If you're seeing this message, it means we're having trouble loading external resources on our website.

If you're behind a web filter, please make sure that the domains *.kastatic.org and *.kasandbox.org are unblocked.

Main content

Mass spectrometry

AP.Chem:
SPQ‑1 (EU)
,
SPQ‑1.B (LO)
,
SPQ‑1.B.1 (EK)
In the analytical technique of mass spectrometry, atoms or molecules are ionized using a high-energy electron beam and then separated based on their mass-to-charge ratios (m/z). The results are presented as a mass spectrum, which shows the relative abundances of the ions on the y-axis and their m/z ratios on the x-axis. This data can be used to calculate the exact masses of the atoms or molecules in the sample. Created by Sal Khan.

Want to join the conversation?

  • stelly orange style avatar for user BootesVoidPointer
    Sal says mass spectrometry can be used to measure the abundance of isotopes of a certain element in nature. But when he says in nature does he mean in all of nature or just 'the nature of the sample'? If the former, then how can we make inferences of the abundance of isotopes in nature from a single sample? If the latter, is there any technique that chemists use to infer the abundance of isotopes in all of nature?
    (13 votes)
    Default Khan Academy avatar avatar for user
    • hopper cool style avatar for user Iron Programming
      In statistics, we can go around, take samples, and make an inference from it.

      Whenever we take a random sample in nature & use mass spectrometry, we are finding the percentage that different isotopes occur in a sample of that element.

      If we just take one sample & use the percentages from that one sample, then I would think that isn't the best inference for all of nature, though we can still make a claim for it.

      Though I obviously did not find the relative abundances of the elements myself, here is what I have done.

      Take multiple samples, use mass spectrometry to to find the relative abundances of each isotope, and then write that down.
      Continue this process for n=K samples, & then I can calculate the mean relative abundances for each isotope.

      If K (our number of samples) is small then we wouldn't be as accurate as if we take more samples.

      If anything I said here is incorrect, please someone let me know. I don't have experience using mass spectrometry so I don't actually know the process they used, but from my experience in AP Statistics that is the type of process we would use.

      Hope this helps,
      - Convenient Colleague
      (15 votes)
  • leaf green style avatar for user CavCave
    This is probably way too advanced of a question
    But how do we guarantee that the electrons we shoot will knock off the electrons of the atoms? Electrons are very tiny, aren't we just gonna miss? Also, what if some of the electrons we shoot end up joining the atom to form negative ions?
    (4 votes)
    Default Khan Academy avatar avatar for user
    • leaf red style avatar for user Richard
      When we ionize a sample in MS, we bombard it with quite a lot of electrons. So even if most of the electrons we fire from the electron source miss the target, enough are making contact for us to be able to measure it in MS.

      The electrons that we fire are of a high enough energy (~70 eV) that they ionize a sample by knocking existing electrons from the sample rather than simply add to the neutral species to create anions. In either case though whether we produce positive cations or negative anions, we can still detect them in MS since they are charged. Hope that helps.
      (8 votes)
  • old spice man blue style avatar for user sidharth
    when you says in nature does he mean in all of nature or just 'the nature of the sample'? If the former, then how can we make inferences of the abundance of isotopes in nature from a single sample?
    (4 votes)
    Default Khan Academy avatar avatar for user
    • hopper cool style avatar for user Iron Programming
      In statistics, we can go around, take samples, and make an inference from it.

      Whenever we take a random sample in nature & use mass spectrometry, we are finding the percentage that different isotopes occur in a sample of that element.

      If we just take one sample & use the percentages from that one sample, then I would think that isn't the best inference for all of nature, though we can still make a claim for it.

      Though I obviously did not find the relative abundances of the elements myself, here is what I have done.

      Take multiple samples, use mass spectrometry to to find the relative abundances of each isotope, and then write that down.
      Continue this process for n=K samples, & then I can calculate the mean relative abundances for each isotope.

      If K (our number of samples) is small then we wouldn't be as accurate as if we take more samples.

      If anything I said here is incorrect, please someone let me know. I don't have experience using mass spectrometry so I don't actually know the process they used, but from my experience in AP Statistics that is the type of process we would use.

      Hope this helps,
      - Convenient Colleague
      (4 votes)
  • stelly blue style avatar for user Waleed Dahshan
    How do we know that most of the elements will have the same charge after they get bombarded by electrons? Does the beam knock off every single electron in the atoms, so that they have no electrons? Otherwise, why would the charge be predictable?

