The Nanoscale World

Sample roughness and appropriate probe...

rated by 0 users
Answered (Verified) This post has 1 verified answer | 8 Replies | 4 Followers

Top 75 Contributor
11 Posts
Points 125
mariam posted on Sat, Jun 8 2013 11:03 AM

Dear AFM users,

I have samples which have roughnesses from 0.5 to 2 microns and I am interested in obtaining their elastic modulus using the peak force QNM. I tried to use the scanAsyst probe but from the roughness the tip broke before I could get any measurements done...could any one suggest what type of probe that would be effficient for this particular task. thank you.

Mariam

 

 

  • | Post Points: 12

Answered (Verified) Verified Answer

Top 10 Contributor
280 Posts
Points 6,221
Bruker Employee
Verified by mariam

Dear Mariam,

  1. How stiff (ballpark) is your ceramic sample?
  2. When you say the roughness is 0.5-2um what is the scan size that gives you that roughness?

If you want to measure the modulus of a stiff sample, you must use a stiff cantilever.  Question 1 will allow you to look in the manual and find the recommended probe.  However, even with the recommended probe, you may not be able to get an accurate modulus map if the sample has too much roughness.  

The model that we apply to fit the data assumes that the tip is spherical (Hertzian or DMT model), the surface is flat, and the force is applied perpendicular to the surface.  This model may not work for most of the area of your samples, but there may be regions on the sample where the model is still OK.  The model will not work well if the sample surface is locally sloped too steeply or if the side of the tip interacts with the slope (instead of the end of the tip where it is approximately spherical).  

For example, I've recently collected some images of bacteria that are >1um tall, but have a flat area on top.  The sides of the bacteria are too steep and appear to be softer than the top surface which is approximately constant in modulus.  This suggests that the model is failing for the sides, but is OK for the top surface.

--Bede

 

All Replies

Top 25 Contributor
30 Posts
Points 397
Bruker Employee

Dear Mariam,

A 2 micron surface roughness may be quite difficult to image - especially given the height of a typical AFM probe is ~3 microns. There are some probes that have longer tips but what is your sample that you are trying to image - is it soft or hard?

Best,

Andrea 

  • | Post Points: 12
Top 75 Contributor
11 Posts
Points 125
mariam replied on Mon, Jun 10 2013 3:27 PM

Dear Andrea,

 

My samples are ceramic and they are hard I guess that's why ScanAsyst Air probe could not work out.

 

Best,

Mariam

 

  • | Post Points: 12
Top 25 Contributor
30 Posts
Points 397
Bruker Employee

Dear Mariam,

I would think that a ceramic sample is definitely out of the modulus range of a ScanAsyst Air probe. There are other probes available that are better suited for higher modulus samples. However, I am still not sure how well they will do with a surface roughness of ~2 microns. I will pass on your post to one of my collegues who may be able to better help you with this.

Best,

Andrea  

  • | Post Points: 12
Top 75 Contributor
11 Posts
Points 125
mariam replied on Tue, Jun 11 2013 4:16 PM

Dear Andrea,

Thank you for your for the help.

Best,

Mariam

  • | Post Points: 12
Top 10 Contributor
280 Posts
Points 6,221
Bruker Employee
Verified by mariam

Dear Mariam,

  1. How stiff (ballpark) is your ceramic sample?
  2. When you say the roughness is 0.5-2um what is the scan size that gives you that roughness?

If you want to measure the modulus of a stiff sample, you must use a stiff cantilever.  Question 1 will allow you to look in the manual and find the recommended probe.  However, even with the recommended probe, you may not be able to get an accurate modulus map if the sample has too much roughness.  

The model that we apply to fit the data assumes that the tip is spherical (Hertzian or DMT model), the surface is flat, and the force is applied perpendicular to the surface.  This model may not work for most of the area of your samples, but there may be regions on the sample where the model is still OK.  The model will not work well if the sample surface is locally sloped too steeply or if the side of the tip interacts with the slope (instead of the end of the tip where it is approximately spherical).  

For example, I've recently collected some images of bacteria that are >1um tall, but have a flat area on top.  The sides of the bacteria are too steep and appear to be softer than the top surface which is approximately constant in modulus.  This suggests that the model is failing for the sides, but is OK for the top surface.

--Bede

 

Top 75 Contributor
11 Posts
Points 125
mariam replied on Thu, Jun 13 2013 2:19 AM

Dear Bede,

 

Thank you for your response. I have not looked into the stiffness of the samples yet, so that may be a starting point for me. The roughness values were previously obtained from profilometry measurements and not during the scanning.

From what you mentioned about your measurements what method can you suggest for obtaining the elastic modulus? Would you recommend the use of point and shoot or first I should obtain force curves and then do post analysis in order to obtain the modulus. The reason this came to my mind is because my sample's morphology may not be very uniform throughout due to some agglomeration and this may affect calculation directly.

 

Best,

Mariam

  • | Post Points: 12
Top 10 Contributor
280 Posts
Points 6,221
Bruker Employee

I would start by getting some high quality topographic images of your sample to identify areas taht are relatively flat and the length scale of those areas.  If the contact area of tip on sample is small compared to the relatively flat area, you should be able to get decent modulus numbers from either PF-QNM or point and shoot force curves (really the model is the same for both).

The other strategy is to go for a macroscale measurement that will incorporate many grains.  For that you would probably want to use an depth sensing indenter (DSI) such as those available from Hysitron so that you can apply larger loads to make bigger deformations.

--Bede

  • | Post Points: 12
Top 75 Contributor
11 Posts
Points 125
mariam replied on Thu, Jun 13 2013 5:30 PM

Thank you Bede for your suggestions. our system was upgraded so I am new with PF-QNM I'm still in the process of learning the system and any shortcuts available. New things keep coming up during measurements but I am grateful for the topics  in this forum they have helped alot...

 

  • | Post Points: 10
Page 1 of 1 (9 items) | RSS
Copyright (c) 2011 Bruker Instruments