Volume 61, Issue 4, October 2017 Published by Johnson Matthey © Copyright 2017 Johnson Matthey

ISSN 2056-5135

Johnson Matthey’s international journal of research exploring science and technology in industrial applications

Johnson Matthey Technology Review is published by Johnson Matthey Plc.

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License. You may share, copy and redistribute the material in any medium or format for any lawful purpose. You must give appropriate credit to the author and publisher. You may not use the material for commercial purposes without prior permission. You may not distribute modified material without prior permission.

The rights of users under exceptions and limitations, such as fair use and fair dealing, are not affected by the CC licenses. www.technology.matthey.com www.technology.matthey.com

Johnson Matthey’s international journal of research exploring science and technology in industrial applications

Contents Volume 61, Issue 4, October 2017

266 Guest Editorial: Johnson Matthey and The United Nations Sustainable Development Goals 2015–2030 By Sean Axon and David James

269 Light-Duty Vehicle Emissions Control: A Brief Introduction to the China 6 Emissions Standard By Huiming Gong, Yunshan Ge, Junfang Wang and Hang Yin

279 Radiolytic Conversion of Platinum, Rhodium, Osmium and Palladium Salts into Metal Coatings and Metal Nanoparticles By Takalani Cele, Philip Beukes, Thomas Beuvier, Elvia Chavez, Malik Maaza and Alain Gibaud

290 “Advances in Industrial Mixing: A Companion to the Handbook of Industrial Mixing” A book review by Li Liu

293 Artificial Photosynthesis: Faraday Discussion A conference review by Joshua Karlsson

297 Reducing the Carbon Intensity of Methanol for Use as a Transport Fuel By Alan Ingham

308 Lithium Sulfur Batteries: Mechanisms, Modelling and Materials Conference A conference review by Amina Touidjine

311 Progress and Outlook on Gasoline Vehicle Aftertreatment Systems By Ameya Joshi

326 Johnson Matthey Highlights

329 Influence of Three-Way Catalyst on Gaseous and Particulate Matter Emissions During Gasoline Direct Injection Engine Cold-start By Maria Bogarra, Jose Martin Herreros, Cruz Hergueta, Athanasios Tsolakis, Andrew P. E. York and Paul J. Millington

342 “Green Catalysts for Energy Transformation and Emission Control” A book review by Catherine Davies https://doi.org/10.1595/10.1595/205651317X696270 Johnson Matthey Technol. Rev., 2017, 61, (4), 266–268

www.technology.matthey.com

Guest Editorial: Johnson Matthey and The United Nations Sustainable Development Goals 2015–2030

It has long been recognised that sustainability and pharmaceutical products; and the efficient is crucial to business and Johnson Matthey's use of natural resources through the use of sustainability goals for 2017 reflected this (1). renewable feedstocks, the application of catalysis But the reverse is also true. Developments in and expertise in re-using, recycling and refining sustainability will be heavily reliant on business of valuable metals. The UN SDGs provide a frame involvement, and private sector companies are of reference to allow the industry to measure how expected to play a leading role in this. it can have a global positive impact. At the United Nations (UN) summit in September Although the SDG programme is only in its 2015, the UN announced the launch of the second year, many chemical companies are Sustainable Development Goals (SDGs). These already aligning their strategy with the goals. were developed after lengthy consultations For example, Dow has publically declared its involving many stakeholders including commitment through alignment of its sustainability international business, non-governmental 2025 goals with the 17 UN SDGs (3). BASF has organisations (NGOs), policy makers and civil stated that it welcomes the SDGs and contributed society. Their aim is to tackle the most pressing to their development (4). AkzoNobel provides an global issues of our time and this requires collective update on its position with respect to the SDGs in action. The UN SDGs are a universal call to action its 2016 annual report (5). to end poverty, protect the planet and ensure that all people enjoy peace and prosperity. There are Johnson Matthey and the UN SDGs 17 universal goals for 2030 (Figure 1) (2), each providing guidelines and targets for all countries Within Johnson Matthey, we have evaluated and organisations to adopt in line with their own and aligned our priorities against the SDGs priorities. They are intended to be collaborative to understand where our products, services, and bring about systemic change in the world. technologies and the way we run our business have the potential to make the greatest contribution. The Chemical Industry and the UN There are six SDGs where we believe we can make the biggest positive impact: SDGs • Good health and wellbeing So how can the chemical industry contribute to • Decent work and economic growth the SDGs? • Industry, innovation and infrastructure Overall the chemical industry and associated • Sustainable cities and communities industries have for many years provided science • Responsible consumption and production and technology solutions that address global • Climate action. challenges such as: environmental protection We have considered our ability to contribute through pollution prevention and greenhouse gas to these in the context of our areas of material abatement technologies; healthcare and human importance, risks, strategic themes and key well-being through medical devices, components performance indicators (Table I) (6).

Fig. 1. The United Nations Sustainable Development Goals (2)

Table I Johnson Matthey Performance Criteria (6)

Financial Social Environmental Governance performance

Material • Financial • Community • Air quality • Climate change risk topics sustainability volunteering • Greenhouse gas • Ethical business • Diversity and emissions practices and inclusion • Product lifecycle compliance • Employee retention management • Sustainability and recruitment • Resource scarcity leadership • Health and safety • Responsible sourcing • Modern slavery and • Water use child labour • Responsible sourcing Principal • Existing market • Environment, health • Environment, health • Environment, health risks and outlook and safety and safety and safety uncertainties • Future revenue • Sourcing of strategic • Sourcing of strategic • People growth materials materials • Ethics and • Maintaining • People • Intellectual capital compliance competitive • Security of metal management advantage and highly regulated • Failure of significant • People substances sites • Business transition • Ethics and • Product quality compliance

Strategic • Creating • Diversity and • Greenhouse gas • Community themes shareholder value inclusion emissions engagement • Sustainable growth • Zero harm • Increase positive • Responsible sourcing impact of products Key • Sales excl. precious • Lost time injury and • Carbon footprint • Gross R&D performance metals illness rate (LTIIR) expenditure indicators • Underlying earnings • Annual incidence of per share occupational illnesses • Return on invested • Voluntary employee capital turnover • Capital expenditure • R&D expenditure UN SDGs

Fig. 2. Four global drivers identified by Johnson Matthey

Johnson Matthey, Sustainability and through Johnson Matthey’s employee Our Markets volunteering programme. These themes will provide a framework as we Customer demand for sustainable, technological continue to build a sustainable business. Goal solutions to global issues defines our markets. 5 specifically will use the UN SDG framework to Johnson Matthey’s markets are driven by four track progress and the overall global impact of the global sustainability drivers (Figure 2). These four products and technologies that we provide to the drivers present significant global challenges which market and the world. Johnson Matthey believes can be addressed by SEAN AXON* science-led solutions; they also strongly align with Johnson Matthey, PO Box 1, Billingham, TS23 specific UN SDG indicators. 1LB, UK *Email: [email protected]

Johnson Matthey’s Focus to 2025 DAVID JAMES Johnson Matthey, Blount’s Court, Sonning We are continuing to apply sustainable business Common, Reading, RG4 9NH, UK principles as the basis of our strategy and in 2017/18 we will introduce new targets to drive References further progress over the next eight years to 2025. These will align to the strategic themes 1. N. Carson, Johnson Matthey Technol. Rev., 2017, 61, (1), 2 with emphasis on our contribution to sustainable development through our value chains, and 2. United Nations, Sustainable Development through our science and technology. They will also Knowledge Platform, Sustainable Development Goals: https://sustainabledevelopment.un.org/ allow us to demonstrate how Johnson Matthey uses sdgs (Accessed on 23rd August 2017) inspiring science to enhance life. Drawing on these material areas and an 3. A. N. Liveris, ‘Redefining the Role of Business to Achieve UN Sustainable Development Goals’, The understanding of the contribution we can make in Dow Chemical Company, Michigan, USA: http:// the world, we have defined six themes which form www.dow.com/en-us/science-and-sustainability/ part of our sustainable business goals to 2025. redefining-the-role-of-business-to-achieve-un- These are: sustainable-development-goals (Accessed on 23rd 1. For health and safety, aspire to zero harm August 2017) 2. Improve sustainable business practices in our 4. ‘UN Sustainable Development Goals’, BASF, supply chains Lugwigshafen, Germany: https://www.basf. 3. Reduce our greenhouse gas emissions per unit com/en/company/sustainability/employees-and- of production output society/goals.html (Accessed on 23rd August 4. Ensure that Johnson Matthey is truly 2017) inclusive, fostering employee engagement 5. ‘AkzoNobel supporting the UN Sustainable and development within a diverse and global Development Goals’, AkzoNobel, Amsterdam, The workforce Netherlands: http://report.akzonobel.com/2016/ 5. Increase the positive impact that our products, ar/case-studies/sustainable-development-goals. services and technologies contribute to a html (Accessed on 23rd August 2017) cleaner, healthier world 6. “2017 Annual Report and Accounts”, Johnson 6. Increase our work within our local communities Matthey Plc, London, UK, 2017

www.technology.matthey.com

Light-Duty Vehicle Emissions Control: A Brief Introduction to the China 6 Emissions Standard The key regulation improvements and areas for further developments are reviewed

Huiming Gong, Yunshan Ge Introduction National Laboratory of Auto Performance and Emissions Test, Beijing Institute of Technology, China has been the world’s largest vehicle market Beijing 100081, China; and Collaborative since 2009 and in 2016 annual new vehicle sales Innovation Center of Electric Vehicle in Beijing, exceeded 28 million, representing 13.7% growth Beijing 100081, China compared to 2015 despite the poor economy, as shown in Figure 1 (1, 2). As a result, the vehicle Junfang Wang and Hang Yin* population on the road reached 194 million (not Vehicle Emissions Control Center, China including about 100 million motorcycles) by the Research Academy of Environmental Sciences, end of 2016 (3, 4), among which more than 87% Beijing 100081, China were LDV. Vehicular emissions have become the leading contributor to air pollution in Tier I cities *Email: [email protected] such as Beijing, Shanghai, Guangzhou, Shenzhen and Hangzhou, and are increasingly contributing to pollution in other cities due to industry relocation and coal consumption control (5, 6). In order to control vehicle emissions and reduce the impacts on air quality and public health, China has been the world’s largest new vehicle the Ministry of Environmental Protection (MEP) market since 2009 and new vehicle sales designated the task of developing the China 6 exceeded 28 million in 2016, among which more emissions standards for both LDV and heavy- than 87% were light-duty vehicles (LDV). In order duty vehicles (HDV) to China Research Academy to reduce emissions and control air pollution of Environmental Sciences (CRAES) in 2015. This China has recently adopted the China 6 emissions paper introduces the emissions standard for LDV standard for LDV which is 50% more stringent and summarises the key technical contents. than China 5. Besides strengthening the tailpipe In order to complete the task in a short emissions limits, China 6 changes the emissions period CRAES formed a core working team test driving cycle to the Worldwide Harmonised including Beijing Institute of Technology, China Light-Duty Vehicle Test Cycle (WLTC), adds real Automotive Technology and Research Center, road emissions requirements and significantly Xiamen Environment Protection Vehicle Emission strengthens evaporative emissions control. This Control Technology Center and Beijing Vehicle paper introduces the standard development Emissions Management Center, and established background, summarises the key technical five technical task groups with more than 40 improvements and discusses the areas for further automotive manufacturers participating. The improvements in future. five technical tasks included test technologies

8 Number of vehicles, millions Number of vehicles, 6

2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 Year

Fig. 1. Annual new vehicle sales in China since 2000

and facilities, reference fuels, fuel evaporation, China 6 emissions standard for LDV, i.e. limits and test cycles, procedure and limits, and the on- measurement methods for emissions from LDV board diagnostics (OBD) system. The standard (China 6). The standard will be implemented in two development process could be divided into four phases: by 1st July 2020 all new LDVs sold and phases: the first phase was aimed at aligning registered in China must meet China 6a and by opinions on the necessity of developing China 6 1st July 2023 all new LDVs sold and registered in and conducting preliminary analysis on some China must meet China 6b. technical issues such as particulate number (PN) and OBD before 2015; the second phase Key Regulation Improvements finished the working group formation, working plan development, task allocation, international 1. Emissions Test Driving Cycle emissions regulations collection and translation, capacity building and information and experience China adopted the New European Driving Cycle exchange with field stakeholders by training, (NEDC) as the test cycle from China 1 to China 5 workshops and study tours from January 2015 since 2000. NEDC is simple and easy to duplicate, but to June 2015; the third phase had more than 40 does not well represent real road driving conditions workshops and meetings, established solutions in China. Therefore, China considered changing the for most technical problems, and developed the test cycle either by developing its own driving cycle draft standard from July 2015 to April 2016; and or by choosing an alternative. The working group the fourth phase completed the standard public compared the differences between the NEDC, the commenting, technical review and administration US Environmental Protection Agency (EPA) Federal approval from May 2016 to December 2016. Test Procedure (FTP-75) and the WLTC, and found On 23rd December 2016 the MEP and General that the WLTC widely represents various driving Administration of Quality Supervision, Inspection conditions and covers a much broader engine and Quarantine (AQSIQ) jointly published the final speed and loading range (Figures 2 and 3).

26 WLTC

24 NEDC

Engine load, bar 10

0 500 1000 1500 2000 2500 3000 3500 Engine speed, rpm

Fig. 2. Comparison between NEDC and WLTC

25 WLTC FTP-75

10 Engine load, bar

0 0 1000 2000 3000 4000 5000 Engine speed, rpm

Fig. 3. Comparison between FTP-75 and WLTC

In order to test how well the WLTC represents consumption. In summary, WLTC resulted in higher driving conditions in China, the working group fuel consumption than NEDC and there was also a collected driving cycles in 20 representative cities smaller difference in the fuel consumption between with different city scales, altitudes, locations and WLTC and the real road at about 14.0% compared terrain. Five cars were used to chase the traffic to about 22.5% between NEDC and real road fuel in each city during working days covering both consumption (Figure 6). peak hours and off-peak hours and weekends in In addition to the analysis above, it was taken different areas including urban, urban-rural and into account that China had participated in the rural areas on different roads for five days. Based discussion and development of the WLTC and on the collected data, the working group built a procedure and committed to deploy the standards. China Light-Duty Vehicle Driving Cycle (CLDC) Therefore the working group confirmed that WLTC and compared the characteristics with the WLTC would be used to replace NEDC in China 6. (Figure 4 and Table I). Across the ten characteristics at four speed ranges 2. Tailpipe Emissions Limits and the results showed the frequency with less than Implementation ±10% variance between WLTC and CLDC was 58% and the frequency with less than ±20% variance China is facing growing challenges to clean up was 80%. But some characteristics, such as idle its air. If China continued following the European ratio at ultra-high speed, had as high as 83% regulation, which will phase in the WLTC from variance, which will require attention in future. 2017 but keep the Euro 6c limits without further The working group further tested the emissions strengthening, it would be difficult for China to of seven cars meeting China 4 and above with five reduce emissions from its fast growing LDV fleet. different driving cycles including NEDC, WLTC, FTP- Considering the Tier 3 emissions standards in 75, Vehicle Emission Control Center (VECC) (1) the USA as a benchmark as well as available and CLDC. The results showed that the emissions technologies on the market, the working group during the WLTC were in the middle compared therefore suggested 40–50% more stringent to the emissions with all other driving cycles. limits (Figure 7). Diesel vehicles and gasoline Compared to the FTP-75, the WLTC produced a vehicles were combined to comply with the same higher concentration of gaseous pollutants despite emissions limits. In addition, the standard adds lower particulate concentration, indicating that particle number limits for gasoline vehicles for the WLTC is more stringent than the FTP-75 to the first time. In order to provide enough leading control gaseous pollutants (Figure 5). The fuel time China 6 is going to be implemented in two consumption of two cars were also tested with the phases: China 6a by 1st July 2020 and China 6b by WLTC and the NEDC and compared to real road fuel 1st July 2023.

140

120 CLDC WLTC 100 –1

60 Speed, km h 40

0 0 300 600 900 1200 1500 1800 Time, s

Fig. 4. Comparison between WLTC and CLDC

Table I Characteristics of the World Harmonised Light-Duty Vehicle Test Cycle (WLTC) and China Light-Duty Vehicle Driving Cycle (CLDC) –1 –1

Speed Cycle range –2 –2 Driving distance, km Average speed, km h Maximum speed, km h Maximum acceleration, m s Maximum deceleration, m s Relative positive acceleration Acceleration, % Deceleration, % Idle, % Steady, %

WLTC 3.09 18.9 56.5 1.47 –1.47 0.205 28.4 31.1 24.5 15.8 Low CLDC 2.85 17.4 52.4 1.24 –1.93 0.205 28.4 23.8 29.2 18.7

WLTC 4.76 39.5 76.6 1.58 –1.50 0.196 36.0 30.3 10.6 23.1 Middle CLDC 5.15 42.8 74.3 1.03 –1.67 0.190 31.4 30.9 12.0 25.6

WLTC 7.16 56.7 97.5 1.58 –1.50 0.132 29.0 27.7 6.4 36.9 High CLDC 7.08 56.1 93.7 1.31 –1.89 0.169 31.2 25.9 4.4 38.5

WLTC 8.25 92.0 131.3 1.03 –1.22 0.125 37.2 32.2 1.5 29.1 Ultra-high CLDC 8.45 94.1 128.5 1.21 –1.29 0.174 35.3 31.6 2.8 30.3

WLTC 23.27 46.5 131.3 1.58 –1.50 0.152 31.9 30.2 12.6 25.3 Overall CLDC 23.54 47.1 128.5 1.31 –1.93 0.179 31.1 27.4 14.1 27.4

(a) (b)

–1 0.08

y = x –1 R2 = 1 0.06 y = x 1 R2 = 1 0.04

FTP-75, g km 0.02 FTP-75, g km 0 0 0 0.5 1 1.5 2 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 WLTC, g km–1 WLTC, g km–1

–1 y = x 0.04 R2 = 1

0.02 FTP-75, g km

0 0 0.01 0.02 0.03 0.04 0.05 0.06 WLTC, g km–1 Fig. 5. Emissions comparison between FTP-75 and WLTC: (a) CO; (b) hydrocarbons; and (c) NOx

(a) (b)

6 8.5

8 5.5 7.5

7 5 per 100 km per 100 km 6.5 Fuel consumption, litres 4.5 Fuel consumption, litres 6 Real-life NEDC WLTP Real-life NEDC WLTP

Fig. 6. Fuel consumption comparison between WLTC, NEDC and real road driving: (a) petrol car, 1.2 l, manual; (b) petrol car, 1.6 l, manual

180

160

140

–1 120

100 China 5

80 China 6a China 6b 60 Euro 6c HC + NOx, mg km 40

0 0 200 400 600 800 1000 1200 CO, mg km–1

Fig. 7. Limits comparison of China 5, China 6a, China 6b and Euro 6c

The feasibility of meeting the standard was the potential China 6b limits directly. Therefore, evaluated by analysing China 5 vehicle type it is clear that there is no significant difference approval data and conducting emissions tests on between the compliance rates of China 6a (18%) 44 cars following the World Harmonised Light-Duty and 6b (16%). In addition, although the compliance Vehicle Test Procedure (WLTP) and the WLTC. rates for each category of air pollutants might Analysis of type approval emissions data of 8600 be high, considering all six pollutants, only 16% LDV (complying with China 5) after 160,000 km cars could meet China 6b. Not all the cars had a deterioration showed 74% and 33% of these gasoline particulate filter (GPF) and only 30% of vehicles could meet China 6a and 6b respectively. the cars met the PN limits, which suggests that Table II summarises the China 6 compliance rates GPF will be essential in future for gasoline vehicles of 44 gasoline cars (complying with China 5, Euro to meet the standard. 6c and US Tier 2/3) based on the emissions tests; The standard also includes the control of greenhouse

18% and 16% of these 44 cars could meet China gas emissions such as nitrous oxide (N2O) and

6a and 6b respectively. The results proved China 6 carbon dioxide (CO2). Automakers only need to is stringent but also achievable. report the CO2 value for the vehicle type and do not Automotive manufacturers sent these 44 cars need to meet any emissions limits because there with the goal to test whether their cars could meet is already a mandatory fuel efficiency requirement

Table II China 6 Compliance Rates of Selected Gasoline Cars

HCa CO NOx NMHCb PMc PNd Overall

6a 95% 80% 96% 96% 81% 30% 18%

6b 95% 72% 81% 93% 70% 30% 16% aHC = hydrocarbon bNMHC = non-methane hydrocarbons cPM = particulate matter dPN = particulate number regulated by the Ministry of Industry and Information their RDE results to the laboratory test emissions Technology (MIIT). The main purpose is to avoid with different driving cycles. Results showed the any policy leakage, such as optimising vehicles to RDE of nitrogen oxides (NOx) were significantly meet different regulations with separate tests, due higher than most laboratory test results and could to separate administration of tailpipe air pollutants be as high as 8.6 times (Figure 8). emissions by MEP and fuel efficiency by MIIT. Because RDE is a new emerging regulation requirement and may still need further evaluation 3. Real-Road Emissions and demonstration, automotive manufacturers only need to monitor and report the RDE results of It is real emissions on the road that matter to air NOx, PN and carbon monoxide before 1st July 2023 quality. In order to reduce the potential risks of and the proposed conformity factor for both NOx high emissions on the road despite good emissions and PN, the value of which is 2.1, needs further test results in the laboratory, the working group confirmation before 1st July 2022. referred to the European regulations and included real driving emissions (RDE) requirements in the standard. However, it expands the altitude boundary 4. Evaporative Emissions Control from 1300 m in the European regulation to 2400 m considering many land areas in China have much With more stringent tailpipe emissions control of higher altitudes. Evaluation was conducted with the evaporative emissions of reactive organic gases several gasoline and diesel vehicles by comparing from refueling, running loss, permeation, hot soak

0.07

0.06

0.05

–1 0.04

0.03 RDE

NOx, g km Laboratory 0.02

0.01

Vehicle A Vehicle B Vehicle C Vehicle D Vehicle E Vehicle F Vehicle G Vehicle H Vehicle I Vehicle J

Fig. 8. NOx emissions comparison between RDE and lab tests

275 © 2017 Johnson Matthey http://dx.doi.org/10.1595/205651317X696199 Johnson Matthey Technol. Rev., 2017, 61, (4) and diurnal are getting much more important. to 0.7 g per test, increases the test temperature China has followed the European regulations and from within the range 20ºC–30ºC to 38±2ºC and requires all LDVs to control evaporative emissions. simplifies the diurnal emissions test from 72 h in However, the regulation is not stringent and does the USA standard to 48 h. In addition, borrowing not include refueling evaporative emissions. As a experience from the USA the new standard adds result, the evaporative emissions in China could be control of refueling evaporative emissions and sets as high as 8800 g per year per vehicle compared the limits at 0.05 g l–1, equal to the US Tier 2 level. to 500 g per year per vehicle in the USA. The new The working group tested the vehicle evaporative standard significantly strengthens the evaporative emissions of ten gasoline LDVs which meet the emissions limits and adds refueling emissions China 5 standard with the new test procedures control. Compared to the previous standard, and found not all the vehicles could meet the new the new standard reduces vehicle evaporative emissions limits, with the highest emissions at emissions limits from 2.0 g per test previously about 5.8 g per test (Figure 9). Two of the ten

4 Hot soak

3 24 hours

48 hours 2

Evaporation, g per test Evaporation, Results 1

Vehicle A Vehicle B Vehicle C Vehicle D Vehicle E Vehicle F Vehicle G Vehicle H Vehicle I Vehicle J

Fig. 9. Evaporative emissions of ten gasoline cars

0.7

0.6

0.5

Hot soak 0.4 24 hours

0.3 48 hours

Results 0.2 Evaporation, g per test Evaporation,

0.1

0 Vehicle AA Vehicle BB

Fig. 10. Evaporative emissions after retrofitting improvements

276 © 2017 Johnson Matthey http://dx.doi.org/10.1595/205651317X696199 Johnson Matthey Technol. Rev., 2017, 61, (4) vehicles were tested again after being retrofitted and the fuel efficiency test, there could be a risk by increasing the size of the carbon canister and of policy leakage and automotive manufacturers improving heat insulation. The results showed the could optimise their vehicles to meet the two evaporative emissions were reduced significantly, regulations separately, while on-the-road air even lower than 0.35 g per test as required in pollutants emissions and fuel efficiency might not US Tier 3 (Figure 10). These results proved the be able to meet both regulations. limits are achievable in China and could be further Another group in CATARC is now developing the strengthened in future. Phase V fuel economy standard for light-duty passenger vehicles, which will phase in from 2021 5. Other Key Improvements and aim to achieve an average of 4 l per 100 km by 2025. MIIT plans to complete CATC in 2017 Besides the major improvements mentioned above, and deploy it in this new fuel economy standard. the China 6 standard has additional changes as Having a consistent test cycle and procedure below: (a) increasing the emissions durability will not only help achieve both goals of reducing mileage from 160,000 km in China 5 to 200,000 air pollutant emissions and fuel consumption km in China 6b while keeping the mileage the same at the same time, but will also help automotive as in China 6a and allowing standard catalyst bench manufacturers reduce the cost of complying with ageing for durability testing; (b) strengthening cold two regulations. start emissions at low temperature by requiring Considering the technical capacity of domestic NOx emissions control at 0.25g km–1 and 33% automotive manufacturers and vehicle emissions more stringent limits for other air pollutants; (c) test organisations and significant changes to improving the OBD system by preventing trouble previous standards, with some new requirements codes from being deleted without fixing the introduced for the first time, the standard still has problem, including in-use performance ratio (IUPR) the potential to be more stringent than the US Tier for OBD, adding monitoring items with specific 3 in some aspects, such as OBD requirements, conditions and requiring key components of hybrid evaporative emissions and durability mileage. The vehicles, air conditioning systems and cold start RDE also need close monitoring and evaluation to emissions control to be monitored; (d) including ensure the effectiveness of the regulation. hybrid vehicle emissions testing and limits; and (e) simplifying the process of deciding new vehicle Acknowledgement product conformity and in-use vehicle compliance. The China 6 standard development has funding Discussion support from the MEP and the Energy Foundation China. The authors would like to thank the China In order to control the oil consumption and Automotive Technology and Research Center, greenhouse gas emissions from the transportation Xiamen Environment Protection Vehicle Emission sector, China’s MIIT is adopting mandatory light- Control Technology Center, Beijing Vehicle duty passenger fuel economy standards and aims Emissions Management Center, and automotive to achieve an average of 5 l per 100 km by 2020. manufacturers for their support. The current fuel efficiency test methodology is based on the NEDC, which is different from the WLTC used in the China 6 emissions standard. At References the same time, MIIT has designated the China 1. “Economic Operation of Automobile Industry Automotive Technology and Research Center in 2016”, Ministry of Industry and Information (CATARC) to develop a new driving cycle, known Technology, Beijing, China, 17th January, 2017 as the China Automotive Testing Cycle (CATC), 2. Annual Statistic Data, National Bureau of Statistics which could better represent the driving conditions of the People’s Republic of China, Beijing, China: in China so that the fuel consumption consumer http://www.stats.gov.cn/tjsj/ (Accessed on 12th experience on the road could be similar to the test July 2017) results in the laboratory. Whether the test cycle 3. “Director of the Department of Environmental used is NEDC or CATC, if a different driving test Protection of the Ministry of Environmental cycle is used for the air pollutant emissions test Protection on the Issue of Light Vehicle Country

Six Standards”, Index Number 000014672/2016- Protection: Motor Vehicles Become the Primary 01459, Ministry of Environmental Protection of the Source of Fine Particles in Many Large and People’s Republic of China, 23rd December 2016 Medium-Sized Cities”, ed. C. Xi, Xinhau Network 4. “2016 National Motor Vehicles and Drivers Co Ltd, Beijing, China, 7th January 2010 Maintained Rapid Growth: The New Registered Car 27.52 Million New Drivers 33.14 Million People”, 6. J. Mengwei, “Ministry of Environmental Protection Traffic Management Bureau of the Public Sector Security Ministry, Beijing, China, 10th January, 9 City Pollution Source Analysis: Motor Vehicle 2017 into the Air Pollution Culprit”, Beijing Business 5. D. Jun and G. Jing, “Ministry of Environmental Daily, Beijing, China, 2nd April, 2015

The Authors

Mr Huiming Gong is a PhD candidate in the Beijing Institute of Technology, China. He also works as the Program Director of Transportation Program under the Energy Foundation China. Before joining the Energy Foundation China in 2005, he served as the Program Officer of the Auto Project on Energy & Climate Change under the Global Environmental Institute. Mr Gong received two Masters’ degrees, one from the University of California, Riverside, USA, majoring in Environmental Toxicology and the other in Peking University, China, majoring in Environmental Sciences.

Yunshan Ge De, a university professor, is the director of the Laboratory of Auto Performance and Emission Test at Beijing Institute of Technology, China. He is primarily engaged in research and development work, such as motor vehicle emission standards and testing technology, vehicle emission regulations and measurement technology, alternative fuels, vehicle noise and emissions control.

