1.   Introduction of Electrochemistry

A fuel cell is based on a rather simple design (hence the popularity of use in various industries) and is made up of multiple components pressed together to form the entire unit – commonly referred to as the Membrane Electrode Assembly or MEA. The key components that a fuel cell is made up off includes but is not limited to: gas diffusion electrodes (GDEs) – alternatively referred to as the anode and cathode chambers; electrolyte – ion carrier/transporter from one electrode to the other (solid or liquid); electrically conducting wire; electro-catalyst – catalyst used to speed up one or more half reactions; the gasses used – although not technically a component of the fuel cell, they are required in order to function. The focus of this research lies with the electro-catalyst. The basic principle of a fuel cell is converting chemical energy into electricity. This is accomplished by a REDOX reaction – An electron (e^-) is extracted from the alcohol (fuel – typically hydrogen based) and passed through an external circuit, generating a minor current, and flows to the other GDE where it combines with oxygen. The site at which both extraction and attachment of the electron occurs, is the catalytic surface – usually platinum. The by-product formed in this process is water – made up of the oxygen present in the cathode bonding with the H⁺ ions passing through the electrolyte and e^- through the external circuit ^[1]. This is explained in greater detail in section 2.

 2.   Literature Review

a.    Basics of a Fuel Cell of Electrochemistry
                                                             i.      Overview (Components and Design) of Electrochemistry

It has been observed that the underlying technology and components of the Breathalyzer have not been updated for decades and the current generation of units are still based on 1970s technology.^[2] This design involves a high use of platinum as the electro-catalyst, a very expensive metal. Due to many important advancements in technology since then, it is believed that a far cheaper alcohol gas sensor can be designed, minimizing the platinum concentration or replacing it completely and thus making a more feasible sensor without inhibiting sensitivity, reliability, and durability.^[3] This will allow for them to be readily and/or easily available in low/middle income countries that make up for 90% of the world’s road accidents ^[4]. Palladium has shown to be a viable electro-catalyst. According to the study performed by Holton and Stevenson ^[5], palladium in its pure form was observed to only be marginally inferior as a bonding agent to oxygen than platinum. It was also observed that bi-metallic compounds could potentially be suitable electro-catalysts if the core metal itself was viable. Thus, it is proposed that a compound containing palladium as its core, and possibly supported by ceramics or carbon nanofibers, could theoretically replace platinum as the leading electro-catalyst. This would drastically reduce both the manufacturing cost as well as the maintenance cost of the Alcohol Gas Fuel Cell Sensor (AGFCS) unit.

Although many other forms of alcohol detection such as gas chromatography, infrared, and semi-conductor ^[6,7,8] techniques are commercially available, breath ethanol detection is usually performed by fuel cell sensors owing to their linearity, accuracy, sensitivity, size, rapid response time, and cost. ^[9]

Several different kinds of Fuel cells exist as shown in Figure 1. There are many ways to categorize these devices, but the simplest way is looking at the reactant fuel and/or the electrolyte used ^[10]. Each type of fuel cell has their own advantages and disadvantages and are chosen based on the need and working conditions of the user. The alkaline fuel cell is best suited for the objectives this research is set out to achieve – due to the alcohol reactant and temperature range.

An example of the mechanism and theory behind the alkaline fuel cell is shown in Figure 2 Since the by-product is water, fuel cells prove to be a great source of energy generation and chemical sensors. The electrocatalyst studied in this research will reside on the anode while the cathodic electrocatalyst will be comprised of some derivative of iron and cobalt (Future work).

The underlying purpose of all fuel cells is to generate a current. The current density (the amount of electric current flowing per unit cross-sectional area of a material) is determined by the concentration and/or amount of reactants that react. The faster the reaction rates of both AOR and ORR, the greater the expected current density^[11]. For industrial based fuel cells, a high current density (relative to the amount of material in the electrode) is sought after. This is because the primary purpose of industrial fuel cells is current generation. The aim is to generate current from chemical energy (label 3 in Figure 2).

In this research, although current generation is a necessary step, the aim is not to achieve a high charge density nor utilize this electrical energy for other uses but rather to simply measure the electricity generated as it is an indication of the concentration of alcohol. Here sensitivity plays a rather important role as opposed to raw power output. The current and charge densities would be plotted against the present breath alcohol concentrations (BrAC). The slope produced by these curves would indicate the catalyst’s sensitivity and combined with the regression of the curves, the linearity could be calculated. Once the required formulae are derived, they can be used to translate the current generated when used by the subject into the appropriate BrAC. This concentration of alcohol in the breath can then be used to determine the amount of alcohol present in the user’s blood and thus one can determine if the user is above the alcohol limit or not. A simplified schematic diagram of the process is shown in Figure 3. The displayed number is related to the user’s alcohol breath, and therefore blood, concentration. Based on Henry’s law, the ratio of exhaled alcohol to blood alcohol is 2100:1. More simply, if x amount of alcohol exists in 2100 millilitres of exhaled air, then x amount of alcohol exists in 1 millilitre of blood. 

1.1.1      Alkaline vs Acidic Electrolyte of Electrochemistry

Fuel cells are sometimes categorized not by the reactants but rather by the electrolyte used. The electrolyte is responsible for transporting the necessary ion from one electrode to the other. This is a purely concentration-based diffusion step. Should the electrolyte fail to transport the ion (H⁺ in most cases but could potentially be OH^- in an alkaline fuel cell or even O₂), the whole process would be at a standstill and reactants would build up in their respective electrodes. This selective transportation of molecules thus has a significant impact on the functionality of the cell and links to reaction rates, current density generated, and peak current generated. ^[10]

Electrolytes have primarily, in the past, been liquids. However, due to advancements in technology as well as the need to have mobile fuel cells, solid electrolytes are now the preferred medium ^[11]. Nafion can help attach the catalyst layer to the membrane, and it also helps increase the ionic conductivity of the catalyst layer ^[11]. According to H₂/air PEMFC testing, Nafion 112 (one of the many industrially used forms of Nafion) has reached a maximum lifetime of just over 10,000 hours - a significantly acceptable number of cycles. ^[13]

The research focuses on the electrocatalyst involved in the PEMFC production and would thus produce results that are comparable to other catalysts, not other electrolytes. However, considering that multiple researchers also used a version of Nafion as their electrolyte. Nafion is likely the most viable material for comparative reasons in order for this research to minimize variations. ^{[5, 11, 13]}

Table 1: Acidic vs Alkaline Electrolytes

 [https://www.doityourself.com/stry/a-complete-guide-to-alkaline-fuel-cell-electrolytes] and PEMFC textbook

1.1.2      Reaction Kinetics of Electrochemistry

According to Breeze et al (2009), two half (redox) reactions occur in two separate regions of the fuel cell. The cell is primarily comprised of two separate chambers called electrodes – The anode and the cathode. In each of these chambers, a “half-reaction” occurs, i.e. the products of one and reactants of the other are intermediates and both reactions combined produce a stable overall reaction.

