Forum for ecoinvent Version 3

Here you find all boards with the latest posts of the Forum for ecoinvent v3

Written on 20.11.2017 by tmiketer see context

Dear Avraam,

Thank you for your reaction and help!

I am using version 3.3 in combination with Sytem model 'Allocation, cut-off by classification'. For this example, I used the following dataset: 'battery cell production, Li-ion' (CN)

Best regards,



Written on 20.11.2017 by Avraam.Symeonidis see context

Dear Tom,

Thank you for your post. In order for me to assist you I need to know:

What version of the database are you using?

What dataset are you looking at? (activity name)

What system model?

Best regards, 

Avraam Symeonidis,

Data Analyst, ecoinvent

Written on 20.11.2017 by guillaumebourgault see context

Dear Praneeth, 

Before I can help you with your question, I need to know a few more information.  

What software do you use?

What version and system model of ecoinent do you use?

What is the exact name of the dataset you are analysing?  Is it a copy or did you modify/added something to it for the purpose of your LCA?

On what coefficient exactly are you applying the 2.16 standard deviation?  

It is possible that randomly, a diverging loop appears in the technosphere matrix.  

Kind regards, 

Guillaume Bourgault, project manager for ecoinvent

Written on 20.11.2017 by tmiketer see context

Dear EcoInvent team and others,

I am doing an assessment to determine the material inventory of a LCI. But I have some questions on how to allocate companion fractions of the metals obtained “From the environment” of an LCI. Some materials from the LCI:

1.       ‘Cadmium, 0.30% in sulfide, Cd 0.18%, Pb, Zn, Ag, In, in ground’:                              0.000113 kg

2.       ‘Ni, Ni 2.3E+0%, Pt 2.5E-4%, Pd 7.3E-4%, Rh 2.0E-5%, Cu 3.2E+0% in ore, in ground’         0.000894 kg

3.       ‘Tellurium, 0.5ppm in sulfide, Te 0.2ppm, Cu and Ag, in crude ore, in ground’     7.877E-12kg

4.       ‘Iron, 46% in ore, 25% in crude ore, in ground’   0.053274kg


1.       In the first flow (Cadmium): is the amount of Cadmium 0.000113 kg? or 0.18%*0.000113 kg? And how to calculate the amount of Pb, Zn, Ag (looks like this is impossible as there is no documentation on these metals? Or have these metals the value of 0.18% as well?)

2.       In the second flow (Ni): is the amount of Nickel 0.000894 kg? Or 2.3%*0.000894 kg? In the second case, the amount of Pt should be 2.5E-4%*0.000894 kg, if I understand it right.

3.       In the third flow (Tellerium): Is the amount of Tellerium 7.877E-12kg, or is it 7.877E-12kg * 2E-5%?

4.       In the fourth flow (Iron): Is the amount of Iron 0.053274 kg? Or must it be allocated as well?

In other words, is the amount in kg given the amount of the substance itself or should it be allocated with the mentioned percentage in their name? Or does it depend? I cannot find any documentation on this issue.

Thank you so much for you time and kind help in advance!

Best regards,



Written on 20.11.2017 by guillaumebourgault see context

Dear Amy, 

Ecoinvent has invested resources in maintaining up to date implementations of certain methods: IPCC, ILCD and ecological scarcity.  We publish detailed reports on the implementation, available to the users via ecoQuery, in the File section.  The other methods have last been updated in 2010.  If you are looking for climate change scores, the IPCC implementation published by is the most recent (2013), is thoroughly documented and has been reviewed by an external expert.  

Unfortunately, we lack the resources to keep up with the volume of methods.  The software are doing a better job keeping the method implementation up to date.  

The priority of ecoinvent is to provide its users with high quality life cycle inventories.  With more involvement from the method developers, it will be possible for us to publish LCIA results with the most up to date methods.  

Software's situation is different from ecoinvent's.  They deal with multiple database at once.  They sometimes have to rename exchanges, and have to harmonize multiple nomenclature conventions for elementary exchanges, coming from databases and LCIA methods.  You will therefore find characterisation factors in a software's implementations that are not part of ecoinvent.  

If you have more questions about LCIA implementations, you can contact me directly at  

Guillaume Bourgault, project manager for ecoinvent

Written on 18.11.2017 by praneeth.annam see context

 Hello all, 

I am performing uncertainty analysis for - conversion of biomass to p-Xylene and I faced an issue. I used a lognormal distribution with a standard deviation of 2.16 and tried running with 50, 100, and 1000 runs using ReCiPe midpoint method. 

Unfortunately, all these ended up in errors that read - maximum number of iterations reached (one or more loops might have a cumulated circular coefficient equal to or more than 1). 

Could you please help me how to solve this error? Any approaches to fix this issue. 


Any help is highly appreciated. 

Thanks a lot

Written on 04.11.2017 by marcel see context

Dear all

In the frame of an EPD of a paint I would have one doubt about CO2 emissions in the next process "Quicklime, in pieces, loose {CH}/production". Any help will be really welcomed!!!

The case is that in the process "Quicklime, in pieces, loose {CH}/production" it is stated an emission of 1,078 Kg CO2/Kg of quicklime. During the trandformation from CaCO3 to lime CO2 are emitted because of decarbonatation. My question is: is decarbonatation the origin of this 1,078 Kg CO2/Kg of lime or this emissions of CO2 have another origin?

