The convention in scientific writing, at least in the hard sciences, is to avoid "I" even for single-author papers. I suspect (but can't prove) that this is why you see so much passive voice in such papers ("the doohickey was then frobitzed to induce a somethingorother reaction").
According to this well-received answer on Academia, you can view use of "we" as an editorial "we" or "we, as in the author and the readers". The latter approach works better for descriptive writing ("we see the following results...") than reporting ("we did X").
Ultimately you should base your decision on the submission requirements of the institution where you intend to publish the paper. But in general, "I" is uncommon, "we" is used even for single-author papers, and you can use "we" in a way that doesn't have to seem weird.