Industrial gases like oxygen, nitrogen, carbon dioxide, hydrogen and others are among the most widely used commodity chemicals, with applications in nearly every industrial sector. They are used in the steel industry, metal processing, the chemical and pharmaceutical industries, and in diverse production processes for motor vehicles, electronic devices, solar cells, flat screens, glass and food. In 2008 the world market for industrial gases is expected to reach a volume of 52 billion dollars. ACHEMA 2009 will present equipment, plant and technologies for industrial gas production in Frankfurt am Main from May 11-15. This leading event for the process industries is set to attract some 4,000 exhibitors and 180,000 visitors from all over the world.

Historically, there have been two fundamentally different approaches to air separation — the use of ultra-low-temperature cryogenic distillation which is typically reserved for applications requiring tonnage quantities of oxygen or nitrogen, and the use of so-called non-cryogenic approaches, which carry out air separation at ambient temperatures using either molecular sieve adsorbents via a process called pressure swing adsorption (PSA), or polymeric membranes.
More recently, a third category of air separation has emerged. This novel alternative — which is in the process of being scaled up for commercial availability — relies on specialized ceramic membranes, which separate air into oxygen and nitrogen at high temperatures, not the super-cooled temperatures required by traditional cryogenic air separation units (ASUs). While commercialization is still years away, extensive demonstration-scale testing to date suggests that this approach will allow for the development of compact systems that produce tonnage oxygen with significant advantages, in terms of reductions in capital cost and energy requirements, over cryogenic ASUs.

Despite the maturity of the tried-and-true air separation techniques that are in use today throughout the world, these methods are constantly being re-engineered. These ongoing optimization efforts are aimed at raising overall operating efficiencies and increasing the production capacity that is possible from a single air separation train, in order to reduce the notoriously high capital costs and energy requirements associated with producing oxygen and nitrogen. And many stakeholders are eagerly anticipating the eventual commercial-scale availability of the fundamentally different high-temperature air separation techniques. Advances in all of these areas are discussed below.

Versatile gases

Nitrogen makes up roughly 78 % of the air we breathe. Because nitrogen is inert to most materials, it is widely used to eliminate the risk of fire and explosion, in numerous industrial inerting systems and in pneumatic conveying operations. Nitrogen is also being increasingly used to improve oil and gas recovery (enhanced oil recovery, EOR; enhanced gas recovery, EGR).

A nitrogen atmosphere is also used in packaging technology to protect food and other perishable products from atmospheric contamination, using modified atmosphere packaging (MAP). And liquid nitrogen is used for food freezing, process cooling, and cryogenic grinding of plastics and other materials.
Oxygen constitutes 21 vol.% of the air. Industrial oxygen is primarily used for enhanced combustion, melting and smelting operations in steel, aluminum and copper production, and for enhanced combustion during cement and glassmaking operations, as well. Additionally it is applied as an oxidizing agent during the production of many chemicals, fuels, as well as in pulp-and-paper operations as a safer and environmentally friendly alternative to chlorine for bleaching, and an agent for delignification. It is also used to support fermentation processes in biotechnology and pharmaceutical applications, and to provide biological oxygen demand during aerobic wastewater treatment.

More recently, strong demand for very large quantities of oxygen (so-called tonnage quantities) has been spurred by steady growth in chemical process operations, for instance, that rely on oxygen-blown gasification to convert coal, petroleum coke, biomass, municipal solid waste and other feedstocks into an intermediate synthesis gas, which can then be further processed to yield electricity, chemicals or transportation fuels.

LNG developments: Driving nitrogen demand

One key driver for large-scale nitrogen demand is the expansion of the worldwide infrastructure to enable the production, transport, and use of liquefied natural gas (LNG). As long as natural gas prices were relatively low, natural gas reserves in remote or geographically unfavorable locations were typically not exploited, but flared, due to a lack of proximity to any viable markets.

