just a selkie in the sea

(I also go by Liz)

avatar by @PotechiPon on twitter


vogon
@vogon

and for a while I've been wondering -- especially in light of my local natural food store's comical attempt to banish plastic packaging from their beverage aisle which they then had to walk back a couple weeks later -- is plastic actually worse than other forms of beverage packaging, in terms of carbon footprint?

I was working on doing the research myself but in the middle of writing the post I found this LCA that was done by Franklin Associates/ERG (a sustainability consultancy) on behalf of -- caveat emptor -- the PET beverage container industry a few months ago:

in its defense it seems to use basically all the same sources I was digging up on my own and is a lot more thorough than I could ever be1. the numbers it comes up with:

  • 1,000 gallons of soda worth of PET bottles, at prevailing recycling rates (29.1% recycling rate, new bottles made out of 10% recycled content), produce between 296 kg CO2e (2-liter bottles) and 623 kg CO2e (16.9 fl oz bottles) of global warming potential depending on bottle size;
  • 1kgal worth of 12oz aluminum cans at the prevailing recycling rates (50.4% recycling rate, 62.3% recycled content) produce 1,241 kg CO2e;
  • 1kgal worth of 16oz cans produce 969 kg CO2e;
  • 1kgal worth of 12oz glass bottles at the prevailing recycling rates (39.6% recycling rate, 38% recycled content) produce 2,608 kg CO2e.

glass and aluminum also do (somewhere between a little and a lot) worse on energy and water consumption, solid waste production, production of nitrogen and sulfur oxides and ozone. the only place where they win is on ozone depletion, where aluminum and glass produce ~milligrams of CFC-11 equivalent per 1,000 gallons, instead of ~centigrams.

the main drivers seem to be:

  • while manufacturing glass packaging produces less carbon dioxide per kilogram of packaging material, a glass bottle weighs an order of magnitude more than either an aluminum can or plastic bottle of the same size;
  • manufacturing finished aluminum cans from aluminum ingot consumes almost 3 times as much energy as manufacturing plastic bottles from PET pellets.

anyway, the real solution here is obviously for americans to get over our phobia of reusing drink containers and establish glass bottle collection, sterilization, and reuse programs. but it's wild that the contemporary hippie wisdom is actively harmful, to the tune of a potential fourfold increase in carbon footprint.


  1. I care a lot about soda bottles. I do not care this much about soda bottles.


vogon
@vogon

imagine a world where the only thing that's different from Earth Prime is that instead of 12 fluid ounce glass bottles, soda comes in 2-liter glass growlers; besides this world being slightly more whimsical than our own, the decrease in the amount of dead weight per kilogram of packaged soda alone decreases the carbon footprint of glass by 700 kg CO2e/kgal -- about 25%.

thanks to @JhoiraArtificer for weighing a bunch of glass bottles she had around her kitchen for the original version of this post


JhoiraArtificer
@JhoiraArtificer

what am I supposed to say to "can I get you to do some science for me"? no???

bonus:
official @vogon science rating: thorough (it was only 9 bottles!)
discord message from vogon reading "that's more than enough thank you!"


vogon: hey can I get you to do some science for me
Liz: most likely
vogon: you've got a kitchen scale, right
Liz: Yeah
It hates 15 grams tho1
vogon: do you have any glass beverage containers? specifically thinking of soda bottles and beer growlers
Liz: Full or empty? I've got empty growlers for sure
vogon: empty
Liz: Yeah
vogon: could you weigh them for me and give me the weight and nominal volume


  1. it really does, it almost always skips straight from 14 to 16 grams


You must log in to comment.

in reply to @vogon's post:

thanks! can't stress enough that this would've been a lot easier if I had found the NAPCOR LCA before spending hours going cross-eyed at all of its primary sources, but hey, at least I didn't have to spend a bunch of extra time crunching numbers on top of that

I wonder how feasible it'd be to do the whole reuse and sterilization thing with tougher plastic bottles instead of glass. Even if they're not as infinitely reusable as glass, the weight savings might make it worthwhile. I don't have high hopes for any sort of bottle-reuse program becoming widespread but I like the idea. Might write it into one of my more optimistic near-future sci-fi settings as a little worldbuilding detail, if I ever get back to that project.

you would need a sterilization method that doesn't use heat, microwaves, UV, or a solvent that ends up attacking it on the microplastics level, bc all of those cause microplastic shedding and or other forms of PET degradation, unfortunately. even with home reuse that Dasani bottle only lasts like 10 days of hard use

Locally here, one of the major milk brands does glass bottles you return to the grocery store to be washed and reused. I think about how there used to be pop brands that had the same model and, you know - imagine how much less waste we could manage if that came back in a sustainable way?

it would be such a relief, even reusing a glass bottle 3 or 4 times instead of having to turn it into glass cullet and recast a bottle in between every time the glass gets used would more than make up for the increase in material usage

glass reuse programs have to be pretty optimized in order to beat plastic/metal in terms of carbon footprint, though my numbers are from Germany where return rates for all drink containers are pretty high because of bottle deposits – but we also have a lot of drinks (primarily beer and fizzy water) that share the same bottles between brands so sometimes distances can be kept pretty low and the numbers for reusable glass start looking way better

Yeah, I was wondering about how this would apply in Germany (because I like glass containers). But the impact of weight seems like it makes sense. However, I think the CO₂ cost of the increased weight seems like it would be incredibly hard to measure.

fwiw this LCA estimates that transporting 1,000 gallons of filled glass containers from the bottler to the store emits about 34kg of CO2; a refill lifecycle would presumably constitute two of those trips, plus the 69.8kg(!) emitted from making a new steel bottle cap and the 235kg(!!) emitted from making new paperboard bottle multipacks, plus whatever emissions are created by the process of cleaning the containers for reuse.

so without any work on the packaging you could cut the lifecycle emissions from 2600kg to (say) 400kg, but you could cut that in half again by making the multipacks durable and reusable, and maybe cut another third off the top of that by switching to a durable bottle closure.

