A commercial rocket flying cargo to the International Space Station exploded just 14 seconds after lift-off last night, creating a spectacular fireball and shockwave as the fully fuelled rocket crashed back down onto its launch pad.
No one was hurt when the uncrewed spacecraft was destroyed, but the loss will dent attempts by the spacecraft's builder, Orbital Sciences Corporation of Dulles, Virginia, to build confidence in its ability to supply ISS cargo services to NASA.
After a one-day launch delay owing to an unauthorised boat breaching the no-go area around the Mid Atlantic Regional Spaceport at Wallops Island, Virginia, Orbital's Antares rocket was launched at 18.22 EDT (22.22 GMT) on 28 October. It seemed to take off normally but its progress quickly slowed as it lost thrust and the rocket came to a halt after 14 seconds, just after clearing the launch tower. It became engulfed in flames and exploded as it plummeted back down to earth.
"The rocket's ascent stopped and there was some disassembly of the first stage" before the explosion, says former NASA astronaut and ISS commander Frank Culbertson, now vice president of Orbital Sciences, in a post-incident press briefing at the coastal spaceport.
Culbertson also says the spaceport's range safety officer – whose job it is to destroy the rocket if it veers off-course and threatens to hit a populated area – is thought to have sent a signal to destroy the crippled Antares when it became clear that the launch was not proceeding as planned – so it blew up before hitting the ground.
It is not yet certain whether the range safety officer did take such action.
Anomaly investigation
Detailed analysis of high-resolution launch video, coupled with examination of telemetry data from the rocket itself, will now play a central role in analysing what happened, Culbertson said. The company will now lead an with NASA to find out what happened.
At NASA, William Gerstenmaier, associate administrator for human exploration and operations, defended Orbital's record: "Orbital has demonstrated extraordinary capabilities in its first two missions to the station. Launching rockets is an incredibly difficult undertaking, and we learn from each success and each setback."
The Cygnus cargo freighter atop the doomed rocket was carrying astronaut supplies, science experiments and a raft of CubeSat satellites, including 26 earth observation satellites from Planet Labs of San Francisco, California, and the first asteroid mining technology demonstrator satellite from Planetary Resources of Redmond, Washington.
Culbertson says Orbital regrets the loss of the valuable research hardware aboard Cygnus – but is relieved there were no injuries despite people some miles away ducking for cover as debris flew around, as can be seen in this video.
One focus of the investigation will almost certainly be the way the Antares is powered. Its first stage is powered by two reconditioned Russian NK-33 rocket engines – made and kept in storage since the Soviet era – but when refurbished by US firm Aerojet Rocketdyne they are sold as AJ-26 engines. The first two Antares/Cygnus flights to the ISS used such motors without any problems. However, in May, one of the Russian-derived motors failed in a NASA ground test for a 2015 mission – and just what happened has not yet been explained.
Use of Russian engines as the US tries to wrest human spaceflight away from Soyuz – by replacing the space shuttle with new US-made manned vehicles from Boeing and SpaceX – is controversial in the US.
Elon Musk, founder of SpaceX, described the use of old, refurbished Russian engines as akin to "the punchline to a joke" in a 2012 interview. However he tweeted last night that he hopes Orbital – his rival in cargo-to-the-ISS market – gets back on track soon.
Motor problems?
Another angle for investigators may involve the second-stage technology on Antares. The Antares that exploded was lofting a first-of-its-kind second stage that had not flown before. Powered like previous second stages by a solid rocket motor, the new one was a higher-powered model designed to loft heavier Cygnus vehicles.
But adjusting spacecraft capabilities has caused issues in the past when guidance software was not adjusted to take account of these new capabilities. For instance, in 1996 ESA lost its four-satellite Cluster mission when the first flight of the Ariane 5 rocket failed – because software designed for the previous, much lighter Ariane 4 had not been adjusted to the new rocket's demands.
However, spaceflight engineering consultant Rand Simberg of Jackson, Wyoming thinks a second-stage issue an unlikely cause. "It looked like a first-stage failure. But they have to consider every possibility," he says.
But the explosion may not have any serious, long-term implications because Antares's first stage is slated to be replaced a few years down the road, says Jonathan McDowell of the Harvard-Smithosonian Center for Astrophysics. "Both NASA and Orbital's commercial rivals have had close calls from time to time and it could easily have been another company's rocket going bang."
Planetary Resources says yesterday's destruction of its Arkyd 3 space telescope technology testing satellite will still be followed, as planned, by the launch of the Arkyd 6, a twice-as-big test spacecraft, in the third quarter of 2015. This will test the momentum wheels that will let the company's space telescopes track asteroids, says chief engineer Chris Lewicki, as well as trialling the laser and infrared imaging technology that will assess potential for mining.
If you would like to reuse any content from New Scientist, either in print or online, please contact the syndication department first for permission. New Scientist does not own rights to photos, but there are a variety of licensing options available for use of articles and graphics we own the copyright to.
(Image: Richard Wilkinson) 
– which are increasing as the world warms and pollution makes the oceans become more nutrient-rich – on marine biodiversity. It means that carbon and other nutrients in the carcasses return to the deep-sea food chain after all, challenging the assumption that "jelly lakes" are ecological dead ends that remove carbon from food webs.