Dr Junfang Wang graduated from Beijing Institute of Technology. She has worked for the Vehicle Emission Control Center of the Ministry of Environmental Protection of China since 2004. Her research focuses on emission inventories and pollution control from diesel vehicles. Dr Wang has experience of drawing up national emission standards for mobile sources and policy evaluation research on mobile sources standards and regulations.

Dr Hang Yin graduated from Beijing Institute of Technology. He joined the Vehicle Emission Control Center of Chinese Ministry of Environmental Protection in 2002. He currently leads policy and regulatory research department of VECC. During his 10 years working for VECC, Hang Yin’s work mainly focused on developing mobile source emission standards and in-use vehicle management. Hang Yin received a master degree in Environmental Science from the Chinese Research Academy of Environmental Sciences and a Bachelor’s degree in Mechanical Engineering from Tianjin University in Tianjin, China.

www.technology.matthey.com

Radiolytic Conversion of Platinum, Rhodium, Osmium and Palladium Salts into Metal Coatings and Metal Nanoparticles Using intense gamma ray irradiation of precious metal salts to synthesise nanoparticles

Takalani Cele* UNESCO-UNISA Africa Chair in Nanosciences- We herein report on the effect of gamma Nanotechnology, College of Graduate Studies, ray radiation on platinum, osmium, rhodium University of South Africa, Muckleneuk Ridge, and palladium salt solutions for synthesis of PO Box 392, Pretoria, 0001, South Africa; and nanoparticles. Pt, Os, Rh and Pd salt solutions iThemba LABS-National Research Foundation, were exposed to intense gamma ray irradiation PO Box 722, Somerset West, 7129, with doses varying from 70 to 120 kGy. The South Africa metal ion salt solutions were easily converted into metal nanoparticles using this radiolysis method. Philip Beukes The radiolytic conversion effect produced metal nanoparticles suspended in solution. For Pt, Pd and iThemba LABS-National Research Foundation, Rh a metal coating on the edges of the polypropylene PO Box 722, Somerset West, 7129, tube used as a container was unexpectedly observed South Africa but not for the Os solution. X-Ray diffraction (XRD) and high resolution transmission electron Thomas Beuvier, Elvia Chavez microscopy (HRTEM) analyses confirmed that both L’Université Nantes Angers Le Mans (L’UNAM), the coating and the metal nanoparticles correspond Institut des Molécules et Matériaux du Mans to the pure metal coming from the reduction of the (IMMM UMR 6283 CNRS), Avenue Oliver initial salt. Quantitative analysis of the XRD patterns Messiaen, 72085 Le Mans, Cedex 9, France shows information about the size and stress of the converted metals. The production of a metal Malik Maaza coating on polypropylene plastic tubes by gamma UNESCO-UNISA Africa Chair in Nanosciences- ray irradiation presents an interesting alternative to conventional techniques of metal deposition Nanotechnology, College of Graduate Studies, especially for coating the inner part of a tube. University of South Africa, Muckleneuk Ridge, PO Box 392, Pretoria, 0001, South Africa; and iThemba LABS-National Research Foundation, 1. Introduction PO Box 722, Somerset West, 7129, South Africa Plastic materials are omnipresent in our daily life and are used in many industrial areas such Alain Gibaud** as packaging, microelectronics, medical devices, L’UNAM, IMMM UMR 6283 CNRS, Avenue Oliver automobile parts or simply as plastic bags. In Messiaen, 72085 Le Mans, Cedex 9, France many high technology applications, plastics cannot be used on their own as they do not Email: *[email protected], meet the necessary requirements for expected **[email protected] applications. Metallisation is a way to modify

279 © 2017 Johnson Matthey http://dx.doi.org/10.1595/205651317X696207 Johnson Matthey Technol. Rev., 2017, 61, (4) the surface of plastics for aesthetic, mechanical, To produce metal nanoparticles from metal salt – biocidal and electrical purposes (1). The packaging solutions, the well-known reducing agents are e aq and microelectronics industries are major users of and H•. Unfortunately, the production of hydroxyl metal-coated plastics. Metallisation can increase radicals OH• hampers the efficiency unless some their gloss and reflectivity but also provide barriers specific hydroxyl scavenger such as isopropanol is for ultraviolet (UV) light degradation and enhance used. Note that as pointed out in the early research plastics’ physical properties. Properties, such studies (25, 26), the free radical (CH3)2C•OH as abrasion resistance and electric conductivity, produced by scavenging has the peculiar property of which are not innate characteristics of plastic, can being a strong reducing agent and also contributes be produced through metallisation. To metalise a to the reduction of metal ions (27). piece of plastic common methods such as vacuum Radiolysis has been widely used to produce gold metallisation, arc and flame spraying or plating are and silver nanoparticles. A number of studies utilised (2–5). have been conducted to investigate these metal In this paper, we present an alternative method nanoparticles using UV-visible spectroscopy with based on the use of intense gamma ray irradiation the aim of analysing their plasmonic absorption of metal salt aqueous solutions using a radiolysis band (28–30). In other studies, gamma ray method. This method was initiated in the 1980s irradiation was used to trap metal nanoparticles by the pioneering work of Belloni and Delcourt’s inside polymers or porous frameworks, for instance group using the reducing radicals produced in water mesoporous silica (31–33). There has also been a radiolysis and the method was applied to some study of the catalytic properties of Pt nanoparticles noble metals (6–9). Deposits of metal clusters were on a glass support prepared by radiolysis (34). also obtained on smooth transparent tin(IV) oxide However, this technique has not been reported

(SnO2) and indium tin oxide (ITO) electrodes (10). to fabricate thick coated films on the surface of This study is focused on the radiolysis of platinum vessels in which solutions are gamma irradiated group metals (pgms) that are far less studied given and almost nothing has been published on the their high price compared to gold or silver. Four reduction of Os and Rh salts by this technique metals were chosen in this group: Pt, Os, Rh and (35–37). In addition, the literature on the use of Pd (11–13). The radiolytic conversion of metal highly concentrated solutions is also quite sparse salt solutions into metal nanoparticles is a very very likely because of the very expensive cost of well-known process of great interest as it does not the precursors. necessitate the use of polluting solvents (14). It is In this paper, we report on the reduction of indeed a versatile and powerful way to synthesise metal salts solutions of precious metals (Pt, Rh, metal nanoparticles of controlled size and shape Os and Pd) which were stored inside polypropylene with the possibility to synthesise bimetallic sealed tubes during gamma irradiation. It was nanoparticles and composite materials (15–20). observed that above a threshold of 5.0 × 10–3 M, The main advantage of radiolysis is that it operates Pt and Rh solutions yielded a shiny metal deposit in very simple physico-chemical conditions (room on the surface of polypropylene tubes below the temperature, absence of contaminants, diluted air-solution surface. For Os solutions, the entire aqueous solutions). During radiolysis the reducing propylene tube was covered with a black deposit agents are either radicals or solvated electrons and the intensity of the deposit increased with that are generated by solvent excitation. The the concentration of the solution. This deposit choice of the absorbed dose is critical in order to is reminiscent of osmium oxide which is black control the cluster size and crystal structure by and volatile (38). Pd solutions also yielded some precise modification of nucleation and growth steps deposit on the surface of the tubes but not as particularly for multi-metal clusters (21–23). The shiny as those for Rh and Pt. We present in the primary effects of the interaction of high-energy following the details of this experiment with some gamma photons with a solution of metal ions are characterisation of the materials formed after the excitation and ionisation of the solvent. The gamma irradiation. different reactions that are observed are well explained (24, 25). In particular, water can produce 2. Materials and Methods upon irradiation a series of reducing and oxidising agents as shown in Equation (i): The salts used in this study were potassium

radiolysis tetrachloroplatinate (K2PtCl4), osmium tetroxide – + H2O e aq, H3O , H•, OH•, H2, H2O2 (i) (OsO4) and palladium(II) chloride (PdCl2)

280 © 2017 Johnson Matthey http://dx.doi.org/10.1595/205651317X696207 Johnson Matthey Technol. Rev., 2017, 61, (4) purchased from BDH Laboratory Chemical Division, detector. XRD patterns were collected in reflection

UK, and rhodium trichloride hydrate (RhCl3.xH2O) geometry either on powder spread out on a glass obtained from Chemica in South Africa. Isopropyl substrate or directly on the coated surface of the alcohol (IPA) 95% that was used in the experiment polypropylene. The X-ray tube was working at 40 kV was obtained from BDH Laboratory Chemical and 30 mA. Raman spectra were recorded at room Division, UK, and sodium dodecyl sulfate (SDS) temperature in the backscattering configuration was obtained from Merck (Pty) Ltd South Africa. on a T64000 Jobin-Yvon (Horiba) spectrometer Ionic solutions of metal salts (Pt, Rh, Os and under a microscope with a 50× objective focusing Pd) were prepared at different concentrations the 514 nm line from an argon-krypton (Ar-Kr) (1.0 × 10–3 M, 2.5 × 10–3 M, 5.0 × 10–3 M, ion laser (Coherent, Innova). HRTEM experiments 7.5 × 10–3 M and 1.0 × 10–2 M). were carried out on a Fei Tecnai G220 operated at Isopropanol was added at a concentration of 200 kV. HRTEM samples were prepared by placing 2.0 × 10–1 M in each solution to prevent OH• a drop of irradiated solution on a copper and nickel radicals and counterbalance the effect of solvated grid, followed by drying under light for 5 min at electrons in radiolysis. Isopropanol is a very good room temperature. scavenger of such radicals. In some experiments a surfactant (SDS) was added at a weight fraction of 3. Results and Discussion 2%. Solutions were then poured into polypropylene tubes and sealed for irradiation with gamma rays. 3.1 Visual Observations The tubes were 13 mm inner diameter and 100 mm long. In all experiments the amount of the prepared Figure 1(a) shows the tubes with different solutions inside each tube was fixed at 5 ml. In concentrations of Rh solutions after irradiation order to avoid the presence of dissolved dioxygen at a dose of 90 kGy. One can easily observe the inside the solutions, each tube was further flushed remarkable appearance of a shiny metal deposit with nitrogen gas for 1 min prior to the gamma on the surface of the polypropylene tubes as ray irradiation. Irradiation was conducted using soon as the concentration of Rh solution is above a 60Co source of activity 600 Gy min–1. Solutions 5.0 × 10–3 M (tube (iii)). It is thus obvious that with were irradiated with 70, 90 and 120 kGy doses for the increase in concentration, there is an increase of a duration of about 2 to 3.3 h. metal deposits. The deposit is very likely the result XRD experiments were performed on a PANalytical of the conversion of Rh metal ions into metallic Rh. q/q Bragg-Brentano Empyrean diffractometer A threshold concentration of about c = 5.0 × 10–3

(CuKa1+2 radiations) equipped with a Pixe M is the limit above which the deposit seems to

(a) (b)

(i) (ii) (iii) (iv) (v) (i) (ii) (iii) (iv) (v)

Fig. 1. Effect of irradiation on solutions of: (a) Rh and (b) Pt ions with concentrations (i) 1.0 × 10–3 M, (ii) 2.5 × 10–3 M, (iii) 5.0 × 10–3 M, (iv) 7.5 × 10–3 M and (v) 1.0 × 10–2 M irradiated at 90 kGy. Tubes are photographed vertically, head down

281 © 2017 Johnson Matthey http://dx.doi.org/10.1595/205651317X696207 Johnson Matthey Technol. Rev., 2017, 61, (4) be continuous and complete to the naked eye. The Rh. The solutions of Os were prepared in the same plastic tubes did not change their colour under volume as those of Pt and Rh. They nevertheless the effect of gamma irradiation unlike glass tubes produced a very significant effect during irradiation. which darkened when irradiated. At the bottom of Instead of yielding a metal film located below the the tubes one can see that the solutions are slightly air-liquid interface as is the case for Pt and Rh, Os dark evidencing the presence of small particles. solutions with a concentration above 5.0 × 10–3 M There was also no evidence of any modification of tubes were coated from top to bottom with a deep the tubes when irradiated although they were likely black deposit as shown in Figure 2(a), tube (iii). to be sensitive to gamma irradiation. This coating is very likely to be osmium(IV) oxide

The particles appeared not to be strongly (OsO2) which is known to be a dark compound aggregated. A similar effect was observed in the and which would be a reduced form of the initial case of Pt solutions, where above 5.0 × 10–3 M, precursor since its valence is downgraded from 6 the tubes were covered with a shiny metal deposit to 4 (39–41). as shown in Figure 1(b), tube (iii). In addition, In the case of Pd solutions, the behaviour is the interior of the tubes contained a small amount again different from the other pgms, as shown in highly agglomerated black particles (Pt black). At Figure 2(b). It can be seen that a slight deposit formed the two lowest concentrations (1 × 10–3 M and on the part of the tube which contained the solution. 2.5 × 10–3 M), the solutions were observed to However, the solution was observed to be colloidal contain suspended particles which were certainly at c = 2.5 × 10–3 M and very dark in colour. Above quite small in size and could be considered as a this concentration there was a yellow transparent colloidal solution. solution forming which contained black particles. At In order to estimate the maximum thickness c = 1.0 × 10–2 M the solution was transparent and of the coating, a rough calculation was made of contained aggregated black particles. the surface of the coating on the tube walls. In In all the above experiments, it was found that the case of the 10–2 M Pt solution, which was the the deposit on the tube walls was extremely stable most concentrated, it can be assumed that the either with or without the solution kept inside the thickness of the coating would be more intense. tube. Deposits were fully preserved a year after The maximum thickness of the coating (Pt film) the irradiation. was estimated to be 3.5 mm assuming that all the The same experiments were also conducted on Pt was converted into the shiny deposit (which is 5.0 × 10–3 M solutions of these metal salts, in not exactly the case as a few Pt particles were still which the SDS was added as a surfactant in order in the solution). to reduce the aggregation of metal particles. It is worth noting that the solutions of Os and As seen in Figure 3, the addition of SDS to the Pd behave very differently from those of Pt and solution was observed to have a prominent effect.

(a) (b)

(i) (ii) (iii) (iv) (v) (i) (ii) (iii) (iv) (v)

Fig. 2. Effect of irradiation on: (a) Os and (b) Pd solutions of concentrations (i) 1.0 × 10–3 M, (ii) 2.5 × 10–3 M, (iii) 5.0 × 10–3 M, (iv) 7.5 × 10–3 M and (v) 1.0 × 10–2 M irradiated at 90 kGy

Raman scattering using a confocal microscope was performed in situ (in the solution) and ex situ (on dried particles). No special effect of the gamma irradiation was seen on the polypropylene tube itself. As expected from the cubic structure of these noble metals, no clear signal was measurable (a) (b) (c) (d) from either the inner part of the tube or the black particles located inside the solution in the Raman analysis. XRD patterns of the coatings and metal nanoparticles after irradiation with gamma rays clearly revealed their metallic nature, except for Os, as shown in Figure 4. One can clearly see in Figure 4 that the diffraction patterns are fully consistent with the Fm3m structures of bulk pgm except for the case of Os. The diffraction peaks are Fig. 3. Effect of irradiation on solutions (a) Pt, (b) Rh, (c) Os and (d) Pd with concentrations of extremely broad in the case of Pt evidencing the 5.0 × 10—3 M with SDS added at 2% w/w and fact that the size of the particles that coherently irradiated at 90 kGy diffract is quite small. The broadening is less for Rh and even less for Pd, which might be due to the fact that in the order Pd < Rh < Pt, it is In all cases colloidal solutions were obtained with easier to reduce the ions because of four chloride 2− a weak deposit on the walls of the propylene tube ligands on [PtCl4] that are able to stabilise small in the case of Rh and Os. It is thus straightforward size particles (21). In the case of Os, the XRD to state from a visual point of view that the effect pattern did not reveal any features showing that of adding the surfactant was quite drastic. For the deposit was mainly amorphous. A fit to the instance Pd solutions without the surfactant yielded data was carried out with the Materials Analysis a yellowish colour at this concentration, in marked Using Diffraction (MAUD) program developed by contrast with the dark colouration in presence of Lutterrotti (42). Very good agreement can be seen the surfactant. Similar effects were observed for Rh between the calculated and the observed patterns. and Pt, while for Os it appeared that the blackening The advantage of using this program is that it allows of the polypropylene tube was reduced. the size of the crystallites that scatter coherently Note that in agreement with our observations, in to be determined, together with the strain of the the absence of any surfactant, the precursor metal lattice. The parameters obtained reveal that the ions were mostly adsorbed on the polyethylene lattice parameter for Pt is very close to that of the walls, as is known for porous supports. The low dose bulk. The size of the particles corresponding in this rate and the long irradiation time used here favour case to the size of the domains which coherently the diffusion of ions towards the walls. Moreover, it scatter the beam is about 8.8 nm and the lattice is is likely that the irradiation of the support induces clearly strained. electrons that contribute to the reduction of The Pt deposit on the tube walls was also analysed adsorbed ions in contact. Then supplementary ions by XRD. The tube was cut along its principal axis and adsorb on the initial reduced metal layer and are the inner part of the tube was mounted on the stage also reduced in situ. In contrast, when the solution of the XRD. It was observed that this deposit was contains SDS, the ion binding with the surfactant is shiny and quite fragile when subjected to scratches stronger than the adsorption on the polyethylene with tweezers. The shiny aspect of the deposit was and mostly inhibits the deposit. a clear signature of its metallic nature. However when a scan in conventional θ-θ geometry was 3.2 X-Ray Diffraction and High carried out, no measurable signal of the Pt deposit was obtained. The main reason was attributed to Resolution Transmission Electron the very small thickness of this deposit that was Microscopy Analysis precluding the measurement of a correct signal over The polypropylene tubes and their contents were noise ratio. For this reason a 2θ scan (detector scan) further analysed using Raman and XRD to identify was performed at a fixed grazing incident angle of the nature of the phases formed upon irradiation. about 0.9º using a parallel beam coming from a

4000 [111] Pt powder 0.01 M 3000 a = 0.392 nm Crystallite size 8.8 nm 2000 [200] Intensity

1000 [220] [311]

700 [111] 600 Rh powder 0.01 M 500 a = 0.380 nm 400 Crystallite size 17 nm 300 [200]

Intensity 200 [220] [311] 100 0 [111] 6000 Pd powder 0.01 M a = 0.389 nm 4000 Crystallite size 40 nm

Intensity [200] 2000 [220] [311]

700 600 Os powder 0.01 M 500 400 300

Intensity 200 100 0 40 60 80 2θ, º

Fig. 4. XRD patterns of Pt, Rh, Pd and Os particles

W-C mirror (Empyrean diffractometer PANalytical). also complement the XRD results. It can clearly It was thus possible to identify all the reflections be observed in these images that particles are of the Fm3m structure, clearly evidencing that the agglomerated. thickness of the deposit was extremely small and The rings observed in the electron diffraction that the deposit was pure Pt (see Figure 5). Similar scattering patterns are in good agreement with conclusions could be drawn for Rh and Pd. the calculated d spacing values from Pt, Pd and Rh Note that in the case of the coating, experiments diffraction data from the literature. The set of hkl were carried out at grazing incidence and that the indices corresponding to these rings are shown in scattering by the tube itself was subtracted. One Figures 6(b), 7(b) and 8(b). can see clearly the perfect concordance between From this present study it can be confirmed that the two patterns. gamma ray irradiation of salt solutions containing Table I gathers some typical information Pt and Rh ions at dose rates above 70 kGy can concerning the metals studied hereafter together be used to convert the pgm salts into metal with the values measured for some of the nanoparticles in the bulk and produce a shiny parameters (i.e. lattice parameters, size and metal coating on polypropylene tubes. The Pd strain) after refining the measurements using the salt solution did not yield such a shiny coating MAUD Software (42). whereas the Os solution produced an intense As shown in Figures 6–8, HRTEM analysis of black coating, the structure of which was found to the metal nanoparticles present in the solution be amorphous.

4000 [111] Pt powder 0.01 M 3000 a = 0.392 nm Crystallite size 8.8 nm 2000 [200]

1000 [220] [311]

0 [111] Intensity 4000 Pt tube a = 0.392 nm Crystallite size 11 nm 2000 [200] [220] [311] 0

30 45 60 75 90 2θ, º

Fig. 5. Comparison between the X-ray powder diffraction patterns collected on Pt powder and on the coating of the polypropylene tube

Table I Space Groups of Platinum, Palladium, Rhodium and Osmium Together with Bulk Lattice Parameters and the Measured Values of the Lattice Parameters in this Study, the Size of the Particles in Powder and the Strain

Bulk lattice Measured lattice Size of particles, Measured Metal Space group parameter, a, parameter, a*, nm strain nm nm

Pt Fm3m 0.39242 0.39222 8.8 0.0057 (0.0001) (0.2) (0.0002)

Rh Fm3m 0.38034 0.38066 16.9 0.0038 (0.0005) (0.6) (0.004)

Pd Fm3m 0.3807 0.38947 41.7 0.0015 (0.0001) (1.1) (0.0001)

Os P63/mmc a = 0.27311 N.A. N.A. N.A. c = 0.43173

(a) (b)

(111) (200) (220) 2 nm (311) (222) Fig. 6. (a) HRTEM image of Pt particles synthesised by gamma irradiation at a dose of 120 kGy; (b) selected area electron diffraction (SAED) of Pt nanoparticles

(a) (b)

(111) (200) (220) 2 nm (311) (222) Fig. 7. (a) HRTEM image of Rh particles synthesised by gamma irradiation at a dose of 120 kGy; (b) SAED of Rh nanoparticles

(a) (b)

(111) (200) (220) (311) 2 nm

Fig. 8. (a) HRTEM image of Pd particles synthesised by gamma irradiation at a dose of 120 kGy; (b) SAED of Pd nanoparticles

4. Conclusion an alternative to other deposition techniques such as metal evaporation or sputtering. The coating The noble metal salt solutions of Pt, Os, Rh and Pd was found to be extremely stable in solution since were successfully reduced into metal nanoparticles no modification was observed even one year after using gamma ray irradiation at doses ranging from irradiation. However when the tubes were cut, the 70 to 120 kGy and even into a metal coating of coating could be easily scratched with tweezers. the propylene tubes, except for Pd solutions. This was showed clearly that although the coating The structure of the metal nanoparticles and the was stable as a function of time, the adhesion of coating was ascertained by a quantitative analysis the metal particles to the tube was not extremely of the XRD patterns. It is shown that these patterns good. Atomic force microscopy (AFM) studies of are fully consistent with the Fm3m structure of the coating are planned in the future to evidence these noble metals except for Os for which it was the quality of the adhesion. impossible to define any crystalline structure. This study revealed that gamma irradiation can be used Acknowledgements to produce beautiful shiny metal coatings with Pt and Rh on the inner surface of plastic tubes provided The authors are particularly grateful to the iThemba that the concentration of the initial solution is LABS, National Research Foundation (NRF), French high enough (in this case >5 × 10–3 M). Such an embassy in Pretoria for their support and to the effect was unexpected and is of clear interest as Protea programme for funding.

References 18. M. Mirdamadi-Esfahani, M. Mostafavi, B. Keita, L. Nadjo, P. Kooyman and H. Remita, Gold Bull., 1. “Processing and Finishing of Polymeric Materials”, 2010, 43, (1), 49 in 2 volumes, John Wiley & Sons, Inc, New Jersey, 19. Y. Yang, P. Cheng and S. Huang, J. Alloys Compd., USA, 2011, 1488 pp 2016, 688, (Part A), 1172 2. T. Volova, “Polyhydroxyalkanoates – Plastic 20. T.-E. Fan, T.-D. Liu, J.-W. Zheng, G.-F. Shao and Materials of the 21st Century: Production, Y.-H. Wen, J. Alloys Compd., 2016, 685, 1008 Properties and Applications”, Nova Science Publishers Inc, New York, USA, 2004, 282 pp 21. J. Belloni, M. Mostafavi, H. Remita, J.-L. Marignier and M.-O. Delcourt, New J. Chem., 1998, 22, 3. R. Klein, “Laser Welding of Plastics: Materials, (11), 1239 Processes and Industrial Applications”, Wiley-VCH Verlag & Co KGaA, Weinheim, Germany, 2012, 22. B. Soroushian, I. Lampre, J. Belloni and 260 pp M. Mostafavi, Radiat. Phys. Chem., 2005, 72, (2–3), 111 4. “Handbook of Polyethylene Pipe”, 2nd Edn., Plastics Pipe Institute, Texas, USA, 2009, 620 pp 23. J. Belloni, Catal. Today, 2006, 113, (3–4), 141 5. R. J. Crawford, ‘General Properties of Plastics’, 24. K. Roy and S. Lahiri, Green Chem., 2006, 8, (12), in “Plastics Engineering”, 3rd Edn., Elsevier 1063 Butterworth-Heinemann, Oxford, UK, 1998, 25. M. K. Temgire, J. Bellare and S. S. Joshi, Adv. 522 pp Phys. Chem., 2011, 249097 6. M.-O. Delcourt, J. Belloni, J.-L. Marignier, C. Mory 26. A. Henglein, Chem. Rev., 1989, 89, (8), 1861 and C. Colliex, Radiat. Phys. Chem., 1984, 23, 27. S. Choofong, P. Suwanmala and W. Pasanphan, (4), 485 ‘Water-Soluble Chitosan-Gold Composite 7. J. L. Marignier, J. Belloni, M. O. Delcourt and J. P. Nanoparticles: Preparation by Radiolysis Method’, Chevalier, Nature, 1985, 317, (6035), 344 The 18th International Conference on Composite 8. J. Belloni (née Cofler), J.-L. Marignier, M.-O. Materials, Jeju, Korea, 21st–26th August, 2011, Delcourt (née Euverte) and M. Minana (née 6 pp Lourseau), National Center for Scientific Research, 28. Y. N. Rao, D. Banerjee, A. Datta, S. K. Das, R. 'Non-noble Metal Micro-aggregates, Process for Guin and A. Saha, Radiat. Phys. Chem., 2010, 79, their Manufacture and their Use in the Catalysis (12), 1240 of the Photoreduction of Water', French Patent 29. J. Biswal, S. P. Ramnani, R. Tewari, G. K. Dey and 8,409,196; 1985 S. Sabharwal, Radiat. Phys. Chem., 2010, 79, 9. A. J. Swallow, “Radiation Chemistry: An (4), 441 Introduction”, Longman Group Ltd, London, UK, 30. E. Gachard, H. Remita, J. Khatouri, B. Keita, L. 1973, 275 pp Nadjo and J. Belloni, New J. Chem., 1998, 22, 10. J. Bruneaux, H. Cachet, M. Froment, J. Amblard, J. (11), 1257 Belloni and M. Mostafavi, Electrochim. Acta, 1987, 31. A. Krklješ, “Radiolytic Synthesis of Nanocomposites 32, (10), 1533 Based on Noble Metal Nanoparticles and Natural 11. ‘Platinum-Group Metals’, in “Mineral Commodities Polymer, and their Application as Biomaterial”, Summaries 2006”, US Geological Survey (USGS), Report No. IAEA–RC–1207.1, International Atomic Reston, Virginia, USA, 2006, p. 126 Energy Agency (IAEA), Vienna, Austria, 2011, 12. L. B. Hunt and F. M. Lever, Platinum Metals Rev., 12 pp 1969, 13, (4), 126 32. Q. Chen, J. Shi, R. Zhao and X. Shen, “Radiolytic 13. T. J. Colacot, Platinum Metals Rev., 2002, 46, Syntheses of Nanoparticles and Inorganic-Polymer (2), 82 Hybrid Microgels”, Report No. IAEA–RC–1124.2, 14. K. Roy and S. Lahiri, Anal. Chem., 2008, 80, (19), International Atomic Energy Agency (IAEA), 7504 Vienna, Austria, 2010, 10 pp 15. J. Amblard, O. Platzer, J. Ridard and J. Belloni, J. 33. V. Hornebecq, M. Antonietti, T. Cardinal and M. Phys. Chem., 1992, 96, (5), 2341 Treguer-Delapierre, Chem. Mater., 2003, 15, (10), 1993 16. M. Treguer, C. de Cointet, H. Remita, J. Khatouri, M. Mostafavi, J. Amblard, J. Belloni and R. de 34. A. D. Belapurkar, S. Kapoor, S. K. Kulshreshtha and Keyzer, J. Phys. Chem. B, 1998, 102, (22), 4310 J. P. Mittal, Mater. Res. Bull., 2001, 36, (1–2), 145 17. C. M. Doudna, M. F. Bertino, F. D. Blum, A. T. 35. J. V. Rojas and C. H. Castano, J. Radioanal. Nucl. Tokuhiro, D. Lahiri-Dey, S. Chattopadhyay and J. Chem., 2014, 302, (1), 555 Terry, J. Phys. Chem. B., 2003, 107, (13), 2966 36. X. Zhang, Y. Ye, H. Wang and S. Yao, Radiat. Phys.