As lime is the base of this paint, in this EPD this is a very important issue...


Thanks in advance!



Written on 30.10.2017 by acaff see context

Hi ecoinvent staff -

I'm bumping this post up, as I noticed a similar discrepancy.

In completing an LCA project (for a wood product), I ran impact assessment methods for two versions of TRACI – one associated with Ecoinvent v3.3 and one associated with openLCA LCIA methods 1.5.6.  The results were drastically different, with the TRACI LCIA 1.5.6 results for GWP being almost 4 times greater than using Ecoinvent v3.3 TRACI.  After digging into the impact assessment methods impact factors, I saw that Ecoinvent TRACI only included 4 flows for carbon dioxide, with no biogenic flows.  The LCIA TRACI includes 30 flows for carbon dioxide, including biogenic carbon. 

I also saw similar differences in results/impact factors between the two CML (baseline) versions (Ecoinvent v3.3 and LCIA 1.5.6). 

I ended up using openLCA LCIA v1.5.6 since its flows were more comprehensive, however I would still like to know why the flows are different between LCIA v1.5.6 and ecoinvent 3.3 if the impact assessment method is the same (TRACI).

After posting in the openLCA forum, I learned that openLCA's LCIA method pack reflects other databases and therefore other elementary flows that are not captured in ecoinvent.  Is ecoinvent working to resolve this?




Written on 17.10.2017 by Avraam.Symeonidis see context

Dear Nigel and Romain,

Regarding Romain comment: Following the recycling path of steel, “iron scrap” is the flow someone should follow in order find the recycling rates. “scrap steel” is rather a low quality waste that has no other fate than incineration or landfill.



Regarding Nigel’s question: Please let me rephrase and apologies for the confusion. Recycling is not excluded, it is cut-off from the producing activity. The recycling activity is then represented by activities "product name, Recycled Content cut-off" which come burden free. Recycling is under represented in ecoinvent, this has been the case since version 2, but it is a place where ecoinvent would like to extend their coverage. Finally, for post-consumer issues, in some cases, we distinguish post-consumer materials; not in the steel case though. In aluminium we do, for example.

Best regards,

Avraam Symeonidis, Data analyst, ecoinvent



Written on 16.10.2017 by Avraam.Symeonidis see context

Dear Praneeth Annam,

Thank you for your thread. Regarding the specific chemicals you are looking you might wish to check “alkyl sulphate (C12-14) production”. Through the synonym tab it corresponds to Sodium Lauryl Sulfate. We unfortunately do not have the second one. However, I would advise you to download the file named “Activity Overview for ecoinvent 3.4, Undefined” which you can directly access from here. There, in tab “IntermediateExchanges” you can look for the specific chemical you wish using their CAS number. On top of that, in tab “activity overview” you can filter column E for an ISIC classification you are interested in. For instance “2023” yields results related to “Manufacture of soap and detergents, cleaning and polishing preparations…..”. By doing this you will be able to find at least proxies that match your case.

Best regards,


Avraam Symeonidis, Data analyst, ecoinvent

Written on 14.10.2017 by praneeth.annam see context


I am a student of Chemical Engineering working on the project - Conversion of Biomass to Chemicals. As the part of my project, I am performing the Life Cycle Assessment for our plant. And I am looking for some data for to the production of Surfactants. (e.g. Sodium Lauryl Sulfate, Sodium Lauryl Ether Sulfate etc.) 

Any help is highly appreciated. 

Many thanks!

Written on 14.10.2017 by Nigel Howard see context

Dear Avraam,

In the Pre Listserv discussion, other matters were raised and i would appreciate your clarifying response on these questions also:

How is post-industrial recycled content treated compared to post-industrial recycled content - increases in post-industrial content imply greater waste and should penalise the product's profile, whereas post-consumer content has served humanity in one or more previous cycles and should benefit a product's profile?  (Post Industrial can usually be effectively modelled by expanding the system boundary).

Why is recycling excluded from the system model?


Nigel Howard

Written on 13.10.2017 by romainsacchi see context

Dear Avraam,

thank you for taking the time to answer.

I seemed to have understod that the recycling rate was mostly defined on the input side by the respective share of steel from eletric arc and oxygen blast furnace.

What drew my attention was rather the way the "end-of-life" was modeled by the "market for steel scrap" dataset. Almost all of it goes to landifll, and a minor share goes to incineration. That was what motivated my question really.


Best regards,


Written on 13.10.2017 by Avraam.Symeonidis see context

Dear Romain Sacchi,


Thank you for our email. Though the topic has already been addressed I hope my reply clarifies your situation a bit further. To begin with, the graphical representation you made envelops methodological inconsistencies that prevent an observer from the whole picture. You compare an ideal (closed) loop system which does not reflect the complexity of reality, with a cut-off system model which is a methodological approach that considers recycling to be a burden free process.
To directly reply your questions, the recycling rate is not 0.5% and by no means you can calculate a recycling rate given the fact "recycling" as a process is excluded due to the system model. If you are interested in the recycling rates of this process please have a look into the UPR (before linking and allocation) datasets of "steel production, electric" and "steel production, converter". The input of recycled material in those datasets is what matters when it comes to recycling rates.
On top of that, in case you wish to dive more into the details of the system models, I would direct you to following page (link) our website and have a look in the explanation we offer related to recycling process in the cut-off system model.


Best regards, 

Avraam Symeonidis, Data analyst, ecoinvent