I'm having difficulty organizing my thoughts on this subject.

Game is HU and rake per hand is 10bb/100.
Room policy is "no flop, no drop."

Because you only pay rake when seeing a flop, it is preferable to open 3 big blinds.

However, you face either of these opponents:

1) Player A, against a 3x, defends 40% of his big blinds.
2) Player B 3bets 50%.

Both players give you incentive to decrease your opening size. However, if you start minraising with a wide range and either player adapts, all players are now probably playing a -EV game. It can take a long time before your stats suggest your opponent adjusted. By the time you realize it, the rake potentially kills any positive win rate you may have had.

So what's the solution? Should you take advantage of an opponent's perceived leak and risk playing a -EV game by minraising? Should you stick with the original plan of 3x-or-folding and exploit in other ways?
I have my own hypothesis but I'll wait before influencing.

By 10bb/100 per hand I'm assuming you mean 10% rake. Not sure what else it could mean so correct me if I'm wrong.

Rake is a percentage of what is in the pot so the raise size doesn't matter. I'd Stick with whatever you normally do.

Player A and player B are very exploitable. I'd play both of them. Player B is especially making a mistake.

Versus player B you can raise 3x wait for the 3-bet. Jam all Ax hands and pocket pairs. Flat everything playable. Fold the garbage. Be slightly tighter preflop, but not too much tighter. That would be my game plan. I don't think there is too much incentive either way to raise smaller or bigger. If you raise bigger you probably get a larger 3-bet hence a bigger pot with the same weak range. It actually makes it easier to not make mistakes.

Versus player A I'd just min-raise my regular range. I may end up not folding T4o and other such hands that I may otherwise just fold. I wouldn't raise everything just because I don't want them to adjust. On low card flops, especially ones that paired I'm going to give player A a hard time because chances are they aren't playing many low cards.

To answer the last question. Exploitation is always a risk that you could be playing a -EV game, but you are taking a chance that your reads are correct. I am a fan of exploitation, but you don't want to go overboard. Don't over adjust and you don't really have to risk playing a -EV game, especially against these players. If you are attentive you'll notice change, especially with player B. In fact, after a few stack offs. I would expect player B to adjust.

if you have a player that tries to counter your exact strategy that's just what good poker is.

Play the strategy that makes the most money despite the fact that you're countered. This is the core tennet of GTO in a heads up scenario.

FWIW rake affects the BB by making him fold more. the increased fold equity that BTN has is cancelled out when he sees a flop - he wins less because of rake.

For clarity, do you mean he folds 60% of the time?

Rake structure is say 5% with a 1bb cap. It works out to be 10bb/100 when you look in PT4.

Yes. He distributes his call/3bet in the remaining 40%.

You can obviously open any two for 3x if that's the case.

I don't see why we'd want to open 2x instead of 3x vs players like these though. I don't see any logic in it.

He is too tight out of position. You can potentially increase your EV by stealing his blind for cheaper.

If you decrease your open size, he won't just keep defending the same range, right? He might be playing much better vs 2x than vs 3x as far as you know.

PL or NL? You almost definitely want to adjust against the 50% 3bettor by transforming the bottom of your raising range into folds and limps.

vs. the underdefender, obv. want to raise 100% of hands but it's not clear that you should use a smaller raise size. Almost all opponents will defend more hands against smaller raises and determining the optimal size requires a solver.