Chem., 2010, 79, (10), 1058 40. J. W. Arblaster, Platinum Metals Rev., 1995, 39, (4), 164 37. J. V. Rojas and C. H. Castano, Radiat. Phys. Chem., 2012, 81, (1), 16 41. I. C. Smith, B. L. Carson and T. L. Ferguson, Environ. Health Perspect., 1974, 8, 201 38. W. P. Griffith, Q. Rev. Chem. Soc., 1965, 19, 42. L. Lutterotti, S. Matthies and H. R. Wenk, ‘MAUD: (3), 254 A Friendly Java Program for Material Analysis Using Diffraction’, Newsletter, No. 21, International 39. J. W. Arblaster, Platinum Metals Rev., 1989, 33, Union of Crystallography, Commission on Powder (1), 14 Diffraction, Stuttgart, Germany, 1999, p. 14

The Authors

Takalani Cele has an MSc in Physical Science from University of the Western Cape. She is currently doing her PhD in nanotechnology at the University of South Africa. Her research interest is pgm nanoparticles by radiolysis, jointly supported by iThemba Laboratory for Accelerator Based Science, South Africa, and Université du Maine, Le Mans, France. She is also working on other projects in the South African Department of Trade and Industry’s Innovation and Technology unit.

Philip Beukes obtained an M(med)Sc in Nuclear Medicine at the Stellenbosch University, South Africa. He is currently head of the Radiation Safety Division at iThemba Laboratory for Accelerator Based Science. His research interest lies in radiation biodosimetry with a focus on quality factors for high energy neutron sources.

Thomas Beuvier completed his PhD in Material Science in 2009 at Université de Nantes, France, and at the Institut des Matériaux de Nantes (IMN), France. His post-doctoral research is focused on polymer thin films, inorganic materials and biomimetic/biogenic mineralisation using synchrotron X-rays.

Elvia Anabela Chavez Panduro received her PhD in Physics from Université du Maine, Le Mans, Institut des Molécules et des Matériaux du Mans (IMMM), France, in collaboration with the European Synchrotron Radiation Facility (ESRF), France. Currently she is working at the Norwegian University of Science and Technology (NTNU), Norway, in collaboration with

SINTEF Petroleum, Norway, on the CO2PLUG project which is financed by the Norwegian Research Council. Some of her research topics in this project are on the nano and micro scale study of cements (using X-ray imaging techniques) and on the problems of cement

integrity when exposed to CO2-brine at high pressure and temperature.

Malik Maaza is a Professor, a Senior Scientist and staff member of the National Research Foundation of South Africa. He holds a PhD in neutron quantum optics from Université Pierre et Marie Curie, France, and an MSc in Photonics and lasers from Université Pierre et Marie Curie and Université Paris-Sud, France. He is the current holder of the UNESCO Africa Chair in Nanosciences and Nanotechnology. He pioneered world-class nanoscience research within the continent and at the international level. He has scientific expertise in the fields of surface and interface two-dimensional (2D)-phenomena, solar energy, neutron quantum optics specifically and neutron and X-ray scattering in general. He has authored and coauthored about 350 international peer reviewed ISI-CSI publications in reputable journals. He is the sole African author who has been invited to publish in Physics Report (2016 Journal Impact Factor 17.425, Clarivate Analytics 2017) and has recently coauthored a publication in Nature’s Scientific Reports (Nature group series). He is one of the rare African scientists serving on several editorial committees of ISI-CSI international journals such as the Journal of Materials Chemistry A and the International Journal of Nanotechnology.

Alain Gibaud is a Professor at Université du Maine, France. He received his PhD in Physics and an MSc in Acoustics at Université du Maine Le Mans. His research interest is in the multidisciplinary fields of nanoscience and X-ray physics.

www.technology.matthey.com

“Advances in Industrial Mixing: A Companion to the Handbook of Industrial Mixing” Edited by Suzanne M. Kresta (University of Alberta, Canada), Arthur W. Etchells III (DuPont, USA), David S. Dickey (MixTech, Inc, USA) and Victor A. Atiemo- Obeng (The Dow Chemical Company, USA), John Wiley and Sons Inc, Hoboken, New Jersey, USA, 2016, 1044 pages, ISBN: 978-0-470-52382-7, £160.00, €192.00, US$200.00

• 21b ‘Magnetic Drives for Mixers’ Reviewed by Li Liu Section 1b introduces the concept of tracer Johnson Matthey, PO Box 1, Belasis Avenue, mean age and local residence time distributions Billingham, Cleveland TS23 1LB, UK (RTD). RTD theory does not consider the spatial distribution of tracer concentration and cannot Email: [email protected] define the local mixing states inside the mixing system. The essence of the mean age theory is a “Advances in Industrial Mixing” is an updated spatial distribution of time-integrated local mean version of the “Handbook of Industrial Mixing” age. The transport equation of the mean age is (1). The unchanged text of the “Handbook of similar to the Navier-Stokes (NS) equation, so Industrial Mixing” is provided electronically (on can be solved using computational fluid dynamics the accompanying DVD), and only the new or (CFD). The relationship between the mean age substantially revised contents are provided in the theory and the residence time theory is described. hard copy. In comparison with the RTD, the advantage of the The order of the chapters in the “Handbook of mean age theory is its ability to quantify the state Industrial Mixing” is retained in the new version. of local mixing in a flow system. New contents are added in new subsections. Section 2b updates the turbulence length scales There are 10 chapters with additional sections: and turbulence dissipation from new experimental • 1b ‘Mean Age Theory for Quantitative Mixing data from different mixing systems. The interaction Analysis’ between turbulence and both solids (lifting of • 2b ‘Update to Turbulence in Mixing Applications’ solid particles from the tank bottom) and gases • 3b ‘Microstructure, Rheology, and Processing (break-up of bubbles) is briefly discussed. of Complex Fluids’ In Section 3b, complex rheology of different • 5b ‘CFD modelling of Stirred Tank Reactors’ types of fluids are described. The relationship • 6a ‘Mechanically Stirred Vessels’ between the rheology of multiple complex fluids • 6c ‘Vessel Heads: Depths, Volumes, and Areas’ (shear-thinning fluid, shear-thickening fluid, • 7b ‘Update to Mixing in Pipelines’ viscoelastic fluid) and scales from microstructure • 7c ‘Introduction to Micromixers’ (polymers, nanoparticles) to macroscale • 9b ‘Laminar Mixing Processes in Stirred (powders) is scrutinised. Vessels’ The current status of CFD, including both • 13b ‘Scale-up Using the Bourne Protocol: advantages and limitations, is examined in Reactive Crystallization and Mixing Example’ Section 5b. The effects of mesh type, mesh • 14b ‘Heat Transfer in Stirred Tanks – Update’ size, discretisation schemes on predicted flow

290 © 2017 Johnson Matthey https://doi.org/10.1595/205651317X696225 Johnson Matthey Technol. Rev., 2017, 61, (4) and turbulence diffusion are described. It raises An example of a reaction with particles produced the question of how to select laminar or different in a continuous process is analysed to solve the turbulence models for transitional flows. Modelling problem of smaller particle generation in Section multi-phase flows is limited with regard to both 13b. This example is used to show why moving incorporating accurate models and high computing from batch to continuous processing causes a resources, and further development is therefore smaller-sized product. required. In Section 14b, additional correlation and In Section 6b, the flow pattern provided in the calculation examples are provided for heat “Handbook of Industrial Mixing” is expanded to transfer in stirred tanks. The original “Handbook include the effects of more factors on mixing: of Industrial Mixing” covered fundamental heat effects of liquid volume, off bottom clearance, transfer, heat transfer geometries and heat bottom shape, viscosity, baffles, impeller diameter, transfer coefficient correlations. This volume adds impeller off-set and angle shafts. Section 6c a few specific aspects of heat transfer, for example discusses the contribution of tank head, head overall heat transfer coefficient calculations. A volume and areas on the liquid height and liquid detailed calculation for a jacketed stirred tank is volume and how to include them in the calculation explained using an example. of liquid volume and height. Section 21b introduces different types of Section 7b focuses on mixing in pipelines, magnetic mixers such as laboratory magnetic covering some modest progress in this area, for stirrers, top-entering magnetic mixer drivers, and bottom-entering magnetic mixer drivers. The example the development of compact turbulent magnetic mixers are more advantageous for high- mixers to obtain a high degree of radial uniformity. pressure, toxic applications and pharmaceutical Low-pressure drop mixer development for laminar applications. flow (Sulzer SMX mixers) and low-pressure Instead of adding new material in a subsection in drop mixer development for turbulence flow the chapter, as mentioned above, Chapter 10, on (orifice-style mixers) are described. New methods solid-liquid mixing is completely revised to include for liquid-liquid and gas-liquid dispersions in a significant expansion and update. For example, laminar, transitional and turbulent flow are besides the traditional Zwietering method for provided. Micromixers are preferable for high calculating the minimum agitator speed, a new rate of heat and mass transfer applications, and method is developed using more extensive data. when specific product properties are requested. A Regarding the effect of solid physical properties, fundamental description of mixing and transport in the impact of particle size distribution and shape, micromixers is provided in Section 7c. solid wettability and stickiness are included. Solid Section 9b presents blending highly viscous fluids suspension with multiple impellers in vessels of in laminar or low transitional regimes where viscous different base shapes is also described. forces are dominant. The complexities of rheology Besides the above mentioned materials that cause blending challenges regardless of mixing are added into the chapters of the “Handbook of device. A broad range of laminar mixing equipment Industrial Mixing”, this volume also includes six is reviewed, such as open impellers, close-clearance new chapters: Chapters 23–28. impellers and multiple impellers. Power numbers Chapter 23 ‘Commissioning Mixing Equipment’ for different non-Newtonian fluids in laminar and gives instructions for commissioning new mixing low transitional regimes are analysed. Various equipment, including inspection, installation, experimental measurements for characterisation operation and maintenance. Safety and risk are explored, including particle image velocimetry management have become priorities in industry. (PIV), laser Doppler anemometry (LDA), planar Chapter 24 ‘Mixing Safety’ uses two case studies laser-induced fluorescence (PLIF) and electrical of incidents to discuss the practice of safety and resistance tomography (ERT). Modelling approaches risk management. such as finite element method (FEM) and finite Chapter 25 ‘Mixing Issues in Crystallization and volume method (FVM) are also briefly introduced. Precipitation Operations’ describes the relationships For mixing systems of high aspect ratio of liquid between mixing and crystallisation and precipitation height and tank diameter, multiple impellers are processes. In the crystallisation and precipitation usually installed but may result in the generation of operations, a number of factors need to be segregated zones. An approach of reversal rotation considered. Identifying the interactions between is introduced to eliminate this. key mixing parameters with the crystallisation and

291 © 2017 Johnson Matthey https://doi.org/10.1595/205651317X696225 Johnson Matthey Technol. Rev., 2017, 61, (4) precipitation processes, such as nucleation, growth to the issues raised in industry, a wide range of and agglomeration, is crucial. new materials are added in this volume, such as Chapters 26, 27 and 28 examine mixing in water, health and safety, and mixing in water, food and food and the pharmaceutical industry, respectively. the pharmaceutical industry. Equipment types, calculation examples for treating "Advances in drinking water, wastewater and sludge are given in Industrial Mixing: A Chapter 26. In Chapter 27, multiple types of mixers Companion to the such as double-motion mixer and high-shear Handbook of Industrial mixers are discussed for food mixing. Different Mixing" types of food groups are briefly summarised. In the pharmaceutical industry, activities and equipment are required to be validated according to a regulatory agency. The validating issues need to be included in the mixing design. Chapter 28 discusses the concept of validation, and where mixing issues may occur.

Conclusion

In summary, “Advances in Industrial Mixing” Reference provides an expansion to the “Handbook of Industrial Mixing” (1), including new developments 1. “Handbook of Industrial Mixing: Science and in both experimental and numerical approaches and Practice”, eds. E. L. Paul, V. A. Atiemo-Obeng and new methods developed based on more extensive S. M. Kresta, John Wiley and Sons Inc, Hoboken, data for assessing mixing quality. With regards New Jersey, USA, 2004

The Reviewer

Li Liu is a Fluid Engineering Core Scientist at Johnson Matthey, Billlingham, UK. She completed her PhD (2009–2013) in Chemical Engineering at the University of Birmingham, UK, with her PhD project on ‘Computational Fluid Dynamics Modelling of Complex Fluid Flow in Stirred Vessels’. She works on projects with various groups at different business units within Johnson Matthey, using advanced experimental and modelling techniques to improve understanding of complex fluid processes such as mixing, porous flow, catalytic reactor and reaction-diffusion.

www.technology.matthey.com

Artificial Photosynthesis: Faraday Discussion Platinum group metals still important for superior photocatalytic activity

Reviewed by Joshua Karlsson approaches, fundamental processes to molecular Molecular Photonics Laboratory, School of and inorganic assembly catalysts, as well as Chemistry, Newcastle University, Bedson integration of systems for practical devices. Building, Newcastle Upon Tyne, NE1 7RU, UK This review will discuss the role of platinum group metals (pgm) in the continuing drive Email: [email protected] to develop new catalysts for molecular and photoelectrochemical approaches to water splitting and carbon dioxide reduction.

Introduction Molecular Catalysts for Water Splitting The Royal Society of Chemistry Faraday Discussions are a series of meetings focusing on The ultimate goal of reducing the world’s rapidly developing areas of physical chemistry. dependence on fossil fuels cannot be realised by Contrary to typical conferences, Faraday laboratory-scale solutions alone, an increased Discussions rely on the active participation of collaboration with industry is required. To that speakers and audience alike. Topics for each end a number of industry experts attended the session are based on new research papers conference, bringing their perspective on the submitted specifically for the meeting. Audience technical challenges in implementing large- participation is important because short scale production of solar fuels. Tohru Setoyama presentations on each paper are followed by (Mitsubishi Chemical Science and Research lengthy discussions, with all delegates free to ask Centre, Japan) detailed the state of the solar questions and make comments. All questions and hydrogen industry. With a target of 10% solar to answers are recorded and later published in a hydrogen conversion efficiency by 2021 industry Faraday Discussions volume. These conferences has set researchers an ambitious goal (1). This are a useful forum for documenting the current will require greater focus in a field which has state of the art within a field of research. become divergent in its approaches to generating The 2017 Faraday Discussion on artificial solar fuels. photosynthesis took place in Kyoto, Japan. This In the context of molecular catalysts for water was noteworthy as it was the first time a Faraday splitting it is apparent that pgms still offer the best Discussions meeting was organised in Japan. The overall performance. Current work has sought to three-day event from 28th February to 2nd March build a better understanding of the mechanisms 2017 was hosted at the Ritsumeikan University behind photocatalytic water splitting (Figure 1) Suzaku Campus, Kyoto. Some 100 delegates and carbon dioxide reduction. Ruthenium-based attended with approximately 30 speakers and catalysts continue to be favoured as a starting close to 50 poster presentations. Topics covered point for new catalysts. Now that the mechanistic all areas of artificial photosynthesis from biological understanding has improved, several research

was concluded that ligand substitution lability was a crucial factor in the design of future catalysts.

H+ Revisiting Solar Fuels Production via H+/H 0 V 2 Heterogeneous Catalysis e– H 2 Hyunwoong Park et al. (3) (Kyungpook LUMO hn National University, South Korea) demonstrated photocatalytic production of molecular hydrogen + HOMO h using a ternary composite of platinum, cadmium sulfide and sodium trititanate nanotubes (TNT), H O 2 Pt/CdS/TNT. The impetus for this work was to revisit solar fuels production via heterogeneous catalysis. H2O/O2 +1.23 V O2 Although CdS has in the past been studied as a photocatalyst, problems with decomposition and the need for a sacrificial electron donor or hole- Fig. 1. Simplified scheme for photocatalytic water scavenger prevented these systems from being splitting using a molecular catalyst effective. With their sodium TNT composite the authors groups are optimising the ligand environment observed a six-fold increase in hydrogen evolution around Ru in particular. This has led to several compared to a more traditional heterogeneous promising molecular systems for water splitting photocatalytic system, a Pt/CdS/TiO2 composite. and CO2 reduction. H2 evolution was shown to be exclusively from Ru-based water-oxidation catalysts (WOC) have water splitting in an aqueous system containing been extensively studied since the first example 5 vol% 2-propanol (IPA) as a sacrificial electron of an artificial WOC, in 1982, came in the form donor. The Pt/CdS/TNT composite essentially of a dinuclear Ru-complex, the well-known ‘blue- protected the metallic Pt from oxidation and CdS dimer’. Over subsequent years there have been from decomposition. Pt plays a key role as a co- a large number of dinuclear and mononuclear catalyst in the system, promoting charge transfer. of Ru-based WOCs reported. In a systematic Pt is known to effectively catalyse H2 evolution comparison of Group 8 metal complexes bearing in CdS composites under visible light irradiation. the same pentadentate and monodentate ligands However, the interplay between CdS, Pt and the (Figure 2), Shigeyuki Masaoka et al. (2) (Institute sacrificial electron donor has led to disparate for Molecular Science, Japan) contributed more reports on the effectiveness of Pt as a co-catalyst. evidence for understanding the basis for the The Pt/CdS/TNT system is more durable with excellent water-oxidation activity of Ru-based respect to CdS decomposition. Isotopic analysis of catalysts against iron and osmium complexes. It the gasses evolved from the system gives a better

(a) (b)

Cl N N N N N H2O O Ru N Ru N Ru N N N H2O N N N

Fig. 2. First reported WOC: (a) the ‘blue dimer’; and (b) pentadentate mononuclear Ru WOC from Masaoka et al. (2)

+ understanding of the mechanism for H2 evolution. used the [Ir(tpy)(ppy)Cl] catalyst as a starting It is particularly important to note that at higher point for new variants with electron-rich ligands. concentrations of sacrificial electron donor (>5 Previous studies had identified the strength of vol% IPA) not all the H2 produced is from water hydride donating species as key intermediates in splitting, but also from deprotonation of IPA. This the catalytic process. might lead to an overestimation of H2 produced in On that basis the aim was to add electron- similar systems. rich substituents (bis(benzimidazole)-phenyl or -pyridine) in place of bidentate (ppy) and tridetate Photochemical Reduction of Carbon (tpy) ligands with a view to creating strong hydride donors for products further reduced beyond Dioxide formate and CO2. It was found that while CO2 Developing photochemical systems for the could be selectively reduced to form CO, turnover reduction of CO2 to carbon monoxide, methanol numbers did not increase with increasing hydricity. and other useful fuels remains highly desirable, The work highlights the difficulty in rational design but very difficult. Given the sheer stability of CO2 of photocatalysts for CO2 reduction. It is postulated (one-electron reduction potential of –1.91 V vs. that effects of the intermediate singly reduced and normal hydrogen electrode (NHE)), this remains doubly reduced catalyst species need to be better extremely challenging to realise photochemically. understood going forward to reduce CO2 more Reduction to CO and formic acid requires two efficiently to CO, formate and beyond. electrons, therefore catalysts must be capable of Osamu Ishitani and Yusuke Tamaki (5) (Tokyo multi-electron reduction in order to be effective. Institute of Technology, Japan) also addressed

Etsuko Fujita et al. (4) (Brookhaven National the problem of CO2 reduction to CO and formic Laboratory, USA) investigated photochemical acid, but with a supramolecular approach. In + reduction of CO2 to CO using [Ir(tpy)(ppy)Cl] their supramolecular system an asymmetric derivatives. These iridium complexes are bis-tridentate Ru(II) complex functions as a one of only a few known compounds able to photosensitiser and a Ru(II) carbonyl complex as photochemically reduce CO2 without a sensitiser, the catalyst (Figure 3). The work builds on previous as they absorb light in the ultraviolet (UV)-visible studies where supramolecular catalysts were region. These complexes still require a sacrificial coupled to semiconductor surfaces. The authors electron donor in order to function. The authors aim to attach the catalyst to a semiconductor and

2+ N Cl CO N n+ Ru H2O3P N N N Cl CO – e N Ru N O N N N N N H2O3P e– N N Ru N Catalyst 4+ H2O3P N N Semiconductor N N Cl N N N N H2O3P N Ru N N Ru N Cl N N N N CO

Fig. 3. Concept of a hybrid supramolecular catalyst system with photocatalyst/sensitiser. Coupling the sensitiser to [Ru(Clbpy)(CO)] (blue) selectively reduces CO, or [Ru(CO)2Cl2] (red) selectively reduces formic acid. The remaining ligand on the catalyst is the bridging ligand (Adapted from (5) with permission of The Royal Society of Chemistry)

295 © 2017 Johnson Matthey https://doi.org/10.1595/205651317X696234 Johnson Matthey Technol. Rev., 2017, 61, (4) increase the oxidative power in a future study. A recurring theme during the meeting was the Here the properties of the catalyst-sensitiser unit effects of temperature on catalytic activity in have been characterised. Coupling the sensitiser electrocatalytic and photocatalytic systems. This to a catalyst via a bridging ligand helps overcome is very closely linked to practical device design. previous problems with back electron transfer. There is more to be learnt about the dynamics Furthermore, use of the asymmetric tridentate of molecular catalysts at different temperatures. Ru(II) complex as a sensitiser was important since Indeed practical devices may operate at a range it has an excited state lifetime long enough (3MLCT of elevated temperatures, where catalyst activity t ≈ 30 ns) to transfer an electron to the catalyst. may be quite different. Finally, as noted by Daniel This integrated supramolecular approach means Nocera and others, while solar fuels production even weak electron donors such as methanol and must overcome the current reliance of fossil water can be used to selectively reduce CO2 to fuels, no one-size-fits-all solution is capable of formic acid and CO. overturning the overbearing economic sway of the fossil fuels industry. Change may well come with Concluding Remarks government intervention in the form of carbon taxation for example. Until then the enormous cost The meeting clearly demonstrated that, as of replacing fossil fuel infrastructure will continue a community, the approaches to artificial to be an obstacle. This shows the need for closer photosynthesis have become quite varied. Over co-operation with industry, which no doubt will a number of years now there has been a drive continue in future meetings. In the meantime the towards replacing pgms as catalysts for water fundamental research into artificial photosynthesis splitting and light-harvesting molecules with seen at meetings like Faraday Discussions will Earth-abundant alternatives such as iron and continue to underpin future innovations. manganese. Researchers keep returning to pgms “Faraday Discussion: Artificial Photosynthesis” though, because of their superior catalytic activity 2017 is currently in press. The volume will be made and to gain further mechanistic insight into water available in due course. splitting and CO2 reduction. Discussions during the sessions noted that if there is to be progress References towards some of the goals industry has set for solar fuels production on a mass-scale, then the 1. T. Setoyama, T. Takewaki, K. Domen and T. scientific community is going to have to agree on Tatsumi, Faraday Discuss., 2017, 198, 509 a more concentrated direction forward. The cost of 2. M. Yoshida, M. Kondo, M. Okamura, M. Kanaike, the pgms cannot be the main impetus for seeking S. Haesuwannakij, H. Sakurai and S. Masaoka, Earth-abundant alternatives if the alternatives Faraday Discuss., 2017, 198, 181 fall greatly short in terms of performance in real 3. H. Park, H.-H. Ou, M. Kim, U. Kang, D. Suk Han devices. In the future there may well need to and M. Hoffmann, Faraday Discuss., 2017, 198, be far greater focus on one class of catalysts. 419 The viewpoint of representatives from industry 4. G. F. Manbeck, K. Garg, T. Shimoda, D. J. Szalda, highlighted infrastructure considerations and full M. Z. Ertem, J. T. Muckerman and E. Fujita, lifecycle analysis as driving forces for industrial- Faraday Discuss., 2017, 198, 301 scale solar fuels production. 5. Y. Tamaki and O. Ishitani, Faraday Discuss., 2017, 198, 319

The Reviewer

Joshua Karlsson graduated from the University of York, UK, in 2013 with a BSc in chemistry, and an MRes in Green Chemistry from Imperial College London, UK, in 2014. He is now in the final year of a PhD project under the supervisor of Professor Anthony Harriman at the Molecular Photonic Laboratory, Newcastle University, UK. His research revolves around investigating the fundamental photophysical properties of fluorescent dyes in the far-red spectral region and applying this knowledge in the contexts of super-resolution fluorescence microscopy and organic solar cells.

www.technology.matthey.com

Reducing the Carbon Intensity of Methanol for Use as a Transport Fuel Impact of technology choice on greenhouse gas emissions when producing methanol from natural gas

Alan Ingham use in existing passenger cars (1). Methanol is an Johnson Matthey, 10 Eastbourne Terrace, affordable alternative transportation fuel due to London W2 6LG, UK its efficient combustion, ease of distribution and wide availability around the globe. Methanol is a Email: [email protected] high octane fuel that enables very efficient and powerful performance in spark ignition engines. Engines optimised for methanol could provide Methanol is increasingly being looked at as a way an energy based efficiency gain of 50% over a to reduce the emissions potential of transport fuel. standard (port fuel injected, non-turbo) gasoline It may be used in place or in addition to gasoline engine in a light-duty vehicle (2). fuel, for example. The amount of greenhouse Two different methods are used to compare the gas (GHG) emitted in producing methanol can emissions from the flowsheets, the first is the vary hugely according to the syngas generation direct GHG emissions from the methanol plant as a technology selected and the choice of electrical carbon dioxide flowrate per hour and the second is or steam turbine drive for compressors and the carbon intensity of producing methanol based pumps. This paper looks at the impact of these on the total carbon emitted from the process per technology choices on GHG emissions and how unit of energy, and is expressed as grams of CO2 the carbon intensity of methanol used as a equivalent per megajoule of methanol on a lower –1 transport fuel compares to the carbon intensity of heating value (LHV) basis (gCO2e MJ MeOH). other hydrocarbon fuels. It is found that methanol produces lower well to wheel emissions than 2. Natural Gas to Methanol gasoline under all production methods studied and Flowsheets can even produce lower GHG emissions compared to ethanol as a fuel supplement. However, the To produce methanol from natural gas, the same is not always true if methanol is used to natural gas must first be reformed to syngas produce gasoline from natural gas. before converting this syngas to methanol, further details of the Johnson Matthey reforming 1. Introduction options can be found elsewhere (3). In order to generate a syngas with the correct stoichiometry Many countries around the world are either using for methanol production there are four main or looking to use methanol as a fuel. China is process flowsheets for reforming the natural gas: currently leading the way and in 2015 used as 1. steam-methane reforming (SMR) much as 12 million metric tonnes of methanol 2. SMR with maximum CO2 addition (SMR + CO2) to fuel its cars, trucks and buses. Methanol now 3. combined reforming (CR), with SMR and makes up 8% of the Chinese fuel pool and in autothermal reforming (ATR) over a dozen provinces fuel blends such as M15 4. gas heated reforming (GHR) and ATR (GHR + (15% methanol and 85% gasoline) are sold for ATR).

Each of the reforming options listed above has include utilities other than the air separation unit advantages and the choice of flowsheet depends on (ASU), where applicable. The natural gas efficiency, a number of parameters, with the most influential on a LHV basis, has been split out to show where being the natural gas composition, operating cost the natural gas is used within the ISBL plant and is and capital cost. There are several other factors that quoted on a per tonne of methanol basis. also have a significant influence when assessing As an alternative flowsheet option, it is also the benefits of each process and the environmental possible to minimise the amount of natural gas impact of the plant is becoming increasingly burnt in the auxiliary boiler by maximising the more important. This is most noticeable in North number of compressors that are driven by motors, America where the cheap natural gas price has led allowing an improvement in the natural gas to numerous methanol projects being developed, efficiency of the ISBL plant as well as reducing the all of which require a Title V environmental permit CO2 emissions. The values in Table II are based on before construction can begin (4). maximising the import electricity while maintaining Figure 1 is an overview of the flow of carbon the minimum load on the auxiliary boiler. and the emission points from the methanol plant Two important trends are displayed in Tables I for Flowsheets 1 to 3. Figure 2 shows the same and II. The first is that the CO2 emissions in overview but for Flowsheet 4, the GHR + ATR Table I move in line with the natural gas efficiency flowsheet, which due to the nature of the reforming of the flowsheet, with the exception of the SMR + section has a different layout. CO2 flowsheet. This stands to reason because, as Using a typical North American pipeline natural Figures 1 and 2 show, again with the exception of gas composition from a recent methanol project in the SMR + CO2 flowsheet, natural gas is the only the USA, a comparison of the natural gas efficiency, carbon input into the ISBL plant, with methanol electrical power consumption and CO2 emissions and CO2 emissions the only output. Therefore, for the four flowsheets is shown in Table I based any carbon in the natural gas not converted to on a capacity of 5000 mtpd. These flowsheets are methanol will eventually leave the plant as CO2. based on driving all compressors and large pumps The SMR + CO2 flowsheet is the exception to this with steam driven turbines and utilising import rule as additional carbon is added to the process in electricity to drive the air cooler fans and smaller the form of CO2 injected upstream of the reformer. pumps only. This is the minimal electrical import This additional carbon helps improve the natural to the inside battery limit (ISBL) plant without gas efficiency but at the expense of increasing the the addition of a turbo generator, where the ISBL CO2 emissions from the ISBL plant. The increase in plant refers to the methanol unit only and does not CO2 emissions for the SMR + CO2 flowsheet is due

(a)

(b)

CO2 emissions CO2 addition Lights and fusel oil (SMR + CO2 only)

Loop purge

Crude Product Natural gas Syngas Methanol MeOH MeOH Reforming Distillation synthesis

CO2 emissions

Auxillary boiler

Fig. 1. Methanol plant overview for Flowsheets 1–3: (a) diagram of the unit operations for Flowsheets 1–3; (b) picture of a SMR + ATR used in Flowsheet 3

(a) (b)

Fusel oil

Crude Product Natural gas Syngas Methanol MeOH MeOH Reforming Distillation synthesis

CO2 emissions

Loop purge Auxillary boiler Lights

Fig. 2. Methanol plant overview for Flowsheet 4: (a) diagram of the unit operations for Flowsheet 4; (b) picture of a GHR + ATR used in Flowsheet 4

Table I 5000 mtpd Methanol Plant Comparison for Minimal Electrical Import

Units SMR SMR + CO2 CR GHR + ATR

Overall natural gas efficiency (LHV) GJ mt–1 32.6 31.6 30.8 31.0 Process 29.6 24.0 27.0 25.5 Reformer 1.7 6.4 3.0 0.0 Auxiliary boiler 1.3 1.2 0.8 5.5

Electricity MW 5.0 5.0 3.6 4.5 (MMBtu) (17) (17) (12.3) (15.4)

a –1 CO2 emissions mt h 92.8 144.9 [80.9] 71.7 77.3 (st h–1) (102.3) (159.7 [89.2]) (79.0) (85.2) a Based on using captured CO2 as a feedstock, the net CO2 emissions are shown in [ ] brackets

Table II 5000 mtpd Methanol Plant Comparison for Maximum Electrical Import

Units SMR SMR + CO2 CR GHR + ATR

Overall natural gas efficiency (LHV) GJ mt–1 32.4 31.4 30.7 25.5 Process 29.6 24.0 27.0 25.5 Reformer 1.7 6.4 3.0 0.0 Auxiliary boiler 1.1 1.0 0.7 0.0

Electricity MW 13.4 12.9 8.3 90.5 (MMBtu) (45.7) (44.0) (28.3) (308.6)

a –1 CO2 emissions mt h 90.4 142.8 [78.8] 70.9 13.9 st h–1 (99.6) (157.4 [86.9]) (78.2) (15.3) a Based on using captured CO2 as a feedstock, the net CO2 emissions are shown in [ ] brackets

299 © 2017 Johnson Matthey https://doi.org/10.1595/205651317X696216 Johnson Matthey Technol. Rev., 2017, 61, (4) to both the increase in natural gas fuel required in The GHR + ATR flowsheet incorporates a GHR in the reformer because of the reduced LHV of the series with an ATR, with an interchanger on the methanol loop purge gas as well as an increase feed to the GHR, as shown in Figure 3. in CO2 concentration in the recycled fuel from the The GHR consists of a refractory lined vessel methanol loop and distillation. Therefore, with any containing vertically supported tubes filled with nickel catalyst. The feed gas is preheated by CO2 injection flowsheet aside, the better the natural the GHR shell-side effluent gas before it passes gas efficiency of the ISBL plant the lower the CO2 down through the tubes where the endothermic emissions. If captured CO2 is used as a feedstock reforming reaction takes place (Equations (i)–(iii)). to the ISBL plant for CO2 injection flowsheets then Tables I and II show that the net CO emissions 2 Reforming CH4 + H2O ↔ CO + 3H2 (i) fall back in line with this trend.