1.1.2.1       Anode Half Reactions of Electrochemistry

A reactive fuel (usually hydrogen gas or some alcohol) enters the anode. At the anode, AOR occurs. The basic pure hydrogen-based reaction occurring the anode indicated by Zhang in Equation 1 can be adapted to fit primary alcohols as shown in Equation 2^[11]. This initial reaction converts the primary alcohol into an aldehyde, however this is not necessarily the end, the recently formed aldehyde can once again be oxidised into a carboxylic acid: this is dependent on the strength of the oxidising agent. If this occurs, the reaction process would proceed as shown in Equation 3.

It should be noted that for the example above, water was assumed to be present in the fuel cell as the human breathe would contain moisture and can be modelled as saturated air - 100% Relative Humidity (RH) - at 32^oC as stated by Wilamed^[12]. The water found here should not be confused with the water exiting the fuel cell at the cathode side, which is a by-product of the ORR.

1.1.2.2       Cathode Half Reactions of Electrochemistry

Air (assuming 21 volume % oxygen) or pure oxygen gas is fed into the cathode. At the Cathode, ORR occurs. Oxygen is broken down into individual oxygen atoms and adheres to the catalyst surface. The flowing hydrogen and electron then merge with these surface oxygen atoms/ions to form a hydroxide and then pure water - a safe by-product. The mechanism of this reaction is Taken from Nickle article

The ORR could potentially experience minor complications. This is due to high-energy reaction kinetics / pathways and the relatively strong O-O bond in pure oxygen, not all the O₂ molecules break on the catalyst surface. This could result in a small concentration of H₂O₂ forming in the cathodic chamber or electrode. This is a highly corrosive compound and fortunately has a much smaller selectivity (the probability of a reaction to form the compound of interest against all possible products). The better the catalyst, the higher the selectivity of H₂O ^[5]. A basic diagram can illustrate the possible alternate pathways oxygen gas may react through and is shown in Figure 4.

Although ruthenium (Ru) and iridium (Ir) are the most efficient catalysts for oxygen evolution (refer to Figure 7), platinum (Pt) is regarded as the best catalyst for both hydrogen evolution as well as oxygen reduction. ^{[nickle article]}Even though Pt is the best for ORR, it is still not at a significantly remarkable level currently. Efforts should be made to enhance the ORR process significantly in the future.

1.1.1      Mass Transfer of Electrochemistry

A critical aspect in the efficiency and overall functioning of a fuel is the mass transfer (MT) stage. Mass transfer is achieved in multiple forms and include; Diffusion (region A) and Convection (Region B). The basic principal behind this is like that of heat transfer in the sense that materials (compounds) will move from a high state of energy (indicated by a large concentration since molecular interactions are more significant) to a low state. This is commonly referred to as the concentration gradient. Thus, the two state conditions effecting MT is the concentration of the bulk material and concentration at the surface of the electrode where the reaction occurs. It is also affected by temperature and pressure gradients but that is less impactful and can be assumed negligible for our research. The formula depicting MT is known as Fick’s Law and is as follows;

Where;

J_i = Mole flux (kg.mol/m².s)

D_ij= Diffusion coefficient (for two species)

∙∇C_i = concentration gradient (kg.mol/m³)

For multi-component diffusion systems (once again neglecting the pressure and temperature contributions), the Stephan-Maxwell equation can be used and is depicted as;

Where;

J_i = Mole flux (kg.mol/m².s)

C_i = Concentration (kg.mol/m³)

D_ij= Diffusion coefficient (for two species)

X = Mole fractions (unitless)

Convective MT would take place as a bulk flow is initiated – in this case, the channel from the anode and cathode chambers (see figure below), dragging species regardless of the concentration differences.

1.1.1      Fuel Cell Efficiency of Electrochemistry

Fuel Cell efficiency, just like any other is based off the ratio between the useful energy output against the energy input. Note that the actual efficiency of the system will either be equal to or lower than the theoretical efficiency calculated. For the fuel cell in general (and for this research as an alcohol gas sensing micro fuel cell), the useful output energy is the electrical energy produced in the external circuit explained above. The energy input is based of the chemical potential energy of the reactants – enthalpy of hydrogen gas. We can simplify the calculations by assuming that all the Gibbs-free energy can be converted into electrical energy without any waste. Since both the enthalpy and Gibbs-free energy are known, the maximum theoretical efficiency of a fuel cell at STP is;

he Gibbs-free potential and enthalpy potential can be obtained by dividing the respective term by the product of n.F. This yields values of 1.23V and 1.48V for the theoretical cell potential and hydrogen’s Higher Heating Value (HHV) – aka thermoneutral potential respectively. PEMFC plotted the efficiency of a FC against temperature (based on HHV) and is shown below. At higher operating temperatures for the FC, the efficiency is lower and could be attributed to wasted energy going into the formation of gaseous products for one. This provides justification to attempt operating at lower temperatures.

The above equation can be specified to any FC by inputting the measured cell potential (V_cell)

OR

However, considering that 100% conversion of the inputted fuel is not realistically achieved, the conversion factor must be considered to accurately determine the efficiency. Thus, the final equation is as follows;

Where;

1.1      World’s Leading Catalysts of Electrochemistry
1.1.1      Platinum of Electrochemistry

The catalyst primarily has two functions to perform without any change to itself (i.e. no leaching or degradation of any kind):

1)      The catalyst adsorbs the elements that need undergo oxidation and/or reduction and hold them for a short period of time.

2)      Once the molecules/elements react, the catalyst releases the product back into the Gas Diffusion Electrode (GDE).

Platinum has the best fit to both these functions. Platinum is a transition element that could easily form bonds with the hydrogen and oxygen molecules present in the GDE. However, some materials such as titanium and iridium form even more stable bonds yet are not considered. This is due to the second function of the catalyst, energy is needed to break up the bonds between catalyst and the given species – a bond too strong would require too much energy to break and would thus result in an accumulation of oxygen and/or hydrogen on the respective electro-catalyst ^[5]. Plots (known as volcano diagrams) obtained from Holton and Stevenson (2013) can be seen in Figure 7 and Figure 8^[5]. A simple explanation is that the top of the peak is the best catalyst, those elements on the left bond too weakly (does not satisfy condition one above) while those on the right bond too strongly (does not satisfy condition two above). Conversely, bonds too weak would not be able to maintain the catalyst-hydrogen/oxygen bond and thus the reaction would proceed very slowly or not at all.

1.1.2        The Need for Alternatives of Electrochemistry

The electrochemical-based fuel cell sensor was introduced in the 1970s ^{[14, 15]}. According to Modjtahediet al.^[3], the current generation of breathalysers commercially used today is still based on the 1970s technology. Although this technology is acceptable to measure the gas concentration of an individual in terms of sensitivity, the manufacturing costs incurred during the production of these units are very high with Pt accounting for nearly 40% of the total production cost. With the advancements in technology made over nearly half a century since then, the various components that make up a fuel cell are being re-examined to design a more feasible option – note that these advancements will also aid in sensitivity, durability and many other factors that will be discussed in further detail below.

As a result, the electro-catalyst (made of nearly 100% platinum) is currently under much scrutiny. Platinum is a costly material driving up manufacturing costs and is significantly poisoned on its surface by strongly adsorbed organic material – the reactant. This results in rapid degradation of the platinum catalyst and in turn requires regular replacement further adding to the costs ^[16].

