Drift of a nanoindentation instrument may also depend on the type of Instrument you use. E.g. we have a Hysitron Triboscope where the indentation tip is attached to an piezo-scanner for imaging purposes and the piezo show creep behavior after changing position of the tip. The only possibility to avoid this is to make a pause before each new indent to let the piezo creep finish (typically 3-5min).

Beside that a housing of the Instrument is essential to avoid thermal changing and a sufficient long thermalization time is nessecary after closing the housing and before starting measurement. That can take a whole night (or day) to wait.

Beside that , naturally,  some materials themselves show creep, especially soft metals. This is the normal plastic behavior of these materials and has to be considered in the measurement, so measurement time has to be (in contrast to what Jonas Paul wrote) long enough to consider material creep, otherwise hardness measured will be too high.

Finally, each nanoindentation should include a drift measurement and a off-line (or on-line) drift correction which will minimize the drift effects.

Best regards

Kirsten

Sadeem -

The other environmental factors that I think might cause drift are changes in air pressure and fluctuating electromagnetic fields (for sensitive devices). These things can change after you have performed your calibration. You can only hope to eliminate them in the sense that their effects will be small compared to your signal.

How do you measure the drift on your instrument? When I did nanoindentation, after the indent, I would withdraw the tip until the force level was, say, 95% of peak and halt it there. While the tip was just barely touching, I would monitor the displacement signal for 20 seconds or so; this would give me an idea of the drift level and, if the drift was uniform, I could correct for it.

-Jon

Ways of minimizing it (I am afraid that it can never be completely avoided) are:

- purchasing a device having an active continuous correction, based on a reference tip

- purchasing an acoustic chamber, in addition to a special antivibration table

both expensive, of course.

Alain

Thanks Jonathan and Alain

I would like to ask are there any models that can describe this phenomenon quantitatively or there are just interpretations that are based on expectations?

This is caused by temperature localized change, misalignment of load and materials behavior.

Are there any experimental studies that try to relate this effect quantitatively with temperature?

Hi, Sadeem, thermal drifts are in nanometrology (atomic force microscope, nano indenter...) one of the common source of error. Their effect may be minimized but rarely canceled. The reason lies in their origin, your system is made of different materials having different coefficient of thermal expansion. Plus you have electronic parts which are heat sources. Basically your system always has thermal drifts but their effect may be minimized by waiting for a thermal equilibrium, using an enclosure to minimize air movements around the apparatus, the room temperature is also important in order to minimize the temperature difference between the equilibrium temperature and the one of the room (it buys you precious time to reach the equilibrium state).

Now although thermal drift always exist there are some ways to cancel their effect  by measuring them in x-y-z directions with for example an interferometric system (fabry perrot). This is used for metrologic AFM. Some other techniques have been developped for nanoindentation measurement,measuring the thermal drift before a measurement and while its value is below a threshold it is supposed to stay cosntant during the experiment, having a capacitive transductor near the tip that is able to measure the Z drift during the test (good for creep measurement)...

I hope this has been of some help

Thanks Jean

Your answer takes me to another point to study the effect of thermal drift, Have any one tried to do nano indentation in low temperatures and see what's happened at these temperatures for the drifting curve?

if any please provide references

Hi, Sadeem

There are number of studies related to causes of drift in nano indentation. few links, but not least are provided below for your reference. 

http://inside.mines.edu/~cpackard/papers/Nanoindentationandcontact-modeimagingathightemperatures.pdf

http://www.csm-instruments.com/en/Ultra-Nanoindentation-Tester-new-generation-of-thermal-drift-free-indentation

http://www.azonano.com/article.aspx?ArticleID=3228

http://www.google.ca/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&uact=8&ved=0CC8QFjAA&url=http%3A%2F%2Fdspace.mit.edu%2Fopenaccess-disseminate%2F1721.1%2F69648&ei=3IAIVNzLG8n9igL2qYHYDg&usg=AFQjCNGuH24mk87OSGE-qJg3hjd0l9Bqeg&bvm=bv.74649129,d.cGE

thanks Shailesh K. Dhoke.

How about low temperature, are there any work that study the trend at low temperatures?

Hi, Dr. Fadhil. Yes, the evil "thermal fluctuations". Sure, they can back up these fluctuations with the Stochastic model, can't they? Are they really so smart? Bye!

Drift comes to picture when there is deformation in the tip and material due to temperature change. We can avoid it by stabilizing the system at a constant temperature. Better results can be obtained if you maintain the temperature at 16-degree centigrade. 

An easy way to reduce the effect of drift is to minimize measurement time.

Drift of a nanoindentation instrument may also depend on the type of Instrument you use. E.g. we have a Hysitron Triboscope where the indentation tip is attached to an piezo-scanner for imaging purposes and the piezo show creep behavior after changing position of the tip. The only possibility to avoid this is to make a pause before each new indent to let the piezo creep finish (typically 3-5min).

Beside that a housing of the Instrument is essential to avoid thermal changing and a sufficient long thermalization time is nessecary after closing the housing and before starting measurement. That can take a whole night (or day) to wait.

Beside that , naturally,  some materials themselves show creep, especially soft metals. This is the normal plastic behavior of these materials and has to be considered in the measurement, so measurement time has to be (in contrast to what Jonas Paul wrote) long enough to consider material creep, otherwise hardness measured will be too high.

Finally, each nanoindentation should include a drift measurement and a off-line (or on-line) drift correction which will minimize the drift effects.

Best regards

Kirsten