However, in recent years, as natural gas prices have soared and these high prices have been sustained, many of these previously discarded natural gas reserves are now being monetized, by compressing and liquefying the gas to produce liquefied natural gas. So it can be transported by tankers as a liquid to a terminal, and distributed from there by an existing pipeline network. This allows new sources of formerly abandoned natural gas to be sold into lucrative markets that would otherwise be beyond pipeline reach. In this way, the growing LNG infrastructure helps to address the geographic imbalance between where the natural gas reserves are, and where they are being consumed.

One example is Europe’s largest LNG plant in Hammerfest, Norway. This facility, commissioned by the Norwegian petroleum company, StatoilHydro, was engineered and built by the Linde Group and came onstream at the end of September 2007. The liquification process first entails the separation of components, such as carbon dioxide, nitrogen and water, before the extracted raw natural gas, which is composed almost entirely of methane, can be liquefied. Only then can the natural gas be cooled down to its boiling point of -162 °C using liquid nitrogen. Once liquefied, it can then be transported by ship to customers all over the world.

Nitrogen is used at every stage of the LNG supply chain: for chilling, purging and liquefaction of the natural gas, and for inerting the LNG during shipboard transportation. Moreover, it is applied for blending the revaporized LNG stream to adjust its composition and maintain a desirable Btu value in order to meet both pipeline and customer specifications.

Gasification: Driving oxygen demand

Sharp increases and persistent volatility associated with crude oil and natural gas prices have led to new facilities that have been built or are under development to convert relatively low-cost feedstocks, such as coal, petroleum coke, natural gas in isolated or geographically unfavorable sites, biomass and even municipal solid waste into a synthesis gas comprised primarily of hydrogen and carbon monoxide. This “syngas” stream is then fired in a combined-cycle power plant, i.e., one that combines one or more gas turbines and steam turbines in succession, to produce electricity. This set-up is called an integrated gasification combined cycle, or IGCC, power plant. Alternatively the syngas can be further refined to produce a varied slate of petrochemicals in a coal-to-liquids (CTL) plant, or reacted catalytically using Fischer-Tropsch synthesis in a gas-to-liquids (GTL) plant to produce liquid transportation fuels, such as low-sulfur diesel, jet fuel and naphtha.

At the heart of any gasification-based facility is the partial-oxidation reactor that reacts the feedstock with a high-pressure oxygen stream to produce the syngas stream. Today, the dominant gasification technique used relies on pure oxygen. Air-based gasification techniques are also being developed; however, this approach is not discussed further here.

Today’s commercial gasification facilities are being built on increasingly larger scales, and as a result, they have enormous oxygen requirements, which continue to push the envelope in terms of the technology requirements to economically produce oxygen to meet such large-scale demand. For instance, while the first generation of commercial-scale IGCC facilities (two facilities in the U.S. and two facilities in Europe, which have now been in operation for a dozen years or more) are all in the range of 250-300 MW capacity, the pending wave of modern IGCC plants that are under development today are all slated to have power generation capacity in the order of 450 to 630 MW. While the earlier (smaller) IGCC facilities require roughly 2,000 tons/day of oxygen, the modern fleet of 630-MW IGCCs will require 5,000 tons/day of oxygen or more.

Similarly, many of the world-class gasification-based facilities that are under development today to produce chemicals or transportation fuels — including the Sasol CTL facilities in South Africa, the Oryx GTL facility of Sasol and Qatar Petroleum in Ras Laffan, Qatar, Shell and Qatar Petroleum’s Pearl GTL facility in Ras Laffan, Qatar, the Escravos facility of Chevron and Sasol in Nigeria, and others — the gigantic cryogenic ASUs that are required to supply the needed oxygen represent about 10-15 % of the total capital requirements, and have a correspondingly enormous demand for power to carry out compression and chilling for cryogenic air separation. Admittedly, this energy demand is compensated by the fact that the GTL/CTL process operates exothermally, and as a result energy in the form of released process steam can be used for the air separation.

