Depolarisation of pzt with heating above curie temp

[szzuk]

Hi,

I'm wondering how I can check if I have depolarised pzt?

I have a piezo that consists of pzt with a curie temperature of 200 degrees C that needs to be glued to metal with an epoxy resin that needs to be heated for 1 hour at 120 degrees C.

I can't risk damaging the piezo, so I was thinking how does pzt change when depolarised and what is the best way of checking.

Regards,

Szzuk.

 

[Sceadwian]

You're 80 degrees away from the curie point, what are you worried about?

Can't really answer any of your questions unless you can define the exact type of piezo material you're using.

You're asking questions without providing basic information.

 

[Jaguarjoe]

PZT is lead(Pb) Zirconium Titanate.

Most things I know of don't recover when heated above the Curie point but then again, I'm not a physicist, so some things may behave differently.

 

[szzuk]

It is material 552 on the link below, PZT Navy type VI. So it has a curie temp of 200 degrees C and a maximum working temperature of 120 degrees C.

**broken link removed**

I was asking in an awkward way how does pzt behave? does it switch off like a light at 200 degrees C or does it degrade slowly as it approaches the curie temp? because I've read 'general' info that says don't heat pzt above half the curie temperature.

My other request was for a test to see if I'd altered the condition of the pzt, at a guess I'd say depolarising pzt will reduce its capacitance, but what if I've heated it above half working temperature, will it lose some capacitance but not all?

Regards,

Szzuk.

 

[Sceadwian]

JaguarJoe, I didn't even know that about PZT materials until you mentioned it, I looked it up and there's a good reason; most commonly used PZT material in electronics tends to be Quartz, and that has a Currie temperature of 573 degree's C, so it's not a worry.

I'm a little fuzzy on the specifics but the reason the change occurs is because at those temperatures the crystal structure changes which causes the loss of polarization. Something you might want to look into a little more, because I haven't (in the few minutes I spent) read anything that suggests that the material won't repolarize when it's cooled back down.

Like hardening of metals if you heat it up fast and snap cool it it may not be okay, but if you heat it up beyond the Currie point and allow it to cool very gradually the original crystaline structure should reform, but this being a ceramic it may be slightly more complicated than that as I'm not familiar with the formation of LZT ceramics, metallurgy is only a hobby =) And one of my least knowledgeable ones.

As long as you don't exceed the Currie temperature; and you have a healthy margin of error by the looks of it no damage can possibly occur anyways. I would mostly avoid thermal shock by making sure that cooling is gradual.

 

[szzuk]

PZT won't re-polarise after it has been heated above its curie temp and then cooled, but it can be re-polarised with a strong electric field. I'm just going to assume depolarisation won't occur below the curie temperature in any significant way, and there is another reason it isn't good to heat it over half the curie temp, probably cracking.

Regards,

Szzuk.

 

[Sceadwian]

szzuk, I believe you, but can you provide any links to the technical reason behind this? I'm curious. I'm sure it has something to do with the formation of the ceramic, I'll look into that presently just to satisfy my curiosity.

Any heating or cooling should be done gently this is basic mechanical common sense, thermal shock is bad. The slower you heat and cool anything the less stress can be induced from things like difference in thermal conductance and resistance of the various components, or common thermal expansion/contraction.

As long as you stay away from the Currie temperature the only effects from heating would be from long term diffusion, and of course it's effective capacitance will change but that I'm certain will recover at a lower temperature. I would say 10 degrees is okay, 20 is better. You're 80 away no bulk depolarization should be possible.

 

[Sceadwian]

My next search found this which looks after a 10 second glimpse to provide all the information I was looking for =) Reading more now.

 

[Sceadwian]

Based on that PDF if you know the devices structure and it's polarization direction it's possible to repolarize it even if it becomes depolarized, which again should not even occur, you're 80 degrees from the Currie point. Maximum piezo effects peaks at around 2kv per mm, all you need from high field DC generator is sufficient insulation of the target to prevent arcing. So if the device is in an electric field that is aligned with it's native polarization and it's raised above it's Currie temperature, it will not lose polarization so long as it's in that same field when it's cooled.