The second important trend is that as the Water-gas-shift CO + H2O ↔ CO2 + H2 (ii) comparison between Tables I and II shows, for the SMR, SMR + CO2 and CR flowsheets there is no Heavy hydrocarbon reforming CnHm + significant scope to maximise the electrical import nH2O ↔ nCO + (½m + n)H2 (iii) while maintaining the minimum auxiliary boiler The heat required to drive the reaction is provided load. The SMR, SMR + CO and CR flowsheets 2 by reformed gas from the ATR which flows all generate high pressure (HP) steam as a way counter-currently on the shell-side of the reactor. of cooling the process gas after reforming. This The partially reformed gas leaves the tube-side of steam is a useful byproduct of the cooling process the GHR at approximately 700°C. because it can be used to power the turbines of The product from the GHR is fed to the ATR, which the large compressors on the plant. In addition, is also a refractory lined vessel. Oxygen is fed to the all flowsheets have an auxiliary boiler, whose burner gun of the ATR and this then mixes with the primary purpose is for start-up and shut-down. In hydrocarbon feed and burns in the upper section of normal operation the boiler is kept running but it the ATR. In the middle section the hot gas passes has a minimum turndown and so this steam also over a fixed catalyst bed, where the temperature has to be utilised within the ISBL plant. After all drops as the endothermic reactions proceed. this steam has been consumed, the additional Sufficient oxygen is fed to produce a temperature power requirements of the smaller compressors exiting the catalyst bed of 1020°C and at these are minimal and hence there is no real benefit conditions the reformed gas contains low levels of in switching from steam turbine driven to motor methane slippage. The hot reformed gas from the driven compressors for reducing the ISBL exit of the ATR passes to the shell-side of the GHR plant emissions and improving the natural gas where it flows counter-currently to the tubes and efficiency. In contrast, the GHR + ATR flowsheet uses the high temperature process gas to provide Feed gas heat for the reforming reaction in the GHR, Syngas which then allows all the compressors and large pumps to be electrically driven if required. The Interchanger ability to decouple the power requirement for the compressors and large pumps from the ISBL plant, Oxygen and the fact that the GHR + ATR flowsheet does not contain a SMR, means that the CO2 emissions of the ISBL plant can be reduced significantly for normal operation, as shown in Table II.

3. Gas Heating Reforming and Autothermal Reforming Flowsheet

To understand why the GHR + ATR flowsheet allows ATR for increased flexibility in choosing the power to GHR drive the rotating equipment, a more detailed Fig. 3. GHR + ATR flowsheet arrangement description of the flowsheet is given below.

300 © 2017 Johnson Matthey https://doi.org/10.1595/205651317X696216 Johnson Matthey Technol. Rev., 2017, 61, (4) provides sufficient heat for the reforming reaction reducing GHG emissions for the ISBL plant and in the GHR tubes. The reformed gas, now known also for providing a natural gas efficient flowsheet. as synthesis gas (syngas), exits the shell-side of Importing electricity allows the ISBL emissions the GHR and passes to the interchanger where it to be reduced but it doesn’t give a complete preheats the incoming feed gas. The syngas exits representation of the carbon intensity of producing the interchanger then passes to the downstream methanol using the GHR + ATR process. For heat recovery. certain states in the USA and Canada, for example No steam generation is required as all the high California, there has been a drive to reduce the grade process heat is recycled directly back into the carbon intensity of the fuels they use and this process which provides the ability to decouple the has resulted in the implementation of legislation power requirement for the GHR + ATR flowsheet in California called the low carbon fuel standard and move it outside battery limits (OSBL). This is (LCFS), a summary of which is given in Appendix A. an effective method of reducing the emissions and This standard looks at the total carbon emissions of improving the natural gas efficiency of the ISBL a fuel from well to wheels and so tries to capture plant. However, typically the imported power to the the total carbon intensity of that fuel over its whole plant will be from the grid, where the electricity life cycle. So taking gasoline as an example, the is generated from a portfolio of technologies, with LCFS aims to take into account the GHG emissions the largest contribution generally from fossil fuels during the extraction and refining of the crude oil, burnt in a power plant. A typical North American transporting the gasoline to the pump as well as portfolio of grid electricity is shown in Figure 4 and the emissions from the combustion engine in the this shows that 68% of the electricity is generated vehicle. In order to enable the carbon intensity of through burning carbon fuels. these fuels to be determined from well to wheels, The imported power means that the source of software has been developed to calculate the GHG the CO2 emissions generated by producing the emissions over the whole life cycle of the fuel. This electrical power is transferred from the ISBL software can therefore also be used to determine plant to the existing producers, so essentially the the carbon intensity of producing methanol on a emissions are just being moved from one location well to product basis, thus incorporating the GHG to another. When building a new methanol plant, emissions from transporting the natural gas to the this is advantageous as the emissions required for plant, the electricity used in the plant and from the Title V environmental permit in the USA are storing the methanol. only those for the new plant and do not include those for the existing producers supplying the 4. The Greenhouse Gases, Regulated import electricity. Therefore, in areas where GHG Emissions and Energy Use in emissions are restricted, the GHR + ATR flowsheet Transportation (GREET) Model with imported power offers the best flowsheet for GREET is the software developed by Argonne National Laboratory, USA, in conjunction with

Biomass power the Californian government’s LCFS to enable the generation 0.3% Oil fired power generation 0.5% calculation of GHG emissions for fuels produced and imported into the state of California (6). The software uses pathways to break each step of the product life cycle down and enables the emissions Nuclear power from each section of that process to be determined. generation 19.5% Using the GREET software, the figures generated Coal fired power below in Tables III and IV show the well to product Renewable power generation values for the four flowsheets based on steam generation 12.2% 41.5% driven turbines for the compressors and large pumps, as Table I. The first section of the table Natural gas fired power generation is divided into three parts for the GHG emissions. 26.2% The first is the processing and transportation of natural gas from the well to the methanol plant, the second is the emissions from the ISBL plant Fig. 4. A typical North American electricity mix (5) and the third is the storage of the methanol. The second section shows the GHG emissions for the

Table III GREET Numbers for Minimum Electrical Importa

Stage Units SMR SMR + CO2 CR GHR + ATR

–1 (a) Natural gas to plant gCO2e MJ methanol 13.0 12.6 12.3 12.4

b –1 (b) Methanol plant gCO2e MJ methanol 23.1 36.0 (20.1) 17.8 19.2

–1 (c) Methanol storage gCO2e MJ methanol 1.3 1.3 1.3 1.3

b –1 Subtotal gCO2e MJ methanol 37.4 49.9 (34.0) 31.4 32.9 Electricity

–1 North America mix gCO2e MJ methanol 0.76 0.76 0.54 0.67

–1 Renewable mix gCO2e MJ methanol 0.005 0.005 0.003 0.004

b –1 Total (North America mix) gCO2e MJ methanol 38.1 50.7 (34.8) 32.0 33.6

b –1 Total (renewable mix) gCO2e MJ methanol 37.4 49.9 (34.0) 31.4 32.9 aThe GREET values quoted in Tables III and IV have been peer reviewed but have not been confirmed as official GREET numbers by the Californian government b The net CO2 GREET GHG emissions are shown in brackets

Table IV GREET Numbers for Maximum Electrical Importa

Stage Units SMR SMR + CO2 CR GHR + ATR

–1 (a) Natural gas to plant gCO2e MJ methanol 12.9 12.5 12.2 10.2

b –1 (b) Methanol plant gCO2e MJ methanol 22.5 35.5 (19.6) 17.6 3.5

–1 (c) Methanol storage gCO2e MJ methanol 1.3 1.3 1.3 1.3

b –1 Subtotal gCO2e MJ methanol 36.7 49.3 (33.4) 31.2 15.0 Electricity

–1 North America mix gCO2e MJ methanol 2.03 1.95 1.23 13.7

–1 Renewable mix gCO2e MJ methanol 0.012 0.012 0.008 0.083

b –1 Total (North America mix) gCO2e MJ methanol 38.7 51.3 (35.4) 32.4 28.7

b –1 Total (renewable mix) gCO2e MJ methanol 36.7 49.3 (33.4) 31.2 15.1 aThe GREET values quoted in Tables III and IV have been peer reviewed but have not been confirmed as official GREET numbers by the Californian government b The net CO2 GREET GHG emissions are shown in brackets distributed electricity to the ISBL plant. There indication of the total possible reduction in carbon are two figures relating to the import electricity: intensity of producing methanol. the first is based on the standard North American The units for the values in Tables III and IV electricity mix, as shown in Figure 4, and the are grams of CO2 equivalent per megajoule of –1 second is based on a standard renewable energy methanol on a LHV basis (gCO2e MJ MeOH). electricity mix, as shown in Figure 5. The GREET GHG emission values in Table III, As Figure 6 shows, the USA and China are leading for flowsheets with the minimum electrical import, the way in the installation of renewable energy follow the same trend as the CO2 emissions in and therefore being able to use electricity where Table I. This is because for the minimum electrical the majority or all of the energy comes from a import flowsheets the contribution to the GHG renewable source is a distinct possibility in the near emissions from the import electrical power is minimal future. This real possibility of access to electricity and so the total emission figures are dominated by from a renewable source is why this option has the emissions from transporting the natural gas to been considered. In addition, it also gives a good the ISBL plant and from the ISBL plant itself.

the SMR, SMR + CO2 and CR flowsheets when using Biomass Solar the typical North American electricity mix compared 4.1% 0.3% to a reduction in emissions for the GHR + ATR flowsheet centres around the plant heat integration and utilisation of the steam from the auxiliary Wind 24.5% boiler. For the SMR, SMR + CO2 and CR flowsheets the generation of HP steam in the reformed gas cooling train means that there is only sufficient heat remaining in the reformed gas to provide approximately 55% of the distillation duty, with Geothermal Hydroelectric 3.9% 67.2% the remaining duty provided by low pressure (LP) steam. There is therefore a large LP steam demand, which typically has been satisfied by using medium pressure (MP) steam in back pressure turbines, with the LP steam header topped up by letting Fig. 5. Standard renewable energy mix (7) down a small amount of MP steam. This therefore maximises the amount of work performed by the However, the GREET GHG emission values in MP steam. When, however, the compressors driven Table IV, for flowsheets with the maximum by these turbines are switched to motor driven, the electrical import, show a different trend. For the LP steam demand remains the same and so the

SMR, SMR + CO2 and CR flowsheets, moving to the shortfall in LP steam is made up by letting down maximum electrical import actually increases the more of the MP steam. This then results in the use overall well to product GHG emissions compared to of MP steam becoming less efficient and so the the values in Table III when using the typical North GHG emissions for the combined ISBL plant and American electricity mix and only a small reduction import electricity actually increase. For the GHR when using the renewable electricity mix. This is + ATR flowsheet, the LP steam demand is small compared to the GHR + ATR flowsheet which shows because all the distillation duty is provided by the a reduction in GHG emissions of 15% and 54% reformed gas train cooling so the flowsheet does when using the typical North American electricity not need to incorporate backpressure turbines to mix and the renewable electricity mix respectively. satisfy the LP steam demand. Therefore, switching The reason for the increase in GHG emissions for the compressors from turbine to motor driven does

Hydro Solar PV (grid-tied) STEG Geothermal Wind Biomass China Germany Spain USA China USA Brazil Italy USA Philippines USA Brazil USA USA Algeria Indonesia Germany China Canada China Egypt/Morocco Mexico Spain Germany Russia Japan Australia Italy India Sweden

Fig. 6. Top countries with installed renewable electricity by technology in 2012 (8). PV = photovoltaic; STEG = solar thermoelectric generator

303 © 2017 Johnson Matthey https://doi.org/10.1595/205651317X696216 Johnson Matthey Technol. Rev., 2017, 61, (4) not mean additional MP steam has to be let down emissions over half that of the CR flowsheet, which to the LP steam level and so removing the steam has the second best emission figures. The GHR + driven turbines has a direct impact on the load of ATR flowsheet is the only flowsheet that doesn’t the auxiliary boiler, in proportion to the increase in generate HP steam as a byproduct of the process, electrical load and hence allows a total reduction in allowing a large portion of the energy requirement emissions. of the ISBL plant to come from electricity import. For the GHR + ATR flowsheet, running all the This in turn allows a large portion of the energy compressors, pumps and air coolers on imported required to make methanol to come from a electricity shows a modest saving on the GHG renewable source. emissions if the supplied electricity is from the In addition to calculating the well to product GHG grid with a typical North American electricity mix. emissions using GREET it is also possible to go one However, using a renewable energy source to step further and calculate the well to wheels value provide the electrical import power to the plant which allows methanol as a fuel to be compared to has a significant impact on the GHG emissions for all the other available transportation fuels. Table V producing methanol from natural gas, with the shows the comparison between the methanol well

Table V Well to Wheel Greenhouse Gas Emissions (9)

Vehicle Well to Fuel Vehicle Total operation product

–1 –1 –1 gCO2e MJ gCO2e MJ gCO2e MJ Methanol (85%) + Reformulated (a) 36.7 (a) 63.2 gasoline E10 (15%). Methanol (b) 47.3 (b) 73.9 produced using maximum North Methanol flexible-fuelled car 26.6 (33.8) (60.4) America mix electrical import (c) 31.3 (c) 57.9 (Notes (i) and (ii)) (d) 28.1 (d) 54.7

Methanol (85%) + Reformulated (a) 35.0 (a) 61.5 gasoline E10 (15%). Methanol (b) 45.7 (b) 72.3 produced using maximum Methanol flexible-fuelled car 26.6 (32.2) (58.7) renewable mix electrical import (c) 30.3 (c) 56.8 (Notes (i) and (ii)) (d) 16.6 (d) 43.1

Reformulated Gasoline E10 Gasoline car 66.3 25.0 91.3 (100%)

Low sulfur diesel (100%) Diesel car 75.7 17.1 92.8

Compressed natural gas (100%) Compressed natural gas car 57.6 18.6 76.2

Liquefied petroleum gas (100%) Liquefied petroleum gas car 64.7 12.5 77.2

Ethanol E85 (100%) (Note (iii)) Ethanol flexible-fuelled car 12.6 57.7 70.4

Gaseous hydrogen (100%) H2 car 0.8 94.5 95.3 Fischer-Tropsch diesel (100%) Fischer-Tropsch diesel car 73.1 36.5 109.6

Electricity (100%) (Note (iv)) Electric car 0 174.4 174.4

Notes for Table V i) The numbering for well to product and total GREET GHG emissions refers to the following flowsheets: 1. SMR

2. SMR + CO2 3. CR 4. GHR + ATR The GREET values quoted for the methanol (85%) + reformulated gasoline E10 (15%) fuel have been peer reviewed but have not been confirmed as official GREET numbers by the Californian government ii) The net CO2 GREET numbers are shown in brackets iii) Based on USA ethanol produced from corn iv) Electricity based on typical North America mix

304 © 2017 Johnson Matthey https://doi.org/10.1595/205651317X696216 Johnson Matthey Technol. Rev., 2017, 61, (4) to wheels carbon emissions and some of the other incorporate renewable electrical energy into the standard fuel types. process to enable a reduction in carbon intensity What Table V shows is that methanol as a fuel of methanol. The heat integration in the GHR + has a lower carbon intensity than gasoline over ATR flowsheet allows the flexibility to significantly its full life cycle, irrespective of which flowsheet is increase the electrical power input into the ISBL used to produce the methanol. It also highlights plant. This not only allows a large reduction in the that methanol as a blend stock for gasoline is GHG emissions from the ISBL plant but also allows less carbon intensive than using ethanol, unless a total reduction in the carbon intensity of the non-captured CO2 injection is used on the flowsheet. process over its entire life cycle and significantly When producing gasoline from crude oil, the so if the source of electricity is from renewable well to product value for reformulated gasoline energy. –1 E10 in Table V is 25.0 gCO2e MJ . Therefore, to From well to wheels, methanol produced from reduce the carbon intensity the well to product natural gas provides a significant reduction in GHG GHG emissions for producing gasoline from emissions when compared to standard gasoline. natural gas via methanol would need to be below Even when compared to ethanol, methanol –1 25.0 gCO2e MJ . As Tables III and IV show, shows a modest reduction in GHG emissions with the exception of the GHR + ATR flowsheet, and emphasises why methanol is such a good the GHG emissions for producing methanol from supplement to gasoline fuel for the reduction of –1 natural gas range from 31.2–51.3 gCO2e MJ GHG emissions. –1 which is already higher than the 25.0 gCO2e MJ If the intended destination of the gasoline is to for refining crude oil. Therefore, even if the carbon a state or country that has implemented a LCFS, intensity of producing gasoline from methanol was then in general making gasoline from natural gas zero, it would not be possible to produce gasoline via methanol does not reduce the overall carbon with a lower carbon intensity from natural gas via intensity of the gasoline and in fact would increase methanol. The only exception to this is the GHR the carbon intensity over the whole life cycle. + ATR flowsheet using the maximum electrical The exception would be processes that are able import from a renewable energy source which to utilise both renewable energy and the GHR + –1 has a well to product value of 15.1 gCO2e MJ ATR flowsheet in order to produce a low carbon and there are companies that are currently intensity gasoline. developing novel flowsheets, incorporating the GHR + ATR process and renewable energy Acknowledgements sources to produce low carbon intensity gasoline from natural gas. This article is an extended and updated version of the International Methanol Technology Operators Conclusions Forum (IMTOF) London 2015 presentation (10). Amelia Cook, Process Engineer at Johnson Matthey, Through raising HP steam in the SMR, SMR + is acknowledged for her contribution to the data

CO2 and CR flowsheets it is not possible to easily collection and processing.

Glossary CR Combined reforming, with steam methane reforming and autothermal reforming GHG Greenhouse gas GHR + ATR Gas heated reforming and autothermal reforming LCFS Low carbon fuel standard M15 15% methanol and 85% gasoline fuel blend MTPD Metric tonnes per day OSBL Outside battery limits SMR Steam methane reforming

SMR + CO2 Steam methane reforming with maximum CO2 addition

Appendix A What is the Low Carbon Fuel Standard? As further background surrounding the LCFS, the following is a summary (11). In California, USA, they have developed a method for determining the carbon intensity of a fuel for the whole of its life using the concept from well to wheels. In January 2010 the Californian state government implemented the LCFS which calls for a minimum 10% reduction in emissions per unit of energy by 2020. The policy focuses on decarbonising fuels for transportation and is a performance standard that is based on the total amount of carbon emitted per unit of energy. This crucially includes all the carbon emitted in the production, transportation and use of the fuel. In America, transportation accounts for two-thirds of all the oil consumed and causes approximately one-third of all the GHG emissions. In an attempt to address this, the LCFS assigns a company (for example an oil refiner, importer or blender) a maximum level of GHG emissions per unit of fuel energy it produces. This level then declines each year with the intention of putting the state on a path to reducing total emissions. There are several ways that regulated parties can comply with the LCFS and in the Californian model there are three compliance strategies available: (a) Refiners can blend low GHG fuels, for example biofuels made from cellulose or wastes, into gasoline and diesel. (b) Refiners can buy low GHG fuels, for example natural gas, biofuels, electricity and hydrogen. (c) Refiners can buy credits from other refiners or use banked credits from previous years. The LCFS in California is not the only fuel standard that has been implemented. A similar scheme is in place in British Columbia in Canada and others have been proposed in Ontario, Canada, several other states in North America as well as the European Union.

References 6. “CA-GREET 2.0 Model”, Air Resources Board, California Environmental Protection Agency, 1. Demonstration Projects, Learn More About California, USA, 6th May, 2016 Exciting Demonstration Projects using Methanol 7. H. Cai, M. Wang, A. Elgowainy and J. Han, as a Vehicle Fuel from Around the World, Methanol “Updated Greenhouse Gas and Criteria Air Fuels, Methanol Institute, Singapore: http://www. Pollutant Emission Factors and Their Probability methanolfuels.org/on-the-road/demonstration- Distribution Functions for Electric Generating projects/ (Accessed on 30th August 2017) Units”, ANL/ESD/12-2, a Report by Argonne 2. Automotive Fuel, Methanol Institute, Singapore: National Laboratory, Tennessee, USA, 1st May, http://www.methanol.org/automotive-fuel/ 2012, 142 pp (Accessed on 30th August 2017) 8. Z. Shahan, “18 Fun Renewable Energy Charts from 3. H. A. Claxton, ‘Horses for Courses’, International NREL Director Dan Arvizu & Ren21′s Renewables Methanol Technology Operators Forum (IMTOF), 2013 Global Status Report”, CleanTechnica, London, UK, 19th–22nd June, 2011 Important Media Network, Sustainable Enterprises 4. “Operating Permits Issued under Title V of the Media, Inc, 7th November, 2013 Clean Air Act”, United States Environmental 9. GREET.net v1.3.0.10631, Argonne National Protection Agency, North Carolina, USA: https:// Laboratory, Illinois, USA, 2015 www.epa.gov/title-v-operating-permits (Accessed 10. A. Ingham, ‘Low Carb Methanol, the Lighter Way on 30th August 2017) to Produce Methanol’, International Methanol 5. “Annual Energy Outlook 2013 with Projections Technology Operators Forum (IMTOF), London, to 2040”, DOE/EIA-0383(2013), US Energy UK, 7th–10th June, 2015 Information Administration, Washington, DC, 11. D. Sperling and S. Yeh, Issues Sci. Technol., 2009, USA, April, 2013 25, (2), 57

The Author

Alan Ingham is a Licencing Manager for Johnson Matthey’s methanol technology. Before assuming this role he spent over 10 years working in the methanol department as a chartered Senior Process Engineer undertaking roles as both a Lead Process Engineer and Site Commissioning Engineer. Alan graduated from Nottingham University, UK, in 2005 with a first class Masters’ degree in Chemical Engineering, following which he joined Johnson Matthey Process Technologies and has been working for the company ever since.

www.technology.matthey.com

Lithium Sulfur Batteries: Mechanisms, Modelling and Materials Conference Recent advances in lithium-sulfur batteries research

Reviewed by Amina Touidjine locally changes the solvent system to the benefit Johnson Matthey, Blounts Court, Sonning of polysulfide (PS) solubility and kinetics. PVP is

Common, Reading RG4 9NH, UK soluble but forms an insoluble complex with Li2Sx. Ketjenblack® is the most challenging carbon Email: [email protected] adhesion-wise but shows the best electrochemical performance. 1000 mAh g–1 was obtained for 50 cycles using electrodes with a S loading of Lithium Sulfur: Mechanism, Modelling and 2 mg cm–2. Their conclusion is that PEO:PVP is a Materials (Li-SM3) was organised by Oxis Energy suitable water-based binder for electrodes based Ltd, UK, Imperial College London, UK, and on highly porous C black. the Joint Center for Energy Storage Research Diana Golodnitsky (Tel Aviv University, Israel) (JCESR), USA. It was held at the Institution of gave a talk about the strategies to impede PS Engineering and Technology (IET), Savoy Place, shuttle by using barrier membranes between London from 26th–27th April 2017. More than cathode and separator or by depositing a thin 150 researchers from around the world attended barrier over them. Toray and SGL carbon papers this event, 44 of them delivered talks and about are the barrier membranes used which are also 30 people presented their posters. A significant electrochemically active. Cells containing these number of delegates from the private sector were barrier membranes delivered high S utilisation present at this event including representatives and excellent columbic efficiency, which suggests from Airbus Defence and Space, France; Arkema, that the PS shuttle has been reduced. Electrodes France; Daimler/Mercedes, Germany; LG Chem, or cathodes deposited with thin PEO-XC72 or South Korea; National Aeronautics and Space lithium tin phosphorous sulfide (LSPS)-PEO based Administration (NASA) Jet Propulsion Laboratory, layers also showed this positive effect. However, USA; Renault, France; Sony Corporation, Japan; the cells with barrier layers had capacity fading Toyota, Japan and Umicore, Belgium. problems. Markus Hagen (Fraunhofer Institute for Chemical Electrochemistry Technology (ICT), Germany) studied electrolyte decomposition. This decomposition is monitored Matthew Lacey (Uppsala University, Sweden) gave by fitting a pressure sensor and applying mass a talk about water-based electrodes and more spectroscopy (MS) to a LiS cell. The pressure particularly about the effects of polyethylene increases during discharge due to the release of oxide (PEO) as an electrolyte additive or as gases such as carbon disulfide (CS2) and nitrogen, a binder. His group prepared a carbon-sulfur produced by the decomposition of the electrolyte. ® material by melting carbon (Ketjenblack N2 gas is produced due to the decomposition

EC-600JD) and sulfur under air at 150°C. For the of lithium nitrate (LiNO3) in the electrolyte, an water based electrode they added C65 and 4:1 additive which is widely used to suppress the PS PEO:polyvinylpyrrolidone (PVP). PEO swells and redox shuttle. In contrast, electrolytes without it

did not show any N2 gas evolution; therefore, the PS. For encapsulated S in a C micropore system speaker concluded that LiNO3 is not beneficial since a pseudo-solid-state reaction takes place, where it promotes decomposition of the electrolyte. the signature of the galvanostatic discharge curve

is different from the conventional Li/S8 cell (1). Characterisation Techniques Barchasz also talked about the tomography studies done in her group. An ex situ X-ray phase contrast Alice Robba (French Alternative Energies and tomography technique (2) was used to characterise Atomic Energy Commission (CEA), Innovation the evolution of the cathode morphology upon Laboratory for the Technologies of New Energies cycling and to track the loss of S from the conductive and Nanomaterials (LITEN), France) gave an C network. A special cell was used for the operando interesting talk about Li2S particle size influence X-ray absorption tomography which consists of an on the first charge working mechanism of lithium aluminium can that houses the Li-metal anode, a sulfide (Li2S) based Li-ion batteries. In order to separator containing electrolyte and a C paper as overcome the safety issue with the use of Li metal the cathode. The solid S8 was initially placed over in a LiS system, her group used Li2S as a positive the C paper, which disappears during discharge electrode and silicon as a negative electrode. and redistributes throughout the cathode during

Replacing S8 by Li2S as the active material allows the subsequent charging sequence. The images the use of a safer negative electrode. However, S8 were two-dimensional (2D) cross sections of the and Li2S have different conductivity and solubility cell and each constituent material was shown in a properties. The focus of the work presented was different grey-scale. the effect of Li2S particle size on the charge Due to their high stability with no risks of mechanism of Li2S based Li-ion batteries. When vaporisation and leakage, solid-state batteries

