How to measure pH of a Gas

[RADT]

I am currently in the process of trying to find a good way to measure the pH of a gas stream and I am searching for alternatives.

The application is in an oil refinery, the gas stream is the flare header gas, made up of a variety of hydrocarbons, basically anything and everything in the refinery depending on what is happening in the various units.

The idea that we are going with right now is to draw the gas through a liquid (water) and measure the pH as a result. However, I have been unable to find anyone in the industry who has done this sort of thing before, I have checked with local lab techs, engineers, vendors, etc.

Our process engineer seems to think that this is going to work without a problem however I am not so sure.

Any information hints or history would be appreciated.

 

[IRstuff]

Probably because it doesn't really make sense. pH has to do with the disassociation of hydrogen specifically in water, so the "pH" of a gas is not possible to measure.

So, the REAL question is what question are you actually trying to answer by this measurement?

TTFN

 

[RADT]

Wording in first section should have been "measure the ph of the water as a result of the gas (vapor) passing through it"

 

[Cat3347]

Just make sure you are using distilled water. That is the key factor in a successful measurement.

 

[RADT]

Why distilled water? consistency? I was hoping to use utility water which is basically river water minus some turbidity. The pH measurement doesn't have to be that accurate it is more just to tell if there is acid present or not.

 

[dcasto]

OK, You are looking for molecules that will ionize in the gas phase Co2, Cn, H2S, HCl...), So, fom chemistry, it's a mole/mole ratio that presents it's self as a pH measurement. So, you pass the gas through distilled water (distilled because ther are no othe H+, OH- ions to interfeer). But, how do you relate this pH to an actual situaion where the flare header has free water and the H2S removal unit has an upset and sends a ton of H2S to this same header?
Go ahead and ask the question, "SHOULD THE HEADER BE MADE OF CORROSION RESISTANT MATERIAL, OR CAN WE ACCURATELY PREDICT WHEN THE CS PIPING WILL FAIL".

 

[IRstuff]

Then it depends on the specific acid and the overall concentration. Your river water may have sufficient bases to neutralize the acid.

Additionally, have you actually determine how MUCH gas is required to get a positively unambiguous pH reading?

Generally, you'd need a sizable fraction of a molar to get a meaningful result.

TTFN

 

[RADT]

I'll elaborate more on the purpose of this instillation, initially I was brief because I was unsure of the feedback that I would receive (first timer).

Anyway, all of the gas in the flare header is sent to a flare gas compressor which pushes the gas back to an amine unit where it is cleaned and turns into fuel gas to be burned in boilers / furnaces throughout the refinery. The flare gas compressor runs roughly 75% of the time when the conditions are right, i.e. sufficient mole weight and volume.
The idea of the pH analyzer is to give the board operator an automated indication of the acidity of the gas going to his compressor. We have had an instance where we were compressing acidic gas for some time with out knowing it until one of the sight glasses on the compressor sprung a leak, obviously very costly rebuild.

Appreciate all the feedback!! Keep it coming!!

 

[dcasto]

Why not put an H2S or CO2 analyzer, more expensive, put you could also feed forward data to your amine system.

 

[IRstuff]

I agree with dcasto, directly measure the quantity of interest, instead of indirectly measuring a conglomerate of variables?

TTFN

 

[RADT]

I appreciate what you are saying but it's not one component that we are interested in, it's the conglomerate of variables that do the damage to the compressor so that is what we would like to measure (regardless of cost). If we only measure one component of the stream there could still be others doing damage, we are no further ahead. There has been talks of using a fluoride ion analyzer, since we use a hydrofluoric acid unit this is presumable the largest contributor to the acidic flare gas However, there could still be the instance where the HF acid unit is not flaring anything but one of the other units could be dumping some other form/type of acid. Again, not quite what we want.

 

[IRstuff]

So what about a mass spectrometer? It'll allow you to simultaneously measure all components of the gas.

TTFN

 

[RADT]

I could look into that, I don't know much about mass spec so I guess I should learn.
I'll let you know how I make out.
Thanks!!

 

[fribby]

Might be worth giving Siemens Process Analytics a call, they do a nice line in (relatively) resaonably priced, industrial Mass Specs. I have seen one in action and it seemed pretty good.

I would like to point out that I don't work for Siemens, lol! Therefore have no vested interest, just a customer of theirs.

 

[JLSeagull]

Process mass spectrometers are costly and rather complex for some facilities. I also need to learn more about them. A few manufacturers offer online mass spectrometers. These include the following:
ABB Extrel
Hamilton Sundstrand AIT MGA iScan.
Siemens Quantra
Thermo Onix VG Prima ?B
Ametek
Monitor Instruments

 

[randi]

How about a bypass manifold with some water in it and a retractable pH sensor - could this work.

Otherwise do a complete on-line process stream monitoring, continuous or on a sampling basis.

 

[dcasto]

Change out the compressor and get it over with. I've installed acid gas compressors running on 100% water saturated H2S and CO2. Ariel will fix you up a unit made of mostly carbon steel and there are lots of packagers that will put together all the other parts. You might be able to retrofit the unit you already have.

 

[RADT]

You'll never guess what we ended up doing for this application. We decided to give a corrosion transmitter a try, we are just in the process of hammering out the specifics but it looks like we are going to go with the Honeywell Smart CET5000. We thought that since it is corrosion that we are most concerned with why not the actual rate of corrosion. This will be a new technology to our site so we will be doing some experimentation to see how it will work best, i.e. we are not sure what probe materials will give us the best response and still last as long as we would like.
The cost is pretty much what drove us this way, budgetary quote came back at between 7 and 10 K for the hardware, this is about 1/3 the cost of the cheapest analyzer that we looked at.
Anyway, I'll try to write back in a few months and let you know how it worked out, thanks for all the feedback.

 

[Agfa]

use the total hydrocarbons analyzer, such as a FID analyzer,to measure the gas content. The gas content can show the relation with PH value of your gas stream, if it has the PH value.