It's identical but on the electric side of things in comparison to the loss of a permenant magnets magnetic properties when it's heated above it's currie point. Which obviously yet oddly I never realized it is exactly the way rare earth magnets work and they're ceramics just like the Lead/Zircon/Titanium ceramic that's used in this PZT.


Hope I didn't confuse anyone too much thank you very much for this post and allowing me the learning experience from trying to figure out what's going on, even if I haven't alleviated your concerns.


Now I'm cursious about the link between what makes a good PZT ceramic vs what makes a good magnetic ceramic.

 

[Jaguarjoe]

JaguarJoe, I didn't even know that about PZT materials until you mentioned it, I looked it up and there's a good reason; most commonly used PZT material in electronics tends to be Quartz, and that has a Currie temperature of 573 degree's C, so it's not a worry.

PZT is a ceramic, quartz is a mineral.

I haven't looked in a few years, but the PCB Piezotronics web site might be worth a look. I don't know if the OP is using this for a sensor or not, but it's very old technology. Waaay back in the day, PCB offered me a job when it was 3 guys in a 2 room 2nd floor 'building' over a truck depot. Of course I 'knew better'. Look at them now. One of dumber mistakes.

This may be valuable:

https://www.electro-tech-online.com...ic_ceramics_28pzt-4_pzt-5a_pzt-5h_pzt-829.pdf

This stuff is even flakier than I remember 40 years ago.

 

[szzuk]

The replies tell me I don't have anything to worry about. It's best to check if not sure.

PZT has been around for a good while but there are high tech areas, for example lead is a hazardous substance to the environment and some would like to ban pzt, the only reason they haven't is that there is no cheap and practical alternative, when someone invents it they'll make a few dollars. Also multilayer pzt devices are very expensive > $1000 for just one part. By multilayer I mean the layers are 20 microns thick, so you can have 50 layers in just 1mm thickness. You don't get much improvement in displacement from those parts but crucially you get that displacement from say just 100V. This matters for electronic engineers and products because 1KV drive circuits are much more complicated and expensive than 100V. When the price of the multilayer devices drop there will be lots of cool new products from pzt, assuming it hasn't been banned...

 

[Sens]

You should be okay. At 80% of curie you will start to see some depoling. You will see some cap drop due to CTE miss-match when coupling to a metal substrate. One thing to monitor is cap decay rate after bonding. If your application is a sensor, you will see significant sensitivity loss if it was depoled.

 

[szzuk]

Ah thank you. I was wondering roughly when depoling would begin. What does the acronym CTE stand for?

Regards,

Szzuk.

 

[Sens]

Coefficient of thermal expansion. There is a miss match between most metals and pzt. When using a heat cured epoxy, the higher the temp to cure, the more compression on the pzt at ambient.

 

[Sceadwian]

Sorru Jaguar, I mistook PZT for generic use for Piezo material. didn't even connect it with being releate to LEad zircontae titanium (I still always find it odd that leads atomic letters are pb)

 

[Jaguarjoe]

Pb is the symbol for lead because plumbers in ancient Rome used lead for pipes. The association lasted a long time. I think lead is properly called plumbium.

 

[Jaguarjoe]

Coefficient of thermal expansion. There is a miss match between most metals and pzt. When using a heat cured epoxy, the higher the temp to cure, the more compression on the pzt at ambient.

I thought Invar compensated for that. I could be wrong though cause that is a 50yo thought

 

[Sceadwian]

I think lead is properly called plumbium.

hehe, that much at least I know is not true at least anymore. Plumbum is latin for lead but it's usage is obsolete, they probably just used the latin abbreviation to avoid confusion with all the other elements that start with L