Titanium foil would probably need ribs to get the camber.
Properly ($$$$$) done, a block of it would be machined into the profile for the tunnel.
Getting a sheet 1 molecule thick would be a challenge...
But in essence the need to do this wouldn't exist as we know the limits of a typical microfilm model, and could adjust the tunnel speed to suit.
Getting a turbulence 2 mph flow, and a balance sensitive enough to measure grams of lift or milligrams of drag would be interesting.