InLCA-LCM 2004

Lesage Pascal · Post subject: **The recycling process in CLCI** Posted: Wed Jul 21, 2004 4:57 pm

Dear authors.

Thank you for your paper. Although it provoked much questioning of my own use of CLCI modelling, I will limit my intervention on this forum to a specific aspect of the model presented: the recycling of collected scrap lead in the CLCI.

*For attributional LCI, the lead recycling process is considered part of the next life cycle stage (cut off). However, it seems to be also cut from the CLCI: I did not find reference to it other than in the model structure schema (it is absent from Table 1). If this is indeed a cut-off, my understanding is that it would imply that the demand D for collected scrap lead, which flows to recycling facilities, is considered constant. Is my understanding correct?

*In cases where (a) the marginal production technology using recycled material is known and (b) recycling processes can be differentiated for specific types of subsequent use of the material, would it be acceptable to
1) calculate the magnitude of the effect on the flow of material from the market to recycling processes (ΔD) using the economic allocation method proposed; and
2) subsequently consider that ΔD affects a specific (or marginal ?) type of recycling process Ri?

*And in cases where the recycling processes do not significantly differ among themselves, or when the subsequent use of the material is unknown, would it be acceptable to allocate an increase (or decrease) of ΔD to average recycling processes Ro?

Thank you in advance
_________________
Pascal Lesage
Student
CIRAIG
Dept of Chemical Engineering,
École Polytechnique de Montréal

Lesage Pascal

Hi Tomas
Thanks for doing away with my uneasiness concerning the recycling process.

As we have discussed together on a previous occasion, my investigation is of the environmental consequences of the management of a brownfield, i.e. an urban site which was occupied, normally by industrial facilities, and which requires rehabilitation for reinsertion in the economy.
The parallel between site rehabilitation and open loop recycling (OLR) of material has led me to be very interested in your economic allocation procedure. I am in the process of adapting your technological whole-system model of OLR in order to express land as a recoverable flow resource used by and exchanged between unit processes (e.g. land transformation, use, rehabilitation and “passive occupation”).
I am grateful for the interest you show in this work and will send you shortly a summary of this adaptation. There are issues I have come across, however, that perhaps have been encountered by others whose insight would surely be of value to me:

1) The tracked cascade material in my study (a given site) has two types of characteristics:
-A set that can be modified (positively or negatively) by economic activities. In the case of land, this can include the site’s “environmental quality” (using LSF, biodiversity and (eco)toxicity indicators) and “appropriateness for a given use” (determined by the site quality and by policy-derived minimum site quality requirements for specified uses of the land);
-A set of characteristics that are considered unalterable. For example, the location of the tracked site cannot be altered by any unit process. The quality that makes it distinguishable (location) strongly influences its relative worth on the market.
Therefore, as opposed to the lead in your paper, the tracked recycled material may not be perfectly interchangeable with other recycled material nor with virgin material. Also, the definition of the market will inevitably group sites which are not economically equivalent. I have yet to determine what level of simplification will not introduce unacceptable uncertainty and yet be feasible.

2) The collection and recycling phases are highly sensitive to the type of reuse which will follow on the tracked site.

3) The collection phase is highly sensitive to the state of the material (land) as it exits the previous use phase.

4) In some decision contexts, the same decision maker is thought to have influence on decisions from the collection phase all the way to the reuse phase. It intuitively seems to me that the influence on the market is different in such a case – but I have found nothing to justify this inkling.

Thank you for the opportunity.
Pascal
_________________
Pascal Lesage
Student
CIRAIG
Dept of Chemical Engineering,
École Polytechnique de Montréal

tomase

I understand that specific methodological issues must be addressed when the market-based approach to open-loop recycling (OLR) is applied to land rather than material recycling. I have thoughts to offer only on the 4th issue. If a land-owner decides to upgrade a square kilometer industrial site to allow for residential buildings, without selling the land, the site will not enter the market for "collected" land. My market-based OLR model might not be appropriate in this case. If it is applied anyway, it should take into account that the decision increases the supply of and demand for "collected" land by the same amount (1 km2). In such cases the market-based OLR model is equivalent to a model of closed-loop recycling: there is no net effect on the demand for and supply of land area elsewhere. However, it seems reasonable to assume that the decision will increase the demand for industrial sites and reduce the demand for residential sites elsewhere. There will be no effect on the demand for land area, but a change in demand for land quality. This line of reasoning reinforces your statement that different types of land are not substitutable (at least not in regions where upgrading of land is an economically interesting option).
_________________
Tomas Ekvall, Associate Professor
Energy Technology
Chalmers University of Technology
Gothenburg, Sweden
tomas.ekvall@me.chalmers.se

tomase

Hi Pascal,

Thanks for your accurate reflections. They make it clear to me that Table 1 should include the lead recycling. Our intention is to include, in the CLCI, also effects on the recycling of lead, as indicated by Figure 2. The initial data goal for this part of the system would be the marginal recycling process. Other types of data can be acceptable, depending on the precision required in the results and on the data availability. If different recycling processes do not differ significantly, average data would be sufficient in most cases.

