Life Cycle Analysis – Review of the tool

Introduction

Life Cycle Analysis can be a useful tool for assessing the environmental impacts of processes and products in society, and making strategic decisions based on this assessment[1]. However its use should be approached with a thorough understanding of the tool itself, and its component parts. In addition, the specifics of each LCA model must be taken into account before an informed decision can be made, particularly if any comparisons are to be made between reports. These include the location of the report, the technologies assessed, the data sources used, and any assumptions that are being made about the future, among other factors.

This paper takes a sample of LCA reports on different aspects of hydrogen production and use, and briefly discusses their findings, highlighting the similarities and differences between them, with the intention of drawing conclusions on sustainable hydrogen generation pathways. This will highlight the areas in which LCA analysis can assist with decision-making for resource optimization and waste management. The paper will look at them using the framework provided by ISO Documents 14040:2006 (Environmental Management - Life Cycle Assessment - Principles and Framework) and 14044:2006 (Environmental Management - Life Cycle Assessment - Requirements and Guidelines), so as to more easily compare and contrast the reports.

By comparing the reports, a more thorough understanding of the LCA tool itself can be obtained. This allows a more accurate interpretation to be made of the reported results, and ultimately an informed assessment can be made of the impact of the processes in question. Decisions are more likely to be reached and will be more valid if based on a sound understanding of all these elements.

ISO Standards

There are four core components of an LCA within the ISO 14040 standards. These are “Goal Definition and Scoping”, “Inventory (LCI)”, “Impact Assessment (LCIA)”, and “Interpretation”. Additionally, there are four optional elements which may be included within the LCIA. These are normalization, grouping, weighting and data quality analysis. These components will be outlined briefly below.

Goal Definition and Scoping

The goal of an LCA outlines the primary aims of the study in question, including the reasoning behind it and the target audience [2]. The scope outlines how this goal is to be achieved, and identifies the depth and breadth of the model that is to be constructed. Particularly, it should outline the reference unit and the boundaries of the study. The reference unit chosen in the study is a key element, as it is a common end-product of two or more systems. In addition, the boundary of the study identifies the limits to which the study has been taken. Any LCA attempts to model a process in the real world, however trying to model every element that could have an impact is unrealistic. As such, the boundaries identify the cut-off points of the model, and so also help identify those areas of the process that have not been modelled. The aim of an LCA should be to model all components of a process, from cradle to grave.

Life Cycle Inventory (LCI)

The LCI stage encompasses data collection, data entry and subsequent calculations to allocate resource and waste flows to specific parts of a process. It is generally undertaken with the aid of software and there are a range of data sources available. Databases are available of a range of process inputs and outputs, from a number of sources - including government and industry - and this can lead to uncertainty over its transparency and impartiality.[3]  ISO 14044[4], states that allocation that is not based on hard data should be avoided where possible by system expansion, and provides guidelines for dealing with allocations. Allocation procedures introduced by the study itself should therefore be considered when analysing and comparing reports.

Impact Assessment (LCIA)

In the LCIA, the allocated flows are classified according to specific environmental impact categories and category indicators, chosen to assist with the goal and scope of the study.  Examples of these categories are eutrophication, acidification and global warming potential, and studies will often use a number of them in order to assist with the LCIA. The processes in the study are then characterized according to their level of impact.

The ISO standards outline four optional elements in an LCIA. “Normalization” converts the results to a common unit for comparison. “Grouping” sorts impact categories according to the goals of the study, and this stage can also include ranking based on value judgements. “Weighting” can be used to numerically adjust and aggregate results across impact categories, and is necessarily based on value judgements, however unweighted results should also be made available. “Data Quality Analysis” encompasses three main approaches to allow further analysis of the results of the LCIA, to better understand their uncertainty, sensitivity and significance.

Care must be taken when comparing LCIAs, as impact categories and indicators will have been chosen specifically for the study. In particular, this phase can introduce value judgements, which must be considered. The ISO standards specify guidelines that any LCA intended for public consumption and comparison should follow, including the need for internationally recognised impact categories, and the exclusion of weighting.

