I was wondering why is so frequent to hear explanations in science relying on what -in philosophical terms- is called reductionism.
Do you have any idea why explain and reduce it to minimal part of matter is used so frequently?
I'm not very sure this is always the case. Take the explanation for why a house burnt: it was made out of inflammable materials and there was a spark. These are not constituents of the fire. Or take the evolutionary explanation of why giraffes have a long neck. This involves environmental selection, which are not part of the giraffes'neck. It's even more easy to find non reductionist explanations outside of science (involving, say, God).
There's a huge literature on explanations in philosophy of science. Some think that explaining is giving a causal mechanism for a phenomena. Others that it appeals to a theoretical framework that unifies the phenomena to be explained with other phenomena in a common scheme. In any case, reductionism seems to be part of the way causal mechanisms are generally presented, or a characteristic of unifying theoretical frameworks, but not specific to explanations. I would say that what gives you this impression is that most theories function this way: they unify various phenomena by analysing their constituents, or they allow one to describe mechanisms where part constituents interact. But that's a characteristic of contemporary science that is not specific to explanations in general.
As you say in your preceding comment above, "because it works" is the sine qua non for scientific explanation (or, for that matter, for any conjectures that purport to be "explanations"). Beyond that, Occam's razor, whereby reductionism comprises the least complex scientific conjectures that "work".
Moreover, reductionism is the only approach that works. Can you suggest any other? In the extreme, you could individually describe every observable phenomenon in complete detail. But that's just a description, not an explanation. Indeed, an explanation has to be simpler than a straightforward description of the phenomena it purports to explain. If not, what do you mean by "explanation"?
And if explanations must be simpler, then why wouldn't even simpler be better, and simplest best? And that's all reductionism is saying, simpler=better. Maybe you can argue that's just a human intellectual bias, but then you still have to suggest an alternative definition of "explanation", and then come up with explanations that actually "work". Offhand, I see no reasonable way to do either, much less both.