You may first want to look at this at a larger scale before you consider SEM. Thermal cracks often occur early in the age of a structure, while sulfate-induced cracking takes time usually to develop. When you say cracks caused by sulfates it could mean many different things: classic sulfate attack expansion due to formation of ettringite from the intrusion of external sulfates, delayed ettringite formation (which can for example happen in mass concrete and occur at the same time with thermal cracks), or it could be physical salt distress. So if you see abundant ettringite needles and gypsum using SEM it could hint at a sulfate-related issue. But the SEM will not tell you much about thermal cracks. For thermal cracks, you need to look at the mixture design, geometry of the structure, boundary conditions, solar radiation, wind, and ambient temperature history. It is a different type of analysis. Good luck.
The crack system you described is perfectly correct, but we have a series of cracks in the concrete when
Concrete setting is visible under the SEM machine and is spread over the entire volume of concrete.
In my opinion, the shape of the crack is very important. The setting cracks are spreading from the center to the periphery, but the sulfate cracks are linear and continue from the outer edge of the concrete into the concrete, and it takes a long time to reach the center of the concrete. The issue also depends on the amount of sulfate.
In forensic engineering, reading cracks is important. But you must capture the mixture design, curing conditions, exposure conditions, etc., in addition to Non-Destructive Testing and Laboratory Testing to find the root cause of the problem with a decent degree of reliability. The information you provided is not sufficient to reach robust conclusions. I was trying to suggest that you capture those important aspects and not just the pattern and direction of cracks.