1. Not finding help here? Sign up for a free 30min tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Taylor series expansion of an exponential generates Hermite

  1. May 6, 2015 #1
    1. The problem statement, all variables and given/known data
    "Show that the Hermite polynomials generated in the Taylor series expansion
    e(2ξt - t2) = ∑(Hn(ξ)/n!)tn (starting from n=0 to ∞)
    are the same as generated in 7.58*."

    2. Relevant equations

    *7.58 is an equation in the book "Introductory Quantum Mechanics" by Richard Liboff.
    (ξ-(∂/∂ξ))n × e-(ξ2/2) = Hn(ξ) × e-(ξ2/2)

    ∑ƒn(a)/n! × (x-a)n

    3. The attempt at a solution

    To check if I am doing things correctly, I chose n=2 and according to the book I should get

    A2(4ξ2 – 2)e-(ξ2)/2

    where A2 is a normalization constant.

    I am told to Taylor Expand e(2ξt - t2)
    Now the Right Hand Side tells me that Hn is a function of ξ so I believe I am supposed to to apply Hn to e(2ξt-t2) with respect to ξ.

    So from the equation 7.58 from the book, if I choose n=2 then I get H2
    (ξ-(∂/∂ξ))2 which equals2 - ξ(∂/∂ξ) - (∂/∂ξ)(ξ) - (∂2/∂ξ2)) so if I now perform the operation
    [(ξ2 - ξ(∂/∂ξ) - (∂/∂ξ)(ξ) - (∂2/∂ξ2)) × e(2ξt - t2)](t2/2!)

    I should get A2(4ξ2 – 2) × e-(ξ2)/2 but with e(2ξt - t2) in place of e-(ξ2)/2 correct? Well I do not.

    Instead I get

    2e(2ξt - t2) - ξ(2t)e(2ξt - t2) - e(2ξt - t2) - ξ(2t)e(2ξt - t2) + (2t)2e(2ξt - t2)) × t2/2

    I can factor out e(2ξt - t2) but it doesn't do anything that would lead me to an answer. I mean, it's obvious this is incorrect since I have the variable t but it doesn't show up anywhere in the table that the Taylor expansion is supposed to correspond to.

    I have looked up videos and checked textbooks for performing a Taylor expansion but they just show me how to evaluate a polynomial at a point a on the function ƒ(x) but I am explicitly given the Taylor expansion to have tn instead of (t-a)n so I don't think I should just pick a random point.

    I also tried using (t-(∂/∂t))n to see if maybe I was supposed to infer a change of variable to t instead of ξ but that just ended in a huge long equation that did not seem to simplify.

    Is my approach wrong? Please let me know.

    Also, please, please, please, please.....do not be rude. To say I love physics is a complete understatement. I have been and will continue to put in the work to learn as much as possible. I do not like to take shortcuts, but my calculus knowledge is lacking because I didn't realize that physics was what I wanted to do until a little bit later in life. So I ask anyone who may help me to not treat me like someone that doesn't value the process of learning by saying something along the lines of "Just look at the definition! Did you even read the book??" or one that I have seen here many times "why would you even do that?!" I would not like to get berated by those whom I automatically respect because of their knowledge of physics.

    Thank you in advance.
  2. jcsd
  3. May 7, 2015 #2

    Ray Vickson

    User Avatar
    Science Advisor
    Homework Helper

    Last edited: May 8, 2015
  4. May 7, 2015 #3


    User Avatar
    Staff Emeritus
    Science Advisor
    Homework Helper
    Education Advisor

    If you set n=2 in 7.58, you get
    $$\left(\xi - \frac{\partial}{\partial \xi}\right)^2 e^{-\xi^2/2} = H_2(\zeta)e^{-\xi^2/2}.$$ Calculate out the left hand side and compare it to the righthand size to identify what ##H_2(\xi)## is equal to.

    I'm not sure what you mean when you say "apply ##H_n## to … with respect to ##\xi## , but I suspect it's the reason for your confusion.

    You're being asked to expand ##f(t) = e^{2\xi t-t^2}## in a Taylor series about t=0. Don't let the presence of ##\xi## derail you. Just treat it like a constant while finding the expansion.

    I suspect your lack of confidence in your math skills is causing you to focus on and read more into some comments than is really there. Students are urged to look up definitions and to consult their textbooks, not to berate them, but because it's what they should be doing at a minimum. As hard as it may be to believe, many students don't realize that these are things they can and should do on their own.
  5. May 27, 2015 #4
    Ah ok I got it!

    I must have been making mistake when taking numerous derivatives.

    Thank both of you.

    As for the comment I had made about being berated, I think that because I am forced to type this out, as opposed to speaking, it is difficult to imitate the rude way that some tutors speak to other students. Sure there are ways to misinterpret what people say but some words used by tutors leave no other interpretation.

    A tutor could have said "did you even read the book?" but instead wrote a paragraph about how students want to take shortcuts and use wikipedia and/or not put in effort. The tutor even expressed frustration at "this generation that doesn't appreciate the learning process". Now if I were a student that was putting in some effort, and I hit a wall and turn to a community for help, it doesn't help (it almost chips away at the student's confidence) for someone to assume that the effort you put in is so pitiful that the only conclusion a person can make is that you must have just glanced at wikipedia.
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted

Similar Discussions: Taylor series expansion of an exponential generates Hermite
  1. Taylor series expansions (Replies: 12)