'Mass Spectrometry (MS) is an analytical technique that sorts ions based on their mass (or "weight"). Mass spectrometry is used for many chemical analyses, ranging from the analysis of a complex mixture of petroleum, to the products of genetic engineering.'

Is it possible to atomise some garbage by applying a very high voltage and a very high current across it, thereby obtaining high-temperature sparks, and then apply the principles of mass spectrometry, to separate out the individual elements for use in manufacturing?

  • 5
    $\begingroup$ Theoretically yes but it is highly uneconomical. $\endgroup$
    – bon
    Commented Feb 22, 2016 at 15:41
  • 1
    $\begingroup$ The garbage part of question makes it look weird. Besides that, I consider it quite interesting. $\endgroup$
    – ssavec
    Commented Feb 22, 2016 at 16:02
  • $\begingroup$ That's exactly how uranium enrichment worked as well. An even for uranium enrichment, the process was considered to be too expensive, so many effort was spend on developing other (chromatography based) methods that are used today. If it is already too expensive for nuclear weapons, I guess you wouldn't consider using it for waste. Bacteria do a much more efficient and inexpensive job at recycling waste for a long time now. $\endgroup$ Commented May 12, 2017 at 14:34
  • $\begingroup$ So what precisely makes the process so uneconomical? Besides, the target is not uranium, but common lighter elements. Doesn't that save some electricity? $\endgroup$ Commented Sep 18, 2018 at 8:17
  • $\begingroup$ Think of it like this: You'd have to evaporate megatons of material. Does that sound like a clever idea? And actually evaporation is not quite enough. $\endgroup$
    – Karl
    Commented Sep 18, 2018 at 10:01

2 Answers 2


Yes it is possible, but is very expensive and would be orders of magnitude more costly than what people are willing to pay for recycled materials. Let me give two data points to explain why it is expensive.

  • Some commercial material is manufactured by mass spectrometry. The starting material isn't garbage, but reasonably chemically purified metals of various types. The product(s) are isotopically separated metals. Iron 57 ($\ce{^{57}Fe}$) is an isotope that is naturally present in terrestrial iron supplies at an abundance of about 2%. It is useful to researchers that study iron materials using various spectroscopic techniques. Pure iron 57 costs hundreds of dollars per milligram.

  • Mass spectrometry requires very high voltage, a very high vacuum, and large amounts of purified gases; additionally, the throughput of material separated by mass is very low. I think that for commercial isotope purification, magnetic sector-type instruments are used. I'm not too familiar with them, so let me go through some rough numbers for a different type of mass spectrometry: time-of-flight: Several thousand volts are applied to packets of ions which leads to their separation. Total ion currents are as high as say $10^9$ particles per second, which corresponds to about 0.1 nanoamperes or $10^{-15}$ moles per second. So to purify one mole of iron would require $10^{15}$ seconds, or several million years of continuous operation of very expensive machinery! Even if other types of mass spectrometry do better than time-of-flight by three orders of magnitude, it will still be cost-prohibitive.$%edit$

  • 2
    $\begingroup$ The Calutrons at Oak Ridge in WW2 used about 1/7th of the US electricity to mak 10s of kg of enriched uranium. That is a data point to consider. $\endgroup$
    – Jon Custer
    Commented Feb 22, 2016 at 16:09
  • 2
    $\begingroup$ One mole of iron-57 would be 57000 milligrams; assuming \$1000 per milligram, that means you could make a mole of iron-57 for less than \$60,000,000. How does that match the $10^{15}$ machine-seconds needed? $\endgroup$ Commented Feb 22, 2016 at 19:30
  • $\begingroup$ Excellent point immibis. I think that the types of mass spectrometry that I am not familiar with are cheaper than TOF. Even if it is a few orders of magnitude cheaper it still isn't close to economical, unfortunately. $\endgroup$
    – Curt F.
    Commented Feb 22, 2016 at 21:44
  • $\begingroup$ Is this a theoretical limitation, or merely a limitation of human ingenuity? I understand bond breaking takes energy, but has someone performed an in-depth analysis on the energy cycle involved? Why precisely does the process require so much electricity? We can burn the garbage first, which I think will release some heat. The burning should be able to provide part of the energy required. To increase the throughput, can't we simply build a bigger recycling plant? As hydrogen, oxygen etc. are separated out, they can also be used to fuel the process. $\endgroup$ Commented Sep 18, 2018 at 8:46
  • $\begingroup$ @ChongLipPhang, if you burn something, energy is released because you are forming stronger bonds than before. These stronger bonds take more energy to break, so how do you propose to break even? Furthermore, MS typically does not break all bonds, so in any case you will be left with a soup of random (molecular) ions, not a collection of atomic ions nicely sorted by mass. $\endgroup$ Commented Sep 18, 2018 at 8:50

The preparative mass spectrometry is field of active research, trying to overcome many obstacles. One of them is capturing the species (not only atoms, but almost anyting, undestroyed), with the hot candidate being soft landing

Another problem is producing high enough flow of ions and removing the strict vacuum requirements of the traditional MS setup. This resulted in high-flux electrospray ionization source, where they report the deposition rate of $\ce{\approx 1 \mu g/day}$.

You can see that it is usable for preparation of very small amounts of very valuable substances, but definitely not for garbage recycling.

  • $\begingroup$ This is a very interesting answer. +1 ! But the question did say it is focused on elemental separation, not necessarily of intact ions. That said, if the deposition rate for electrospray-based preparative MS is comparable to the deposition rate for high-flux magnetic sector instruments, it goes some way to answering @immibis's comment to my own answer. $\endgroup$
    – Curt F.
    Commented Feb 22, 2016 at 21:46
  • 1
    $\begingroup$ @CurtF. I intentionally ignored the elements, because that renders it more of a mining issue, extraction, separation, etc... Regarding the flux, using the estimate of 1 ug/day, for one mole of iron it is around 150 000 years, so much faster than your estimate, but I'd better postpone the start of project by 10 years, waiting for better technology ;) $\endgroup$
    – ssavec
    Commented Feb 22, 2016 at 22:08
  • $\begingroup$ So far the assumption seems to be that the yield is one of a few valuable substances needed in a small amount. I am not sure if the same can be said about the large-scale elemental separation of garbage. $\endgroup$ Commented Sep 18, 2018 at 8:47

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.