MEASURING NITRIC ACID TEMPERATURE W/ T-COUPLE???

[ctroyp]

Can anyone tell me why this happens? Why does my electronic temperature device display go unstable when trying to read the temperature of nitric acid. I am using a "J" type thermocouple wire twisted at the end. I am sure there are better ways, but why does this not work? I did try a keyword search with no luck. Thanks for any help...

 

[hacksaw]

sounds like you've made a thermocouple for direct immersion in the acid.

several issues:

1. you need to make certain that you are using wires that are made for thermocouple service (purity and accuracy issue).

2. most measurements require that the sensing element be placed in a pocket or sensing well to protect it from corrosion.

3. use of twisted end-thermocouples went out with the dark-ages. either silver solder, fuse(welding tourch), or spot weld the ends together. you can also use a crimping sleeve.

4. direct immersion of the sensor is discouraged if you want an accurate reading. Type J is a Cu-Fe couple. both are attacked by nitric solutions.

Now that you've done all that, now you have to confirm the calibration against a set of standards. you also have to worry about the junction emf's you might be creating in wiring your sensor to the display device.

The simplest thing to do is to purchase a factory made t/c in a stainless steel sheath, possibly with a suitable thermowell.

if you have an industrial measurement there are another set of issues that have to be included.

 

[ctroyp]

Thank you hacksaw, very helpful.

 

[azg]

popco - in addition to what hacksaw said, you need to sheath your thermocouple - or use a non-contact temperature measurement.

The reason for the loss of stability in readings is the electrochemical processes occurring in the nitric acid around the thermocouple: on one hand, you are trying to generate a thermopotential against the cold end; on the other, you have Cu and Fe in an acidic environment, which will inevitably cause you to lose some charges into the acid (more so from the Cu, as it is more reactive).

I could write out the exact chemistry of the processes you have, but I would need to know the concentration of the acid and the temperatures you are trying to measure.

As Cu is losing Cu+ and / or Cu++ (at a higher rate than Fe is losing Fe++ and / or Fe+++), and as the nitric acid is taking the cations, it loses some electrons off the Nitorgen, and therefore turns into an oxide of nitrogen (NO / NO2), which is a gas. That occurs on the surface of your TC wire, which is now a set of electrodes. The gas blocks the wire from contacting the bulk of the liquid nitric acid, and therefore, you are losing stability as the bubbles form and leave the surface of your thermocouple.

In addition, as copper and iron are getting etched out, they redeposit themselves as oxides / nitrites / nitrates on the surface, which degrades the sensitivity of your temperature measurement.

My best advice - once again, use a glass-sheathed thermocouple. A thin layer of glass will be sufficiently thermoconductive, and all you need is to look at the chart for this specific sheathed thermocouple.