How do you get the pressure of 6 bars at the end of the reaction?
I don't have SW so I can't look at your model and, anyway, I don't know this tool... On the other hand, the result of Flowsimulation seems pretty good to me.
Sorry but I basically come back to my ideal gas equation P.V=n.R.T. Here, V, n and R are constant so P=a.T. At t0, P0 = a.T0 with P0 = 3 bars and T0 = 273 °K. A tf, Pf = a.Tf with Tf = 273+90=363 °K. As a result, Pf = (P0/T0). Tf = 4 bar approx. (3.99 exactly). So, if you have a result at 4.1 bars, it would be consistent.
OK, I understand your mistake better. In fact, it's your starting point that is wrong because you took it in the "liquid+gas" zone (inside the black curve) and not in the gaseous state only (right part of the graph). So, if you recalibrate your starting point and follow the green line, you fall back on about 4 bars. I've put all this back on the graph.
On the other hand, if you have to use this graph for your dissertation, be careful because it also tells you that your initial hypothesis is not entirely accurate. Indeed, if you look closely, you see that you can't be purely gaseous at 0° and 3 bars. Either you keep 3 bars but your starting temperature will be slightly higher or you keep 0° but the starting pressure must be lower.
The data you found are probably more likely to be obtained at constant enthalpy and not at constant volume, which explains the difference in values.
In a refrigerated circuit, the opposite phenomenon is used to the one we are interested in here. In other words, the pressure (compression or expansion) of the gas is used to increase or lower its temperature.
Indeed, I had forgotten this detail. As a result, if there is a phase transition, all this is no longer valid.
On the other hand, given the consistency of the results obtained by taking only a gaseous state, it seems quite clear to me that FlowSimulation made the same error of initial condition.
So, I think you need to look for ways to better define your starting point (liquid/gas proportion, enthalpy, ...) so that the change of state is properly taken into account.