Li2S was prepared by electrodeposition (nano-Li2S), are a promising alternative to liquid electrolytes. two plateaus were visible for S8 evolution (≈2.3 V Jessica Lefevr (Technical University of Denmark) and 2.45 V). However, when commercial Li2S gave a talk about solid electrolytes for LiS systems

(micro-Li2S) was used, a huge difference was based on lithium borohydride (LiBH4), which is an seen in the electrochemical signature. In order to interesting salt for LiS due to its light weight and its understand this difference between nano-Li2S and electrochemical stability. While the orthorhombic micro-Li2S, the respective batteries were analysed phase (Pnma) is stable at room temperature and by operando X-ray absorption and emission has a low ionic conductivity (10–5 mS cm–1 at 30°C), spectroscopies coupled with operando X-ray the hexagonal phase (P63/mmc), which is stable diffraction. When nano-Li2S was used, a smaller above 110°C, has a much higher ionic conductivity –1 polarisation and a smaller activation barrier for (1 mS cm at 120°C). Confinement of BH4 in

Li2S oxidation were observed. Operando studies mesoporous silicon dioxide (SiO2) allows fast showed a very nice PS evolution and a complete ionic conductivity even at room temperature. disappearance of Li2S. However, when micro-Li2S Operando X-ray diffraction, tomography and was used, a higher polarisation was seen upon Raman spectroscopy measurements were used to charging and there was no PS detection while understand the electrochemistry and improve the + E>3.3 vs. Li/Li . There was consumption of Li2S but batteries’ performance. never a total disappearance and there was an early Deyang Qu (University of Wisconsin Milwaukee, appearance of S. Therefore, Robba’s conclusion USA) reported an in situ characterisation technique was that use of nano-Li2S is better because the in which electrolyte retrieved from the cell at polarisation can be lowered and there is a smaller different states of charge (SOC) were analysed

‘activation’ barrier for Li2S oxidation. using high performance liquid chromatography Céline Barchasz (CEA LITEN) gave an overview (HPLC) and MS. 4-(dimethylamino)benzoyl chloride of different types of LiS systems. In a Li-ion/S (3) is added to the electrolyte prior to HPLC and MS system the cathode initially contains Li2S and the measurements, in order to prevent the dissolved anode is graphite, which has better cyclability since PS in them from undergoing disproportionation the parasitic reactions involving Li metal anode reactions. HPLC is used to separate different are prevented. In a catholyte cell, comprising a PS and MS is used to identify their speciation. high concentration (7 M) of PS in electrolyte, a Based on the in situ HPLC and MS results Qu has special current collector with vertical C fibres is proposed reaction mechanisms for both discharge used. However the specific energy is limited to and charge processes. According to him, the two only <180 Wh kg–1 due to the limited solubility of plateaus characteristic of the charge and discharge

–1 1000 >550 km Applications 800 >400 km For aerospace applications, gravimetric energy 600 >225 km density is more important than the volumetric Today 400 >200 km energy density. NASA and Airbus are currently 200 160 km using LiS technology and both gave a talk about 100 km Specific energy, Wh kg Specific energy, 50 km80 km the importance of the research in this field. 0 Pb-acid Ni-Cd Ni-MH Li-ion Future Zn-air Li-S Li-air The challenge is still to have higher S loading Li-ion (m = 13 mg cm–2 ideally). (S) Price, US$ kW h–1 The Zephyr aircraft is a high-altitude 200 600 900 600 <150 <150 <150 <150 long-endurance unmanned aerial vehicle that holds Available Under R&D development the world record for the longest flight duration of any aircraft: 14 days non-stop. It has been in Fig. 1. Practical specific energies for some rechargeable batteries, along with estimated development since 2004 and uses LiS batteries driving distances and pack prices (Reprinted by to store the energy captured by its solar panels permission from Macmillan Publishers Ltd: Nature and power the aircraft at night. Sarah Bassett Materials (4), copyright (2012)) (Airbus Defence and Space) gave a talk about the model ‘Zephyr T’ that is under development. This model will provide communications and internet connectivity to developing world countries without separators and new electrolyte formulations. the need for installing expensive infrastructure Anode protection was often mentioned as a key on the ground. She insisted on the need for challenge (often in the conclusions slide) but no ultra-lightweight batteries and explained why LiS is one gave a talk about it. Very little was about the perfect for this application. cathode preparation. Engineering is also crucial for the development of this technology. Conclusion

LiS batteries are a most promising candidate for References the next generation of electrical energy storage. 1. E. Markevich, G. Salitra, A. Rosenman, Y. They are made of cheap and abundant elements Talyosef, F. Chesneau and D. Aurbach, J. Mater. and they have a theoretical specific energy of Chem. A, 2015, 3, (39), 19873 ca. 2500 Wh kg–1 or 2800 Wh l–1, which is much 2. L. Zielke, C. Barchasz, S. Waluś, F. Alloin, J.-C. higher than that of Li-ion batteries (Figure 1). Yet Leprêtre, A. Spettl, V. Schmidt, A. Hilger, I. Manke, their cycling stability shows limitations and scaled J. Banhart, R. Zengerle and S. Thiele, Sci. Rep., production technologies need to be established to 2015, 5, 10921 meet cost targets. The LiSM3 was a very interesting conference 3. D. Zheng, D. Qu, X.-Q. Yang, X. Yu, H.-S. Lee where many issues with LiS were discussed. LiS and D. Qu, Adv. Energy Mater., 2015, 5, (16), is of interest where light weight is important such 1401888 as in aircraft. The main current research activity 4. P. G. Bruce, S. A. Freunberger, L. J. Hardwick and is to control the PS via the development of new J.-M. Tarascon, Nature Mater., 2012, 11, (1), 19

The Reviewer

After a Masters in Materials Science at Université de Montréal, Canada, and a PhD in electrochemistry at Université de Picardie Jules Verne, France, working on the optimisation of the negative electrode based on silicon for Li-ion battery applications, Amina Touidjine went to the United Nations Environment Programme (UNEP) in Geneva, Switzerland, to do a traineeship on chemical waste management. She joined Johnson Matthey in October 2016 where she is currently working on LiS battery technology.

www.technology.matthey.com

Progress and Outlook on Gasoline Vehicle Aftertreatment Systems Meeting the tightening limits for criteria pollutants and greenhouse gas emissions for China, Europe and the USA

By Ameya Joshi On the regulatory changes, the focus is on Corning Incorporated, One Riverfront Plaza, Europe and China, where Euro 6 regulations Corning, NY 14831, USA are being implemented along with changes to the certification test cycle and RDE targets. The Email: [email protected] combined non-methane organic gas (NMOG) + NOx target of 30 mg mile–1 in the USA will be the tightest standard globally by 2025, accompanied Driven by concerns on deteriorating ambient air by the GHG reductions being discussed for quality, measures are being taken across the world light-duty vehicles. A variety of advanced to adopt and enforce tighter vehicular emission vehicle technologies are being developed and regulations to minimise tailpipe unburned implemented to meet these targets, and some of hydrocarbons, nitrogen oxides (NOx) and these are discussed along with the implications particulate matter (PM). In regions with advanced from an aftertreatment systems perspective. regulations, the focus is on limiting the pollutants As gas temperatures continue to decline as the under real-world or in-use driving conditions. engine improvements lead to heat rejection to Given the intensified effort to curb global warming the exhaust, innovative approaches are focused and limit fossil fuel use in the transportation on lowering cold-start emissions. Particulates sector, several countries have adopted targets on from gasoline engines are also regulated and tailpipe carbon dioxide emissions. This confluence various control measures are under investigation. of stringent regulations for both criteria pollutant Gasoline particulate filters seem poised for wide and greenhouse gas (GHG) emissions is leading adoption in major markets. Electrification of the to a rapid adoption of advanced powertrains and powertrain is proceeding rapidly as well, and while aftertreatment technologies. This is a review this can deliver impressive gains in fuel economy, of some of these recent advances pertinent to cool-down on aftertreatment components and reducing vehicular emissions and developing emissions associated with engine restarts are improved aftertreatment solutions. The scope is being studied. On the other hand, the internal limited to gasoline vehicles where the adoption combustion (IC) engine continues to evolve, of gasoline direct injection (GDI) and hybrid and advanced gasoline engine technologies are powertrain technologies is leading to significant challenging the well-to-wheels CO2 emissions shifts in the aftertreatment solutions. There is of electric powertrains. Research continues on significant work being done to improve diesel low-temperature combustion with the aim of aftertreatment systems especially in light of avoiding soot and NOx formation altogether, and real-world driving emission (RDE) regulations. researchers are pursuing several strategies. By These are not covered here, rather the reader definition, these lead to low exhaust temperatures is referred to a previous article in this journal’s which pose a challenge to the aftertreatment archive (1), and to a more recent review (2). system.

1. Light-Duty Regulations tighter limits and full compliance with RDE. CFs will be finalised by 2022. Major regions such as Figure 1 gives an overview of the latest regulations Beijing may adopt regulations earlier, but the for gas emissions implemented in Europe, China and details are yet to be clarified. There are some the USA. The US Environmental Protection Agency important differences compared to the European (EPA) Tier 3 standards will be the toughest in the regulations. The regulations are fuel neutral world, with reductions phased in starting 2017 and make no distinction between GDI and port through 2025 and ending with a combined limit on fuel injected (PFI) vehicles. This is an important NMOG + NOx at 30 mg mile–1. Figure 1(a) shows provision, as PFI vehicles have high PN emissions the concurrent targets for emissions of criteria under cold ambient conditions, and PFI hybrids pollutants and GHGs. Figure 1(b) compares this emit particulates even beyond cold-start (5) with the tightening of emissions limits happening (likely due to transients associated when the elsewhere in Europe and China through the Euro 6 engine turns on). As shown in Figure 1, the and China 6 regulations, respectively. gas emissions in the final China 6b stage will be Euro 6 light-duty (LD) regulations introduce much tighter, by roughly a factor of 2 for carbon several important changes. A particle number limit monoxide and hydrocarbons (HC), 40% lower for of 6 × 1011 km–1 applies to GDI vehicles. Starting NOx and 33% lower for PM. There is also a limit –1 in September 2017, new type approvals are to be of 20 mg km on nitrous oxide (N2O) emissions, certified using the new World Harmonised Light absent in Europe (N2O is also limited in the USA at Vehicle Test Procedure (WLTP) (3). The third RDE 10 mg mile–1). A recent study by Ricardo (6) shows package has been recently published (4), which that managing N2O emissions at such low levels confirms the conformity factor (CF) of 1.5 for is not trivial, and that catalysts will have to be particulate number (PN). The CF is expressed as optimised to meet such tight limits. The durability (1 + margin PN), with the margin set at 0.5, requirement is also increased to 200,000 km for providing for measurement uncertainties China 6b, as compared to 160,000 km in Europe, associated with the portable emissions monitoring and an emission warranty and defect-reporting system (PEMS) equipment, and subject to annual requirement has been implemented for the first review. The CF for NOx was confirmed at 2.1 in time. China is serious about strict enforcement previous packages. of these regulations, and this is reflected in the China 6 LD regulations will be implemented new framework for certification, conformity of in two stages: China 6a starting in 2020 with production and in-use compliance testing (7). RDE monitoring, and China 6b in 2023 with Much of the onus for these tests lies with the

(a) (b) GHG emissions 100 120 38% reduction Euro 6d - Diesel (2017)

–1 80 100 2015 China 6a (2020) Euro 6d - Gasoline (2017) 80 Tier 2 + GHG-1 60 –1 Gas emissions 42% 56% 60 70% reduction (Gasoline) (Diesel) 2020 40 Tier 3 + GHG-2 (phase-in) 44% (Diesel) 40 China 6b (2023) US Tier 3 (2025) NOx, mg km 20 50% (Gasoline)

NMOG + NOx, mg mile 20 2025 Tier 3 + GHG-2 0 0 0 50 100 150 200 250 300 0 20 40 60 80 100 –1 –1 CO2, g mile Total HC, mg km

Fig. 1. Light-duty regulations for tailpipe emissions in major markets: (a) the concurrent tightening of criteria pollutant and CO2 emission limits in the USA; (b) by 2023, the proposed limits in China will be tighter than those in Europe, roughly by a factor of two

312 © 2017 Johnson Matthey https://doi.org/10.1595/205651317X696306 Johnson Matthey Technol. Rev., 2017, 61, (4) original equipment manufacturers (OEMs), while vehicle technologies are rapidly falling, such the regulatory authorities maintain the right to that compliance with the 2025 GHG standards check the results. could be up to 40% lower than anticipated in the India will skip one level of regulations, the joint assessment. Some of the advanced engine first of its kind for a major market. It will move technologies which are already being adopted from Bharat Stage (BS) 4 directly to BS6 for LD rapidly include direct injection, cooled exhaust gas regulations, and BS VI for heavy duty (HD) in recirculation (c-EGR), high compression ratio (CR) 2020. RDE test procedures are being developed to Atkinson cycle, stop-start, cylinder deactivation, reflect the local driving conditions, and compliance advanced turbocharging and mild hybridisation. is expected to be enforced by 2023. Other countries A detailed review of all these technologies is not such as Australia and Thailand are also rapidly the focus of this paper. But we will highlight some implementing stricter regulations with a view to of the recent work. c-EGR is effective in reducing Euro 6 levels in the coming years. These global the tendency to knock at high loads, therefore changes have implications for rapid development eliminating fuel enrichment. At part loads, both and deployment of advanced engine technologies c-EGR and cylinder deactivation improve fuel and engine emissions aftertreatment systems, efficiency through reduced pumping work. A which are covered in the following sections of this recent study (12) looked at the incremental benefit review. of adding these technologies to a 2 l naturally In contrast to the other major regions listed aspirated Atkinson engine with high CR (14:1). above, PM regulations in the USA are mass based: The brake thermal efficiency (BTE) improvements starting in 2017, both EPA Tier 3 and California over the engine map were used to assess the Air Resources Board (CARB) Low-Emission Vehicle improvements over the Federal Test Procedure III (LEV III) standards set limits on the allowable (FTP)/Highway Fuel Economy Test (HwFET) –1 tailpipe particle mass at 3 mg mile for light-duty two-cycle CO2 reductions. For a future vehicle, vehicles (8). The LEV III regulations continue to the improvements using c-EGR were predicted at tighten to 1 mg mile–1 starting 2025. 7.6%, while cylinder deactivation gives another 2% improvement. 2. Advances in Engine Technologies Variable compression ratio (VCR) technology was commercialised recently by Nissan, achieved via The broader technology trends are being shaped its ‘multi-link’ system (13). The CR can be varied by the concurrent needs to deliver on vehicles between 8:1 and 14:1. A combination of VCR, with high fuel economy and near-zero tailpipe multi-port injection, direct injection and downsizing gas and particulate emissions. One study (9) helped realise a 30% fuel economy improvement discussed pathways to meet the CARB target of when moving from a conventional 3.5 l V6 engine –1 48 gCO2 km by 2050, an 80% reduction over to a 2.0 l engine. The technology has benefits for 2010 levels. It was shown that for the goals to reduced emissions: high CR leads to improved fuel be met independently through IC engines, fuel consumption and lower CO2 emissons, while lower cell vehicles or pure-electric vehicles, a large shift CR during cold start helps with higher exhaust to low-carbon fuels is needed: 60–80% of fuel temperatures for early catalyst light-off and lower will have to be derived from cellulosic ethanol, particulates due to reduced fuel impingement on

48% of H2 and >36% electricity generation from the piston. renewables, respectively. This points to a long- Dedicated-EGR, where one cylinder’s exhaust term shift, already underway, to renewable fuels is fed into the intake manifold for constant 25% and electrification of the powertrain. In the short EGR for a four-cylinder engine, was demonstrated term, meanwhile, various engine improvements to deliver brake specific fuel consumption (BSFC) are being deployed to improve the efficiency of <200 g kWh–1 from 12–14 bar brake mean modern vehicles. effective pressure (BMEP) and 1500–3500 rpm. In their joint Draft Technical Assessment Vehicle testing showed the potential to reach Report (10), the EPA, National Highway Traffic NOx + NMOG emissions of 31 mg mile–1, just above Safety Administration (NHTSA) and CARB the LEV III limit. Lower exhaust temperatures pose concluded that the GHG standards for model year a challenge for HC emissions, and HC traps are 2022–2025 LD vehicles can be met using a wide being considered (14). range of advanced gasoline technologies. A recent Water injection is gaining some interest as analysis (11) concludes that the costs of advanced a potential technology to mitigate knock and

313 © 2017 Johnson Matthey https://doi.org/10.1595/205651317X696306 Johnson Matthey Technol. Rev., 2017, 61, (4) fuel enrichment at high loads. Studies show the ignition engines. Oak Ridge National Laboratory, potential to achieve ~5–7% reduction in fuel USA, has summarised the various levels of fuel consumption over test cycles, with much higher stratification strategies being pursued (17), as gains up to 17% observed at full load (15, 16). shown in Figure 2. The overall goal is to achieve: The amount of water needed can be high, the (a) sufficient premixing of fuel and air to avoid soot studies cited above have noted the best fuel formation associated with fuel rich combustion, efficiency obtained at water-to-fuel ratios upwards and (b) reduced peak combustion temperatures of 50%. The exhaust temperature is reduced (by via dilution with air or EGR to avoid NOx formation. ~50–100°C), such that NOx emissions are reduced, While promising high fuel efficiency, the lower but HC emissions increase. combustion temperatures also imply high unburned Much progress is also being made on achieving HC and CO emissions and the challenges with lean, low-temperature combustion via compression achieving downstream catalyst warm-up. Other

(a) 4.5 Fig. 2. Low-temperature combustion strategies 4.0 (14). aLTGCI = low-temperature gasoline Soot 3.5 Diesel compression ignition combustion formation 3.0 2.5 2.0

1.5 Incomplete combustion Low-temperature Local equivalence ratio Local equivalence 1.0 combusion (LTC) 0.5 NOx formation 0 1000 1400 1800 2200 2600 3000 Local temperature, K

(b) Reactivity controlled Premixed charge Conventional diesel compression ignition (RCCI) compression ignition (PCCI) combustion (CDC)

Gasoline vapour Gasoline spray

Diesel vapour Diesel spray

–360 –300 –240 –180 –120 –60 0 Homogeneous LTGCIa spectrum Heterogeneous combustion combustion

Increase fuel stratification Increase ignition delay to improve controllability, to avoid NOx and improve combusion soot emissions while efficiency and reduce maintaining control combustion noise over combustion

314 © 2017 Johnson Matthey https://doi.org/10.1595/205651317X696306 Johnson Matthey Technol. Rev., 2017, 61, (4) challenges include operability over a wide load 3. Stoichiometric Gasoline Emissions range and combustion timing control. Aftertreatment An example of the progress made is the development of the gasoline direct injection New regulatory developments outlined in the first compression ignition (GDCI) engine targeted section – lower limits on tailpipe criteria pollutants, to meet Tier 3 Bin 30 limits (18). BSFC of the introduction of particulate number and mass 211–214 g kWh–1 is demonstrated over a wide load regulations for gasoline engines and challenging range, utilising a high CR of 16:1 and injection CO2 or fuel economy targets – are stimulating work pressure of 350 bar. The difficulty of achieving the on new and improved aftertreatment systems for very high conversion efficiencies (>99%) needed gasoline engines. The underlying challenge for for HC have translated into a rather complex catalysts is achieving early light-off despite the aftertreatment system, which at the latest stage lowering exhaust temperatures as a consequence includes a pre-turbo catalyst, HC trap, selective of efficient combustion. Improved combustion and catalytic reduction (SCR) and a passive gasoline the introduction of gasoline particulate filters is particulate filter (GPF) for off-cycle particulates. imminent to meet the particulate regulations under As this article was being written, Mazda (19) real-world driving conditions. has announced its plans to introduce the first supercharged homogeneous charge compression 3.1 Three-Way Catalysts ignition (HCCI) engine to market in 2019, projected The cold-start challenge is highlighted in a recent to cut fuel consumption by 20–30% over its current study (22) comparing the emissions across gasoline engines. This is in line with its vision 82 LD vehicles with both PFI and GDI technology, (20) for achieving 25% improvement in IC engine complying with various levels of regulations (Tier 0 efficiency for well-to-wheels CO2 equivalence with to super ultra-low emissions vehicle (SULEV)). electric vehicles (EVs) in many electrical grids. Tailpipe emissions were measured on the As mentioned at the beginning of this section, cold-start unified cycle (UC). The ratio of cumulative despite the advances in engine improvements cold-start emissions in the first bag divided by listed above, there is a growing consensus that the distance-specific emissions after engine these may be insufficient to meet CO targets 2 warm-up in the second bag was used to quantify the beyond 2025, and OEMs are increasingly adopting cold-start contributions to overall emissions. For various levels of electrification to lower fuel HC, this ratio increased from 15.4 miles for Tier 1 consumption. Start-stop technology is becoming vehicles to 101.5 miles for SULEV vehicles. Put in more prevalent, full hybrids and plug-in hybrid words, modern vehicles will have to drive over 100 electric vehicles (PHEVs) can offer >30% reduction miles after engine warm-up for the emissions to in fuel consumption, while mild hybrids using a equate to the first few seconds after a cold-start. 48 V battery look particularly attractive, providing This points to the tremendous progress which has many of the benefits of a full hybrid at a fraction been made in catalyst technology – the conversion of the cost. The actual benefit of hybrids is tied is near complete once the catalyst is hot and closely with the driving conditions and state of active – but it also points to the challenge ahead to charge (SOC) of the battery. In a study simulating address the cold-start emissions. a LD vehicle with advanced engine configurations Light-off in three-way catalysts (TWCs) is not (21), it was shown that using a hybrid powertrain, a new topic. The various factors which impact the fuel economy almost doubled over the cold-start conversion have been studied extensively, New York city cycle (compared to a naturally both experimentally and using simulations (23). aspirated 1.6 l, four-cylinder baseline), while the For TWCs, there are several approaches being benefit was only 12% on highway and real-world explored to achieve lower cold-start emissions. driving conditions. Pure EVs remove the use of One is increasing the precious metal content, but IC engines altogether and as such can offer the that adds to the cost and also produces limited highest benefits from electrification, although improvements: the additional catalyst will still the well-to-wheels benefit is closely tied with the be ineffective at very low temperatures. Another carbon footprint of the fuel used to generate the is optimising the location of the platinum group electricity. metal (pgm). According to one analysis (24),

the pgm costs could vary by a factor of two from the first layer of an underfloor Pd/Al2O3/OSC

(US$35–US$70) depending on the catalyst catalyst with Ba/Zn/CeO2 led to a 10% reduction in location for achieving non-methane hydrocarbon NMOG + NOx emissions over the Los Angeles Route (NMHC) + NOx emissions of about 42 mg mile–1. Four (LA4) test cycle using a 2011 model year Conversely, a US$40 pgm loading might deliver Honda Civic (partial zero emissions vehicle (PZEV), NMHC + NOx emissions between 105 mg mile–1 and 1.8 l L4). The Zn reduces the oxygen binding 30 mg mile–1, depending on the catalyst design. energy, increasing the O-availability and CO Continuing along the pgm route, another approach oxidation at low temperatures, while the barium is the catalyst design: choice of precious metals, reduces CO2 adsorption which limits the oxygen their concentrations and the choice of support. availability. This catalyst was commercialised and One such study (25) aims at lowering the light-off applied to the 2016 model year Civic. Another temperature to achieve 90% conversion of HC, CO example (29) is the use of iron-based catalysts as and NOx on aged catalyst at 150°C (T90). Ongoing a way to reduce pgm. The combination of Fe/CeO2 work has considered several design levers: was shown to work: the CeO2 reduces the binding (a) catalyst preparation: the results suggested energy of oxygen to Fe, and helps the catalyst post-impregnation may provide better access to respond to oscillating redox conditions. Normally pgm than catalysed slurry, under the conditions Fe would react with alumina to form aluminates tested; which leads to loss of aged performance, so a

(b) palladium level: the best performance was perovskite oxide (LaFeO3) structure was adopted

found at 2% concentration on alumina (Al2O3) to inhibit structural changes. At only 10% of the Rh support, while higher loadings helped on loading compared to a conventional catalyst, the

zirconia (ZrO2) and titania (TiO2) supports; aged performance of this catalyst was found to be

surface area than ZrO2 and TiO2. Fresh ZrO2 The reduction of pgm is going to be especially

has lower fresh T90, but Al2O3 has better aged critical for meeting stringent upcoming regulations performance; in cost sensitive markets, as is the case with BS6 in (d) impact of ageing: the performance actually India. The viability of using advanced spinel oxides

improved with ageing (lower T90) for Al2O3 due with low pgm loadings has been demonstrated (30) to better Pd dispersion, while it deteriorated on in both underfloor and close-coupled TWCs. the other supports. The choice of substrate has played a critical role Much progress has been made via optimising these in facilitating cold-start performance of TWCs. variables, so that T90 for HC was lowered from Reducing the thermal mass of the substrate is 345°C on a commercial catalyst to 253°C on a an effective method to enable early heat-up of combination of 0.5% rhodium/TiO2/Al2O3, under the catalyst. This has been achieved through aged conditions (26). While this is impressive, increased porosity of the substrates from the there is clearly more work to be done to reach the traditional 27–35% up to 55%, while meeting the ambitious goal of T90 at 150°C. Another study (27) strength requirements for canning and designing focused on improving the oxygen storage capacity the right microstructure to maintain on-wall (OSC) and thermal durability using an advanced coating to provide good access for the reactants alumina and OSC material. An improvement in to catalyst sites (31). The substrates were shown OSC of 15–30% was achieved over a wide range of to reduce light-off time and cold-start emissions temperature and mass flow rates, translating into by up to 24%, and have been commercialised (32). conversion improvements for CO up to 20% and A different approach taken recently (33) is the use NOx up to 10% over wide air:fuel ratios. of substrates with varying cell design aimed at gas A very different approach being considered is the flow redistribution. The idea is to use higher cell reduction or elimination of pgm to reduce costs. density in the centre which increases resistance New catalysts are being developed, aimed primarily to flow, redistributes it towards the outer channels at the underfloor location where the performance and improves overall gas catalyst contact. Vehicle requirements are not as stringent as at the testing on the US06 Supplemental Federal Test close-coupled position, and the lower temperatures Procedure (SFTP) using the new substrate design pose milder conditions for the catalyst ageing. in the underfloor position led to similar or better

A new barium/zirconium/ceria (Ba/Zn/CeO2) NOx conversion despite 20% lower pgm. However, catalyst formulation was shown (28) as a viable as compared to the close-coupled position, the alternative to Al2O3. The replacement of Al2O3 pgm content and the extent of flow non-uniformity

316 © 2017 Johnson Matthey https://doi.org/10.1595/205651317X696306 Johnson Matthey Technol. Rev., 2017, 61, (4) are both smaller in the underfloor position. Other In another study (36), the emissions of a GDI approaches, such as better packaging, are often vehicle were measured over the FTP-75 test cycle used to address this issue in the close-coupled with and without a catalysed GPF in the underfloor position. position. Through the use of the coated GPF, the Other than addressing cold-start and pgm tailpipe emissions of CO, total hydrocarbon (THC) content, there are some new system layout and NOx were reduced further by 86%, 38% and choices emerging to address the tightening limits 34% respectively. The catalysed GPF also helped of both criteria pollutants and fuel economy. One reduce NOx preferentially under the aggressive example (34) is the use of a ‘fuel-cut NOx trap driving conditions of the US06 test cycle, leading (FCNT)’ replacing the underfloor TWC and placed to 88% reduction in NOx over baseline. downstream of the muffler. Fuel-cut events help The durability of coated GPFs was demonstrated for improve fuel economy, but lead to excessive NOx a 1.4 l GDI engine in China (37). After 160,000 km emissions due to lean conditions. NOx emissions testing, the engine-out emissions were unchanged. were found to increase by a factor of three due The TWC performance was robust, with only a slight to such fuel-cut events on the FTP-75 test cycle. (~15°C) increase in light-off temperature, and also The NOx trap addresses this issue. Moving the trap maintaining filtration efficiency (~85%) over its behind the muffler helped reduce the temperature lifetime. A study evaluating the impact of RDE on by 60–70°C and facilitate an optimal temperature emissions from two Euro 6b certified turbocharged window for the lean NOx trap (LNT). The FCNT GDI vehicles highlights the challenge ahead (38): helped upgrade an ultra-low emissions vehicle one of the vehicles tested showed a NOx emissions (ULEV) vehicle to SULEV standards, while retaining increase under real-world driving conditions by a the fuel economy benefits of fuel-cuts. factor of two or three times compared to WLTP The biggest change for TWCs from a systems- values. The NOx emissions exceeded the limit of level perspective is the introduction of catalysed 60 mg km–1 even with a fresh exhaust system. GPFs. As will be discussed in more detail in the Replacing the underfloor TWC with a coated GPF subsequent section, GPFs are being developed for helped to reduce both CO and NOx emissions. catalysed applications in both the close-coupled and underfloor position, where they combine the 3.2 Gasoline Particulate Filter functionality of filtration and TWC. This is similar in many respects to the combination of SCR deNOx Boosted, downsized GDI engines offer improved functionality on diesel particulate filters (DPFs) in fuel economy over the traditional PFI engines. The diesel aftertreatment. Several studies have now charge cooling associated with direct injection of shown that it is possible to maintain high reaction fuel in the cylinder reduces the risk of engine knock performance with catalysed GPFs replacing the and enables operation at higher compression ratios. traditional TWC coated flow-through substrates. Also, coupled with electronic controls, the amount One study (35) showed that an aftertreatment and timing of fuel injected into the chamber can system was able to meet the SULEV30 emission be more precisely controlled. However, similarly targets of 30 mg NMOG + NOx via the use of a to diesel, insufficient time for mixture preparation close-coupled high-porosity substrate mentioned and fuel impingement on the relatively cooler in the previous paragraph and an underfloor GPF combustion chamber surfaces leads to pockets of catalysed with Rh. While most of the NOx conversion fuel-rich combustion and an increased propensity occurred over the close-coupled TWC, the GPF for particulate formation. Europe and China have PN helped to increase conversion by 18–30% over the regulations to limit the fine particulates emitted, and US FTP-75 test cycle. These have also been tested there is accordingly considerable focus on engines, in conjunction with an underfloor catalysed GPF, fuels and aftertreatment systems (for example the where a 10–12% reduction in NMHC and NOx over GPF) to control particulates. As mentioned earlier, the FTP cycle was found over standard substrates. particulate matter regulations in the USA are mass The system tested had two close-coupled catalysts based, and while studies show that it may be and most of the conversion was found to occur over possible to meet the 3 mg mile–1 standard using the first close-coupled catalyst. The best system engine methods, the 1 mg mile–1 limit may require tested achieved 25 mg mile–1 of NMOG + NOx, the use of GPFs. Several studies have shown below the US Tier 3 Bin 30 limits, with the use a direct correlation between solid particulate of high-porosity substrates in both close-coupled number and mass emissions. Broadly, the studies positions. (for example (39–41)) report a correlation of

