There is technical data on their website, under Downloads there is a white paper.
However, their own white paper comes up with numbers which are internally inconsistent.
They mention a typical "connected device" with a CR2032 battery with a lifetime of 6 months, and calculate its power consumption at an average of 165uW. That's about right, and it corresponds to 14.25J/day.
Later, they say:
the average RF density measured in an office or external environment ranges from 20 to 35 nW/cm2. This power density yields an estimated harvested energy between 5.5 and 10 Joules over a one-year period.
That also seems realistic. So for a realistic product in a realistic use case, Freevolt can provide less than 0.3% of the product's energy requirements.
They also say this
a low power commercially available temperature and humidity sensor requires approximately 3.2 ?J to take a reading
And on this basis calculate that Freevolt can provide 4000 additional readings per day. But this is entirely misleading. It is true that e.g. the Silicon Labs Si7021 only requires about 2uJ to take the lowest-resolution reading, but the energy required to read the value over I2C dwarfs that figure, as does the energy required to transmit those readings using a radio link.
There are numerous papers about, and even commercial development kits for ambient RF energy harvesting. The technology certainly is real, and it has its place, but I have my doubts about the claim that the CleanSpace tag will never need recharging.