The largest of these “mega-syngas” complexes (Pearl GTL facility) requires 30,000 tons/day of oxygen. In contrast, other standard-sized gasification-based facilities producing chemicals or fuels require around 5,000 tons/day of oxygen. For comparison, a typical steel mill consumes oxygen at a rate of 500 to 3,000 tons/d of oxygen.

Today the Linde Engineering Division plays a leading role in the GTL facility business. In 2006 Linde was contracted by Qatar Shell GTL Ltd., which belongs to the Royal Dutch Shell Group, and Qatar Petroleum to build eight large air separation facilities for the Pearl gas-to-liquids facility (see above) in Ras Laffan Industrial City, Qatar. This plant will represent one of the world’s largest integrated GTL complexes of this kind. For this large-scale project the Linde Engineering Division will provide the enormous oxygen demand of around 30,000 tons/day. Consequently, the Qatar contract is the largest ever tendered for ASUs. To date, four cold boxes have already been delivered. By mid-2009 the remaining four will also be on site.

To meet the enormous oxygen requirements of today’s modern gasification-based facilities, the pressure is on to build larger and larger cryogenic ASUs, and today, operators tend to seek ASUs with the largest possible single train size, in order to take advantage of economies of scale and thereby reduce overall capital cost requirements associated with air separation at their facilities. The industrial gas industry has responded to these market pressures, and delivered advanced systems that meet the needs of the end user. For instance, just a decade ago, the largest single-train cryogenic ASUs available could deliver 3,500 short tons/day of oxygen. Today, single-train ASUs that deliver between 3,500 and 5,000 short tons/d of oxygen are commercially available, and even larger systems are already under development.

Oxyfuel Combustion: Another driver for oxygen

Ongoing developments in a process called Oxyfuel Combustion (also called Oxy-Coal Combustion technology) are also expected to drive demand for tonnage oxygen in the years to come. Using Oxyfuel Combustion, coal is burned in a high-oxygen environment, rather than in an air-based environment. By eliminating nitrogen (from the combustion air) from the outset, the resulting combustion fluegas stream becomes more highly concentrated in carbon dioxide (CO₂) — the main industrial pollutant that is implicated in global warming — and virtually nitrogen-free. Thus the Oxyfuel Combustion process makes it easier to capture, compress and liquefy the concentrated CO₂ stream for use in underground sequestration EOR applications. By comparison, when coal and other fossil fuels are combusted in air, the resulting fluegas stream is relatively rich in nitrogen, but relatively dilute with respect to CO₂.

The process has the potential to enable the capture of 98 % of CO₂ from coal-fired operations, according to Air Products, which is one of several international companies that is involved with the development and commercialization of the Oxyfuel Combustion process. As pressure mounts worldwide to curb greenhouse gas emissions, this oxygen-enhanced combustion process is being eyed as a potential retrofit option to make existing coal-fired and oil-fired power plants and other industrial combustion facilities, blast furnaces and cement plants more amenable to cost-effective CO₂ capture (by eliminating nitrogen from the fluegas and creating instead a fluegas that is concentrated in CO₂ and water vapor).

In September 2008 the Linde Group and Vattenfall Europe Technology Research GmbH, a subsidiary of the Vattenfall energy group, agreed on a comprehensive technology partnership for CO₂ separation in coal-fired power plants. This cooperation targets testing the Oxyfuel Combustion process for lignite and anthracite and scaling up the technology for future application in large power plants. The investigations are being carried out at a research installation for a coal-fired power plant with CO₂ capture at Schwarze Pumpe in Lausitz (Brandenburg), which Vattenfall recently put onstream. Linde has already built an air separation and CO₂ liquefaction unit for this 30 MW pilot plant. As part of this cooperation, Linde is backing Vattenfall with extensive scientific and technical support during the first test phase, ending in 2011. To date, this power plant is the world’s first Oxyfuel plant for carbon capture and sequestration (CCS) of its kind in operation.

