Re: Strange peaks from grainy Si (data re-attached)
Dear Congwu, Maybe your Mo K-Alpha contribution gets a lot of stronger if you would range the kV settings of your X-ray generator. As a rule of thumb, you need an acceleration voltage that is about 4 x higher than the characteristic energy in keV. In other words, I would advise to run your source at 60 keV instead of 35 keV. Your Mo K-Alpha lines are probably very low compared to the bremsstrahlung. With best regards, Martijn Dr. Martijn J. Fransen Product Manager XRD PANalytical Lelyweg 1 7602 EA Almelo, The Netherlands Phone ++31 546 534 345 Fax ++ 31 546 534 598 PANalytical The Analytical X-ray Company The information contained in this message is confidential and may be legally privileged. The message is intended solely for the addressee(s). If you are not the intended recipient, you are hereby notified that any use, dissemination, or reproduction is strictly prohibited and may be unlawful. If you are not the intended recipient, please contact the sender by return e-mail and destroy all copies of the original message. gregor 01/25/2007 05:06 PM Please respond to rietveld_l@ill.fr To rietveld_l@ill.fr cc Subject Re: Strange peaks from grainy Si (data re-attached) Dear Congwu, as you put it, it definitely seems that you don't get out the MoK radiation from your collimator. Control the alignment and make sure the beam hits your sample. best miguel On 25 Jan 2007 at 9:55, [EMAIL PROTECTED] wrote: Hi, Miguel, the unit is the so-called moment transfer that is calculated as sin(theta)/lambda. When I did the simulation, I use the Zr K-emission as the first wavelength and the Mo-K-alpha as the second, because the spectrum I measured with an AMTREK spectrometer shows that the Zr line is higher than Mo line. Yes, with powdercell, you are able to see the peaks from other wavelength when the width of the calculated profile (# of FWHM) is large enough. the attached figure is another simulation with the parameters of: wavelenth 0.79Å, 0.44Å, I1/I2=0.5, Width of calc. profile: 4000FWHM. The 0.44angstrom can only come from the bremsstrahlung in my experiment if it is true. The problems bothering me are the measured peak is much wider than the simulation and the bremsstrahlung at 0.44angstron is ~1000times weaker than the main line. congwu Quoting [EMAIL PROTECTED]: Congwu, I'm getting confused with the units of your diffraction patterns, could you just indicate 2theta which is probably the primary observation without any assumption about lambda? Also, I've trouble with your simulated Si pattern: there should be no traces of Zr radiation generated peaks, and the absorption edge in the bremsstrahlung would probably be difficult to be detected at the current level of resolution. So I would prefer to see the simulation for plain MoKa radiation and the full peak height of the 111, 220 and 311 reflections. Observed intensities are very low, are you sure you get the Mo radiation out of your collimator? Another problem is how you can manage to calibrate the detector distance using such broad peaks, but I don't know about the purpose of your experiment. BTW, ground Si wafers are fine, just control the grain size under a microscope to make sure there are no grains exceeding 5um or so (relatively easy to achieve with smalll sammple size). The rings actually look a bit grainy. best miguel On 24 Jan 2007 at 15:17, [EMAIL PROTECTED] wrote: I just did an estimation: assuming the unknown peak is the [111] peak, if it is another wavelength, it corresponds to a wavelength of 0.44 angstrom (28keV), consistent to the spectrum of Mo with 35kV voltage. I do not think it could be the contamination of W, because the voltage is still far from the W characteristic line to activate it. probably it is worthy of doing a broad spectrum simulation. does anybody know where I can find the formulas that powdercell uses to calculate the pattern? btw I re-attached the data. congwu Quoting Andrew Payzant [EMAIL PROTECTED]: Peter, That is my guess too either additional (weak) characteristic lines such as Mo-K beta, or contaminant lines such as W L, or, as you point out, some wavelength in the Bremmstrahlung that meets the Bragg condition for some silicon crystallites. However, I would not have expected a few large crystallites to yield a continuous ring, as appears in his CCD image, so perhaps there is another explanation. Andrew -- E. Andrew Payzant Senior RD Staff Member High Temperature Materials Laboratory Oak Ridge National Laboratory 1 Bethel Valley Road PO Box 2008, MS 6064 Building 4515, Room 113 Oak Ridge, TN, 37831-6064 ph: (865) 574-6538 FAX: (865) 574-3940 web: http://html.ornl.gov/dtpgrp/staff/payzant.html
