Average valuables content in e-Waste

Gold Refining Forum

Help Support Gold Refining Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

JoeyJoystick

Well-known member
Joined
May 25, 2019
Messages
53
Location
Ban Chang, Rayong, Thailand
Hi All,

Like most newbies investigating the recycling of e-waste to extract precious metal content I also went looking for some numbers as to how much PMs and PGMs and Copper e-Waste contains.

This is not an easy task. Not because there is no information available, quite the opposite actually. And the numbers that you find vary a lot. It seems to me that there are several reasons for the wide variety in the available numbers. The differences between the various processes, the kind of feed stock and the feed stock sample size are but a few of the variables causing these numbers to vary.

And then I came across some numbers from Umicore in Belgium in a research paper from 2014. I am guessing that many of you have heard of this plant, but for those who have not here I go. The Umicore plant in Belgium is (was?) the largest e-Waste recycling plant in the world. They have invested heavily in both the research and the building of the plant. It is as environmentally friendly as you can reasonably get with todays technology and the emissions are well below local and European standards.

Out of the 16 metals which Umicore produces I will highlight just a few of them, though arguably the most interesting and valuable ones. Here are the numbers: 250.000 metric tons of e-Waste is processed per year. The production rate on a yearly basis is 50 tons of PGMs, 100 tons of Gold and 2400 tons of Silver. At today's market prices this translates to well over 7 Billion (USD), and with a feed of 250.000 tons annually this translates to something like 28.000USD per ton. What makes the number a good average is the 250.000 tons feedstock because this is not a sample anymore.

Additionally, this does not include any of the other metals which they extract, most notably copper which is a significant part of e-Waste. If indeed the copper content is higher than 10% as often reported, this would easily add an additional 140.000.000USD to the tally.

p.s. I have been conservative with the prices I took, just to be on the safe side. I took 5.5USD/kg for Copper, 1.400USD/Oz for Gold, 0.5USD/g for Silver and 1.000USD/Oz for PGMs.

I guess those numbers make even the die-hard pros on this forum look like amateurs. Absolutely no insult intended, because the bottom line is that I truly envy you guys.

Anyways, I thought it was interesting and just wanted to share this with you.


Joey
 
Joey

Thanks for the OP, however your logic isn't based upon complete data. From their released figures you cannot make the link between 250,000 tonnes and $28,000 per tonne.

There are holes in that logic that I could drive a Sherman tank through. Sideways.

Don't work on that number.

Jon
 
anachronism said:
Joey

Thanks for the OP, however your logic isn't based upon complete data. From their released figures you cannot make the link between 250,000 tonnes and $28,000 per tonne.

There are holes in that logic that I could drive a Sherman tank through. Sideways.

Don't work on that number.

Jon

Hi Jon,

Thanks for the feed-back. I would not work with these numbers for several reasons. The main reason being, that I would work with the lowest reliable numbers I can find and it would be plain stupid to base a small operation on the numbers of the largest operation in the world.

I need to go to work now but I will investigate a little further once I have returned back to my hotel. But can you give me an indicator as to where I have gone off track. You made me curious.

Cheers,


Joey
 
Sure. Where you've gone wrong is taking the headline figures and applied a direct average across them.

In reality the range is enormous, from $2,000 per tonne upwards. Also these figures were produced in 2014 when the PM and PGM content of the product being refined included all the older equipment which is only available now in very small parcels.
 
we have this saying:

the average temperature of a human body in the hospital is 36.6 Co - including the morgue :)
 
Hi Jon,

Thanks.

I was very aware of the 2.000USD/Ton and upwards, but that is why it was an average. What did puzzle me was why the average was so high. But you explained that very well with 'older equipment' comment. But I figured after your earlier post that there must be more to it.

And so I went to look into this a little further as I had indicated in my earlier post. Here is where I went wrong. Yes, the numbers are ok actually. However, I was a little naïve to assume that they only recycle e-waste. The also accept waste from various other industries. Think of jewellers waste, Catalyst (Cars), and they even process ores from a silver mine they own/run in Brazil. They actually opened a plant for catalyst only, just a few kilometres form where I live in Thailand. But that is not applicable to the mentioned numbers. Those numbers are Belgium only.

oh, and by the way. The production has been increased and they are about to expand again now to a total of 500.000tons for this particular plant.

Anyway, my intentions were good, but the picture is clearly wrong. Very wrong. In part of what you mentioned of course, but also various other reasons.

Still, it's quite interesting that people are doing this on such a scale now and recovering so much metal from all our waste. 13 plants worldwide.

