Ati: I hope the biphasic schedule works well for you! (I can't see any reason why it wouldn't.) Be sure to post your results.
In answer to your question, I've heard about the 90-minute cycle in sleep from a few different places, but I got the idea of using it to design a sleep schedule here
Granted, the author is really exaggerating when he says "Leonardo Da Vinci, Thomas Jefferson, and Buckminster Fuller used this *exact* technique"--Da Vinci and Fuller slept polyphasically, but with naps of 20 minutes and 30 minutes, respectively (I don't know about Jefferson). The reason they were able to still get all the stages of sleep is because of the nature of polyphasic sleep adaptation: after two or three weeks, a polyphasic sleeper's naps cease to follow the normal sleep cycle and instead visit whatever sleep stages are necessary in order to get the right proportion of each stage of sleep (75-80% Stages 1-4 NREM, the rest Stage 5 REM).
But in a person not currently adapted to polyphasic sleep, a nap interrupted between 40 and 70 minutes is at high risk for waking the person in the middle of Stage 3 or 4 (deep sleep), which is where waking up is most difficult and disruptive. For that reason, and because a 90-minute period of sleep has room for all the stages of sleep, I would agree with the author of the article that it's best to stage your sleep around multiples of 90 minutes. If, on a given day, you decide to take two naps instead of one, it would actually be better to do two 30-minute naps than two 45-minute naps.
I admit that most of my information comes from reading on the Internet: some from sleep experts, some from Wikipedia, but also a lot from bloggers and ordinary people. I tend to assign the most credibility to the information that's most consistent with what I know, but I'm often surprised when new information gives me a better explanation for phenomena.