Yes you can. See Standard methods for examination of water and waste water, 2017. If 0.45 micrometer is acceptable, then why not 0.22 which is even a smaller pore size.
Ideally, your sample pre-treatment is determined by the question you want to answer. Do you want to know the total amount of heavy metals in the water phase, then you should not filter at all. Do you want to compare your results to maximum acceptable levels, then you should pre-treat your samples according to the way the level is determined (is the level determined for 0.45 um filtered samples, then try to stick to 0.45 um). Furthermore, the size of the difference between 0.45 en 0.22 um filtered samples will depend on the type of water. If I remember correctly, the few times that we used both kind of filters on Dutch surface water, the difference was small.
Gerlinde Roskam thank you for your answer, in fact I am analyzing the heavy metals in the soil, I made acid attack, when I have to pass it on ICP, I have to filter it
Nouha Khiari Do test different membrane materials. I would not be worried about the size of the pores (0.22 will have flow issue for viscous sample). I would be worried if your samples are acidified prior to filtration. Filter membrane do contain traces of metals and acidified samples might elute some of the metals from the membrane or in the worst case scenario, the membrane might adsorb appreciable metals. Teflon or PTFE would be less leaching or less metal-adsorbing, I guess. also see