Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.
Picosecond pulsed laser heating problems
Posted 26.03.2015, 06:02 GMT-4 Heat Transfer & Phase Change Version 5.0 6 Replies
Please login with a confirmed email address before reporting spam
Hello,
I am trying to model the laser heating of silicon using a picosecond pulsed laser but have been encountering some problems that I can't seem to solve.
Firstly the model:
I have a 3D rectangular geometry for the silicon, and a 40 um diameter circular beam incident upon the top surface. The laser is modeled as a heat source using
Qin = Q0 * (1-Rc) * Ac * (1/ pi * sigx * sigy) * an1(x,y) * exp(-Ac * abs(z)) * an2(t)
Q0 is the input laser power, Rc the reflection coefficient, Ac the temperature dependent absorption coefficient, sigx & y are the radius of the laser beam in x and y. The exp(-Ac * abs(z)) factor accounts for the exponential decay of the beam intensity within the material as per the Beer-Lambert law.
an1 is a function which describes the Gaussian distribution of the laser beam in x and y
an2 is a function which describes the time varying power input to make it pulsed. Currently an2 is set as a triangular pulse shape of variable pulse duration, and a periodic repetition such that one pulse is incident every 500 ns (2 MHz repetition rate)
I am using the standard heat transfer in solids option to model the heat flow, with all properties taken from the material database on silicon. I am using a time step of 100 ns and running the simulation over 10 us.
The simulation produces believable temperature distributions for pulse durations of 7 ns and above; however when I decrease the pulse duration any further the model shows no heating at all!
I have tried changing the periodic function in an2 as I thought that perhaps, with a shorter pulse duration, the gradient of the change in laser power with respect to time was too high for the software to handle, but that has had no effect on the results. I have also tried decreasing the time step from 100 ns to 50 ns but that has also had no effect on the lack of heating below 7 ns.
Has anyone got any ideas, or successfully modeled picosecond laser pulse heating of silicon and got any ideas as to what might be causing my problems and how to fix them?
Thanks,
Matt
I am trying to model the laser heating of silicon using a picosecond pulsed laser but have been encountering some problems that I can't seem to solve.
Firstly the model:
I have a 3D rectangular geometry for the silicon, and a 40 um diameter circular beam incident upon the top surface. The laser is modeled as a heat source using
Qin = Q0 * (1-Rc) * Ac * (1/ pi * sigx * sigy) * an1(x,y) * exp(-Ac * abs(z)) * an2(t)
Q0 is the input laser power, Rc the reflection coefficient, Ac the temperature dependent absorption coefficient, sigx & y are the radius of the laser beam in x and y. The exp(-Ac * abs(z)) factor accounts for the exponential decay of the beam intensity within the material as per the Beer-Lambert law.
an1 is a function which describes the Gaussian distribution of the laser beam in x and y
an2 is a function which describes the time varying power input to make it pulsed. Currently an2 is set as a triangular pulse shape of variable pulse duration, and a periodic repetition such that one pulse is incident every 500 ns (2 MHz repetition rate)
I am using the standard heat transfer in solids option to model the heat flow, with all properties taken from the material database on silicon. I am using a time step of 100 ns and running the simulation over 10 us.
The simulation produces believable temperature distributions for pulse durations of 7 ns and above; however when I decrease the pulse duration any further the model shows no heating at all!
I have tried changing the periodic function in an2 as I thought that perhaps, with a shorter pulse duration, the gradient of the change in laser power with respect to time was too high for the software to handle, but that has had no effect on the results. I have also tried decreasing the time step from 100 ns to 50 ns but that has also had no effect on the lack of heating below 7 ns.
Has anyone got any ideas, or successfully modeled picosecond laser pulse heating of silicon and got any ideas as to what might be causing my problems and how to fix them?
Thanks,
Matt
6 Replies Last Post 17.06.2016, 17:36 GMT-4
Please login with a confirmed email address before reporting spam
Posted:
10 years ago
Matt,
the time stepping is obviously not resolving the pulses. So I am not surprised you get inconsistent results. The pulses have to be properly sampled by the time stepping, i.e. something like 10 time steps per pulse. You probably have to use strict time stepping.
Cheers
Edgar
--
Edgar J. Kaiser
emPhys Physical Technology
www.emphys.com
the time stepping is obviously not resolving the pulses. So I am not surprised you get inconsistent results. The pulses have to be properly sampled by the time stepping, i.e. something like 10 time steps per pulse. You probably have to use strict time stepping.
Cheers
Edgar
--
Edgar J. Kaiser
emPhys Physical Technology
www.emphys.com
Please login with a confirmed email address before reporting spam
Posted:
10 years ago
Hi Edgar,
Ok thanks, I'll try that, although it is going to take a very long time to run when my pulse duration is of the order of tens of ps! I don't understand why it was still apparently working with a time step much greater than the pulse duration for 7 ns rather than 6 ns though
Matt
Ok thanks, I'll try that, although it is going to take a very long time to run when my pulse duration is of the order of tens of ps! I don't understand why it was still apparently working with a time step much greater than the pulse duration for 7 ns rather than 6 ns though
Matt
Please login with a confirmed email address before reporting spam
Posted:
9 years ago
Merci
Please login with a confirmed email address before reporting spam
Posted:
9 years ago
Hi Matt,
Did Edgar's time stepping solution solve your problem?
Kevin
Did Edgar's time stepping solution solve your problem?
Kevin
Please login with a confirmed email address before reporting spam
Posted:
9 years ago
Hello,
Rather than using Strict Timestepping, use Explicit Events for thermal loads that turn on and off over time. See, for example:
www.comsol.com/blogs/modeling-a-periodic-heat-load/
Also, if you're wanting to model laser light penetrating into a material and heating it, see:
www.comsol.com/blogs/modeling-laser-material-interactions-with-the-beer-lambert-law/
and also more generally:
www.comsol.com/blogs/modeling-laser-material-interactions-in-comsol-multiphysics/
Rather than using Strict Timestepping, use Explicit Events for thermal loads that turn on and off over time. See, for example:
www.comsol.com/blogs/modeling-a-periodic-heat-load/
Also, if you're wanting to model laser light penetrating into a material and heating it, see:
www.comsol.com/blogs/modeling-laser-material-interactions-with-the-beer-lambert-law/
and also more generally:
www.comsol.com/blogs/modeling-laser-material-interactions-in-comsol-multiphysics/
Please login with a confirmed email address before reporting spam
Posted:
8 years ago
Hello Matt,
Thank you very much for your post! I found it very helpful. However, I just have one brief question. I think you were trying to simulate your heating laser as a volumetric element. So did you just use the "Heat Source" function under the Physics of "Heat Transfer in Solids"? Did you have to use the Physics of "General Form PDE" as this blog is saying?
www.comsol.com/blogs/modeling-laser-material-interactions-with-the-beer-lambert-law/
I am trying to simulate a system where a polymer bead sits on a glass slide, and a laser hits both of them. Any suggestions will be greatly appreciated!
Best,
Jeremy
Thank you very much for your post! I found it very helpful. However, I just have one brief question. I think you were trying to simulate your heating laser as a volumetric element. So did you just use the "Heat Source" function under the Physics of "Heat Transfer in Solids"? Did you have to use the Physics of "General Form PDE" as this blog is saying?
www.comsol.com/blogs/modeling-laser-material-interactions-with-the-beer-lambert-law/
I am trying to simulate a system where a polymer bead sits on a glass slide, and a laser hits both of them. Any suggestions will be greatly appreciated!
Best,
Jeremy