The extreme cold of the night before had chilled the rubber insulating O-rings in the right solid rocket booster's aft field joint (the joint near the SRB's lower supporting strut that attached it to the external fuel tank) well below their specified operating temperature.  Because no shuttle had ever been launched in such cold weather, engineers at Morton Thiokol (the company that manufactures the boosters) had no data to say just what would happen in such conditions, although their examination of retrieved boosters from previous launches had led them to conclude that lower launch temperatures directly affected the O-rings' ability to seal the joints properly upon SRB ignition.  This sealing failure was called "blow-by" and was indicated by evidence of hot gases from the firing of the SRBs having eroded the O-rings during flight.  The lower the temperature, the greater the blow-by.  In fact, one SRB used on a 1983 Challenger launch had actually sustained a complete O-ring failure resulting in hot gases escaping from the joint, but the moment of failure had occurred after the SRBs had separated from the external tank.  If it had happened half a minute or so earlier, we would today be talking about the 1983 Challenger disaster.  Nevertheless, Morton Thiokol engineers could not categorically say that the seal would fail if the shuttle were launched in sub-freezing temperatures, only that it might, and so their "no-go" recommendation was overruled by Morton Thiokol managers, who in turn gave the "go" to NASA managers.  In fairness to those who approved the launch, evidence of "blow-by" had been observed during launches when the air temperature was as high as 73 degrees, and they argued that "blow-by" appeared to happen no matter what the temperature was (of course, this should have been taken as evidence that the entire O-ring and joint configuration needed to be redesigned anyway).

When the SRBs were ignited at liftoff on January 28, both the primary and the secondary O-rings at the right SRB's aft field joint failed as engineers had feared they would.  Evidence of this failure took the form of high-speed video images that showed seven discrete puffs of dark black smoke emanating from the area of the right SRB's aft field joint during the first two seconds after liftoff.  However, the leak apparently sealed itself after that, probably plugged by debris from burned fuel and the failed O-ring seals.

This temporary plug appeared to stay in place for the next 56 seconds until the shuttle entered the zone of maximum dynamic pressure, called Max Q, during which shuttles ordinarily experience increased atmospheric stress as they climb toward orbit (later data showed that the stress Challenger endured that day was greater than on any previous launch but still within design specifications).  At 58 seconds after launch, however, long-range video showed a flicker appearing in the area of the right SRB's aft field joint and quickly expanding to a continuous plume of flame; the stress of going through Max Q coupled with the throttling up of the shuttle's main engines apparently had broken the debris plug at the joint, allowing a 6,000 degree flame to escape through the side of the booster.  The slip stream caused by the shuttle's nearly 1,500 mile-per-hour speed deflected the plume toward the SRB's lower supporting strut and the external fuel tank.  The increasing loss of chamber pressure in the right SRB due to the leak caused the shuttle's guidance system to compensate for the loss of thrust on the right side by swiveling the shuttle's engines and the SRBs' nozzles in an attempt to keep Challenger on course.  At 66 seconds after liftoff, data showed a significant loss of fuel pressure from the external tank to the main engines, indicating a growing hydrogen fuel leak, which fed the growing flame from the right SRB's failed joint.  By this time the SRB's supporting strut had been severely weakened by exposure to the flame, as had the surface of the external tank.  At 70 seconds after liftoff, long-range video showed a circumferential leak of hydrogen gas about a third of the way up on the external tank, indicating that the hydrogen innertank had failed.  A bright, sustained glow also appeared between the external tank and the underside of the shuttle.  At 72 seconds, data showed extreme movement of the right SRB relative to the left booster and the shuttle, indicating that the lower supporting strut had broken away completely.

At this point several simultaneous events occurred that resulted in the destruction of the shuttle.  Probably at the same time that the SRB supporting strut failed, the lower third of the external tank fell away, releasing the hydrogen innertank's remaining load of liquid hydrogen.  This release propelled the upper part of the hydrogen innertank upward into the liquid oxygen tank above it.  At about the same moment, the right SRB pivoted around its remaining upper strut, its nose cone smashing into the top of the external tank.  This resulted in the release of all of the remaining liquid hydrogen and oxygen fuel, which vaporized instantly in the thin atmosphere nine miles up.  This sudden fuel vaporization produced what appeared to be a fireball or fiery explosion but was really a combination of reflected sunlight, radiance from the brightness of the SRBs' exhaust nozzles and some local burning of gases within the expanding vapor cloud.  Because of the right SRB's pivoting around its upper attachment, its motion suddenly pointed Challenger to the left, with the result that the shuttle was no longer pointed in the same direction that it was flying.  The resulting aerodynamic stress from this "broadside" effect at nearly 2,000 miles per hour was more than the shuttle was built to withstand, and the forward part of the shuttle broke away from the payload bay.  The nose of the shuttle became separated from the crew cabin, the steering rockets in the nose releasing their fuel in an explosive burn.  The rest of the shuttle, its forward end suddenly opened like a tube to the supersonic wind, blew apart from the inside out.  All of this happened within the space of a second or so.  Contrary to initial speculation, there was no actual explosion (in spite of Tom Brokaw's citing of scientists at the time who likened the force of the "explosion" to that of "a small nuclear blast"); what we saw was the dramatic vaporization of Challenger's liquid fuel in the thin atmosphere as the shuttle broke up under severe aerodynamic stress.  Challenger was not blasted to pieces by an explosion; it was blown apart by aerodynamics.

The crew cabin's momentum after the breakup quickly carried it upward to an altitude of around twelve miles before aerodynamics and gravity slowed its ascent and the cabin began the long fall to the ocean.  What happened to the crew after the breakup and during the fall will never be fully answered.  When the cabin broke away from the rest of the shuttle, it lost all its electrical power and oxygen supplies.  If the cabin depressurized due to the breakup, then the crew would have quickly begun losing consciousness due to lack of oxygen.  The fact that three out of four recovered PEAPs (Personal Egress Air Packs) had been activated and partially used indicates that at least some of the crew survived the breakup long enough to take some action to try to stay alive.  Their having turned on the PEAPs does not prove that the cabin lost pressure but does show that at least some of the crew, all of whom were wearing air-tight flight helmets, believed that pressure had been lost or was being lost (otherwise, the packs wouldn't have been activated).  If the cabin indeed lost pressure after the breakup, then the crew in all likelihood lost consciousness, although the short time between breakup and the cabin's impact with the ocean means that, barring cardiac arrest, they were alive but unconscious when the cabin hit the water.  If, on the other hand, the cabin maintained its pressure, then they were likely alive and awake until the end.  A third possibility is that the cabin depressurized at the altitude of shuttle breakup but repressurized as it fell into the denser atmosphere near sea level, which raises the nightmarish possibility that the astronauts passed out after the breakup only to regain consciousness in time to see the ocean's surface racing toward them at 200 miles per hour.  Regardless of whether they were conscious or unconscious during the fall, any astronauts still alive died instantly upon impact with the ocean's surface.  Damage to the cabin from its hitting the ocean's surface made it impossible to determine what damage, if any, had happened to the cabin during the breakup, which is why it's impossible to say just what happened to the crew during the fall.  (The alleged "transcript" of the crew's fall to the ocean that has been published in tabloid magazines and on some Internet sites is only a hoax.)