I asked them the same question, and they said that any "heavy" metal might enable this polymer to be both fluorescent and phosphorescent, platinum was just the best candidate. And in the next phase of their research they plan to try out others.
The research has shown a very positive results in terms of efficiency and colour control, but infact this will still require many to find out the real white out of it. As none of the experiments have resulted in a white colour outputing LED. If a cheap metal can replace platinun it can be really an affordable light.
These researchers are shooting for OLED applications--backlights, light bulb replacements and such. They also claim their approach will be lower cost, because of the extreme simplicity of their design--no pixels or phosphors, but its too early to say whether longevity will be enhanced too.
White is "white" no matter how achieved. That said I don't really see how their design is "whiter" than other white LEDs with multiple band-gaps that create overlapping spectrums that add up to white. Effectively that is all they are doing. It just sounds like academic double-speak, not to downplay the accomplishment.
One of the posters talked about getting past 50%. We are already there. State of the art white LED (blue pumped phosphor) are already at or greater than 50% wall plug efficiency. The blue die on its own is hitting 60%.
Why? There is really no justification for that at all.
Without intimate knowledge of a scene and that same scene having lots of red, most people would not be able to tell if the light source was 85 or 95CRI. Side by side comparison, maybe, but without a reference no.
CRI is not a great measurement of color fidelity anyway. CRI is only a measure effectively of how well a light source either matches a blackbody radiator at some CCTs and sunlight at others. Given the vast amount of 4000K lighting, there is no blackbody reference comparison or sunlight for that matter than anyone could really make.
You can make lower CRI sources look better than high CRI sources depending on the color gamut, and spectral pattern.
If you want to mimic an incadenscent bulb perfectly great, but the prevalence of 2700-3000K sources in terms of evolutionary biology are pretty near nil. Even when sunlight is in the CCT range, it does not look blackbody.