2 × 1012 particles mg–1, based on which the PN of most modern GDI vehicles and also some PFI limit of 6 × 1011 km–1 translates to ~0.5 mg mile–1. vehicles will include a GPF. The potential system A detailed review of fuels’ impact on particulates layouts are shown in Figure 3 (46), and include is beyond the scope of this article, but there is bare (uncoated) or TWC-coated GPFs. In the former, sufficient evidence that components with higher the main functionality is filtration only, while the resistance to volatilisation – such as aromatics latter provides the added functionality of gaseous – increase particulates, while other components emissions conversion. The filters can be added in such as ethanol – which adds oxygen – decrease the close-coupled or in the underfloor position. particulates (36). The increasing particulates with Several studies have now shown that the aromatics is especially important and is being backpressure with the addition of a GPF can be studied in China (42) given the variability in fuel maintained relative to the base OEM system, quality. A PM Index (PMI) model has been proposed thereby having little or no impact on fuel economy. (43), which quantifies the relationship between fuel There are several design levers which can be turned composition and particulate emissions. The model to avoid incurring a CO2 penalty due to addition of accounts for the double bond equivalents, vapour GPF. The choice of larger filter diameter, optimised pressure and weight fraction of each component washcoat loading, cell design of both the filter and in the fuel, and several studies (44) have shown a upstream TWC, are all factors which can help keep strong correlation between the PMI and measured tailpipe CO2 at levels close to the original system. particulate emissions. In a previously cited study (32), fuel economy Many studies have confirmed that particulate with a bare GPF was found to be equivalent to emissions are highest during cold starts, due the OEM system with a traditional underfloor to lower fuel volatility, fuel impingement on TWC, while particulate emissions reduced to colder surfaces and less time for evaporation. 0.27–0.4 mg mile–1 over the FTP-75 test cycle, Cold ambient temperature has a similar effect, comfortably meeting the 1 mg mile–1 proposed and one study (45) has shown that even PFI standard. Other than proper design of the GPF, and hybrid vehicles exceeded the EU regulated the pressure drop reduction was also assisted by PN limit of 6 × 1011 km–1 on the New European replacing the 900/2 close-coupled catalyst with a Driving Cycle (NEDC), when tested under sub-zero 600/3 cell design. Moreover, the GPFs are designed ambient temperatures. The average PN during the for low pressure drop taking the soot and ash initial 180 s was almost identical for both GDI and accumulation over the vehicle lifetime. In an analysis PFI vehicles. In the EU, cold-start emissions are on high-mileage GPFs (47), no significant increase included in the RDE analysis following the latest in fuel consumption was seen after 130,000 miles (third) RDE package, and the PN limits apply of vehicle ageing, despite the approximately two- to both GDI and PFI vehicles in China, so these fold increase in pressure drop due to accumulated findings are important. ash. Meanwhile, the nature and role of ash is getting In light of the particulate regulations discussed increasing attention. Generally, the ash amounts are above, it appears that the aftertreatment systems much lower than accumulated by filters in diesels.

(a) (b)

c-GPF

TWC c-GPF TWC GPF

Fig. 3. Light-duty gasoline system architecture with GPF. Various options shown, broadly classified into: (a) TWC-coated c-GPF and (b) uncoated GPF systems for filtration only (46)

Lambert et al. (47, 48) reported 0.95–1.62 g of ash PFI truck, and approximately four times more in the first 3000 km under real and random driving gas-phase PAH emissions (52). Through a study on conditions and 58–61 g after 150,000 miles. This three Euro 5 GDI vehicles on the WLTC (53), it is translates to ~0.3–0.5 mg km–1 in the low mileage surmised that PAH are likely associated with soot, conditions and ~0.25 mg km–1 over the lifetime. as uncoated GPFs placed after the TWC were found Even a small amount of ash accumulation on the to reduce PAH by up to 90% for the larger species; channel walls helps increase filtration efficiency and 98% for benzo[α]pyrene, a five-ring known for GPFs, although with a pressure drop penalty, carcinogen. Smaller two- and three-ring PAH are as already known from DPF experience. In the reduced 40–60% with the uncoated GPF. Looking study mentioned above, an increase in filtration ahead, Europe would like to continue to revise the efficiency from ~60% to 80%, that is an increase of conformity factors downwards as measurement ~20%, associated with a pressure drop penalty of accuracy improves, and include particles below 20–22%. Studies are also converging on the source 23 nm in the PN measurements considering health and ultimate location of the ash within the filter. impacts of fine particles (54). Studies show the The study found 50% of ash derived from engine latter looks feasible, and that GPFs are effective oil consumption, the rest related to possible engine in capturing these sub-23 nm particles, given that wear and some washcoat loss from upstream TWC. filtration improves with Brownian motion of smaller An ash accumulation study (49) on an underfloor particle sizes. In an investigation of particulate GPF after 150,000 miles on a 3.5 l GDI vehicle emissions from five GDI vehicles (55), high PN also showed that only 50% of ash was collected emissions (~6 × 1012 km–1) were recorded when from engine oil sources. After accumulating sub-23 nm particles were included, and GPFs were 150,000 miles on vehicles, ash was distributed found to be effective in near total capture of these ~60% on channel walls and 40% as plugs. small particles. Another study (56) on a 1.4 l Euro Accumulation of soot leads to an increase in 6b GDI engine under real-world driving conditions pressure drop, so the periodic regeneration found an increase in engine out PN of >50% when of filters via soot oxidation is important. The particles >7 nm and <23 nm were included, while possibility of active regeneration is an area still tailpipe emissions after a catalysed GPF increased being explored. Under conditions of high engine out by only ~20%. All tailpipe emissions with the use soot and low temperatures (for instance underfloor of GPF comfortably met the limit of 6 × 1011 km–1 GPF location), active strategies may be required. even with the inclusion of particles below 23 nm. However, in other cases, passive regeneration could be sufficient, given high exhaust temperatures 4. Lean-Burn Gasoline NOx Control typical of gasoline engines coupled with the low soot accumulation rates (compared to diesel). One Lean-burn gasoline engine technology has the study (50) has shown that soot oxidation can occur potential to improve fuel economy by 5–15% passively through fuel-cut events, characterised compared with a stoichiometric baseline. TWCs are by high O2 concentration and low flow rate, and ineffective to convert the high NOx generated under that an equilibrium is reached such that incoming lean (l>1) conditions and as such lean gasoline soot from the engine is balanced by the soot that requires additional deNOx aftertreatment. Similarly is oxidised. to diesel, NOx could be treated using active SCR, Overall, GPFs are now a robust technology and but the additional cost and urea consumption would are expected to be prevalent in upcoming gasoline negate some of the fuel efficiency improvements. vehicle aftertreatment systems. Other than their Passive SCR is one solution, wherein an upstream benefits in capturing particulates, GPFs also help TWC generates ammonia under periodic short reduce polyaromatic hydrocarbons (PAH), several of rich conditions, which is utilised by a downstream which are known carcinogens. In an analysis (51) of SCR catalyst to reduce the previously stored the exhaust plumes of seven different GDI vehicles NOx. Studies focus on catalyst formulations and under idle, cruise and accelerating conditions, rich-lean modulation conditions for generating aromatic compounds using the benzene, toluene, sufficient 3NH for deNOx, without incurring a fuel ethylbenzene and xylene (BTEX) proxy were found penalty due to excessive NH3 slip. One study (57) in the 62–96th percentile (higher in winter) of found an NH3:NOx ratio of 1.13 at the TWC outlet all vehicles on the road. Similarly, a 2014 GDI optimum for achieving 99.5% NOx conversion, while pick-up was found to have fourteen times more total also delivering 11.5% reduction in fuel consumption. PM-bound PAH emissions than a similarly powered Reduced cycle time was found to be beneficial to

reduce NH3 oxidation during lean periods. The there is also more attention being given to the oxidation history of the catalyst was also shown real-world driving emissions from various levels to strongly affect the NOx reduction performance of hybridisation of the powertrain. While hybrids (58). Rh was identified as the key catalyst for NO emit less pollutants due to the reduced fuel usage, reduction and also responsible for the hysteresis there are also some unique challenges which need between the oxidised and reduced states in the to be addressed for meeting the stringent gas and catalyst. Future challenges for LNTs include HC and particulate emission standards. One key issue CO slip and sulfur tolerance (59). is the lower exhaust temperatures and catalyst A novel pgm-free underfloor NOx catalyst was cool-down which occurs when the engine is off and developed (60) using a copper/zeolite upper layer the energy demand is being met by the battery. and a nickel/CeO2 bottom layer. The former helps Much has been said earlier about the cold-start with SCR-based conversion of NOx while the latter problem, and this is in some ways exacerbated helps with reduction of NO using exhaust CO. The in the case of hybrids, where temperatures can

NH3 required for SCR conversion is generated via remain near or below light-off even during cruising the upstream TWC, which is a trilayer formulation conditions. The actual decrease in temperature of Pd/Rh/Pd. On a 2.4 l naturally aspirated PZEV and catalyst effectiveness depends on the driving engine, the new underfloor catalyst converted 85% conditions and battery SOC. One study analysed NOx on the LA4 test and 40% on the US06 test. fuel consumption, CO and NOx emissions as a Despite the efficient combustion of lean gasoline function of the SOC and engine-off duration, engines, GPFs will likely be needed to meet the using the Toyota Prius full hybrid vehicle (62). For particulate regulations. Particulate emissions engine-off periods longer than 30 s, the CO emissions remain high in both the lean stratified and lean index (EI, emissions divided by fuel consumption) homogenous combustion modes (61). The increased up to 63%, while NOx EI increased up to modulated rich periods required to make ammonia 73%, likely due to catalyst cooling after engine-off for passive NOx control make the particulate and the resulting loss in its efficiency. emissions even worse. GPFs are very effective, As with gaseous emissions, there is evidence capturing particulates with >95% filtration that particulate emissions for hybrids can be efficiency even with the highly transient nature of equivalent or even higher than vehicles without any this combustion strategy. electrification. In a study in Japan (63), particulate As major regions continue to reduce the emissions were measured on the JC08 test cycle allowable tailpipe CO2 emissions, lean-burn from several gasoline and diesel vehicles, including gasoline engines may provide one approach one GDI and one PFI hybrid. As has been described to meet these future regulations. Ricardo has earlier, most of the particulate emissions for analysed potential pathways for achieving further GDI and PFI engines were found to occur during reductions in tailpipe NOx and CO2 from gasoline 100–200 seconds after a cold start, with almost no vehicles, beyond those mandated in the latest emissions after the engine warmed up. However, in Euro 6 and Tier 3 regulations (6). With a view the case of hybrids, emissions were found to occur to achieving improvement in fuel economy, even after engine warm-up and through the entire lean stratified combustion was considered, test cycle. Overall, emissions from hybrid PFI vehicles and several aftertreatment system choices were found to exceed those from diesels fitted with were simulated. The study concludes that for a DPFs. The issue here is the emission of particulates C-segment vehicle, a twin LNT system and a GPF associated with transient engine turn-on events is the most cost-effective way of keeping tailpipe which occur throughout the drive cycle. Another NOx at 60 mg km–1 (40 mg km–1 + CF = 1.5). study used a GDI engine to simulate hybrid electric Exhaust temperatures are predicted to decrease vehicle (HEV) operation by stopping and starting with efficient combustion, thus challenging the the engine during NEDC testing (64). Although the catalysts to convert CO and HC effectively. Also, engine only operated 28% of the time during the the challenge of meeting the N2O regulations in cycle, the PN emissions were 4.5 times higher than the USA and China is highlighted. when the engine was run in conventional mode. The above problem is especially acute with PHEVs, 5. Hybrid Vehicles where ‘high-powered cold-starts’ have been shown to greatly increase the particulate emissions Powertrain electrification is proceeding rapidly due (65). This is the scenario when the engine turns to the improved fuel economy it offers. Accordingly, on for the first time when the vehicle is already

320 © 2017 Johnson Matthey https://doi.org/10.1595/205651317X696306 Johnson Matthey Technol. Rev., 2017, 61, (4) moving at a high speed or under high load, and 6. Summary results in cold-start emissions which are larger than they would be under normal cold-starts. A In response to mandates of lower fuel consumption recent study focused on this issue and measured worldwide, the IC engine continues to evolve and real-world emissions from a 1.5 l GDI PHEV deliver impressive gains in fuel efficiency. Along certified to Euro 6b standards, and driven with with powertrain electrification, advanced gasoline various levels of charge on the battery (66). As vehicle technologies are being adopted rapidly, at expected, the PN emissions in urban conditions lower cost points, and are seen as important tools were totally eliminated when the battery was to lower GHG emissions. New engine and vehicle fully charged as the battery alone propelled the technologies also present new challenges to the vehicle. Under charging mode though (empty aftertreatment systems: efficient combustion and battery), the demand on the IC engine was high hybridisation lead to reduced exhaust temperatures, and PN emissions were the highest. For the total producing further demands on the catalyst to RDE drive cycle though, for all cases with various work more efficiently at lower temperatures. And levels of charge including a 100% charged battery, regulations continue to tighten globally, such that the PN emissions were lowest for the reference new vehicles will be required to reduce criteria GDI conventional vehicle with a GPF. For the pollutants to near-zero levels, and under real-world fully charged battery, there were no cold-start driving conditions. Finally, gasoline vehicles are emissions in the urban part, but the emissions receiving attention for their particulate emissions: were much higher when the engine turned on later there is a growing body of knowledge that gasoline along with a high power demand. Hybrids and vehicles can produce more particulates than their power management strategies are evolving diesels equipped with DPFs (67). In response to and clearly there is room to improve on some of these challenges, progress continues on all fronts: the above shortcomings. In the meantime robust engine improvements, fuels and aftertreatment aftertreatment solutions are seen to be necessary systems. Advanced catalysts, substrates, improved to cover the operation of hybrids under widely packaging and gasoline particulate filters are being varying real-world conditions. This would likely developed and implemented, and are well poised to involve some of the advanced emission control mitigate vehicular pollution. solutions mentioned through the paper: cold start strategies, advanced substrates and catalysts which promote early heat-up and emissions Acknowledgement conversion, improved heat retention to avoid rapid cool-down when engine is turned off, and the use The author would like to express his gratitude to of filters to capture particulates during regular Tim Johnson, Corning Inc, USA, for his invaluable engine operation and transients associated with guidance and suggestions that have greatly helped the switching from battery to engine. improve the quality and rigour of this article.

Abbreviations c-EGR cooled exhaust gas recirculation BMEP brake mean effective pressure CF conformity factor BS Bharat Stage CO carbon monoxide BSFC brake specific fuel consumption

CO2 carbon dioxide BTE brake thermal efficiency CR compression ratio BTEX benzene, toluene, ethylbenzene and xylene DPF diesel particulate filter

CARB California Air Resources Board EI emissions index

EPA Environmental Protection Agency NMOG non-methane organic gas

EUDC Extra Urban Driving Cycle NOx nitrogen oxides

EV electric vehicle OEM original equipment manufacturer

FCNT fuel-cut NOx trap OSC oxygen storage capacity

FTP Federal Test Procedure PAH polyaromatic hydrocarbons

GDCI gasoline direct injection compression PEMS portable emissions monitoring ignition system

GDI gasoline direct injection PFI port fuel injected

GHG greenhouse gas pgm platinum group metal

GPF gasoline particulate filter PHEV plug-in hybrid electric vehicle

HC hydrocarbons PM particulate matter

HCCI homogeneous charge compression PMI particulate matter index ignition PN particulate number HD heavy duty PZEV partial zero emissions vehicle HwFET Highway Fuel Economy Test RDE real-world driving emission IC internal combustion SCR selective catalytic reduction LA4 Los Angeles Route Four SFTP Supplemental Federal Test Procedure LD light-duty SOC state of charge LEV Low-Emission Vehicle SULEV super ultra-low emissions vehicle LNT lean NOx trap THC total hydrocarbon LTGCI low-temperature gasoline compression ignition TWC three-way catalysts

NA naturally aspirated UC unified cycle

NEDC New European Driving Cycle ULEV ultra-low emissions vehicle

NHTSA National Highway Traffic Safety VCR variable compression ratio Administration WLTP World Harmonised Light Vehicle Test NMHC non-methane hydrocarbon Procedure

References Powertrain Conference, Namyang, Korea, 25th– 26th October, 2016 1. T. Johnson, Platinum Metals Rev., 2008, 52, (1), 23 15. J. Kim, H. Park, C. Bae, M. Choi and Y. Kwak, Int. 2. T. Johnson and A. Joshi, ‘Review of Vehicle Engine J. Engine Res., 2016, 17, (7), 795 Efficiency and Emissions’, SAE Technical Paper 16. F. Hoppe, M. Thewes, H. Baumgarten and J. 2017-01-0907 Dohmen, Int. J. Engine Res., 2016, 17, (1) 86 3. Commission Regulation (EU) 2017/1151, Official 17. A. B. Dempsey, S. J. Curran and R. M. Wagner, Int. J. Eur. Union, 2017, 60, (L175), 1 J. Engine Res., 2016, 17, (8), 897 4. Commission Regulation (EU) 2017/1154, Official 18. M. C. Sellnau, ‘Aftertreatment for Low-Temperature J. Eur. Union, 2017, 60, (L175), 708 Combustion and US Tier3-Bin 30 Emissions’, SAE 5. H. Yamada, S. Inomata and H. Tanimoto, Emiss. 2017 Light Duty Emissions Control Symposium, Control Sci. Technol., 2017, 3, (2), 135 Washington DC, USA, 23rd–24th January, 2017 6. M. Christie and A. Ward, ‘Aftertreatment and 19. ‘Mazda Announces Long-Term Vision for Technology Emissions Control for Improved GHG and Air Development: Sustainable Zoom-Zoom 2030’, Quality’, ERC 2017 Symposium: Impact of Future Mazda, Hiroshima, Japan, 8th August, 2017 Regulations on Engine Technology, Madison, USA, 20. R. Okita, ‘Mazda SKYACTIV-G Engine with New 14th–15th June, 2015 Boosting Technology’, Advanced Clean Cars 7. H. He and L. Yang, ‘China’s Stage 6 Emission Symposium: The Road Ahead, Air Resources Standard for New Light-Duty Vehicles (Final Rule)’, Board, Sacramento, USA, 27th–28th September, The International Council on Clean Transportation 2016 (ICCT), Washington, USA, 16th March, 2017 21. S. Mamikoglu, J. Andric, and P. Dahlander, ‘Impact 8. ‘Control of Air Pollution from Motor Vehicles: Tier 3 of Conventional and Electrified Powertrains on Fuel Motor Vehicle Emission and Fuel Standards: Final Economy in Various Driving Cycles’, SAE Technical Rule’, EPA-HQ-OAR-2011-0135, Part II, Federal Paper 2017-01-0903 Register, Environmental Protection Agency, 22. G. T. Drozd, Y. Zhao, G. Saliba, B. Frodin, Washington, USA, 28th April, 2014, Vol. 79, Issue C. Maddox, R. J. Weber, M.-C. O. Chang, H. 81, p. 23414 Maldonado, S. Sardar, A. L. Robinson and A. H. 9. C. Gearhart, MRS Energy & Sustain., 2016, 3, E8 Goldstein, Environ. Sci. Technol., 2016, 50, (24), 13592 10. Office of Transportation and Air Quality, US Environmental Protection Agency, National Highway 23. T. Watling and J. Cox, SAE Int. J. Engines, 2014, Traffic Safety Administration, US Department of 7, (3), 1311 Transportation and California Air Resources Board, 24. D. Ball and D. Moser, ‘Impact of LEV-III and Tier- ‘Draft Technical Assessment Report: Midterm III Emission Regulations’, SAE 2017 Light Duty Evaluation of Light-Duty Vehicle Greenhouse Gas Emissions Control Symposium, Washington DC, Emission Standards and Corporate Average Fuel USA, 23rd–24th January, 2017 Economy Standards for Model Years 2022-2025’, 25. J. Theis, A. Getsoian, and C. Lambert, SAE Int. J. EPA-420-D-16-900, Environmental Protection Fuels Lubr., 2017, 10, (2), 583 Agency, Washington, USA, July, 2016 26. C. Lambert, ‘Next Generation Three-Way Catalysts 11. N. Lutsey, D. Meszler, A. Isenstadt, J. German for Future, Highly Efficient Gasoline Engines”, and J. Miller, ‘Efficiency Technology and Cost 2017 Annual Merit Review and Peer Evaluation, Assessment for US 2025–2030 Light-Duty US Department of Energy, Washington DC, USA, Vehicles’, International Council on Clean 5th–9th June, 2017 Transportation (ICCT), Washington, USA, White 27. J. Schoenhaber, J. M. Richter, J. Despres, M. Paper, 22nd March, 2017 Schmidt, S. Spiess and M. Roesch, ‘Advanced TWC 12. C. Schenk and P. Dekraker, ‘Potential Fuel Economy Technology to Cover Future Emission Legislations’, Improvements from the Implementation of cEGR SAE Technical Paper 2015-01-0999 and CDA on an Atkinson Cycle Engine’, SAE 28. M. Hashimoto, Y. Nakanishi, H. Koyama, S. Inose, Technical Paper 2017-01-1016, SAE International, H. Takeori, T. Watanabe, T. Narishige, T. Okayama Warrendale, USA, 28th March, 2017 and Y. Suehiro, ‘Development of Low Temperature 13. Y. Fujimoto, ‘Introduction of Variable Compression Active Material for Three Way Catalyst’, SAE Turbo Engine’, Advanced Clean Cars Symposium: Technical Paper 2016-01-0932 The Road Ahead, California Air Resources Board, 29. Y. Hanaki, M. Fujimoto and J. Itou, ‘Alternative Sacramento, USA, 27th–28th September, 2016 Technology for Platinum Group Metals in 14. T. Alger, ‘New Developments in Dedicated Automobile Exhaust Gas Catalysts’, SAE Technical EGR Engines’, 16th Hyundai Kia International Paper 2016-01-0930

30. S. Golden, Z. Nazarpoor and R. Liu, ‘TWC Using Technologies for Advanced Gasoline Applications’, Advanced Spinel Materials and Prospects for BSVI 15th Hyundai Kia International Powertrain Compliance’, SAE Technical Paper 2017-26-0126 Conference, Seoul, Korea, 27th–28th October, 31. C. Tanner, K. Twiggs, T. Tao, D. Bronfenbrenner, Y. 2015 Matsuzono, S. Otsuka, Y. Suehiro and H. Koyama, 47. C. K. Lambert, M. Bumbaroska, D. Dobson, ‘High Porosity Substrates for Fast-Light-Off J. Hangas, J. Pakko and P. Tennison, SAE Int. Applications’, SAE Technical Paper 2015-01-1009 J. Engines, 2016, 9, (2), 1296 32. S. Otsuka, Y. Suehiro, H. Koyama, Y. Matsuzono, C. 48. C. K. Lambert, T. Chanko, M. Jagner, J. Hangas, X. Tanner, D. Bronfenbrenner, T. Tao and K. Twiggs, Liu, J. Pakko and C. J. Kamp, SAE Int. J. Engines, ‘Development of a Super-Light Substrate for LEV 2017, 10, (4), 1595 III/Tier3 Emission Regulation’, SAE Technical 49. N. Custer, C. J. Kamp, A. Sappok, J. Pakko, C. Paper 2015-01-1001 Lambert, C. Boerensen and V. Wong, SAE Int. J. 33. T. Yoshida, H. Suzuki, Y. Aoki, N. Hayashi and K. Engines, 2016, 9, (3), 1604 Ito, SAE Int. J. Engines, 2017, 10, (4) 50. T. Boger, D. Rose, P. Nicolin, N. Gunasekaran and 34. M. Y. Choi, ‘A New Catalyst Technology for T. Glasson, Emiss. Control Sci. Technol., 2015, 1, Improving Fuel Economy’, 16th Hyundai Kia (1), 49 International Powertrain Conference, Namyang, 51. N. Zimmerman, J. M. Wang, C.-H. Jeong, M. Korea, 25th–26th October, 2016 Ramos, N. Hilker, R. M. Healy, K. Sabaliauskas, J. 35. A. Craig, J. Warkins, K. Aravelli, D. Moser, L. Yang, S. Wallace and G. J. Evans, Environ. Sci. Technol., D. Ball, T. Tao and D. Ross, SAE Int. J. Engines, 2016, 50, (4), 2035 2016, 9, (2), 1276 52. G. Karavalakis, D. Short, D. Vu, J. Yang and T. 36. T. W. Chan, M. Saffaripour, F. Liu, J. Hendren, K. Durbin, ‘Monoaromatic and Polycyclical Aromatic A. Thomson, J. Kubsh, R. Brezny and G. Rideout, Emissions from GDI and PFI Vehicles on Ethanol Emiss. Control Sci. Technol., 2016, 2, (2), 75 and Iso-butanol Blends’, SAE 2016 International Powertrains, Fuels and Lubricants Meeting, 37. A. Zhou, ‘Legislation Trend and GPF Development Baltimore, USA, 24th–26th October, 2016 Trend in China’, 16th Hyundai Kia International Powertrain Conference, Namyang, Korea, 25th– 53. M. Muñoz Fernandez and N. Heeb, ‘PAH and 26th October, 2016 Nitro-PAH Emissions from GDI Vehicles’, 19th ETH-Conference on Combustion Generated 38. J. Schoenhaber, N. Kuehn, B. Bradler, J. M. Nanoparticles, Zürich, Switzerland, 28th June–1st Richter, S. Bauer, B. Lenzen and C. Beidl, ‘Impact July, 2015 of European Real-Driving-Emissions Legislation on Exhaust Gas Aftertreatment Systems of 54. ‘Transposition of GTR15 (WLTP) into UN Turbocharged Direct Injected Gasoline Vehicles’, Regulations: Update from WLTP Transposition SAE Technical Paper 2017-01-0924 Task Force’, Informal Document GRPE-75-18, 75th GRPE Session, Working Party on Pollution and 39. I. A. Khalek, T. Bougher and J. J. Jetter, SAE Int. J. Energy, UNECE, Geneva, Switzerland, 6th–9th Fuels Lubr., 2010, 3, (2), 623 June, 2017 40. M. M. Maricq, J. Szente, M. Loos and R. Vogt, SAE 55. J. Czerwinski, P. Comte, N. Heeb, A. Mayer and Int. J. Engines, 2011, 4, (1), 597 V. Hensel, ‘Nanoparticle Emissions of DI Gasoline 41. T. W. Chan, E. Meloche, J. Kubsh and R. Brezny, Cars with/without GPF’, SAE Technical Paper 2017- Environ. Sci. Technol., 2014, 48, (10), 6027 01-1004 42. W. Yinhui, Z. Rong, Q. Yanhong, P. Jianfei, 56. J. Andersson, J. Demuynck and H. Hamje, “AECC/ L. Mengren, L. Jianrong, W. Yusheng, H. Min and Concawe 2016 GPF RDE PN Test Programme: S. Shijin, Fuel, 2016, 166, 543 PN Measurement Above and Below 23nm”, 21st 43. K. Aikawa, T. Sakurai and J. J. Jetter, SAE Int. ETH-Conference on Combustion Generated J. Fuels Lubr., 2010, 3, (2), 610 Nanoparticles, Zürich, Switzerland, 19th–22nd 44. G. Karavalakis, ‘Fuel and After-Treatment Effects June, 2017 on Particulate and Toxic Emissions from GDI and 57. V. Y. Prikhodko, J. E. Parks, J. A. Pihl and T. J. PFI Vehicles: A Summary of CE-CERT’s Research’, Toops, SAE Int. J. Engines, 2016, 9, (2), 1289 Workshop on Effects of Fuel Composition on PM, 58. J. Li, N. Currier, A. Yezerets, H.-Y. Chen, H. Hess Chicago, USA, 8th December, 2016 and S. Mulla, SAE Int. J. Engines, 2016, 9, (3), 45. H. Badshah, D. Kittelson and W. Northrop, SAE 1615 Int. J. Engines, 2016, 9, (3), 1775 59. J. Parks, T. Toops, J. Pihl and V. Prikhodko, 46. A. Joshi, D. Bronfenbrenner, C. Tanner, R. ‘Emissions Control for Lean Gasoline Engines’, Ogunwumi, D. Rose, P. Nicolin, B. Coulet and Vehicle Technologies Office Merit Review 2016: T. Boger, ‘High Porosity Substrate and Filter Emissions Control for Lean Gasoline Engines,