Re: Strange peaks from grainy Si (data re-attached)
Congwu, I'm getting confused with the units of your diffraction patterns, could you just indicate 2theta which is probably the primary observation without any assumption about lambda? Also, I've trouble with your simulated Si pattern: there should be no traces of Zr radiation generated peaks, and the absorption edge in the bremsstrahlung would probably be difficult to be detected at the current level of resolution. So I would prefer to see the simulation for plain MoKa radiation and the full peak height of the 111, 220 and 311 reflections. Observed intensities are very low, are you sure you get the Mo radiation out of your collimator? Another problem is how you can manage to calibrate the detector distance using such broad peaks, but I don't know about the purpose of your experiment. BTW, ground Si wafers are fine, just control the grain size under a microscope to make sure there are no grains exceeding 5um or so (relatively easy to achieve with smalll sammple size). The rings actually look a bit grainy. best miguel On 24 Jan 2007 at 15:17, [EMAIL PROTECTED] wrote: I just did an estimation: assuming the unknown peak is the [111] peak, if it is another wavelength, it corresponds to a wavelength of 0.44 angstrom (28keV), consistent to the spectrum of Mo with 35kV voltage. I do not think it could be the contamination of W, because the voltage is still far from the W characteristic line to activate it. probably it is worthy of doing a broad spectrum simulation. does anybody know where I can find the formulas that powdercell uses to calculate the pattern? btw I re-attached the data. congwu Quoting Andrew Payzant [EMAIL PROTECTED]: Peter, That is my guess too either additional (weak) characteristic lines such as Mo-K beta, or contaminant lines such as W L, or, as you point out, some wavelength in the Bremmstrahlung that meets the Bragg condition for some silicon crystallites. However, I would not have expected a few large crystallites to yield a continuous ring, as appears in his CCD image, so perhaps there is another explanation. Andrew -- E. Andrew Payzant Senior RD Staff Member High Temperature Materials Laboratory Oak Ridge National Laboratory 1 Bethel Valley Road PO Box 2008, MS 6064 Building 4515, Room 113 Oak Ridge, TN, 37831-6064 ph: (865) 574-6538 FAX: (865) 574-3940 web: http://html.ornl.gov/dtpgrp/staff/payzant.html From: [EMAIL PROTECTED] Reply-To: rietveld_l@ill.fr Date: Wed, 24 Jan 2007 12:51:57 -0500 To: rietveld_l@ill.fr Subject: Re: Strange peaks from grainy Si I'll mail you some fine Si powder if you send me your address. My best guess is that you have a relatively big lump of Si in your sample that happens to be lined up to make a bright spot from the bremstrahlung part of the spectrum. It happens to meet some Si Bragg reflection condition for some wavelength in the brems. spectrum. ^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~ Peter W. Stephens Professor, Department of Physics and Astronomy Stony Brook University Stony Brook, NY 11794-3800 fax 631-632-8176 -- Miguel Gregorkiewitz Dip Scienze della Terra, Università via Laterino 8, I-53100 Siena, Europe fon +39'0577'233810 fax 233938 email [EMAIL PROTECTED]
Re: Strange peaks from grainy Si (data re-attached)
... The problems bothering me are the measured peak is much wider than the simulation Peak widths should be 400 microns (your beamsize) divided by 3.5 cm (your distance) in radians, which is about 180*0.4/3.5/PI = 6.5 degrees. HTH, Jon
Re: Strange peaks from grainy Si (data re-attached)
Dear Congwu, as you put it, it definitely seems that you don't get out the MoK radiation from your collimator. Control the alignment and make sure the beam hits your sample. best miguel On 25 Jan 2007 at 9:55, [EMAIL PROTECTED] wrote: Hi, Miguel, the unit is the so-called moment transfer that is calculated as sin(theta)/lambda. When I did the simulation, I use the Zr K-emission as the first wavelength and the Mo-K-alpha as the second, because the spectrum I measured with an AMTREK spectrometer shows that the Zr line is higher than Mo line. Yes, with powdercell, you are able to see the peaks from other wavelength when the width of the calculated profile (# of FWHM) is large enough. the attached figure is another simulation with the parameters of: wavelenth 0.79Å, 0.44Å, I1/I2=0.5, Width of