SEIBERSDORF, Austria (Reuters) - The self-styled "Sherlock Holmeses" of the U.N. nuclear watchdog, now seeking access to a major Iranian base, say they have the capability to find tiny traces of atomic material at a site even if a country were to try to cover it up.
In talks later this month, the International Atomic Energy Agency will once again press Tehran to allow its inspectors to visit Iran's sprawling Parchin military complex. That would enable them to bring back swab samples for thorough checks at the IAEA's high-tech laboratory near Vienna.
Western diplomats have accused Iran of trying to cleanse the Parchin site of possible signs of tests relevant for the development of nuclear weapons, casting doubt on whether U.N. investigators would discover anything even if they could go.
Iran says Parchin, located south-east of the capital Tehran, is a conventional army base. It dismisses allegations that it has carried out atomic bomb research and says its disputed nuclear programme is for energy and other peaceful aims only.
Experts say that while it may now be difficult to find any evidence, it would still be possible to locate traces of nuclear materials with equipment that can study particles 10,000 times smaller than a grain of sand.
Tell-tale particles could not be removed completely from a facility where uranium was used, said Stephan Vogt, a senior IAEA official, who emphasised that he was speaking generally and not specifically about Iran or Parchin.
"You cannot get rid of them by cleaning, you cannot dilute them to the extent that we will not be able to pick them up. It is just a matter of time," Vogt, who heads the IAEA's Environmental Sample Laboratory, said.
"We won't find it maybe the first time we go there," he said. But, "the more often we go, the higher the probability that we will pick up (traces) in some corner, at some table, in some plumbing".
Former chief IAEA inspector Olli Heinonen said any attempt by Iran to purge Parchin of clues would make the agency's task considerably harder, but "complete sanitisation is very difficult to achieve if nuclear materials were actually used".
Like others at the IAEA's Seibersdorf laboratory complex outside the Austrian capital, Vogt was not authorised to discuss Iran, Syria or any other specific cases which have made the agency a key player in international nuclear diplomacy.
But he made clear his confidence in the sophisticated techniques at the scientists' disposal, including a new 3.8 million euro instrument to study tiny particles.
Likening the IAEA's investigative work to that of fictional detective Sherlock Holmes, Vogt said: "You are running around, looking for the right spot to sample and then you look for microscopic particles, they can tell you stories."
Installed in a purpose-built building, the Large Geometry Secondary Ion Mass Spectrometer can analyse 100-150 samples per year - up from 30-40 previously - collected around the world by inspectors using small pieces of cotton on surfaces.
"We have a much larger magnifying glass, we see much smaller particles," Vogt said, showing the machine, which occupies a room of its own. It "opens brand new doors into what we can see and what we can interpret."
The Seibersdorf facility gained a more prominent verification role in the 1990s after the first Gulf War when the IAEA was given wider powers to detect undeclared activity following the discovery of Iraq's clandestine nuclear programme.
Iraq shut down its nuclear, chemical and biological weapons programmes under orders from the United Nations. Suspicions that it was not cooperating with investigators were used by the United States and Britain to justify their invasion in 2003.
The IAEA has shown it can find particles even when a country has worked hard to hide them. It picked up tiny traces of enriched uranium at Kalaye Electric in Tehran in 2003, even though Iran had removed equipment and renovated parts of the facility.
The IAEA also uses a network of member states' laboratories to help it study samples taken during its inspections globally.
In order to strengthen its capabilities, it is now modernising Seibersdorf, housed in an anonymous-looking complex of white, low buildings.
As part of an 81 million euro upgrade, the IAEA is building a new Nuclear Materials Laboratory where uranium and plutonium samples will be checked to make sure that materials that can be used for bombs are fully accounted for.
Hundreds of samples from nuclear reactors and fuel plants are pored over every year by experts dressed in white coats and protective gear in the present 1970s-era building. The vast majority of tests turn up nothing suspicious.
"People working here don't know where the sample comes from. We are not doing politics here. We are only doing technical analysis," said chemistry team leader David Amaraggi.
But despite the IAEA's insistence that it is a technical organisation serving 159 member states, its monitoring of Iran's nuclear programme can have geopolitical implications.
It regularly inspects Iran's declared nuclear facilities - including the Natanz and Fordow uranium enrichment sites - but has so far failed to persuade Tehran to enable it to resume a stalled investigation into suspected nuclear weapons research.
In a tenth round of talks since early 2012, an IAEA team led by chief inspector Herman Nackaerts will meet Iranian officials in Vienna on May 15 to try to end the deadlock.
The IAEA's priority is to visit Parchin, where it believes Iran built a steel chamber for explosives tests more than a decade ago, possibly using non-nuclear materials like the metal tungsten as substitutes for uranium.
Citing satellite imagery, Western diplomats have said that Iran appeared to be rebuilding the specific part of Parchin the IAEA wants to see, after earlier razing smaller buildings and removing soil. Iran denies it has anything to hide.
IAEA Director General Yukiya Amano says the U.N. agency still wants to inspect Parchin, even though it fears that the suspected clean-up will have seriously undermined its ability for "effective verification" at the site.
Robert Kelley, a former IAEA inspector in Iraq, said there would be a good chance to discover particles of man-modified uranium if such tests were conducted at Parchin, but if substitutes were used they would be harder to find.
"Environmental sampling is thousands of times less sensitive for detecting non-radioactive things like tungsten," Kelley said. (Additional reporting by Mark Heinrich in London; Editing by Peter Graff)