“In these periods of major change, the established points of reference are being swept away, even in so-called traditional industries” (2).

Sustainable mobility is already on the agenda of every government in the world. The concept of sustainable development, defined by the Brundtland Commission as “development that meets the needs of the present without compromising the ability of future generations to meet their own needs” (3) has permeated and is now a standard objective in all spheres of life and at government level. It is also at the core of the Sustainable Development Goals (4), recently launched by the United Nations as global objectives for 2030.

Transport plays a fundamental role for economic development and social welfare of a country. The movement of individuals and goods facilitates production and trade, enhances labour mobility and provides customers with access to goods. Transport externalities jeopardise sustainability. Transport externalities include environmental externalities (mainly climate change, air pollution and noise), but also extend to accident externalities and congestion externalities (5–7). The environmental impact of transport is substantial and ”based on continuing current rates of growth for passengers and freight, and if no mitigation options are implemented to overcome the barriers, the current transport sector’s GHG emissions could increase by up to 50% by 2035 at continued current rates of growth and almost double by 2050” (8).

Most policies in place and most proposed policies by design focus on existing externalities. However, the transport externalities we know today may be replaced by other problems. The world is being shaken up by new technologies and the speed of change is unprecedented. The term ‘disruptive technologies’ is becoming widespread, as shown by recent reports produced by McKinsey and Company, USA (9) and Deloitte LLP, UK (10). With the help of a comprehensive literature review, the aim of this paper is to understand the impact that current and proposed policy could have on the technological change in the road transport sector and how this will change the nature of the problems encountered and the sustainability.

The paper is organised as follows. Section 2 concentrates on alternative energy vehicles, with particular attention to electric and fuel cell vehicles based on their likely preponderance in the vehicle fleet by 2050. Section 3 concentrates on the UK policy in supporting the development and manufacture of electric vehicles (EVs) and Section 4 on the sustainability considerations resulting from current policy measures. Section 5 brings together the key findings and Section 6 concludes with final thoughts and direction for policy recommendations.

In terms of sustainability and emissions in particular, the transport sector is coming under increasing scrutiny. The ‘Paris Agreement’ of 2015 aims to hold the increase in the global average temperature to well below 2°C above pre-industrial levels and to pursue efforts to limit the temperature increase to 1.5°C above pre-industrial levels, recognising that this would significantly reduce the risks and impacts of climate change (11). Transport, as the source of nearly a quarter of all Europe’s greenhouse gas (GHG) emissions (Figure 1), has become one of the focal points. This section focuses on technological improvements that are possible for passenger cars up to 2050 rather than on behavioural change or significant modal shift. The basis for this is that although their modal share would decrease by about 7% between 2010 and 2050, passenger cars will still represent about 67% of total passenger transport activity in 2050 based on European Union (EU) projections (13), whilst a UK study predicted a growth in overall road traffic demand of between 37% and 61% by 2050 (14).

Fig. 1.

In looking to reduce emissions from the road transport sector, the EU has taken regulatory action, which commits the automotive industry to reach a fleet average of 95 g CO₂ km^–1 by 2020 (15). Whilst the 2020 target can still be achieved without a radical industrial transformation, the 10 g CO₂ km^–1, calculated as the tolerable maximum in 2050 to stay below 2°C global warming (16), will require a much more radical departure from current technological trajectories. Technological innovation will play a major role in taking on this challenge.

Beyond 2020 and towards 2050, road transport vehicles are very likely to be propelled by a range of low-carbon technologies: battery electric and fuel cell electric propulsion; and varying degrees of hybridisation. Electromobility, either battery or fuel cell electric, will increasingly challenge the paradigm of internal combustion engine (ICE)‐based mobility, simply because it is technically impossible to increase the efficiency of ICEs to the levels needed to achieve the emissions requirements (17). However, due to the various political and technological uncertainties, it is far from clear how fast and how radical the market penetration of these alternative energy vehicles will be, even though most predictions and forecasts give them a preponderant role in 2050 (18).

2.1 Vehicle Penetration by 2050

2.2 Electricity Generation and Supply

2.3 Hydrogen as an Option?

The UK Climate Change Act, which became legislative in 2008, aims to reduce the emissions of all GHGs by 80% by the year 2050 (from a 1990 baseline). The importance of the transport sector in achieving this target is illustrated (Table II), with transport contributing one third of all UK CO₂ emissions in 2018 compared to just over one fifth in 1990.

To reduce transport related CO₂ emissions, the UK Government plans to phase out ICEs from new vehicle sales by 2040 and “has set ambitions to ensure that almost every car and van in the UK is a zero-emission vehicle by 2050” (37). However, these ambitions come with much uncertainty and the feasibility has been questioned.

Several risk factors will determine how quickly and deeply alternative energy vehicles will penetrate the UK vehicle mix and whether it will become a sustainable market segment. It is of strategic importance that industry understands these risks that can inform their research and development (R&D) investments. Alternative energy vehicles are a new product in a new industry and their radically different composition potentially means substantial change to production systems and value chains. The risk for industry in investing in the nascent value chain is compounded by competing alternative-vehicle technologies. Even though in the UK the government stance is to be technology neutral, government policies play a key role in how new technologies are supported by the wider stakeholder community (38). This will affect the quantitative nature of the risk and its perception in a significant way.

3.1 Creating a Competitive Electric Vehicle Manufacturing Sector

3.2 UK Government Policy in Support of Battery Development

The highly developed car industry is capable of producing sophisticated cars at low production costs. To reach the targets required to meet the Paris Agreement will require alternative drivetrain technology and for the industry the BEV is at present the most market viable solution. However, it takes courage to start the production of large numbers of EVs and the decision is not purely a technical one. It is a combination of science, technology, engineering and public policy that defines the type of EV that will be successful in the marketplace.

The current policy framework allows for a number of potentially divergent pathways. The one discussed in the previous section focused on improving the value proposition by reducing the cost of the high value components, in this case the battery, with the objective of aligning the cost of the EV to the present combustion engine incumbent. Examples of original equipment manufacturers (OEMs) that have adopted this pathway include Jaguar, UK, with its I-PACE, Tesla, USA, with the Model S, Model X and Model 3 and Chevrolet, USA, with the Bolt. Each combines existing approach to vehicle manufacture (materials and processes), hence realising a low-cost base vehicle platform, combined with a battery that has a high energy capacity and relative low cost (achieved through economies of scale associated with the battery manufacture). A further approach, exemplified by BMW, Germany, with its i3, is to increase the overall efficiency by reducing the vehicle weight through innovative manufacturing methods and material choices. This approach recognises that the customer requirement of increasing range and reducing cost can potentially be achieved by focusing on reducing the size of the battery: a lightweight vehicle can cover longer distances with the same battery capacity. A further, and more extreme approach to lightweighting, is the Ped-elec (Coventry University, UK). The dichotomy is that mobility concepts used in urban areas are, at present, extensions of those used outside of the urban environment. They are inherently less efficient. Ped-elec responds to a call for new personal mobility based on energy used per unit mass moved (50).

Based on adoption rate (sales of each vehicle type) it is clear that the industry is gravitating to one particular pathway, reducing the cost of the high value battery whilst retaining the existing approach to manufacture of the vehicle (materials and processes). The option of weight reduction (focusing on energy used per unit mass moved) is a higher cost approach relative to providing additional battery capacity to overcome the lower vehicle efficiencies. Indeed, the need to realise increasing economies of scales in the area of battery manufacture are worrying national governments (UK included) concerned that the battery manufacture will act as the nucleus around which the rest of the industry gravitates; presently the industry gravitates around the location where final assembly of the vehicle takes place. However, whilst this is the preferred option, is it the most sustainable?

EV manufacturing requires more energy and results in more carbon emissions than manufacturing a conventional car (51). A study conducted by the American Chemical Society (ACS) estimated that the Ford Focus EV (Ford Motor Company, USA) has 39% higher ‘cradle to gate’ emissions then a conventional Ford Focus (52). In fact, Ellingsen et al. stated that EVs of all sizes may require 70,000 km to become cleaner than conventional vehicles to make up the manufacturing debt (53).

Various studies on the growth in EV and hence the demand for raw materials required in battery manufacture highlight that certain key materials (such as cobalt, nickel and copper) are at risk from supply constraints. In response, development has begun looking at materials such as iron to replace the cobalt commonly found in batteries (54) whilst research activity into the recycling of battery packs is also a priority area of research. At present there are no facilities for recycling EV batteries in the UK. Processes such as hydrometallurgical recycling and leaching are currently seen as energy efficient methods of recovering spent battery materials, aiming to reduce the cost of recycled batteries metals. Currently research is being undertaken to recycle larger percentages of battery material, with some promising results. Natarajan reports that 99.9% of lithium, 98.7% cobalt and 99.5% of magnesium were leached out of a cathode with a purity of between 98.7% and 99.4% (55). Another study, related to lithium-ion phone batteries, saw 90.02% of cobalt and 86.04% of lithium restored to maximum concentration (56). These tests are currently resigned to laboratories and not available in the UK on a commercial scale. Whilst the metals recycled from EV batteries are deemed to be of sufficient quality to be used in new EV batteries with no performance issues, due to issues of cost, recycled lithium costing three times that of new lithium, and the individual material compositions of each EV battery, bulk battery recycling on a commercial scale is currently not considered economically viable.

Achieving the Sustainability Goal

In Section 2, the case for alternative energy vehicles as a response to meeting policy objectives was made. Although there is some uncertainty of the share of each technology in the powertrain portfolio, it is clear going forward that ICEs will represent only a small percentage of the total vehicle fleet or disappear altogether. It is further evident that there are multiple interest groups in the alternative energy vehicle market and that in preparation for the new mobility paradigm envisaged for 2050 investments will need to be made in new infrastructure and connectivity. Hence, there needs to be an orchestration of policy intervention to integrate issues of energy policy, transport policy and social policy.

In Section 3, there was a discussion around the policy support that the UK Government has in place to realise its ambition of a world leading UK alternative energy vehicle sector. It is clear that the industry, in transitioning to electrification, faces considerable risk. The industry chooses to leverage existing competencies in vehicle design and manufacture, and to achieve cost reduction and range improvements through a focus on the battery. In response the UK Government has put in place support for battery development, leveraging existing research competencies in this area by coordinating activities, and for battery manufacture by looking to attract inward investment and securing the future of automotive manufacturing in the UK.

In Section 4, the policies in support of transitioning to an alternative energy vehicle fleet on the one side and supporting the development of the UK capability in response were brought together in order to explore sustainability. The issue is that the way in which the industry responds to the challenge of emissions reduction creates a cleaner vehicle fleet, but does not necessarily consider optimising the efficiency or sustainability. The problem is that to square the circle – to meet the customer demand of increased range at reduced cost – the industry has looked to economies of scale at the manufacturing level and at the same time look for incremental improvements in the batteries. This enables vehicles to utilise larger batteries at less cost, but at the same time leads to heavier vehicles that fail to optimise efficiency and with increased energy demand can lead to stressing of the energy grid. A further problem is that larger batteries consume more materials and there is risk that certain material supply chains are being stressed and may not be able to respond to future demand, posing critical challenges regarding sustainability and security of supply chain. Whilst interventions, for example greater recycling and the retention of previously processed materials in the value chain, could influence this, the costs associated with these interventions would go counter to the objective of reducing the cost of the battery through economies of scale. Whilst lighter vehicles would be a move in the right direction, and a pathway exists for such vehicles within existing policy framework, the existing requirements for measuring the environmental performance of vehicles focus on emissions at the tailpipe and the move to electric drive removes a check on vehicle weight. Policy intervention is required to correct the above. This policy can target control of vehicle mass directly or can influence it indirectly through a move towards life cycle analysis of CO₂, each approach having its merits and challenges.

The transition to electrification of the vehicle drivetrain represents a considerable risk to the vehicle manufacturing sector. The UK has put in place policies to support both the production and research parts of the equation, but at the same time there is potential mismatch between the direction that is set by these policies and creating a sustainable road transport sector. New policies are required that orchestrate closer coordination across the separate policy areas: promoting lighter vehicles will reduce the stress on raw material supply chains; development of recharging networks will reduce range anxiety and align with the drive to reduce mass through enabling smaller batteries; and improvements in connectivity will lead to greater leverage of both vehicle and energy network capability.

The world is at the start of an energy revolution: the biggest energy transformation since the Industrial Revolution, during which the use of fossil fuels drove growth and prosperity, with global temperature increase implications that we have only started to understand relatively recently. This energy revolution will drive the world towards a lower carbon, more sustainable future, with major implications for energy and electricity generation, heating, industrial power and transportation. Governments, states and regions are proposing, and in some cases (such as the UK) committing to, net zero greenhouse gas (GHG) or carbon dioxide emission targets over the coming years. To date, 15 countries have set defined targets to become net zero economies by 2050 or earlier, with over 50 others, including Germany and Canada, discussing when to implement such a target. Perhaps most significantly, the European Union (EU) intends to be net zero by 2050: this objective is at the heart of the European Green Deal and in line with the EU’s commitment to global climate action under the Paris Agreement.

Interestingly, at the time of writing, around 49% of global gross domestic product (GDP) derives from nations and regions discussing, or with legislated, net zero emissions targets to be achieved by 2050 at the latest (1). Significantly, eight months previously this figure was only 16%, demonstrating the rapid rate at which such commitments are being made. Companies are also making net zero commitments, and this pace is accelerating too: in 2017, 87 companies made such commitments, in 2018 this rose to 174, and in 2019 398 companies announced net zero targets.

Figure 1 summarises the proportion of global fossil-derived CO₂ from each of the major sectors. While electricity generation is the largest contributor, the transportation sector comes second, accounting for around 23% of global CO₂ emissions.

Fig. 1.

Figure 2 looks into the transport sector in a little more detail, revealing that passenger road vehicles contribute almost half (45%) of transport CO₂, with freight vehicles accounting for another 30%, so road transport accounts for almost 75% of the emissions. The aviation and shipping industries also release large levels of CO₂, each at around 11% of global transport-derived emissions, with rail being a relatively minor contributor, at 1%. But to add a little context, these rail CO₂ emissions are around 0.1 billion tonnes per year, the same level as those of Belgium or Austria. Therefore, it is clear that transport has a critical role to play in the global decarbonisation agenda.

Fig. 2.

This special edition of the Johnson Matthey Technology Review looks at the challenges faced in the decarbonisation of the transport sector, and highlights the likely solutions that will be implemented to enable this transition. The articles consider the regulatory frameworks already in place, and those likely to come in the near future, to accelerate the moves to net zero across the transport ecosystem. The current status of the technologies that will play a key role in this transition is discussed, along with expected future developments and performance targets. It is clear that both battery-based and hydrogen fuel cell-based electric vehicles will make major contributions to the decarbonisation of ground transportation, across cars, vans, buses and trucks, and these technologies are discussed in detail. Challenges with the roll-out of the infrastructure for these new vehicles are also assessed, and the likely paths forward are presented.

Battery and fuel cell technologies are unlikely to see large scale uptake in the marine and aviation sectors in the foreseeable future, so here the focus is on the development and deployment of alternative, sustainable, lower carbon fuels which will replace the existing heavy fuel oil and aviation fuels, to mitigate carbon emissions from these two very large sectors.

Meeting net zero GHG emission targets within the transportation sector can only be achieved alongside clean generation of electricity and hydrogen, since these will be the fuels for the highest volume future transport modes (cars and trucks). Therefore, articles in this special edition also look at the changes required in electricity and hydrogen generation to enable the move to clean transport. Recall that electricity generation currently accounts for around 38% of global CO₂ emissions, so increasing the use of renewables and nuclear power are an essential piece of the net zero jigsaw puzzle.

This brings me to the final message: reducing transport CO₂ or GHG emissions to zero is not in itself enough to stabilise earth’s climate. It is a critical step, but needs to take place alongside the decarbonisation of the other major sectors: power generation, industry and building heating and cooling. There is a need for a cross-sector, systems-based approach, rather than looking at individual large emitters in isolation. The article on hydrogen looks at the role that it can play as an energy vector, enabling cross-sector coupling to facilitate the decarbonisation of transport, as well as other key areas such as domestic heating, industrial processes, and as a feedstock for low or zero carbon chemicals and fuels. It also discusses hydrogen’s use as a source of low carbon dispatchable power, as well as how it is a key enabler of significant increases in renewable energy or electricity generation. To reach net zero GHG emissions all the key sectors need to work together, and this special edition, though focused on transport, considers the other changes in the future energy ecosystem that link to, and in some cases enable, clean mobility.

For the past few decades, energy demand in the world has been rising considerably due to an increase in global population and demands for industrial production. The world has been undergoing a massive shift away from fossil fuels towards cleaner energy sources and hydrogen could be an excellent alternative for this purpose (1–3). PEMFCs powered by hydrogen have attracted much attention as a promising candidate for eco-friendly vehicles, i.e. FCEVs, owing to their high power density, high efficiency and zero emission features (4–10).

Since the world’s first mass production of Tucson ix35 FCEV by Hyundai Motor Company (hereinafter abbreviated as Hyundai) in February 2013, global automotive OEMs have also focused on commercialising FCEVs (11–15), including the latest manufacturing of the second generation FCEV (i.e. Nexo) by Hyundai in March 2018 (2). Specifically, Toyota Motor Corporation, Japan, unveiled a mass-produced FCEV, i.e. Mirai, in December 2014 (11, 12). The Mirai FCEV with a seating capacity of four persons employed two hydrogen storage tanks and hydrogen compression pressure of 70 MPa. To reduce contact resistance and improve water management in fuel cells, Mirai adopted three-dimensional fine mesh flow fields which were different from conventional flow fields composed of ribs and channels. In 2016, Honda Motor Company, Japan, also deployed a mass-produced FCEV, i.e. Clarity (13). The Clarity FCEV offered a seating capacity of five persons, two hydrogen storage tanks and hydrogen compression pressure of 70 MPa. In 2017, Daimler, Germany, launched a new generation of FCEV, i.e. Mercedes-Benz GLC F-CELL, with a hydrogen storage system similar to that of other automotive OEMs (14). In June 2018, Audi, Germany, teamed up with Hyundai to share intellectual property and components of fuel cells, with the aim of accelerating the commercialisation of FCEVs and expanding the global market (15). Accordingly, the global market for PEMFCs has been expanding to a broad range of applications including not only for vehicles such as passenger (i.e. sport utility vehicles (SUVs) and sedans) and commercial vehicles (i.e. buses and trucks) but also for trains, trams, forklifts, power generators and vessels. To meet the needs and requirements for these various applications, it is essential to develop more durable and cost-effective materials, components and systems for PEMFCs. Here we present the recent advances in hydrogen and PEMFCs technologies and address remaining technical challenges and barriers to be resolved, which are critical to commercialise next-generation PEMFCs and thus power the future society.

Hydrogen has been regarded as a promising candidate for alternative energy to fossil fuels because it is versatile and can be used in a broad range of applications such as transportation, chemicals, synthetic fuels and metals processing (16–30). Figure 1 shows the concept of wide-scale hydrogen production and utilisation suggested by the US Department of Energy (DOE) (21).

Fig. 1.

In addition, hydrogen is abundant in that approximately 70 million tonne-H₂ year^–1 is used today in pure form, mostly for oil refining and ammonia manufacture for fertilisers. A further 45 million tonne-H₂ year^–1 is used in industry without prior separation from other gases (22). One of the key features of hydrogen production is its diversity. Hydrogen can be produced by a variety of resources including fossil fuels such as natural gas and coal (with carbon capture and storage (CCS)), nuclear energy and renewable energy sources such as wind, solar, biomass, geothermal and hydroelectric power (16–30). Figure 2 shows an illustration of hydrogen production technologies suggested by the US DOE (25, 26). Most hydrogen is currently being produced by conventional ways based on fossil fuels (i.e. steam methane reforming (SMR) or natural gas reforming) which generate a significant amount of carbon dioxide emissions, resulting in ‘brown’ or ‘grey’ hydrogen. However, if the carbon dioxide emitted from the conventional SMR process can be captured and stored, or reused, the hydrogen produced is cleaner than grey hydrogen, which is often referred to as ‘blue’ hydrogen. The cleanest version of all is ‘green’ hydrogen which is produced by renewable energy sources such as wind or solar power, without generating carbon dioxide emissions (23–26). Although the share of green hydrogen produced by clean technologies is now relatively low, the production amount of green hydrogen is expected to increase considerably through water electrolysis powered by renewable energies, photoelectrochemical (PEC) and solar thermochemical hydrogen (STCH) techniques in the future as the hydrogen economy grows (3, 16–20, 23–26).

Fig. 2.

Hydrogen can serve as a versatile energy carrier and plays an essential role in decarbonising major sectors of the economy (27–29). In the power sector, the timing of variable electricity supply and demand is not well matched over the day nor between seasons, which increases the need for operational flexibility. For instance, the production amounts of renewables vary considerably between seasons. Solar generation in Europe is approximately 60% lower in winter than in summer, which coincides with higher electricity demand of about 40% as days become shorter and colder in winter than in summer (27, 28). Therefore long-term energy storage is necessary for large-scale renewable power integration and in this context hydrogen enables large-scale and efficient renewable energy integration through cost-effective long-term storage capability. Figure 3 shows the electricity supply and demand simulation results for Germany in 2050 (27). In this scenario of 90% renewables in Germany, curtailment of more than 170 TWh year^–1 is predicted for 2050, which is equivalent to approximately half the energy needed to fuel the German passenger vehicles with hydrogen. As shown in Figure 3, summer has curtailed periods of electricity oversupply, whereas winter has periods of electricity deficits, indicating strong mismatch between supply and demand of electricity produced by renewable energy sources (RES). Therefore, if we use water electrolysis to convert excess renewable electricity into hydrogen during times of power oversupply, the produced hydrogen can be used to provide back-up power during power deficits or can be used in other sectors such as transport based on fuel cells, industry or residential applications (21, 22, 27–29). In this way, hydrogen can bridge gaps in supply and demand of power and thus can serve as a long-term carbon-free seasonable energy storage medium (27, 28).

Fig. 3.

Hydrogen enables international energy distribution, linking renewable energy-abundant regions (for example, Australia or Norway) with those being deficient in renewable energies and thus requiring energy imports (for example, Japan or South Korea) since hydrogen can store and transport renewable electricity efficiently over long periods of time (27–33). For instance, Japan plans to launch the first technical demonstration of a liquefied hydrogen carrier ship to enter international trade in the near future (27–33). To date, hydrogen pipelines and gaseous or liquefied tube trailers are the most common ways of transport. As the distribution of hydrogen increases, the costs for liquefaction and transport are expected to drop by 30–40% by 2032 (27).

Here we address the recent paradigm shift of global automotive industry before going deep into the details of FCEVs powered by PEMFCs. Recently, while hydrogen has been receiving great attention worldwide and facilitating the energy transition from fossil fuels to renewable energies, the global automotive industry has been experiencing a paradigm shift from traditional internal combustion engine vehicles (i.e. gasoline- and diesel-powered vehicles) to next-generation vehicles based on future mobility concepts such as connected, autonomous, shared and electric (CASE) vehicles (also called autonomous, connected, electric and shared (ACES) vehicles) (34–36). Figure 4 represents an illustration of future mobility concept of Hyundai which is intended to provide ‘connected mobility’, ‘clean mobility’ and ‘freedom in mobility’ for customers.

Fig. 4.

The CASE technologies are closely interlinked with each other and to implement the future mobility concept, in particular, the combination of both autonomous and electric vehicles should be inevitable. The level of autonomy ranges from level-0 (i.e. no automation) to level-5 (i.e. full automation) (37–39). The electric vehicles powered by either batteries or fuel cells are generally well suited to autonomous vehicles. However, as the autonomous vehicles encounter the need for a higher level of autonomous driving technologies which normally require a rapid energy consumption of electric vehicles, the vehicles need more frequent electricity-charging for battery electric vehicles (BEVs) or hydrogen-refueling for FCEVs. In this case, FCEVs could be a better candidate for the platform of autonomous vehicles owing to their longer driving range: over 600 km (i.e. Nexo FCEV) and shorter refueling time, usually less than 5 min (40, 41).

To meet various demands and requirements for customers in the world, as shown in Figure 5, Hyundai has been developing a variety of clean and eco-friendly vehicles over the last decade, i.e. gasoline- and diesel-powered vehicles with improved fuel economy, hybrid electric vehicles (HEVs), plug-in HEVs and pure electric vehicles such as BEVs and FCEVs. And Hyundai has been increasing the share of electrification of the vehicles. In general, BEVs and FCEVs have different strengths that complement each other in that BEVs are more adequate to shorter driving range applications, while FCEVs have a more competitive edge in heavier and longer driving range applications such as buses and trucks.

Fig. 5.

Recently, Hyundai has been actively increasing its commitment to commercialising FCEVs due to their versatile potential in the future power systems, which will be discussed in detail in the following section.

Figure 6 shows the history of FCEV development of Hyundai since 1998. Hyundai developed a proprietary in-house 80 kW stack system in 2004 and since then Hyundai has achieved significant advancements in FCEV commercialisation technologies, finally launching the world’s first mass-produced FCEV (i.e. Tucson ix35: the first generation FCEV) in February 2013, followed by the manufacturing of the second generation FCEV (i.e. Nexo) in March 2018, whose features will be discussed later in more detail.

Fig. 6.

Figure 7 shows a photo and a package layout of the world’s first mass-produced Tucson FCEV of Hyundai. The Tucson FCEV employed an existing internal combustion engine vehicle’s platform. A 100 kW fuel cell stack was located in the engine bay. The vehicle adopted a battery system with 24 kW and two hydrogen storage tanks with a capacity of 5.64 kg-H₂, leading to a driving range of 415 km according to fuel economy tests in Korea. The Tucson FCEVs were deployed in 18 countries worldwide.

Fig. 7.

Through the technical expertise for manufacturing Tucson FCEV since 2013, Hyundai had improved significantly the PEMFC technologies and finally commercialised the second generation of the mass-produced Nexo FCEV in March 2018, with improved performances and durability compared with its predecessor. Figure 8 shows an overview and general features of the Nexo FCEV. In contrast to the Tucson FCEV which had to use an existing internal combustion engine vehicle’s platform, the Nexo FCEV was built on a newly developed and fully dedicated vehicle platform, which renders it higher power and improved driving dynamics than the Tucson FCEV. Figure 8(a) shows the new design of Nexo which was optimised to reduce the drag coefficient from 0.35 (Tucson) to 0.33 (Nexo). Multiple aerodynamic features were discreetly integrated into the front, side and rear areas of the Nexo. As shown in Figure 8(b), the Nexo also performs a remote smart parking assist function which allows the vehicle to autonomously park or retrieve itself from a parking lot.

Fig. 8.

On top of that, a variety of advanced driver assistance system technologies such as the blind-spot view monitor, the lane-following assist and the highway driving assist systems were implemented into the Nexo FCEV to facilitate safe driving. As shown in Figures 8(c) and 8(d), the interior of Nexo features the wide black dashboard that houses two large liquid-crystal displays to hold the digital instrument cluster (left) and the navigation system (right). Figure 8(e) shows the overall package layout of the Nexo FCEV. It primarily consists of an integrated power module with a fuel cell stack and a balance of plant (BOP) system, a motor with maximum torque of 395 N m, three hydrogen storage tanks with a capacity of 156.6 l and 6.33 kg-H₂ and a battery system with a power of 40 kW and an energy capacity of 1.56 kWh.

Figure 9 shows an enlarged view of the integrated power module of the Nexo FCEV. The integrated power module is mainly composed of a 95 kW fuel cell stack and a BOP system consisting of fuel (hydrogen) processing, thermal management and air processing systems. The fuel cells in the Nexo’s stack employ advanced membrane-electrode assemblies (MEAs) with perfluorinated sulfonic acid ionomer-based reinforced membranes and platinum-based electrodes, carbon fibre paper-based gas diffusion layers (GDLs) with microporous layers, metallic bipolar plates and elastomeric sealing gaskets. The BOP system is also of great importance to achieve improved performances, enhanced durability and reduced cost of the Nexo FCEV. The fuel processing system mainly consists of hydrogen supply lines and hydrogen-related sensors, and the air processing system is primarily composed of air humidifier, air compressor and other components. The thermal management system includes cooling-related valves and sensors.

Fig. 9.

Table I summarises key features between Tucson and Nexo FCEVs of Hyundai. Both FCEVs placed their stacks in the front engine bay instead of under the floor and employed hydrogen compression pressure of 70 MPa, hydrogen refuelling time of less than 5 min and a seating capacity of five persons. The Nexo FCEV adopts a variety of proprietary fuel cell components and systems as well as advanced vehicle operation technologies as summarised in Table I. In comparison with its predecessor Tucson FCEV, as listed in Table I, the motor power of the Nexo FCEV increased significantly from 100 kW to 120 kW. Most importantly, the durability of the Nexo FCEV approximately doubled from 4 years/80,000 km to 10 years/160,000 km and the driving range on a single charge increased considerably from 415 km to 609 km, to the authors’ best knowledge, which should be unprecedented among all mass-produced electric vehicles commercially available to date. The cold start-up capability in wintertime had been limited due to the freezing of water produced intrinsically during the oxygen reduction reaction (ORR) at the cathode of PEMFCs and thus challenging to a wide adoption of FCEVs on the real road worldwide. As for the Nexo FCEV, however, the cold start-up capability was greatly improved from −20°C to −30°C, facilitating the vehicle’s market penetration in the world. The system efficiency of the Nexo improved from 55% to 60% as a result of enhanced performances of fuel cell components and systems. The acceleration time from 0 to 100 km h^–1 of the Nexo decreased by 3.3 s, i.e. from 12.5 s to 9.2 s and the maximum vehicle speed increased from 160 km h^–1 to 177 km h^–1. Thanks to the newly developed and fully dedicated vehicle platform, the Nexo FCEV can adopt three hydrogen storage tanks, which enable a larger internal volume of hydrogen tanks from 140 l to 156.6 l and a higher hydrogen storage capacity from 5.64 kg to 6.33 kg, which has contributed to the long driving range of Nexo.

One of the biggest obstacles standing in the way of wider adoption of FCEVs worldwide is the safety concern about hydrogen. Therefore it is of paramount importance to verify the safety of hydrogen storage system in FCEVs. For the past two decades, Hyundai has done a lot of front, rear and side crashworthiness tests on FCEVs as shown in Figure 10. Figures 10(a), 10(b) and 10(c), 10(d) represent the front and rear collision tests of the Nexo FCEV, respectively. In the rear collision or crash test, the vehicle was placed on the transparent test plate underneath which a camera was located. A mobile barrier crashed against the FCEV at the rear end, which caused damage and deformations of hydrogen storage system in the FCEV. Despite the deformations after the collision test, there was no leakage out of the tanks, verifying the safety of the hydrogen storage system. In 2018, the Nexo FCEV was awarded the highest rating in safety from the European crashworthiness test, i.e. European New Car Assessment Programme (Euro NCAP).

Fig. 10.

To date the global markets for PEMFCs for a variety of FCEV applications have been growing very rapidly in terms of both passenger vehicles and medium- and heavy-duty vehicles such as buses and trucks, which require much higher durability than passenger vehicles, i.e. 5000 h for passenger vehicles vs. 25,000 h for heavy-duty vehicles (21, 42, 43). In addition to automotive applications, the PEMFCs are also in demand for other applications such as fuel cell electric trains. Figure 11 shows transportation applications of BEVs, FCEVs, biofuels and synthetic fuels-powered vehicles suggested by Hydrogen Council (27). The FCEVs are expected to occupy the markets of medium- to large-sized passenger vehicles, commercial vehicles including buses and trucks, and even trains.

Fig. 11.

Recently, the concept of commercialising fuel cell electric trains and trams has been materialising in the world. For instance, as an alternative to diesel-powered trains, Alstom Company, France, launched the world’s first passenger train powered by hydrogen fuel cells, i.e. Coradia iLint^TM, to offer commercial passenger service in Germany in September 2018 (44, 45). The Coradia iLint^TM fuel cell electric train was specially designed for operation on non-electrified lines, enabling clean and sustainable train operation while ensuring high performances with a maximum speed of 140 km h^–1. Another commercialisation project of eco-friendly trams powered by PEMFCs has been underway by Hyundai Rotem Company in collaboration with Hyundai Motor Company in South Korea since June 2019 (46). The project plans to develop a low-floor fuel cell electric tram which can travel up to 200 km at a maximum speed of 70 km h^–1 on a single charge by late 2020.