In July 2008, Praxair Deutschland GmbH & Co. KG, a subsidiary of Praxair, partnered with Vattenfall AB of Sweden, one of Europe’s leading energy companies, to develop advanced power-generation technology that would reduce CO₂ emissions from conventional coal-fired power plants. Initially, the team will carry out a conceptual study for a possible 500-MW combined heat and power plant in Germany, which would incorporate Praxair’s oxy-coal technology, and would enable more than 90 % of the CO₂ generated by the coal-fired boilers. This project would require roughly 8,000 tons/day of oxygen. Praxair has several other demonstration-scale projects under way to demonstrate its Oxy-Coal Combustion technology.

High-temperature air separation

While cryogenic distillation remains the current paradigm for producing tonnage quantities of oxygen and nitrogen, as noted above, this approach is inherently capital- and energy-intensive. Using cryogenic air separation, the inlet air must be filtered, compressed and chilled to about -185 °C, and the liquefied stream must then be distilled in large distillation towers to separate air into its component phases (78 vol.% nitrogen, 21 vol.% oxygen, 1 vol.% argon and other trace gases), based on differences in their boiling points.

To reduce energy consumption, streamline project execution, and reduce both capital and operating costs, aggressive re-engineering efforts are under way. For instance, some industrial gas producers are already using packed towers instead of the traditional trayed towers, moreover others are pursuing heat exchangers, high-efficiency compressors and control systems that provide real-time optimization. The use of higher operating pressures in order to reduce overall column diameter is also being investigated.

Meanwhile, several of the major industrial gas producers are pursuing a variety of promising — fundamentally different — alternatives to conventional cryogenic distillation for the separation of air into oxygen and nitrogen. The newer techniques carry out air separation at elevated temperatures — on the order of 800-900 °C — using novel ceramic membranes or molecular sieves.

Furthest along the commercialization spectrum is the Ion Transport Membrane (ITM) system, which has been under development by Air Products and Chemicals (Allentown, Pa.; www.airproducts.com) for the past several years. ITM relies on patented, high-temperature ceramic membranes, which separate oxygen from air.

The ITM system relies on a series of compact modules — each about the size of a loaf of bread — according to the company. Each module contains a stack of high-temperature, selectively permeable ceramic wafers. In the existing prototype demonstration unit, each module contains tens of dozens of wafers. The early prototype of the system has demonstrated considerable capital and energy savings compared with conventional cryogenic distillation for air separation.
While a typical cryogenic distillation column stands 60 meters tall, and a given ASU system used for today’s large-scale IGCC power plants and GTL and CTL chemical and fuel facilities typically requires several such columns (all housed within insulated cold boxes), a comparably sized ITM Oxygen system, including all vessels and piping, will be about the size of just one distillation tower, according to its developer Air Products. As a result, the ITM system is expected to reduce the capital costs associated with tonnage oxygen production by 35 % or more, and require 35-60 % less energy compared to cryogenic air separation.
Air Products has been operating a 5-ton/d pilot plant since 2005, and the company is working with the U.S. Dept. of Energy to bring onstream a 150-ton/d test facility by 2009. The company’s goal is to eventually commercialize ITM Oxygen systems that can produce high-purity oxygen (>99 %) at a rate of 1,000-2,000 tons/day or more.