    If some atoms end up having a different charge after bombardment by electrons, then the particular isotope that is represented by the mass to charge ratio would be hard to identify on the x axis. Suppose for example the mass is 4 and the charge is 2+, that wouldn't be distinguishable from a mass of 2 and a charge of 1+, right?
    (3 votes)
    Default Khan Academy avatar avatar for user
    • leaf red style avatar for user Richard
      Good question. It requires energy to remove electrons from atoms or molecules in MS. Since it takes more energy to remove more than one electron from your sample particles, most cations produced in the ionization chamber carry a +1 charge. There will be a small number of atoms or molecules with a +2 charge but they will be too small to make much an impact on the results.

      And yes you are correct since the MS only records a mass to charge ratio, m/z, those two ions would be indistinguishable on the spectrum.

      Hope that helps.
      (5 votes)
  • piceratops seedling style avatar for user DangerAlzahrani
    i don't get it, doesn't this just changes the electrons in the sample? where does the neutrons come from? do they attach to each other depending on their charge?
    (3 votes)
    Default Khan Academy avatar avatar for user
  • leaf green style avatar for user Abby
    At , Sal says that the atoms are ionized so that they can pass through an electric plate. He says that the purpose of the electric plates is to accelerate their speed. Why does the speed of the ions need to be accelerated? Another question — why are heavier atoms deflected less, as Sal says at ?
    (1 vote)
    Default Khan Academy avatar avatar for user
    • leaf red style avatar for user Richard
      The whole point of mass spectrometry is to separate ions based on their mass. Ions need to be accelerated or else they wouldn't be able to reach the detector at the end of the spectrometer. For mass spectrometry to work the ions need to enter the magnetic field with velocity.

      When they are accelerated and pass through the magnetic field they are deflected based on their masses. We need this or else we wouldn't be able to sort out ions with the same charge but different masses. Ions with higher masses have higher inertia and therefore require more force to change their direction of motion, and vice versa for ions with lower masses. So if the magnetic field is held constant then ions with a lot of mass and inertia will therefore be slightly deflected, while ions with less mass and inertia will be deflected to a higher degree.

      Hope that helps.
      (4 votes)
  • orange juice squid orange style avatar for user Zeta Sky
    Is there a name for this particular set-up? I've seen other types of mass spectrometer in textbooks and online, and this one has a slightly different process to what I've read about. It doesn't look like TOF either.
    (2 votes)
    Default Khan Academy avatar avatar for user
  • duskpin ultimate style avatar for user Isabel
    If the Mass Spectrometry machine thing ionizes the atoms, how will the number of neutrons be detected? If you're trying to find the atomic mass of the atom, that's the protons+neutrons, but I don't get why shooting off electrons will have anything to do with it.
    (1 vote)
    Default Khan Academy avatar avatar for user
    • leaf red style avatar for user Richard
      In MS we ionize a sample by essentially firing electrons at the sample which has the effect of knocking other electrons bound to the sample particles off. The loss of these electrons creates cations with a 1+ charge most often. When these cations are finally detected at the end of MS, they are detected as a mass-to-charge ratio, or m/z. Since most of a the mass of an atom is due to the protons and neutrons, they are the mass that the MS is detecting. And since the charge is usually 1+ the denominator of the m/z ratio is usually 1, so the m/z ratio is often just the atomic mass, m.

      Hope that helps.
      (3 votes)
  • mr pink red style avatar for user Cara Goodman
    How do you determine the molecular formula from MS? Which peak do you look at?
    (1 vote)
    Default Khan Academy avatar avatar for user
    • leaf red style avatar for user Richard
      Depends on which type of MS you're doing.

      If you're just sending a sample with a single element in it like Sal's example then you use that to find the isotopes and their abundances.

      But if you're doing something in organic chemistry where we send organic compounds with multiple carbons of various chain lengths then we usually look at the M+, M+1, and M+2 peaks. The M+ peak is the largest atomic weight peak referred to as the molecular ion peak. It's essentially the entire molecule, minus a single electron, turning it into what we call a radical cation. This is more or less your molecular weight of your molecule. The M+1 peak would just be the peak with an atomic mass 1 greater than the M+ peak. From the M+1 peak we can usually determine the number of carbons, oxygens, and nitrogens in the molecule using their known isotope abundances. The M+2 peak is similar to the M+1 in that it is 2 atomic mass units greater than the M+ peak. If there's any sulfurs or halogens in your molecule, they'll show up in the M+2 if there is one.