Interpretation

This final step is where conclusions are drawn from the previous stages, identifying main findings, and providing recommendations and suggestions for further investigation within the scope and goals of the study. Additionally, this section should evaluate the study’s completeness, consistency and sensitivity, and identify areas where another iteration of the study could assist in providing a more complete model of the processes studied. As the conclusive part of an LCA report, particular care must be taken when assessing the importance of the findings, taking into consideration all the uncertainties and possible biases mentioned above.

Criticisms of ISO standards

Finnveden et al.[5] highlight three main criticisms of the ISO standards. From a design perspective, the standards do not account for the relationship between the method and goal of an LCA[6]. Another criticism of the design is the absence of any reference to the ISO standard for dealing with uncertainty, when discussing uncertainty in an LCA[7]. The lack of LCA methods for dealing adequately with certain flows, including freshwater depletion, land use and chemical flows are also mentioned[8].

Report Adherence to ISO

The degree to which LCAs follow the standards laid out in the ISO documentation is a factor to be taken into consideration when interpreting them. ISO adherence means that reports are more easily comparable to each other as they have a similar structure and terminology, however there are a number of optional elements within the guidance[9], so complete uniformity between reports, even if desirable, is unlikely. It has been noted by Geerken et al. in a review of LCAs that a large proportion of current LCAs do not adhere to ISO standards, or even mention them[10].

LCA Reports

The following LCA reports on various aspects of hydrogen production and use are summarised within the framework of the ISO guidelines:

1.    Koroneos, C et al.,(2004) Life cycle assessment of hydrogen fuel production processes,   International Journal of Hydrogen Energy 29, 1443 – 1450

2.    Ji-Yong Lee et al.,(2010) Life cycle environmental and economic analyses of a hydrogen station with wind energy, International Journal of Hydrogen Energy 35, 2213–2225

3.    J. Ally, T. Pryor, (2007) Life-cycle assessment of diesel, natural gas and hydrogen fuel cell bus transportation systems, Journal of Power Sources 170, 401–411

1. Life cycle assessment of hydrogen fuel production processes[11]
Koroneos et al. (2004) analysed the impact of generating 1 MJ of energy from hydrogen using various hydrogen production methods. These consisted of one conventional source of hydrogen - steam reforming of natural gas - and a number of renewable energy systems connected to a high-pressure alkaline electrolyser. The renewables included photovoltaic panels, solar thermal energy, wind, hydro and biomass, and the model included subsequent liquefaction of hydrogen. All the data on the renewable energies was taken from a previous study based on GEMIS data[12].

The LCIA categories used were Global Warming Potential (GWP), using CO2 as an indicator, Acidification Effect, based on SO2, Eutrophication Effect, based on PO4, and Winter Smog Effect, indicated by solid particulate matter. This was the only study to use a weighting mechanism, in the form of Eco-indicator 95. This method emphasises environmental effects on eco-systems and human health.

The LCI results and interpretation indicated that the photovoltaic system had the worst overall performance due to the resource intensity of the manufacturing process, with biomass combustion marginally worse than conventional steam reforming. Solar thermal had half the impact of steam reforming, while hydro was half of this again. Wind was seen as having the least impact, at roughly 15% of the overall impact of conventional methods.[13]

2. Life cycle environmental and economic analyses of a hydrogen station with wind energy[14]

The reference unit chosen by Lee et al. (2010) was 160,000 km of driving range. The hydrogen production model encompassed the water extraction and use for electrolysis, electrolysis, compression and storage of H2, followed by its use in a fuel cell vehicle (FCV). Electricity for electrolysis was provided by either the Korean national grid or a wind power plant. These pathways were compared with conventional gasoline and diesel production processes. Vehicle manufacture and infrastructure was excluded from the study.

Data for fossil fuels was taken from Korean National databases. Wind power construction data came from NREL databases, while wind plant operational and maintenance data was from two Korean wind plants. Electrolyser figures were from operational data of a 11.2 Nm3/h Teledyne unit.[15]

The categories used were Abiotic Resource Depletion (ARD), Fossil Fuel Consumption and GWP, also detailing the regulated air pollutants of each pathway. The resultant data and interpretation ranked the wind-generated hydrogen as having the least impact, while grid hydrogen was only marginally better than conventional fuels in most categories[16]. This was mainly due to the grid’s high proportion of thermal power generation methods, and crude oil sensitivity to ARD analysis. FCV’s were noted as being particularly efficient compared to internal combustion engines which accounted for their high performance[17].

