6 replies to this topic

[AndyChemEng]

Hi, 

 

I'm looking at installing activated carbon on a process facility that has variable outlet concentration (1000-50000 mg/m³) @ 300m³/h gas flow. The VOC consists of toluene and normally operates around 5000mg/m³ with short duration spikes (10 mins) near 40-50g/m³. 

 

When I look at the heat of adsorption for toluene (approx 500kJ/kg) along with the heat capacity (1kJ/kg) of the carbon and the maximum loading (25%), then I calculate that the maximum conservative adiabatic temperature rise to be approximately 125^oC for instant adsorption. 

 

My query is, for the case of toluene, it appears it can't cause hot spots >125^oC for any inlet concentration in a carbon bed or am I missing something silly, some kind of desorption and absorption process that cycles without providing cooling to the same spot?

 

I've not even factored in the cooling effect of the gas stream which is also significant. 

 

There is plenty of literature on activated carbon beds but a lot of these cases involve aldehydes, ketones, sulphur compounds etc but I feel like I'm missing something from the mechanisms involved as I can't understand how higher temperatures are reached.

 

Note - the vent stream has very low oxygen <1%. Relatively humidity 25%. Temp 20 ^oC Carbon stable below 250 ^oC

 

Regards

Andy

[breizh]

Hi,

My experience with this technology in Resins process was a continuous follow up of the temperature within the beds to alarm the operators of the risk of self-ignition (critical mass).

 

Note 0: the auto ignition temperature of the other components could be much lower.

 

Note 1: activated carbon needs to be regenerated and is very complicated to dispose (my experience in China).

This technology was replaced by biofiltration units (experience in Australia).

 

I will check with local EPA  for approval of the AC technology and will consider other means to treat the VOC, TO for example.

 

Don't minimize the risks of using AC , what will you do if the bed is going under fire? Check also for OPEX.

 

https://cdn.intechop.../pdfs/29382.pdf

 

Note 2: Toluene is a carcinogenic substance, avoid accumulation and risk for people to be in contact with.

 

Good luck

Breizh 

[Pilesar]

My experience with carbon beds is ancient and limited to amine systems. Toluene has a low flash point. Autoignition temperature is high. Fire concern for me would be 'how could the bed be exposed to oxygen to get a flammable mixture'. I don't think you could get too high bed temperature unless there is oxygen and combustion. But if oxygen were available and a fire did start, there is a lot of combustible material there.

[shvet1]

 Note - the vent stream has very low oxygen <1%. Relatively humidity 25%. Temp 20 ^oC Carbon stable below 250 ^oC 

 

1/ AC is planned to adsorb toluene, but are you sure that toluene will stay stable all the time? Is toluene (or impurities) able to condense and form tar, for example during steaming? Note that toluene derivatives have low self ignition temperature and are able to self warming up in contact with air.

2/ Are you sure that air is not able to get in contact with AC? Is inbreathing possible caused by liquid level drop or VOC condensation inside of vent system? How is such controlled and ensured?

3/ Imagine - a worker opens filter for AC replacement and tar caught fire. What a worker shall do? How the whole system will work while personnel fight with burning AC?

4/ AC adsorbs organics layer by layer. Inlet layer the first and outlet layer the last. This means this hot spots and tar will be concentrated at front layer and this front layer is critical for adiabatic heating calculation. Other layers will not take part in adsorbtion&heating, those will provide only final polishing.

5/ If a thin front layer got tar then all AC bed becomes inoperable.

6/ What is planned period of operation of this bed between replacement/regeneration? Who has calculated this time and how and by what criteria/tests? 

Edited by shvet1, 04 August 2023 - 12:47 AM.

[AndyChemEng]

Thank you for your responses.

 

The final system that will be installed will incorporate all standard safety measures, including temperature readings, CO monitoring, SIS bypass/nitrogen cooling/flushing, and oxygen monitor/vent header alarm, in accordance with best practices. I was more curious about the specific mechanism, but based on the above information, I assume that it is due to the unknown or complex chemistry that can occur in the bed of typical chemical facilities, in addition to the usual presence of ketones, aldehydes, and sulphates. Therefore, the standard protection setup approach that is recommended.

 

Thank you again,

Andrew.

 

 

1/ AC is planned to adsorb toluene, but are you sure that toluene will stay stable all the time? Is toluene (or impurities) able to condense and form tar, for example during steaming? Note that toluene derivatives have low self ignition temperature and are able to self warming up in contact with air. 

The solvents in the system are relatively inert VOC, but complex unknown chemistry couldn't be ruled out.

2/ Are you sure that air is not able to get in contact with AC? Is inbreathing possible caused by liquid level drop or VOC condensation inside of vent system? How is such controlled and ensured?

The system is operated with a positive pressure and has absence of flammable atmosphere via less than MOC as it's basis of safety - nitrogen purge flow with oxygen analyser.

3/ Imagine - a worker opens filter for AC replacement and tar caught fire. What a worker shall do? How the whole system will work while personnel fight with burning AC?

Workers are not to open the AC as per supplier instructions (acute exposure risk) and the modular unit is removed as one item. The system will have a nitrogen cooling purge system if high temperature, CO or high oxygen is detected.

4/ AC adsorbs organics layer by layer. Inlet layer the first and outlet layer the last. This means this hot spots and tar will be concentrated at front layer and this front layer is critical for adiabatic heating calculation. Other layers will not take part in adsorbtion&heating, those will provide only final polishing.

Is the above suggesting that the adsorption on the carbon surface shouldn't consider the specific heat of carbon as it's localised on the surface? I'm not sure how activated carbon can do that with it's large surface area being the mechanism?

5/ If a thin front layer got tar then all AC bed becomes inoperable.

Does that safety risk or is this more of an operational issue - backpressure etc?

6/ What is planned period of operation of this bed between replacement/regeneration? Who has calculated this time and how and by what criteria/tests?

The beds are replaced based on breakthrough. Time is estimated based on carbon loading for toluene and as mentioned above, the bed is replaced as a modular unit and chemical break-in is done externally by the supplier.

Edited by AndyChemEng, 21 August 2023 - 06:43 AM.

[dfallon56]

Toluene is used in the printing industry and activated carbon vapor recovery is common. 

 

Will steam be used to regen the carbon? The flowrate if continuous may warrant regen using steam or hot Nitrogen. 

 

Are there other organic gas components?  

 

Very low O2 is good because toluene LFL is 11,000 ppm and UFL is 71,000 ppm.  Relative humidity is important for minimizing carbon temperature rise.  25% RH may be on the low end, max would be 40%.  The water adsorbed onto carbon is evaporated by toluene adsorbing exotherm. 

 

I don't have experience with toluene, but a good company to contact is AMCEC- see dalal engineering link. They can better advise on toluene chemical stability, operating parameters, safety instrumentation. Also shown an EPA link that is helpful. Working capacity of the bed in EPA's report is variable depending on inlet concentration, gas flowrate, inlet gas temp, and max allowed emissions.  A toluene example is included. 

 

 

https://www.epa.gov/..._7thedition.pdf

 

 

http://www.dalalengi...nt_Recovery.pdf

 

 

[breizh]

Hi,

Good to read:

https://nepis.epa.go...ey=2000F9HO.PDF

Breizh