😌😌😌😌😌

for what it's worth all of our beers are also in glass bottles, though they're still sold in paperboard multipacks and most states don't have bottle deposits (the states that do (unsurprisingly!) have a recycling rate nearly twice the others.) plastic bottles are almost entirely a soft drink thing, with the exception of some beer bottles at sporting events, which are plastic so you can't use them as a weapon

hell of a country

We're not that different, though sporting events usually sell beer off the tap in plastic cups (there's stadiums with beer pipelines here which is a weird thing to say). I think if we didn't have the deposit system for years today you'd stir up a lot of controversy trying to introduce one, despite the obvious benefits. I personally really like the side-effect of not having to fill up my recycling with bottles and cans

looking at the source, their data for glass bottles seems to be of lesser quality than that for PET and aluminum containers. The transport distance is much higher, and there are no details on the glass bottle manufacturing process (there are for PET and aluminum cans).

in reply to @vogon's post:

The thing that makes me a little skeptical here is the recycling rates, especially as vary by locale, but I guess living in the southwest and being burned by "Basically, none of our plastic is recycled, but some of our aluminum is".

Where I live now, the local estimated capture rate is like 85% for paper, 62% for plastics, and 50% for metal though so maybe this is more reasonable.

Of course all of this pales compared to plentiful places offering cold brew growlers in bougie AZ suburbs

yep! you're right about local recycling yields; this LCA attributes 352-396kg CO2e to aluminum raw material manufacturing, compared to 173-348kg for PET raw material. extrapolating up to the carbon footprint of purely virgin PET from the 10% recycled content they use here, PET's raw material footprint increases to ~184-369kg CO2e, and extrapolating down to purely recycled aluminum from the 62% recycled content they use here, aluminum's raw material footprint decreases to ~58-65kg CO2e. this would make aluminum's footprint somewhat better (680kg for 16-ounce cans, 890kg for 12-ounce cans) and plastic's slightly worse (307kg for 2L bottles, 647kg for 16.9-oz bottles).

but unfortunately, a can and a fresh aluminum ingot take the same amount of energy to melt, cast into sheet, form, and fold into a can again -- about 435kg CO2e worth, compared to the 85-175kg CO2e required to injection mold a PET preform and blow a bottle out of it -- and that more than makes up for any advantage it might have. the specific context that I abandoned my original post in, in fact, was trying to figure out whether the footprint of forming raw aluminum and plastic into finished containers was a negligible energy expenditure compared to raw material extraction -- that's not true for either plastic or aluminum but in aluminum's case it's more than 60% of the entire lifecycle emissions!

I should also point out though that these numbers are contingent upon the carbon footprint of the electricity used (here, they're using Average American Electricity) -- extracting the raw material for and manufacturing aluminum cans uses 13-15GJ of energy per 1,000 gallons, whereas plastic bottles use 6-12GJ of energy -- so aluminum's footprint will also tend to decrease preferentially to plastic's as the average joule of American electricity gets greener.

Yeah, this is about what I was expecting when you start to fiddle with the assumptions / data on current use patterns!

I think that I've also been focused more on the material reuse than the energy costs, headed into a future with greener sources for electricity. That just feels like a more solvable problem compared to the proliferation of microplastics and reliance (for now) on petrochemical original material.

as a fan of soda and packaging materials and also finding more eco friendly ways to do things, I think that relying entirely on carbon emissions as a score to determine Which Packaging Method Is the WINNER paints a very incomplete and somewhat misleading picture. if anything, it's one small piece of a huge ecology puzzle in regards to material usage for packaging material. but yeah glass reuse programs could do wonders for soda (and beer! and wine! etc) packaging sustainability

totally! but even though CO2 emissions were the thing that prompted me to think about this, the LCA also points out (and I reiterated) that other forms of packaging result in, all up, more solid waste being thrown away, more sulfur oxides and low-level ozone being emitted, and more water and electricity being consumed as well.

So you're saying that, like almost like every other feel-good superficially green personal consumptive choice, the actual results are somewhere between dubious and incredibly counterproductive?

Greening electricity generation at a societal scale is like step 0 in making any appreciable difference but instead we've apparently decided to just burn more coal to manufacture new Teslas and new reusable shopping totebags.

yep, sure is! (one counterfactual I wish this report had in it is how the numbers work out if electricity is 100% renewable. 15-20% of the cumulative energy demand for aluminum and glass is renewable, whereas almost zero for plastic is, since the energy content of the plastic itself dominates the total energy usage.)

Very interesting! Container reuse seems to be the only sensible option here.

I completely buy the efficiency argument but, I always thought the main issue with plastic containers and other single-use plastics was not carbon footprint but that they’re so hard to get rid of? (Eg microplastics, giant garbage heaps floating around in the ocean etc).

yep! though the LCA also shows that at contemporary American recycling rates less solid waste ends up in landfill over the plastic container lifecycle than with either aluminum or glass as well -- at least by mass; obviously, plastic waste is less dense than aluminum or glass, but assuming all of the waste is discarded containers, PET ends up generating roughly 0.11 m^3 of landfill per 1kgal of packaged beverage, aluminum roughly 0.14 m^3, and glass 0.67 m^3.