The Ceramic Autothermal Recovery (CAR) Process

The gas and engineering company, The Linde Group, has also been developing a high-temperature air separation process, to enable the production of tonnage oxygen more cost-effectively compared to the use of cryogenic distillation. Unlike Air Products’ ITM Process, Linde’s CAR Process is not based on ceramic membranes; instead, it carries out the adsorption and storage of oxygen at high temperatures (600-800 °C), using a fixed-bed vessel whose multiple beds are filled with extruded pellets of the ceramic perovskite, sandwiched between layers of alumina beads.
Linde’s CAR process operates much like a pressure-swing adsorption (PSA) system, in that multiple beds are cycled back and forth between adsorption and desorption modes, to enable the production of a continuous stream of oxygen. During desorption, the oxygen that was collected by the pellet bed is released by partial pressure reduction, using either hot recycled fluegas or superheated steam to flush the bed. Since oxygen adsorption on perovskite is exothermic (meaning that it yields heat), while oxygen release is endothermic, the process, once initiated, operates autothermally, so it requires little or no additional heat input, says the company.

The CAR process has been further developed over the past two years as part of a project funded by the US Dept. of Energy. Linde has constructed a 0.7-ton/day CAR pilot plant to test and validate the CAR technology with integrated Oxyfuel Combustion in partnership with the Western Research Institute (Laramie, Wyoming; westernresearch.org) at its coal combustion test facility. Together with a third partner, Alstom Power Plant Laboratories in Windsor, Connecticut, Linde devised a detailed process and profitability analysis for an Oxyfuel power plant using the CAR process for oxygen supply. The analysis has revealed that the CAR process is more efficient and cost-effective than cryogenic air separation processes. The commercial use of the process still faces considerable challenges, however, for instance with respect to efficiency, the cost of perovskite and the impact of impurities on this material.

Oxygen Transport Membranes (OTM)

Praxair is also developing oxygen-conducting ceramic membrane systems that are designed to separate oxygen from air in a high-temperature environment. One of the company’s main developmental goals for this new technology is to enable it to be integrated with the Oxyfuel Combustion Process (discussed above).

During operation, an electric current is used to separate oxygen from low-pressure, high-temperature inlet air as it travels across the membrane. Oxygen is adsorbed on a porous, electrically conductive coating that is applied to the surface of the membrane. The oxygen dissociates to form oxygen ions, which are transported through the non-porous ceramic electrolyte. Once through the membrane, the oxygen ions lose electrons, forming molecular oxygen, which is then desorbed from the membrane’s surface. Demonstration-scale testing is under way.

Non-cryogenic air separation, via membranes and molecular sieves

Many industrial applications require smaller-than-tonnage amounts of oxygen and nitrogen. Since the early 1990s, the widespread commercial availability of air separation methods based on the use of molecular sieve adsorbents, or hollow-fiber polymeric membranes to carry out air separation at ambient pressures and temperatures has helped to provide cost-effective ways to produce these valuable industrial gases onsite, right at the point of use.
Unlike cryogenic distillation systems, non-cryogenic systems separate air into oxygen and nitrogen as a function of differences in the molecular structure, size and mass of these molecules. Not surprisingly, therefore, ongoing efforts by all industrial gas companies continue to improve the efficiency and reduce the costs associated with non-cryogenic air separation.

Improvements in membrane-based air separation

Since it was first commercialized, the basic system design of today’s membrane-based air separation systems has not changed much. These systems typically involve modules, much like shell-and-tube heat exchangers, that are loaded with thousands of hollow-fiber membrane strands (each strand being thinner than a human hair). As cleaned compressed air passes through the modules and travels along the entire length of each fiber, it covers a tremendous surface area and is separated into nitrogen and oxygen by the selective membranes.

Typically, membrane-based systems are used to produce industrial quantities of nitrogen (which is the more voluminous discharge stream, since nitrogen comprises roughly 78 vol.% of air). The smaller-volume oxygen permeate stream (oxygen comprises roughly 21 vol.% of air) is typically just vented to the atmosphere, although the customer may find some use for this high-purity oxygen stream onsite.

Today, all of the major gas producers continue to seek and incorporate proprietary advances into their systems, to improve the selectivity and throughput of their membranes, and to improve module designs. When such advanced systems allow operation at lower pressures, power requirements (and costs) for air compression can be reduced. Similarly, when throughput capacity can be increased without having to increase the number of modules, such an advance helps to reduce both capital cost and operating complexity.

