by Vaclav Smil
Figure 5.4 Chinese pipelines.
Annual volume of 38 Gm3 is not particularly large considering either the total Russian output or the huge Chinese demand, but the deal is to last 30 years and hence the total volume will surpass 1.1 Tm3 which is more than, for example, the total reserves of Azerbaijan or the Netherlands. Nevertheless, the CEO of Gazprom claimed that the deal will influence the whole gas market (Itar-Tass, 2014a), but the undisclosed price (between $350 and 380/1,000 m3) is no higher than the discounted long-term contracts signed with major European importers during 2013, and the construction of nearly 4,000 km pipeline to Vladivostok on the Pacific Ocean (to feed Russia’s future LNG exports) will cost close to $50 billion, and the branch to China will add another $20 billion, totals that make the deal less appealing than the sales to Europe.
The rest of Asia is still far behind the emerging supernetwork of long-distance trunk lines connecting China with the Central Asia and Siberia. There are a number of shorter lines for selling gas to neighbors: Egypt to Jordan, Syria, and Lebanon, with an undersea branch to Israel (with deliveries severely limited due to Egypt’s high domestic demand, result of a growing urban population and high fuel price subsidies); Turkmenistan to Iran; Iran to Armenia; Myanmar to Thailand; and, since 2013, Myanmar to China and Malaysia to Thailand. Perhaps the most remarkable large-scale plan is the trans-Afghanistan line bringing the Central Asian gas from Turkmenistan to northern Pakistan and India. Completion of this project remains (for obvious security reasons) particularly uncertain, but potential to export to India is, in the long run, no smaller than selling to China: if there were no supply and payment constraints, they could be on the order of 100 Gm3 by 2025 and double that a decade later.
Development of natural gas networks in Western Europe has had three aims: bringing the North Sea gas by undersea pipelines to the continental market; distributing the imported LNG inland; and building many new interconnections, including bidirectional links, that allow a greater flexibility of intracontinental supply. The oldest subsea pipeline for natural gas export, the Vesterled system from the Norwegian Heimdal Riser platform to St. Fergus in the United Kingdom, was completed in 1978, and the 1990s was the decade when most of the existing export lines went into operation: Europipe I and II in 1995 and 1999 (from Norwegian waters to Germany), Franpipe to Dunkerque in 1998, and Zeepipe from the Troll processing plant near Bergen to Belgium (Subsea Oil and Gas Directory, 2014). Langeled, the region’s longest (1,166 km) export pipeline taking gas from Nyhamma via Sleipner field to Easington in the United Kingdom, was finished in 2006.
Europe got its first connection with North Africa in 1983 when the Trans-Mediterranean pipeline (2,476 km) began to transport Algerian gas from Hassi R’Mel via Tunisia to Sicily and then across the center of the island and the Strait of Messina to Italian mainland and all the way to Slovenia. Thirteen years later came the exports of Algerian gas westward via Morocco and the Strait of Gibraltar to Spain, to Cordoba distance of 1,620 km. The gas is then moved to Madrid and northward to a low-capacity interconnection with France. By 2013, Europe also had 19 regasification terminals for natural gas imported (leaving aside small intra-European LNG exports from Norway) from North Africa, Nigeria, the Middle East, and Trinidad and Tobago. I will take a closer look at the development of Europe’s LNG imports in the next section of this chapter when I will follow the industry’s long evolution.
Europe has been looking beyond its four current major supply streams (Russia, North Sea, Groningen, and LNG) to pipeline imports from Azerbaijan, Turkmenistan, and even Iraq. The Nabucco pipeline was originally conceived to supply Azeri gas to Europe via Turkey, with Iraqi gas added later. The original plan was for 3,893 km line from Ahiboz in Turkey (where it would receive gas from South Caucasus line and an Iraq line) to Baumgarten in Austria, but after more than a decade of plans and negotiations and contracts, the line’s fate remains uncertain. A competing project is the trans-Caspian line (capacity of 30 Gm3/year) that would run from Turkmenistan to Azerbaijan and then across Turkey to Europe, completely bypassing Russia. Opportunities for further pipeline extension and grand integration of Russian, Chinese, Middle Eastern, and European networks would change radically should Iran join the ranks of normal countries, but there is no chance that now so deeply embedded (soon to be 40 years) Shiite theocracy would yield its control voluntarily and that Iran will emerge as natural gas export power rivaling Russia.
