Had an interesting discussion around that with Thomas from NCM in Vietnam when I was reviewing the Bella. Apparently harder surfaces (or more porous ones) alter the resonance peaks in the soundwaves being bounced off them. He was using stainless steel tubing on his flagship (Bella) rather than titanium as he felt the very notional sonic improvement from steel to titanium wasn't worth the additional complication and difficulty in machining the titanium.
The guys at Flare Audio have done stuff on their site in the past about why housing material matters - believe that was all related to resonance as well. If I remember their thoughts correctly, they go for the material with the densest atomic structure, as this introduces the least resonance and therefore gives the least colouration.
I don't know anywhere near enough about the underlying science, but it does sound plausible enough to me. I always equated it with using tonewood on guitars or closed back headphones - the ZMF range and Fistex/Massdrop TH-X00 models all use different wood for the enclosures on their models, and each imparts a different tonality or tuning to the sound, according to multiple comparative reviews.
Pretty sure ZMF have done explanation up as well on what woods do what - imagine if certain resonances either amplify or dampen certain frequencies, this could either introduce or remove a masking effect covering one or more frequencies, adjusting perception of clarity of sound as well as just tonality. Pretty sure that was the whole sales pitch of the Sennheiser IE800 as well - ceramic housing to reduce masking effect.
Apologies for the rambling response - this has always been an area that interests me in terms of the science of sound, so if anyone has any useful links as to why (or why not) material affects sound, would love to see them.