Thermal Stress Analysis of Low-Temperature Distillation Columns: Methods, Testing, and Applications

Thermal stress analysis of low-temperature distillation columns is a critical engineering evaluation for vessels in air separation, LNG, and petrochemical service, where operating temperatures range from approximately -30°C down to -196°C. By combining transient finite element simulation with low-temperature material testing, this analysis predicts stress distributions across the shell, nozzles, supports, and internals, helping engineers prevent brittle fracture, thermal fatigue, and buckling while demonstrating compliance with international pressure vessel codes. This article outlines the technical scope, sample types, testing items, methods, and typical applications of a professional thermal stress analysis program.

Technical Overview

Low-temperature distillation columns, frequently referred to as cryogenic columns, are tall, slender pressure vessels used to separate air, natural gas liquids, and light olefins at operating temperatures between approximately -30°C and -196°C. When such a column is cooled from ambient conditions to its operating temperature, the shell, internals, supports, and connected piping contract by different amounts and at different rates. Because this contraction is restrained by skirts, nozzles, external piping, tray support rings, and stiffeners, substantial thermal stresses develop in the shell wall, at nozzle-to-shell junctions, at skirt attachments, and within welded joints.

The magnitude of thermally induced stress can be estimated from the classical relationship for a fully restrained biaxial wall section, where the stress is proportional to the modulus of elasticity, the coefficient of thermal expansion, and the restrained temperature difference, divided by one minus the Poisson ratio. For austenitic stainless steel at cryogenic temperature, a restrained temperature difference of only 50 K can generate nominal stresses exceeding 200 MPa, which is above the secondary stress limits permitted by many design codes. This illustrates why formal analysis, rather than rule-of-thumb judgment, is required for low-temperature service.

Thermal stress analysis evaluates several distinct failure mechanisms that are either initiated or aggravated by low-temperature operation:

• Brittle fracture caused by reduced material toughness at cryogenic temperatures, particularly at welds and stress concentration features.
• Thermal fatigue resulting from repeated cool-down and warm-up cycles, start-stop operation, and process upsets.
• Local yielding, shakedown failure, and ratcheting under combined pressure and cyclic thermal loads.
• Buckling of vacuum-insulated inner vessels and thin shells under combined external pressure and through-wall thermal gradients.
• Distortion and cracking at dissimilar metal joints where aluminum, stainless steel, and nickel-alloy components with different expansion coefficients are joined.

The analysis is typically performed within the framework of internationally recognized pressure vessel codes, including the design-by-analysis procedures of ASME BPVC Section VIII Division 2 Part 5, EN 13445-3 for unfired pressure vessels, GB/T 150 for Chinese pressure vessel design, and GB/T 18442 for static vacuum-insulated cryogenic vessels. Material property inputs and verification testing are carried out according to ASTM and ISO standards, and independent verification can be provided by accredited third-party laboratories.

Testing Samples

Base Material Specimens

Specimens are taken from certified plate and forging heats of the materials most commonly used in cryogenic column construction. These include austenitic stainless steels such as 304L and 316L, aluminum-magnesium alloys such as 5083 used in air separation columns, low-temperature carbon and nickel-alloy steels for moderately cold service, and 9% nickel steel for liquefied natural gas applications. Sampling plans consider plate thickness, product form, rolling direction, and heat treatment condition so that the tested properties represent the actual vessel shell.

Welded Joint Specimens

Welds are usually the governing locations in thermal stress evaluation because they combine stress concentrations, residual stresses, and potential toughness degradation. Sample groups include longitudinal and circumferential butt welds, nozzle-to-shell welds, skirt-to-shell attachment welds, and dissimilar metal transition joints. Specimens are removed from qualified welding procedure qualification records so that the analysis reflects as-built fabrication conditions.

Representative Column Sections

For instrumented physical testing or mock-up validation, representative sections of the column are prepared. These may include a shell course containing major nozzles, a skirt support assembly, tray support rings, liquid distributor baffles, and internal piping manifolds. Full-thickness sections are used to establish through-wall temperature gradients during transient thermal testing.

Ancillary Components

Auxiliary samples complete the material property database required for simulation. These include bolting materials for flanged joints at cryogenic temperature, gasket and insulation materials for thermal property characterization, and short piping spools representative of the connected loads acting on column nozzles.

Testing Items

A comprehensive thermal stress analysis program combines physical material testing with computed stress and deformation results. The principal testing items are:

• Low-temperature tensile properties, including yield strength, ultimate tensile strength, elongation, and reduction of area at design minimum temperature.
• Charpy V-notch impact toughness at or below the design temperature to verify resistance to brittle fracture.
• Coefficient of thermal expansion over the full operating range, measured for each distinct material and weldment.
• Thermal conductivity and thermal diffusivity as functions of temperature, which control transient gradient development.
• Elastic modulus and Poisson ratio at cryogenic temperature, noting that the modulus of most metals increases as temperature decreases.
• Weld residual stress measurement at critical joints prior to and after thermal cycling.
• Strain-controlled low-cycle fatigue properties for components subject to frequent start-up and shutdown.
• Computed stress categories from finite element analysis, including general and local membrane stress, bending stress, secondary stress, and peak stress at weld toes.
• Predicted and measured displacements at nozzle connections, support interfaces, and internal attachment points.
• Stability margins against buckling for inner vessels subject to external pressure combined with thermal gradients.

