The Origin Forum
File Exchange
Try Origin for Free
The Origin Forum
Home | Profile | Register | Active Topics | Members | Search | FAQ | Send File to Tech support
Username:
Password:
Save Password
Forgot your Password? | Admin Options

 All Forums
 Origin Forum
 Origin Forum
 Problems with Fitting Multiple Lorentzians
 New Topic  Reply to Topic
 Printer Friendly
Author Previous Topic Topic Next Topic Lock Topic Edit Topic Delete Topic New Topic Reply to Topic

Christoph Frommen

1 Posts

Posted - 03/16/1999 :  10:04:00 PM  Show Profile  Edit Topic  Reply with Quote  View user's IP address  Delete Topic
I am trying to use Origin 4.1 for fitting my Mössbauer spectra. The experimental lines are all Lorentzians but the fit does not look too good. For some lines the fitted amplitudes are less than the experimental amplitudes. I have supplied the following sample file so that you can check out what I am talking about. The dataset should be fitted as 6 Lorentzians with independent linewidths. The x-values correspond to velocities (mm/s) and the y-values are counts registered by the detector.
Does anybody have an idea why the fit looks bad???

By the way, I can fit the data with our Mössbauer fitting software without any problem...it is just that Origin is much easier to use than our old program and I would like to see it work.
Thanks!

-6.4815 532478
-6.4295 533355
-6.3776 533304
-6.3257 532478
-6.2738 533355
-6.2219 533304
-6.1699 533109
-6.1180 531406
-6.0661 532614
-6.0142 533089
-5.9623 533371
-5.9103 533943
-5.8584 532071
-5.8065 530363
-5.7546 530986
-5.7027 529874
-5.6507 529306
-5.5988 527441
-5.5469 523823
-5.4950 518847
-5.4430 510573
-5.3911 499325
-5.3392 487472
-5.2873 486880
-5.2354 497405
-5.1834 510164
-5.1315 517765
-5.0796 523106
-5.0277 527321
-4.9758 527900
-4.9238 527804
-4.8719 529550
-4.8200 529719
-4.7681 530379
-4.7162 531261
-4.6642 531971
-4.6123 532050
-4.5604 532107
-4.5085 533123
-4.4566 532368
-4.4046 532302
-4.3527 531707
-4.3008 532154
-4.2489 533795
-4.1969 531388
-4.1450 533345
-4.0931 532337
-4.0412 533191
-3.9893 532510
-3.9373 532461
-3.8854 532426
-3.8335 532001
-3.7816 532193
-3.7297 531459
-3.6777 531707
-3.6258 530619
-3.5739 530295
-3.5220 529527
-3.4701 528857
-3.4181 526953
-3.3662 525733
-3.3143 521305
-3.2624 513590
-3.2104 502978
-3.1585 491197
-3.1066 486772
-3.0547 498477
-3.0028 509798
-2.9508 519281
-2.8989 523039
-2.8470 526298
-2.7951 528381
-2.7432 527197
-2.6912 530517
-2.6393 530075
-2.5874 530240
-2.5355 531628
-2.4836 531907
-2.4316 531457
-2.3797 532710
-2.3278 531605
-2.2759 531286
-2.2239 532429
-2.1720 532536
-2.1201 532365
-2.0682 531594
-2.0163 532517
-1.9643 533422
-1.9124 531592
-1.8605 532895
-1.8086 532264
-1.7567 532137
-1.7047 532512
-1.6528 531440
-1.6009 532398
-1.5490 532243
-1.4971 532615
-1.4451 531972
-1.3932 531792
-1.3413 532777
-1.2894 531099
-1.2374 531452
-1.1855 530798
-1.1336 529742
-1.0817 527183
-1.0298 525187
-0.9778 520049
-0.9259 514977
-0.8740 514308
-0.8221 520623
-0.7702 524268
-0.7182 527702
-0.6663 529496
-0.6144 530563
-0.5625 531041
-0.5106 530903
-0.4586 530205
-0.4067 531058
-0.3548 531546
-0.3029 532484
-0.2509 531923
-0.1990 531693
-0.1471 532589
-0.0952 532010
-0.0433 532622
0.0087 532378
0.0606 531571
0.1125 532060
0.1644 531519
0.2163 532327
0.2683 532589
0.3202 531957
0.3721 531724
0.4240 531069
0.4759 530553
0.5279 530955
0.5798 528640
0.6317 527616
0.6836 524876
0.7355 521198
0.7875 516262
0.8394 513955
0.8913 518366
0.9432 523852
0.9952 525721
1.0471 529568
1.0990 529527
1.1509 531325
1.2028 530938
1.2548 531964
1.3067 531289
1.3586 531301
1.4105 533276
1.4624 532067
1.5144 532480
1.5663 531829
1.6182 533489
1.6701 532112
1.7220 532234
1.7740 532862
1.8259 532728
1.8778 531582
1.9297 531864
1.9817 532309
2.0336 532037
2.0855 532356
2.1374 532933
2.1893 532906
2.2413 533108
2.2932 531550
2.3451 532236
2.3970 532041
2.4489 532215
2.5009 531096
2.5528 530697
2.6047 530992
2.6566 529861
2.7085 528892
2.7605 529382
2.8124 525341
2.8643 522982
2.9162 516946
2.9682 506174
3.0201 493482
3.0720 486848
3.1239 493949
3.1758 504945
3.2278 514865
3.2797 521062
3.3316 524863
3.3835 526294
3.4354 529141
3.4874 530935
3.5393 529060
3.5912 531414
3.6431 529986
3.6950 530699
3.7470 532692
3.7989 532358
3.8508 532103
3.9027 532645
3.9547 533085
4.0066 532482
4.0585 532278
4.1104 532495
4.1623 532843
4.2143 532630
4.2662 532347
4.3181 533503
4.3700 531992
4.4219 532933
4.4739 532899
4.5258 532557
4.5777 532477
4.6296 532067
4.6815 531857
4.7335 532067
4.7854 532123
4.8373 533034
4.8892 530898
4.9412 529555
4.9931 529861
5.0450 528723
5.0969 525989
5.1488 522900
5.2008 516092
5.2527 506631
5.3046 494620
5.3565 485152
5.4084 488261
5.4604 501472
5.5123 511973
5.5642 517938
5.6161 523826
5.6680 525948
5.7200 527273
5.7719 529738
5.8238 530375
5.8757 530906
5.9276 531115
5.9796 532403
6.0315 531303
6.0834 532519
6.1353 533637
6.1873 532995
6.2392 532145
6.2911 531130
6.3430 532826
6.3949 532268
6.4469 532508
6.4988 532371
6.5507 532499
6.6026 532489
6.6545 533786
6.7065 534023
6.7584 533404