I remember I did some inspection work in a PGM refiner in Belgium some 15 years ago. I do not remember though if this was Umicore. It was quite an experience though. Just getting in there is more difficult than getting into a bank after closing time. :lol: And getting out is even more difficult. Nice to see 5 reactors on a row producing gold, platinum, silver and the likes. We did a Video Inspection of the reactor internals and some associated piping. But details are vague after such a long time.

Joey
 
I would say that if you could break down the scrap into both age and source you could get a better average but then it comes down to the fact that a specific separation as would be required would probably cost more than the recovery.

It reminds me of an experience I had 15 years ago in New York City. I was approached to do assys on "sewer sludge" which was actually solids settled in storm drains and periodically cleaned out by the city. The first sample I received ran in excess of 15 ounces per ton. (466 PPM) The guy thought he would get rich quick and I cautioned him to get a larger sample. He ended up with a sample from a drain further down the block which was a fraction of the first drain. These drains were on 47th street, the heart of the jewelry district. I did 4 samples from the street and the average was under 0.15 ounces per ton. (4.6 PPM)

He went from a (self assumed) rich man, to owing me for the assays pretty quickly. In the end he gave me all of the sludge he had, (dried) which amounted to about 10 kg in payment and it all came from the highest yielding drain which was assayed first. Total gold content was about 1/3 of a gram. That wouldn't even cover the in weight fee's at a smelter. So I got skunked as the guy never came back, but this did serve to point out to me once again the value of sampling properly.

And proper sampling makes the difference between a successful operation and a mediocre operation.
 
4metals said:
Proper sampling makes the difference between a successful operation and a mediocre operation.

Hi 4metals,

I fully agree with you here. I think most people with comparatively small businesses do exactly this. Hence you see many 'small' operation which specialise in a particular kind of waste.

My thinking was that on a large scale it would be un-economically to go to deep into the separation of different kind of wastes. As pointed out in the previous post, I had obviously not taken into account that this company did much much more than just electronic waste though...

Though far from claiming to be an expert on this, I find it hard to believe that a large operation, such as Umicore, will go to separation details to the extend of separating processors from 8-24 pin DIL housing chips. Or even separating chips from capacitors. I actually do not even believe that they will separate cell phones from computer boards and telecom boards and low grade boards such Washing machine and TV boards. Again, I might be wrong, but I really think that all this is treated in one big heap of waste. When working on a small scale this would not be viable because of the equipment required to treat a mixed bag of waste, hence the reason people running small scale operations separate and specialize. The first separation (maybe collection is a better word in this context) starts with the purchase of the waste. The large operators still separate though, but in a much grander scheme. As in main categories such as jewellers waste, electronic waste, ores and automotive catalyst, to name just a few.

Anyway, these were my initial thoughts. both you and Jon made it clear I needed more info though. So what does Umicore do? They get there waste delivered in pre-sorted loads. They have suppliers which Umicore refers to as 'pre-processors'. [edited for grammar] And they will get delivered batches of cell phones, computer boards, automotive catalyst etc. And they, Umicore, still do assays them selves on some or all of there wastes.

I guess at the end of the day it all comes down to economics.

I still think it's amazing that with a relatively low investment they can create such a massive turn over. That, of course, is only half the picture, because the operating costs for them are considerable. Still for their 'green' plant they have invested several hundred million and they have created a turnover of billions. Which is a bit unusual. The high operating costs reduce the profits considerable though, but they still end up with hundreds of millions of profit. I do not know of many industries where you can (not saying the do though) write of your investment in less than a year. And all that makes it an impressive operation. I think few will doubt this.

It also shows that, all my good intentions aside, how easy it is to go wrong with numbers. Not so much the mathematics (in this case), but more what the numbers being used actually represent. And this is exactly what you were saying 4metals.

Lesson well learned on my side I guess.

Joey
 
JoeyJoystick said:
4metals said:
Proper sampling makes the difference between a successful operation and a mediocre operation.


Anyway, these were my initial thoughts. both you and Jon made it clear I needed more info though. So what does Umicore do? They get there waste delivered in pre-sorted loads. They have suppliers which are Umicore refers to as 'pre-processors'. And they will get delivered batches of cell phones, computer boards, automotive catalyst etc. And they, Umicore, still do assays them selves on some or all of there wastes.

Proper sampling, as 4metals calls it, is how they know how much to pay you.

You send a 40,000 lb load to them in a container. They dump it out and push it in to a pile with front loaders. Then they take a scoop out of it with the front loader. That is your sample. They burn that separate, process it, determine how much gold, silver, palladium and copper you have...and they apply that fraction to the entire 40,000 lb load.

