The primary change in the calculation with CO2/H2S acid gas is that the mixture can be highly non-ideal and may even cross into the supercritical zone where it might perform more like a liquid than a gas (depending upon temperature and pressure). You therefore need to watch the fugacity constant, which can vary a long way from one for acid gas. By comparison, sweet, low CO2 natural gas behaves much more like an ideal gas. Fugacity and non-ideal behavior is less of a problem unless the composition is so heavy that you are near the critical conditions for one or more of the component compounds.

Compressor performance and behavior depends on the composition of the inlet stream. If you have a look at basic compressor equations you will find parameters such as the gas constant, compressibility factor, adiabatic index (or polytropic index as the case maybe). This factors all depend on inlet gas composition. When the inlet gas composition changes, you can expect that all these parameters will change and this will impact on overall compressor performance including the input energy rate and the delivery pressure. How significant the change will be will depend on the amount of change in the inlet composition other parameters being the same. Going back to your question, they will definitely be a change with the two compositions you mentioned and what can give you an idea of which way the change will go is the molecular weight of the inlet gas stream....

@Stephen Smith are you suggesting the presence of the acid gases will lower the critical point of the mixture. Sounds likely but I think the problem of entering the critical point can only rise if the gas is compressed to a pressure above or around the critical point. Otherwise, I do not imagine that the issue of entering the supercritical region will be any problem. Assuming the critical point is 40 bar and the gas is compressed to just 20 bar, there will be no issue about entering the supercritical zone. Also, I do not see fugacity as a problem in gas (single phase) compression. Can you throw more light on how fugacity affects gas (single phase) compression. Thank you

@ Eni - The term "acid gas" is generally used to describe a gas that is comprised primarily of CO2 and H2S. The critical T/P for CO2 are 304K/7.4MPa and for H2S are 373K/9MPa, so it is quite probable that a compressor will exceed the critical conditions for a CO2/H2S mixture. As for fugacity, CO2 and H2S in the gas phase tend to be far less ideal than hydrocarbon mixtures or steam. I have frequently seen fugacity constants for these chemicals on the order of 0.2 to 0.4.

Thank you Steve for your response. My point however is that exceeding the critical condition for the gas mixture depend on process design. If the compressor maximum delivery pressure is 4MPa for instance, do you think they will be such possibility to entering the supercritical zone given that the maximum delivery of the particular compressor is very far way from the boundary. Again, the critical T/P you gave is for CO2 and H2S. It will be interesting to know the critical T/P of the mixture and the impact of CO2 and H2S on the critical T/P of the mixture i.e if the presence of CO2 and H2S increases or decreases the mixture critical T/P. I think it's rather hasty to conclude that critical condition will be exceeded without considering what the mixture critical point is and design delivery pressure of a compressor.

Having said that, on fugacity I see your point. This will be the case on the grounds of a real system in contrast to an ideal system. I agree with you. But when like it is usually the case in many situations, the compressor is designed based on ideal gas law, I am not sure this will be a problem.

@ Eni, I have the advantage of having worked with the design (materials selection) of a couple of acid gas compression trains. In the first stage of compression, criticality is only an issue with respect to temperature (which will be far above Tc at the outlet) so it is not as important. However, these systems are generally designed as multistage compression so that the "gas" is compressed to a high enough pressure for injection back into the earth, which can be 13 to 35 MPa (or more). In the case of systems that I worked on, the CO2/H2S stream at the fourth (and final) stage exceeded both the Tc and Pc and had fluid characteristics that allowed it to be pumped to the final pressure rather than compressed. A pump was far less expensive than another compressor.

As for the changes in Tc and Pc, I agree that these will change with the fluid composition. It would be best to have a thermodynamic modeling program optimized for acid gas conditions calculate this change. However, an old procedure described in the Gas Processor Suppliers Engineering Data book indicates that Tc can drop as much as 19C for a high H2S, low CO2 stream.

So the moral of the story is that you cannot treat acid gas compression like you would a hydrocarbon or steam stream if you want to get "optimal" results. The design needs to be carefully adjusted to account for non-ideality.

Thanks a lot for all your technical response. I think we need to have a series of comparative calculations using a technical simulator. Do you have any suggestion for design of simulation runs?

@ Stephan, What type of pump was suitable to be replaced by a compressor for the stages that exceed the critical conditions?

The pump was a centrifugal pump. The properties of the supercritical fluid were near the critical pressure and the temperature was low enough following cooling that the properties of the supercritical fluid were more like a liquid than a gas. If that does not make sense, then you really need to consult with someone who is very familiar with the properties of supercritical fluids during your design.