The hydrogen production unit has the problem that the outlet level of trace carbon dioxide is higher than the inlet level

Although I’m not sure what it means by the imported carbon dioxide content being 5 ppm, in terms of the PSA purification process, carbon dioxide is an adsorbed component; therefore, the carbon dioxide content in the resulting hydrogen should not be higher than that in the incoming gas. It is recommended to verify the method used for sampling and analysis. Inadequate replacement of the sampler during sampling can lead to errors. Additionally, it is advisable to calibrate the analyzer using standard gases. Whenever the carrier gas used in the analyzer is changed, it needs to be recalibrated in order to avoid errors in the analyzer’s readings.

It’s the same analyzer; currently only one column has been replaced. Every time this column is used for adsorption, the carbon dioxide level increases, while it remains unchanged when the other columns are used for adsorption.

In this case, it is determined that the carbon dioxide adsorption capacity is insufficient, allowing carbon dioxide to penetrate through the adsorption bed and reach the product gas; in other words, the amount of silica gel or activated carbon used is too low. Molecular sieves have a higher adsorption capacity for carbon dioxide than silica gel or activated carbon, but desorption is more difficult (or slower). The imported carbon dioxide level you mentioned is 1 ppm, while the exported level is 5 ppm; such parameters are still not understandable. The adsorption of impurity gases by the adsorbent is relative and depends on the adsorption time and space velocity; if the adsorption time is long, the impurity gases will penetrate the bed once they reach saturation, and an excessively high space velocity will also cause them to penetrate the bed as they cannot be adsorbed in time.

I suspect that the adsorbent filled this time is a regenerated one. Previously, the adsorption towers were filled mainly with silica gel, with activated carbon as a supplement. Since methane was introduced into the system, some of the silica gel was removed and replaced with activated carbon. So far only one tower has been replaced, and this problem has occurred. Subsequently, the tower containing the replaced adsorbent was removed, evacuated for a few days, and then reintegrated into the system. During the first few operating cycles after it was reintegrated, the carbon dioxide concentration at the outlet increased from 1.0 ppm to 1.2 ppm; as time went on, this concentration gradually rose back to its original level. Therefore, I suspect there is an issue with the quality of the adsorbent. I was wondering, is it possible for my hypothesis to be true?

Generally, the adsorbent should be replaced entirely with new material; if reused, the existing filler that is removed must first be activated before it can be used again. This is especially true for molecular sieves, which absorb moisture from the air when material is drawn out, and it is very difficult to remove this moisture from molecular sieves at room temperature. If the molecular sieve is reused, its adsorption capacity may decrease. As the methane content in the feed gas increases, the original design team should recalculate based on the composition of the modified feed gas in order to determine the volume of packing required.