Office of Energy Efficiency and Renewable Energy, 65. M. Nicholas, G. Tal and T. Turrentine, ‘Advanced US Department of Energy, Washington DC, USA, Plug-In Electric Vehicle Travel and Charging 9th June, 2016 Behavior’, Advanced Clean Cars Symposium: The 60. H. Nakayama, Y. Kanno, M. Nagata, and X. Zheng, Road Ahead, California Environmental Protection SAE Int. J. Engines, 2016, 9, (4), 2194 Agency, Air Resources Board, Sacramento, USA, 61. J. E. Parks, J. M. E. Storey, V. Y. Prikhodko, M. 27th–28th September, 2016 M. Debusk, and S. A. Lewis, ‘Filter-Based Control 66. C. Favre, ‘Real-Driving Emissions Test Programme of Particulate Matter from a Lean Gasoline Direct Results from a Plug-In Hybrid Electric Vehicle Injection Engine’, SAE Technical Paper 2016-01- (PHEV)’, 13th Integer Emissions Summit and 0937 AdBlue® Forum Europe 2017, Dresden, Germany, 62. G. O. Duarte, R. A. Varella, G. A. Gonçalves and 27th–29th June, 2017 T. L. Farias, J. Power Sources, 2014, 246, 377 67. S. M. Platt, I. El Haddad, S. M. Pieber, A. A. Zardini, 63. H. Yamada, S. Inomata and H. Tanimoto, Emiss. R. Suarez-Bertoa, M. Clairotte, K. R. Daellenbach, Control Sci. Technol., 2017, 3, (2), 135 R.-J. Huang, J. G. Slowik, S. Hellebust, B. Temime- 64. S. Zinola, S. Raux, and M. Leblanc, ‘Persistent Roussel, N. Marchand, J. de Gouw, J. L. Jimenez, Particle Number Emissions Sources at the Tailpipe P. L. Hayes, A. L. Robinson, U. Baltensperger, C. of Combustion Engines’, SAE Technical Paper Astorga and A. S. H. Prévôt, Sci. Rep., 2017, 7, 2016-01-2283 4926

The Author

Dr Ameya Joshi is Director, Emerging Technologies and Regulations at Corning Incorporated, USA. He holds a PhD in Mechanical Engineering from the University of Delaware, USA, and completed his postdoctoral research in Chemical Engineering at the Colorado School of Mines. His technical focus is on reaction engineering as applied to combustion and emissions aftertreatment. His current role is to follow advances in engine technologies and regulations pertinent to vehicular emissions. Previously he has been a Regional Technology Manager responsible for introducing innovative products to customers in Japan and Korea, prior to which he was Research Manager, Modeling & Simulation, in the Reaction Engineering group where he was responsible for developing a suite of modeling tools to predict the in-use performance of cellular ceramic products.

www.technology.matthey.com

Johnson Matthey Highlights A selection of recent publications by Johnson Matthey R&D staff and collaborators

The Influence of Gas Composition on Pd-Based The Effect of Water on Methane Oxidation over Catalyst Activity in Methane Oxidation − Inhibition Pd/Al2O3 under Lean, Stoichiometric and Rich and Promotion by NO Conditions N. Sadokhina, G. Smedler, U. Nylén, M. Olofsson O. Mihai, G. Smedler, U. Nylén, M. Olofsson and L. and L. Olsson, Appl. Catal. B: Environ., 2017, Olsson, Catal. Sci. Technol., 2017, 7, (14), 3084 200, 351 The effect of oxygen concentration and the Methane oxidation on Pd and PtPd-based catalysts presence of water on methane oxidation over a under lean conditions in the presence of either H2O Pd/Al2O3 catalyst was investigated. This catalyst or NO is studied. 5 vol% H2O severely inhibited was characterised by BET, XRD, STEM, TPO the catalytic activity. Dry NO also suppressed and TPR. Various ramping experiments from catalytic activity. NO had a promotional effect on 150–700ºC were carried out using rich, the activity when co-fed with water, compared stoichiometric and lean gas mixtures with and to pure H2O. It is proposed that NO reacts with without water. Raising the oxygen concentration hydroxyl species forming HNO2 which reduced in a dry atmosphere resulted in higher methane the deactivating effects of water present in the oxidation activity which can be linked to the reaction mixture. assistance of palladium oxide formation. Only small amounts of PdO up to 700ºC were Selective Hydrogenation of Acetylene over decomposed as shown by the TPO data but in Cu(211), Ag(211) and Au(211): Horiuti–Polanyi the stoichiometric and rich reaction mixture, Mechanism vs. Non-Horiuti–Polanyi Mechanism PdO was still decomposed due to the oxygen B. Yang, R. Burch, C. Hardacre, P. Hu and P. limitation. Hughes, Catal. Sci. Technol., 2017, 7, (7), 1508 A Parametric Evaluation of Powder Flowability DFT calculations are used to analyse and compare Using a Freeman Rheometer Through Statistical the Horiuti–Polanyi and non-Horiuti–Polanyi and Sensitivity Analysis: A Discrete Element hydrogenation mechanisms for acetylene Method (DEM) Study hydrogenation to ethylene over Cu(211), Ag(211) S. K. Wilkinson, S. A. Turnbull, Z. Yan, E. H. Stitt and Au(211). Hydrogen molecules dissociate first and M. Marigo, Comp. Chem. Eng., 2017, 97, 161 followed by the subsequent addition of hydrogen atoms to the hydrocarbon in the Horiuti–Polanyi A case study simulating a Freeman rheometer to mechanism whereas in the non-Horiuti–Polanyi characterise powder flowability is presented. DEM mechanism, hydrogen molecules directly react was used with input parameters to describe static with the hydrocarbon. The authors found that and rolling coefficients, coefficient of restitution, the Horiuti–Polanyi mechanism is preferred Young’s modulus and cohesion energy density. on Cu(211) for the hydrogenation reactions of DoS principles were used to create a simulation acetylene to ethylene whereas the non-Horiuti– matrix to explore these. Basic flowability energy Polanyi mechanism is preferred for the reactions and specific energy were assessed. Static and over Ag(211). In contrast, the hydrogenation rolling friction were found to play a critical role of C2H2 and C2H3 on Au(211) follows the in determining powder basic flowability energy Horiuti–Polanyi mechanism, while the hydrogenation and specific energy while cohesion energy density of C2H4 follows the non-Horiuti–Polanyi mechanism. affected basic flowability energy.

t Porous Zinc and Cobalt 2-Nitroimidazolate {Pd(µ-Br)(P Bu3)}2. The development of dinuclear t Frameworks with Six-Membered Ring Windows and Pd(I) as opposed to the Pd(0) complex ( Bu3P)2P d a Layered Cobalt 2-Nitroimidazolate Polymorph was demonstrated to be dependent on the A. Orsi, D. J. Price, J. Kahr, R. S. Pillai, S. Sneddon, stoichiometry of Pd to phosphine ligand, the order S. Cao, V. Benoit, M. M. Łozińska, D. B. Cordes, A. M. of adding the reagents and, most importantly, Z. Slawin, P. L. Llewellyn, I. Casely, S. E. Ashbrook, the nature of the Pd precursor and the option of G. Maurin and P. A. Wright, CrystEngComm, 2017, the phosphine ligand used. Mechanistically vital 19, (10), 1377 additional Pd- and phosphine-containing species were identified by experiments on gram scale in Pd. Polymorphs of Zn(2-nIm)2 and Co(2-nIm)2 (2-nIm = 2-nitroimidazole) were prepared by Nanoscale Ion Intermixing Induced Activation of solvothermal synthesis or recrystallisation of Fe2O3/MnO2 Composites for Application in Lithium ZIF-65(Zn/Co). The compounds produced were Ion Batteries isostructural, with a tetrahedrally-connected S. Hao, B. Zhang, J. Feng, Y. Liu, S. Ball, J. Pan, M. framework topology similar to tridymite Srinivasan and Y. Huang, J. Mater. Chem. A, 2017, (lonsdaleite). Single crystal XRD analysis showed 5, (18), 8510 that Zn(2-nIm)2 has rotational disorder for two of the three crystallographically-distinct linker types. A facile method to produce hollow-structured Computation and solid-state NMR spectroscopy oxygen-vacancy-rich Fe2O3/MnO2 nanorods is analysis were carried out. The compounds were demonstrated. The results show that oxygen vacancies are induced by nanoscale ion intermixing tested for their uptake of CO2 and selectivity for CO2 between Fe and Mn ions during annealing. Due over CH4 and N2. to their distinct core-shell hollow nanostructure Effect of Crystallite Size on the Performance and and the presence of oxygen vacancies the

Phase Transformation of Co3O4/Al2O3 Catalysts Fe2O3/MnO2 nanorods display excellent During CO-PrOx – an in situ Study electrochemical performances as anode material for T. M. Nyathi, N. Fischer, A. P. E. York and M. Claeys, lithium ion batteries and a reversible capacity higher Faraday Discuss., 2017, 197, 269 than 700 mA h g–1 after 2000 cycles.

The effect of crystallite size on the mass- and Superoleophobic Surface Modification for Robust surface area-specific CO oxidation activity and the Membrane Distillation Performance reduction behaviour of Co3O4 were studied. The N. G. P. Chew, S. Zhao, C. Malde and R. Wang, J. reverse micelle technique was used to synthesise Membrane Sci., 2017, 541, 162 model Co3O4 catalysts with average crystallite sizes between 3 and 15 nm. During the catalytic Robust membranes with anti-fouling and tests, it was found that reducing the size of Co3O4 anti-wetting qualities are being investigated for crystallites raised the mass-specific CO oxidation produced water treatment from oil and gas industry activity between 50–200ºC. Moreover, in the same by direct-contact membrane distillation (DCMD). temperature range the surface area-specific CO In this work, a composite Janus membrane oxidation activity demonstrated a volcano-type was prepared by single-step co-deposition of behaviour where crystallites with an average size polydopamine (PDA)/polyethylenimine (PEI) onto of 8.5 nm were the most active. The reduction of the outer surface of a commercial hydrophobic

Co3O4 was examined in situ by a magnetometer and polyvinylidene fluoride (PVDF) substrate. Its a PXRD capillary cell while recording kinetic data. performance was tested by feeding a series of low surface tension solutions. This modified PVDF Understanding the Unusual Reduction Mechanism membrane, which was inspired by mussel adhesive, of Pd(II) to Pd(I): Uncovering Hidden Species and could potentially be used for long-term water Implications in Catalytic Cross-Coupling Reactions recovery from produced water via DCMD. C. C. C. Johansson Seechurn, T. Sperger, T. G. Scrase, F. Schoenebeck and T. J. Colacot, J. Am. Influence of Sb on the Structure and Performance of Chem. Soc., 2017, 139, (14), 5194 Pd-Based Catalysts: An X-ray Spectroscopic Study S. Gatla, O. Mathon, A. Rogalev, S. Pascarelli, J. In a large number of Pd-catalysed processes the Radnik, M.-M. Pohl and A. Brückner, J. Phys. Chem. reduction of Pd(II) intermediates to Pd(0) is a vital t C, 2017, 121, (7), 3854 elementary step. In the case of P Bu3, which is one of the most powerful new generation phosphine XPS and XAFS investigations were carried out on ligands, oxidation state Pd(I) and not Pd(0), is 10 wt% Pd–16 wt% Sb/TiO2 catalyst for gas-phase generated upon reduction from Pd(II). Experimental acetoxylation of toluene to benzyl acetate. The impact and computational studies were used to evaluate of the co-component Sb on the active Pd species the mechanism of the reduction of Pd(II) to Pd(I) was assessed. Excess electron charge was found on for the emergence of the highly active precatalyst metallic Pd species after several hours on stream.

This phenomenon may be due to electron transfer 5–6 were present in the most active catalysts and an from metallic Sb. TEM-EDX analysis confirmed that atomic ratio of 3 in deactivated samples. intermixed Pd–Sb particles with an atomic ratio of Synthesis and Characterization of Boron Carbon Oxynitride Films with Tunable Composition using Catalysis Valence electrons Methane, Boric Acid and Ammonia

100 d+ Amount of Pd 4d electrons B. J. Matsoso, K. Ranganathan, B. K. Mutuma, T. Lerotholi, G. Jones and N. J. Coville, New J. Chem., 80 2017, 41, (17), 9497

60 2D boron carbon oxynitride (BCNO) films were synthesised by atmospheric pressure chemical 40 vapour deposition (APCVD) using carbon-rich methane, nitrogen-rich ammonia, boron- and 20 oxygen-rich boric acid as precursors. Various Sb Pd – atomic compositions were accomplished by altering

Conversion selectivity, % selectivity, Conversion e 0 d– the vapourisation temperature of boric acid by Sb:Pd ratio in active particles changing the distance (i.e. 2 cm to 12 cm) between Time on stream boric acid and the growth substrate. The XPS survey spectra showed that the atomic compositions of the Reprinted with permission from S. Gatla, O. Mathon, BCNO films formed differ as follows: C 48–71 at%, A. Rogalev, S. Pascarelli, J. Radnik, M.-M. Pohl and A. B 2.34–12.8 at%, N 1.98–7.9 at% and O Brückner, J. Phys. Chem. C, 2017, 121, (7), 3854. 33–34 at%. The films also indicated vibrational Copyright 2017 American Chemical Society modes from h-BN, B–C and graphene domains from Raman spectra.

www.technology.matthey.com

Influence of Three-Way Catalyst on Gaseous and Particulate Matter Emissions During Gasoline Direct Injection Engine Cold-start Analysing emissions to meet Euro 6c legislation

Maria Bogarra, Jose Martin engine cold-start. Large concentrations of carbon Herreros, Cruz Hergueta, monoxide, propane, acetaldehyde, formaldehyde, Athanasios Tsolakis* ethanol, toluene and ethylene were emitted Mechanical Engineering, University of during the first 70–90 seconds from the engine Birmingham, Edgbaston, B15 2TT, UK start. Gaseous emissions were reduced on the catalyst at temperatures higher than 290°C, *Email: [email protected] with the catalyst reaching almost 100% removal efficiency at 350°C. The effect of the TWC on PM Andrew P. E. York and Paul J. emissions has been analysed for the different PM Millington diameter ranges. A reduction of particles smaller Johnson Matthey, Blount’s Court, Sonning than 20 nm was observed as well as a reduction Common, Reading, RG4 9NH, UK in the accumulation mode. In order to understand the nature of the particles emitted during cold-start, transmission electron microscope The development of gasoline direct injection (TEM) grids were used for particulate collection (GDI) engines has provided a strong alternative to at the engine start and after 80 seconds and port fuel injection engines as they offer increased 140 seconds of engine operation. A peak of 8 power output and better fuel economy and carbon 1.4 × 10 particles was produced at the engine dioxide emissions. However, particulate matter start and this steadily reduced to 3 × 107 in (PM) emission reduction from GDI still remains 50 seconds. The TEM micrographs showed solid a challenge that needs to be addressed in order particles with similar fractal-like shapes. to fulfil the increasingly stricter environmental regulations. A large number of the total 1. Introduction particulate emissions during driving cycles are produced during the engine cold-start. Therefore, The fleet of GDI engines will continuously controlling PM during cold-start events will increase for years to come due to their significant significantly reduce the final PM output. advantages in terms of fuel consumption and

This research work provides an understanding CO2 reduction when compared to older port fuel of PM characterisation from a 2 l four-cylinder injection engines, and the negative publicity GDI engine during cold-start. Gaseous emissions diesel vehicles are receiving (1–3). However, GDI including hydrocarbon (HC) speciation studies are engines have been associated with an increase also carried out pre- and post- a Euro 6 compliant in PM levels when compared to earlier gasoline three-way catalyst (TWC). In addition, particulate powertrains. Euro 6c emission legislation limits size distribution and total particulate number the number to 6 × 1011 particles km–1 and were recorded for the first 280 seconds after the came into force in September 2017 (4). The

329 © 2017 Johnson Matthey http://dx.doi.org/10.1595/205651317X696315 Johnson Matthey Technol. Rev., 2017, 61, (4) main contributor to the total particulate number is reduce the aggressive spikes of the HCs and small particles (i.e. those formed in the nucleation NOx produced during the cold-start even if the mode), which are more hazardous to human catalyst temperature was still low, although the CO health than larger particles as they are able to reduction was less noticeable. The TWC reached penetrate deeper into the human respiratory 300°C after 21 seconds of operation. The same system (5). Due to the difficulties when measuring authors in a different study compared the raw the nucleation mode, the Particle Measurement engine emissions during cold-start with the results Programme (PMP) established the cut-off to be 23 obtained using a PMP compliant dilution tunnel nm particles in order to assure the repeatability (cut-off of 23 nm) (15). The difference between of the results (6). However, sub-23 nm particles raw and PMP results are more than one order of have been reported during GDI normal operation magnitude. Rich conditions during cold-start led (7, 8) and particles around 6 nm diameter have to several peaks in PM number. The presence of been found using TEM (9). The high variability of two differentiated PM modes, nucleation and cold-start events, as well as the low temperature accumulation, during the cold-start suggested the in the exhaust and TWC, hinders the measurement coexistence of particles formed under different of PM. In addition, during the engine cold-start conditions in the combustion chamber. Samuel et a large amount of fuel is being injected and this al. (16) analysed the performance of the TWC on increases the emission of larger particles (10). PM reduction during cold-start. According to their During urban journeys, the engine is subjected to results, particles between 5 to 25 nm represented several cold-starts. Although the engine can be 99% of the total emissions until the engine was warmed, the low exhaust temperature reduces the warmed-up. The TWC was capable of reducing effectiveness of the aftertreatment system as well the number of particles emitted, especially those as increasing the likelihood of nucleation and the ranging between 5–25 nm and 50–100 nm. The formation of small particles in the exhaust. Engine effect of the TWC on particulate size distribution cold-starts represent 50% of urban driving emissions at steady-state condition has been analysed (17). and contribute to 80% of the total emissions for A reduction in the particle concentration around some species such as volatile organic compounds 5–50 nm after the catalytic converter was reported. (11). The emission of unburnt HC during engine However, particles above 100 nm were increased. cold-start is also an area of concern. Amongst the The authors claim that the TWC is able to remove species found during the engine start is methane, and oxidise some of the PM, but particles may a potent greenhouse gas with radiation trapping coagulate in the catalyst increasing the diameter efficiency of approximately 25 times higher than of particles post-TWC. Soot oxidation behaviour,

CO2 (12), while polyaromatic HC such as benzene nanostructure and Raman analysis during and toluene are carcinogenic to humans (13). This cold-start was analysed (18). The authors is particularly worrying as in urban areas human reported higher reactivity of the particles collected exposure is high. Cold-starts are estimated to last during the engine cold-start with respect to hot for 120 seconds, which is equivalent to a one-mile steady-state soot samples. This effect was attributed journey (11). Although the TWC is an efficient to the high percentage of unburned ash precursors way of removing CO, unburned HCs and nitrogen found in cold-start soot samples. The nanostructure oxides (NOx) from the exhaust, during cold-starts and Raman analysis also showed that during cold- its temperature is far from the ideal operating start the soot had an ordered structure similar to conditions, reducing the TWC’s efficiency. In warm steady-state conditions meaning that the addition, as PM control is not the main objective of carbon crystalline structures during cold-start are the TWC, its effect on PM during cold-starts is not similar to those during steady-state operation. yet well documented. In this research, gaseous emissions including HC Peckham et al. (14) studied the legislated speciation and PM characterisation during engine emissions using fast-response analysers during cold-start have been measured on an air-guided the first 100 seconds of the engine cold-start. 2 l four-cylinder GDI engine equipped with a Approximately 0.31 g of HC, 0.08 g of NOx Euro 6 compliant TWC. In addition, total particulate and almost 2 g of CO were emitted during this number was recorded in the first 280 seconds of period. The main contribution to the cumulative the engine cold-start. The particle size distribution tailpipe emissions were the first 20 seconds of the and the fraction of particles per diameter have been cold-start, during which several particulate number analysed at different stages during the engine start spikes were emitted. The TWC was reported to (cold-start and after 80 and 140 seconds of engine

330 © 2017 Johnson Matthey http://dx.doi.org/10.1595/205651317X696315 Johnson Matthey Technol. Rev., 2017, 61, (4) operation). Additional PM characterisation studies Table II Gasoline Properties were carried out using TEM. PM was collected from the engine exhaust gas at different time intervals Analysis (test method) Result using copper grids and analysed by TEM in order to Density at 15°C, kg m–3 743.9 understand qualitatively their morphology. Initial boiling point, °C 34.6

20% v/v, °C 55.8 2. Experimental Setup and Method 50% v/v, °C 94.0 2.1 Experimental Setup Final boiling point, °C 186.3

The engine used for this study was a 2 l, C m/m, % 84.16 four-cylinder, air-guided stoichiometric GDI. The H m/m, % 13.48 details of the engine specification can be found in Table I. Standard EN228 gasoline with 5% (v/v) O m/m, % 2.36 ethanol content provided by Shell was used for Paraffins, vol% 43.9 this research. Fuel properties are presented in Olefins, vol% 11.7 Table II. 5W30 fully synthetic lubricating oil was employed. Naphthenes, vol% 7.8 Gaseous emissions were measured using a Aromatics, vol% 26.9 Fourier-transform infrared spectroscopy (FTIR) Oxygenates, vol% 7.7 2100 MKS. Legislated emissions (CO, CO2, NOx and total HC (THC)) as well as HC speciation were Sulfur, ppm 6 sampled before and after the TWC during the Calorific values, MJ kg–1 42.22 cold-start. The sample was filtered to avoid any damage of the optical part by PM and pumped at Motor octane number 85.3 –1 a rate of 1 l min through the equipment. The Research octane number 96.5 heating lines and the pump were maintained at 191°C to eliminate any condensation of HCs and water in the pipes. 2.2 Test Procedure and Method Total particulate number and size distributions were obtained using a Dekati Electrical Low All the experiments were carried out at cold-start Pressure Impactor (ELPI®+). An ejector diluter conditions. At least 24 hours were left between system was used to precondition the sample. The tests to soak the engine. The engine warm-up dilution ratio (DR) was set at 10 and recorded process was established by the test control system throughout the test based on nitric oxide (NO) (CADET) following standard vehicle operation, and CO2 concentration on the exhaust and diluted the torque and speed trace during this process streams, Equation (i), using the FTIR. The air is provided in Figure 1. In addition, to further dilution temperature was ambient. understand the behaviour of gaseous and PM emissions, the lambda trace during this period ppmNOraw exhaust %CO2 raw exhaust DR = ≈ (i) is also shown in Figure 2. The engine starts at

ppmNOdilute sample %CO2 dilute sample 20 seconds, a sudden increase to 1200 rpm and 5 Nm is produced and the overall lambda is rich at this point to start the combustion. High fluctuations in lambda are produced to stabilise the engine. Table I Engine Specifications From this point, different changes in speed and Compression ratio 10:1 torque are programmed to warm-up the engine Bore × stroke 87.5 × 83.1 mm leading to changes in lambda that affect gaseous and PM emissions. At the end of the sequence, Turbocharger Borg Warner K03 lambda tends to one. Rated power 149 kW at 6000 rpm Gaseous and PM emissions were measured during the first 280 seconds after the engine start pre- and Rated torque 300 Nm at 1750–4500 rpm post-TWC. The location of the ELPI®+ and the FTIR Engine Bosch Me17 sampling point pre- and post-TWC was swapped management in each test until a total of five measurements

2500 70 Speed Torque 60 2000

50 Torque, Nm 1500 40

1000 30

Speed, rpm 20 500 10

0 0 0 50 100 150 200 250 300 Time, s

Fig. 1. Torque and speed trace during engine cold-start

1.6 1.4 1.2 1.0 0.8 0.6 Lambda 0.4 0.2 0 0 50 100 150 200 250 300 Time, s Fig. 2. Lambda trace during engine cold-start before and after TWC were obtained for statistical 3. Results and Discussion analysis. For morphology analysis 3.05 mm, TAAB 3.1 Gaseous Emissions Formvar coated copper grids were loaded directly from the exhaust pipe at the engine start and after 3.1.1 Regulated Emissions 80 seconds and 140 seconds of engine operation. A schematic of the experimental setup is provided The regulated gaseous emissions measured in in Figure 3. the engine exhaust before and after the TWC in the first 280 seconds are presented in Figure 4. No apparent changes in the gas composition pre- and post-catalyst can be observed in the first 150 seconds. A peak of CO of 10,000 ppm is produced just after the engine start, corresponding

Intake Exhaust TWC to overall rich lambda conditions and this is rapidly decreased to around 1300 ppm. At 235 seconds, engine acceleration leads to an increase in CO. Alternate days On the other hand, HC emissions remained Direct exposure stable at 4000 ppm during this cold-start period, of TEM grids therefore HC emissions are not influenced by the engine conditions. No oxidation of CO and THC was observed during the first 120 seconds of the engine operation. At this point the exhaust ELPI®+ FTIR pre-TWC reached temperatures higher than Fig. 3. Experimental setup for measuring 300°C and the oxidation of HC and CO started. NO emissions during engine cold-start in the present reduction did not start until the TWC temperature study exceeded 300°C.