Testing Methods

Numerical Thermal-Structural Simulation

The core analytical method is a coupled thermal and structural finite element analysis. First, a transient thermal model computes the temperature field during cool-down, steady operation, and warm-up, using realistic internal fluid conditions, ambient heat gain, and insulation performance. The temperature history is then mapped onto a structural model that includes the shell, nozzles, skirt, internals, and appropriate boundary conditions representing piping and support restraint. Results are classified into primary, secondary, and peak stress categories in accordance with design-by-analysis procedures, and acceptance is judged against the allowable limits of the applicable code. Fatigue life is subsequently evaluated using design fatigue curves for the expected number of thermal cycles over the service life.

Low-Temperature Material Characterization

Tensile testing at low temperature is performed in accordance with ISO 6892-3, with room-temperature reference testing per ASTM E8/E8M or ISO 6892-1. Impact toughness is measured by ASTM E23 or ISO 148-1 using specimens oriented to intercept the weld fusion line and heat-affected zone where applicable. Thermal expansion is determined by push-rod dilatometry per ASTM E228, while thermal diffusivity is measured by the laser flash method per ASTM E1461 and converted to thermal conductivity. Fatigue testing follows strain-controlled procedures such as ASTM E606 or force-controlled methods per ASTM E466.

Experimental Stress and Temperature Measurement

Physical measurements validate the simulation. Surface and embedded thermocouples record through-wall and axial temperature distributions during controlled cool-down. Cryogenic strain gauges mounted at nozzle welds, skirt attachments, and shell courses capture actual strain histories, from which stresses are derived using temperature-corrected gauge factors. Residual stresses in welds are quantified by the hole-drilling strain gauge method per ASTM E837 or by X-ray diffraction. Data are typically acquired and reviewed through accredited third-party laboratories when independent certification is required for regulatory acceptance.

Thermal Cycling and Validation Testing

Mock-up sections or instrumented field cool-downs are used to verify predicted stress and displacement behavior. Specimens and component sections undergo repeated thermal cycling between ambient and design minimum temperature at controlled rates, typically on the order of a few tens of degrees Celsius per hour, followed by non-destructive examination for crack initiation at weld toes and dissimilar joints.

Applications

Air Separation Units

Cryogenic distillation columns for producing oxygen, nitrogen, and argon operate close to -183°C to -196°C. Thermal stress analysis establishes safe cool-down rates for the high-pressure and low-pressure columns, evaluates brazed aluminum internals and copper piping interfaces, and confirms the integrity of long vessel trains with complex support schemes.

LNG and Natural Gas Processing

In LNG liquefaction and fractionation service, as well as in demethanizers, nitrogen rejection units, and helium recovery columns, thermal stress analysis verifies 9% nickel steel and aluminum column designs, evaluates vacuum-insulated inner vessel stability, and controls stress at interfaces between cold equipment and ambient-temperature structures.

Ethylene and Petrochemical Cold Sections

Ethylene plant cold boxes, demethanizers, and ethylene and propylene fractionators experience severe cyclic thermal loads during start-up, decoking, and turnaround sequences. Analysis of these columns addresses nozzle loadings, skirt flexibility, and fatigue accumulation over hundreds of major operating cycles.

Design Verification, Revamp, and Failure Investigation

Beyond new construction, thermal stress analysis supports capacity revamps where faster cool-down is desired, relocation of columns to colder climates, life-extension assessments, and root-cause investigation of cracking, distortion, or support distress observed in service.

FAQ

What temperature range requires a dedicated thermal stress analysis?

Formal analysis is generally mandated whenever the design metal temperature is below the toughness-verification threshold of the governing pressure vessel code, typically below approximately -29°C for carbon and low-alloy steels, and whenever cyclic or rapidly changing temperature operation is expected regardless of the absolute temperature.

Which materials are suitable for low-temperature distillation columns?

Austenitic stainless steels such as 304L and 316L, aluminum alloys such as 5083, and nickel steels are the most common choices because they retain toughness at cryogenic temperature. Material acceptance is confirmed by impact testing at or below the minimum design metal temperature.

How are cool-down and warm-up rate limits determined?

Rate limits are derived from transient finite element analysis by evaluating the through-wall and axial temperature gradients that produce secondary and peak stresses. The permissible rate is the highest value that keeps stress ranges and fatigue damage within code allowables, and it is verified experimentally with thermocouples and strain gauges.

Is finite element analysis alone sufficient for code acceptance?

Simulation must be supported by verified material properties tested at the design minimum temperature, and governing load cases should be validated by measurement. Many purchasers and regulators additionally require documentation reviewed by accredited third-party laboratories before the analysis is accepted.

How often should thermal stress analysis be reviewed?

Analysis should be revisited whenever operating conditions change materially, such as faster start-up requirements, altered feed composition, additional thermal cycles, insulation degradation, or vessel relocation, and it forms part of fitness-for-service and remaining-life assessments for aging columns.