Finding Your Way In Multifunctional Processes And Recycling
Since the beginning of LCA, there have been two major issues on which we will probably never reach consensus: how to deal with multifunctional processes and with recycling. This article tries to explain the concepts underlying this question.
No One True Way
Because there is no obvious solution to many impact allocation problems, the ISO standards for LCA leave a large degree of freedom, and they can be interpreted in many ways. For the problems of recycling and multifunctional processes, there are also multiple good points of view, as you’ll understand after reading this article.
This flexibility has led to different approaches, such as the British PAS2050 for carbon footprinting, the Greenhouse Gas (GHG) Protocol, the International EPD System, and the Product Environmental Footprint (PEF) Guide.
LCI data providers also have different views on how to deal with these issues. For example, the Agri-footprint database is available in three versions with three different allocation keys to divide the upstream burdens of multifunctional processes between the co-products. The ecoinvent data is also available in three different versions, based on very diverse approaches.
Important concepts in LCA
In this article, I’ll try to give an overview of the different options. I will do this by describing six important concepts in LCA:
- The consequential approach
- Physical allocation
- Economic allocation
- Allocation at the point of substitution
- Recycled content or cut-off approach
- Closed-loop scenarios
1. The Consequential Approach
The consequential approach to allocation is probably the most theoretically correct model to deal with multifunctional processes and recycling. In practice, however, you need to make a lot of assumptions about the so-called avoided burdens.
Let’s take beer as an example. The life cycle of a can of beer includes co-production of straw (from growing barley) and spent grains (from brewing the beer). Straw and spent grains can be used as animal feed, and their co-production means the impact of producing other sources of animal feed is avoided. After the beer is consumed, recycling of the beer can means the impact of virgin aluminium production is avoided.
The consequential approach forces users to make assumptions about quality: do the other sources of animal feed have the same function/quality as the straw and spent grains? And does the recycled aluminium have the same quality as the virgin aluminium?
2. Physical Allocation
If you want to avoid having to make such assumptions and concentrate on the life cycle of the beer can itself, you can choose an attribution approach. However, this approach is not free of arbitrary choices either. In the case of multifunctional processes, you first need to classify the outputs as waste, recyclable material or marketable co-products. Then you need to determine an allocation key for the marketable products. This can be based on physical characteristics, such as mass, dry mass, volume, exergy content, energy content, and energy input associated with each co-product.
For example, if the beer cans are transported together with other beverages, you may divide the burdens related to transport by the relative volume of each type of beverage in the truck. When looking at the spent grains, volume is not an appropriate measurement, but allocating by mass would mean that about 15% of the burden of beer brewing is allocated to the spent grains, while the revenue for brewers from this by-product is almost negligible compared to the revenue from the beer.
3. Economic Allocation
As seen above, physical allocation can be considered unfair for users of the lower-valued by-products. To prevent this unfairness, it is not uncommon to use the price of the product and by-products to calculate the allocation keys. Although prices are often confidential and volatile, solutions can be found. To avoid communicating the real prices, for example, you can use average prices over several years and divide them by the price of the main product.
4. Allocation At The Point Of Substitution
Spent grains are not a typical by-product, because they are available in different forms, depending on dry matter percentage. Straight from the brewing process, spent grains are rather wet and have very little market value. You could therefore choose to not allocate any burdens to the wet spent grains and only attribute the burdens from drying the spent grains.
Alternatively, you could include the drying of the spent grains in the brewing process and allocate the burdens at the point that production of spent grains could potentially cause the burden of production of alternative animal feed to be avoided. This way, the beer also carries some burden from drying the spent grains and the spent grains carry some burden from beer brewing.
You could do the same with the used aluminium cans. The burdens from recycling the used cans into aluminium for new cans can be partly allocated to the beer and the burdens from brewing beer can be partly allocated to the recycled aluminium.
This complex approach is called allocation at the point of substitution and is used in the ecoinvent system model called allocation, ecoinvent default. It has some theoretical advantages, but in practice, it is very complex and sometimes leads to strange results. For example, recycled metals may have higher impact scores than their virgin equivalents. If you use this approach, you need to check if the results make sense compared to the simpler and more conventional allocation approach.
5. Recycled Content Or Cut-off Approach
The most common approach for situations like the wet spent grains and used aluminium cans is the recycled content or cut-off approach. This model allocates burdens at the point where a product is sold and applies a cut-off at the point the recyclable material leaves the product system. So, the wet spent grains are sold for a negligible price and no burdens from brewing are allocated to them. Similarly, the used cans carry no burden from before the point they were collected for recycling.
6. Closed-loop Scenario
But what if there is a great demand for recycled aluminium, there is limited supply in your country, and you use relatively little recycled aluminium because of the limited availability at the production site? If you’d use the recycled content approach, your beer cans would get a large burden from using virgin aluminium, while the used cans are almost all collected for making new beverage cans. In such situations, several LCA and carbon footprint specifications allow you to apply a closed-loop scenario. This means that you assume that all the cans that were recycled will be used for the same purpose again. The percentage of recycled aluminium for making the cans is assumed to be equal to the percentage of cans that is recycled at the end of the beer can’s life.
The problem with this approach is that it is not always easy to see when the closed-loop scenario applies. How do you determine whether the market for a recycled material is saturated? There is a high demand, for example, for cardboard to create the cartons around beer cans, but there is also a large supply of recycled cardboard. This difficulty is why the PEF Guide introduced a sort of 50/50 approach, where you combine the two approaches with equal weights. This seems arbitrary and unfair for some industrial sectors, but at least it is consistent.
Accepting Different Choices
In an ideal situation, we would all use the consequential approach. Then we could discuss our assumptions about which product is actually substituting which product in practice, rather than endlessly disagree on the allocation methods and recycling formulas we should use. However, I don’t think we are ready for this utopian situation. Life cycles are complex and full of multifunctional processes and recycling. It takes a lot of time to go through all the options, collect the additional data of substituted life cycles, and convince stakeholders that you made the best decisions for solving each situation.
We need to accept that we disagree on which allocation method and recycling formula should be applied. The most basic attributional approach can be found in the International EPD system and the allocation, recycled content or cut-off system model of ecoinvent. The PAS2050 and the GHG Protocol allow the choice between recycled content and closed loop. The PEF Guide and the allocation at the point of substitution system model of ecoinvent introduce complex compromises.
These are just a few of the leading initiatives. There are many more approaches out there. And the bottom line is: all these approaches are correct in their own way. All you can do as user is decide which point of view makes most sense for your situation, and choose accordingly.
Tommie worked for PRé as a Technical Consultant from 2012 until 2015. As a part of the Consultancy Team, he worked with databases and methods. Tommie collaborated in projects such as Prosuite and improving the ReCiPe method.