In addition to the role of PEMFCs for transportation applications, another interesting potential of PEMFCs is their capability to produce electricity as a power generator using hydrogen energy. Accordingly, over the last few years, the potential of PEMFCs in FCEVs as distributed power suppliers has received great attention worldwide. Figure 12 shows an illustration of the distributed power generation concept by PEMFCs in FCEVs. The FCEVs can produce approximately 10 kW under idling conditions, which can be used to provide electricity for houses and buildings.

Fig. 12.

To validate and demonstrate extensively the distributed power generation concept by PEMFCs in FCEVs and thus increase public awareness on this aspect, a vehicle-to-grid (V2G) demonstration project (i.e. Hydrogen Electric House project) using Hyundai’s Tucson and Nexo FCEVs has been progressing in South Korea since August 2017. Figure 13 shows the Hydrogen Electric House project in South Korea. The FCEVs can supply electricity, heat and water for the Hydrogen Electric House.

Fig. 13.

The FCEVs also can provide back-up power for people in emergency regions such as earthquake and typhoon disaster areas. In addition it can be used as an electricity charger for BEVs and plug-in HEVs.

Recently, a similar demonstration project showing the V2G technology through integrating an FCEV with photovoltaic power and a residential building was reported in the Netherlands to implement a net zero-energy residential building concept (47). This project showed that utilising an FCEV working in V2G mode could reduce the annually imported electricity from the grid by approximately 71% over one year and aid the buildings in the microgrid to implement the net zero-energy building target.

Another feature of interest of FCEVs differentiating themselves from other types of vehicles is their capability to clean the outside air and thus mitigate air pollution in society (2). Similar to BEVs, the FCEVs do not emit any air pollutants and particulate matter (PM) out of the vehicles while driving on the road. Unlike the BEVs, however, the Nexo FCEV of Hyundai employs an advanced air filter to filter out most of the fine dusts and micro-sized PM in the outside air, enabling to provide purified oxygen from air for the cathode in fuel cells.

On a basis of this positive perspective on hydrogen and PEMFCs, Hyundai announced its investment plan for PEMFCs and FCEVs to the public as the ‘FCEV Vision 2030’ in December 2018. According to this plan, Hyundai will invest US$6.9 billion and produce 700,000 PEMFC systems by 2030: specifically, 500,000 PEMFC systems for automobiles and 200,000 PEMFC systems for other applications such as forklifts, trams, trains, power generators and vessels, as shown in Figure 14.

Fig. 14.

To realise the vision for hydrogen economy through hydrogen and PEMFCs, the fuel cell industry, investors and governments in the world will need to ramp up and coordinate their efforts (27–29). And in order to facilitate the commercialisation of next-generation PEMFCs for a broad range of applications, from the technical point of view, it is of paramount importance to develop more durable and cost-effective fuel cell materials, components and systems as well as advanced fuel cell operation techniques. Here we address several key challenges to be overcome in the future.

Even though there have been extensive efforts to increase hydrogen refuelling infrastructure worldwide over the last decade, the infrastructure is still scarce to deploy the FCEVs, in particular passenger vehicles, sufficiently on the real road. Therefore, to reduce the dependence on the hydrogen refuelling infrastructure, it is necessary to turn our attention to other applications that are less dependent on the number of hydrogen refuelling stations (HRSs). These applications include trams, trains and medium- to heavy-duty commercial vehicles such as fuel cell electric buses (FCEBs) and trucks. For instance, the ideal locations for HRSs of FCEBs are regarded as the bus depot, which allows to estimate the HRS location precisely and thus minimise the cost for HRS construction, indicating no infrastructure requirements on the operation routes (23, 48–50). Fuel cell electric trucks, trains and trams appear to be in a similar condition. For an FCEB to become commercially competitive, however, it is of great importance to develop highly durable fuel cell materials, components and systems first, followed by a drastic reduction of cost, since the durability requirements for FCEBs are much higher than those of passenger vehicles such as SUVs and sedans. It was reported that ultimate lifetimes of an FCEB and its power plant should be approximately 800,000 km and 25,000 h, respectively (42, 43, 48–50), which are five times longer than that of ordinary passenger vehicles. Among core components of PEMFCs for FCEBs, the membrane failure due to pinhole formation seemed to be critical to the lifetime of FCEBs (42), requiring highly durable membranes in terms of both chemical and mechanical durability.

In the case of cathode catalysts for PEMFCs, over the last two decades, extensive research works have been performed to develop durable and cost-effective ORR catalysts with lower Pt loadings (4–6), i.e. highly active Pt-based core-shell catalysts. As pointed out clearly in the literature (5), however, intensive research efforts on developing more durable and reliable electrodes using these novel catalysts should be further exerted, since not all promising ORR activity of catalysts based on typical rotating-disk electrode (RDE) test results have translated into real-world MEA performance, causing a great mismatch between RDE and fuel cell data.

As for the anode catalysts for PEMFCs, it is necessary to develop more effective cell voltage reversal-tolerant anode (RTA) based on oxygen evolution reaction (OER) catalysts. Figure 15 shows a schematic illustration of PEMFC operation under normal conditions with sufficient hydrogen supply for the anode and abnormal conditions of hydrogen starvation at the anode (52).

Fig. 15.

As reported in the literature (51–59), the durability of FCEVs can be significantly reduced by insufficient hydrogen oxidation reaction due to hydrogen starvation at the anode at both normal (i.e. 60~90°C) and subfreezing operation temperatures, which would eventually cause cell voltage-reversal problems. To mitigate the cell voltage-reversal degradation, a variety of system and operation control strategies, i.e. gas purging of anode compartment to remove accumulated nitrogen or water at the anode (60, 61), have been developed over the past decades. However, these techniques could limit the vehicle performance and make the vehicle system and operation more complicated. Therefore, as an alternative, material-based approaches have been suggested through adding OER catalysts to the anode, leading to an RTA (51–59). However, despite the recent progress on reducing cell voltage-reversal degradation through various techniques described above, it is not still sufficient to guarantee long-term reversal-tolerant durability and thus requires more robust and stable RTAs under acidic operation conditions of PEMFCs as well as much simpler and more effective system control technologies.

It is also critical to understand better the difference between the pristine and aged structures of fuel cell materials and components, i.e. membranes and electrodes in MEAs, GDLs and bipolar plates, on both micro- and nanoscales since the performance and durability of PEMFCs are closely related with these structural features. Therefore it is essential to develop more advanced imaging techniques, i.e. three-dimensional nanoscale X-ray computed tomography (62–64) and electron tomography performed in a high-angle annular dark-field scanning transmission electron microscope (65, 66) and correlate the imaging results with the performances and durability of actual fuel cells.

PEMFCs powered by hydrogen have received much attention as a promising candidate for FCEVs owing to their high power density, high efficiency and zero emission features. Hyundai commercialised the world’s first mass-produced Tucson ix35 FCEV in 2013, followed by the manufacturing of the second generation Nexo FCEV in 2018. To date, other global automotive OEMs, i.e. Toyota, Honda, Daimler and Audi, have also focused on commercialising FCEVs, which leads to an expansion of the global market of PEMFCs for a broad range of applications. Hydrogen is regarded as an excellent alternative to fossil fuels. In comparison with the existing grey hydrogen produced by conventional fossil fuels, the share of green hydrogen produced by excess renewable energies is expected to increase considerably in the future. Hydrogen can serve as a versatile energy carrier and plays an essential role in decarbonising major sectors of the economy.

Recently the global automotive industry has been experiencing a paradigm shift from traditional internal combustion engine vehicles to next-generation vehicles based on future mobility concepts such as CASE. These technologies are closely interlinked with each other. The FCEVs could be a strong candidate for the platform of autonomous vehicles owing to their longer driving range over 600 km and shorter refueling time usually less than 5 min.

Over the last decade, Hyundai has been actively increasing the commitment to commercialising FCEVs due to their versatile potential in the future power systems. In comparison with its predecessor Tucson ix35 FCEV, the durability of the Nexo FCEV approximately doubled from 4 years/80,000 km to 10 years/160,000 km and the driving range on a single charge increased considerably from 415 km to 609 km. The cold start-up capability of the Nexo FCEV was greatly improved from −20°C to −30°C. The Nexo FCEV was also awarded the highest rating in safety from the European crashworthiness test.

The global markets for PEMFCs have been growing very rapidly in terms of both passenger vehicles and medium- and heavy-duty vehicles such as buses and trucks, which require much higher durability than passenger vehicles. The PEMFCs are also in demand for other applications such as trains, trams, power generators and vessels. Hyundai will produce 700,000 PEMFC systems by 2030. To realise this vision, it is of paramount importance to develop more durable and cost-effective fuel cell materials, components and systems as well as advanced fuel cell operation techniques. It includes the development of highly durable membrane, more cost-effective cathode catalysts, RTA and advanced imaging techniques.

Lithium molybdate (LMO) is a hygroscopic material which is congruently soluble in water and was one of the first materials used to demonstrate cold sintering (4, 13, 18). LMO is conventionally sintered at 540°C but via cold sintering it can be densified at 120°C with the addition of 2–10 wt% distilled water under applied pressure (12, 17, 18). The properties of cold sintered LMO are comparable with conventional samples but a slight increase in dielectric loss is thought to relate to residual hydroxyl groups at the grain boundaries (12).

Materials related to LMO such as sodium molybdate (NMO) can also be densified via cold sintering. Whilst NMO is not hygroscopic, it is highly soluble in water making it another ideal candidate for cold sintering. NMO is conventionally sintered at 610°C, whilst the material can be cold sintered at 150°C with the addition of 5–10 wt% water and the application of 200 MPa of pressure. Wang et al. achieved relative densities of 87% after conventional sintering but 96% after cold sintering. The dielectric properties of NMO are also comparable between conventional and cold sintered. An increase in permittivity (ɛ_r) is observed (conventional: 11.6, CSP: 12.7), due to increased density whilst residual hydroxyl groups increased dielectric loss (19, 20).

As LMO and NMO are readily densified via cold sintering, they have been used as a starting point to create many composites with other materials with more favourable properties, but which are harder to cold sinter such as barium hexaferrite, sodium bismuth molybdate and bismuth lithium vanadium molybdate (19, 21, 22).

Zinc oxide has also gained interest as a material that can be densified via cold sintering. ZnO is wide band-gap (3.4 eV) semiconductor traditionally used in electronics such as varistors and requires temperatures in excess of 1100°C to sinter (23, 24). The high temperatures for conventional sintering leads to grain coarsening and other effects deleterious to the electrical properties and therefore methods of reducing sintering temperature have been widely investigated. Several studies have successfully cold sintered ZnO at ≤300°C. Unlike molybdate compounds, ZnO has limited solubility in water and therefore an alternative transient solvent such as aqueous acetic acid or zinc acetate (Zn-Ac) is utilised to achieve sufficient dissolution to promote densification (25, 26).

Recent work at The University of Sheffield has investigated the effects of pressure and temperature on the cold sintering of ZnO. Samples of ZnO were produced at temperatures 125–300°C and pressures of 187–375 MPa with 25–30 wt% of 1 M acetic acid. Scanning electron microscopy (SEM) of cold sintered ZnO showed that temperature and pressure have a significant effect on grain growth and morphology, helping to corroborate previous work by Funahashi et al. and Kang et al. (25, 26).