1.2      Alternate Catalysts of Electrochemistry

There are various factors that go into consideration when a system requires the “best catalyst”. The needs of the system need to be weighed. What is more important? Cheap production cost or raw performance? Sensitivity of the readings or durability of the device? In this sense, talking about an optimum catalyst is more apt than mentioning the best catalyst. Even though platinum reigns superior in most regard, there are other catalysts well worth being studied as possible replacements. The 2 major categories being production cost and durability. Looking at Figure 7, it can be seen that 3 other elements reside near platinum. Another group involving nickel, cobalt, and iron depict slightly lower current densities for hydrogen evolution at a significantly lower adsorption energy of the metal-hydrogen bond.  

1.1.1      Palladium (Pd) of Electrochemistry

Research for an alternative electrocatalyst is already underway. Many academics have performed individual AOR and ORR on multiple pure elements and the results from Holton and Stevenson^[5]are displayed in figure 8 and 9 respectively. The closer the element is to the peak, the greater its suitability to be an electro-catalyst. It is worth mentioning that elements to the left of the peak bond too weakly and elements on the right bond too strongly. Based on the grouping of Platinum Group Metals (PGMs) near the peak on Figure 8and Palladium far exceeding the potential of the other elements in Figure 8, Palladium – a PGM – has been chosen as the core metal electro-catalyst for this project. It has similar electrochemical and physical properties as platinum but is denser and much more affordable. Palladium can prevalently oxidize most alcohols to their corresponding carboxylic acid – the exception to this is methanol which proceeds through intermediary of formate (CHOO^-) or carbon monoxide (CO). ^[1

The Volcano diagram in Figure 8 shows Palladium as a great alternative as previously mentioned. However due to Nickel’s incredible conductivity and ease of synchronisation as an alloy, it has seen countless studies as a filler catalyst. It also boasts great results on the volcano plots, but previous studies suggest that merely alloying the metals near the top is not sufficient, as nickel has already proven to compete with copper and silver as platinum’s core-shell.

1.1.2      Nickel (Ni) – very good paper for pt/Ni

With regards to water-splitting, a catalyst that would perform hydrogen evolution and oxygen evolution in the same electrolyte is desired. This would result in a simpler system design, improved practical applications, and lower cost. Ni is by far the most efficient bifunctional element used to catalyse this reaction in basic media^[n1,n2]. So far multiple Ni-based catalyst have already been developed. The downfall of these catalysts resides in the fact that they are not very stable and/or durable. Merging these catalysts in order to eliminate these shortcomings produce new challenges such as lower electrical conductivity and thus water splitting catalysts remain a huge challenge to this day.^[n]

[n] concluded that Ni-based electrocatalysts have been extensively studied in the past and introducing other transition metals as an alloy has been established as the most promising approach in enhancing the number of active sites for the desired reactions. However, despite very high catalytic efficiencies and thermal stabilities of these alloys, the stability under highly concentrated electrolyte media proves to be a setback – This is something that needs to be resolved by enhancing our support material.

Many scientists propose to replace Pt with Ni due to similar chemical properties, identical group number on the periodic table of elements, and the abundance of Ni found on earth which results in a significantly lower cost.[^n3]

Source:

[n1] = Ahn, S. H.; Manthiram, A. J. Mater. Chem. A 2017, 5, 2496-2503.

[N2] = You, B.; Jiang, N.; Sheng, M.; Bhushan, M. W.; Sun, Y. ACS Catal. 2015, 6, 714-721.

[n3] Furuya, N.; Motoo, S. J. Electroanal. Chem. Interfacial Electrochem. 1978, 88, 151-160.

[most other things] = Nickel–Based Electrocatalysts for Energy Related Applications: Oxygen

Reduction, Oxygen Evolution, and Hydrogen Evolution Reactions

1.2      Support Materials
1.2.1      Carbon Black of Electrochemistry

Previous reports have studied the (intrinsic) electronic interaction between the catalyst and the carbon support to determine the potential influence on methanol ^{[19, 20]}. The binding energies of the 4f orbital for platinum and palladium were found to be higher for smaller sparticle sizes - in the case of Pt it was ~71.6eV for clusters and particles and ~71.1eV for the bulk catalyst. The presence of carbon weakens the 4f electron binding making it more inclined to interact with and/or adsorb organic compounds. Heating the Pt/C catalyst also increased the 4f binding energy from 71.3eV to 72.2eV (673K for 4 hours). This is a result of the metal catalyst donating an e^- to the carbon support, lowering the overall activity of the system. Compared to the usual carbon black electrocatalysts, the nanotube-supported Pt catalyst resulted in up to 140% improvement in efficiency for a proton exchange membrane fuel cell (PEMFC) ^[21]. It had also been shown that the performance of a carbon nanotubes (CNT)-based MEA is superior to that of conventional Pt/C MEA ^[22]. Thus other derivatives of carbon should be tested. Although a high surface area would prove useful on many fronts, [fd] showed that electrochemical stability is often sacrificed for this cause.^[cb1]Thus, a balanced surface area would be viable for both a stable yet conductive catalyst support.

1.1.1      Onion-Like Carbon of Electrochemistry

Carbon onions or Onion-like carbons (OLCs) are quasi-spherical nanoparticles ranging from 5-10nm. They consist of concentric graphite shells which enclose fullerene-like carbons. Due to their highly symmetrical nature, they host different properties compared to more commonly known carbon derivatives such as nano-diamond, carbon tubes, vulcan carbon, and nanotubes amongst others.^[x1]They have a range of applications from magnetic recording systems to aiding in the attachment of desirable functional groups^[x2]. Their lack of interest in commercial research was due to the inability to mass produce OLCs – However, recently viable methods such as the heating of nano-diamond could produce OLCs in large quantities.

     

1.1.1      Cerium Oxide (CeO₂)

Almost all electro-catalysts seem to degrade over their lifetime. Some leach into the electrolyte while others leach into the GDEs. For this reason, cerium oxide has been chosen to stabilize the catalyst and decrease the rate of degradation. CeO₂ will be fused with all electro-catalyst synthesized in this project to increase the life-span of the catalyst and/or AGFCS. A previous experiment and report showed that hydrogen and CO oxidation activity was significantly dependent on the cerium concentration ^[11]. A large concentration would increase activity and selectivity of the desired product H₂O but too much would lower the surface area of the catalyst and thus lower overall activity. They concluded that CeO₂ was a promising additive to PEMFCs ^[18].

1.   Material Characterization
2.   Methodology

Chemicals

Cerium Nitrate (Ce(NO₃)₃)

Carbon Black (CB)

Onion-Like Carbon (OLC)

Ultrapure Water (UPW)

Palladium (Pd) nanoparticles

Nickle Chloride (NiCl₂)

Stirrer used: Stuart SB 162 heat-stir

Making the ink

1) In small vial, weigh 2mg of catalyst powder

2) Add 2uL of Nafion and 1mL of absolute (99%) ethanol

3) Sonicate for 1 hour

4) Mix contents using a pipette.

