My name is Felipe and this is my first post here. I'm working on a project to calibrate an electrochemical ammonia sensor. But I have a lot of doubts because I am a layman in chemistry. I'll list them and I'd be grateful if you could respond them.

The system I have here is basically a cylinder with ammonia gas of known concentration connected by a 0.5 lpm regulator and a tygon tube to a chamber where the sensor is. Here are my doubts:

  • When I open the regulator the gas starts to flow into the chamber, the concentration of oxygen(or other gases) in the chamber can affect the concentration of ammonia gas? Or even if the gases mix, there is a fixed concentration of ammonia?

  • I've seen a lot of calibration videos. Why does nobody care about the tube length? I think as long as the tube length more is the waste of gas, right?

  • When I stop with the ammonia gas flow and disconnect the tube from the chamber. What are the precautions I should take with the contaminating ammonia in the air? Let's say I'm using a calibration gas of 50ppm


The calibration gas will purge the air out of your chamber, so that it contains 50 ppm NH3. This assumes that there are no reactions that change the mixing ratio of the ammonia (e.g. reaction with water in your chamber). The length of the tube doesn't matter unless the gas reacts with your tubing, though length will decrease the pressure of the gas coming out of the tube.

  • $\begingroup$ You do need to let it flow long enough so that the volume of (tube + chamber) is thoroughly purged first. If no dead volumes or recirculation paths that could be perhaps 10 volumes or more of gas flow. But, 500 sccm is a pretty good flow rate given some Tygon tubing and a small chamber. As for wasting, the bottle has lots of gas in it... As for safety, a glance at the MSDS shows you should not breathe directly from the gas line (but then, you don't say what the balance in the cylinder is so you could just asphyxiate yourself without worrying about the NH3 TLV). $\endgroup$ – Jon Custer May 30 '18 at 22:02
  • $\begingroup$ Of course, I guess you could calculate a volume required to flush the tubing, but it's just easier to wait until the reading is steady. $\endgroup$ – Mark Jun 1 '18 at 18:40
  • $\begingroup$ Thank you for your answers. Does anyone know how to calculate the volume needed to flush the tubing? $\endgroup$ – Felipe Quadros Jun 4 '18 at 17:42
  • $\begingroup$ @JonCuster I know the waste is really too low. But I keep wondering how it is done in companies that have a lot of sensors. You agree if I have to repeat the process thousands of times, at some point the waste begins to be relevant. $\endgroup$ – Felipe Quadros Jun 4 '18 at 17:45
  • $\begingroup$ Well, I’d use a load lock and cycle many sensors in at a time... $\endgroup$ – Jon Custer Jun 4 '18 at 20:54

Not the answer you're looking for? Browse other questions tagged or ask your own question.