rtoomey

USA
184 Posts

Posted - 03/19/1999 :  7:31:00 PM  Show Profile  Edit Reply  Reply with Quote  View user's IP address  Delete Reply


Untitled Document


Christoph Frommen,


I have been able to successfully fit your data. The way I did it was as follows:



  1. Create a new Y column next to the Y column with the original data and fill it with the negative of the original Y values. To do so, click on the new column and select Column:Set Column Values. Enter the line -(col(b)) (assuming the original data is held in a column named 'b') and click the Do It button.

  2. Plot these newly calculated values as a line plot. Notice that the peaks display in the positive direction. This is important since the initial settings in the fitter do not work too well with negative peaks.

  3. Selected Analysis:Fit Multi- peaks:Lorentzian to fit the data. This provides a quick way to get a decent fit, but usually requires some parameter value adjustment to get the fit just right.

  4. To get the fit just right, select Analysis:Nonlinear Curve Fit and alter the parameter values corresponding to the centers of the peaks. These would be the parameters called (xc1, ..., xc6). If you are unsure how to alter them, try using the Data Reader tool on the Tools toolbox to estimate better center values for your peaks.

  5. Fix xc1, ..., xc6 once you are satisfied with their positions. To do so, uncheck the Vary? check boxes next to each.

  6. Perform several more iterations until the fit looks good. To do so, click the 10 Iter. button.

  7. Go back to the worksheet and plot your original Y data.

  8. Go back to the fitter and change the Y dataset assignment to your original Y dataset. To do so, select Action:Fit. If you do not see any menus in the fitter, you are not in the advanced mode. To get there, click the More button. Once you have selected Action:Fit, click on the Y Dep line in the Variables Datasets Fitting Range interface, select the original Y dataset from the Available Datasets list box, and click the Assign button.

  9. Select Action:Fit and add negative signs in front of all the amplitude parameter values (A1, A2, ..., A6).

  10. Click the 10 Iter. button to perform some iterations. You should see a pretty good fit. The red fit line is the actual fit line and the green fit lines are the individual fits of each of the peaks. These can be removed by going into the Layer Properties (double-click on the layer 1 icon in the top left corner of the graph window) and moving them out of the Layer Contents.


This may sound complicated, but it actually isn't that difficult. If I have misunderstood your question, or you have trouble following the steps, feel free to contact Microcal Technical Support for further assistance. You can either call the following number:


(800)969-7720 (press option 2)


or send an e-mail by clicking on the link above. Please be sure to provide the following information:



  • your serial number,

  • the operating system name,

  • the dataset you provide in the Forum, and

  • a step-by-step description of how you perform the fit in Origin (from start to finish).


Thank you for your interest in Origin.


Sincerely,


Ryan Toomey

Microcal Software, Inc.



[This message has been edited by rtoomey (edited 03-19-99).]

Go to Top of Page
  Previous Topic Topic Next Topic Lock Topic Edit Topic Delete Topic New Topic Reply to Topic
 New Topic  Reply to Topic
 Printer Friendly
Jump To:
The Origin Forum © 2020 Originlab Corporation Go To Top Of Page
Snitz Forums 2000