Has anyone here experienced a flash explosion in a PSA pressure swing adsorption tank for hydrogen production using gas? How to deal with it?
Thread Content
Has anyone here experienced a flash explosion in a PSA pressure swing adsorption tank for hydrogen production using gas? How to deal with it?
This post was last edited by altpage on 2022-9-17 at 11:14. Everything exploded; an emergency shutdown was initiated, the emergency shut-off valves were closed, the flare valve was opened, and the material was sent to the flare system. Personnel were evacuated from the area. Remote fire-fighting equipment (fire trucks, fixed fire pumps) was used to extinguish the fire, and help was sought through alarms. PSA hydrogen production units require strict control of leaks; unless there is a large-scale hydrogen leak, the on-site hydrogen detectors often fail to detect it. It is necessary to regularly use portable hydrogen detectors to conduct close inspections of all static and dynamic sealing points, as well as to carry out regular dark-test checks.
It’s not that extreme; the oxygen content in the upstream gas is too high, and the prolonged PSA adsorption period leads to the accumulation of oxygen in the desorption tank. The high-speed airflow during desorption in the adsorption tower generates static electricity, resulting in a flash explosion inside the desorption tank and a sudden sharp increase in pressure (0.15–0.5 MPa). All other PSA processes operate normally.
Oh, an explosion caused by a small amount of oxidizer resulting in limited energy – it does not meet the criteria for an accident, and no equipment was damaged; therefore, it is not an accident to be dealt with, but rather a process adjustment.
Reduce the adsorption time and shift the adsorption front downward to ensure sufficient purity of the new hydrogen
Ensure proper grounding of the unloading tank; install a control valve before allowing the unloading gas to enter the tank in order to regulate its flow rate
After experiencing a flash explosion, it is best to conduct non-destructive testing on the vertical storage tank to check for any damage. The safety valve should have activated as well; it needs to be recalibrated. I’m really puzzled; why would a flash explosion occur if there shouldn’t be any oxygen in the tank? Check whether the upstream feed components contain oxygen ； Check whether the equipment’s static grounding and the flange cross-wire grounding are properly in place ； Reconduct an external leakage check to determine if there are any leaks in the surrounding pipeline and equipment. I tend to think it’s caused by external leakage leading to fire/sparking; I welcome everyone’s discussions and corrections.
On the first floor, it was mentioned that hydrogen is produced from gas, and gas contains oxygen. A few days ago, a performance test was conducted at a plant that produces LNG from gas; the oxygen content in the gas received there ranged from 0.31% to 1.07%. The oxygen concentration increased as a result of oxygen enrichment during the PSA process. At the same time, the lower explosion limit decreases in an oxygen-rich environment, and if the gas is released too quickly, static electricity may be generated, which leads to flash explosions.
This is also due to the excessive oxygen content in the upstream coke oven gas, as well as an overly long PSA adsorption time; this results in oxygen accumulating inside the tank during discharge, thereby causing a flash explosion
After the flash explosion in the transfer tank, the pressure dropped to around 0.5 MPa in an instant. The bonding and grounding were fine; it was still the uncontrolled raw materials that caused the oxygen level to exceed the limit. The long and fast airflow during transfer generated static electricity, leading to a flash explosion inside the tank.