3. Life-cycle assessment of diesel, natural gas and hydrogen fuel cell bus transportation[18]

Ally and Prior (2007) used a functional unit of vehicle kilometres[19] in three different fuel buses, to compare traditional diesel production pathways with compressed natural gas, and hydrogen via four main production routes. Two electrolysis systems were considered, drawing power from a wind plant and the grid. Steam reforming from natural gas was another hydrogen source. The final source was from a local refinery which produces hydrogen as a by-product[20]. The system boundary is identified as including the fuel infrastructure to a cut-off criteria of 1%.

The study used Gabi 4 software and datasets[21] and previous studies on the oil refinery to provide an LCIA based on GWP, Acidification Potential, Eutrophication Potential and Photochemical Ozone Creation Potential[22]. These results were normalised using the impact of diesel as 100%[23].

The findings in relation to GWP showed a fuel cell from wind power as being the only option for the bus that was better than diesel, and it was significantly better. The hydrogen by-product from the refinery was on a par with diesel, while compressed natural gas was slightly worse. Steam reforming was seen as having nearly a 300% impact, while hydrogen production using the grid had nearly ten times the impact of diesel[24].

The authors stressed how the relative infancy of hydrogen technologies had a negative impact on the overall performance of the hydrogen pathways due to the 1% cut-off criteria, which apportioned a high impact to the infrastructure of hydrogen[25]. In addition, inefficiencies in the fuel cell bus were mentioned, which further reduced their performance. Future improvements in fuel cell technology were hypothesised which would improve the overall picture for hydrogen use and production[26]. Uncertainties over the lifetime of a fuel cell were also raised, and it was noted that improvements in this area could also lead to large improvements in life-cycle profile for all hydrogen routes[27].

Assessment of the Results

The common conclusion reached by all three studies discussed above highlight one clear pathway to sustainable hydrogen generation, through electrolysis powered by wind. It is part of all the studies, and ranks highest in all three. Indeed, other studies reach similar conclusions when hydrogen through wind is analysed[28][29][30]. However, before any decisions could be made based on this knowledge, an obvious factor that would need to be taken into consideration is geography. There would be little point making decisions based on wind-generated hydrogen if wind power was not available in the reader’s locale.

All studies rank hydrogen generation via fossil fuels in any capacity, either steam reforming, or using fossil-mixed grid electrolysis, quite badly, however Ally and Pryor (2007, p.407) introduce the idea that hydrogen that is a by-product of the fossil fuel industry can compete environmentally with conventional fuels. A similar proposition is put forward by Hwang and Chang (2010)[31] who conclude that coke-oven gas hydrogen is the most favorable pathway in their well-to-wheel analysis on hydrogen scooters.

However it would be difficult to base long-term decisions on such a pathway without completely assessing the local resource and its particular processes. In particular, a study applying Abiotic Resource Depletion methods would be advisable for any processes reliant on fossil fuels, so as to minimise the risk of further dependency on fossil fuels and the entrenchment of fundamentally unsustainable processes.

Although LCA can help provide a picture of the relative impacts of different processes, there are still uncertainties attached to novel technologies and future scenarios. In the referenced papers, these centre around the electrolysis and fuel cell technologies, which are still in their infancy, and are presumed to get progressively more efficient over time. Each of the studies focused on one particular electrolysis or fuel cell process and it would be necessary to study a number of systems before strategic decisions could be made.

As such, reading LCAs can provide a general benchmark by which to gauge technologies or processes, assuming a solid understanding of the LCA process, and an awareness of all the possible uncertainties. Biases, either through the application of a weighting method, or through omissions or inaccurate data, require caution. LCAs are in fact increasingly being used at macro-policy level, particularly in conjunction with Input Output tables, though Finnvenden et al. point out that many authors argue that these studies are too coarse.[32]

In order to make solid strategic decisions based on LCA however, a new study should always be designed to answer questions specific to the situation and to model all the relevant processes.

