Breaking the ice

Antartica. An icy world of whiteouts, bone-chilling cold and eight-month winters of permanent darkness.

But it is here at the South Pole, the most inhospitable place on the planet, that a group of scientists and engineers from over 20 institutions in seven different countries will next month begin work on one of the world’s largest, strangest and most inaccessible scientific instruments.

IceCube is a neutrino detector. When it is finally completed in around 2010 the $250m (£146m) observatory will use 1km3 of ice, more than 2km below the surface at the South Pole, to detect these messengers from some of the most violent events in the universe.

Very little is known about neutrinos, but they are believed to carry information about the birth of our galaxy and the mystery of black holes.

Physicists think that they are born when violent cosmic events, such as colliding galaxies or distant black holes, occur at the very edges of the universe. Able to travel billions of light years through space without being absorbed or deflected either by magnetic fields or by atoms, these mysterious high-energy particles could provide answers to some of the most fundamental questions about the universe.

But first you have to find them.

IceCube is the elder brother of the Antarctic Muon and Neutrino Detector Array (AMANDA), an earlier, smaller experiment led by the universities of Wisconsin and California.

AMANDA will be integrated into IceCube but will be soon be dwarfed by its successor.

As a proof-of-concept design completed in 2000, AMANDA excited scientists and astronomers enough to convince them to take the next big step.

The concept of a 1km3-sized facility that could observe neutrinos bombarding the Earth from the most distant of cosmic sources was underway. Funded predominantly by the US National Science Foundation (NSF),  IceCube will consist of sophisticated neutrino detectors encased in 1km3of ice.

Whereas AMANDA used 677 of these glass light detectors, IceCube’s detector array will eventually use around 4,200. Like giant rosary beads, 60 bowling ball-sized sensors are attached to a long electrical cable that is lowered into a hole in the ice nearly 3km deep. Even the cube’s uppermost modules will lie more than 1km beneath the surface.

This process will then be repeated as 70–80 holes are filled with sensor strings until the entire 1km3 of ice is filled with evenly spaced, hi-tech sensing equipment.

The project is hugely ambitious both in scale and in terms of the engineering challenges that it presents those willing to spend the next five to six years working in such an unforgiving landscape.

But while it would be hard to find a more remote or logistically difficult place to locate an observatory, there are a number of very important reasons why IceCube is being built where it is.

Buried at the South Pole, the instrument will be looking for the traces left by neutrinos when they crash into atoms of ice. This is a relatively rare occurrence because neutrinos are so small and usually slip between individual atoms. Every second, trillions of neutrinos stream through the human body, but it is unlikely that any will ever leave a trace.

The collision between a neutrino and an atom produces particles known as ‘muons’ in a flash of blue light called ‘Cherenkov radiation’. In the ultratransparency of the Antarctic ice, IceCube’s optical sensors detect this blue light.

The trail left in the wake of the subatomic collision allows scientists to trace the direction of the incoming neutrino, back to its point of origin, be it a black hole or a crashing galaxy. However, it is more complicated than simply detecting muons.

For every muon created by a cosmic neutrino, a million more are produced by cosmic rays in the atmosphere above the detector. Therefore, to counter this interference, IceCube’s sensors are actually directed downwards, through the Earth’s core, to detect neutrinos as they pass through the planet.

As neutrinos are the only known particles that can pass through matter unobstructed, IceCube and AMANDA use the planet as a filter to select only the muons that result from collisions with neutrinos.

The elusive nature of neutrinos also dictates the positioning of the observatory. A neutrino telescope must be transparent enough so that light can pass through a widely spaced array of sensors, and dark enough to avoid interference from natural light. It must also be deep enough to avoid interference from southern hemisphere cosmic rays.

The Antarctic ice fulfils all these criteria and has a number of advantages over deep-ocean neutrino detectors such as the Baikal Detector in Siberia. Using ice means that the sensors can be deployed from the surface rather than from a ship and data can be collected by instruments located directly above the detector rather than at a distant shore station. Additionally the polar ice is extremely clean and free of radioactivity, allowing the best possible view of neutrino collision on the planet.

But before IceCube can begin to reveal the secrets of the universe, its engineers are faced with the rather more down-to-earth challenge of drilling very deep holes in the ice as quickly as they can.

Drilling using a jet of high-pressure hot water has long been used in the Antarctic for research and is the cheapest and quickest way of making holes in ice.

Keith Makinson works for the British Antarctic Survey, and has used the technique in his oceanography research for a number of years. He is studying the interaction between the icy Antarctic waters and the ice shelves at the Filchner-Ronne Ice Shelf and Rutford Ice Stream, and he acted as a consultant on the IceCube project.

The technology and equipment used in IceCube is, he said, on a different scale from anything that he has worked with in his own experience of hot-water drilling.

‘The concept is really very straightforward,’ he explained. ‘You have extremely hot water, a hosepipe and ice. You keep feeding the hose into the ice and melt a nice big hole. It sounds beautifully simple. But when you’re in a freezing environment it poses a lot of problems in terms of equipment. On top of that, ambient temperatures at the South Pole for IceCube are so much lower than anywhere else I’ve worked in Antarctica.’

Bob Paulos is IceCube’s project manager and his background speaks volumes about the scale of the project. Having previously worked as a program manager for NASA, developing its space mission technology and hardware, he likens the difficulties posed by IceCube to launching satellites into space.

According to Paulos, the equipment used in AMANDA did not have enough power. Whereas the AMANDA team took more than 100 hours to drill a hole, IceCube’s engineers aim to penetrate 2.4km into the ice within 30 hours.