It's not quite that simple, but that's the idea.

It's in your best interest to have a homogenous sample...for a lot of reasons.

Short of someone who has been sending lots to a large smelter, you won't find this information in a public setting...because it's never about how much gold is in your lot, it's about how much gold is in your sample. Thee sampling that 4metals describes is what defines how well metals in your lot correlate to the metals in your sample. It's not a quantitative payout.
 
snoman701 said:
You send a 40,000 lb load to them in a container. They dump it out and push it in to a pile with front loaders. Then they take a scoop out of it with the front loader. That is your sample. They burn that separate, process it, determine how much gold, silver, palladium and copper you have...and they apply that fraction to the entire 40,000 lb load.

snowman

per the underlined - that's not how they take the sample from your load

The load is first shredded - it is then incinerated - it is then run in a mill like a ball mill - then run through a mixer to get it well mixed - they then pull their sample's from that mix for their assay's - the assays then determine the pay out on the load

Kurt
 
That’s just me relaying what I was told by someone that has been there.

It doesn’t much matter how they sample...it’s going to be proven out by umicore to be in umicore’s favor.

And it’s only through multiple loads and multivariate statistical analysis that you start to figure out what’s in your various grades, and just how good your guys are at sorting.

And by the time people or firms have data back on multiple shipping container loads....they aren’t going to want to share that data for the purpose of academics....nor is umicore going to offer that data up to anyone that doesn’t NEED to know it.

I don’t personally care how much gold is in small socket boards, or power boards or .......I care about the per pound price. I hope that the guy I sell to is making an absolute killing....and I mean that...because his risk is a heck of a lot higher than mine. I buy right so I can sell and cover my costs to grow. Until I am able to source 500 lbs of boards per business day.....that’s all I can hope to do.

For some products....I’m the end of the line. For others....I’m just a middle man...and I want both sides of the transaction to be happy with what they get. When I started this Lou told me very specifically, don’t try and do it all....pick a couple things and get good at it. Need to crawl before you can walk.






Sent from my iPhone using Tapatalk
 
I find it hard to believe that a large operation, such as Umicore, will go to separation details to the extend of separating processors from 8-24 pin DIL housing chips. Or even separating chips from capacitors. I actually do not even believe that they will separate cell phones from computer boards and telecom boards and low grade boards such Washing machine and TV boards.

I did not say that any company, Umicore or any other e-scrap processing refiner will do the separation. I was just saying that if you, the owner of the scrap, wanted to quantify different classes of material, my suggestion is what you have to do.

And no decent refiner will not just do as snowman suggests!
You send a 40,000 lb load to them in a container. They dump it out and push it in to a pile with front loaders. Then they take a scoop out of it with the front loader. That is your sample.

A refiner is also concerned that a sample is representative so it is in their best interest to get a good sample, not to maximize your return but to make sure they don't over pay you.

Along time ago when I was new to this forum, I posted about sampling on this thread;
https://goldrefiningforum.com/phpBB3/viewtopic.php?f=72&t=3166&hilit=stream
If you scroll down to the section I described stream sampling and the pitfalls. That is what is done if the customer wants a sample. If you have a 20 ton load as snowman suggests it may be processed as a single load without sampling and you will be paid on out-turn. That is never an option I would take. I am a show me kind of guy and I made money in refining by refining what fit my niche and shipping the rest, but I never shipped a lot I couldn't quantify first unless I was going with it.
 
Damn, recently there's an influx of good "e-scrap business learning" threads here, I'm guilty of finding that much more interesting than chemistry ;)

snoman701 said:
You send a 40,000 lb load to them in a container. They dump it out and push it in to a pile with front loaders. Then they take a scoop out of it with the front loader. That is your sample. They burn that separate, process it, determine how much gold, silver, palladium and copper you have...and they apply that fraction to the entire 40,000 lb load.
kurtak said:
The load is first shredded - it is then incinerated - it is then run in a mill like a ball mill - then run through a mixer to get it well mixed - they then pull their sample's from that mix for their assay's - the assays then determine the pay out on the load

