Material Flow Analysis (MFA): Metal Demand and Climate Change.
(Note: This post, and many of my earlier posts in this space, contains some graphics which may not be accessible to Chrome users because of a recent upgrade to that browser, but should work in Firefox and Microsoft Edge. When my son has time, he will adjust the file system for a website he's building for me to make these graphics usable in Chrome, but he seldom has that much time on his hands. Interested parties, should they exist, can still read my posts including the graphics, but regrettably must use a browser other than Chrome. Apologies - NNadir)
The paper I'll discuss in this post is this one: Global Metal Use Targets in Line with Climate Goals (Takuma Watari, Keisuke Nansai, Damien Giurco, Kenichi Nakajima, Benjamin McLellan, and Christoph Helbig
Environmental Science & Technology 2020 54 (19), 12476-12483)
Depending on the source and the focus, these periodic tables vary, but the American Chemical Society's version of periodic table of "endangered elements" is here: ACS Endangered Elements Web Page. The ACS lists 44 elements of economic importance for which supply concerns have been identified. Some of those identified as being of most concern are elements in common experience, generally not thought of as being "precious." These elements, in the darkest red color in the table at the site include helium, zinc, and arsenic for example. Others have high technological importance but are less well known by the general public, indium, tellurium, gallium, tellurium and germanium fall into this class.
These lists and tables - and there are many around, some focusing on elements for the defense industry, others on elements for the electronics industry and still others about metallurgy, others relating to catalysts for the chemical industry - are generally based on existing technology and do not generally discuss possible alternatives that might displace the requirements for them.
For example, if you'd asked me, until recently, whether the element Europium should appear on the list, I would have said "yes," but the chief application for europium was until about ten years ago, to make red phosphors in cathode ray tube based color televisions. Following that application, was its use in fluorescent light bulbs. The recent rapid growth in the LED industry however has eliminated much of this residual demand. I had the pleasure of leafing through a paper late last night, in a current issue of the same journal from which the paper under discussion comes, of learning that once having been a "critical" element, stocks of europium are rising (and prices falling) since it is a side product of the lanthanide (rare earth) industry, and demand no longer matches the supply: Byproduct Surplus: Lighting the Depreciative Europium in China’s Rare Earth Boom (Qiao-Chu Wang, Peng Wang, Yang Qiu, Tao Dai, and Wei-Qiang Chen Environmental Science & Technology 2020 54 (22), 14686-14693)
Over the years, I have been spectacularly successful at being around people who are smarter than I am. A few years ago, I attended a lecture by Dr. Yueh-Lin (Lynn) Loo of Princeton University - she is the director of the Andlinger Center for Energy and the Environment - on the interesting subject of conducting polymers for which many possible applications exist, including energy saving windows, flexible solar cells, etc., etc. The lecture was recorded and may be viewed here: Science on Saturday: Plastic Electronics.
Dr. Loo's lecture was quite interesting, and involved some very beautiful organic chemistry, about which I commented in the Q&A. (I wasn’t entirely convinced by her answer to my question, but, again, she’s smarter than I am.) My son - then a high school senior - picked up the fact that electrodes in some of her glasses utilized ITO (indium tin oxide) electrodes, and noted in his question that indium is an element of concern. (That's my boy!) Dr. Loo's response, if I recall correctly was along the lines of "Don't worry about it! We'll find something to replace ITO!" She offered a few possibilities.
Indium tin oxide is widely used because it is a material that can conduct electricity and transmit light; it's transparent and it can easily be layered into very thin films.
Now, Dr. Loo is obviously a highly intelligent woman, a professor and the administrator of a technological think tank at Princeton University, one of the world's great universities. She should be taken seriously when she suggests that alternatives to indium tin oxide exist, but on the other hand, there are many cases where extremely brilliant people have searched for alternatives to materials and failed to find something that works better. (I recently made brief remarks about the difficulty we've had with displacing cobalt in "lithium" batteries: This should go a long way to ending modern day human slavery to mine cobalt in the Congo region)
Suppose, with the world now scraping 420 ppm concentrations of the dangerous fossil fuel waste carbon dioxide in the atmosphere, this after half a century of wild cheering for the "solar energy will save us" meme, that, as brilliant as she is, Dr. Loo is wrong and we never find an alternative that works as well as indium tin oxide as a transparent conducting material. Will thin film devices, in particular organic solar cells and the famous CIGS (copper indium gallium selenide) thin film cells prove to be even more unable to scale as the solar industry in general has been demonstrating for half a century, during which its energy output with respect to the growing use of dangerous fossil fuels has been laughably (or perhaps terribly) trivial?
The rote response to this line of inquiry is to mumble about recycling. But let's think a little more deeply about this issue with respect to indium. The chief technological use for indium is to make touch screens on cell phones and on computer monitors. The operative point about indium tin oxide is that it present as extremely thin films. A typical formulation of ITO will be roughly 75% indium by weight and will be present in very thin layers deposited on the surface of glass using vapor deposition and related techniques.
Consider a "typical" solar cell that is 1.5 square meters in area, for argument's sake a CIGS solar cell. The Andlinger Center has put out a nice surprisingly honest - for a wild eyed institution which seems quite fond of advancing the "solar will save us" meme - discussion of some realities about the solar field: Sunlight to Electricity: Navigating the Field Article 2 gives some key concepts and vocabulary. In particular, it discusses issues that are very seldom discussed in the cheering for the solar industry that one hears, specifically the difference between peak capacity and capacity utilization. As it's linked and open, one can read it for oneself, but I'll summarize: On a clear day, near noon, the solar radiation arrives at the surface of the earth at a power level of 1000 watts per square meter (1500 watts for the area of our putative solar cell). A solar cell is capable of reaching - a good solar cell, and CIGS cells are very good solar cells - capable of conversion efficiency of 20%. Thus we have a peak power generation of 300 Watts for this 1.5 square meter CIGS cell at the peak insolation on a clear day. The Andlinger Center offers some information showing that the capacity utilization of solar cells - the amount of capacity that is actually available as a fraction of the peak theoretical capacity (300 Watts in this case) - of 14%. A day contains 86400 seconds, meaning that the total energy is 0.14 X 1500 Watts X .2 X 86400 seconds per day = 3,628,800 Joules per day, since a Watt is a Joule/second. A kilowatt-hour (kWh) is a unit of energy obtained by generating a power of 1000 Watts for 3600 seconds (1 hour) or 3,600,000 J. Thus our putative solar cell generates about a kWh per clear, sunny day.
A gallon of gasoline is said (depending on the octane level) to have an energy content of about 130,000,000 Joules. This means that it would take our putative solar cell a little over a month of clear sunny days to generate as much energy as is contained in a gallon of gasoline. Let us assume that an installation of 10 solar cells having an area of 1.5 square meters involves – I hope I’m being generous – the consumption, by the workers installing it on the roof of a suburban “green” McMansion involves the consumption of 10 gallons of gasoline. One can say that for a period that is a minimum of a month – longer in a less than perfectly sunny climate, will be lost. Someone will pipe into say this is “only” a short period, and that the famous or infamous “EROEI” will still cover this, without noting that the energy (as things stand now) is in a completely different form, gasoline. The alternatives, electric solar panel trucks powered by batteries, hydrogen, blah, blah, blah remain largely fantasies, much as they have for the last half a century. It is important to note that the means of converting electricity to chemical fuels – a charged battery is as much a chemical fuel as is gasoline or hydrogen – involves energy losses according to the second law of thermodynamics, a law no congress can repeal. So there’s that. And then there is the often ignored issue of time. Then there’s also an economic issues: A putative hydrogen plant that powered by so called “renewable energy” that is operating because the sun isn’t shining and the wind isn’t blowing is a stranded asset, and may have negative value because of depreciation, salaries, maintenance etc. Indeed a battery is a stranded asset whenever it is charging. With this in mind, the 35 days is even more of a minimum figure than the supposition of involving only bright sunny days suggested. Of course, the embodied energy of a solar cell includes many other issues; the gasoline used by the installers is only one item in the list.
It is important to note that when a solar cell (or any other indium containing electronic device) is recycled, all of the environmental costs are pretty much incurred again, as are health costs. "Indium lung disease" is an observed, irreversible and sometimes fatal disease incurred by inhaling indium dusts, very similar in its symptoms and outcome to black lung disease among coal miners.
A fairly sophisticated analysis of indium demand for CIGS type solar cells, which also contain gallium, itself a critical material which is not discussed in any detail in the paper and which I will not discuss further in this post, is given here:
Linking energy scenarios with metal demand modeling–The case of indium in CIGS solar cells (Anna Stamp,?, Patrick A. Wäger, Stefanie Hellweg, Resources, Conservation and Recycling 93 (2014) 156–167).
Similarly, an excellent and detailed overview of total world indium sources is here: The world’s by-product and critical metal resources part III: A global assessment of indium (T.T. Werner Gavin M. Mudd, Simon M. Jowitt, Ore Geology Reviews 86 (2017) 939–956)
Indium has no highly concentrated ores. It is generally obtained as a by-product (as the title of the latter paper suggests), chiefly of zinc, although other metals, including lead and copper, also contain indium. Typical concentrations range from 1 gram per ton to 100 grams per ton of ore, although some ores are known that have higher concentrations. Thus the recovery of indium is energy intensive, with much of the energy incurred in the form of the embodied energy of solvents, extractants, acids, etc. A feel for the quality of ores with respect to indium can be found in the following graphic from the latter paper just cited:
The caption:
It is interesting to note that the current world supply of indium is dominated by Chinese production, using what are apparently low quality ores.
I've discussed indium at some length with a dumb anti-nuke of the "renewables will save us" type who posts on this website, a person who has happily made it to my "ignore list" since ignorant cultists get boring at some point, who liked to pretend that since so called "renewable energy" is so perfect in his little mind the world could not possibly be restrained with respect to indium. (Indium is also a component of wind turbines as well as solar cells, touch screen monitors and cell phones.) Superficially, in the absence of critical thinking, it would seem that there is a lot of indium. The interested reader can glean from the abstract of the paper, that there are 356,000 metric tons of identified indium reserves, as well as another 24,000 tons available from zinc mine tailings around the world for a total of 380,000 metric tons.
The fantasies of "renewables will save us" types notwithstanding, the idea that the world could never run out of indium because so called "renewable energy" is so wonderful, it does appear that people (lots of people if you look) write papers all the time about the concern over world indium supplies. The former of the two papers just cited is just one of those papers, the Resources, Conservation and Recycling paper, discusses something that reflects something called "reality," which is the mass efficiency of recovery processes for recovering indium from ores, which is surely similar to the recovery in the case where a discarded solar cell is the ore, the "recycling scenario." The mass efficiency is shown in this illuminating graphic from that paper:
The caption:
Each of the "residues" can be defined as well as "waste" - for people who believe that the existence of "waste" is acceptable - and, of course, in theory, since atoms, except uranium and thorium, are not destroyed by processing or use, but the issue is whether or not the energy and materials for further capture, which translates into cost, justifies it. That is, in fact, the question with the handwaving about recycling indium from spent solar cells.
It is now worth asking how much indium is required to make solar cells.
This information can be gleaned - even without full access - from the abstract of the Resources, Conservation and Recycling paper, which reads:
This phrase utilizes the common, and seldom challenged, misleading unit of the "GW" - GigaWatt - to describe the scale of this proposed growth in the use of CIGS cell. This is, frankly, the most egregious lie, albeit it common lie, associated with the lexicon associated with "renewables will save us" belief system. Recall from above that the capacity utilization of solar energy is taken by the Andlinger Center - an active participant serving up the "renewables will save us rhetoric - is 14%. This a GW, 1000 MW, of power from solar energy is not the equivalent of a power plant capable of running 24/7 (and in relation to demand). Rather, it is the equivalent of a 140 MW power plant of any kind on average, although the unreliability of these systems, connected not to demand but to weather and season, so much so, that the solar cells might be producing power when the electricity produced by the solar cells (and all other generating capacity) is worthless as a result of oversupply. Conversely the capacity is worthless when the sun is low or absent in the sky when high demand is observed in late afternoon or evening hours, precisely the case where peaks are most frequently observed on most grids.
Using the information in the abstract to see how much energy might be produced in the "best" case, 1401 GW of CIGS solar capacity, and the number of seconds in a sideral year, 86,400 seconds per day X 365.25 days/sideral year, 31,556,736 seconds and the 14% capacity utilization, we can see that the amount of energy produced by 1401 "GW" of CIGS solar cells, "by 2050" would produce 6.2 X 10^(18) Joules of energy, 6.2 ExaJoules (EJ), this on a planet where humanity, as of 2018, was consuming just about 600 EJ per year, or roughly 1% of the world's energy demand, quite possibly at precisely the times the energy is essentially unneeded. If we multiply this number by 100 to pretend that solar energy produced 600 EJ in a sideral year - ignoring the 2nd law energy losses from necessary energy storage - the indium requirement becomes 1,572,400 tons, or three times the world's known supply of the element, excluding all other applications of its applications such as touch screens, cell phones, and alloys.
From the peak power figures given by the Andlinger Center for a 1.5 square meter solar cell, 300 Watts, we can estimate the number of solar cells required to reach 1401 "GW" of solar cells, 1401 X 10^(9) watts/300 watts = 4.6 billion 1.5 square meter solar cells. We can thus estimate how much indium is in each cell: 15,724 X 10^(6) grams/4.6 X 10^(9) solar cells = 3.4 grams per solar cell.
It seems quite reasonable to assume that a spent solar cell is a fairly concentrated "ore" for indium, probably richer, in fact, than most zinc mine tailings. However it differs from zinc mine tailings because the zinc mine tailings all start in one place, the tailing's pile, or residues from zinc refining. To collect the solar cell "ore" trucks have to drive around to McMansions where bourgeois types are removing their solar cells, the solar cells have to be stored in some centralized location until enough accumulate to justify shipping them, then trucked to a port, and finally shipped to some third world country featuring citizens in whose health we are spectacularly disinterested, and have them grind up the solar cells, and reprocess the dust, using oodles of solvents and acids, and considerable energy inputs, to collect the recycled indium.
(The estimated usable lifetime of CIGS solar cells is between 20 to 25 years.)
A description of a putative indium recycling procedure, not for solar cells, but for LCD's, although one should expect the processes would be similar, as solar wastes have very little difference from other forms of electronic wastes, can be found here: Recycling Indium from Scraped Glass of Liquid Crystal Display: Process Optimizing and Mechanism Exploring (Xianlai Zeng, Fang Wang, Xiaofei Sun, and Jinhui Li, ACS Sustainable Chemistry & Engineering 2015 3 (7), 1306-1312) The recovery is not quantitative.
The operative point here is that so called "renewable energy" is metal intensive, and the isolation of metals is neither environmentally nor energetically free. The more mass a system or a set of redundant systems require, the more impact they will have on climate change in particular, and other environmental impacts as well. The low energy to mass ratios of so called "renewable energy" as well as the requirement that many involve elements that are subject to depletion, calls into question whether the word "renewable" is in fact, an honest description of what solar and wind energy are.
This brings me to the paper referenced at the outset of this post. From the introduction:
Technology-rich integrated assessment models are typically used to provide such scenarios by exploring possible technology mixes and their costs.(9?11) However, this approach often fails to reflect the physical interconnection in the series of metal cycles(12) that includes material production, manufacturing, in-use stock, and waste management, resulting in a weak foundation for explaining future demand and scrap availability.(13) Furthermore, existing studies have focused strongly on innovative technology solutions such as carbon capture and storage (CCS)(14) and hydrogen-based production,(15) while metal cycle solutions,(16) including circular economy (CE)(17) strategies, have tended to receive less attention...
For the record, I think that "circular economy" approaches are highly desirable, but, as I noted using references above in the case of indium, they are not all some kind of magical slam dunk.
A little further down in the introduction the authors write:
The authors also consider metal conservation, and focus on the six metals they report as representing 98% of all metals used, iron, aluminum, copper, zinc, and nickel.
They evaluate historical materials flow analysis for these metals from 1900 to 2010, integrating data, they say from 214 countries, to evaluate how future metals use can be consistent with "aligning with the emission pathways of the 2 °C climate goal."
Of course, there is a wide disparity between lifestyles of the rich and the lifestyles of the poor (who end up "recycling" our "stuff" ) about which the authors write:
I trust no one will be surprised with this.