Funahashi studied cold sintering of ZnO using a wide variety of pressures, temperatures and concentrations of acetic acid (0.1–17.5 M) and water for comparison. The presence of the acetic acid was critical to achieve high density, with 1.0 M the optimal concentration. Pressures of 387 MPa combined with 300°C were reported to produce the highest densities from the test pressures and temperatures. The lower pressure of 77 MPa did produce high densities however no neck-growth was observed (26), in agreement with work done in The University of Sheffield. Pellets pressed at 250 MPa at 300°C showed high density (>98% theoretical) and grain growth but no necking, Figure 3(c). When pressure was increased to 374 MPa necking is observed (Figure 3(d)) producing a sample with a microstructure similar to conventional sintering.

Fig. 3.

Kang et al. studied a large range of processing conditions, including the use of various solvent chemicals and the effect of pressure, temperature, pH and ion concentration. Initial findings agreed with Funahashi with pressure promoting necking and temperature having the largest effect on grain growth (25, 26). However, it was proposed that pressure has a threshold below which densities are significantly affected. Beyond this threshold, densification becomes pressure independent. Other chemicals explored by Kang et al. (other than acetic acid or Zn-Ac) include hydrochloric acid, sulfuric acid, zinc chloride and zinc sulfate but only 70–75% densities were achieved, as well as unwanted cement and hydroxide phases (25). Densification was independent of pH due mainly to the presence of Zn²⁺ and acetate ions. The exchange of Zn²⁺ ions through solution enabled by applied pressure is the largest contributor to densification (25). Secondary phases observed by Kang et al. were also present, contradicting earlier suggestions that samples were single phase (26). Raman analysis showed evidence of residual acetate or Zn-Ac (25) which was confirmed in our studies. Figure 4 shows a comparison of ZnO cold sintered at 125°C and 300°C compared to a conventionally sintered sample produced at The University of Sheffield. At 125°C and 300°C, three peaks are observed which are not present in the conventional sample. The peak at ~943 cm^–1 is typical of a C–C bond, ~1435 cm^–1 a C–O bond and finally at ~2930 cm^–1 a mode characteristic of a C–H and the peak ~650 cm^–1 is still under investigation. These organic peaks are weaker in the 300°C sample, suggesting that some of the acetate has been removed. Acetate decomposes at ~225°C consistent with this hypothesis.

Fig. 4.

Gonzalez-Julian et al. densified ZnO via cold sintering and a combined cold sinter-FAST/SPS method, using Kelvin probe force microscopy (KPFM) on resulting samples, to better understand the mechanisms behind cold sintering (27). For both CSP and CSP‐SPS, powders were moistened with 1.6–3.2 wt% distilled water or a 0.5% Zn–Ac solution. Samples were then sintered according to parameters in Table II. The addition of Zn–Ac to the transient solvent was found to significantly reduce the onset temperature of densification from 90–130°C to ~25°C at all pressures. This demonstrates the crucial role of powder dissolution in densification via cold sintering.

KPFM was used to analyse the surface potentials of samples sintered via CSP-SPS at 150 MPa. The addition of water increased surface potential compared to the as-received powder, which implies an increase in defect concentration. A contrasting effect is seen with the addition of Ac-H₂O, where surface potentials are reduced indicating a lower defect concentration, which the authors attribute to the observed grain growth. The increase in surface potential indicates that the solvent phase not only promotes transport but also raises the sintering potential through the creation of activation energy lowering defects, as OH^– and H⁺ ions diffuse into the surface of the crystal structure (25–27).

From impedance spectroscopy, ZnO densified using Ac-H₂O had the highest conductivity with significantly lower total activation energy than other conditions. The bulk activation energy of ZnO was found to be significantly reduced by sintering with both water and acetate, whereas the grain boundary (E_a) was found to be increased with H₂O and decreased with Ac-H₂O. This lowering of activation energy is thought to be due to the manufacture of highly defective diffusion pathways, which helps to encourage sintering at low temperatures.

Overall Gonzalez-Julian et al. theorised the liquid phase has five main roles during the CSP in ZnO: (a) a better initial packing of the powder material due to interparticle friction; (b) dissolution of Zn²⁺ and O^2– ions from the powder surface; (c) formation of defects in ZnO crystals due to H⁺ and OH^– diffusion; (d) formations of highly defective diffusion pathways between grains and (e) elimination of carbonates. These effects are thought to be further enhanced by the presence of the acetate phase by improving dissolution. This paper indicates that the liquid phase has a more complex role than first suggested by initial studies; better understanding of defect chemistry effects is in understanding and improving the results achieved from the CSP.

To unite cold and flash sintering, Nie et al. studied the effect of an aqueous transient liquid phase on flash sintered ZnO. An electrode green body produced by uniaxial pressing was placed in a flash chamber and flowing wet argon + 5% hydrogen was introduced after 1 h. The conductivity of the hydrated pellet increased by a factor of four compared to the unhydrated form (3 × 10^–7 S cm^–1 to ~7 × 10^–3 S cm^–1). The presence of water was found to trigger flash sintering at room temperature due to the higher conductivity, producing relative densities of ~98%. The water was also proposed to assist with densification via mass transport due to partial dissolution of the substrate (28).

As already discussed, Li, NMO and ZnO are relatively easy to cold sinter and coarse powders can be densified with the addition of water or acetic acid as the transient solvent. To cold sinter a wider variety of materials, with a broader range of properties, several methods are employed, such as reducing the particle size to nanoscale, thereby increasing the reactivity of the powder and altering liquid additive to include more complex acids, alkalis or ionic solvents.

In cases where the powder dissolves incongruently, a method of hydrothermal assisted cold sintering is utilised through reactive intermediate phases. The liquid utilised in cold sintering of incongruent materials is often a solution containing a deep eutectic reaction precursor to form the desired products at temperatures below that of a solid‐state process (29–32).

When particles of BaTiO₃ are exposed to water, Ba ions leach from the surface, leaving a titanium‐rich layer (33). To cold sinter BaTiO₃, Guo et al. utilised nanoscale particles of BaTiO₃ and a barium hydroxide on titania suspension in deionised water. This prevents the dissolution of Ba during cold sintering and the Ba(OH)₂ and TiO₂ react to form BaTiO₃ during annealing at 700–900°C.

Strontium titanate is conventionally sintered at over 1400°C (34). Boston et al. developed a method of cold sintering for SrTiO₃ which utilised reactive intermediate phases (29). Nanoscale SrTiO₃ and TiO₂ powders were mixed with 0.2 ml of a 1.5 M strontium chloride aqueous solution with 1.5 M equivalent of anatase nanoparticles. The mixture was pestle and mortared to produce a free-flowing powder which was then pressed at 750 MPa for 10 min at room temperature before increasing to 180°C for 60 min. After cold sintering, a 4 h heat treatment at 950°C was utilised to promote microreactions between SrCl₂ and TiO₂ intermediate phases, forming SrTiO₃. Electrical testing of cold sintered SrTiO₃ showed similar trends to conventionally sintered materials, however the relative permittivity values exhibited frequency dependence. Particle size in the conventionally sintered samples is shown to affect the permittivity and loss (Table III).

Cold sintering is an exciting area for development in ceramic science. There are however a number of challenges which will need to be overcome to improve the commercial prospects of this new technology.

Most developments in cold sintering so far have been on an empirical, material-specific, ‘trial and error’ basis. A greater understanding of the mechanisms and how they relate to processing parameters will allow a wider range of materials to be densified via cold sintering. There are numerous processing parameters which can be altered to tailor the densification of material during cold sintering, including composition of transient liquid phase, volume of transient liquid required, pressure, temperature and powder particle size.

The transient phase should allow for the congruent dissolution of the solid phase or react to form a desirable composition upon heating during sintering or subsequent heat treatment. Therefore, it is important to understand the dissolution behaviour of the ceramic within the temperature range of sintering. The amount of liquid used during cold sintering is mostly quoted in weight percent of the solid phase. This does not consider the effect of surface area, far greater for nano- as opposed to micropowders. The purpose of pressure during cold sintering is the rearrangement of particles, but it also plays a more complex role in dissolution, grain growth and activation of reactions due to inhomogeneous pressure distributions.

The temperatures used during cold sintering are largely dependent on the evaporation point of the solvent. Grain growth has also been observed in some materials cold sintered significantly above the solvent evaporation temperature but below conventional sintering temperatures, this could be used to achieve specific grain sizes and structures.

In some cases, secondary phases can form during cold sintering, due to reactions between the solid phase and transient solvent or residual solvent after sintering completion. Kang et al. used a number of acids in the cold sintering of ZnO to observe their effectiveness as solvent phases. For a ZnCl₂ solution, significant amounts of zinc oxychloride phases were observed, with similar results produced when sulfate and nitrate based solvents were used. While the use of such strong solvents is an extreme example, it demonstrates the importance of correct solvent choice for the sintering process to prevent secondary phase formation. Even for more successful solvent phases used to densify ZnO, such as acetic acid and Zn-Ac, small amounts of residual acetate phases have been detected and affect properties.

When BaTiO₃ interacts with distilled water, Ba is leached from the material leading to an amorphous Ti-rich surface layer. This preferential leaching is overcome using a solution containing high concentrations of Ba and Ti ions but amorphous material forms which requires a further post CSP crystallisation step at high temperatures.

While the amount of energy required to densify material via cold sintering has shown to be significantly reduced, the energy of nanopowder production has not been routinely considered when evaluating the total energy consumed. To produce nanopowders significant amounts of energy or complex chemical reactions are often required, transferring the energy consumption and environmental costs to a different stage of the manufacturing process.

Microwave (MW) dielectric materials show strong interactions with electromagnetic waves, making them extremely important in modern communications as resonators, filters and substrates (22, 35). The three selective parameters of MW dielectric ceramics are high quality factor (Qf), near-zero temperature coefficient of resonant frequency (TCF) and high ɛ_r (19, 22, 35).

With fifth generation (5G) network technologies beginning to be utilised and installed in numerous countries, the material challenge is to develop systems of very high resonant frequency and low latency. Whilst fourth generation (4G) systems operate in the 2–8 GHz range, the operating range of 5G systems will eventually be up to 30 GHz. For these 5G systems, the dielectric loss of polymeric substrates used in 4G is too high and other substrates must be investigated (36–39).

Cold sintering has shown promise within this area at The University of Sheffield with the densification of several known MW ceramics achieved at low temperatures. However, none of the early materials such as LMO and NMO exhibited near zero TCF. Consequently, Wang et al. (17, 19) has developed several, two component temperature stable MW ceramic composites via CSP.

Na_0.5Bi_0.5MoO₄-Li₂MoO₄ (NBMO-LMO) composite samples were produced by mixing NBMO and LMO powders with 5–10 wt% of deionised water and pressing pellets 30 min at 150°C and 200 MPa. Sintered pellets were dried for 24 h at 120°C to remove any residual moisture (22). The NBMO‐LMO ceramic composites in this study showed no chemical reaction between the phases during cold sintering and near zero TCF was achieved at ~20% LMO with ɛ_r = 17 and Qf = 8000 GHz (22) (Figure 5).

