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

Re: carbon steel - Is a crystal the same as a grain?

  1. Mar 21, 2004 #1
    Re: carbon steel -- Is a crystal the same as a grain?

    I'm still a little confused about crystals and grains. Does a crystal become a grain as the metal solidifies?

    I'm also not clear on molecules. For instance with a body centered lattice you have nine atoms. One on each corner and one in the center of the body. So within that lattice -- what determines what a molecule is? Sorry -- I'm sure that's a dumb question.
  2. jcsd
  3. Mar 22, 2004 #2

    Dr Transport

    User Avatar
    Science Advisor
    Gold Member

    In a body-centered crystal, each atom on the corner is only counted as 1/8, or in other words is part of 8 unit cells. Go back to your solid state physics book and reread that chapter.

    As for the grain problem, I do not know. Anyone else?????? I'd also like to know.

  4. Mar 22, 2004 #3
    each atom on the corner is only counted as 1/8

    Sorry -- I've never seen a solid states physics book, much less read one. :smile:

    I know very little about physics -- I'm just searching for a little info regarding a specific area of metallurgy. At first, I didn't understand what you meant by 1/8 -- but after thinking about it -- I remember seeing the illustrations of the corner atom cut to an eight and it finally dawned on me. Then I found this web site, which cleared if up with an animated illustration:
    http://www.kings.edu/~chemlab/vrml/bcc.html [Broken]

    However, I still don't understand what makes up a steel molecule.


    Last edited by a moderator: May 1, 2017
  5. Mar 24, 2004 #4
    392hemi said:

    "....I still don't understand what makes up a steel molecule."

    There's no such thing as a "steel molecule". I'm not sure what the exact definition of a molecule is, but this one should work in almost all cases: For a substance to be made of molecules (a localized group of 2 or more atoms) the vapor phase (or at least the liquid phase) of the substance has to be made of those molecules. This rules out steel because when its heated, the Fe, C and any other atoms that its made of come apart long before they all melt. In fact, it rules out all metals and alloys that I can think of as well as most of the elements (exceptions are O2, N2 etc.).
  6. Mar 26, 2004 #5
    I would say that a grain is a crystal. There are objects that are made up of a single crystal no matter how large they might be. Most crystalline objects however are made up of small crystals joined together at the boundaries. The orientation of each crystal is different. They are called grains because they resemble the grain seen in wood. So whenever a liquid solidifies, if it forms a crystal it is made up of one or more grains.
  7. Mar 26, 2004 #6
    This fits well with my understanding of metallic bonding - the nuclei are surrounded by a 'sea' of electrons so there is no 'molecule' to identify, just a repeating pattern of nuclei sites (the unit cell). Basicallly, the concept of molecule has no meaning for metallic bonding. It's useful for covalent & ionic bonding however. A sodium & a chlorine atom make a 'salt' molecule but two iron atoms don't make a metallic unit cell.
  8. Mar 26, 2004 #7


    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

    http://er6s1.eng.ohio-state.edu/mse/mse205/lectures/chapter3/chap3.pdf [Broken]

    This site has some background on crystal structures, and why these "grains" are only in polycrystalline materials.

    Keep in mind the advise that was given that you do really need to look up a solid state physics text, because the question involved here are the ones covered and answered in a typical solid state physics class.

    Yes, this is a perfectly acceptable model - to the first approximation. Keep in mind that "metallicity" can sometime be a rather vague definition. What you described as your understanding of metals is perfectly valid for the "conventional" metals that obey the straightforward band structure model. However, for metals (or more accurately, conductors) that are actually doped insulators, such picture breaks down severely. In that case, you CAN have "molecules" as the material making up the conductor - the Cu-O molecules in the cuprate superconductors for example.

    Last edited by a moderator: May 1, 2017
  9. Mar 26, 2004 #8
    I'm sure glad I asked.

    I have stacks of books on metalworking, welding and metallurgy. Many of them have mentioned "molecules" when discussing metal. I'm writing a book on hand shaping metal and I wouldn't want to make the same mistake.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook