Doesn't it depend on the context? I don't understand why you think it might be un-scientific. "However, at temperatures above 120C, the solution is liable to explode catastrophically". That might be worth saying in a scientific document.
Other literary phrases have a similar effect when the intent is to inform the reader. They also tend to mentally "flag" important phrases so that they are less likely to be overlooked. The word "however" especially gives the reader an immediate hint that the next phrase or sentence is somehow different from what has come before. A contrast is being presented. It "primes" the reader to consider the distinction. Other such words also have a traditional "flagging" use, though different from that of "however".
The language is what it is. But each field has certain phrasing that is common there and the writing might seem awkward when it is abandoned. I'll guess, without evidence, that "however" is less common in math writing, than in, say, philosophy, where alternative arguments might need to be introduced. In mathematics, a paper will tend to drive inexorably toward a conclusion. However, in philosophy alternate views are often introduced so as to find a way to some truth. (You see what I did there?)