The unsettled state of the Middle East has been the major reason why there have been no serious plans to connect gas from the world’s largest natural gas field to European markets by pipeline: a line from Qatar to the EU would have to cross Saudi Arabia and then either Kuwait and Iraq or Syria before reaching Turkey, routes that have always posing unacceptable risks for capital investment of tens of billions of dollars, and that became impossible with the disintegration of Syria and Iraq. But Qatari gas reserves are too large and too inexpensive to produce to be left in the ground, and that is why Qatargas decided to develop them in two technically more demanding but ultimately rewarding ways, by building the world’s largest GLT project (see Chapter 4 for details) and by acquiring the world’s largest gas liquefaction capacity as well the most modern and largest fleet of supersized LNG tankers.
5.3 EVOLUTION OF LNG SHIPMENTS
Increasing volumes of the fuel are traded in the form LNG. Liquefaction entails cooling the gas down to −162°C and reducing its volume to nearly 1/600th of the gaseous state: its density is 428 g/l (1 m3 of LNG = 0.43 t) compared to 0.761 g/l for methane at ambient temperature.
Liquefaction is done in several independent units (trains) that are typically about 300 m long (Linde, 2013). Liquid gas is stored in superinsulated containers before it is transferred (via articulated pipes) to isothermal tanks on LNG tankers. Typical LNG contains 94.7% CH4, 4.8% C2H6, and mere traces of higher alkanes and nitrogen, and its energy density is 53.6 MJ/kg (22.2 MJ/l, compared to 0.035–0.037 MJ/l for natural gas).
LNG is shipped in special tankers (Figure 5.5). The largest of these vessels (Q-Max ships of Qatar) carry 266,000 m3 of gas (Qatargas, 2014a). Mozah (named after the wife of Qatar’s emir) was the first ship of this class built by Samsung Heavy Industries in 2008, and it is now transporting LNG to the United Kingdom (Marine Traffic, 2014). Qatar now has 14 of these ships whose typical dimensions are a length of 345 m, a width of 53.8 m, and a draft of 12 m. With 22.2 GJ/m3, a Q-Max ship’s cargo amounts to almost 6 PJ of floating energy storage, enough to fuel 70,000 US homes for 1 year. Regasification takes place in seawater vaporizers. Most of the recently commissioned terminals have liquefaction capacities of 4–5 Mt/year, and most of the new receiving terminals rate between 3 and 5 Mt/year (IGU [International Gas Union], 2014).
Figure 5.5 LNG tanker Arctic Voyager.
© Corbis.
Evolution of large-scale LNG industry is a perfect example of gradual advances whose progress has been determined not only by technical capabilities but also by willingness to pay initially very high costs for importing the gas from overseas. The long road to global LNG market began in 1852 when Thomas Joule and William Thompson (later Lord Kelvin) discovered that as a highly compressed air flowing through a porous nozzle expands to the ambient pressure, it cools slightly (Almqvist, 2003). Sequential use of this incremental cooling was put into practice only by 1895 when Carl von Linde patented a reliable process of air liquefaction (Linde, 1916). Two decades later, Godfrey Cabot patented a design for shipping LNG by oceangoing vessels (Cabot, 1915), but as long as the United States remained the only major user of natural gas, there was no incentive to commercialize this possibility.
But there were important developments with LNG storage; the first small LNG project was completed in West Virginia in 1939 and a larger one in Cleveland in 1941, both to store energy-dense fuel for the periods of peak demand. Unfortunately, in 1944, a new larger, and poorly built, storage tank in Cleveland tanks failed, vaporized LNG ignited
, and the resulting explosion killed 128 people. This accident, traced by a subsequent investigation to a poor tank design, came to haunt the industry as a proof that LNG storage, and by implication even more so the LNG transportation, is inherently too risky (USBM [US Bureau of Mines], 1946). Production of liquefied industrial gases expanded rapidly after WWII, but as North America only increased its global primacy as natural gas producers and as Europe and Japan embarked on imports of cheap Middle Eastern oil, there was no incentive to develop expensive LNG imports.