The drill for IceCube is a massively scaled-up version of the one used for AMANDA, with a number of additional tweaks. Weighing 460 tonnes, this huge piece of equipment was transported from Wisconsin to the South Pole over the course of 30 separate Hercules LC-130 flights.

Finding the water to use in the drill was the first problem for engineers on the project; Antarctica is the driest continent on the planet but it does possess snow in abundance. Thus, melted snow is used to feed the drill’s voracious appetite for at least 760 litres of water per minute.

The heating equipment is stored within a series of large shipping containers and consists of massive modified heating units similar to those used in a car wash.

These are powered by jet-fuel brought by the LC-130 planes, fuel that is also used to power the heating system for the occupants of the South Pole Amundsen-Scott Station where IceCube is based.

The process begins with the creation of a ‘ROD well’, an underground cavern that is drilled out of the snow using the hot-water drill which has been started just using snow piled in the heaters. The cavern is made large enough that it acts as a reservoir and collects water at its base, which is then pumped out through to the heaters before moving along the drill hose. The hose, made from specially designed heavy-duty rubber, is more than 3km long and weighs around 11 tonnes.

To enable the project’s engineers to drill each hole as planned, the hose nozzle is equipped with a number of sophisticated electronic devices.

‘The principle is achingly simple but extremely complicated to do in one shot, so we need all the help we can get,’ said Paulos.

A navigation package of software and sensors in the drill head provides information regarding the angle of the drill and real-time data on the hole’s progress. Unfortunately the technology does not extend to being able to control the drill’s journey.

‘We count on gravity to drill straight, but we can’t steer the drill,’ admitted Paulos. ‘If we go off at an angle we will just have to abandon the hole.’

Last season the IceCube team drilled its first hole in the ice, an experience that gave it plenty to think about throughout the Antarctic winter. The drilling had not been progressing very quickly and the drill had started to veer off course.

At 1,000m down, the decision was made to pull the drill out and investigate the problem. It was then that the difficulties posed by the environment were bought into stark focus.

As the drill sat idle, water began to freeze in the return line, but as the drill crew worked on solving this problem a member of the team was hit by a cable and badly injured.

The group also discovered much about the minutiae of hot-water drilling. While the original drill possessed a 1in-diameter nozzle, software modelling carried out since last year indicates that a smaller diameter drill could provide more efficient drilling.

One of the requirements for hot-water drilling is to be able to successfully model the refreezing speed of the ice in the hole.

‘You need to know what the hole will look like after 30 hours of drilling, and work around   that,’ said Paulos. ‘As soon as you stop drilling, the hole begins to close.’

Once the hole is finished, the team is then in a race against the clock to deploy the sensing ‘strings’ before the watery hole refreezes. To do this, the hole needs to be a minimum of 45cm across up to 30 hours after drilling.

Working with Makinson at the British Antarctic Survey, IceCube’s team modelled the ice extensively using fluid flow software to find out exactly what was happening at the nozzle’s tip.

‘It’s a tricky business dealing with things like turbulent flow and you also have a fluid/ice transition that is hard to model,’ said Paulos.

One of the greatest obstacles to any project at the South Pole is the extreme cold, particularly the way it affects sensitive electronic equipment.

According to Paulos, it is relatively easy to find electronics rated for –40°C, but any colder and custom-made equipment is required. A lot of money was spent in the lead-up to IceCube on testing all the sophisticated parts of the system, as anything left at the South Pole over the Antarctic winter is liable to be subjected to temperatures of around –70°C.

For the same reason, custom engineering was required in the development of the cables for the encased sensor string. ‘Cables seem pretty trivial, I know, but for us they are extremely serious equipment,’ said Paulos.

As the cold can make electrical cables very brittle they had to be specially engineered to be able to survive the extreme temperatures they would encounter. More than $25m (£14.5m) of the project’s budget was spent on cables, which came from Swedish firm Ericsson.

The cables were then treated by a specialist US underwater electrical company that had developed a new waterproofing technology that could be injected into both the cable and its connections with the optical modules.

Two tanks containing 2,000 gallons each of frozen water sit on the surface above each hole like small swimming pools frozen solid. Each tub, known as IceTop, contains two modules similar to those encased in the ice below and acts as a calibration device. As they cannot work in 3D like IceCube, they are unable to detect the path of a neutrino but they will hopefully provide additional data in the case of a neutrino ‘event’ being detected below.

The optical modules themselves, although similar in size and design to those used for AMANDA, also represent a considerable step up in technology. Each glass optical module is extremely rugged and is specially constructed from reinforced glass to withstand pressures of up to 10,000psi.

They consist of a reinforced glass pressure vessel, which houses a photomultiplier tube that transforms light into electrical signals. The glass balls also contain electronics, which link via the internet to the computers at the project’s base station.

The digital sensor technology, in comparison with AMANDA’s more primitive analogue signals, will digitise the signal from neutrino ‘events’ and send them instantly to the base station to be analysed.

Every year three crew members will stay behind at the Amundsen-Scott Station over the dark Antarctic winter to check the data, and make sure everything is functioning properly.

An online system will clean the data and then send it via satellite back to the US, where it will be disseminated to all the collaborating institutions, which include Imperial College London and Oxford University.

This year drilling is scheduled to resume on 10 December, and teams of drillers will be working 12-hour shifts in a bid to improve upon last year’s performance, when they managed to drill the first hole in 58 hours.

The goal this season is to finish 10 more holes.

‘When we have finished, IceCube will be the world’s biggest neutrino telescope. It’s a pretty exciting time for all of us but we know we still have a long way to go, and a lot more holes to drill before we get there,’ said Paulos.