Recently, I fabricated gold heater of serpentine shape on polyimide substrate. The heater is about 50 nm thick and the substrate is 7 um thick. However, when I increase the voltage of source to 6V, the resistance raised continuously to 400 Oh. (under room temperature normally is 250 Ohm). This change can be irreversibly. When I apply even higher voltage. The resistance could be ten thousand Ohms. Obviously, the heater was damaged or burned. If I want to heat up a little drop of water, the voltage that could result in damage can be lower. I am not sure what I can do in order to heat up and without damage my heater. I used a IR camera to detect the temperature when 6 V is applied, it shows the maximum temperature to be 34C. Thank you
Do you use a gold ink or some process with grown continuous gold layer?
Your result (steady, irreversible increase of resistance until open-circuit) is typical for the gold-ink process.
The problem is the interface resistance between gold grains in ink-printed layer. These interfaces start to disintegrate at about 1mA/mm2
Even if you used a continuous film method (evaporation, sputtering or electroless gold), 50nm thickness is likely not enough to form continuous film on rough polyimide substrate. If your film is not continuous, you will have the same problems as with gold-ink process - the easily broken interfaces between grains.
Therefore, i recommend to check your conductors under electron microscope. After you will see the actual surface morphology, the answer may become obvious.
Thank you so much Oleg. I think the thickness is indeed one problem. I used ebeam evaporation method to deposit the gold.
There is still one question: for 1mA/mm2, is the area indicate the cross area? I/A=V/R/A=V/r/l, (r is resistivity). Does this indicate that longer length would make the value smaller, and it has no relevance to width and thickness.
Thank you Alex. The camera is actually not reporting the correct value. Sorry for this. Since when I used this, I did not fully focus with the lens, the actual temperature could be 60C. Also, the camera does not have more pixel to fully show the temperature difference between trace and the rest of the area. A more advanced micron lens is required for such detection
Yes, 1mA/mm2 is current per cross-section. This value is independent of length of conductor.
Also, if you have 60C or more trace temperature, your film may be unstable against formation of hot spots (voltage domains). Manual for thin-film transmission lines made of pure metals says what temperature rise above 10-30K above ambient is risky. To reduce the risk of hot spots instability, metal in heaters must be alloyed to reduce temperature coefficient. Few % of silver usually do the trick.
- Badges
- Science topic
- Similar topics
- Entomology
- Resistance