calc. profile: 4000FWHM. The 0.44angstrom can only come from the bremsstrahlung in my experiment if it is true. The problems bothering me are the measured peak is much wider than the simulation and the bremsstrahlung at 0.44angstron is ~1000times weaker than the main line. congwu Quoting [EMAIL PROTECTED]: Congwu, I'm getting confused with the units of your diffraction patterns, could you just indicate 2theta which is probably the primary observation without any assumption about lambda? Also, I've trouble with your simulated Si pattern: there should be no traces of Zr radiation generated peaks, and the absorption edge in the bremsstrahlung would probably be difficult to be detected at the current level of resolution. So I would prefer to see the simulation for plain MoKa radiation and the full peak height of the 111, 220 and 311 reflections. Observed intensities are very low, are you sure you get the Mo radiation out of your collimator? Another problem is how you can manage to calibrate the detector distance using such broad peaks, but I don't know about the purpose of your experiment. BTW, ground Si wafers are fine, just control the grain size under a microscope to make sure there are no grains exceeding 5um or so (relatively easy to achieve with smalll sammple size). The rings actually look a bit grainy. best miguel On 24 Jan 2007 at 15:17, [EMAIL PROTECTED] wrote: I just did an estimation: assuming the unknown peak is the [111] peak, if it is another wavelength, it corresponds to a wavelength of 0.44 angstrom (28keV), consistent to the spectrum of Mo with 35kV voltage. I do not think it could be the contamination of W, because the voltage is still far from the W characteristic line to activate it. probably it is worthy of doing a broad spectrum simulation. does anybody know where I can find the formulas that powdercell uses to calculate the pattern? btw I re-attached the data. congwu Quoting Andrew Payzant [EMAIL PROTECTED]: Peter, That is my guess too either additional (weak) characteristic lines such as Mo-K beta, or contaminant lines such as W L, or, as you point out, some wavelength in the Bremmstrahlung that meets the Bragg condition for some silicon crystallites. However, I would not have expected a few large crystallites to yield a continuous ring, as appears in his CCD image, so perhaps there is another explanation. Andrew -- E. Andrew Payzant Senior RD Staff Member High Temperature Materials Laboratory Oak Ridge National Laboratory 1 Bethel Valley Road PO Box 2008, MS 6064 Building 4515, Room 113 Oak Ridge, TN, 37831-6064 ph: (865) 574-6538 FAX: (865) 574-3940 web: http://html.ornl.gov/dtpgrp/staff/payzant.html From: [EMAIL PROTECTED] Reply-To: rietveld_l@ill.fr Date: Wed, 24 Jan 2007 12:51:57 -0500 To: rietveld_l@ill.fr Subject: Re: Strange peaks from grainy Si I'll mail you some fine Si powder if you send me your address. My best guess is that you have a relatively big lump of Si in your sample that happens to be lined up to make a bright spot from the bremstrahlung part of the spectrum. It happens to meet some Si Bragg reflection condition for some wavelength in the brems. spectrum. ^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~ Peter W. Stephens Professor, Department of Physics and Astronomy Stony Brook University Stony Brook, NY 11794-3800 fax 631-632-8176 -- Miguel Gregorkiewitz Dip Scienze della Terra, Università via Laterino 8, I-53100 Siena, Europe fon +39'0577'233810 fax 233938 email [EMAIL PROTECTED] -- Miguel Gregorkiewitz Dip Scienze della Terra, Università via Laterino 8, I-53100 Siena, Europe fon +39'0577'233810 fax 233938 email [EMAIL PROTECTED]
Re: Strange peaks from grainy Si (data re-attached)
Congwu, If you can move the detector the radius of the silicon rings will change according to the distance and Bragg angles. Moving the sample might be easier for the same check. Does your spurious ring move so that you can trace it back to a point inside the sample? Could it come from the collimator or beamstop? What does the background look like when you remove the silicon sample and just measure the sample holder? If you know the ring only appears when you put the silicon sample then try checking some other materials. Aluminium foil is normally easy to find, either in the kitchen or wrapped around a UHV experiment. Table salt (NaCl) can also be worth a try. If they all give spurious rings coming from the sample and the background is clean then it sounds a bit strange. Good luck, Jon - This mail sent through IMP: http://horde.org/imp/
