Hello! I've been a little confused recently with some obtained data. The related question is, can roughness of a spin coated and highly uniform film be larger than the thickness of the film? For instance, if the film thickness is 20 nm, can roughness be 25 nm?
Hello,
it depends what is your "layer thickness".
If it's measured by a quartz monitor you measure the increase in mass on the sample (Thickness is obtained by dividing the mass by the density - not necessarily the true density of a sputtered film).
You don't know how the mass is distributed.
If your layer consists of pyramids 50nm high (measured by AFM) the average thickness (mass) is much lower.
It's definitely possible. Your average thickness may be 20 nm while your thickness variation is +/- 12.5 nm giving you a minimum thickness of 7.5 nm and a maximum thickness of 32.5 nm and a peak roughness of 25 nm.
Dear Ron Reiserer
The film is actually highly uniform, the shown roughness is significantly larger than the highest value for the thickness. Therefore, I'm a bit confused.
It would be helpful if you could share the data or images with which you have drawn your conclusions. The composition of the film and the substrate as well as the measurement instrument would be useful information as well.
Ron Reiserer thank you for your replies, but unfortunately, I have no permission to share the data as they are a part of a submitted paper in a revision phase and I'm just one of the authors.
Best regards,
Here are some sample illustrations of how you may get a large surface roughness measurement from a thin film. The top two can appear quite uniform to the eye (or instruments) even though the variation could be larger than the average film thickness.
I would be highly concerned... Roughness? Are talking about Ra? This would be the "mean deviation" from a zero line after removing the shape and waviness. You should expect maximium deviations that are 5x larger as the mean value.
I would consider thickness variations as "waviness" (in your viewgraph). Roughness is the small-range modulation. If you expect the roughness in the 20nm-range a Gaussian filter of 80µm should be used (this is used for "technical surfaces", machined surfaces). Length scales are curcial.
Thickness is average of all heights from substrate and roughness is average of differences from the thickness. So a Mathematical formulation approach can help which can show the range or domain for conditions, if one can put here.
Thanks
- Badges
- Science topic
- Similar topics
- Physical Sciences
- Materials Science
- Materials Chemistry
- Polymer Chemistry