CO pre-TWC CO post-TWC pre-TWC temperature, ºC (a) 10,000 500 Temperature, ºC 8000 400 6000 300 4000 200 2000 100 0 0 Concentration, ppm Concentration, 0 50 100 150 200 250 300 Time, s

NO pre-TWC NO post-TWC pre-TWC temperature, ºC (b) 2000 500 Temperature, ºC 1500 400 300 1000 200 500 100 0 Concentration, ppm Concentration, 0 0 50 100 150 200 250 300 Time, s

THC pre-TWC THC post-TWC pre-TWC temperature, ºC (c) 10,000 500 Temperature, ºC 8000 400 6000 300 4000 200 2000 100 0 0 Concentration, ppm Concentration, 0 50 100 150 200 250 300 Time, s

Fig. 4. Emissions during cold-start: (a) CO; (b) NOx; (c) THC

3.1.2 Hydrocarbon Speciation other hand, unsaturated species are beneficial for NO reduction (19). During the cold-start analysed The different HC species measured by the FTIR are in this work, the most abundant species are light divided in three groups: HC such as propane, acetaldehyde, formaldehyde, (a) linear HC species (Figure 5(a)) acetylene and ethylene. The TWC is not able to (b) oxygenated HCs (Figure 5(b)) effectively remove these species during the first (c) unsaturated HCs (Figure 5(c)). 200 seconds due to the low temperature. From The reactivity of individual HC species is affected by this point, the catalyst is closer to its light-off the exhaust composition. In addition, CO oxidation temperature, reducing all HC species, and reached and NO reduction are affected by the different 100% conversion at approximately 230 seconds HC present in the exhaust and the intermediate after engine start. The HC reactivity follows species formed on the catalyst during HC oxidation, the order reported in the literature: alcohols > such as ethoxide, acetate, formate and benzoate aldehydes > aromatics > alkenes > alkanes (21). (19). For instance the higher reactivity of the HC Linear HCs: more than 600 ppm of propane were will lead to an increased inhibition of CO oxidation observed pre-TWC, Figure 5(a). The concentration due to the competition for active sites (20). On the after TWC fluctuated before the catalyst temperature

(a) CH4 pre-TWC CH4 post-TWC Propane pre-TWC Propane post-TWC Dodecane pre-TWC Dodecane post-TWC Dodecane concentration, ppm 1000 200

800 160

600 120

400 80

200 40 Concentration, ppm Concentration,

0 0 0 50 100 150 200 250 300 Time, s

(b) Acetaldehyde pre-TWC Acetaldehyde post-TWC Formaldehyde pre-TWC Formaldehyde post-TWC Ethanol pre-TWC Ethanol post-TWC

140 120 100 80 60 40

Concentration, ppm Concentration, 20 0 0 50 100 150 200 250 300 Time, s

(c) Acetaldehyde pre-TWC Acetaldehyde post-TWC Ethylene pre-TWC Ethylene post-TWC Propylene pre-TWC Propylene post-TWC Toluene pre-TWC Toluene post-TWC 250

200

150

100

50 Concentration, ppm Concentration, 0 0 50 100 150 200 250 300 Time, s Fig. 5. HC speciation during cold-start: pre-TWC and post-TWC: (a) linear HCs; (b) oxygenated HCs; (c) unsaturated HCs reached 300°C. At this point the propane the engine start but as the engine warmed up concentration decreased steadily until reaching the engine output concentration was then rapidly negligible engine-out concentration at 420°C, reduced and stabilised at around 50 ppm. The pre- 250 seconds after the engine start. The second and post-TWC methane concentration was similar highest engine output emission concentration for the entire study and no oxidation activity was was methane: 220 ppm were produced just after observed. Methane emissions are a major concern

334 © 2017 Johnson Matthey http://dx.doi.org/10.1595/205651317X696315 Johnson Matthey Technol. Rev., 2017, 61, (4) due to its global warming potential, its link with molecule’s functional group with the surface of ozone formation (12) and the difficulty of oxidising the catalyst is the reason for the higher oxidation methane in the TWC (22). The oxidation of the activity of ethanol (19). approximately 20 ppm of dodecane was observed Unsaturated HCs: toluene and ethylene are the earlier than propane. The longer the alkane chain, two main unsaturated HC species emitted from the lower the light-off temperature required (19). the engine, Figure 5(c). At the engine start, a Oxygenated HCs: the largest concentration peak of 160 ppm of ethylene was produced. After of measured oxygenated species emitted was 100 seconds of engine operation the ethylene acetaldehyde, followed by ethanol and then concentration started declining steadily, reaching formaldehyde, Figure 5(b). The formation of close to 0 ppm after 230 seconds. Ethylene and acetaldehyde and formaldehyde, unregulated propylene are more reactive than alkane species compounds classified as carcinogenic to humans, and adsorb more easily on the catalyst surface has been reported during the incomplete leading to earlier light-off (20). The toluene peak combustion of ethanol blends (23, 24). More was delayed with respect to the other species than 100 ppm of acetaldehyde were produced and a maximum concentration of 250 ppm was during the engine start, and the start of oxidation observed after 70 seconds from the engine was observed at a temperature of 280°C, earlier start. At this point and similarly to ethylene’s than NO, CO and the rest of the HC species. The behaviour, the concentration decreased almost maximum concentration of formaldehyde emissions linearly. Acetylene peaked at 125 ppm at the during the first 70 seconds of the engine operation cold-start and showed stronger adsorption was around 60 ppm. The TWC seemed to store behaviour than ethylene or propylene; this can part of the acetaldehyde and formaldehyde until lead to CO oxidation inhibition (20). the catalyst was close to its light-off temperature (350°C). The condensation of HC on the TWC 3.2 Particulate Matter during cold-start due to the cold catalyst surface is linked with a delay in CO conversion (20). The 3.2.1 Particle Number During Cold- light-off activity for formaldehyde is similar to that start of acetaldehyde. Aldehydes have been reported to be more reactive than aromatics and alkanes The evolution of PM during the first few seconds (21). However, ethanol emissions reached 80 ppm of the engine cold-start is presented in Figure 6, and started decreasing at 300°C. Ethanol light- note that scales pre- and post-TWC are different. off temperature has been reported to be lower A significant peak of particles, 1.3 × 108 particles than that of CO, despite the higher C–H bonding cm–3, was initially recorded before its rapid energy when compared to propane or toluene. The reduction during the first 70 seconds. These high dipole-dipole interaction of the polar ethanol PM levels correspond to overall rich lambdas. As

1.4 × 108 1.4 × 107

1.2 × 108 1.2 × 107 PN (post-TWC), cm –3 1.0 × 108 1.0 × 107

8.0 × 107 8.0 × 106

6.0 × 107 6.0 × 106

4.0 × 107 4.0 × 106 –3 PN (pre-TWC), cm PN (pre-TWC), 2.0 × 107 2.0 × 106

0.0 0.0 0 50 100 150 200 250 300 Time, s

Fig. 6. Particle number emissions during engine cold-start. Effect of the TWC. Note that scales pre- and post-TWC are different

335 © 2017 Johnson Matthey http://dx.doi.org/10.1595/205651317X696315 Johnson Matthey Technol. Rev., 2017, 61, (4) in the case of CO engine output emissions, at this similar to GDI engine steady-state operation or stage PM levels are highly influenced by lambda even similar to diesel. The majority of the particles fluctuations. The TWC is able to considerably reduce collected in the grid pre-TWC for the TEM analysis the amount of PM emitted even though at this early were solid, a result that has previously been stage it is far from its light-off temperature. Particle observed (15). Different types of particles were losses in catalysts have been mainly attributed to found pre-TWC, fractal-like agglomerates similar particle diffusion and thermophoresis (25). Particle to diesel particles (Figure 8(b) or Figures 7(b) removal by diffusion is significant for particles and 7(c)) and slurry-like particles (Figure 8(c)). below 100 nm. At 10 nm 75% of the particles Furthermore, the presence of small spherules have been reported to be reduced in a catalyst by (Figure 7(c)) was also observed. The presence of diffusion (25). In addition thermophoretic losses different types of PM has been previously reported during transient operation can be around 20% in the literature (9, 28). After 80 seconds, the (26). These results suggest that PM formed by concentration of particles found in the grid heavy HCs are condensed in the TWC on account dropped significantly. At this point the engine of the temperature conditions and can either (a) decelerated, leading to an increase in the air be oxidised or (b) blow-off the TWC when the fraction. Lambda was still rich but approximating temperature reaches the light-off conditions. In to stoichiometric conditions; still the four addition, FTIR is only able to measure some light types of PM aforementioned were observed. At HC species in the gas-phase suggesting that PM are 140 seconds after the engine start, only a few composed by a soot core and heavier polyaromatic solid particles were found under the TEM beam. HCs. The need for enrichment to assure the Although the engine accelerated, the air-fuel engine cold-start, worsened by piston and wall- mixture may be already homogeneous reducing impingement of the fuel, leads to the formation of the overall formation of particles. It is well-known fuel-rich areas promoting PM formation (27). that the main contributor to PM emissions in GDI engines is cold-start and transient operation. It 3.2.2 Particulate Size Distributions has been suggested that the nature of this PM was volatile HC; however, the images clearly show the and Diameter Fractions During Cold- presence of soot at this early stage of the engine Start operation in agreement with where solid particles The particulate size distributions before and after have previously been found (18). the TWC at the engine start and after 80 seconds Figure 10 presents the percentage of particles in and 140 seconds of engine operation are presented each of the measured diameter ranges before and in Figures 7–9, note that scales pre- and post- after the TWC. Sub-23 nm particles account for up TWC are different. TEM images before the TWC to 67% of total PM emitted. The TWC considerably have been included as well for qualitative analysis reduced the amount of particles between 5–10 nm, of soot pre-TWC. All the distributions showed a however, up to 40% of sub-23 nm particles were bimodal shape with nucleation particles smaller than observed at 140 seconds after engine start 20 nm and an accumulation mode centred between (Figure 10(b)). The origin of sub-23 nm particles 50–100 nm. At the engine start the accumulation in GDI engines is thought to be metals from the mode measured pre-TWC was centred at 70 nm. lubricant oil or from the fuel additives. The percentage The catalyst was capable of storing and trapping of solid sub-23 nm is typically below 60%, similar to small particles and HC droplets as well as reducing the results obtained in this work, slightly exceeding the accumulation mode, leading to a shift in the the sub-23 nm fraction reported for diesel engines peak to 30 nm. On the other hand, after 80 and (29). The results show that the TWC itself can have 140 seconds of engine operation, the TWC was a reducing effect on the PM, but that an additional capable of eliminating particles smaller than 20 nm catalyst system will be required for full PM control; and larger than 50 nm without affecting particles this will be in the form of a particulate filter. in the range 20–50 nm. This lack of effect between 20–50 nm has been previously reported in the 4. Conclusions literature (16). Larger particles (accumulation peak centred in 120 nm) were observed pre-TWC. Gaseous emissions, HC species and PM have At first glance, the TEM grids corresponding to the been analysed during cold-start engine conditions first stage of the cold-start were heavily loaded, for the first 280 seconds after engine start. The (Figure 7(b)) and had a fractal-like appearance performance of a Euro 6c compliant TWC in gaseous

(a) pre-TWC post-TWC dN 8 7 –3 1.5 × 10

2.5 × 10 / dlogD 2.0 × 107 8 1.0 × 10 p 1.5 × 107 (post-TWC), cm

7 7 1.0 × 10

(pre-TWC), cm (pre-TWC), 5.0 × 10 p 5.0 × 106 dlogD / 0.0 0.0 –3

dN 5 50 500 Aerodynamic diameter, nm

(b) (c)

500 nm 200 nm

Fig. 7. Engine start: (a) particulate size distribution (PSD); (b) TEM micrograph; (c) TEM micrograph showing particles with a different nature. Note that scales pre- and post-TWC are different

(a) pre-TWC post-TWC dN –3

6 6 / 2.0 × 10 1.0 × 10 dlogD

8.0 × 105 p (post-TWC), cm 6.0 × 105 1.0 × 106 4.0 × 105 (pre-TWC), cm (pre-TWC), p 2.0 × 105

dlogD 0.0

/ 0.0 –3

dN 5 50 500 Aerodynamic diameter, nm

(b) (c)

100 nm 100 nm

Fig. 8. 80 seconds of engine operation: (a) PSD; (b) TEM micrograph; (c) TEM micrograph showing particles with a different nature. Note that scales pre- and post-TWC are different

(a)

pre-TWC post-TWC dN –3

6 5 / 1.5 × 10 8.0 × 10 dlogD

6.0 × 10 p 1.0 × 106 (post-TWC), cm 4.0 × 105 5 (pre-TWC), cm (pre-TWC), 5.0 × 10 p 2.0 × 105 dlogD

/ 0.0 0.0 –3

dN 5 50 500 Aerodynamic diameter, nm

(b) (c)

100 nm 100 nm

Fig. 9. 140 seconds of engine operation: (a) PSD; (b) TEM micrograph; (c) TEM micrograph showing particles with a different nature. Note that scales pre- and post-TWC are different

(a) 100

80 100–1000 60 50–100 23–50 40 10–23 20 5–10 Number fraction, % Number fraction, 0 Engine start + 80 s + 140 s

(b) 100

80 100–1000 60 50–100 23–50 40 10–23 20 5–10 Number fraction, % Number fraction, 0 Engine start + 80 s + 140 s

Fig. 10. Percentage fraction of particles per diameter: (a) pre-TWC; (b) post-TWC

338 © 2017 Johnson Matthey http://dx.doi.org/10.1595/205651317X696315 Johnson Matthey Technol. Rev., 2017, 61, (4) emissions removal during this period has been Fund as part of the Science City Research Alliance assessed as well as its effect on PM emissions. Energy Efficiency Project are also acknowledged The TWC effectively reduced NO, CO and THC for supporting the research work. Louie Chen from for temperatures above 300°C. However, before Scielutions Ltd, UK, and Dekati Ltd, Finland, are the catalyst light-off, large quantities of methane, also acknowledged for the use of the ELPI®+. propane, toluene, ethanol, acetaldehyde and formaldehyde were emitted. The HC reactivity References followed the order: alcohols > aldehydes > aromatics > alkenes > alkanes during the period 1. “Advanced Direct Injection Combustion Engine analysed. Technologies and Development: Gasoline and Gas Total particle number emissions peaked at Engines”, 1st Edn., ed. H. Zhao, Vol. 1, Woodhead 1.4 × 108 particles cm–3 at the engine start and Publishing Ltd, Cambridge, UK, 2010, pp. 1–19 dropped an order of magnitude after 140 seconds of 2. V. Franco, F. P. Sánchez, J. German and P. Mock, the engine warming up. A significant concentration of “Real-World Exhaust Emissions from Modern sub-23 nm was observed pre-TWC. HC and particle Diesel Cars: A Meta-Analysis of PEMS Emissions Data from EU (Euro 6) and US (Tier 2 Bin 5/ULEV deposition on the TWC have been observed during II) Diesel Passenger Cars: Part 1: Aggregated the cold-start which can delay CO oxidation. Particle Results”, The International Council of Clean size distributions showed a bimodal distribution Transportation, Berlin, Germany, 2014 along the 280 seconds of the analysis. The TWC 3. “European Vehicle Market Statistics: Pocketbook was able to reduce the number of particles between 2016/17”, ed. P. Mock, The International Council 5–10 nm, but still a significant concentration of of Clean Transportation, Berlin, Germany, 2016 particles between 10–23 nm was observed. The 4. Commission Regulation (EU) 459/2012, Official J. particles observed during the engine cold-start are Eur. Union, 2012, 55, (L142), 16 solid and fractal-like similar to diesel PM. 5. K.-H. Kim, E. Kabir and S. Kabir, Environ. Int., 2015, 74, 136 Acknowledgments 6. B. Giechaskiel, R. Chirico, P. F. DeCarlo, M. Clairotte, T. Adam, G. Martini, M. F. Heringa, R. The authors would like to thank EPSRC and Richter, A. S. H. Prevot, U. Baltensperger and C. Johnson Matthey for funding the project and Astorga, Sci. Total Environ., 2010, 408, (21), providing a scholarship to Maria Bogarra. Innovate 5106 UK (Technology Strategy Board) is acknowledged 7. I. Khalek, T. Bougher and J. Jetter, SAE Int. J. for supporting this work with the project “CO 2 Fuels Lubr., 2010, 3, (2), 623 Reduction through Emissions Optimisation” 8. T. Rönkkö, L. Pirjola, L. Ntziachristos, J. Heikkilä, P. (CREO: ref. 400176/149) in collaboration with Karjalainen, R. Hillamo and J. Keskinen, Environ. Ford Motor Company, UK; Jaguar Land Rover, Sci. Technol., 2014, 48, (3), 2043 UK, and Cambustion, UK. The Advantage West 9. T. L. Barone, J. M. E. Storey, A. D. Youngquist and Midlands and the European Regional Development J. P. Szybist, Atmos. Environ., 2012, 49, 268 10. C. K. Gaddam and R. L. Vander Wal, Combust. Glossary Terms Flame, 2013, 160, (11), 2517 CO carbon monoxide 11. M. S. Reiter and K. M. Kockelman, Transport. Res. CO2 carbon dioxide D: Transport Environ., 2016, 43, 123 DR dilution ratio 12. “Inventory of U.S. Greenhouse Gas Emissions ELPI®+ electrical low pressure impactor and Sinks: 1990–2015”, EPA 430‑P‑17‑001, US FTIR Fourier transform infrared Environmental Protection Agency, Washington, GDI gasoline direct injection engine DC, USA, 13th April, 2017, 633 pp HC hydrocarbon NOx nitrogen oxides 13. A. J. McMichael, ‘Carcinogenicity of Benzene, PM particulate matter Toluene and Xylene: Epidemiological and PMP particle measurement programme Experimental Evidence’ in “Environmental PSD particulate size distribution Carcinogens: Methods of Analysis and Exposure SMPS scanning mobility particulate sizer Measurement: Benzene and Alkylated Benzenes”, TEM transmission electron microscope eds. L. Fishbein and I. K. O’Neill, IARC Scientific TWC three-way catalyst Publications No. 85, Vol. 10, International Agency VOC volatile organic compound for Research on Cancer, Lyon, France, 1988, pp. 3–18

14. M. S. Peckham, A. Finch and B. Campbell, SAE Int. 22. G. Beulertz, M. Votsmeier and R. Moos, Appl. J. Engines, 2011, 4, (1), 1513 Catal. B: Environ., 2015, 165, 369 15. M. S. Peckham, A. Finch, B. Campbell, P. Price and 23. R. Suarez-Bertoa, A. A. Zardini, H. Keuken and C. M. T. Davies, ‘Study of Particle Number Emissions Astorga, Fuel, 2015, 143, 173 from a Turbocharged Gasoline Direct Injection 24. S. Paz-Estivill, R. Delgado-Ortiz, E. Cirera- (GDI) Engine Including Data from a Fast-Response Domènech and F. Broto-Puig, ‘Vehicle Exhaust Particle Size Spectrometer’, SAE Technical Paper Emissions Characterization by Chromatographic 2011-01-1224, 2011 Techniques Applied to Different Gasoline-Ethanol 16. S. Samuel, A. Hassaneen and D. Morrey, Blends’, SAE Technical Paper 2013-01-1044, ‘Particulate Matter Emissions and the Role of 2013 Catalytic Converter During Cold Start of GDI 25. J. Swanson, W. Watts, D. Kittelson, R. Newman Engine’, SAE Technical Paper 2010-01-2122, 2010 and R. Ziebarth, Aerosol Sci. Technol., 2013, 47, 17. I. Whelan, S. Samuel and A. Hassaneen, (4), 452 ‘Investigation into the Role of Catalytic Converters on Tailpipe-Out Nano-Scale Particulate Matter from 26. J. E. Johnson and D. B. Kittelson, Appl. Catal. B: Gasoline Direct Injection Engine’, SAE Technical Environ., 1996, 10, (1–3), 117 Paper 2010-01-1572, 5th May, 2010 27. H. Badshah, D. Kittelson and W. Northrop, SAE 18. S. Choi and H. Seong, Combust. Flame, 2015, Int. J. Engines, 2016, 9, (3), 1775 162, (6), 2371 28. P. Karjalainen, L. Pirjola, J. Heikkilä, T. Lähde, T. 19. S. B. Kang, S. B. Nam, B. K. Cho, I.-S. Nam, C. H. Tzamkiozis, L. Ntziachristos, J. Keskinen and T. Kim and S. H. Oh, Catal. Today, 2014, 231, 3 Rönkkö, Atmos. Environ., 2014, 97, 262 20. T. Bäroth, A. Drochner, H. Vogel and M. Votsmeier, 29. B. Giechaskiel and G. Martini, ‘PMP: SUB 23 NM Top. Catal., 2017, 60, (3–5), 278 Review’, UNECE, Particle Measurement Programme 21. A. O. Rusu and E. Dumitriu, Environ. Eng. Manage. (PMP), 28th Session, Brussels, Belgium, 21st J., 2003, 2, (4), 273 November, 2013

The Authors

Maria Bogarra is a Research Fellow working on a Horizon 2020 funded project at the School of Engineering, University of Birmingham, UK. She obtained her PhD degree in Mechanical Engineering entitled “Characterisation of Particulate Matter Emitted by Gasoline Direct Injection Engines” at the University of Birmingham in March 2017. She has worked in the physical-chemical characterisation of alternative fuels, the performance of aftertreatment systems and alternative fuels for both diesel and gasoline powertrains and on the development of methods for the characterisation of particulate matter emitted by gasoline direct injection engines.

Dr Jose M. Herreros is a Lecturer in Vehicle Engineering at the School of Engineering at the University of Birmingham. His research focuses on the investigation of clean and efficient powertrain systems based on the energy and emissions efficient integration of various propulsion systems with the ultimate goal to develop energy-efficient and clean powertrains to be used in vehicular applications. He has published journal articles on issues related to fuel design and properties, pollutant emissions characterisation and catalysis.

Cruz Hergueta Santos-Olmo studied Mechanical Engineering at University of Castilla- La Mancha (UCLM), Spain, between 2007 and 2012. In September 2015 he was successful with his scholarship application to fund his PhD studies and since then he is a Doctoral Candidate at the Department of Mechanical Engineering at the University of Birmingham. The research work is focused on gasoline-bio-alcohols blends (gasoline- ethanol and gasoline-butanol) performance on modern gasoline direct injection engines combined with alternative technological solutions such as exhaust gas recirculation strategies, reformed exhaust gas recirculation and gasoline particulate filters for improving efficiency and emissions reduction.

Professor Athanasios Tsolakis has academic and industrial expertise in the field of low carbon energy carriers, environmental catalysts, combustion and pollutant control technologies. He works at the forefront of basic and translational research to improve fuel efficiency and reduce the environmental impact of the transportation and power generation sectors. Prior to his academic appointment at the University of Birmingham in 2005 he worked as a research scientist at Johnson Matthey in the design and characterisation of environmental catalysts for modern aftertreatment systems. Professor Tsolakis has successfully supervised to completion 27 doctoral researchers. In 2009 he was elected Fellow of the Institution of Mechanical Engineers (FIMechE) and in 2011 he was elected Fellow of the Higher Education Academy (FHEA). In 2014 he led the Research Excellence Framework 2014 (REF2014) for the School of Mechanical Engineering and was appointed the Director of Research and Knowledge Transfer for the School. Since 2015 he is the Director of Research for the School of Engineering.

Dr Andy York joined Johnson Matthey, Sonning Common, in 2000 in the Emission Control Research group. He has worked in a variety of roles, including gasoline and diesel catalyst research, and reaction engineering modelling and reaction kinetics. He has also led a collaboration with the University of Cambridge, UK, working on a wide range of academic and business related projects involving catalysis and engineering.

Dr Paul Millington originally joined Johnson Matthey in the Emission Control Research group in 1995. After a short break in the automotive industry he rejoined in 2001. He currently works on all forms of pgm-containing aftertreatment in the Emission Control Research group at Johnson Matthey.

www.technology.matthey.com

Green Catalysts for Energy Transformation and Emission Control Edited by Virender K. Sharma (Texas A&M University, College Station, Texas, USA), Sue-min Chang (National Chiao Tung University, Hsinchu, Taiwan), Ruey-an Doong and Chien-Hou Wu (National Tsing Hua University, Hsinchu, Taiwan), ACS Symposium Series, Volume 1184, American Chemical Society, Washington, DC, USA, 2014, ISBN 9780841230149, eISBN 9780841230156, £112.50, US$160.00

Reviewed by Catherine Davies Removal of Pollutants and Reduction Cardiff Catalysis Institute, School of Chemistry, of Greenhouse Gases Cardiff University, Park Place, Cardiff CF10 3AT, UK The first three chapters of the book cover quite varied topics: use of iron nanoparticles to remove Email: [email protected] organic pollutants in soil, biomass-derived coal‑like fuel to reduce greenhouse gas emissions and synthesis of sensors to detect and remove Introduction environmental mercury contaminants. In the “Green Catalysts for Energy Transformation first chapter of the book, as well as reviewing and Emission Control” is book 1184 in the ACS the use of iron nanoparticles in remediation of Symposium Series, which has been published soil contamination, including full-scale, real‑life since 1974 and is peer-reviewed, consisting of examples, J. Virkutyte (Hammontree and original research papers and review articles. The Associates Ltd, Ohio, USA) and R. S. Varma (US purpose of the series is to publish comprehensive Environmental Protection Agency, Ohio, USA) books based on current scientific research provide a comprehensive introduction to the presented at ACS sponsored symposia. The principles of green and sustainable remediation content of this particular volume was developed and the processes involved, which also acts as an from the symposium of the same title, held at introduction for the rest of the book. the 246th ACS national meeting in Indianapolis, The second chapter is less focused on removal Indiana, USA, in 2013 and sponsored by the of environmental pollutants than preventing ACS Division of Environmental Chemistry. It them. G. K. Parshetti and R. Balasubramanian contains 12 chapters, covering a variety of (National University of Singapore) describe how topics within the subject area, including removal addition of hydrothermally carbonised oil-palm of environmental pollutants from gaseous and empty fruit bunch, the primary solid waste of liquid phase systems. The research interests palm oil processing, to coal at an optimum ratio of the book’s editors include environmental can enhance the fuel’s burning properties, so applications of iron species, use of nanoparticles that less pollutants such as carbon monoxide for decontamination of wastewater, development and methane are produced. In Chapter 3, M. of materials for photocatalysis and method Lee et al. (Tunghai University, Taiwan) review development for trace analysis of environmental current developments in the use of naphthalimide pollutants. The intended audience of this book is derivatives for chemical sensing of mercury, which graduate students who are engaged in research in is a highly toxic environmental pollutant. Such the fields of green chemistry, nanotechnology and fluorochemosensors provide an improvement environmental science. over typical detection methods since they are less

342 © 2017 Johnson Matthey https://doi.org/10.1595/205651317X696333 Johnson Matthey Technol. Rev., 2017, 61, (4) costly, can be more selective and sensitive, and can hydrogen evolution from water splitting, and silver provide results in terms of a visible colour change. doped bismuth iodide as a photoanode in a photo fuel cell for simultaneous electricity generation Photoactivated Materials and degradation of organic wastewater pollutants respectively. In Chapter 8, H. Liao and R.-a. Doong The next six chapters are all related to (National Tsing Hua University, Taiwan) report the photoactivated materials. Chapters 4 to 7 discuss investigation and optimisation of the parameters photocatalysis using a variety of materials for in the preparation of titania nanotubes, and how degradation of aqueous and gaseous compounds, the addition of platinum significantly increases the using ultraviolet (UV) or visible light. Chapter 4 is photoconversion efficiency. In Chapter 9, C. He et concerned with the use of iron-containing silicate al. (Sun Yat-sen University, China; South China glasses as photocatalysts which can use visible Normal University, China and Guangdong Provincial wavelength light to decompose methylene blue dye. Key Laboratory of Environmental Pollution Control Y. Takahashi et al. (Tokyo Metropolitan University, and Remediation Technology, China) synthesise Japan, and Kinki University, Japan) describe and characterise bismuth iodide and silver-bismuth how catalysts were prepared using two different iodide electrodes and test their activity using methods, and through Mossbauer spectroscopy bisphenol A solutions as simulated wastewater. It and X-ray diffraction (XRD) it was determined that was concluded that the addition of silver enhanced good catalytic activity could be attributed to the both the efficiency of electricity generation and presence of α-Fe2O3. degradation of pollutants in the photo fuel cell In Chapter 5, acid yellow 17 is the target molecule under visible light. for decomposition. C. Wu et al. (National Tsing Hua University, Taiwan, and Industrial Technology Environmental Decontamination Research Institute, Taiwan) investigate the activity of soda lime glass-supported thin titania films of The final three chapters of the book are each varying thickness, using UV light. It is concluded concerned with iron species and their use in that there is an optimal film thickness, as a balance environmental decontamination, since iron is between the amount of titania providing active an earth-abundant element. Chapter 10 briefly sites for catalysis, and the ability of the light to reviews the use of iron-enriched mineral oxides penetrate the catalyst material. Various methods for water decontamination before describing two of preparing the films are also reviewed. J. Chen studies carried out by the authors: Y. Li and W. et al. (Tunghai University, Taiwan) investigate in Yan (Texas Tech University, USA), to investigate Chapter 6 how the morphology of anodic aluminium the importance of the amount of iron present. In oxide (AAO) supported zinc oxide catalysts affects Chapter 11, Y. Wanatabe et al. (Tokyo Metropolitan their activity for degradation of gaseous isopropyl University, Japan, and Kinki University, Japan) used alcohol under UV light. Plasma-enhanced chemical a variety of characterisation techniques to analyse vapour deposition (PECVD) was utilised to deposit the effect of trichloroethylene and methylene blue the zinc oxide at different weight loadings, which decomposition on the local structure of mixtures resulted in different morphologies; it was observed of metallic iron and iron oxides, and demonstrated that a rod-like structure was the most active. that the activity of the mixtures was enhanced Chapter 7, by one of the editors S.-m. Chang when nanoparticles were used. In Chapter 12, (National Chiao-Tung University, Taiwan), is a V. K. Sharma (Texas A&M University, USA) et al. review of the effects of bulk and surface doping discuss the use of ferrate(VI) for the treatment of of titania on photocatalytic activity. The review odorous gases, such as organosulfur compounds. focuses more on the catalytic materials themselves The iron is reduced by the reaction, through and the various mechanisms by which doping transfer of oxygen from the ferrate species, affects their activity rather than any particular resulting in iron(II) or iron(III) products, however, reaction, although decolouration of acid naphthol it is stated that iron(VI) could be generated in situ red solution is one that is mentioned. It is concluded electrochemically. that surface doping of titania is more beneficial than bulk doping. Conclusions Chapters 8 and 9 extend photoactivation to electrochemistry, discussing the use of titania Overall, the chapters in this book cover a range and platinum-doped titania as a photocathode for of topics relating to mitigation of environmental

pollutants and are accessible to readers with an “Green Catalysts interest in this broad area, due to introductions for Energy which explain each topic area well, although some Transformation and prior knowledge of catalysis or the analytical Emission Control” techniques used may be an advantage in terms of understanding the data presented. The title of the book is slightly misleading, however, as not all chapters are about catalysts. A substantial portion of the book is concerned with photo-activated materials, particularly photocatalysts for treatment of wastewater, and so it may be of most interest to readers involved in this area of research. These chapters also provide sufficient description of the sometimes complex mechanisms involved that those unfamiliar with the concept should be able to understand their content.

The Reviewer Dr Catherine Davies is a Post Doctoral Research Associate at the Cardiff Catalysis Institute. Her current research interests are in heterogeneous catalysis for automotive applications, focusing on catalysed oxidation of particulate matter.

344 © 2017 Johnson Matthey Johnson Matthey Technology Review is Johnson Matthey’s international journal of research exploring science and technology in industrial applications www.technology.matthey.com Editorial team

Manager Dan Carter Editor Sara Coles Editorial Assistant Ming Chung Senior Information OfficerElisabeth Riley

Johnson Matthey Technology Review Johnson Matthey Plc Orchard Road Royston SG8 5HE UK Tel +44 (0)1763 253 000 Email [email protected] www.technology.matthey.com