Shadow Economy in the Industrial Gas Industry

The industrial gas sector is built on complex, energy-intensive physics that very few people in a company — let alone its owners — fully understand. That asymmetry of information is not incidental. It is the foundation on which a multi-layered system of institutionalised profit extraction is constructed. This analysis maps seven distinct levels of that system, from the thermodynamics of the air separation unit to the criminal falsification of cylinder inspection certificates.

Everything begins at the air separation unit (ASU). Cryogenic air separation — compressing atmospheric air, cooling it to near absolute zero and distilling it by differences in boiling points — is the fundamental technology for producing oxygen, nitrogen and argon at industrial scale. It is also the most energy-intensive process in the facility: energy costs account for 29–32% of operating costs, and the compression stage alone represents 34–36% of capital costs.

The theoretical minimum specific energy consumption for producing gaseous oxygen by cryogenic separation is 0.28 kWh/Nm³. Modern industrial standard is 0.30–0.35 kWh/Nm³. The gap between standard and actual performance on real facilities can be dramatic:

Engineering and management staff have an impeccable scientific term for explaining these deviations to financial control: “exergy destruction.” Exergetic analysis of the compression zone often identifies losses of up to 37% of total electrical energy consumed. The largest individual losses appear in compressors (up to 1,325 kW in documented installations), multi-stream heat exchangers (519 kW) and the high-pressure column (448 kW).

For the CFO, these figures look like an unavoidable technological tax — the objective laws of thermodynamics. What is hidden behind this complexity is a subsidy mechanism. A significant portion of the energy written off as exergy destruction is not lost as heat. It is used to intensify compressor operation beyond the standard process regime, producing high-margin by-products — primarily argon, and in some cases krypton and xenon — that are then removed from official commercial accounting. The owner pays multi-million electricity bills. The inflated cost base is loaded onto the primary product. The net margin from unaccounted product accumulates in the shadow consortium operating within the plant.

Argon constitutes only about 0.93% of atmospheric air by volume. Yet it is a high-value product — irreplaceable in metallurgy, welding, electronics manufacturing and heat treatment. The global argon market, valued at approximately $11.6 billion in 2024, is forecast to reach $20.8 billion by 2033 at a CAGR of 6.0–6.7%. In Europe, spot prices reached $0.54/kg in late 2025 driven by energy costs, planned maintenance and semiconductor demand. Every unaccounted cylinder of argon is a highly liquid asset.

Extracting argon is thermodynamically complex. In a standard ASU, argon concentrates in a specific zone of the low-pressure column (LPC) and is drawn off as a side vapour stream to a dedicated crude argon column. The separation is critically difficult due to extremely small differences in relative volatility and boiling points — temperature differentials in cryogenic heat exchangers measure only a few Kelvin.

Modern plants use Model Predictive Control (MPC) systems to manage this highly coupled, non-linear process with significant inertia. The MPC continuously adjusts manipulated variables to maintain setpoints. It is precisely at the programming level — or via manual override of MPC inputs — that product extraction is implemented.

Institutionalisation reaches its peak when the plant manager or an affiliated party creates a sole-trader entity or intermediary company. This company formally rents two square metres of office space from the plant and receives rights to sell “surplus” gases or provide cylinder-filling services. Officially the plant produces oxygen for steel mills. In practice, its compressors — consuming megawatts of electricity paid for by the owner — run to produce unaccounted argon generating net profit for the plant manager’s private enterprise.

If argon is the mass instrument of shadow enrichment, krypton and xenon represent the elite tier — accessible only to the senior technical management of the largest cryogenic complexes. These gases are present in the atmosphere in microscopic trace amounts. Their extraction is economically viable only on very large ASUs producing over 2,000 tonnes of oxygen per day. Global production is structurally limited: approximately 75,000–90,000 m³ of krypton and 25,000–35,000 m³ of xenon are produced globally each year. These volumes are rigidly constrained by the physical capacity of existing ASUs and cannot be rapidly increased through investment.

The apparent paradox in this table — extraction costs for krypton exceeding its market price by 800 times — requires explanation. If a plant were built exclusively for krypton, it would be deeply unprofitable. But krypton and xenon accumulate in liquid oxygen in the main condenser as an inevitable by-product. All the massive costs of compressing, cooling and distilling millions of cubic metres of air have already been allocated to the primary products (oxygen and nitrogen). These are sunk costs. The extraction of primary krypton-xenon concentrate costs virtually nothing in marginal operating terms.

The shadow scheme exploits reported recovery rates. Management tells the board that krypton-xenon extraction is technically difficult, risks heat exchanger blockage from hydrocarbon ice, or is economically unviable. An artificially low official recovery rate is declared. In reality, extraction runs at maximum capacity. The difference between actual and declared volumes is written off as technological losses during column purging. The unaccounted concentrate is quietly packaged and leaves the facility under the guise of empty vessels or technical nitrogen cylinders, sold through specialist traders to research institutes and medical centres where the product’s provenance is rarely questioned.

Once gases are converted to cryogenic liquid — LOX, LIN, LAR — they are moved in large storage tanks or tanker trucks to customers. Here a physical phenomenon provides ideal cover for product theft: boil-off gas (BOG).

Cryogenic liquids exist at extreme temperatures (liquid nitrogen boils at −196°C). Despite advanced insulation technologies, absolute thermal isolation is impossible — heat from the environment inevitably penetrates the vessel. This heat causes the cryogenic liquid to boil, with vapour vented to atmosphere or recovered. BOG losses during marine transport can reach 6% of cargo volume depending on cycle duration. In ground distribution they are also substantial.