The first demonstration shipment of just 5,000 m3 of LNG (from Lake Charles, LA, to Canvey Island on the Thames) took place in January 1959. Methane Pioneer was not a true tanker but merely a converted Liberty-class vessel (built in 1945 as one of more than 2,700 leading WWII freighters), and it subsequently carried seven other LNG cargoes to Canvey Island before it was discarded in 1967. In 1961, the British Gas Board signed an agreement for long-term imports of LNG from Algeria’s new Camel (Sonatrach) Plant in Arzew (three trains, low capacity of just 0.9 Mt/year) and the editorial in The Gas World (on November 11, 1961) called it “triumph of common sense”. Deliveries of natural gas from Hassi R’Mel via Arzew to Canvey Island began in 1964, and two dedicated tankers (Methane Princess and Methane Progress) served the route, each carrying 27,400 m3 (Corkhill, 1975). In 1972, Sonatrach added another LNG facility at Skikda, and both LNG plants were later expanded with additional trains.
Japan became the next importer LNG. In the absence of domestic hydrocarbon resources, it was motivated by desire to diversify its sources of imported energy (at that time only crude oil and coal) and reduce heavy air pollution in its large coastal cities. During the late 1960s, commercially the most viable opportunity was to bring LNG from Alaska, and in 1969, two tankers, Polar Alaska and Arctic Tokyo (each with capacity of about 70,000 m3), began to peddle between Cook Inlet, Alaska, where Phillips Petroleum and Marathon Oil built a single-train LNG plant (annual capacity of 1.5 Mt) and Tokyo (Marine Exchange of Alaska, 2014). First imports from an Asian country began in 1972 when Brunei LNG was shipped to Ōsaka.
But as new LNG importers were added in the early 1970s (France from Arzew, Spain and Italy from a new Marsa El Brega plant Libya in 1970), the future of European LNG purchases began to look doubtful. Groningen, a giant Dutch natural gas field near Slochteren, was discovered on July 22, 1959, and as additional drilling uncovered its magnitude, it became clear that the find will not only end the age of coal in the Netherlands but that it will make the country a major exporter of natural gas to neighboring European states (Smil, 2010a). Concurrently, search for hydrocarbons in the North Sea had finally yielded both crude oil and natural gas: the first large gas find was in the West Sole Field in September 1965, 3 months later came the Viking Gas Field and more discoveries followed during the late 1960s.
As a result, the British contract for Algerian LNG imports was not renewed after its 1979 expiry, and the future of LNG imports to Europe became even more unlikely in 1984 with the completion of the first high-capacity pipeline (Urengoy–Uzhgorod) to bring Russian (at that time Soviet) gas from West Siberia. Concurrently, American LNG imports fell far short of initial expectations. The first US regasification terminal began to operate in November 1971 at one of the most unlikely locations, in Everett, Massachusetts, on the northern coast of the Mystic River less than 4 km from Faneuil Hall, with tankers moving upstream past the city’s downtown; moreover, the terminal is also less than 5 km from the NW–SE runway of Boston’s Logan International Airport.
This location focused attention on potential risks of LNG transportation and accidents during docking, loading, and offloading or due to earthquakes or a terrorist attack. Potential hazards arise from cryogenic temperatures of the transported gas, its dispersion, and flammability. Wilson (1973) was concerned about these hazards because he thought that their possibilities (particularly the explosion of spilled methane) bring to mind a comparison with potential nuclear reactor accidents. A few years later, Fay (1980) worried about new phenomena that could become evident in large-scale spills but cannot be observed in small-scale tests. And, not surprisingly, after 9/11, there were increased concerns about the LNG tankers as targets or tools of terrorist attacks (Havens, 2003).
But for half a century, the modern LNG industry has had a remarkable safety record demonstrated by Acton et al. (2013) in their study of 328 incidents that took place at peak shaving or LNG receiving facilities worldwide between 1965 and 2007. They found a very low frequency of incidents (0.24/site-year, most recently just 0.14/site-year), with about 75% of hydrocarbon releases being less than 100 kg/event and originating mostly from storage and send-out equipment. Just 7% of these incidents resulted in fire, explosion, or rapid phase transition, gravity of such events has decreased over time, and operation and maintenance shortcomings were the primary cause. The only category of increasing incidents is related to LNG trucks whose operation is often outside the control of LNG terminals.
