IELTS Academic Reading Practice Test 149

On 27 June 1954, a small reactor in a pine forest about a hundred kilometres south-west of Moscow began sending electricity to the local grid. The plant at Obninsk was not large: its output was about five megawatts, a fraction of what a modern station produces. It was, however, the first nuclear power station anywhere in the world to supply power to an electricity network, and it showed in the most public way possible that the heat released by splitting atoms could raise steam, turn a turbine and light ordinary homes.

Behind that achievement lay less than two decades of research. Nuclear fission — the splitting of a heavy atomic nucleus into lighter ones, with the release of energy — was discovered in Germany in 1938 by the chemists Otto Hahn and Fritz Strassmann, and explained in theoretical terms by Lise Meitner and Otto Frisch. Physicists quickly realised that each fission event releases further neutrons, which can split other nuclei and so sustain a chain reaction. In December 1942, beneath the stands of a stadium in Chicago, a team led by Enrico Fermi achieved the first controlled, self-sustaining chain reaction.

A reactor is not a bomb. In a power station the fuel is only slightly enriched, or in some designs not enriched at all, so the chain reaction proceeds slowly and can be halted at any moment. Three components make this possible. Control rods made of boron or cadmium absorb neutrons and so slow the reaction. A moderator — graphite or heavy water in the earliest designs — surrounds the fuel and slows fast neutrons down, because slow neutrons are far more likely to split uranium-235 nuclei. A coolant, which may be water, carbon dioxide or liquid metal, then carries heat away from the core to boil water and drive a turbine.

Governments kept most reactor research secret during the war and in the years that followed, because the same reactors that produced heat could also produce plutonium for weapons. The mood changed in December 1953, when President Dwight Eisenhower addressed the United Nations and proposed that countries share civilian nuclear technology in return for safeguards against military use. The speech, known as Atoms for Peace, is generally credited with encouraging the declassification of reactor designs, and in 1954 the United States passed a law allowing private companies to build and run reactors.

Britain's contribution was Calder Hall, in Cumbria, which Queen Elizabeth II formally opened on 17 October 1956. Its four reactors were served by tall cooling towers, and the site was described at the time as the world's first full-scale nuclear power station. The design was distinctive. The fuel was natural uranium sealed inside cans of a magnesium alloy, which gave the reactor type its name, Magnox; the moderator was graphite; and the coolant was carbon dioxide gas. Each reactor initially generated about fifty megawatts. Calder Hall also produced plutonium, and this was an important part of its purpose.

On the other side of the Atlantic the first large commercial station was Shippingport, on a bend of the Ohio River in Pennsylvania, which began operating in December 1957 with an output of sixty megawatts. Its design was a pressurised water reactor, in which water held under high pressure is prevented from boiling as it passes through the core. The technology was adapted from the engines built for nuclear submarines, the first of which, USS Nautilus, had entered service in 1954.

Other countries followed. France's first reactor to supply electricity, at Marcoule in the Rhône valley, started up in 1956 and, like Calder Hall, was gas-cooled and graphite-moderated. Larger Soviet stations came into service in the early 1960s, and Britain built a series of Magnox stations on similar lines. The first generation of power stations was shaped less by economics than by the experience of wartime weapons programmes and by the particular fuels and skills each country already possessed.

Optimism was enormous. In a speech to American science writers in 1954, Lewis Strauss, then chairman of the United States Atomic Energy Commission, remarked that electricity would one day become too cheap to meter. The remark has been quoted ever since, usually as an example of excessive confidence. Public opinion in the 1950s was largely favourable, and few people outside a small scientific community questioned either the safety of reactors or the problem of what to do with their radioactive waste.

Reality proved harder. In October 1957 a fire in one of the two plutonium-production piles at Windscale in Cumbria, built for the weapons programme rather than for power generation, released radioactive material across Britain and led to a long clean-up. Costs disappointed too: the Magnox stations were expensive to build, their output was modest by later standards, and a design intended partly to make plutonium was not the cheapest way to generate electricity. By the 1970s most countries had turned to larger water-cooled reactors for new stations, although the first generation remained in service for decades.