Today’s membrane-based systems are both compact and lightweight, and this has allowed them to be used in a growing array of useful industrial applications. For instance, membrane-based nitrogen systems are now widely used to produce nitrogen right at the point of use for inerting perishable or flammable cargo on ships, and more recently, systems that produce nitrogen are now used to provide for onboard inerting of fuel tanks on commercial aircraft.

Improvements in adsorption-based systems

Since they were first commercialized, pressure swing adsorption (PSA), vacuum swing adsorption (VSA) systems, and hybrid vaccum-pressure swing adsorption (VPSA) systems have relied almost entirely on activated-carbon-based molecular sieves (also called sorbents) to capture oxygen and produce nitrogen with 95 to 99.5 % purity, or alumina in combination with zeolite silicates to trap nitrogen in order to produce oxygen with purities from 90 to 95 %.
More recently, several industrial gas companies have commercialized systems that rely on improved lithium-based adsorbents, which have greater selectivity and higher mass transfer rates. Such advanced adsorbents, in conjunction with a variety of mechanical innovations, such as the use of radial beds (instead of conventional vertical and horizontal bed designs), improved air movers and compression systems, and simpler control schemes, are allowing for the construction of more compact systems. These advances have helped to enable modern systems whose adsorbent requirements are less than 25 % those of the original systems, whose energy requirements have been reduced by more than 20%.

Such improvements have helped the industry to triple the maximum capacity of a single-train, adsorption-based unit. For instance, 30 tons/d used to be the economic limit of using adsorbent-based systems. Today, onsite PSA systems that produce 200 tons/d of oxygen are widely available.
All of these advances have helped non-cryogenic systems for onsite oxygen production to move beyond traditional, medium-volume markets (i.e., those requiring 20 to 100 tons/d), to serve the needs of applications with larger-volume oxygen needs (i.e., those needing 100 to 300 tons/d and beyond) — customers whose options were traditionally limited to the use of an onsite cryogenic distillation systems. Such improvements continue to enable the rapid expansion of non-cryogenic air separation into non-traditional markets.

Further advances in cryogenic air separation

Hitherto, in order to meet the demands of traditional users (for instance, in the steel industry), large-scale cryogenic air separation plants were designed for the production of high-purity (>99 %) oxygen, whereas there were hardly any applications for cryogenically produced oxygen with about 95 % purity. The proliferation of different applications in the field of CCS and energy production, for example Oxyfuel Combustion, has meanwhile turned the ground rules around. For such processes, 95 % oxygen purity is adequate. Consequently, it is now paramount to optimize the energy efficiency of cryogenic ASUs. Strategies include, for example, multi-tower processes, lower process air pressure and more efficient compression. Such approaches make energy savings of about 20 % already quite realistic. Furthermore they can be fully utilized for the novel applications described. Plans to drastically reduce the energy requirements of cryogenic ASUs in tonnage quantities are well underway.

World market for industrial gases set to reach record volume of 52 billion dollars in 2008

The top ten industrial gas producers have a market share of over 90 %, with four companies controlling over two-thirds of the business worldwide. Industrial gas producers have a huge demand for equipment: the slate ranges from pumps, compressors, valves and fittings to membrane technology, cryogenics and logistics. The ACHEMA 2009 Exhibition and Congress will present a corresponding array of interesting new developments.

Who’s Who in Industrial Gases

The Linde Group (Munich, Germany; www.linde.com)
Air Liquide (Paris, France; www.airliquide.com)
Air Products (Allentown, Pa.; www.airproducts.com)
Praxair (Danbury, Conn.; www.praxair.com)
Taijo Nippon Sanso (Japan, www.tn-sanso.co.jp)
Messer Group (Sulzbach bei Frankfurt am Main, Germany; www.messergroup.com)