Fig. 5.

(Bi_0.95Li_0.05)(V_0.9Mo_0.1)O₄-Na₂Mo₂O₇ (BLVMO‐NMO) composites were also sintered by combining the mixtures with 5–10 wt% of deionised water and hot pressing for 30 min at 150°C and 200 MPa. A post sintering drying step of 120°C for 24 h was performed to remove residual moisture (19). Electrical and MW analysis of the BLVMO-NMO showed similar trends to the NBMO-LMO and near zero TCF was obtained at ~20% NMO with ɛ_r ~40 and Qf = 4000, Figure 6.

Fig. 6.

Although the Qf values of these composites do not compete with conventionally sintered ceramics for resonator applications, their properties, ease of integration and low energy consumption show promise for a wide range of novel devices.

Graded-index (GRIN) lenses (19, 43) are able to convert a point electromagnetic source to a planar wave and vice versa. They are normally used in optics but CSP can be used to fabricate devices from ceramics in MW applications. A MW GRIN lens (19, 22) consists of concentric rings of material, with decreasing ɛ_r towards the outer edge of the structure, preferably reaching values close to ɛ_r = 1. Dimensions and ɛ_r of the layers are tailored to ensure they have the same focal point (O) to convert the incident spherical wave to a plane wave; a schematic GRIN lens design is shown in Figure 7 and a simulation (CST Microwave Studio, Dassault Systèmes, France) of a working GRIN lens is shown in Figure 8.

Fig. 7.

Fig. 8.

The simulated lens was illuminated with an open-ended K_a-band waveguide (7.112 mm × 3.556 mm). Across the whole frequency range the boresight directivity was increased from 26 GHz to 40 GHz, the relative increase compared to the case with no lens was between 4.6 dB and 8.5 dB. The simulated BLVMO-NMO and NBMO-LMO lenses exhibited an aperture efficiency ~70% at 26 GHz and ~78% at 34 GHz respectively (19, 22).

Due to the ability to control lateral dimensions during cold sintering, it is possible to co‐sinter multiple layers of ceramics without the detrimental effects of differential shrinkage and divergent thermal expansion coefficients. Wang et al. co‐sintered three layers of ceramic in the LMO‐NBMO system (LMO, NMBO-10 wt% LMO and NMBO-50 wt% LMO) to create a macroscopic ceramic-ceramic composite (22). This demonstrated the ability to utilise cold sintering to produce graded dielectrics and thus illustrated proof of concept for the fabrication of a MW GRIN lens. Simulations were performed to understand the potential efficiency of lenses composed of six concentric rings of radially reducing ɛ_r, illuminated with a K_α-band waveguide between 26–40 GHz. Peak aperture efficiencies were 78% at 34 GHz for the NBMO-LMO lens and 70% at 26 GHz for the BLVMO-NMO lens. Demonstrating high conversion rates between input and output of the lens.

Microstrip patch antennae (MPA) are a low-profile form of antenna that can be integrated effectively where space and weight restrictions apply and maybe printed onto polymer-based printed circuit boards (PCBs) for mobile device applications (44, 45).

At The University of Sheffield, Wang et al. made use of cold sintering to produce an MPA from a calcium titanite-potassium molybdate (CTO-KMO) composite. Composites were produced at 150°C under a uniaxial pressure of 200 MPa for 30 min, achieving high density for all compositions. From energy-dispersive X-ray spectroscopy (EDS) mapping, the authors showed that two chemically discrete regions are present in the composites after cold sintering indicating that no reaction occurs between the two phases during sintering (45).

For 5G antenna substrates, materials should have low ɛ_r (<15), a near-zero TCF and high-quality factor (45, 46). In previous studies, the dielectric properties of composites approximately follow known mixing laws and can therefore be tailored to suit specific applications. A CTO-KMO composite produced with 92 wt% KMO had TCF ~–4 ppm °C^–1, ɛ_r = 8.5 and Qf ~11,000 GHz. This composition was then used to create a cold sintered MPA which operates at 2.51 GHz with a 62% radiation efficiency (Figure 9). The combination of high antenna performance and low temperature densification demonstrate the potential for the direct fabrication of antenna substrates onto PCBs (45).

Fig. 9.

Multilayer ceramic capacitors (MLCCs) consist of alternate layers of ceramic and metallic electrode and over three trillion are produced every year (20). MLCCs are conventionally sintered at high temperatures which presents several challenges, not least the electrode melting point and chemical compatibility with the ceramic.

Capacitors are used in a wide variety of applications and conditions and they are characterised by the temperature dependency of their properties. C0G (or NP0) Class 1 dielectric materials do not show a significant variation in capacitance over a wide range of temperatures. Materials with positive and negative temperature coefficients of capacitance (TCC) can be mixed to create a temperature stable ceramic composite. Figure 10 compares the TCC of several common Class 2 and C0G capacitors.

The TCC of BLVMO and NMO ceramics are approximately +81 ppm °C^–1 and –99 ppm °C^–1 respectively. The combination of these materials creates temperature stable composites <10 ppm °C^–1 with low dielectric loss (tan δ ≈ 0.001) and ɛ_r = 40 (19, 20). Using BLVMO-xNMO (x = 0.2) composites, researchers at The University of Sheffield have demonstrated the use of cold sintering to produce multilayer ceramic capacitors with comparable properties to conventional calcium zirconate C0G MLCCs manufactured at 1100°C (20). Laminated stacks were made from tape cast BLVMO-NMO with screen printed silver electrodes. After binder burnout, at 180°C for 3 h, stacks were exposed to water vapour in a sealed beaker at 80°C. The moistened stacks were then cold sintered at 150°C under 100 MPa of pressure for 30 min. The SEM image of the cross-section of the cold sintered MLCC in Figure 11 shows the dielectric layers are well densified, well laminated and unwarped. The Ag electrodes also appear well defined indicating no reaction at the metal-ceramic interface (20). The room temperature ɛ_r and loss at 1 MHz were found to be 39 and 0.01 respectively and the TCC was within 0.013% up to 150°C.

Fig. 10.

Fig. 11.

In the crystal lattice, zinc and oxygen are arranged in tetrahedral geometry with each Zn atom surrounded by four O atoms and vice versa. ZnO exists in three crystal structures i.e., wurtzite, zinc blende and rock salt. At ambient conditions, ZnO exists in wurtzite form (11). A stable zinc blende phase can be achieved by growing ZnO on a cubic substrate (32–34). The rock salt structure can be obtained by applying very high pressure to the wurtzite structure (35). For the wurtzite structure, the lattice parameters a and b are equal and in the range 3.2475–3.5201 Å and c is in the range 5.2042–5.2075 Å. The bond between Zn and O in the crystal lattice possesses very strong ionic character. Therefore, ZnO is classed as being between an ionic and covalent compound (11).

ZnO is a direct bandgap material. Figure 2 shows the band structure of ZnO. It can be observed that in the Brillouin zone at k = 0, the lowest of the conduction band and topmost of the valence band lies at the same point. The electron configuration of Zn is 1s² 2s² 2p⁶ 3s² 3p⁶ and O is 1s² 2s² 3p⁴. In a ZnO crystal, the bottom of the conduction band is due to occupied 2p states of O^2– and the top of the valence band is due to the empty 4s states of Zn²⁺. The valence band further splits into three subvalence bands under the influence of spin that can be seen in Figure 2 (36).

Fig. 2.

ZnO exhibits n-type properties due to intrinsic defects. The defects arise because of deviation from stoichiometry. Major defects present in ZnO are oxygen vacancies (V_O) and zinc interstitials (Zn_i). However, which one of the defects dominates is still unclear (20). Due to these major defects, ZnO exhibits n-type characteristics. Figure 3 shows the defects and energy levels associated with it. In the Figure 3, Zn and O stand for zinc and oxygen respectively and V, and i correspond to vacancy and interstitial site respectively. Zn_i and V_O result in a donor level in the forbidden gap whereas Zn vacancies create an acceptor level. The V_O creates deep level donor states while the shallow level donor states are due to Zn_i. The difficulty in achieving p-type conductivity is due to the compensation of acceptor atoms by deep level donors that are the result of V_O (37). The luminescence in green, blue and violet light regions is also attributed to these defects (38). Figure 3 shows possible luminescence from ZnO due to the various defect levels.

Fig. 3.

For materials to be used in optical emitting devices, they should have direct bandgap and high exciton energy. ZnO is a direct and wide bandgap semiconductor with high refractive index (2.008). Its bandgap is around 3.4 eV at room temperature. It has an exciton binding energy around 60 meV as compared to 25 meV of GaN. Due to this, exciton recombination is possible at room temperature and above. Therefore, ZnO is a stable light emitter as compared to GaN. Because of the excitonic process, emission in the UV region (380 nm) is observed from ZnO. However, due to the intrinsic defects of lower energy states, emission of violet, blue and green light has also been observed (39–41). Therefore, ZnO is an efficient material for phosphor applications (42). Stimulated emission under optical pumping has also been observed from ZnO. This phenomenon may be due to excitonic-excitonic scattering or emission (43, 44). Electrically pumped lasing from ZnO nanowires has also been achieved by some research groups (45–47).

The conductivity of a thin film mainly depends on carrier concentration and mobility. The relation between conductivity, mobility and carrier concentration is given by Equation (i):

(i)

where n is the density of electron (hole) concentration in the conduction band (valence band), q is the charge on the electron (1.6 × 10^–19) and μ is the mobility of charge carriers. ZnO exhibits n-type characteristics due to the intrinsic defects (V_O and Zn_i). The carrier concentration and mobility highly depend on the level of defects. In 2011 Torricelli et al. (48) proposed a multi-trapping-and-release-transport mechanism for charge transport phenomena in disordered ZnO. According to this model, the conductivity can be explained as Equations (ii) and (iii):

(ii)

(iii)

where μ_b is band mobility at infinite temperature, N_b is total states per unit volume in the transport band, T_o is the characteristic temperature that accounts for the energetic disorder, N_t is the total number of trap states and n_t is the charge-carrier concentration in the trap states in the disordered ZnO. The authors assumed that the charge carriers n_t in the trap state are much greater than that of the carriers in the transport band n. Therefore, the total carrier concentration n_T was approximated as n_T = n + n_t ~ n_t. The defects and hence the carrier concentration and mobility in the ZnO highly depend on the deposition method and the growth conditions. The concentration and mobility of electrons in ZnO have been found in the range 10¹⁶~ 10¹⁷ cm⁻³ and 20~400 cm² V^–1 s^–1 respectively (11, 25, 49–51).