Almost all of the pioneer stations have gone. Obninsk was shut down in 2002, after nearly half a century of service; Calder Hall's four reactors closed in 2003 and its cooling towers were demolished four years later; Shippingport, the first American commercial station to be fully decommissioned, stopped operating in 1982. Their importance lies less in the electricity they produced than in the proof they offered that nuclear power could work at all.

Every encrypted message begins with a shared secret. For most of history that secret was carried by hand — a codebook, a courier, a sealed envelope — because anyone who intercepted it could read everything that followed. Public-key cryptography, developed in the 1970s, removed much of that difficulty by letting two people who have never met settle on a key over an open channel. The codes used to protect internet traffic, banking and state secrets all depend on it. Their security rests on mathematics, and on the assumption that certain problems remain unsolvable in practice.

Quantum key distribution, or QKD, offers a different kind of guarantee, based on physics rather than arithmetic. Each bit of the key is encoded in the quantum state of a single photon, usually its polarisation — the direction in which the light wave oscillates. A photon can be prepared, for example, with vertical or horizontal polarisation, or with one of two diagonal orientations at 45 degrees to these. These pairings are called bases. If a photon prepared in one basis is measured in the other, the result is random, and the information it carried is lost.

The best-known protocol was published in 1984 by Charles Bennett and Gilles Brassard and is therefore called BB84. The sender transmits a long stream of photons, each prepared in a randomly chosen basis and state. The receiver measures each photon, also choosing a basis at random, so roughly half the measurements are made in the wrong basis and give meaningless results. The two then compare, over an ordinary public channel, which basis they used for each photon — not what value they obtained — and keep only the cases where the bases matched. What remains is a string of identical random bits known only to them.

The scheme detects eavesdropping because of a result known as the no-cloning theorem, which shows that an unknown quantum state cannot be copied exactly. An eavesdropper who intercepts a photon and passes a copy on must therefore introduce disturbances, and these appear as errors in the receiver's results. The two users test for this by sacrificing a sample of their bits and comparing them openly. If more than a small agreed proportion — in practice around one in ten — disagree, the key is discarded. Otherwise, error correction removes the remaining mistakes, and a procedure called privacy amplification compresses what is left, so that any knowledge an eavesdropper may have gained becomes useless.

A variant devised by Artur Ekert in 1991 uses pairs of entangled photons, whose properties are linked in such a way that measuring one immediately fixes what will be found for the other. Its security follows from a famous result in quantum theory, Bell's theorem, rather than from the difficulty of copying a photon. Entanglement-based systems are harder to build, but they allow some assumptions about the equipment to be dropped. In practice, entangled pairs are produced by shining a laser through a special crystal, and the two photons are then sent in different directions.

The principal limitation is distance. Photons are lost as they travel through optical fibre, and the loss grows exponentially with length, so the key rate collapses after a few hundred kilometres. A conventional signal could be amplified, but an amplifier cannot copy an unknown quantum state, so no simple repeater exists. Long-distance networks therefore rely on trusted nodes: intermediate stations where the key is decoded and re-encoded. This works, but each node must be physically secure, which weakens the original promise.

Space offers a way round the problem, because most of the atmosphere lies below the path and free-space losses are far smaller than in glass. In 2016 China launched a satellite called Micius, which distributed keys between ground stations roughly 1,200 kilometres apart and later exchanged keys with the ground as it passed overhead. The atmosphere is thin, and a satellite above most of it can see a ground station clearly for a few minutes on each pass. European groups and several companies have since run similar experiments, and dedicated fibre links now connect a number of cities, including Beijing and Shanghai.

Practical systems have also been attacked through their hardware rather than their theory. Detectors, in particular, can be blinded with a bright pulse of light and then manipulated, so that a device behaves as the attacker wishes while the mathematics remains sound. One response is measurement-device-independent QKD, a design in which the detectors sit in the middle of the network and need not be trusted by either user. QKD also requires an authenticated classical channel in order to compare bases, and without that channel neither user can be sure who the other is.

