I suspect you're getting ion propulsion confused with nuclear fusion. In order for deuterium and tritium to be useful for fusion, you're going to need kilograms of it, not individual atoms. And tritium is required to be radioactive for it to be useful for fusion, but it has a halflife of about 12 years, meaning any fuel we take with us will be useless pretty quickly. And that's even before you take into account the idea of "constant acceleration"... some sort of magical sustained explosive reaction that would go on for 100 years? And that's after sourcing the unobtanium required to shield/support the "constant" detonation.

I suspect you're thinking of an ion drive, which uses fancy magnets to accelerate mass (pea-sized pellets all the way down to individual molecules), which is a slow start but is fine for constant acceleration over tens or thousands of years.

The problem there though is that the acceleration is SLOW, much less than 1G, hence the 20-80-thousand-year estimate to get to our closest star.

>If you go to 1920s and ask a million people if humanity will be able to walk on the moon within their lifetimes, not a single one would say yes.

I'm not sure I agree with that. Spaceflight and lunar (and even Martian!) colonisation was a pretty common scifi idea by 1920. Cars and planes were already 20 years old by that point. If you asked a million people in 1920 I'm pretty confident you'd get a decent chunk of people saying "yes".