More interesting to me is, given that people (I'm not sure who) say that in principle a thermonuclear bomb can have unlimited stages and size, would it be practical to build a "regular" bomb several orders of magnitude bigger than Tsar Bomba that would essentially blow up the whole planet. Let's say > 10x the dinosaur killing projectile.
If you read the article, you'll find that the Earth is extremely hard to blow up. It's an enormous ball of iron, and almost any explosion at the surface is unlikely to affect it much. You'd have to detonate a bomb close to its centre, and that bomb would have to be astronomically large. Remember that the Earth got hit with a presumed Mars-sized object in the past, yet Earth is still around (the debris formed the Moon). Such a collision might eradicate all life, but it wouldn't "blow up the planet".
Well, yeah, I would consider making the top mile or two of crust molten "close enough for government work". I didn't mean dispersing a cloud of debris to infinity.
After the collision that created the Moon, I believe some research suggests that the Earth was literally as hot as the surface of the Sun, deduced from evidence of the effect of "earthshine" on the Moon.
If the Chicxulub impactor resulted in a global rain of molten rock, I'd also consider that scale of explosion to be "enough" even though some living things survived.
It would need to be many, many orders of magnitude bigger. Chicxulub was approximately 10E6 more energetic than Tsar Bomba, and it didn't even manage to kill off all life. And that barely left a dent in the surface of the planet.
Also worth noting is that bigger bombs don't scale particularly well in terms of damage. Double a bomb's size, and much of it immediately just goes to outer space. Another large part of it is reflected from the surface of the planet to go, again, to outer space. Damage at radius r ~ E^(1/5).