Note that our preliminary CLCI model, which is illustrated in Figure 2, does not distinguish between the marginal use of virgin lead and the marginal use of recycled lead. Instead, we include only one alternative lead use. This part of the model is accurate only if virgin and recycled lead are suficiently interchangeable.

I am glad that our paper inspired thoughts about your own research in this area. And I am curious to know more about them, in this discussion forum or through email.
_________________
Tomas Ekvall, Associate Professor
Energy Technology
Chalmers University of Technology
Gothenburg, Sweden
tomas.ekvall@me.chalmers.se

Frydendal

Dear Authors

Based on your experiences so far, I am interested in your recommendations in relation to LCA on products and services in general.

If I were a company person and wanted to carry out an LCA for one of my products with the following goals in mind:
- Comparing my products to other products on the market
- Setting up environmental demands for my suppliers

Which method would you then advice me to use?
_________________
Jeppe Frydendal
M.Sc. in Engineering
Head of LCA Center Denmark

tomase

Dear Jeppe,

Thanks for your question, which is very relevant. In my mind, the consequential methodology is immature and only operational in part. For this reason I would not advice the company person to use the approach without close guidance by someone who has given the approach a lot of thought. Referring to the two cases you mention:

1. If you represent a company and want to compare your product to competing products, I would advice you to apply broad systems thinking to find out if there are potentially important consequential aspects that an attributional LCA would miss. If not, do the attributional LCA. If there are such aspect, my advice would depend on:
1a. if the comparison is made for internal use only, or
1b. if it is made to inform third party.
In the first case, you could do just the consequential LCA, under proper guidance of course. In the second case, my advice would be to do the attributional as well as the consequential LCA. This would allow you to inform the third party of the two different perspectives on the choice of products (the life-cycle perspective and the consequential perspective), allowing them to consider which of the two perspectives they perceive as the most relevant.

2. A consequential LCA will not assist in choosing among suppliers in most cases. The reason is that the consequential LCA will typically not include input data from your supplier; instead it will include input data from the marginal supplier of that good. If you want to use LCA to select supplier, my advice would probably be attributional LCA. But the case you asked about was using LCA to set environmental demands on your suppliers. In this case, the consequential as well as the attributional LCA can help. The input data from a marginal supplier can be used for benchmarking of different suppliers. My advice to you would be to do either an attributional or a consequential LCA depending on what perspective (life-cycle or consequential) you perceive as the most relevant.
_________________
Tomas Ekvall, Associate Professor
Energy Technology
Chalmers University of Technology
Gothenburg, Sweden
tomas.ekvall@me.chalmers.se

natdnomyar

Dear Authors,

Has there been any use of dynamic models in CLCI? It appears to me that methods such as system dynamics might be perfect for modelling technology transitions, but I have found very little of it in LCA literature.
My own work has been on comparative LCA of motor vehicle energy sources, using a "steady state" basis. CLCI would add a useful dimension to my research.
_________________
Raymond R. Tan, MSc, PhD
Associate Professor
Chemical Engineering Department
De La Salle University-Manila, Philippines
E-mail: rgrtan@asia.com

tomase

Dear Raymond,

Thank you for your question. We use a dynamic optimising model of the Nordic energy system to identify the marginal electricity production. Results from this dynamic model will be fed into the static, consequential LCI model. This is the only use of dynamic system modelling in CLCA that I am aware of. I do not know of any dynamic, consequential LCI model.
_________________
Tomas Ekvall, Associate Professor
Energy Technology
Chalmers University of Technology
Gothenburg, Sweden
tomas.ekvall@me.chalmers.se

natdnomyar

Dear Tomas,

Thanks for the quick reply. I have been thinking of using system dynamics (SD) in LCA for some time now - for example to model the gradual phaseout or replacement of a particular technology over time. I will need to study this approach, or perhaps find a collaborator with prior knowledge of SD, as my understanding of the methodology is still very elementary.

Perhaps you will do some work of this nature in the near future?
_________________
Raymond R. Tan, MSc, PhD
Associate Professor
Chemical Engineering Department
De La Salle University-Manila, Philippines
E-mail: rgrtan@asia.com

tomase

Thank you for your message, Raymond. Let us investigate the possibility of cooperation through email.
_________________
Tomas Ekvall, Associate Professor
Energy Technology
Chalmers University of Technology
Gothenburg, Sweden
tomas.ekvall@me.chalmers.se