For high performance electronic and optoelectronic devices, high-quality metallic contact on the ZnO thin film is very important. The electrical properties of semiconductor devices are greatly affected by the contact used. The metallic contact on ZnO can be Schottky barriers or ohmic depending on the difference between the work function of the metal and the electron affinity of ZnO. For a Schottky contact on ZnO thin film, metals with high work function are required. Platinum, palladium, tantalum and gold are high work function metals that are generally used for making Schottky contact with ZnO film. Pd (φ_m = 5.12 eV) and Au (φ_m = 5.1 eV) have been reported to form the most stable Schottky barrier contact on ZnO thin films (52, 53). An ohmic contact plays an important role in the performance of devices like solar cells, TFT, varistors and LED. A good ohmic contact on a semiconductor film is characterised by a linear current-voltage (I-V) relationship and negligible contact resistance. To create an ohmic contact on ZnO, the work function of the metal should be close to the electron affinity of ZnO (χ = 4.35 eV) (54). Al, In and titanium have work function values close to 4.28 eV, their resistivity is very low and the contact resistance formed between these metals and ZnO is also negligible. Therefore, these metals can be a good choice for making ohmic contact with ZnO films.

TCO are widely used as electrodes in optical and electronic devices like displays, solar cells, LED and organic light-emitting diodes (OLED) (55). At present indium tin oxide (ITO) is used as a TCO due to its excellent transparency and conductivity but its availability is limited and this makes it very costly (56). As a result, the cost of devices incorporating ITO as electrodes is very high. ZnO is widely available, cheap and also has very good transparency in the visible region and good conductivity. Therefore, it can be an alternative choice as TCO. Highly crystalline transparent ZnO film with good conductivity is easy to process at low temperatures making it compatible with plastic and glass substrates (55, 57). The electrical conductivity of ZnO is not equivalent to ITO but the conductivity of ZnO can be modified by doping it with elements like Al, In and Ga (58). Agura et al. (59) and Jun et al. (60) have reported the lowest resistivity of Al-doped and Ga-doped thin films respectively. The reported resistivity was 8.1 × 10^–5 Ω cm for Al-doped ZnO (AZO) and 7.7 × 10^–5 Ω cm for Ga-doped ZnO (GZO) thin film. The transparency of GZO and AZO was found to be greater than 90% equivalent to the transparency of ITO (21, 61, 62). Therefore, it can be concluded that ZnO can be a good choice for TCO.

Gas sensors have many important applications like environmental pollution control, fire detection, as an alcohol breath analyser, industrial process controller or for detection of harmful gas leaks in mines and other industries (63). Semiconducting oxide-based gas sensors are easy to fabricate, have low cost and their surfaces have good sensitivity to the adsorbed gases (64). For good sensitivity, the film surface should have high grain density with a porous surface (65). ZnO being physically and chemically stable can be a good choice for thin film gas sensors. Doping ZnO with suitable elements in appropriate amounts increases the surface density of grains and porosity thereby improving the sensing selectivity and response time of the film (66). The sensitivity further improves at high temperature. The conductivity of ZnO thin film surfaces will increase or decrease depending upon the nature of reaction (oxidation or reduction) of the adsorbed oxygen on the surface of the ZnO thin film and the gas under test (65). There are numerous reports of ZnO thin film gas sensors for detecting species such as ammonia, ammonium, nitrogen dioxide, water, ozone, carbon monoxide, hydrogen, hydrogen sulfide and ethanol for various applications (17). Chou et al. (63) reported Al-doped ZnO thin film by rf sputtering method with interdigitated Pt electrodes that can be used as a breath analyser for sensing ethanol. Kim et al. (67) reported a Sn-doped ZnO thin film gas sensor for NO₂ detection with improved selectivity. Other reported works include Pd-doped ZnO gas sensors for H₂ detection by Al-zaidi et al. (68) and a ZnO thin film gas sensor by rf sputtering for H₂, NO₂ and hydrocarbon detection by Sadek et al. (69). Balakrishnan et al. (70) reported the detection of NH₃ gas by a p-type ZnO thin film. The p-type thin film was obtained by co-doping with aluminium nitride and aluminium arsenide and then depositing with rf sputtering method.

The large bandgap of ZnO and high exciton energy makes it an ideal material for blue and UV LED. ZnO is widely available and cheap, so it has an advantage over GaN from the cost point of view. The limiting factor in realising ZnO based LED was the lack of stable and reproducible p-type ZnO. The alternative approach was that n-type ZnO thin film was grown on other p-type materials like Si, GaN, zinc telluride, copper(I) oxide and GaAs (11, 27). Various ZnO based heterojunction LED in the UV and visible ranges (red, blue, green or white) have been reported. Rogers et al. (71) and Alivov et al. (72) have reported n-ZnO/p-GaN and n-ZnO and p-AlGaN LED in the UV range of 375 nm and 385 nm by pulsed laser deposition (PLD) and chemical vapour deposition (CVD) processes, respectively. Yang et al. (73) and Alivov et al. (74) have reported n-ZnO/p-GaN and n-ZnO/p-GaN/Al₂O₃ blue LED by metalorganic chemical vapour deposition (MOCVD) and CVD processes, respectively. An n-ZnO/n-MgZnO/n-CdZnO/p-MgZnO LED emitting red light has been reported by Ohashi et al. (75) by a MOCVD process. Chichibu et al. (76) fabricated greenish-white LED using helicon-wave-excited plasma-sputtering from p-type copper gallium sulfide heterojunction diodes using n-type ZnO as an electron injector. They also reported IR-LED (780 nm) by using a p-CuGaS₂/n-ZnO-Al structure fabricated by the helicon-wave-excited plasma-sputtering method (77). Earlier it was difficult to achieve p-type doping in ZnO but, at present, several researchers have reported p-type ZnO and homojunction LED based on it. Wang et al. (78) reported p-ZnMgO/ZnO/n-ZnMgO p-n junction LED. Tsukazaki et al. (79) represented a p-i-n homojunction structure on a (0 0 0 1) ScAlMgO₄ substrate. The p-type conductivity was achieved by doping ZnO with nitrogen. Ryu et al. (80) fabricated arsenic-doped p-type ZnO and demonstrated (Zn, Be)ZnO/n-ZnO-based LED. Lim et al. (81) fabricated p-ZnO/n-ZnO/sapphire LED by rf sputtering method.

For short-wavelength semiconductor laser diodes, wide bandgap materials are ideal (82). At present blue and UV lasers are based on GaN materials (83). Because of the large exciton binding energy of 60 meV as compared to 25 meV of GaN, ZnO could be a promising material for UV and blue laser applications. The lasing phenomenon in ZnO occurs due to exciton-exciton scattering. Various researchers have observed stimulated emission from ZnO (84–87). Stimulated emission from the surface and edges of a ZnO thin film is observed under optical pumping. ZnO has high excitonic energy hence lasing is observed under moderate pumping. Therefore, ZnO-based lasers have a low threshold value (11). Ozgur et al. (83) in 2004 reported low threshold exciton-exciton scattering-induced stimulated emission in rf-sputtered ZnO thin films. Random stimulated emission from a ZnO polycrystalline thin film was observed by Cao et al. (88). Gadallah et al. (89) in 2013 reported surface and edge emission under optical pumping from a ZnO thin film grown on a sapphire substrate by PLD with the highest gain and lowest loss (at that period). Waveguide assisted random lasing was also observed from an epitaxial ZnO thin film (90). Although there are various reports on lasing through ZnO, there are no reports on ZnO-based laser diode. The limitation in fabricating ZnO-based laser diodes was that a stable p-type ZnO thin film was not realisable. But now with the various reports on p-type ZnO (78–81), it is expected that a ZnO-based laser diode will be available soon.

A biosensor is a transducer that detects a biological response and converts it into an equivalent electrical signal. Biosensors have many important applications especially in the healthcare and food-processing industries. They are used for chemical and biological analysis. They are also used for clinical analysis and environmental monitoring. Materials to be used for biosensors should be biocompatible and non-toxic so that the biological activity of the element to be recognised is retained. It should also present a high surface area to the element to be detected for better sensitivity. The large surface-to-volume ratio and electron and phonon confinement of nanomaterials make them favourable for biosensors. ZnO due to its biocompatibility, non-toxicity and antibacterial properties is a good choice for biosensors. ZnO has a high isoelectric point (9.5) therefore elements with a low isoelectric point can also be immobilised on it through electrostatic interaction. ZnO nanostructures as biosensors can be used to detect species like hydrogen peroxide, urea, protein, glucose, human immunoglobulin G (IgG), DNA (phosphinothricin acetyltransferase (PAT) gene), phenol, catechol or cholesterol (26, 91).

A photodetector is a device that senses electromagnetic waves. It converts the optical signal into an equivalent electrical signal. If a light wave with energy greater than or equal to the bandgap of the semiconductor falls on it, then electron-hole pairs are generated. The charge pairs drift towards the anode and cathode respectively under the influence of the appropriate electric field resulting in the generation of current. This current is called photocurrent, and it is proportional to the intensity of light falling on the semiconductor. Photodetectors can be classified as photoconductors and photodiodes. Photodiodes are further classified as metal-semiconductor-metal (MSM) photodiodes, Schottky photodiodes, p-n homojunction photodiodes and p-n heterojunction photodiodes. In a photoconductor, the conductivity of the semiconductor changes under the influence of light. If light with energy greater than or equal to the bandgap of the semiconductor falls on it, an electron in the valence band absorbs energy and jumps to the conduction band. The concentration of electrons in the conduction band increases hence the conductivity also increases. Thus, the current in the external circuit under biased conditions increases proportionally to the photocurrent. A photodiode works in reverse bias mode. In a diode, a depletion region is formed at the junction of p and n regions and the width of this region increases on increasing the reverse bias voltage. In this region, there are no free charge carriers but because of thermal energy, very few electron-hole pairs may be generated. Under the influence of an electric field in the depletion region, the electrons and holes drift towards the n-region and the p-region respectively. This current has very small magnitude and is known as leakage current or dark current (current in the absence of visible light). This current depends on the ambient temperature, reverse bias voltage and external series resistance. If the diode is exposed to a light wave of appropriate wavelength, more electron and hole pairs generate, and more current flows in the external circuit. That current is the sum of dark current and photocurrent. The photocurrent is proportional to the intensity of light falling on the diode.

For efficient detection of light, the photodetector should have some desirable features. It should be sensitive in the required spectral region with high responsivity, high quantum efficiency, fast response time and small noise equivalent power (NEP). It should have low noise current in the undesired spectral range (92). ZnO is mainly used for detecting UV rays. Mollow (93) in 1940 was the first to observe the UV photoresponse of ZnO thin films. The 3.4 eV bandgap of ZnO makes it very sensitive to UV rays compared to visible and IR rays. The bandgap can, however, be tuned by doping with materials like In, Al or magnesium to make a detector for a specific wavelength. UV sensors have a wide range of applications. They are used in space applications for communication and in the military for missile warning and guiding systems. They can be used for environmental monitoring as an ozone layer monitor and for commercial purposes as a fire detector (92, 94). Materials to be used for space and military applications should be thermally, mechanically and chemically stable and should have high radiation resistance. ZnO is an ideal material with all these properties along with high gain and high photoresponse. ZnO is almost transparent in IR and in the visible region hence ZnO-based UV detectors exhibit less dark current and better sensitivity to UV rays as compared to Si-based UV detectors. Figures 4(a)–4(c) show some important structures of ZnO-based UV detectors.

