Embodied carbon of energy technologies


#1

Hi all, does anyone have a good source for the embodied carbon associated with energy technologies (PV, fossil fuel generators, battery storage, etc.)? E.g. Life cycle assessment databases?

Emissions associated with fuel use is easy to come by in the literature, but for embodied carbon, I can only find the odd questionable source, like this.

Thanks!


#2

Hi @brynpickering The main LCA database is ecoinvent. Mostly stocked with publicly funded research but nonetheless proprietary. Perhaps someone with access can run a query for you. HTH


#3

You could also try mining:

“Understanding future emissions from low-carbon power systems by integration of life-cycle assessment and integrated energy modelling”

https://doi.org/10.1038/s41560-017-0032-9


#4

Hey Bryan,

this Paper by Jens Peters and Marcel Weil evaluates a lot of the Life-Cycle Inventory data that recent studies used, traced them back to primary data sets (such as Ecoinvent) and makes a suggestion for a common basis for further LCAs of battery technologies:

https://doi.org/10.1016/j.jclepro.2017.10.016

A good overview on LCAs of “sustainable energy technologies” but already a little bit older is provided by Singh et al. (2013) here:

Otherwise, I would have also suggested Michaja’s paper referred to by Tom and a predecessor by the same people from PIK and some people from NTNU:

https://doi.org/10.1016/j.envsoft.2017.09.010

More on the database site, I only know projects that have not been updated recently:
http://www.needs-project.org/needswebdb/search.php
https://www.nrel.gov/lci/

If you’re interested in digging deeper into the topic, it may make sense that you / your chair set up your own node in the Life Cycle Data Network maintained by the JRC
http://eplca.jrc.ec.europa.eu/LCDN/

I can provide you with Ecoinvent datasets if you need some :slight_smile:

Cheers
Niklas


#5

Thanks all for the links! It seems that LCA work doesn’t really disentangle embodied vs operational emissions/energy, which makes it difficult to use for optimisation (where both are sets of decision variables).

Saying that, the supplementary information given in https://doi.org/10.1016/j.envsoft.2017.09.010 does break it down pretty well. But, it requires further digging to go from e.g. “freight transport (tkm)”, “Iron (t)”, and “cement (t)” to relevant CO2eq emissions.

I’ll drop you an email about Ecoinvent datasets @accon, thanks for the offer!


#6

Indeed the way datasets are defined in ecoinvent is a bit messy, and it is not straight forward to differentiate embodied and operational burden. There may be ways around it (i.e. derive a classification from isic codes) but it is not straight forward.

I also suggest to have a look to 10.1016/j.spc.2018.07.001.