When ideas are left on the drawing board, it is often with good reason.
But that didn’t stop the BBC testing out one of the craziest proposals of the Second World War...a boat made from ice and wood pulp.
Maverick inventor Geoffrey Pyke claimed his five and a half ton craft would both save on steel and be impossible to sink.
Yet a mock-up of his brainchild took on water and melted within minutes of its launch in Portsmouth harbour yesterday.
Experts said that the experiment for science show Bang Goes The Theory probably failed because the boat was too small, and so less resistant to melting, and because the water they tested in was far warmer than the Atlantic - where the invention was designed to be used.
In the event of steel stocks running out in the 1940s inventor Geoffrey Pyke suggested it was possible to make an unsinkable aircraft carrier using a material called Pykrete, made of both ice and wood pulp.
The bizarre mixture could be moulded into any shape and, with a slow melting rate, it was thought perfect for seafaring vessels.
The BBC decided to put Pyke's theory to the test by mixing 5,000 litres of water with the hefty material hemp and freezing it in a 20 feet-long boat-shaped mould.
It took three weeks to freeze it in one of the UK's largest ice warehouses, in Tilbury, Essex, before it was ready for launch in Gosport, Hants.
The team made it in to Portsmouth Harbour where they were saluted by members of the navy stationed on destroyer HMS Diamond.
But shortly after that, after just over an hour in the water, it began to take on water and capsized.
Four BBC presenters, who had hoped to make it all the way to Cowes on the Isle of Wight, had to abandon ship and swim to rescue craft.
Lynette Slight, of the BBC science show, said: ‘They had just got out of the marina when it began to sink.
‘It was all a little bit strange. I don’t think they realised what would happen. In the end it just tipped upside down. It was taking on too much water at the back and the engine became too low.’
Jon Edwards at the Royal Society of Chemistry said 'It’s hardly a surprise that the boat sank – the temperature in the Solent is probably a fair bit higher than the middle of the Atlantic, where Pyke designed his material to be used.
'He also used enormous cooling units to keep the pykrete in his tests below zero degrees centigrade. If they didn’t use those refrigerators, the intrepid ice-sailors from Bang never stood a chance.'
He added: 'The size of the boat may have added to their problems, too. A huge aircraft carrier, as Pyke envisioned, would have been more resistant to melting – a larger surface area of ice requires a lot more energy to start melting, so the non-surface ice stays cooler for longer.
'A 1000-ton test boat, built out of normal ice on a lake in the Rockies, lasted a whole summer.'
And an Institute of Physics spokesperson said: 'The surface to volume ratio will have been the key to success.
'If too much of the surface of the ice was exposed directly to the water, or if the volume of ice set to melt was not calculated accurately enough, then unfortunately it was always doomed to failure.'
The plan was to sail the boat, complete with outboard, to Cowes in the Isle of Wight with the show's presenters, Jem Stansfield, Liz Bonnin, Dallas Campbell and Dr Yan Wong, on board.
All four presenters had to be rescued from the water and the boat, which seemed to melt beyond recognition in no time at all, had to be towed to shore.
Lynette Slight, production coordinator of the show, said: 'They had just got out of the marina when it began to sink.
'It was all a little bit strange. I don't think they realised what was going to happen. In the end it just tipped upside down.
'It was taking on too much water at the back and the engine became too low. They thought they could get it to Cowes - they couldn't, but you never know until you try it.'
Giles Harrison, director of the show, blamed the failure on a fault, which meant water poured into the vessel sooner than expected.
He said: 'There are a couple of reasons why we did it. There was the proposal in the Second World War, when they were running low on steel to use ice with wood pulp in it.
'It was an idea taken quite seriously, until the war ended and it was forgotten. We were essentially using that concept to see how composite materials work.
'We did anticipate that something would go wrong but we hoped to get further out than we did.
'I think we've proved that Pykrete works but it is unstable.'