiley & Sons.
(5) wetten.nl - Regeling - Protocol van Kyoto bij het Raamverdrag van de Verenigde Naties inzake klimaatverandering, Kyoto, 11-12-1997 - BWBV0001538. (2019). Retrieved 29 September 2019, from
https://wetten.overheid.nl/BWBV0001538/2005-02-16
(6) UNITED NATIONS UN Climate Change Summit 2019. (2019). Retrieved 29 September 2019, from
https://www.un.org/en/climatechange/
(7) Krot, A., Keil, K., Scott, E., Goodrich, C., & Weisberg, M. (2007). Classification of Meteorites. Treatise On
Geochemistry, 1-52. doi: 10.1016/b0-08-043751-6/01062-8
(8) Percak-Dennett, E., He, S., Converse, B., Konishi, H., Xu, H., & Corcoran, A. et al. (2017). Microbial acceleration of aerobic pyrite oxidation at circumneutral pH. Geobiology, 15(5), 690-703. doi: 10.1111/gbi.12241
(9) Hydrogen Evolution Reaction. (2019). Retrieved 30 September 2019, from
http://large.stanford.edu/courses/2012/ph240/wang-ha1/
(10) Miao, R., Dutta, B., Sahoo, S., He, J., Zhong, W., & Cetegen, S. et al. (2017). Mesoporous Iron Sulfide for Highly Efficient Electrocatalytic Hydrogen Evolution. Journal Of The American Chemical Society, 139(39), 13604-13607. doi: 10.1021/jacs.7b07044
(11) Mineral Data Publ. (2000). Handbook of mineralogy. Tucson, Ariz.
(12) Heift, D. (2019). Iron Sulfide Materials: Catalysts for Electrochemical Hydrogen Evolution. Inorganics,
7(6), 75. doi: 10.3390/inorganics7060075
(13) Bhar, S., Jana, S., Mondal, A., & Mukherjee, N. (2013). Photocatalytic degradation of organic dye on porous iron sulfide film surface. Journal Of Colloid And Interface Science, 393, 286-290. doi: 10.1016/j.jcis.2012.10.049
(14) Pyrite Meaning. (2019). Retrieved 7 October 2019, from
https://thehealingchest.com/crystals-stones/pyrite-meaning/
(15) Physics and Properties of Semiconductors-A Review. (2006). Physics Of Semiconductor Devices, 5-75. doi: 10.1002/9780470068328.ch1
(16) Pleskov, Y.V., & Gurevich, Y.Y. Semiconductor photoelectrochemistry. United States.
(17) Semiconductors and Silicon Solar Cells. (2019). Retrieved 9 October 2019, from http://butane.chem.uiuc.edu/pshapley/Environmental/L9/2.html
(18) Ellmer, K., & Höpfner, C. (1997). On the stoichiometry of the semiconductor pyrite (FeS2). Philosophical
Magazine A, 75(4), 1129-1151. doi: 10.1080/01418619708214015
(19) Nakamura, S., & Yamamoto, A. (2001). Electrodeposition of pyrite(FeS2) thin films for photovoltaic cells.
Solar Energy Materials And Solar Cells, 65(1-4), 79-85. doi: 10.1016/s0927-0248(00)00080-5
(20) CédricTard. Proton-coupledelectrontransfer: relays,bondbreakingandcatalysis. Other. Univeristé Paris Diderot Paris 7; Sorbonne Paris Cité, 2015.
