Natural Gas- Fuel for the 21st Century

by Vaclav Smil

  America’s early transition from coal and oil to natural gas was relatively slow for two obvious reasons: the United States was the first major economy to pioneer the shift, and in its early years (as already noted) the pace was limited by technical capabilities (above all the absence of long-distance high-throughput pipelines); and the magnitude of the country’s TPES precluded any rapid fuel substitutions. Natural gas reached 5% of the US primary energy production already in 1924 and 10% in 1935. The economic crisis and WW II slowed down the pace a bit, but 20% mark was reached in 1951 and 25% in 1957. A temporary peak just short of 34% came in the early 1970s, and declining production lowered it to about 25% by 1990. Relative gas contribution stagnated at that level for more than a decade until shale gas extraction pushed the share up to 31% by the year 2012 (Figure 7.3). This increase has been accompanied by a higher share of gas used for electricity generation: volume of gas used for thermal electricity generation had doubled between 1983 and 2013 as the share of all gas-derived electricity rose to nearly 30% (USEIA [US Energy Information Administration], 2014e).

  Figure 7.3 US gas share in primary energy production.

  Canada’s rich natural endowment has made it the country with the lowest specific carbon emissions among the affluent economies. Until the late 1990s, carbon-free hydroelectricity supplied a higher share of its TPES than natural gas; as a result, the share of gas in primary energy consumption has been rising only slowly, taking 50 years to double to nearly 30% by the year 2012 (but for most of that time, at least 30% and up to almost 50% of Canada’s annual natural gas production were exported to the United States). And so it is Mexico that has the highest natural gas share of the TPES in North America, at 40% in 2012, second only to crude oil and 30% above both the US and Canadian shares.

  The post-1950 Soviet exploitation of natural gas deposits was initially delayed by large increases in oil production that pushed the fuel’s share of the TPES from 16% in 1950 to the peak of 35–37% between 1974 and 1983. But between 1970 and 1990, rapid development of supergiant Siberian fields had more than quadrupled the Soviet gas production; by 1983, the USSR surpassed the US output and even its new large export commitments to supply its Communist satellites; and the EU were no impediment to a steadily rising gas share from less than 10% of the TPES in 1960 to more than 20% in 1970, 32% in 1980, and 41% in 1990. And while many performance figures have plummeted in the post-Soviet Russia, natural gas gained relatively greater importance as its supply share rose to about 52% by the year 2000 and to 54% in 2012.

  7.2 METHANE IN TRANSPORTATION

  This section should open by a reminder (see Chapter 3) that natural gas has had for decades a critical role in its own transportation, as efficient and reliable gas turbines power compressors that propel gas through pipelines. But the gas has, so far, been only a marginal source of energy for moving goods and people, and this minor contribution to the transportation sector presents a major impediment to the fuel’s rise to a significantly higher share in the global primary energy supply. Transportation claims about 20% of the world’s TPES and refined oil products supply 93% of that total and natural gas less than 4% (IEA [International Energy Agency], 2013). Transportation’s sectoral share in the United States is higher (28%), but the shares of liquid fuels and natural gas are very similar at 93% and less than 3% (USEIA, 2014e). Of course, the reason for these marginal contributions is obvious: while natural gas is the best fuel for space heating and industrial processing, its low specific density and low volumetric energy density make it an inferior transportation fuel compared to superior densities of liquid fuels refined from crude oils.

  This reality excludes it from ever being a fuel for commercial aviation: even after liquefaction, its volumetric energy density of 21.4 GJ/m is only about 60% that of volumetric energy density of jet fuel (kerosene)—and well-insulated tanks are needed to keep the fuel at −162°C. But even when solving the technical challenge of carrying the cryogenic fuels, the planes would need more than twice as much LNG to cover the same distance, the need that would inevitably limit their passenger and cargo capacity and make flying quite expensive on that account alone. But liquefaction and portable cryogenic storage do not present any insurmountable technical problems for using natural gas in heavy-duty land and water transportation where additional mass necessary to carry the fuel can be readily accommodated.

  7.2.1 LNG

  Indeed, LNG’s benefits in heavy-duty applications—above all in shipping and trucking—make it an excellent choice for markets that are now served by high-efficiency diesel engines. Costs aside, the fuel offers three fundamental advantages when compared to diesel fuel: lower generation of CO2 per unit of useful energy, lower emissions of sulfur and nitrogen oxides, and noise reduction. But LNG is also a fairly cost-competitive fuel choice, particularly for newly built ships and new truck fleets. LNG tankers are obviously the most obvious candidates, and they have used the fuel without mishaps for the past 40 years. They were originally propelled by steam turbines with steam raised by burning the naturally boiled-off gas, but soon, as in every form of mass cargo shipping, highly efficient and inexpensively operating diesel engines became dominant.

  The two basic types of these prime movers are known as diesel low speed and reliquefaction (DLR) and dual-fuel diesel electric (DFDE). DLR uses low-speed diesels for propulsion and four to five auxiliary machines that power cargo pumps, supply onboard electricity, and also reliquefy any boiled-off gas and pump it back to the ship’s tanks, with all DLR machines burning heavy fuel oil. DFDE propulsion consists of four to five identical engines to power the alternators, and the generated electricity is used for propulsion, pumps, compressors, and all auxiliary systems and accommodation needs. The DFDE’s great advantage is that it can burn any mixture of naturally boiled-off gas and liquid fuel (heavy fuel oil or marine diesel). Naturally boiled-off gas amounts to 0.11–0.15% of the cargo load, enough to produce about 20 MW of kinetic energy for a tanker carrying 150,000 m3 of the gas. That could supply roughly two-thirds of the total demand when sailing fully loaded at 20 knots (37 km/h), and it would be far higher than 1.5 MW needed when a ship waits to enter a terminal or 7.5 MW needed when discharging its cargo (Smil, 2010a).

  As the world’s fleet of LNG ships expands, so will the use of LNG in dual-fuel engines of new tankers, but there is a much larger potential as LNG-powered propulsion can be adopted by other vessels: LNG tankers aside, in 2013, more than 30 ships were powered by LNG, and another 30 were to enter operation within 2 years, but that total is still only about 0.1% of all ocean-going cargo vessels. A key factor that will govern the rate of future adoption of (or conversion to) LNG will be the implementation of stricter air pollution rules, globally and for existing emission control areas (ECA) in North America and Western Europe and their eventual ECA extension to Asian coastal waters. The global limit on sulfur in fuel is 3.5%, while the existing ECAs—the Baltic Sea and the North Sea, 200 nautical miles zone along the Canadian and US coastal waters (including southern Alaska and Hawaii) and the US Caribbean—got 1% sulfur limit since 2010, and this demand is met by new low sulfur fuels, 380 CST and 180 CST (Kaur, 2010).

  But standards should become much tougher, down to a mere 0.1% S in the ECAs beginning in January 2015 and globally to just 0.5% in 2020, with latter target contingent on the outcome of 2018 feasibility study (Adamchak and Adede, 2013). That assessment may show that there is not enough clean liquid fuel on the market to meet the target—but the compliance would be easy with LNG whose sulfur content is just 0.004%. In any case, several companies are planning to create LNG supply for a variety of ships, including ferries, barges, tugs, cruisers, and support vessels for offshore hydrocarbon production, laying of undersea pipelines, and construction of wind turbines. Semolinos (2013) anticipated that by 2020 LNG will capture about 5% of the overall marine fuel market, or about 3% of all LNG use, and that these shares will rise, respectively, to 10 and 5% by 2030.

  Analysis of world ship traffi
c data by Burel, Taccani, and Zuliani (2013) shows that roll-on/roll-off vessels and small- and mid-size tankers (10,000–60,000 DWT) spend most of their time sailing in the ECAs and they would be the most suitable candidates for LNG-powered propulsion. Cleaner and quieter operation of LNG-powered engines would be particularly welcome on Europe’s heavily frequented inland waterways. Since 2013, two Dutch-built LNG-powered barges deliver liquid fuels along the Rhine between the Netherlands and Switzerland, and their spark-ignition engines are up to 50% less noisy compared to conventional heavy diesels (Shell, 2013).

  Using LNG as fuel for trucks is the best practical alternative to oil-derived fuels in vehicular transportation and the one that also has several advantages (Linde, 2014). As with LNG in shipping, the combustion of LNG has lower emissions of local pollutants (particulate matter, NOx, and SOx) than diesel fuel while also generating less CO2 per unit of useful energy. Moreover, for high fuel-use fleets, LNG offers lower life-cycle cost than diesel. Truck drivers appreciate that the fuel is nontoxic and noncorrosive, that refueling is as simple and as fast as filling with diesel or gasoline, that the range is good (up to 1,000 km), and that LNG-powered engines are quieter than conventional diesel motors at normal speeds (an advantage to be appreciated by people living along heavily traveled roads; at high speeds and on steeper roads, the noise is similar) and operate with much lower vibration. For countries with plenty of domestic natural gas but with limited or no production of crude oil, LNG trucking obviates imports of refined liquid fuels.

  LNG trucking is a particularly appealing option for new, rapidly expanding fleets, with China and India being the two best examples, but benefits would be also great on congested European and US roads. The European Commission has a demonstration project for LNG-powered heavy-duty vehicles that includes participation of major truck makers (IVECO, Volvo) and eventual creation of four blue corridors (Atlantic, Mediterranean, South–North, West–East) with refueling stations for medium and long trucking distances (Hubert and Ragetly, 2013). Similarly, in 2012, Shell revealed plans to build a corridor of LNG fueling station in Alberta to serve heavy-duty truck fleets delivering materials to the province’s northern oil sands operations.

  Supply chain for road LNG uses (as well as for some coastal traffic and inland waterways) would start with small-scale liquefaction, either on fixed or movable (then preferably modular) systems with annual capacities up to 1 Mt (compared to medium-sized LNG facilities of 2–3 Mt/year and large plants above 4 Mt/year). Shell has developed the moveable modular liquefaction system that can deliver LNG on a smaller scale, enough to meet local and regional transport requirement but requiring much lower capital expenditure than the standard large liquefaction facilities. Trailer trucks (or LNG bunker vessels) would distribute the fuel to LNG refueling stations (Figure 7.4).

  Figure 7.4 LNG filling station.

  © Corbis.

  But the progress has been slow and uneven. Early LNG adopters have experienced mixed results. Some have found that higher maintenance cost and lower-than-expected operating efficiency have greatly extended their expected payback periods; others had higher-than-expected energy needs for diesel fuel required for compression ignitions and for in-cab methane detectors. In 2013, by far the largest North American LNG truck order was by the UPS for 700 vehicles, with the second highest order in the United States for just 36 trucks (Raven) and the largest Canadian order (by Bison transport) for only 15 trucks (Truck News, 2014)—in a market that sells about 250,000 units a year to an expanding fleet of more than 3.5 million units. In 2014, FedEx announced plans to fuel 30% of its fleet by LNG before 2025, Procter & Gamble aims at 20% share within 2 years, but in 2014 Shell pulled back on its LNG expansion plans in Alberta.

  China had about 50,000 LNG-powered trucks in 2012 with plans for nearly 250,000 in 2015, and that would be still less than 5% of the country’s heavy-duty truck fleet (Hong, 2013). But the Chinese–owned Blu LNG scaled down its plans for the US expansion after it had built only half the number of LNG refueling stations it planned to have in place by the end of 2013 (Groom, 2014). Moreover, the newest 12 l gas-powered engine (a joint venture by Cummins and Westport) is best suited for smaller loads on flatter routes, and most of the orders have been for its compressed natural gas (CNG)-fueled configuration rather than for LNG.

  7.2.2 CNG

  At about 20 MPa, CNG is nearly 130 times as dense as ambient methane (volumetric density of 128.2 g/l compared to 0.761 g/l), but it has a much lower density than LNG (428 g/l). Much as other gas-fueled machines, CNG-powered vehicles reduce emissions of urban air pollutants: compared to diesels, CNG delivery trucks generate 95% less particulate matter, 50% less NOx, and 75% less CO, while CO2 reductions for buses are between 13 and 23% (Werpy et al., 2010). Marbek (2010) study found the following reductions of greenhouse gas emissions compared to diesel or gasoline engines: 23% for heavy-duty LNG-fueled and 19% for medium-duty CNG-fueled trucks and 23% for passenger cars running on CNG. In addition, unlike with LNG, refueling with CNG does not require any mask and gloves, and CNG-powered trucks are easier to maintain than LNG vehicles.

