Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

If you scratch a piece of metal, does it weigh more?

  1. Aug 4, 2010 #1
    Say you have a 1kg piece of metal (1kg EXACTLY) at an arbitrary, constant temperature. You make a scratch in it with a needle, and you do it carefully enough that you don't rub off any atoms off the metal onto the needle, that way you still have 1.00000000000000000000000000... kg of metal in your block. Then you wait until the block returns to its original temperature described above (EXACT same temperature).

    Does the block weight more now?
  2. jcsd
  3. Aug 4, 2010 #2
    yes, it has the same mass. If you scratch something without removing any of the metal, then you have basically just changed its shape. You can change the shape of something all you want, and still have it way the same. You could even break the thing in half if you wanted to and it would weigh the same.
  4. Aug 4, 2010 #3
    Assuming you've not lost any of the needle either..

    well, you've changed the distribution of stresses (the store of elastic PE) in the body. So you may have technically altered its mass, one way or another (even when returned to the same temperature). And since you asked about weight, I guess you've also altered how it is exposed to the slightly non-uniform gravitational field..

    What are you trying to get at?

    Is it a riddle? The only piece of metal that could weigh exactly 1kg is in practice going to be the kilogram standard, which will remain exactly 1kg even if you take a bite out of it.. :smile:
    Last edited: Aug 4, 2010
  5. Aug 4, 2010 #4
    It is equivalent to the hard drive question.

    And no, assume the material is "soft" enough so that the stresses are the same and none of the needle gets on the block of metal.
  6. Aug 4, 2010 #5
    What hard drive question?

    Are you asking about information encoded in the arrangement of the metal? (In that case, I'm going to venture the mass hasn't changed, or at least that in principle the mass doesn't have to change, especially not incrementing consistently..)
    Last edited: Aug 4, 2010
  7. Aug 4, 2010 #6


    User Avatar
    Gold Member

    Don't forget about the bonds between atoms in the metal. Breaking the bonds alters the energy... thus the mass is altered.
  8. Aug 4, 2010 #7
    So, like the hard drive, it could weigh more or less and it would be very unlikely it weighs the same.
  9. Aug 4, 2010 #8
    The total surface area increased, so it might contain more oxydation. Many, if not most metals oxydate at the surface, at least slightly. The extra mass is oxygen atoms, but the quantity of metal with metallic properties decreased.

    If it's an oxygen free atmosphere, then you might want to consider the slight amount of radioactivity everythings gives off over time.

    Electrons can also fly in and out, although its feasable that rates are the same.
  10. Aug 4, 2010 #9
    I think all you want is for it to be conceded that in principle your system has multiple distinguishable states with the same total mass (energy). Don't sweat the specifics.

    Then, you can enumerate these states, and encode information by the shifting the system to particular such states (and without requiring net work, at least not if you knew which of the states it was already in). So you can't simply tell by weighing whether it encodes King Lear, or a blank string, or random bits.
    Last edited: Aug 4, 2010
  11. Aug 4, 2010 #10
    If you scratch the surface of aluminum, this will expose bare aluminum under the protective oxide layer, which will then oxidize and add weight to the 1 Kg piece of metal.

    Bob S
  12. Aug 4, 2010 #11
    omg why you all have to come in and grenade the thread with your oxidation BS. I never said the block was in presence of oxygen and it can be a metal which is non reactive. There are the exact same number of molecules in the block before and after, and all molecules are still of the same metal. And no, there were no bonds broken, assume the metal is soft and the scratch doesn't kill anything.

    The "information" in the scratch can equate to a large number, possibly 10^16 digits or so, or more. That information cannot be used to decrease entropy in any system or subsystem, therefore its not "energy".
  13. Aug 4, 2010 #12
    By scratching, you have caused stress in the structure of the metal. So the enery state of the whole piece is different. I suppose this is what you call information. It's now in state, say "10101", whereas it used to be "00000". (each bit-digit might represent the next highest energy state of 5 atoms on the scratch).

    As for the weight, what I can remind you of is that a vertical bar of metal weighs less than the same bar if its lying, because its center of mass is now further from the earth's. Considering the piece now has a different geometry, you might want to take this into account.

    As for the mass, if the number of particles doesn't change, why should the mass? But I don't think this is your question.
  14. Aug 4, 2010 #13
    Clarifying the above : I am assuming the bar of metal is resting on a table when I compare vertical and horizontal. My point is that a change in geometry shifts the center of mass. And the weight depends on distance of the centers of mass of the two attracting objects.
  15. Aug 5, 2010 #14
    Don't sweat it, the reason I made this thread is to logically prove that "adding information" to a system does not mean that its energy is increased.
  16. Aug 5, 2010 #15
    IMO, adding information can either add or remove energy, or neither.
  17. Aug 6, 2010 #16
    IMO in this case it is neither since the "information" is irrelevant. Just like some energy is "useless" and cannot do work, some information is "useless" (thermodynamically speaking) and cannot decrease entropy.
  18. Aug 6, 2010 #17
    How do you scratch metal without breaking bonds? I think you are trying to create an ideal system but have instead created an impossible system.

    Scratching does break and form bonds, this does change the mass. Just like in the hard drive rearranging the small magnets changes the energy which does change the mass. There is no need to appeal to entropy to show that yes, the mass does change (either higher or lower).
  19. Aug 6, 2010 #18
    It's conceivable that for each broken bond, there is a new one, so that there is no change in ENERGY.

    Why do you say changing the energy changes the mass? Which theory is this claim based on?
  20. Aug 6, 2010 #19


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    It depends on what you scratch. If you scratch F= ma, the result will be different than if you scratch, say, =Fma

  21. Aug 6, 2010 #20
    And the probability for that happening exactly with billions of bonds is very, very slim. Its overwhelmingly more likley that the energy after is different than the energy before.

    Relativity, bro! Its only the most famous equation in all of science... :p
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook