The Nanoscale World

Getting started with CAFM

rated by 0 users
Answered (Verified) This post has 2 verified answers | 3 Replies | 2 Followers

Top 500 Contributor
2 Posts
Points 24
Daragh8008 posted on Thu, Apr 17 2014 5:25 AM

Normal 0 false false false EN-IE X-NONE X-NONE MicrosoftInternetExplorer4

HI all,

 

I was hoping someone on this forum might be able to help me. We had a conductive AFM module installed on of dimensions V about two years ago, but no one here has ever used it, as a result we don’t know how to get going. So we are basically working from the manual supplied (nanoscope software V7). Couple of problems though.

1)     The manual say in the scan parameters list for Feedback mode to use the “open” option, but on the software only provides disable or enable. Should this be set to disable?

2)     Again the manual says “Set Feedback > Deflection Setpoint to no more than a few tenths of a volt above the engagement point. This will keep the contact force sufficiently low, thus not causing damage to the tip or sample. This is especially important for metal coated tips or soft samples.”  When we set the deflection set point to 0.1-0.3V for example the tip instantly thinks its engaged when it is not in contact with the surface. When it is set higher say 0.6V it does engage but I think the tips are getting damaged very quickly at this point. Looking at a soft sample I could see a scratch being formed on the substrate. How does one get the tip to engage without damaging the tip or sample?

3)     Whether by tip damage or other means, when it is scanning we don’t seem to get any electrical variation. But we are not particularly sure as to the settings range for the DC sample bias. So we could have melted the metal coating on the tip as well damaging it due to the deflection setpoint. Could anyone suggest a starting range? For example if I was looking at a 10um film of FTO on glass (making a connection to the sample chuck with copper tape) using a undamaged SCM-PIC tip what sort of sample bias and current sensitivity would be considered a reasonable starting point or range to see some features?

Sorry for all the question but any help would be greatly appreciated.

 

Thanks

Daragh  

  • | Post Points: 12

Answered (Verified) Verified Answer

Top 25 Contributor
26 Posts
Points 298
Bruker Employee
Verified by Daragh8008

Hi Daragh,

In C-AFM (or TUNA), one can operate in 'constant voltage' or 'constant current' mode. 'Constant voltage' is the most common & easiest method, and I strongly recommend you use this. You can select this mode by selecting 'Open' in the feedback mode, or 'disable' (softwrae version dependent).

When your 'deflection setpoint' is set to small values, it is posisbel that you will get a so-called 'false engage'. Sounds as if you are facing this. Mkae sure the laser is nicely aligned on the cantilever as poor alignment can contribute to this. You can also increase the 'deflection setpoint' in order to allow engage: just use a slightly higher value (0.1V higher than waht you used before) and click once more the 'engage' button. Repeat as needed until you have a real engage. Once engaged, you can usually reduce the 'deflection setpoint' again by a few times 0.1V (basically to about 0.1-0.2V above the 'free' deflection. You might want to set the scan size to 0nm durign engage, to avoid imaging at setpoints higher than you wish. Increase the scan size only after you have found a low 'deflection setpoint' value, which still provides stbale feedback on the surface.

Often, cantilevers for C-AFM or TUNA have a relatively high spring constant (2N/m). This can be quite high for some softer samples, as we are scanning in contact mode. You could try to use a softer spring constant cantilever or use our more recent Peak Force TUNA mode, in whcih we avoid the contact mode (and therefore perfect for conductivity mapping of soft & fragile samples).

The DC voltage is indeed important. I recommend you start by 0V (during engage, it is good to keep 0V between tip and sample, as any voltage could lead to a snap-to-contact or higher currents than expected, which could cause tip damage). Also set the current gain of your amplifier (C-AFM or TUNA) to the highest sensitivity, so taht you can immediatly observe if there is a small current or not. With 0V applied, optimize your imaging settings for good topography quality. Wihtout a good topogrpahic image (good tracking, no deformation, ..) you will nto be able to get good electrical data. Next, you can start to increase the DC bias. Start with small values (for example 100mV increments) until you start seeing something. On the type of smaple you use, you might have to go to a few Volts. Some samples have a storng non-linear I-V response, so it is a good idea to try both polarities as you are increasing the DC voltage. If you have any doubt if the selected voltage actually gets applied to your sample: use a voltmeter to check it.

Before starting the measurement, make sure your electrical 'back' contact is well connected. I see you use Cu tape. I usually use Silver paint or metal clips to provide the electrical contact to the top surface of samples like yours.

Regards,

Peter

 

 

 

  • | Post Points: 13
Top 25 Contributor
26 Posts
Points 298
Bruker Employee
Verified by Daragh8008

Measurign the conductivity of nanowires which are less conductive than their substrate or matrix they are emdedded into, can be difficult. Usually, it is easier to do the reverse: have them on a substrate whcih is less conductive than the actual nanowire. The best success I had is when the nanowires are only partially on a conductive substrate, and are partially on an isolating area. This allows one to see how the conductivity of th enanowire drops of as you move (along the nanowire) further away from the conductive substrate area. This then allows one to extract the actual resistivity of the nanowire. If the conductiviyt of the nanowire is substantially different (lower) from the substrate, you might have to use DC voltage swhich saturate the current when in the substrate. If these currents are really high, you might want to place a series resistance in serie between the sample and the back contact electrical connection.