Experimental solutions (100mL)

KOH

0,5M KOH = 2.81g KOH + 100ml Ultrapure Water (UPW)

 

KOH (+Ethanol) [1:1]

1M KOH + 1M EtOH = [2.81 KOH + 50ml UPW] + [2.95mL EtOH (99.9%) + 47.05mL UPW]

 

Ferric-Ferrocyanide (1:1) in 0.1M KCl

0.03:0.03:1 = 0.127g K₄(CN)₆.3H₂0 + 0.099g K₃(CN)₆ + 0.746g KCl + 100mL UPW

    

1.1      Synthesising the Support CeO₂/CB (and CeO₂/OLC)

1) Add 1.507g of Cerium Nitrate to 50ml of Ultrapure Water (UPW) in a conical flask

2) Slowly add 1.130g CB (or OLC) to cerium nitrate slurry

3) Fill to 100ml with UPW ensuring to wash the walls of the flask

4) Stir for 30mins

5) In another flask, make 40ml of 2M KOH – stir for 10mins

6) Add mixture (5) to (3) until a pH of 12 is achieved

7) Sonicate for 1 hour

8) Stir vigorously for 2 hours

9) Centrifuge and wash to pH 7

 

Yield after step 9:        CeO₂/CB = 2.0091g

                                    CeO₂/OLC = 2.0225g

 

10) Using a furnace, heat 1.700g CeO₂/CB (or CeO₂/OLC) to 250^oC for 3 hours

11) Cool to room temperature (25^oC) under a flow of Argon (Ar)

1.2      Synthesising the Catalysts

1.   Result and Discussions
1.1      Material Structure Analysis

As-prepared electrocatalyst has been characterized by XRD, SEM, TGA, BET, and Raman spectroscopy. The XRD of the Pd-CB, Pd-CB-CeO2, Pd-OLC, and Pd-OLC-CeO2 are shown in figure 1. XRD pattern of Pd/CB possesses a broad peak at approximate 24^ο, which is corresponding to 002planesof graphitic carbon with a low degree of graphitization. Moreover, the Pd-OLC shows the sharp peak at the top of the broad peak at approximate 24^ο, corresponding to the higher degree of graphitization compared to the CB. The graphitized carbon of Pd-OLC may help for the enhancement of electrochemical activity and stability. Since graphitized carbon possesses good conductivity along with corrosion resistance property.

Further, in the case of Pd-CB+CeO₂ and Pd-OLC+CeO₂the carbon peak (CB and OLC) has been suppressed, and the new peak near 28^ο observed and shown by the square in figure 1. The newly obtained peak was observed and it assignsto the CeO₂.  Along with the carbon and CeO₂, the Pd peak for Pd-CB, Pd-CB-CeO₂, Pd-OLC, and Pd-OLC-CeO₂ was found and shown by star symbol in the XRD spectrum. Among all the as-prepared catalyst Pd-OLC shows the very sharp peak of the Pd,it attributed tothe crystalline Pd particles on the OLC. The highly graphitized carbon and the crystalline Pd of the Pd-OLC, is expected to showthe superior electrochemical performance among all the as-prepared catalyst.

Further, to understand the distribution of Pd nanoparticles on various support, the morphology of the as-prepared Pd-CB, Pd-CB-CeO₂, Pd-OLC, and Pd-OLC-CeO₂ was obtained by scanning electron microscope (SEM) images and results are shown in figure 2. Figure 2a represents the Pd-CB, and Pd particles are agglomerated and highlighted in the circle. It suggests that the Pd particles are not well dispersed throughout the CB surface. However, in figure 2b of Pd-OLC indicates the uniform distribution of Pd particles on the OLC surface. Further, in the case of Pd-CB+CeO₂, the Pd particles are well distributed as shown in figure 2c, but the CeO₂ shows the agglomerated on the CB surface shown in the circle. Moreover, Pd-OLC+CeO₂, CeO­₂ on the OLC surface is less agglomerated comparatively CB as shown in figure 2d.

Thermogravimetric analysis (TGA) was performed on the as-prepared catalyst to determine the amount of Pd catalyst on the support. The TGA analysis was conducted under the air atmosphere in the temperature range from room temperature (RT) to 900⁰C. In this condition, the carbon can get the evaporate and the remaining residue is considered as the metal content.  Figure 3 shows the TGA analysis of the as-prepared catalyst and Pd content is approximately 8% present in the Pd-CB and Pd-OLC. Further, metal content approximately 40% is present in the Pd-CB+CeO₂ and Pd-OLC+CeO₂, it includes the weight percentage of CeO₂ and Pd. The carbon evaporation of the Pd-CB and Pd-OLC shows at higher temperature comparatively Pd-CB+CeO₂ and Pd-OLC+CeO₂. 

Further, the surface area of the active catalysts was obtained from the nitrogen adsorption-desorption curves, and the results are shown in figure 4.  The Pd-OLC shows the highest surface area among all as-prepared material. The highest surface area of the Pd-OLC can be helpful for the adsorption of ions from the electrolyte, and it can help to shorten the diffusion of ions (Bokobza et al, 2015

1.1      Raman Spectroscopy analysis

Raman spectroscopy is a non-destructive technique to understand the structure of carbon material. Hence, herein we have used the carbon and the metal oxide as support for the electrocatalyst of the ethanol oxidation reaction. Raman spectrum of the Pd-OLC and Pd-OLC + CeO₂ shown in figure 5. The presence of D -band peak by the disordered structure of SP² hybridized carbon system, whereas, G-band arises is due to the E2g and exist due to the stretching of carbon-carbon bond in graphitic carbon.  However, in the case of Pd-OLC+CeO₂ small peak at the G peak is observed and known as D’-peak, is due to the impurities or surface charges. It may occur due to the presence of CeO₂ on carbon network.  The small peak at 1088 cm^-1 for Pd-OLC and Pd-OLC+CeO₂ was observed and itsattributes to the presence of Pd. Further I_d/I_g ratios provide information about defects/disorder in the carbon materials.The I_d/I_g ratios of 1.20 and 1.12 were observed for the Pd-OLC and Pd-OLC+CeO₂, respectively (Ghosh et al, 2014).  

1.1      Electrochemical Performance Analysis

Carbon support plays a crucial role not only for the electrocatalyst distribution on carbon support; also, it promotes the electronic structure of the catalyst. Such as degree of graphitization, presence of surface functionalities, heteroatom (N, P, S, B) dopant in carbon matrix influence the electron transfer from electrocatalyst to support.  Besides that, textural and morphological (porosity, surface area, etc.) of support affect the activity of electrocatalyst and selectivity of ethanol oxidation. Therefore, this project mainly focused on the support for the electrocatalyst[Iqubal et al, 2017; Zhang et al, 2018; Li et al, 2017).

The cyclic voltammetry (CV) is the important techniques to understand oxidation and reduction peak for the redox reaction. As seen in figure 6, anodic and cathodic scan proceed the oxidation of K₄Fe(CN)₆ to K₃Fe(CN)₆ and reduction is vice-versa as shown in equation [1].

