Hello,
I know this topic has been covered recently in the Pré LCA forum but looking at the published datasets it seems no definitive and comprehensive position has been taken.
It is indicated in the quality guidelines that the reference flow for a chemical production transforming activity dataset shall refer to 1 kg of the chemical without out water, i.e. in dry state. This seems clear enough but does it mean that the water in the solution, for example for the reference flow "sodium hydroxide, without water, in 50% solution state", should be included as an input for the transforming activity dataset? This does not seem to always be the case. If not, then should this water be included in the transforming activity dataset using this chemical and how should it be entered (e.g. water from environment, tap water, deionised water)?
Thank you,
Jean-François Ménard
Analyst
CIRAIG - Interuniversity Research Centre for the Life Cycle of Products, Processes and Services
www.ciraig.org
Hi,
In the v3, the chemical products are always expressed in 100% active substance.
In your example, 1kg of the product "sodium hydroxide, without water, in 50% solution state" refers to 1kg pure NaOH. The 50% solution is merely given as a hint of the most frequent solution state the chemical might be found.
If in your DS what you need is an input of a solution of NaOH, then you should add an appropriated input of water.
Some DS in the v2 were not consistent with this approach, and some chemicals were already produced as solutions, leading to subsequent errors in using activities. For v3.0, we have corrected all the inconsistencies that we had identified. Please contact us if you identifiy any remaining one.
Kind regards,
Emilia Moreno Ruiz
Hello Emilia,
Thank you for the clarification, but could you also propose a first/best guess as to the type of water that should be added to "create" a solution for a DS that requires such a solution as an input?
Thank you, kind regards,
Jean-Francois
Thanks Jean-François for asking this question.
Nevertheless, it is not possible to give a general adivice on the type of water that should be added to generate a chemical solution. NaOH for example is available in normal solution or ultrapure – depending on the use type.
Usually it can be assumed that depending on the use, deionised to ultrapure is used in a European/North American context, and likely also globally, but here uncertainty might be bigger.
Regards,
Emilia
Hello,
I'm looking for the inventory of KOH pellets.
In Ecoinvent is included KOH but there are not specification about purity. According with a simplify potassium balance based on the inventory data (0.48 kg of KCl expresed as K2O per kg of product) it would be around 57 wt%; very low respect to comertial products >= 85 wt% .
Thanks in advance,
Bernabé
Dear Bernabé,
thank you for pointing this out. The dataset you are referring to (potassium hydroxide production) was originally created in v2.2. According to the report "Life Cycle Inventories of Bioenergy" (Jungbluth et al., 2007), this product is commercialized as 90% purity. Looking more closely at the description, the final product contains 0.144 kg of water per kg of product, thus corresponding to 85.6% purity.
Given the level of purity in the final product, the amount of KCl input is too low. The report states that 823 kg of KCl is required for the production of one metric tonne of caustic potash, which is also too low (maybe the amount was actually referring to the mass of K).
Anyway, since the release of version 3 of the database, the inventories of chemical products are supposed to be given for 1 kg of pure chemical product (100% active substance). Therefore, we will scale up the amounts of the inputs and outputs, and at the same time correct the amount of KCl input. The new input amount of KCl will be calculated based on stoichiometry, taking into account the emissions of KCl and KOH, as well as the KCl and KaOH contained in the wastewater.Moreover, the outputs of chlorine and hydrogen will be added (they had been cut off in version 2, which was in line with the system model implemented at the time).
The new version of this dataset will be made available in the next release of the database.
Best regards,David FitzGerald, Data Analyst at ecoinvent
Hello,
I would like to follow up on the question of JF Ménard regarding concentration in datasets of inorganic acids and salts such as NaOH.
I understand that the datasets are expressed in 100% active substance. But is the evaporation of water to reach these 100% included in the inventory?
If we take the DS "Sodium hydroxide, without water, in 50% solution state {RER}| chlor-alkali electrolysis, mercury cell | Cut-off, U", the only energy mentioned is electricity (for the electrolysis); there is no consideration of the heat or steam used to evaporate the remaining water, in order to reach 100% NaOH from the 50% solution obtained after electrolysis.
So, it means that the inventory actually considers 1 kg of NaOH and 1 (hidden) kg of water. If so, adding the missing water to reach the desired NaOH concentration, as you recommend to do, means that the water mass balance will not be correct anymore.
So, can you confirm whether 1) either the datasets model 100% NaOH after evaporation of water (and in the case, evaporation energy should be added in the DS in the next ecoinvent update), or 2) it only considers the energy of electrolysis leading to 1kg of 50% NaOH and multiplies it by 2 to get 1kg of NaOH (and 1 kg "hidden" remaining water).
Thanks in advance for your clarification.
Victoire Goust, independent LCA consultant.
Dear Victoire,
No, the datasets do not model the evaporation of water to reach 100% of active substance, as this is not what is happening during the production. The inventory corresponds to the production of the chemical product in the actual concentration that it is produced in, and then it is scaled up such that there is 1 kg of pure substance on the output side. So, the alternative 2) is how the datasets are modelled.
Additionally, the datasets that follow this convention do not have "hidden water" in the output. This is due to the fact that any water on the input side that ends up in the solution is subtracted from the inventory after scaling up to 1 kg of pure active substance. So the datasets should in principle be balanced for water, however, as Emilia mentioned above, some datasets were not originally created following this approach, so there may be some that are not balanced for water.
Consequently, whenever you use a chemical product as input in your model, you need to also add an input of water to account for the water that is in the solution. The small challenge here is that the input of water should not come from where the chemical product is consumed, but from the location where the chemical is produced. If you are not sure what type of water to use as input, you may check the producing activities to see which water input is used there. For the NaOH example (produced in the chlor-alkali electrolysis datasets), the water comes directly from the environment. I would however recommend using a water input that comes from the technosphere (e.g. deionised or ultrapure, as recommended by Emilia above), as these also include the treatment process.
Best regards,
David FitzGerald
Data Analyst, ecoinvent
Dear David,
Thanks for your clear and precise answer!
Victoire