As I was recently explained by an industry professional It goes something like that:
Whole lot is shredded, then a small fraction, say 1/10 gets randomly taken (hopefully not just by a guy with a showel, but for example by rotary splitter) and fed to another, smaller shredder, then 1/10 of that gets randomly taken and ball/hammer milled, and so on until you have a finely divided sample which then is fire assayed, I did not discuss it, but it was my understanding that there is a preferred size of few kg for this fire assay, so my guess for the sample size by which you get your pay-out is not a fixed % but rather fixed weight, so If you sent in 3000 kg and your fire assay sample was 3 kg, your lot is represented by 0.1%, but if you sent in 30 000 kg, then your lot is represented just by 0.01%.... I guess if you are sending 30 tons at a time you could witness treatment and sampling yourself or rather hire a professional witness, as sampling is a science on itself, but still - I have a dreadful feeling that during all of the treatment and sampling process there is a lot of ways it could go wrong - improper treatment, mixing of granulate before taking sample, heavier fractions gravity separating due to vibration, etc. I'm cynical by nature so I have a suspicion that whole sampling process is designed by a refiner in such a way that all of the potential random problems to their disadvantage are covered extensively but, any potential deviations to the other side are not that carefully managed... after all they get "all of it" in the end... so that brings up 2 points:
1.
4metals said:
If you have a 20 ton load as snowman suggests it may be processed as a single load without sampling and you will be paid on out-turn. That is never an option I would take. I am a show me kind of guy and I made money in refining by refining what fit my niche and shipping the rest, but I never shipped a lot I couldn't quantify first unless I was going with it.
You are hands down most knowledgeable (or at least most willing to share) person here on technical details while dealing with large refineries, could you clarify this post for me please;
A)Are you saying you would not take an option of pay-out on final yield data after all of the refining is done? Say, they shredded, burned, refined JUST YOUR 20 ton PCB load, got 2 kg gold, 15 kg silver, few Toz of palladium and a couple of tons of copper and pay out you a % of a spot price for that amount, minus their charges? Why would you not take such a deal (heard of any refineries offering such a deal?), wouldn't it take out all of the randomness of the sample selection? Is it because you would not trust their "final" numbers and sampling process is easier for you to oversee(witness)? I'm worried that it takes a few PhD of theoretical knowledge and a few lifetimes of practice to confidently witness the treatment, sampling and assay with the limited access the refinery allows you, and even when taking separate samples for yourself and umpire, you are already limited by any shortcomings of treatment pre-sampling and everything that comes before the final samples are selected...
B)Or are you saying that you would not trust any refiner to send them material that you are not 100% sure on composition?? Why trust their treatment and sampling process but not their recovery and refining? I mean if they are honest, they are honest...

If it were me, an option of "final yield settlement" seems much more attractive, heck, let them smelt a dore bar, then sample & pay out on that, it should be much more easier to properly sample 100 kg of copper dore (and sample slag for traces of values) than a ton of boards.... seems to me that whole system (current business practices) is purposely designed to the advantage of the refiner, I mean you have a little leeway, a little sense of due diligence, but not really, just enough to lull your sense of awareness.... there was this saying here on who does what last ;/

And that brings me to my second point:
2.
snoman701 said:
I don’t personally care how much gold is in small socket boards, or power boards or .......I care about the per pound price. I hope that the guy I sell to is making an absolute killing....and I mean that...because his risk is a heck of a lot higher than mine. I buy right so I can sell and cover my costs to grow. Until I am able to source 500 lbs of boards per business day.....that’s all I can hope to do.
For some products....I’m the end of the line. For others....I’m just a middle man...and I want both sides of the transaction to be happy with what they get. When I started this Lou told me very specifically, don’t try and do it all....pick a couple things and get good at it. Need to crawl before you can walk.
Sent from my iPhone using Tapatalk

Teach me master! Again and again you bring up points about e-waste business I've found to make perfect sense, I want to learn from you :)
 
niks neims said:
Teach me master! Again and again you bring up points about e-waste business I've found to make perfect sense, I want to learn from you :)

My knowledge is based upon what I've been able to garner from industry professionals...some reliable, some not so much. As an example of not so much, see above. I put a lot more weight in 4metals experience as his shared experiences have never let me down.

For me, there's not much to it beyond recognizing different grades of material.
 
I'm on final yield settlement terms.

It's a tough basis to get from a refinery in many cases. That given I can separate a batch into smaller discreet segments with no additional processing charges and this helps to get yields on particular types of boards. Which in turn allows purchase prices to be developed from hard data rather than blending and guessing (as I call it.)

The other benefit is that we can put a load together of varying products, and given the weight of each type we can get to within a few percent of the out turn before it is even refined.
 
4metals said:
Before the material is put into a ball mill, check to see that there are only balls in the mill. Have them install the dump grate and spin the mill to make sure no left over material is in the mill. Add your material to the mill and let it crush for the required time, usually 1 hour.

Why is that?

Is there something that can be added that's going to throw off your sampling of the fines later?

And I know that this was not necessarily in reference to e-scrap.
 
Back
Top