                                                    [1]

The CV curves of Pd-CB, Pd-CB+CeO₂, and Pd-OLC+CeO₂ shows the broad peak of oxidation and reduction and the sharp peak has appeared for the Pd-OLC in figure 6. It suggests that the Pd-OLC possess high reversibility and fast kinetic for electrochemical performance. Hence, it is expected that the Pd-OLC could have the fast kinetics for ethanol electrooxidation reaction. It may occur due to the faster electron transfer from the catalyst surface to the support. This indicates that the OLC could be the good support for the fast electron transfer and better ethanol electrooxidation reaction.

Figures 7a and b show the cyclic voltagramm (CV) curvesof the Pd supported on the various carbon support in 1.0 M KOH withand without 1.0 M ethanol electrolyte, respectively.  Generally, the Pd surfaceis considered tobeanactive site for electrochemical redoxreactions. Three potential peaks were observed in Figure 7a. Peak I in the potential range of -0.9 to -0.7 may attribute to the adsorption of hydrogen on the Pd surface. Among Pd-CB, Pd-CB-CeO2, Pd-OLC, and Pd-OLC-CeO2, the peak-I current of Pd-OLC-CeO2 are higher, and it indicates that the hydrogen adsorption occurs only on the Pd surface. Pd-OLC also shows the sharp peak, but the peak for the Pd-CB and Pd-CB-CeO2 is the broad peak, it may suggest that the CB also helps to the hydrogen adsorption.  Further, during the cathodic scan, the hydrogen desorption peak -V was observed at potential -0.7 to -0.8 V(Vračar et al, 1998). The current density for the hydrogen desorption peak of all the material is almost the same, it indicates that the hydrogen desorption occurs a similar way for all the as-prepared materials.

Peak – II and III in Figure 7a correspond to the adsorption of OH^- and formation of the oxide layer on the catalyst surface, respectively. The Pd-OLC and Pd-OLC-CeO2 show the better electrochemical double layer formation among all the as-prepared material. Peak – III shows the higher current density compared to the Pd-OLC-CeO2; it indicates that oxide layer formation was more prone to the Pd surface of the Pd-OLC catalyst. It may contribute that the OH- was adsorbed on the Pd surface of the Pd-OLC catalyst and OH^- may be adsorbed on carbon or CeO2 of Pd-OLC-CeO2 catalyst. The reduction of Pd (II) oxide during cathodic sweep confirms byan increase in current density of peak – IV of Pd-OLC is higher than the Pd-OLC-CeO2. These results may conclude that the Pd-OLC-CeO2 shows the superior electrochemical performance due to catalyst support provides a platform for the OH^- adsorption restrict the formation of the oxide layer on the catalyst surface (Takamuraand Ken'ichi, 1965; Prabhuram et al, 1998; Liang et al, 2009)..

Two main peaks A and B have shown in figure 7b, which attribute to the positive and negative peak related to the ethanol oxidation activity of the electrocatalyst.  The peak – D in figure 7b corresponds to the hydrogen adsorption and desorption, and the current density intensity was suppressed compared to the peak intensity in 1.0 M KOH. It attributes to the as-prepared catalyst shows the ethanol oxidation reaction. The positive oxidation peak-A attributed to the direct oxidation of ethanol and the peak – B attributed to the intermediate product of ethanol oxidation and/or incompletely oxidized carbonaceous material produced in the system(Sahin and Hilal, 2013; Li et al, 2019).

The higher current density of the Pd-OLC among the all other as-prepared catalyst indicates the superior catalyst activity towards the ethanol oxidation. The current density of the peak – A and B has been decreased for the Pd-OLC-CeO2. It may associate that the electron transfer from catalyst to the support has been hindered due to the non-conducting behaviour of CeO₂ and lower surface area of the Pd-OLC-CeO2.  The Pd-CB shows the least ethanol oxidation performance may be due to the least surface area.  Further, CeO₂ has been added with the carbon support, the other peak – C has been observed, it may attribute to the oxidation of CeO2. The CV of as-prepared catalysts in 1.0 M KOH and 1.0 M KOH +EtOH confirms that the Pd-OLC shows the superior activity due to the synergistic effect of metal with carbon supportsfor the ethanol oxidation reaction.

Further, the effective catalyticsurface area (ECSA) by using the integral area of the hydrogen adsorption from CV plots and equation 2.

                                                                                                                              [2]

Where, Q = Integral of hydrogen adsorption from CV plot,

S = Charge density constant for Pd (0.424 mC/cm²), and

L = mass loading of Pd on electrode.

The ECSA values of each electrocatalyst were compared in Figure 8. Among the as-prepared 

     various electrocatalyst, Pd-OLC shows better electrochemical activity with higher ECSA towards the ethanol oxidation. The support OLC helps to faster electron transfer during the electrochemical oxidation.

To further study the EOR, a typical liner sweep voltagramm at the sweep rate of ______ mV/s has been carried out on as-prepared electrocatalyst in 1.0 M KOH + 1.0 M EtOH, and the result shown in figure 9.

The current density at potential – 0.2 V is gradually increasing from Pd-CB, Pd-CB+CeO₂, Pd-OLC+CeO₂, and Pd-OLC in figure 9. It may attribute to the electron transfer from catalyst surface to support is faster. Therefore Pd-OLC shows the superior performance among all the as-prepared electrocatalyst. The Pd-OLC and Pd-OLC+CeO₂ show the approximately same onset potential compared to the Pd-CB and Pd-CB+CeO₂. However, the current density of the Pd-OLC is much higher than that of Pd-OLC+CeO₂, it indicates that the Pd-OLC possess the highest ethanol oxidation kinetics. It confirms that the carbon support is helpful to enhance the kinetics and lower the overpotential for the EOR. The high surface area of the Pd-OLC and graphitized carbon is responsible for the superior EOR.  

In addition, the compared Tafel plots of ethanol oxidation were obtained in 1.0 M KOH+1.0 M EtOH and the values of each catalyst are shown in figure 10. The lowest value Pd-OLC was observed among all the as-prepared catalyst, it indicates the ethanol oxidation is more favourable on the Pd-OLC. Tafel slope values of Pd-OLC 121 mV/dec which is very close to the ethanol electrooxidation controlled by the adsorption of OH^- value is 120 mV/dec(Jiang et al, 2010; Shen et al, 2010). 

(Note: The lowest value of Pd-OLC is expected, please recalculate for the Pd-CB) 

Further to understand the charge transfer resistance for prepared catalyst theNyquist plots are made and shown in figure 11. The charge transfer resistance (R_ct) provides a resistance between the surface of the electrode and the electrolyte when the ethanol oxidized at the surface of electrocatalysts. The lower the R_ct signifies the faster the charge transfer and hence lower the obstacle for electron transfer from electrocatalyst surface. Hence, the lower the R_ct value of the Pd-OLC indicates that the better the electrocatalytic performance of the catalyst. Figure 11 shows the lower R_ct of the Pd-OLC among all the other catalyst and impedance results are consistent with CV, LSV, and Tafel plots. Therefore, it can be concluded that the carbon support is important to enhance the performance of the ethanol electrooxidation (Yang et al, 2019). 

