How Do I Estimate the Number of Injected Holes in a Pulsed Laser Experiment?

[pazmush]

I've just completed an experiment using a laser to excite electons and hole in a germanium sample, and i want to fit a curve to some of the dat I've collected n origin, but i need to set up some initial parameters and i don't know how to work out what the original number of injected holes.

So does anyone know, round about, the number of injected holes from a pulsed laser light?

Or how to do my fit in origin

f(x) = h*(exp(-a*x))*(1/((b*(x+(1/a)))^0.5))*(exp((-(c-(d*x))^2)/(b*(x+(1/a)))))

thanks

 

[Cthugha]

I've just completed an experiment using a laser to excite electons and hole in a germanium sample, and i want to fit a curve to some of the dat I've collected n origin, but i need to set up some initial parameters and i don't know how to work out what the original number of injected holes.

So does anyone know, round about, the number of injected holes from a pulsed laser light?

Well, how large was your spot size and your laser power? Was it bulk germanium or some nanostructure (are you looking for free electrons and holes or excitons)? What was your exciting laser wavelength? What was the pulse duration? Which temperature did you work at? This question is really not trivial.

Or how to do my fit in origin

f(x) = h*(exp(-a*x))*(1/((b*(x+(1/a)))^0.5))*(exp((-(c-(d*x))^2)/(b*(x+(1/a)))))

I assume you have a newer version of Origin (7 or newer).

Plot your data points and go to analysis->nonlinear curve fit->advanced fitting tool. Check, whether the function you need is one of the ones included or choose function->new and define your own function.

 

[pazmush]

ok thanks I've done that but id doesn't seem to fit a curve to my data, it just outputss a curve with the values of my initial parameters, do i have to do some parameter initialization?

 

[Cthugha]

Ok. Did you define your own function?
If so, could you just paste in, how your function is defined exactly, what the names of your parameters are, how many parameters you use and what your dependent and independent variables are?

 

[pazmush]

thanks for this

y = h*(exp(-a*x))*(1/((3.14*b*(x+(1/a)))^0.5))*(exp((-(c-(d*x))^2)/(b*(x+(1/a)))))

independent = x
independent = y
parameters = h,a,b,c,d

h is the the total number of holes injected? = 1E8?
a = 1000?
b = 0.012?
c = 0.005?
d = 75?

i'm using origin 8 by the way

 

[Cthugha]

Looks ok so far. I usually use Origin 7.5, but I hope user defined fitting is still the same.

So how do you try to fit? You switch to Action->Fit, choose a dataset, check vary vor all parameters and click 1 iteration (or 100)?

Sometimes it is sensible in Origin 7.5 to use "edit function in code builder" directly after defining the function and to save and most important compile it there. The user defined fitting sometimes won't work, if you don't compile it this way. Other possible error sources are using "." as the decimal seperator, but having "," set as the predefined decimal seperator or just using the chi sqr button by accident.

 

[pazmush]

can i give you some data and you see if you can fit it please?