by Vaclav Smil
Before WWII, natural gas was the preferred raw material for carbon black in the US: plants in or near oil fields took advantage of inexpensive natural gas associated with oil production to meet the rising rubber demand created by new automobile age. In 1920, 8% of the US natural gas production was used by carbon black industry, and by 1940, that share rose to almost 18% (Schurr and Netschert, 1960). The industry relied on the thermal black process, with the gas injected into a reactor that contained a limited supply of air and the ensuing thermochemical decomposition produced carbon and hydrogen-rich combustion gas; the gas was burned off to generate heat required for the decomposition; and the carbon was separated, densified, and processed to specifications by screenings or pelletizing. Once the gas became more valuable, the industry switched to oil: since the mid-1970s, most of the material (now about 8 Mt/year) has been derived from heavy aromatic oils by the oil furnace process, and by the year 2000 only one out of 23 US carbon black plants was using the natural gas-based carbon black process (Crump, 2000).
Hydrogen produced by steam reforming of methane now accounts for roughly half of all H2 derived from fossil fuels (the other half comes mostly from steam reforming of liquid hydrocarbons and coal, while the electrolysis of water, done with inexpensive hydroelectricity, supplies less than 5% of the global demand). Hydrogen’s principal uses are in crude oil refining, for cracking, dearomatization, and desulfurization (Chang, Pashikanti, and Liu, 2012). Hydrogen now has a role in processing almost 4 Gt of crude oil every year, and with the average demand being roughly 0.5% (or 60 m3/t of oil) of the total refinery crude throughput, total consumption is on the order of 20 Mt H2 a year. The rest of some 50 Mt on the global market is roughly split between chemical syntheses (producing methanol, polymers, solvents, and pharmaceuticals) and a variety of industrial applications ranging from glass making to food processing (unsaturated fatty acids are hydrogenated to produce solid fats), with a minor share going for liquid rocket fuel.
4.3.1 Ammonia Synthesis
No use of methane as a raw material is more important than being a dominant source of hydrogen for the synthesis of ammonia (NH3), the world’s most important nitrogenous fertilizer without whose applications the global agriculture would not be able to feed at least 40% of the current humanity (Smil, 2001). Nitrogen, phosphorus, and potassium are the three macronutrients whose adequate supply is essential for high crop productivity. Potassium is supplied simply as potash (KCl, potassium chloride), a relatively abundant mineral that requires just mining and crushing before application to crop fields. Phosphate mining is highly concentrated in a few countries (the United States, Morocco, China), and the mineral is reacted with acids to prepare compounds (superphosphates) that yield the nutrient in forms that are more readily available to growing plants.
In all traditional agricultures, nitrogen came only from the recycling of organic matter (manures, crop residues) and cultivation of leguminous crop (symbiotic with nitrogen-fixing bacteria). The nineteenth century saw a relatively brief and intensive exploitation of guano (accumulated bird droppings on some subtropical islands) and mining and rising sales of the first inorganic form of fertilizer nitrogen in the form of Chilean nitrates. By the beginning of the twentieth century, there were two other minor inorganic contributors (ammonium sulfate from coking operations and synthetic cyanamide), but a truly revolutionary breakthrough was achieved only in 1900 when Fritz Haber succeeded to synthesize ammonia from its element under high pressure and in the presence of a metallic catalyst. Carl Bosch of the BASF (at the time the country’s largest chemical company) led the effort to convert Haber’s bench demonstration to a commercial process, and the world’s first ammonia plant began to operate in October 1913 (Smil, 2001; Figure 4.4).
Figure 4.4 Fritz Haber and Carl Bosch.
This innovation removed what has been always the most common limit on crop yield in all areas receiving adequate precipitation or irrigation, the availability of reactive nitrogen. Obviously, the synthesis of ammonia from its elements requires affordable supply of the two constituent gases, nitrogen and hydrogen. The former has been obtained since the 1890s by liquefaction of air and separation of its most abundant constituent. BASF’s first plant, as well as all other early enterprises built in Europe during and after WWI, based their hydrogen production on coal. The process began with generating CO-rich gas and then, in the presence of metallic catalysts, transforming CO and steam into CO2 and producing hydrogen .
German engineers knew that naphtha, a mixture of light liquid hydrocarbons, would make a much better feedstock than coal, but it was expensive and not readily available in the post-WWI Europe. BASF had eventually (between 1936 and 1939) developed a new process of partial oxidation that made it possible to use almost any hydrocarbon regardless of its molecular weight; as a result, heavy fuel oils became common feedstocks for NH3 synthesis, particularly in India and China (Czuppon et al., 1992). But because partially oxidized heavier hydrocarbons have lower H:C ratio than CH4, their use as feedstocks requires larger inputs as well as larger facilities for handling larger volumes of CO and CO2.
Methane, the lightest homolog of the alkane series with the highest H:C ratio of all hydrocarbons, is the most economic choice for steam reforming, with every molecule of CH4 producing three molecules of H2 . Although no natural gas was available in post-WWI Germany, Georg Schiller, a BASF engineer, discovered during the 1920s how to reform CH4 in an externally heated tube oven using a nickel catalyst. The process was licensed to the Standard Oil of New Jersey which began to produce hydrogen in its Baton Rouge refinery in 1931, but the first ammonia plant using steam reforming of methane was built only in 1939 (Hercules Powder Company in California).
Postwar expansion of ammonia synthesis in the United States and the USSR was based entirely on the steam reforming of natural gas. European synthesis remained coal based until the late 1950s, but then the rising imports of crude oil and the discovery of natural gas in Groningen and in the North Sea and the imports from Siberia brought a fast shift to hydrocarbon feedstocks. Elsewhere, only China kept on building a relatively large number of smaller-capacity coal-based plants. This dependence was further strengthened since the 1960s with the replacement of old reciprocating compressors by much more efficient centrifugal machines.
Until that time, ammonia plants had a separate reciprocating compressor for each of its parallel synthesis loops known as trains. These compressors were first powered by coke-oven gas or by steam, after WWI mostly by electric motors, and although efficient they were also expensive (both to install and to operate) and could not handle very large volumes of air, limiting the daily capacity of a single train to no more than 300 t NH3. A new design was introduced for the first time in 1963 by M.W. Kellogg Company of Houston, now part of Kellogg Brown and Root, a leading engineering, construction, and service company that has been involved (in the licensing, design, or actual building) of roughly half of the global ammonia capacity (KBR, 2014).
In Kellogg’s new plant, all compression needs were supplied by a single centrifugal machine powered by steam turbines, with the steam produced very efficiently by combustion of natural gas. This change—using natural gas as both the feedstock and the fuel energizing the process—allowed integration of the plant’s energy needs, maximized the overall energy conversion efficiency, and made it possible to produce much larger volumes of NH3 at a much lower cost. While the pre-1963 plants operated typically with pressures of 30–35 MPa in the synthetic loop, the latest ammonia plants operate with pressures below 10 MPa. Capacity of the first M.W. Kellogg single-train plant in Texas City was 600 t NH3/day, soon afterward the first plant with daily output of 1,000 t came, and more than two dozen similar facilities were at work by the late 1960s.
During the 1970s, the largest order for these new plants came from China, driven by the existential need to increase the country’s stagnating crop production (Smil, 2004). By the end of the twentieth century, more than 80% of the global NH3 syn
thesis capacity relied on steam reforming of methane, all large plants built since the year 2000 are natural gas based, and the largest single-train plants under construction now having daily output capacities of 4,000 t. But the sequence of ammonia synthesis has not changed: production of feedstock gases is followed by a shift reaction that greatly reduces the volume of CO and yields more H2 and CO2, then the two carbon oxides are removed, and pure N2 and H2 undergo high-pressure catalytic synthesis.
Natural gas delivered by a pipeline is first purified by removing traces of H2S and particulate matter, and then it is preheated and compressed to the reformer pressure and mixed with superheated steam. Its primary reforming (inside heated steel alloy tubes packed with nickel catalyst) yields a mixture of H2 and CO , and it is the largest energy user in any ammonia plant. The gas is then led to the secondary reformer (a cylindrical vessel filled with a suitable catalyst) where the unconverted CH4 (between 5 and 15% of the initial volume) is oxidized to yield CO2 and water. The resulting gas contains 56% of hydrogen, 23% of nitrogen, 12% of CO, 8% of CO2 and less than 0.5% methane as well as residual water. After cooling, a catalytic shift reaction produces more hydrogen , and all remaining CO2 is removed by adsorption using either aqueous ethanolamine solutions or pressure-swing absorbers filled with microporous aluminosilicates (zeolites). Any residual traces of CO or CO2 are eliminated by catalytic methanation .
Synthesis gas (74% H2, 24% N2, 0.8% CH4, and 0.3% Ar) is then compressed (6–18 MPa depending on the process), heated (to 400–450°C), and converted to ammonia in the presence of catalysts (formerly all based on magnetite (Fe3O4) and promoted with Al2O3, KCl, and Ca, more recently also with ruthenium). The exit gas contains 12–18% NH3; ammonia is refrigerated and stored; and the unreacted gas is compressed once again and led back into the converter. Besides America’s KBR, other major licensors, consultants, and builders of ammonia plants are the Danish Haldor Topsøe, Germany’s ThyssenKrupp Uhde GmbH, and the Swiss Ammonia Casale.
The worldwide switch to natural gas as fuel and feedstock for ammonia synthesis has transformed the industry. Its post-1950 growth has not been steady. Steady gains continued until the late 1980s, driven first by expanding crop production in the United States and Europe (both with growing populations and higher demand for animal foods), then by the nitrogen requirements of high-yielding rice and wheat cultivars of Asia’s green revolution (Smil, 2000). The global output of ammonia (with about 80% of it destined for fertilizers, the rest for a variety of further chemical syntheses) rose from just below 5 Mt in 1950 to about 50 Mt in 1975. The USSR displaced the United States as the world’s largest producer, and in turn, it was surpassed by China. Then the record output in the year 1989 was followed by a pullback during the early 1990s (largely due to declining output in post-Soviet states and reduced demand in Western countries brought by more efficient use of nitrogenous fertilizers) and then came renewed expansion.
The global capacity of ammonia plants surpassed 180 Mt in 2010 and it will be about 230 Mt in 2015 (FAO, 2011), and the actual output of ammonia reached 159 Mt in 2010 and 170 Mt in 2013 (equivalent of 140 Mt N), with nearly 56 Mt in China, almost 15 Mt in India, 12 Mt in Russia, and 10.6 Mt in the United States (USGS [United States Geological Survey], 2014). This means that ammonia production has been one of the world’s two most important chemical syntheses when measured by the total output. In mass terms, the production of sulfuric acid and ammonia has been almost identical, but ammonia’s lower molecular weight (17 vs. 98 for H2SO4) means that the gas has been the most important product when compared in terms of synthesized moles. Large-scale natural gas-based syntheses pioneered in the United States of the 1950s and 1960s have left their lasting imprint on the worldwide industry: the largest plants are located near major source of natural gas, and they are often integrated with production of more complex solid or liquid nitrogenous or mixed fertilizers.
Largest US plants (with annual capacities in Mt) are in Kenai in Alaska (0.63), Donaldsonville in Louisiana (four plants totaling 2.04 Mt), and Enid and Verdigris in Oklahoma, each nearly 1 Mt (IFDC [International Fertilizer Development Center], 2008). In Canada, they include Redwater and Carseland in Alberta, but the Western hemisphere’s largest ammonia plant concentration is on the western coast of Trinidad in Point Lisas where the island’s offshore gas fields feed 11 plants with the combined annual capacity of 6 Mt in 2013. Saudi plants are concentrated in al-Jubail on the Persian Gulf near the giant hydrocarbon fields of the country’s Eastern Province, and the world’s largest new plant (1.2 Mt/year capacity) will be completed in 2016 in Ras al-Khair, just north of al-Jubail. Russia’s largest concentration of ammonia plants (seven facilities with the total annual capacity of 3.15 Mt) is in Tolyatti on the Volga, close both to the country’s major (now declining) hydrocarbon basin (Volga–Ural) and a large hydrostation (Kuybyshevskaya on the Volga).
Ammonia has the highest nitrogen content (82%) of all fertilizers, but storage and applications of anhydrous ammonia require special equipment (tanks, hollow knives injecting ammonia into soil) and have been largely limited to the United States and Canada. Various ammonia solutions may be preferable, and because urea (CH4N2O), containing 45% N, is the most concentrated solid nitrogen fertilizer that is easy to store and apply, it has become the world’s leading source of crop nitrogen, mainly because of its dominance in rice-growing Asia. Other common choices include ammonium nitrate (NH4NO3) and ammonium phosphate (NH4H2PO4).
Gas-based synthesis is the least expensive option: plants based on the other two feedstocks (heavy oil and coal) cost 1.4–2.4 times as much as to build, 20–70% more to operate, and need 30–70% more energy. Switching to methane from the original coke-based synthesis was a major reason for impressive energy savings (Smil, 2001). Energy requirements of ammonia synthesis include fuels and electricity used in the process and energy embodied in the feedstocks. The first commercial operation in Oppau began with more than 100 GJ/t NH3 in 1913, and during the late 1930s, coke-based plants consumed around 85 GJ/t NH3. During the 1950s, natural gas-based plants using low-pressure reforming and reciprocating compressors required between 50 and 55 GJ/t NH3, and by the early 1970s, when high-pressure reforming and centrifugal compressors became common, the rate was reduced to just 35 GJ/t NH3.
By 1980, many plants required just 30 GJ/t NH3, soon afterward new plant designs by M.W. Kellogg and Krupp Uhde reduced the rate to just below 29 GJ/t NH3, and the best current performances are 27–28 GJ/t NH3 based on natural gas, only about 30% higher than the stoichiometric minimum of 20.9 GJ/t (Worrell et al., 2008). Obviously, average global performances have been considerably less efficient (mainly due to higher energy intensity of plants that still rely on reforming heavier hydrocarbons or coal) as they declined from around 80 GJ/t NH3 in 1950 to just over 50 GJ/t in 1980, to 45 GJ/t by the year 2000, and, according to the IEA, to a global weighted mean of 41.6 GJ/t in 2005 (ICF International, 2007).
4.3.2 Plastics from Natural Gas
The age of plastics depends heavily on two heavier hydrocarbons present in natural gas, above all on ethane, the second alkane in natural gas and the most voluminous component of natural gas liquids, and on propane. These constituents of natural gas provide the monomers for the production of the three dominant polymers—polyethylene (PE), polypropylene (PP) and polyvinyl chloride (PVC)—whose consumption now accounts for roughly three-fifths of the global market for synthetic materials. Ethylene is also produced, more expensively, by cracking liquid naphtha obtained by the distillation of crude oil, and the research is advancing on methods to derive ethylene from biomass feedstocks—but the dominance of inexpensive ethane-based ethylene will not be dislodged easily.
Steam cracking of ethane produces ethylene, and after purification, it is catalytically converted to different kinds of PE. As already noted (in Chapter 2), ethane concentrations (by volume) range mostly between 2 and 7%, with extremes from 1 to 14% (Groningen 2.0%, Hugoton, TX 5.8%, Hassi
R’Mel 7%, Agha Jari, Iran 14%), natural gas processing reduces high ethane level to concentrations prescribed by pipeline operators, and the separated ethane become a valuable feedstock. The only time this sequence presents problems is when new gas with high ethane levels is produced in regions that lack ethane-based industries. That has been the case in parts of the Marcellus shale in Pennsylvania where some wells produce gas with up to 16% of ethane (Martin, 2010).
About half of the world’s ethylene production is polymerized: polymerization (conversion of small molecules, monomers, into long chains or networks of molecules, polymers) produces a variety of PE, the world’s most important group of plastic. Polymerization process was first patented in 1936 by the British ICI, and commercial production of low-density polyethylene (LDPE) began in September 1939. High-density polyethylene (HDPE) was introduced during the 1960s; its density is only a bit higher than that of LDPE (0.96 g/cm3 compared to 0.93 g/cm3), but its melting temperature is 20°C higher (135°C vs. 115°C), and its tensile strength is more than three times as much as for LDPE. During the 1970s came the commercial synthesis of linear low-density polyethylene (LLDPE) that is stronger than LDPE and can be turned into thinner films.
Global production of ethylene is now on the order of 160 Mt/year, with the Middle East (above all Saudi Arabia, taking advantage of its NGL supply), the United States, and China being the larger suppliers. Extrusion, molding, casting, and blowing turn different kinds of PE into a huge array of products for both ubiquitously visible and hidden uses. Thin PE films are used for shopping and garbage bags, food bags and wraps, and cover sheets (now common for cultivating vegetables); LDPE makes huge impact-resistant water tanks as well as soft bubble wraps; and LLDPE is used as frozen food bags, for heavy-duty liners, for pools, and for geomembranes. Leading hidden uses include house wraps, water pipes, insulation for electrical cables, and materials for knee and hip replacements. Given their widespread use, it is fortunate that PE products (triangle symbols 2 for HDPE, 4 for LDPE) are the most commonly recycled plastics.