Fig. 4.

5.6.1 Photoconductor

5.6.2 Metal-Semiconductor-Metal Photodiode

5.6.3 Schottky Photodiode

5.6.4 p-n Heterojunction Photodiode

5.6.5 p-n Homojunction Diode

The TFT was first patented (in 1952) as a solid-state amplifier (118). It is a three-terminal (source, drain and gate) device similar to the metal-oxide-semiconductor field-effect transistor (MOSFET) with the same working principle. It has a substrate (for providing mechanical support to the structure), a dielectric layer, an active channel layer and source/drain and gate contacts. Charge carriers are injected through the source electrode at one end and collected at the drain electrode at another end. The gate electrode is to control the flow of charge between the source and drain terminal. The dielectric layer between the gate electrode and the channel layer is to prevent the flow of charge carriers between them. The difference between MOSFET and TFT is that the channel in the TFT is formed by the accumulation of charges and in the MOSFET the channel is formed by inversion. Figures 5(a)–5(c) show the working principle of n-channel TFT and Figures 5(d)–5(e) represent the output characteristics of n-channel TFT (119, 120).

Fig. 5.

For n-channel TFT, a positive bias is applied between the drain to source and gate to source contact. The source contact is biased at 0 V. Figures 5(d)–5(e) show the graphs I_D vs. V_G and I_D vs. V_D respectively for an n-channel TFT. In Figure 5(e), Region 1 is known as the linear region. In this region V_D<<V_G and the drain current I_D is given by Equation (iv):

(iv)

where W is the width, L is the length of the channel, μ_FE is the mobility of electrons, V_th is the turn-on voltage, C_i is the capacitance of gate insulator per unit area, V_GS is gate-to-source voltage and V_DS is drain-to-source voltage. Since V_D<<V_G the drain current can be approximated by Equation (iv). It can be observed that in the linear region I_D varies linearly with V_DS, Equation (v):

(v)

Region 2 is known as the saturation region. In this region (V_G–V_th) >> V_D and the drain current is given by Equation (v). It can be seen that the value of drain current in this region is constant and does not vary with V_DS, Equation (vi):

(vi)

Based on the position of the gate terminal and source/drain electrodes there can be four possible TFT structures. Figures 6(a)–6(d) show the possible structures of TFT: (a) staggered bottom-gate, (b) co-planar bottom-gate, (c) staggered top-gate and (d) co-planar top-gate (115–117, 121–123). In the coplanar structure, the source/drain contacts and the gate contact are placed on the same side of the semiconductor/oxide interface. In staggered structures, the gate electrode is placed on one side, and the source/drain contact is placed on the other side of the semiconductor/oxide interface. The bottom gate structure is easy to fabricate, but the disadvantage is that the channel layer is exposed directly to the atmosphere. Therefore, the performance of the bottom gate TFT is easily affected by the presence of light, gases and humidity. Passivation of the channel layer is required to prevent exposure. The passivated bottom gate TFT is more stable, reliable and gives a much better performance as compared to an unpassivated one (124). The top gate structure has the advantage that the active layer is covered by the gate oxide and the gate contact, but an extra masking step is needed to fabricate it. The first TFT was based on cadmium selenide material. In 1962 Weimer et al. (125) reported the first TFT in which he used CdSe as an active channel material. In 1973, Brody et al. (126) demonstrated the use of TFT in the LCD. He used a matrix of 120 × 120 CdSe TFT for switching of pixels. But the very high cost and issues like reliability, stability and the invention of low power CMOS technology limited the research interest in TFT at that time. In 1979 le Comber et al. (127) reported the a-Si:H TFT. The channel layer was deposited using plasma-enhanced chemical vapour deposition (PECVD) and doping of hydrogen was done by a glow discharge technique. After that, it took ten years for TFT LCD to become attractive in the commercial market thereby increasing the research interest in the field of TFT. At present, active-matrix liquid-crystal display (AMLCD) technology is based on a-Si:H. But it has many disadvantages. The field-effect mobility of a-Si is very low (~1 cm² V^–1 s^–1). This makes it unsuitable for ultra-high-definition displays where very high switching rates are required. Poly-Si has a very high field-effect mobility (>50 cm² V^–1 s^–1) but the problem is that it requires a very high processing temperature (>500°C) and the crystallisation process is very time-consuming. The high temperature makes it incompatible with cheap glass and plastic substrates and hence the cost of a poly-Si TFT display is very high. Due to its polycrystalline nature, it exhibits different characteristics across the film area. Therefore, poly-Si TFT is unsuitable for large-area displays. The common problem with Si-based TFT is that they are sensitive to visible light, so a shield in the form of an array is required that blocks the backlight, therefore the resolution of the display is degraded. These limiting factors turned attention toward other materials especially to wide bandgap materials due to their insensitivity to visible light (11).

Fig. 6.

There are various reports on ZnO-based TFT dated from 1968. The insensitivity to visible light, low processing temperature, deposition of a highly crystalline thin film over a large area by conventional processes like sputtering and devices with very high field-effect mobility in the range of 0.2 cm² V^–1 s^–1 to 40 cm² V^–1 s^–1 have made ZnO a very attractive channel material for TFT (128, 129). The first ZnO TFT was reported by Boesen et al. in 1968 (130). Numerous ZnO TFT have since been reported with very high mobility as compared to a-Si TFT. Due to the transparent nature of ZnO in the visible region, it is possible to realise ZnO based fully transparent TFT. In 2003 Hoffman et al., Carcia et al. and Masuda et al. (131–133) reported fully transparent ZnO TFT. In 2008 Hirao et al. (134) demonstrated a 1.46 inch LCD with 61,600 pixels driven by bottom gate ZnO TFT arrays.

The main performance parameters for TFT are turn-on voltage, drain current on-to-off ratio (I_on:I_off) and channel mobility. Turn-on-voltage is the minimum gate voltage required to turn on the TFT. The lower the turn-on voltage, the lower the requirement of biasing voltages leading to lower power consumption. TFT with high mobility and a high I_on:I_off ratio can work at a higher frequency and are suitable for high-resolution displays. ZnO TFT with field effect mobility up to 50 cm² V^–1 s^–1 and I_on:I_off ratio greater than 10⁵ can be obtained. The performance of the ZnO TFT can be further improved by various techniques like the use of high-k dielectric, doping of ZnO and post-deposition treatments.

Memristor is of interest to many research groups as it finds important application in fields like non-volatile memory, neural networks optoelectronics, radiation sensors and neuromorphic systems. There are some reports on ZnO based memristor devices. Patil et al. (135), Fauzi et al. (136), Barnes et al. (137), Santos et al. (138) and Le et al. (139) have reported ZnO based memristor with low power and fast switching activity.

Johnson Matthey has over 200 years of history, creating sustainable technologies, shaped around customers’ needs. Our ambition is to research, develop and innovate solutions to make the world cleaner and healthier, today and for future generations. Much of the underpinning science behind these technologies relies on a knowledge of chemistry and its application. Like most successful organisations, Johnson Matthey reflects on the scientific capabilities that are key to developing these solutions today but also looks to the future to plan which capabilities will be required to meet future challenges and opportunities. Much of this learning comes from external insight by looking at what is happening both within the markets and scientific disciplines we are familiar with but also in parallel disciplines. Today, our core scientific capabilities can be grouped into nine key areas covering catalysis, characterisation and modelling, chemical synthesis, materials design and engineering, electrochemistry, platinum group metal and specialist metallurgy, process optimisation, product formulation, surface chemistry and coatings. Pulling these together forms a powerful toolbox to develop solutions for our customers’ needs.

When looking beyond these capabilities, one useful ‘lens’ to look through is the overlap between scientific disciplines. For Johnson Matthey this might be to look at the interface between chemistry, one of our key underpinning strengths, and other sciences. For example, the interface between chemistry and physics; chemistry and biology or with other enablers such as the digital transformation that is enabling different ways of exploring science. Following this premise, Johnson Matthey Technology Review has devoted this issue to focus on physics and a future edition will look at biology.

Figure 1 shows how Johnson Matthey’s core science capabilities today may overlap with physics and biology. Further insights can then be drawn by mapping how we apply these capabilities to provide customer solutions into our existing markets (pink text) or where they may be aligned with global drivers and world challenges. Such techniques can be a valuable tool to help discuss and identify opportunities and needs for an organisation.

Fig. 1.

Two areas increasingly dependent on capabilities bridging chemistry and physics are characterisation and modelling of materials and processes and the development of functional surfaces and coatings. These topics feature heavily in this edition of the Johnson Matthey Technology Review. For Johnson Matthey, characterisation and modelling are key capabilities to help develop new technology.

Characterisation provides insights into composition, structure and property-performance relationships at all length scales. The latter includes in situ and operando analysis, which is important to understanding how materials may respond in their intended application.

Modelling also encompasses all length scales and includes statistical, empirical and physical models. Modelling has been used for a long time in chemical engineering to design reactors, systems and processes. Examples include designing a new reactor for a chemical reaction, an aftertreatment system for a vehicle or a process flow sheet for recycling waste materials. More recently, advances in modelling are permitting chemists to be more predictive, to be able to design materials, reactions and their performance with far fewer experiments. For example, in this edition, the need for computational modelling methods to replace incremental experimental development to meet the need to design complex new advanced materials is explained (1).

The application of nuclear magnetic resonance (NMR) to characterising activated carbons leads to insights into kinetic exchange of solvent molecules (2). The technique makes use of the magnetic shielding properties of the carbon structure to give insights into molecular level mechanisms which can give information to the chemist about where adsorbed species are in the material’s structure. These techniques enable the industrial chemist to gain a better understanding of the materials being used which leads to faster development and better understood technology. Equally important is the fundamental understanding of new materials and their properties both at the atomic and molecular scales which in time can lead to advances in existing or new technology.

Coatings and surface properties is another area at the interface of chemistry and physics. Johnson Matthey has many examples of products which rely on the functionality of particles deposited onto a surface. Examples include precious and base metal catalysts, advanced energy materials and medical components. As the coating thickness reduces from micron to atomic, the chemist’s traditional toolbox to deposit layers of formulated slurries, pastes and inks changes towards different deposition techniques such as chemical vapour deposition (CVD) and physical vapour deposition (PVD). The ability to design and deposit functional particles of a controlled size and shape onto a surface can find application in many disciplines such as transparent or reflective coatings, semiconductor devices, energy harvesting and sensing. Typically, these applications harness a combination of electronic, optical and chemical functionality. Further examples of applications in areas such as sensing, electronics and renewable energy are explored within this edition (3, 4).

Looking forward, global drivers such as climate change, the energy transition, population growth and longevity and resource challenges will drive the need for new technologies in areas such as more sustainable products, low carbon operations, clean energy and improved health and medical care. To meet these challenges chemists will increasingly need to reach out to adjacent disciplines to develop innovative solutions. In this edition of Johnson Matthey Technology Review, we welcome you to look at some of the advances in physics and explore how they are being used to drive forward R&D.