      The rest of the MS data will be the fragments of your molecule cut off and turned into smaller radial cations. These are only really helpful if your organic molecule is a straight carbon chain, otherwise it's not much help.

      Hope that helps.
      (3 votes)
  • blobby green style avatar for user mmoore27
    Why does the sample need to be vaporized?
    (1 vote)
    Default Khan Academy avatar avatar for user
    • leaf red style avatar for user Richard
      So while it's usually more convenient for a sample to be vaporized before it's ionized, the sample doesn't always need to be vaporized into a gas for mass spectrometry to work. The phase of the sample could be a solid or a liquid too. The phase that you want depends on the ionization method you use. Not all mass spectrometers are of the same construction and use the same method; there's a variety of ionization methods. Why certain ionization methods require certain phase would require me to go into the specifics of each method.

      Sal is describing electron ionization as the ionization method here which does require the sample to be in a gas phase and therefore must be vaporized prior. Now certain samples cannot be vaporized directly since they decompose at high temperatures before vaporizing. To get around this we can subject them to a vacuum so only modest amount of heating is required to make it vaporize (since you lower the vapor pressure).

      Hope that helps.
      (3 votes)

Video transcript

- [Instructor] In other videos, we have talked about the idea that, even for a given element, you might have different versions of that element, and we call those different versions isotopes. And each isotope of an element can have a different atomic mass. And that stems from the idea that, if it's a given element, it's going to have the same number of protons, but you could have a different number of neutrons. Now one question that you might have been asking yourself is how have chemists been able to figure out what the various isotopes of an element are and their relative abundance? What percentage of an element that we find in the universe is of isotope A versus, say, isotope B? And the answer to your question is they use a technique known as mass spectrometry. I can never say it right, mass spectrometry. Sometimes you'll hear the word mass spectroscopy, and they're essentially referring to the same idea. And what this technique is, is that you put a little bit of a sample right over here, let's say we're talking about zirconium in this example, and you heat it up. So you have it, you have a bunch of the zirconium floating around, and then you beam it. You will bombard it with a bunch of electrons. And what the electron bombardment does is, it can knock off electrons from the atoms in your sample, and it can ionize them. And by ionizing some of your atoms, they now have charge. And because they have charge, they can be accelerated through these electric plates. So now you have these ions, in this case, of zirconium, moving quite rapidly through this chamber, and then they enter into a magnetic field. And a magnetic field, a strong magnetic field, can bend the path, can deflect ions with charge. For a given charge, the force of the deflection will be the same. But if you have a larger mass, you're going to be deflected less. And if you have a lower mass, you're going to be deflected more. And so what you see here are the different isotopes being deflected different amounts as they go through the magnetic field. And then you have the detector. And at different points of the detector, you will detect each of these isotopes. And the more ions that hit a certain part of the detector, that means that, hey, I have more of that type of isotope in nature. And then from that, you can generate a chart that looks like this, where you see, on the horizontal axis, sometimes you'll see it labeled atomic mass. And here, it's in unified atomic mass units. And you can see, when you put the zirconium through the mass spectrometer like this, you get a little bit that has a mass number of 96, you have a little bit more that gets a mass number of 94, 92, 91, and most of the zirconium, over 50%, has a mass number of 90. Now in other cases, you won't see it just in terms of atomic mass, given in unified atomic mass units. Sometimes in this horizontal axis, they'll give it in terms of mass-to-charge ratio, where mass is the mass, but then charge is essentially the charge of the ions. Now in a case where your charge is one, for example, if you knock one electron off of the atoms and you have a plus-one charge, well, then the mass-to-charge ratio would be the same thing as atomic mass measured in unified atomic mass units. If your ions have a different charge, well, then you would have to make the appropriate adjustment. But in introductory chemistry class, most of the time you will get things in terms of just straight-up atomic mass. If you happen to get something in terms of mass to charge, just make sure that if the charge is, say, plus two, that you make the appropriate adjustment for the masses. But this right over here will tell you the various isotopes, and it will tell you its abundance. And it all comes from this process of ionizing those atoms, speeding them up, deflecting them through a magnetic field. And the ions that have a higher mass-to-charge ratio will be deflected less, and the ions that have a lower mass-to-charge ratio will be deflected more. And you can use that information to make a graph like this.