The UK's top computational chemistry facility has announced it is to close after a £3.8 million grant proposal that would have funded it until 2011 was rejected in July. Seventy-five research projects - ranging from molecular simulations to electronic structure calculations - using the National Service for Computational Chemistry Software (NSCCS) at Imperial College London now face an uncertain future, with the facility set to close in March 2009.

The decision by the Engineering and Physical Sciences Research Council (EPSRC) has been criticised by researchers who believe that the closure will leave them without access to key software, training and expertise. An online petition set up on 13 July to save the NSCCS has already been signed by over 250 chemists working in a range of disciplines.

'The closure of the NSCCS will leave many researchers, in particular experimentalists, with no alternative resource to use,' says Hazel Cox, the University of Sussex computational chemist who started the petition.

The NSCCS is just one of a number of key facilities that could be under threat thanks to new EPSRC funding rules that mean money for 'national services' is no longer ring-fenced. Instead the grants to run these services now have to fight for funding through responsive mode. But many believe that pitting the facilities' grants against cutting-edge research proposals is unfair.

'A National Service is not at all similar to a single PI research grant,' says David Leigh, an organic chemist at the University of Edinburgh who has signed the petition. 'The continuation - or not - of national services requires strategic decisions that should be decided upon at a high level, tensioned against other funding, but only after due consideration of the pros and cons.'

Other facilities that could be at risk include the Electron Paramagnetic Resonance Service in Manchester and the National Crystallography Service, Southampton. Both are set to apply for more funding in 2009.

The EPSRC, however, defends its decision. 'If the EPSRC were to continue to support national services through directed-mode funding, the result would simply be less available funds to support grants offered through responsive-mode,' says Atti Emecz, EPSRC communications and information director. 'Once established, a national service such as the NSCCS should be able to effectively compete within a responsive-mode panel.'

Mike Robb, a member of the NSCCS management committee, says chemists will now need to set aside grant money to buy any computer software and hardware they need. 'Our traditional users are experimentalists who use computation as an adjunct to experiment,' Robb says. 'It follows that if they are to get the benefit of our type of service, they will need to provide for this, as part of a consortium, within a grant proposal.'

Fred Campbell