As the currents in your nanowires seem to be rather low, keep the current gain in its 'highest sensitivity' setting. If you have TUNA, this should allow you to measure to below 100fA.

Peter

  • | Post Points: 11

All Replies

Top 25 Contributor
26 Posts
Points 298
Bruker Employee
Verified by Daragh8008

Hi Daragh,

In C-AFM (or TUNA), one can operate in 'constant voltage' or 'constant current' mode. 'Constant voltage' is the most common & easiest method, and I strongly recommend you use this. You can select this mode by selecting 'Open' in the feedback mode, or 'disable' (softwrae version dependent).

When your 'deflection setpoint' is set to small values, it is posisbel that you will get a so-called 'false engage'. Sounds as if you are facing this. Mkae sure the laser is nicely aligned on the cantilever as poor alignment can contribute to this. You can also increase the 'deflection setpoint' in order to allow engage: just use a slightly higher value (0.1V higher than waht you used before) and click once more the 'engage' button. Repeat as needed until you have a real engage. Once engaged, you can usually reduce the 'deflection setpoint' again by a few times 0.1V (basically to about 0.1-0.2V above the 'free' deflection. You might want to set the scan size to 0nm durign engage, to avoid imaging at setpoints higher than you wish. Increase the scan size only after you have found a low 'deflection setpoint' value, which still provides stbale feedback on the surface.

Often, cantilevers for C-AFM or TUNA have a relatively high spring constant (2N/m). This can be quite high for some softer samples, as we are scanning in contact mode. You could try to use a softer spring constant cantilever or use our more recent Peak Force TUNA mode, in whcih we avoid the contact mode (and therefore perfect for conductivity mapping of soft & fragile samples).

The DC voltage is indeed important. I recommend you start by 0V (during engage, it is good to keep 0V between tip and sample, as any voltage could lead to a snap-to-contact or higher currents than expected, which could cause tip damage). Also set the current gain of your amplifier (C-AFM or TUNA) to the highest sensitivity, so taht you can immediatly observe if there is a small current or not. With 0V applied, optimize your imaging settings for good topography quality. Wihtout a good topogrpahic image (good tracking, no deformation, ..) you will nto be able to get good electrical data. Next, you can start to increase the DC bias. Start with small values (for example 100mV increments) until you start seeing something. On the type of smaple you use, you might have to go to a few Volts. Some samples have a storng non-linear I-V response, so it is a good idea to try both polarities as you are increasing the DC voltage. If you have any doubt if the selected voltage actually gets applied to your sample: use a voltmeter to check it.

Before starting the measurement, make sure your electrical 'back' contact is well connected. I see you use Cu tape. I usually use Silver paint or metal clips to provide the electrical contact to the top surface of samples like yours.

Regards,

Peter

 

 

 

  • | Post Points: 13
Top 500 Contributor
2 Posts
Points 24

Hi Peter,

 

thank you for the help. I tried the things you suggested and after a little tweaking, managed to get our first C-AFM images so we are now happy that the module is working OK. One last question if you have a moment. One of the samples we are hoping to look at are Si nanowires in a metal film. They lying flat and are half embedded in a silver in the film and half exposed above. I.e their major axis is in the plane of the metal film. I was able to get a conductivity map of the silver film but the wires having significantly larger resistance show up as 0 current spots. Can we increase the sensitivity having the silver go over range to get detail on the wire or is there a better approach? Any suggestions?

 

Thanks Again

Daragh

  • | Post Points: 12
Top 25 Contributor
26 Posts
Points 298
Bruker Employee
Verified by Daragh8008

Measurign the conductivity of nanowires which are less conductive than their substrate or matrix they are emdedded into, can be difficult. Usually, it is easier to do the reverse: have them on a substrate whcih is less conductive than the actual nanowire. The best success I had is when the nanowires are only partially on a conductive substrate, and are partially on an isolating area. This allows one to see how the conductivity of th enanowire drops of as you move (along the nanowire) further away from the conductive substrate area. This then allows one to extract the actual resistivity of the nanowire. If the conductiviyt of the nanowire is substantially different (lower) from the substrate, you might have to use DC voltage swhich saturate the current when in the substrate. If these currents are really high, you might want to place a series resistance in serie between the sample and the back contact electrical connection.

As the currents in your nanowires seem to be rather low, keep the current gain in its 'highest sensitivity' setting. If you have TUNA, this should allow you to measure to below 100fA.

Peter

  • | Post Points: 11
Page 1 of 1 (4 items) | RSS
Copyright (c) 2011 Bruker Instruments