The scheme uses BOG as an accounting black hole. Cryogenic tanks are equipped with sophisticated level measurement systems: differential pressure sensors, radar level gauges (such as Rosemount 5900S systems providing non-contact measurement and temperature profiling), and continuous capacitance sensors. Instrumentation specialists who are part of the scheme introduce software corrections to the process controller — altering the base product density in settings or offsetting the differential pressure zero point. The system begins systematically underreporting the actual physical liquid level in the tank.

The difference between the actual volume and the instrument reading is physically drained into the tanker of an affiliated transport contractor, arriving nominally for planned equipment maintenance. In official records, the shortfall is written off as anomalously high BOG — attributed to degradation of the perlite-vacuum insulation of an ageing vessel, unusual summer heat, or frequent pipeline cool-down cycles. Verifying the true insulation condition without a full shutdown, thaw and re-evacuation of the enormous vessel is impossible. The owner pays natural product losses disguised as physics.

The 75/25 mixture adulteration scheme

In modern MIG/MAG welding of carbon and low-alloy steels, the industry gold standard is a shielding gas mixture of 75% argon and 25% CO₂. The metallurgical basis for this ratio is precise: argon ensures arc stability, spray or fine droplet metal transfer, high-quality bead formation and minimal spatter. CO₂ increases penetration depth. Between these two gases there is a vast economic gap.

CO₂ is cheap — a by-product of chemical industry or combustion. Argon requires energy-intensive cryogenic distillation. At the filling ramp, the classic adulteration scheme is straightforward: instead of the specified 75/25 ratio, operators adjust the mixing equipment to deliver 60% argon and 40% CO₂, or even 50/50. The reduction in expensive argon fraction across thousands of cylinders releases substantial unaccounted argon for packaging in “shadow” cylinders and selling as pure product — at 100% net margin.

Margin leakage in distribution

Official regional gas distributors operate under extreme margin pressure — often earning approximately 1% per transaction. To compensate, sales managers implement hidden revenue maximisation. Part of the volume runs through official contracts at list prices; the remainder is supplied informally through the manager’s sole-trader entity or an affiliated reseller. For the plant’s control systems, this appears as seasonal demand decline or competitive pricing pressure. The customer receives needed volumes at a cash discount. The manager takes his share. The plant loses both revenue and control over its cylinder fleet.

Cylinders are expensive assets. Plants generate significant revenue from daily demurrage charges. In margin leakage schemes, thousands of cylinders are fictitiously written off as lost by customers, deemed beyond repair, scrapped, or moved to long-term storage warehouses. In practice they form a shadow pool rented to grey customers for cash. Auditing a fleet of hundreds of thousands of units without universal RFID tracking is meaningless — shortfalls are discovered only years later.

Every industrial gas cylinder is a pressure vessel at 150–300 atmospheres — with the destructive force of an artillery shell if it ruptures. Laws in all developed countries, alongside international standards (DOT, ISO, PHMSA regulations), mandate regular technical inspection and hydrostatic testing of every cylinder — typically every 3–5 years for steel and aluminium industrial cylinders.

The hydrostatic testing procedure is labour-intensive and costly: complete cylinder evacuation, valve removal, visual borescope inspection for internal corrosion, water-filling, placement in a pressure chamber (water jacket) and pressurisation to measure volumetric metal expansion. The legal market cost is $30–40 per cylinder. The counterfeit alternative is simple:

The inspection station supervisor — acting in collusion with distributors and resellers — converts his station into a money press. A worker takes a metal punch with the registered inspection mark and stamps the cylinder neck with a fresh date: five more years of legal service life. The shadow fee: $15–20 per cylinder in cash — half the official rate. With thousands of cylinders monthly requiring re-certification, the shadow income is substantial.

Every ASU is equipped with an air prepurification unit (APU). Before atmospheric air enters the turbines and cryogenic heat exchangers, it must be completely freed of moisture, CO₂ and hydrocarbon traces — which would freeze at −78.5°C (dry ice temperature), instantly blocking heat exchanger channels and triggering emergency shutdown. Purification uses large adsorbers filled with molecular sieves — synthetic zeolites (3A/13X) or carbon molecular sieves. Quality industrial zeolite costs $2–10/kg. Specialised carbon sieves: $5–20/kg. Loading the adsorbers of a large ASU requires tens of tonnes, making planned replacement (every 3–5 years) a tender of hundreds of thousands to millions of dollars.

The scheme: the procurement director or chief engineer colludes with an affiliated intermediary. The tender’s technical specification is written with parameters — bulk density, dynamic capacity, particle size distribution — that nominally match only a specific supplier. The supplier procures the lowest-grade Asian-market zeolite at $1.50/kg, delivers it to the plant at $9.00/kg as an “innovative premium adsorbent with enhanced adsorption kinetics.” The margin flows to shell companies.

The cascade production damage

Low-quality molecular sieve has poor mechanical strength — it rapidly abrades to dust under cyclic pressure changes in the adsorber. This dust clogs dust filters and enters the cold box. Inferior sieve also has reduced CO₂ and moisture absorption capacity. To prevent CO₂ breakthrough into the cryogenic section, operators are forced to shorten adsorption cycle times (accelerating costly valve wear), increase purge nitrogen consumption, and raise electrical heater output for regeneration gas heating. Specific energy consumption begins climbing inexorably. We return to Level 1: efficiency falls, exergy losses grow. Management presents reports attributing the efficiency decline to natural equipment ageing, high summer humidity or atmospheric dust. The true cause — a corrupt procurement cycle — remains safely hidden behind complex chemistry.

Diagnosis and detection

The question hanging in the air at owner meetings — “Revenue is there, capacity is there, so where is the profit?” — has a precise, mathematically and thermodynamically grounded answer. Profit is not dissolving into market inefficiency or currency fluctuations. It is flowing systematically and with technical sophistication into a parallel ecosystem operating inside the legitimate body of the enterprise.