Bibliography

• Finnveden et al. (2005) Life Cycle Assessment of energy from solid waste-part 1: General methodology and results, Journal of Cleaner Production 13
• British Standards Institution (2006)  BS ISO/IEC 14040:2006. Environmental Management - Life Cycle Assessment - Principles and Framework. London, British Standards Institution
• Finnveden et al. (2009) Recent Developments in Life Cycle Assessment, Journal of Environmental Management 91
• British Standards Institution (2006)  BS ISO/IEC 14044:2006. Environmental Management - Life Cycle Assessment - Requirements and Guidelines. London, British Standards Institution
• T Geerken et al. (2004) Review of hydrogen LCAs for the Hysociety project – Final report, Hysociety Project
• Koroneos, C et al.,(2004) Life cycle assessment of hydrogen fuel production processes,   International Journal of Hydrogen Energy 29
• Lee. J et al. (2010) Life Cycle environmental and economic analyses of a hydrogen station with wind energy, International Journal of Hydrogen Energy 35
• J. Ally, T. Pryor (2007) Life-cycle assessment of diesel, natural gas and hydrogen fuel cell bus transportation systems, Journal of Power Sources 170
• M. Granovskii et al. (2006) Life cycle assessment of hydrogen fuel cell and gasoline vehicles, International Journal of Hydrogen Energy 31
• U. Wagner et al. (1998) Energy Life Cycle Analysis of Hydrogen Systems,  International Journal of Hydrogen Energy 23
• M. Granovskii et al. (2007) Exergetic life cycle assessment of hydrogen production from renewables, Journal of Power Sources 167
• ^[1]Hwang, J and Chang, W (2009) Life-cycle analysis of greenhouse gas emission and energy efficiency of hydrogen fuel cell scooters, International Journal of Hydrogen Energy 35

References

---

[1]Finnveden et al. (2005) Life Cycle Assessment of energy from solid waste-part 1: General methodology and results, Journal of Cleaner Production 13, p213

[2]British Standards Institution (2006)  BS ISO/IEC 14040:2006. Environmental Management - Life Cycle Assessment - Principles and Framework. London, British Standards Institution, p11

[3]Finnveden et al. (2009) Recent Developments in Life Cycle Assessment, Journal of Environmental Management 91, p7

[4]British Standards Institution (2006)  BS ISO/IEC 14044:2006. Environmental Management - Life Cycle Assessment - Requirements and Guidelines. London, British Standards Institution, p14

[5]Finnveden et al. (2009) Recent Developments in Life Cycle Assessment, Journal of Environmental Management 91, pp6-15

[6]ibid., 6

[7]ibid., 15

[8]ibid., 13-14

[9]British Standards Institution (2006)  BS ISO/IEC 14044:2006. Environmental Management - Life Cycle Assessment - Requirements and Guidelines. London, British Standards Institution, p20

[10] T Geerken et al. (2004) Review of hydrogen LCAs for the Hysociety project – Final report, Hysociety Project, p10

[11]Koroneos, C et al.,(2004) Life cycle assessment of hydrogen fuel production processes,   International Journal of Hydrogen Energy 29, pp1443 – 1450

[12]ibid., p1445

[13]ibid., pp1449-50

[14]Lee. J et al. (2010) Life Cycle environmental and economic analyses of a hydrogen station with wind energy, International Journal of Hydrogen Energy 35, pp2213–2225

[15]ibid., 2217

[16]ibid., 2221/2

[17]ibid., 2219

[18]J. Ally, T. Pryor (2007) Life-cycle assessment of diesel, natural gas and hydrogen fuel cell bus transportation systems, Journal of Power Sources 170, pp401–411

[19]ibid., 406

[20]ibid., 404

[21]ibid., 403

[22]ibid., 406

[23]ibid., 405

[24]ibid., 408

[25]ibid., 403

[26]ibid., 409

[27]ibid., 407

[28]M. Granovskii et al. (2006) Life cycle assessment of hydrogen fuel cell and gasoline vehicles, International Journal of Hydrogen Energy 31, p350

[29]U. Wagner et al. (1998) Energy Life Cycle Analysis of Hydrogen Systems,  International Journal of Hydrogen Energy 23, pp1-6

[30]M. Granovskii et al. (2007) Exergetic life cycle assessment of hydrogen production from renewables, Journal of Power Sources 167, pp461–471

[31]Hwang, J and Chang, W (2009) Life-cycle analysis of greenhouse gas emission and energy efficiency of hydrogen fuel cell scooters, International Journal of Hydrogen Energy 35, pp11947 - 11956

[32]Finnveden et al. (2009) Recent Developments in Life Cycle Assessment, Journal of Environmental Management 91, p8