  But road transportation powered by CNG faces more obstacles than does LNG for trucking, and it would not be enough just to remove or to lower one or two of them before CNG-fueled vehicles could find a wider acceptance in countries with high degree of automobilization and highly developed car market that offers highly competitive combinations of vehicles, fuels, and engines, with hybrids, plug-in hybrids, purely electric vehicles, and new clean diesels in the mix. The gap in model choice is enormous: North American market now offers more than 250 car models sold by more than 40 manufacturers, but only four US companies are making natural gas engines (conversions of diesels), and in 2013, American Honda was the only car manufacturer offering CNG vehicle (Civic GX costing about $5,000 more than gasoline-fueled Civic EX), while GM and Chrysler began to sell CNG-fueled pickups (with $11,000 premium) in 2012 (Fraas, Harrington, and Morgenstern, 2013).

  Model choice is also limited in a few low-income countries that have promoted CNG vehicles and set up their domestic manufacturing: Toyota and Suzuki in Pakistan are the best example, but most Brazilian CNG vehicles are, as is the case in the United States, after-market conversions. Conversions (done by certified outfitters) include pressurized storage tanks (which take up useful space and increase vehicle’s tare), pressure reducer (to the level of the engine’s fuel-management system), shut-off valve, and fuel injectors. Conversion are expensive: Marbek (2010) study estimated additional capital investment of $8,000 for cars, about $50,000 for buses, and $90,000 for long-haul trucks. Obviously, these costs reduce the advantage of cheaper fuel, lengthening the payback period for a typical passenger car (driven less than 20,000 km/year) to 6–8 year compared to about 3 years for long-haul trucks. Higher maintenance requirements (also due to the need for periodic high-pressure tank testing) add to the lifetime ownership cost.

  Refueling remains a challenge in most countries: in December 2012, Pakistan had some 3,300 stations and China nearly 2,800—but the US total was just 1,120 (compared to 160,000 serving gasoline); France, 149; and Canada, just 47 (IANGV [International Association of Natural Gas Vehicles], 2014). Moreover, a large share of these stations is not open to public. Putting in place a requisite refueling infrastructure poses a chicken-and-egg dilemma that is shared by electric vehicles: what come first, investment in widely accessible refueling infrastructure or a large vehicle base?

  The cost of an additional refueling infrastructure would not be low. Gallagher (2013) estimated that the investment needed for CNG fuel dispensing in the US would be $100–200 billion (but only $10–20 billion for widely available LNG fleet refueling). And a limited range of CNG vehicles (about 150 km on a full tank) is another inconvenience.

  All of these barriers explain why centrally fueled urban and suburban vehicle fleets operating within a limited area are the best candidates for fueling by CNG, either in bifuel mode (two separate fuelin
g systems) or as vehicles powered only by gas. Model choice is largely irrelevant while suitable engines are readily available, conversion costs are repaid more rapidly for vehicles driven at least 50,000 km/year, and centralized refueling is no problem. That is why most CNG vehicles now operating in modern economies belong to high fuel-use urban fleets (Figure 7.5). Almost 20% of America’s urban buses run on CNG, and in 1998, the Supreme Court of India ordered complete switch to CNG, later enforced by heavy fines to operators of diesel buses who refused to accept new vehicles (Marbek, 2010). Other common CNG-powered vehicles include garbage and delivery trucks, taxis, and shuttles.

  Figure 7.5 CNG bus in New Delhi.

  © Corbis.

  Given these realities, it is not surprising that CNG-powered transportation accounts for only a tiny fraction of vehicular traffic: in 2012, the global share was 1.28%, ranging from negligible shares in affluent economies (the United States, 0.05%; Japan, 0.06%; France, 0.03%) to 77% in Armenia, 65% in Pakistan, and 37% in Bolivia (IANGV, 2014). In absolute terms, the two leaders were Iran and Pakistan (3 and 2.9 million vehicles, respectively) followed by Argentina (2.1 million), Brazil (1.75 million), and China (1.6 million). The US total was about 128,000 (overwhelmingly fleet) vehicles (of 253.7 million) served by about 1,100 refueling stations.