1.   Conclusion of Electrochemistry

In summary, the Pd dispersed on the various support such as CB, CB+CeO₂, OLC, and OLC+CeO₂ and tested for the ethanol electrooxidation. The presence of graphitized carbon in OLC and the non-graphitized carbon of CB has been identified by the XRD spectrum. The amount of mass loading on the support has been detected by using the TGA. The Pd-OLC possess the highest surface area of 357.73 m²/g has been observed through the Nitrogen adsorption-desorption isotherm technique. The highest electrochemical surface area and the hydrogen adsorption and desorption peak were analysed by the CV technique. The as-prepared Pd-OLC shows the highest ECSA among all the as-prepared catalyst. The lowest onset potential -0.58 V and the highest current density of the Pd-OLC indicates the highest ethanol electrooxidation kinetics than as-prepared electrocatalyst. The lowest Tafel slope value of 121 mV/dec suggest the ethanol electrooxidation through the OH^- adsorption mechanism. The lower R_ct value of the Pd-OLC indicates the better electrocatalytic performance of the catalyst. The superior electrochemical performance of the Pd-OLC could occur due to the graphitized and high surface area of OLC among other support.

2.   References of Electrochemistry
Bokobza, L., Bruneel, J. L., and Couzi, M. (2015). “Raman spectra of carbon-based materials (from graphite to carbon black) and of some silicone composites.” C—Journal of Carbon Research, 1(1), 77-94.
Ghosh, S., Ganesan, K., Polaki, S. R., Ravindran, T. R., Krishna, N. G., Kamruddin, M., and Tyagi, A. K. (2014). “Evolution and defect analysis of vertical graphene nanosheets.” Journal of Raman Spectroscopy, 45(8), 642-649.
Iqbal, M., Li, C., Jiang, B., Hossain, M. S. A., Islam, M. T., Henzie, J.,and Yamauchi, Y. (2017). “Tethering mesoporous Pd nanoparticles to reduced graphene oxide sheets forms highly efficient electrooxidation catalysts.” Journal of Materials Chemistry A, 5 (40), 21249-21256.
Jiang, L., Hsu, A., Chu, D., and Chen, R. (2010). “Ethanol electro-oxidation on Pt/C and PtSn/C catalysts in alkaline and acid solutions.” International Journal of Hydrogen Energy, 35 (1), 365-372.
Li, Z., Yang, R., Li, B., Yu, M., Li, D., Wang, H., & Li, Q. (2017). “Controllable synthesis of graphene/NiCo2O4 three-dimensional mesoporous electrocatalysts for efficient methanol oxidation reaction.” Electrochimica Acta, 252, 180-191.
Li, Z., Zhang, L., Yang, C., Chen, J., Wang, Z., Bao, L., Wu, F., and Shen, P. (2019). “Graphitized carbon nanocages/palladium nanoparticles: Sustainable preparation and electrocatalytic performances towards ethanol oxidation reaction.” International Journal of Hydrogen Energy, 44 (12), 6172-6181.
Liang, Z. X., Zhao, T. S., Xu, J. B., and Zhu, L. D. (2009). “Mechanism study of the ethanol oxidation reaction on palladium in alkaline media.” Electrochimica Acta, 54 (8), 2203-2208.
Prabhuram, J., Manoharan, R., &Vasan, H. N. (1998). “Effects of incorporation of Cu and Ag in Pd on electrochemical oxidation of methanol in alkaline solution.” Journal of applied electrochemistry, 28 (9), 935-941.
Sahin, O., &Kivrak, H. (2013). “A comparative study of electrochemical methods on Pt–Ru DMFC anode catalysts: the effect of Ru addition.” International journal of hydrogen energy, 38 (2), 901-909.
Shen, S. Y., Zhao, T. S., & Xu, J. B. (2010). “Carbon supported PtRh catalysts for ethanol oxidation in alkaline direct ethanol fuel cell.” international journal of hydrogen energy, 35 (23), 12911-12917.
Takamura, T.; Minamiyama, K. I. (1965). “Anodic oxidation of methanol at palladium electrode in alkaline solution.” Journal of The Electrochemical Society, 112 (3), 333-335.
Vračar, L. J., Burojević, S., &Krstajić, N. (1998). “The surface processes at Pd–Ni alloy in acid and alkaline solutions.” International journal of hydrogen energy, 23 (12), 1157-1164. 
Yang, Y., Yu, S., Gao, L., Wang, X., and Yan, S. (2019). “The Properties of PdRu/C with respect to the Electro-oxidation of Methanol and Ethanol.” INTERNATIONAL JOURNAL OF ELECTROCHEMICAL SCIENCE, 14 (1), 1270-1282.
Zhang, Q., Jiang, L., Wang, H., Liu, J., Zhang, J., Zheng, Y., Li, F., Yao, C., and Hou, S. (2018). “Hollow graphitized carbon nanocage supported Pd catalyst with excellent electrocatalytic activity for ethanol oxidation.”  ACS Sustainable Chemistry & Engineering, 6 (6), 7507-7514.
Project II: Bimetallic catalyst is used as PdNi with various carbon support
Introduction:
In ethanol oxidation reaction (EOR) various strongly adsorbed intermediate such as acetic acid, acetaldehyde, and carbon monoxideused to form, which can poison the electrocatalytic surface and reduced the ethanol oxidation efficiency (Kowal et al,2009;  Du et al, 2012).  On the other hand, in alkaline fuel cell adsorption of OH^- on the catalyst surface leads to the poisoning of the catalyst surface. However, Ni would activate the dissociation of a water molecule (H₂O  ---à H⁺ + OH_ads + e^-) at lower potential and provide an active site for the OH adsorption (Sulainman et al, 2017). Hence, the alloy formation of Pd with Ni can protect the active site of the Pd surface from the adsorption of OH^- and the ethanol oxidation efficiency of the PdNi would be expected to increase.  Therefore, in the present report PdNi alloy used as an anode electrode for the EOR.
According to the previous report, the carbon support helps to transfer the electron from the electrocatalyst to the support, hence herein we have used the various carbon support for the PdNi catalyst. 
Material Characterization:
Figure 1 shows the XRD pattern of the PdNi-CB, PdNi-CB-CeO2, PdNi-OLC, and PdNi-OLC-CeO2. XRD pattern of PdNi-CB possesses a broad peak at approximate 24^ο, which is corresponding to 002 planesof graphitic carbon. Whereas, PdNi-OLC shows a sharp peak on top of the broad peak at 24^ο, corresponds to graphitic carbon. Hence, CB possess the non-graphitic carbon, however, OLC suggests the graphitic carbon. The carbon peak in figure 1 shown by the triangle symbol and the square symbol at 28^ο suggest the CeO₂. The peak at 40⁰ indicated Pd since the peak is not very sharp, and broadening of peak suggests the Ni incorporation in the crystal of Pd (Yang et al,2005).  

The morphology of the as-prepared electrocatalysts PdNi-CB, PdNi-CB+CeO₂, PdNi-OLC, and PdNi-OLC+CeO₂ is shown in figure 2. The PdNi on CB are well dispersed and shown by the arrow in figure 2a, however, PdNi particles on OLC also distribute and the size of the particles is lesser than the PdNi-CB shown in figure 2b. Figure 2c and d shows the presence of CeO₂ and highlighted by the circle. It suggests that the CeO₂ well adsorbed on the CB and OLC in figure 2c and d respectively. The well-dispersed PdNi particles may help to enhance the electrochemical performance.

