I believe it is not necessary to put more taking point at hysteresis area as well. The main point is, make sure adsorption and description is well overlapping each other at the beginning and end of graph?
Do you have any opinion or suggestion if looking at this graph?
Note: This is silica alumina standard (MSDS)
It seems you are using a Micromeritics (Gemini or TriStar?).
For BET multipoint surface area you have enough points, as Christian said above the region of interest is 0.1 < P/P0 < 0.3
Check your BET plot if the linear fit is good, it should be.
About the crossing of the branches, Christian is also correct: try longer or stronger degassing first. Probably, it is not a problem with your equipment since overlap is very small.
Attached is how the same sample looks in our ASAP 2020. As you can see , it is almost the same.
Usually if the sample is non-porous the linear región for application of the BET equation is from 0.05 to 0.35 p/p°. When the samples are mesoporous the upper limit is reduced, especially as the mesopores become smaller. To see where the upper limit of your BET linear portion is Rouquerol has suggested plotting Vads(1-(p/p°)) versus (p/p°) which gives you a curve where the maximum is the highest (p/p°) value that you should consider.
If the sample is microporous the linear región is normally located from 0.02 to 0.12 p/p°, thus shorter and displaced to lower relative pressure values. Again, the method of Rouquerol confirms this.
Best regards,
Malcolm