Painstaking reconstruction of the 9/11 attacks reveals much detail about the process of destruction.
America's National Institute for Science & Technology (NIST) revealed more detail than ever of its computer based analysis of the Twin Towers disaster of 11 September 2001 last week.
The ndings were presented and discussed at a conference on its campus at Gaithersburg near Washington DC. Not everyone agreed with everything that was said, but the findings of one of the most comprehensive structural failure investigations in history were fascinating to witness at first hand.
Million node structural and fire performance models of both towers and of the Boeing twin engined, wide bodied passenger jets which smashed into them enabled NIST to reproduce almost all the observed phenomena of the collapses.
And what became evident was the complex role of the fuel in the tanks of the Boeings at the moment of impact and in the microseconds that followed.
Without the fuel the thin skinned, largely aluminium planes would have been shredded by the protective 'fence' of high strength steel perimeter columns on the Leslie Robertson-designed towers. Undoubtedly, the engines could have caused serious damage - NIST's simulations showed that a single engine had enough kinetic energy to smash through the perimeter 'fence' and still take out up to two columns in the core. And the tough steel landing gear was capable of considerable penetration as well.
But on impact each plane's wings and fuselage would have been reduced to aluminium confetti.
This, according to NIST, is precisely what happened to the outer wings and the front section of the fuselage. But inside the inner wing sections and the fuselage fuel tanks of both aircraft were some 38,000 litres of fuel.
The concentrated inertia of this fluid mass converted the fragile tanks into 800km/h sledgehammers. The perimeter columns yielded and failed, creating the gaping wounds whose iconic images still grip the attention of the world.
The same impact ruptured the fuel tanks. NIST's recreation of these moments assumes around 30% of the fuel released was consumed in the enormous fireballs which roared through both towers. Its model shows that most of the fuel cloud was confined within the impact floors.
NIST made no comment on why both Boeings were banking sharply to the left at the moment of impact. The result of the angled impacts was to spread fire and ruin over a number of floors, greatly increasing the initial loss of life.
The impacts and fireballs produced shockwaves and high velocity debris. NIST believes that fire-proofing in the impact zones was largely scoured away by the debris and the shockwaves, and that this loss of protection was the key factor in the ultimate collapses of both towers.
Other experts disagree (see News). But what is largely unchallenged so far is NIST's contention that on the impact floors the violent releases of kinetic and chemical energy blasted much of the floors' contents across to the far side of the buildings, where they were churned together with the debris from the planes and significant quantities of unburnt jet fuel.
This debris pile effect was more pronounced on WTC2 than on WTC1, NIST believes, due to the different geometries of impact. Add to that the greater number of windows blasted open by the fireball on WTC2 and the probable extra damage to the core columns, and NIST's explanation of why WTC2 collapsed so much earlier than WTC1 - even though it was struck 18 minutes later - is complete.
The debris pile in WTC2 was bigger and had plenty of oxygen available when it burned, so it burned hotter. Floor trusses and columns in this area were affected by the fire sooner. As a result, the hat truss at the top of WTC2 had a much harder job to redistribute loads from severed and softened columns, and failed sooner.