For many security agencies these complications are decisive. Rather than distributing keys with photons, they favour post-quantum cryptography: ordinary mathematical algorithms, chosen so that quantum computers are thought to offer no advantage against them, which run on existing hardware and can be deployed as software updates. The United States National Institute of Standards and Technology announced the first algorithms of this kind in 2022 and published standards in 2024. QKD remains valuable where its particular guarantee is worth the cost, but it now looks less like the future of cryptography than one specialised tool within it. The two approaches are not really rivals: one protects the key, the other the conversation that the key protects.

In the first months of 2020 almost every business trip in the world was cancelled at once. Executives who had spent much of their working lives in airports found themselves sitting in front of a screen, and meetings that were supposed to require flights went ahead without them. Deals were signed, projects continued, and a question that would have seemed eccentric a year earlier became respectable: had business travel been exposed as an expensive habit? My answer is that it has been exposed as something narrower. Much of it was indeed habit, but some of it does work that no screen can do.

The industry built around the trip is enormous. Airlines have long depended on business travellers out of all proportion to their numbers, because someone who books a flexible ticket a few days before departure pays several times what a holidaymaker pays for the same seat, and often occupies a premium cabin as well. Hotels, conference centres, exhibition halls and taxi firms are arranged around the same traffic. Anything that reduces it therefore threatens not only airline profits but a great many livelihoods, which is one reason the argument about video calls is not conducted calmly.

The case for the screen is nevertheless strong. Most internal meetings exist to share information, and information travels perfectly well through a camera. Video also widens access: a parent who cannot leave home, a disabled employee, a junior colleague in another country — all can now be present in a room they could never have afforded to fly to. And a return long-haul flight can produce a tonne of carbon dioxide for every passenger on board, which is why a growing number of companies have promised to cut their travel, and some have written the target into their climate plans.

The costs of the screen are real as well. Researchers at Stanford University have described the exhaustion that many people feel after a day of video calls, tracing it to prolonged close-up eye contact, the constant sight of one's own face, the difficulty of moving about, and the extra effort required to read expressions without the peripheral vision that a real room provides. Beyond tiredness lies a subtler loss. Much of the value of a business trip is created in its margins — the walk to the meeting, the coffee afterwards, the remark overheard in a corridor — and those margins disappear when the agenda is a screen and a timetable.

The margins matter most when people do not yet know one another. Trust is difficult to build at a distance, particularly with a new client or a team assembled from several companies, and negotiations that involve genuine disagreement are generally easier in person, where hesitation and posture can be read. Training and creative work benefit in the same way. It is striking that companies which moved most enthusiastically to remote working in 2020 still flew their senior people across the world when a relationship was at stake. Presence is expensive, but it is also a signal: it says that the other party is worth several days of somebody's time.

The honest conclusion is that the two sides of the argument are describing different kinds of trip. A great deal of business travel before 2020 was ritual: the conference that could have been a report, the visit that happened because it always had, the meeting whose real purpose was to demonstrate that someone had turned up. Such journeys wasted money and time, and firms are right to cut them. But the trip that builds a relationship is not ritual, and no amount of bandwidth substitutes for it.

Fairness is rarely mentioned in this debate. If being seen matters, then those who live near head office will be advantaged over those who work at a distance, and staff may be passed over for promotion without anyone intending it. In many parts of the world, too, a willingness to appear in person is still read as a sign of seriousness, and a supplier who never visits may be quietly discounted. A firm that reduces travel without thinking about this may find that it has created two classes of employee.

The environmental arithmetic is not straightforward either. Money that a company saves on flights may be spent on something else that produces emissions, and video calls are not free: they run on devices and data centres that consume electricity. Aviation is one of the fastest-growing sources of emissions and one of the hardest to decarbonise, since there is as yet no practical alternative to liquid fuel for long-haul aircraft. A firm that counts only the flights it did not take is flattering itself. Even so, the saving is far larger than the cost, and it would be foolish to pretend otherwise.

The sensible policy, then, is neither a return to the old volume of travel nor a retreat into permanent remoteness. Companies should be selective: they should ask what a particular journey is for, send people when presence genuinely adds something, and measure travel in carbon as well as in money. Video calls will not replace business travel. They will discipline it, which is a less dramatic claim than the one made in 2020 but a more durable one. The question is not whether to travel but what for.