(21) Duan, H., Zheng, Y., Dong, Y., Zhang, X., & Sun, Y. (2004). Pyrite (FeS2) films prepared via sol–gel hydrothermal method combined with electrophoretic deposition (EPD). Materials Research Bulletin,
39(12), 1861-1868. doi: 10.1016/j.materresbull.2004.06.012
(22) Wan, D., Wang, Y., Zhou, Z., Yang, G., Wang, B., & Wei, L. (2005). Fabrication of the ordered FeS2 (pyrite) nanowire arrays in anodic aluminum oxide. Materials Science And Engineering: B, 122(2), 156-159. doi: 10.1016/j.mseb.2005.05.003
(23) Bessergenev, V., Pereira, R., & Botelho do Rego, A. (2007). Thin film sulphides and oxides of 3d metals prepared from complex precursors by CVD. Surface And Coatings Technology, 201(22-23), 9141-9145. doi: 10.1016/j.surfcoat.2007.05.017
28
(24) Wang, H., Xu, B., Liu, J., Zhuang, D., Wei, S., & Jin, G. (2007). The iron sulfide coatings prepared by plasma spraying and their friction-reduction performance. Surface And Coatings Technology, 201(9-11), 5286-5289. doi: 10.1016/j.surfcoat.2006.07.229
(25) Wu, R., Zheng, Y., Zhang, X., Sun, Y., Xu, J., & Jian, J. (2004). Hydrothermal synthesis and crystal structure of pyrite. Journal Of Crystal Growth, 266(4), 523-527. doi: 10.1016/j.jcrysgro.2004.02.020
(26) Kar, S., & Chaudhuri, S. (2005). Synthesis of highly oriented iron sulfide nanowires through solvothermal process. Materials Letters, 59(2-3), 289-292. doi: 10.1016/j.matlet.2004.10.005
(27) Kawai, S., Yamazaki, R., Sobue, S., Okuno, E., & Ichimura, M. (2014). Electrochemical deposition of iron sulfide thin films and heterojunction diodes with zinc oxide. APL Materials, 2(3), 032110. doi: 10.1063/1.4869035
(28) Yu, J., & Liu, W. (2012). The Study of Pyrite Surface Properties and Effect Mechanism with Xanthate Caused by the Microscopic Crystal Structure and Defects. Advanced Materials Research, 538-541, 363-367. doi: 10.4028/www.scientific.net/amr.538-541.363
(29) Guidotti, Ronald & Reinhardt, Frederick & Dai, Jinxiang & Roth, Jeff & Reisner, David. (2002). Characterization of Plasma-Sprayed Pyrite/Electrolyte Composite Cathodes for Thermal Batteries. Journal of New Materials for Electrochemical Systems. 5.
(30) Shukla, S., Loc, N., Boix, P., Koh, T., Prabhakar, R., & Mulmudi, H. et al. (2014). Iron Pyrite Thin Film Counter Electrodes for Dye-Sensitized Solar Cells: High Efficiency for Iodine and Cobalt Redox Electrolyte Cells. ACS Nano, 8(10), 10597-10605. doi: 10.1021/nn5040982
(31) Kushwaha, R., Srivastava, P., & Bahadur, L. (2013). Natural Pigments from Plants Used as Sensitizers for TiO2Based Dye-Sensitized Solar Cells. Journal Of Energy, 2013, 1-8. doi: 10.1155/2013/654953 (32) Steinhagen, C., Harvey, T., Stolle, C., Harris, J., & Korgel, B. (2012). Pyrite Nanocrystal Solar Cells:
Promising, or Fool’s Gold?. The Journal Of Physical Chemistry Letters, 3(17), 2352-2356. doi: 10.1021/jz301023c
(33) Miao, R., Dutta, B., Sahoo, S., He, J., Zhong, W., & Cetegen, S. et al. (2017). Mesoporous Iron Sulfide for Highly Efficient Electrocatalytic Hydrogen Evolution. Journal Of The American Chemical Society, 139(39), 13604-13607. doi: 10.1021/jacs.7b07044
(34) Heift, D. (2019). Iron Sulfide Materials: Catalysts for Electrochemical Hydrogen Evolution. Inorganics,
7(6), 75. doi: 10.3390/inorganics7060075
(35) Liu, S., Li, M., Li, S., Li, H., & Yan, L. (2013). Synthesis and adsorption/photocatalysis performance of pyrite FeS2. Applied Surface Science, 268, 213-217. doi: 10.1016/j.apsusc.2012.12.061
(36) Lin, S., Padilla, C., Ischay, M., & Yoon, T. (2012). Visible light photocatalysis of intramolecular radical cation Diels–Alder cycloadditions. Tetrahedron Letters, 53(24), 3073-3076. doi: 10.1016/j.tetlet.2012.04.021
(37) Schlögl, R. (2015). Heterogeneous Catalysis. Angewandte Chemie International Edition, 54(11), 3465-3520. doi: 10.1002/anie.201410738