Like the IPCC reports, and the reports of the International Energy Agency's "World Energy Outlook" repots, this paper speaks in terms of "scenarios." Experience teaches that these "scenarios" are always overly optimistic. The degradation of the planetary atmosphere is accelerating, not decelerating or even remaining constant. Even the "BAU" (Business as Usual) scenarios usually prove to be overly optimistic.
Some figures from the text:
The caption:
The caption:
The "CE" scenario is the "circular economy" scenario, with which, there will be, inevitably, in the best cases, material losses as well as energy demands. A "circular economy implies cheap and sustainable energy.
The caption:
The caption:
From the conclusion of the paper:
"If science-based policy instruments work properly..."
Oh well then, good luck with that...
For my money, one of the clearest thinkers on the subject of material flow analysis and energy analysis is Vaclav Smil, whose thinking has influenced me greatly in these areas. Since I choose to be eclectic in what I take, and do not take, from the great thinkers - and Smil is a great thinker - this should not imply that I endorse his libertarian free market rhetoric which one sees in some Czech thinkers who lived under Czechoslovakia's communist government and escaped from it, but in considering energy and mass flows, he is simply one of the best there is.
Recently in another post elsewhere, I cited and quoted from one of his works:
What I see when I see a wind turbine (Numbers Don't Lie) (Vaclav Smil, IEEE Spectrum Volume: 53, Issue: 3, March 2016)
If wind-generated electricity were to supply 25 percent of global demand by 2030 (forecast [pdf] to reach about 30 petawatt-hours), then even with a high average capacity factor of 35 percent, the aggregate installed wind power of about 2.5 terawatts would require roughly 450 million metric tons of steel. And that’s without counting the metal for towers, wires, and transformers for the new high-voltage transmission links that would be needed to connect it all to the grid...
...A 5-MW turbine has three roughly 60-meter-long airfoils, each weighing about 15 metric tons. They have light balsa or foam cores and outer laminations made mostly from glass-fiber-reinforced epoxy or polyester resins. The glass is made by melting silicon dioxide and other mineral oxides in furnaces fired by natural gas. The resins begin with ethylene derived from light hydrocarbons, most commonly the products of naphtha cracking, liquefied petroleum gas, or the ethane in natural gas.
The final fiber-reinforced composite embodies on the order of 170 GJ/t. Therefore, to get 2.5 TW of installed wind power by 2030, we would need an aggregate rotor mass of about 23 million metric tons, incorporating the equivalent of about 90 million metric tons of crude oil. And when all is in place, the entire structure must be waterproofed with resins whose synthesis starts with ethylene. Another required oil product is lubricant, for the turbine gearboxes, which has to be changed periodically during the machine’s two-decade lifetime.
Undoubtedly, a well-sited and well-built wind turbine would generate as much energy as it embodies in less than a year. However, all of it will be in the form of intermittent electricity—while its production, installation, and maintenance remain critically dependent on specific fossil energies. Moreover, for most of these energies—coke for iron-ore smelting, coal and petroleum coke to fuel cement kilns, naphtha and natural gas as feedstock and fuel for the synthesis of plastics and the making of fiberglass, diesel fuel for ships, trucks, and construction machinery, lubricants for gearboxes—we have no nonfossil substitutes that would be readily available on the requisite large commercial scales...
Numbers don't lie. This year, the annual maximum concentration of the dangerous fossil fuel waste carbon dioxide in the planetary atmosphere was reached in May, 417.43 ppm. This was 24.80 ppm higher than the figure ten years earlier, in May of 2009, one of the worst ten year increases in recorded history. Whatever it is we think we're doing about climate change hasn't worked, isn't working and frankly, won't work.
No participant on this website can feel anything but extreme relief that Joe Biden and Kamala Harris will soon hold the reigns of power in government. It is not enough, however, to obtain power; it is necessary to govern well.
In 1862, in a time of divisiveness that dwarfs even our own times, Abraham Lincoln wrote:
That is every bit as true in 2020, 158 years later, as it was in 1862.
One hopes that will be on the agenda and in the back of the mind of the new President, that the "dogmas of the quiet past are inadequate to the stormy present..."
I trust you enjoyed a pleasant and safe Thanksgiving Holiday, and that you will similarly enjoy the upcoming December holidays.
= new reply since forum marked as read