Since the beginning of international LNG, there were no reports of damage outside peak shaving or receiving facilities, and hence, it is correct to conclude that imports, regasification, and distribution of LNG have not been a public hazard. As for the liquefaction plants, there was only one catastrophic explosion resulting in substantial loss of life and large economic loss. This most serious accident took place on January 19, 2004, in an LNG plant in Skikda on the Mediterranean coast of Algeria when a steam boiler (used to produce steam for turbines supplying power to compressors) exploded and triggered a massive secondary explosion of gas vapor cloud (Hydrocarbons Technology, 2014). Explosions and the subsequent fire (that took 8 hours to extinguish) killed 26 people, injured 74, destroyed three LNG trains and damaged another one, but left two trains, as well as all storage tanks, intact. Most remarkably, since 1964, there has been never any serious shipping accident resulting in loss of life and cargo or damage to port facilities as LNG tankers have crisscrossed the oceans delivering gas from every continent.
Two additional US terminals, at Cove Point in Maryland and Elba Island in Georgia, began to import LNG in 1978, but they were shut down just 2 years later. Lake Charles terminal in Louisiana opened in 1981 and was shut down the next year. LNG prospects dimmed further with the deregulation of US gas prices, with the 1985 collapse of high crude oil prices, and in 1986, no exported LNG came to the United States, but the shipments from Algeria to Lake Charles were resumed in 1988. Little had changed during the 1990s, the decade of relatively low and stable energy prices. This left Japan as the only major LNG buyer, and the country concluded new long-term import contracts in 1977 with the United Arab Emirates (Abu Dhabi) and Indonesia (LNG from Bontang field in eastern Kalimantan; a year later, shipments also began from the Arun plant) and in 1983 with Malaysia (Bintulu LNG plant), and by 1984, 75% of all traded LNG were sold to Japan. In 1991, LNG from Australia’s North West Shelf was added to Japanese imports.
That began to change soon afterward as two East Asian countries, whose economic development was based on the Japanese path, became LNG importers: South Korea in 1986 and Taiwan in 1990 (both from Indonesia). And as in Japan’s case, their purchases were in the form of long-term supply contracts, with LNG shipped from facilities with modest annual capacities and using relatively small tankers. During the late 1990s, two companies began to export LNG from the world’s largest natural gas field: three Qatargas trains began shipping in November 1996 and two RasGas trains came on line in August 1999. Qatargas under the leadership of Faisal al-Suwaidi accepted Shell, ExxonMobil, and Total as full partners and went on to develop the world’s largest LNG export capacity: 77 Mt/year is an energy equivalent of one quarter of Saudi Arabia’s crude oil exports in 2012.
In 1999, Trinidad and Tobago’s Atlantic LNG was the first company in Latin America to complete its liquefaction facility at Point Fortin on the southwest coast, and it began exporting to the United States (to Everett), while Nigeria’s first two LNG trains began operating at Bon
ny Island in the Niger Delta. At the end of the twentieth century, the LNG industry was maturing, but it still had a long way to go before it could be seen as a key global player. There were 12 LNG-exporting countries, and a global fleet of about 100 tankers was used to export almost 140 Mt of the fuel to nine importing nations; that was about 20% of all natural gas exports but less than 6% of the global natural gas consumption (Castle, 2007).
Most LNG plants still had annual train capacities of just 1–2 Mt/year, the largest one could ship 3.5 Mt/year, and the volume of the largest LNG tankers had not changed for a quarter of the century: the record volume was 126,227 m3 in 1975, and after surpassing 130,000 m3 in 1978, the size of the largest tanker grew only marginally during the next 25 years, reaching 145,000 m3 in 2003, a capacity stagnation mostly explained by the fact that Japan, the world’s largest LNG importer, limited the maximum size of LNG tankers allowed to dock in its ports. Evolution of LNG exports during the twentieth century is thus an excellent example of gradual advances characteristic of global energy transitions (Smil, 2010a).
If we consider 5% of a market as a threshold of global importance, then it took LNG exports more than a century after the first commercial liquefaction to reach that threshold which came 50 years after the first 1959 test shipment from Louisiana to the United Kingdom. And while in the year 2000 LNG imports supplied significant portions of primary energy in Japan, South Korea, and Taiwan, they had only marginal roles in Europe and in the United States, and everywhere, they were dwarfed by imports of crude oil and refined oil products. But the first decade of a new century broke that long spell of slow advances, both in terms of expanding export base and fundamental technical improvements.