 Thermogravimetric analysis (TGA) helps to estimate the metallic percentage in the as-prepared catalysts and is shown in figure 3. Therefore, the TGA analysis was performed in air medium so the carbon can be evaporated the remaining residue can consider as the metallic part. The TGA analysis of PdNi-CB and PDNI-OLC shows the presence of the approximately 18% of a metallic content and 50% is present in the Pd-CB+CeO₂ and Pd-OLC+CeO₂. From the percentage of PdNi-CB, PdNi-CB +CeO₂, PdNi-OLC, and PdNi-OLC+CeO₂, it can estimate that around 30% is the CeO₂ and remaining carbon is present in the as-prepared catalysts.

Raman Spectroscopy analysis:

Raman spectroscopy is a non-destructive technique to understand the structure of carbon material. Hence, herein we have used the carbon and the metal oxide as support for the electrocatalyst of the ethanol oxidation reaction. Raman spectrum of the PdNi-OLC and PdNi-OLC + CeO₂ shown in figure 5. The presence of D -band peak by the disordered structure of SP² hybridized carbon system, whereas, G-band arises is due to the E2g and exist due to the stretching of carbon-carbon bond in graphitic carbon (Ghosh et al, 2014).  However, in the case of PdNi-OLC+CeO₂ small peak at the G peak is observed and known as D’-peak, is due to the impurities or surface charges. It may occur due to the presence of CeO₂ on carbon network.  The small peak at 1088 cm^-1 for PdNi-OLC and PdNi-OLC+CeO₂ was observed and its attribute to the presence of Pd and Ni. Further I_d/I_g ratios provide information about defects/disorder in the carbon materials(Bokobza et al, 2015). The I_d/I_g ratios of 1.18 and 1.13 were observed for the PdNi-OLC and PdNi-OLC+CeO₂, respectively.

 Electrochemical Performance Analysis:

The electrochemical activity of the as-prepared catalyst initially tested in potassium ferrocyanide solution and the results are shown in figure 6. The CV curves of PdNi-CB, PdNi-CB+CeO₂, and PdNi-OLC+CeO₂ shows the broad peak of oxidation, a small peak at top of the oxidation peak confirming the multielectron oxidation reaction occurred and the cathodic scan shows the single reduction peak in figure 6. However, the sharp oxidation and reduction peak is appeared for the PdNi-OLC in figure 6, indicating the one-electron transfer process and higher current suggest the faster electron transfer through the OLC support. It may suggest that the PdNi-OLC faster kinetics for the ethanol oxidation reaction.

Further, we have investigated the catalytic activity of the as-prepared PdNi-CB, PdNi-CB+CeO₂, PDNi-OLC, and the PdNi-OLC+CeO₂ towards the ethanol oxidation reaction by using the cyclic voltagramm (CV) in 1.0 M KOH and 1.0 M KOH +1.0 M EtOH. In figure 7a, peak – I and V represents the hydrogen adsorption and desorption peak respectively, the current density of the peak – I and V has been increasing from PdNi-CB, PdNi-CB+CeO₂, PdNi-OLC, and to PdNi-OLC+CeO₂. Further peak – IV also present for all the as-prepared catalyst, it suggests that the adsorption of OH^- or formation of an oxide layer on the catalyst surface and peak – V corresponds to the reduction of oxide or desorption of OH^- ion from the catalyst surface (Na et al,2015).

Generally, the mechanism of EOR on Pd can be ascribed as following equations (1-4)(Tripkovic et al, 2001; Liang et al, 2009; Liu et al, 2007).

Equation (1) and (2)have basically occurred on the Pd surface, however, the generated hydroxyl radical (OH^-) block the Pd active surface as shown in equation (3) and hence the ethanol oxidation activity of monometallic Pd may decrease.

Ni is active metal for the adsorption of OH^-, as shown in equation (3) adsorption of the CH₃CO and OH^- is the determine the ethanol oxidation activity.  Therefore, Ni could be involved for the OH^- adsorption and can convert it into Ni(OH)₂ and Pd surface sites are used for the ethanol oxidation reaction. Henceforth, Ni can provide a site for the OH^- adsorption and therefore, Pd facilitate the ethanol oxidation at lower onset potential ( Zhang et al, 2011).Therefore, it is expected that the similar behaviour of all the catalyst, but in figure 7b the current density of each catalyst is different since the carbon support is varied. It suggests that carbon support plays an important role in electron transfer from the catalyst surface. In figure 3b, the current density of the peak – A is not much increase from PdNi-CB, PdNi-CB+CeO₂, PDNi-OLC, and PdNi-OLC+CeO₂ since it indicates the direct oxidation of ethanol. Whereas, peak – B assigned to the removal of the carbonaceous species and the regeneration of Pd active for the fresh ethanol oxidation reaction. Figure 7b shows the monotonical increment of the current density of the peak – B as the catalyst support is varied. Although the current density of peak – B of the Pd-OLC+CeO₂ is better than the Pd-OLC the current density of peak – A is similar for both the catalyst. It may attribute that, the generation of the carbonaceous product is higher for Pd-OLC+CeO₂ and incomplete ethanol oxidation.  It indicates that the carbon support plays an important role along with catalyst surface for the ethanol oxidation.

Therefore, estimation of the effective catalytic surface area (ECSA) of as-prepared catalysts is essential and it was estimated through the CV and following equation (5).

                                                                                                                  [5]

Where, Q = Integral of hydrogen adsorption from CV plot,

S = Charge density constant for Pd (0.424 mC/cm²), and

L = mass loading of Pd on electrode.

From figure 8, it observed that the ECSA of the Pd-OLC is 515.8 cm²/mg and highest among of all the as-prepared catalyst.  It suggests that Pd-OLC possess the high surface area for the electrochemical performance and hence the CV in figure 3b shows the similar current density of anodic peak of the direct-ethanol oxidation for the Pd-OLC and Pd-OLC+CeO₂. It may indicate that the Pd-OLC promote the ethanol oxidation reaction. 

Further, the linear sweep voltammetry (LSV) curve of the various catalyst has been tested and the results are shown in figure 9. The as-prepared catalyst shows the very little increment in current density from the PdNi-OLC and the PdNi-OLC+CeO₂, but the onset potential is found to be -0.62 and -0.58 V, respectively and shown by I and II in figure 9. The lowest onset potential of PdNi-OLC suggests that the superior electrochemical performance among all the as-prepared catalyst. 

Further, to understand the electrochemical performance of as-prepared catalyst Tafel slop has been shown in figure 10. The lowest Tafel slope value of 95 mV/dec was obtained by the PdNi-OLC catalyst and the other values are shown in figure 10. It confirms that PdNi-OLC is the best catalyst among all the as-prepared catalyst. Although all catalyst possesses the bimetallic PdNi alloy, but the support is varied and hence the ethanol oxidation activity of the catalyst is varied. PdNi-OLC shows the best catalytic activity towards the ethanol oxidation because of the high surface area and graphitic nature of the OLC. The Tafel value is close to the 120 mV/dec, it suggests that the ethanol oxidation was controlled by the OH^- adsorption mechanism. The as-prepared PdNi-CB, PdNi-CB+CeO₂, and PdNi-OLC+CeO₂ show the 127, 184 and 186 mV/dec values, respectively. It suggests that the PdNi-CB is also controlled by the OH^- adsorption mechanism. The CeO₂ added to support does not follow the OH^- adsorption mechanism.

Further, to understand the charge transfer between the electrode-electrolyte surface, the electrochemical impedance spectroscopy (EIS) was studied and the result is shown in figure 11. In the case of EIS, semicircle in figure 11 shows the charge transfer (R_ct) and lower the R_ctfaster the charge transfer and lesser the obstacle for electron transfer from the catalyst surface to support. Figure 11 shows the lower R_ct of the Pd-OLC among all the other catalyst and impedance results are consistent with CV, LSV, and Tafel plots. Therefore, it can be concluded that the carbon support is important to enhance the performance of the ethanol electrooxidation. 

References

 [1] Webpage – http://www.news24.com/SouthAfrica/News/sa-one-of-the-worst-in-alcohol-related-road-deaths-20151203 (Date accessed: 14/10/2017)

[2] Gasteiger, H., Kocha, S., Sompalli, B. & Wagner, F. 2005. Activity benchmarks and requirements for Pt, Pt-alloy, and non-Pt oxygen reduction catalysts for PEMFCs. Applied Catalysis B: Environmental. 56(1-2):9-35. DOI: 10.1016/j.apcatb.2004.06.021.

[3] Modjtahedi, A., Amirfazli, A. &Farhad, S. 2016. Low catalyst loaded ethanol gas fuel cell sensor. Sensors and Actuators B: Chemical. 234:70-79. DOI: 10.1016/j.snb.2016.04.108.

[4] Webpage - http://www.who.int/mediacentre/factsheets/fs358/en/ (Date accessed: 14/10/2017)

[5] Holton, O. & Stevenson, J. (2013). The Role of Platinum in Proton Exchange Membrane Fuel Cells. Platinum Metals Review. 57(4):259-271. DOI: 10.1595/147106713x671222.

[6] Ito, S., Chibana, C., Nagashima, K., Kameoka, S., Tomishige, K. &Kunimori, K. (2002). CO hydrogenation over RhVO₄/SiO₂, Rh/V₂O₃ and Rh/SiO₂ catalysts: reduction and regeneration of RhVO4. Applied Catalysis A: General. 236(1-2):113-120. DOI: 10.1016/s0926-860x(02)00283-1.

[7] Liu, Z., Gan, L., Hong, L., Chen, W. & Lee, J. 2005. Carbon-supported Pt nanoparticles as catalysts for proton exchange membrane fuel cells. Journal of Power Sources. 139(1-2):73-78. DOI: 10.1016/j.jpowsour.2004.07.012.

[8] Chai, G., Yoon, S., Yu, J., Choi, J. & Sung, Y. (2004). Ordered Porous Carbons with Tunable Pore Sizes as Catalyst Supports in Direct Methanol Fuel Cell. The Journal of Physical Chemistry B. 108(22):7074-7079. DOI: 10.1021/jp0370472.

[9] Jiang, J. &Kucernak, A. 2002. Nanostructured platinum as an electrocatalyst for the electrooxidation of formic acid. Journal of Electroanalytical Chemistry. 520(1-2):64-70. DOI: 10.1016/s0022-0728(01)00739-2.

[10] Breeze, P. et al (2009). Renewable energy focus handbook. Amsterdam: Academic Press/Elsevier. ISBN: 9780123747051

[11] Zhang, J. 2008. PEM fuel cell electrocatalysts and catalyst layers. London: Springer.

[12] Webpage - http://vingmed.se/wp-content/uploads/2014/03/Broschyr_EN.pdf(Date accessed: 17/10/2017)

[13] LaConti AB, Hamdan M, McDonald RC. Mechanisms of chemical degradation. In. Handbook of fuel cells: fundamentals, technology, and applications, vol. 3. Vielstich W, Lamm A, Gasteiger H, editors. Chicester, England: John Wiley and Sons; 2003;647–62

[14] Jones, A. & Andersson, L. (1996). Variability of the Blood/Breath Alcohol Ratio in Drinking Drivers. Journal of Forensic Sciences. 41(6):14025J. DOI: 10.1520/jfs14025j.

[15] Zuba, D. (2008). Accuracy and reliability of breath alcohol testing by handheld electrochemical analysers. Forensic Science International. 178(2-3):e29-e33. DOI: 10.1016/j.forsciint.2008.03.002.

[16] Lamy, C., Rousseau, S., Belgsir, E., Coutanceau, C. & Léger, J. (2004). Recent progress in the direct ethanol fuel cell: development of new platinum–tin electrocatalysts. Electrochimica Acta. 49(22-23):3901-3908. DOI: 10.1016/j.electacta.2004.01.078.

[17] Bianchini, C. & Shen, P. (2009). Palladium-Based Electrocatalysts for Alcohol Oxidation in Half Cells and in Direct Alcohol Fuel Cells. Chemical Reviews. 109(9):4183-4206. DOI: 10.1021/cr9000995.

[18] Lin, R., Cao, C., Zhang, H., Huang, H. & Ma, J. (2012). Electro-catalytic activity of enhanced CO tolerant cerium-promoted Pt/C catalyst for PEM fuel cell anode. International Journal of Hydrogen Energy. 37(5):4648-4656. DOI: 10.1016/j.ijhydene.2011.05.021.

[19] HALL, S. (2004). Influence of metal?support interaction in Pt/C on CO and methanol oxidation reactions. Solid State Ionics. 175(1-4):809-813. DOI: 10.1016/j.ssi.2004.08.030.

[20] Şen, F. &Gökaǧaç, G. (2007). Different Sized Platinum Nanoparticles Supported on Carbon:  An XPS Study on These Methanol Oxidation Catalysts. The Journal of Physical Chemistry C. 111(15):5715-5720. DOI: 10.1021/jp068381b.

[21] Cao GZ. Nanostructures & nanomaterials: synthesis, properties & applications. Singapore: World Scientific Publishing, 2004.

[22] Wang, X., Waje, M. & Yan, Y. 2005. CNT-Based Electrodes with High Efficiency for PEMFCs. Electrochemical and Solid-State Letters. 8(1):A42. DOI: 10.1149/1.1830397.

[23] Webpage - https://en.wikipedia.org/wiki/Alkaline_fuel_cell#/media/File:Alkalinecell.svg

[24] Qian W, Wilkinson DP, Shen J, Wang H, Zhang J, Architecture for portable direct liquid fuel cells, Journal of Power Sources, 154(1), 202–13.

[25] Miller, H., Lavacchi, A., Vizza, F., Marelli, M., Di Benedetto, F., D'Acapito, F., Paska, Y. & Page, M. et al. (2016). A Pd/C-CeO2Anode Catalyst for High-Performance Platinum-Free Anion Exchange Membrane Fuel Cells. AngewandteChemie International Edition. 55(20):6004-6007. DOI: 10.1002/anie.201600647.