The complexity of many metal structures makes it impractical to form their shape from a single piece of metal. Instead, they are constructed by forming simpler components that can then be joined together.

The strength and endurance of a completed structure therefore depends on the properties of the joints that hold them together. Engineers during the 18th and 19th centuries preferred bolted or riveted joints, but modern welding processes have been proved reliable in a wide range of applications and are now almost universally employed.

With such a universal adoption, naturally comes the demand for a whole range of industrial gases and gas mixtures.

Welding processes
The many welding processes that have been developed share a common objective of causing metals to coalesce at the joint, by applying sufficient heat to raise the temperature above the melting point.

Depending on the process selected, heat may be applied by gas flame, electric current through resistance, electric arc, friction under pressure, laser light, electron beam or other energy source, and often a compatible filler metal is introduced into the joint.

Unless the process is performed in a vacuum, contact between the heated metal and the surrounding air would result in atmospheric gases like oxygen, nitrogen, carbon dioxide and water vapour being absorbed into the molten weld pool.

After solidification, these would form bubbles trapped in the metal causing the defect known as porosity.

Some processes rely on the use of fluxes that generate a protective gas and/or slag to cover and protect the molten weld metal. Others employ a shielding gas that is fed at a controlled rate over the weld zone to prevent contact with the air.

Properties of shielding gases
Five properties determine how gases contribute to a welding process:
Chemical reactivity determines whether the gas is inert or active
Thermal conductivity affects the ability of the gas to transfer heat from the electric arc to the metal surface
Density relative to air affects the minimum flow rate required to form an effective gas shield
Ionisation potential is the minimum voltage required to initiate an electric arc and this determines the rate at which energy is converted into heat or the arc energy
Oxidising potential affects the ability to cope with contaminants that enter the weld metal and the amount of slag formed

Inert components
Only two of the ‘noble’ gases are cost effective for practical use and these provide the major component of all weld shielding gases.

Processes that employ a non-consumable electrode generally require an inert shielding gas to prevent oxidation of the fixed electrode, with the exception of hydrogen, which is a reducing agent.

The first inert gas to be used commercially as a shielding gas, helium has very low density and requires relatively high flow rates.

A bi-product extracted from natural gas, helium’s extremely low boiling temperature of -268°C means that it can only be transported over distance in vacuum insulated containers with liquid nitrogen refrigeration.

Naturally, this makes it very expensive compared to argon, unless used close to its source.

Its high ionisation potential produces a somewhat unstable arc with high energy concentration and this produces a deeper, broader weld penetration than argon and enhances sidewall fusion, especially when welding highly conductive metals like copper and aluminium.

Helium is often blended with argon to achieve a useful compromise of arc stability and energy transfer, but this is only justified where the application demands additional heat input.

Argon is denser than air and forms an effective atmospheric shield at fairly low flow rates. A bi-product of oxygen during cryogenic air separation, argon is available commercially in most industrial centres around the world.

The low ionisation energy of pure argon produces a stable arc that tends to stray and is difficult to control, without the addition of an oxidising compound like oxygen or carbon dioxide. Low arc energy results in narrow weld penetration and joints that can be prone to lack sidewall fusion.

The high current and relatively low voltage arc produced by argon-rich shielding gases encourages the deposition of metal as fine droplets known as spray transfer and usually results in wide weld beads of excellent appearance and minimum spatter, requiring little post weld cleaning. Argon is usually the base component in most blended welding gases.

Active components
Welding processes that use a consumable electrode can often be enhanced by small additions of active gases, including carbon dioxide, oxygen, nitrogen and hydrogen to the shielding gas.

The unique properties of these active components is exploited to improve the quality or cost efficiency of the welding process.

Carbon dioxide
Carbon dioxide has around 1.5 times the density of air and provides excellent heat transfer properties because in its relatively high energy arc, the molecule dissociates and recombines after releasing heat to the cooler metal surfaces.

The presence of dissociated oxygen in the weld zone enables contaminants to be oxidised, but this also results in the deposition of more slag on the weld surface.

These characteristics mean that the addition of carbon dioxide to a shielding gas will increase the depth and width of weld penetration, while scavenging impurities to produce welds with excellent sidewall fusion.

The oxidising potential of carbon dioxide when added to shielding gas blends is only about 10% that of pure oxygen and therefore, higher concentrations can be tolerated.

Pure carbon dioxide can be used as a low cost shielding gas for consumable electrode processes, but because it does not permit metal deposition by spray transfer, the problem of spatter at high welding speeds generally limits its use to light gauge fabrication – where aesthetic appearance is relatively unimportant.

Additions of oxygen are limited to a maximum of about 5%, because it is strongly oxidising. It improves the performance of shielding gases by stabilising the arc and reducing the surface tension of the liquid metal, to improve wetting and flow into the prepared joint.

In some markets crude argon with about 2-4% oxygen content has been used as a low cost bulk shielding gas, requiring no additional components – provided that the level of nitrogen contamination can be tolerated and the oxygen content is acceptably consistent over time.

Nitrogen additions of up to 0.5% are used for welding nitrogen-containing stainless steel alloys, because it improves arc stability and increases weld penetration, while improving the mechanical properties and resistance to pitting corrosion.

These mixtures should not be used on carbon steel because of the risk of porosity.

The addition of hydrogen tends to narrow the arc, increase its temperature and improve the fluidity of molten metal.

Chemically it acts as a reducing agent that can offset the oxidising effect of carbon dioxide and oxygen.

It is often added to argon up to concentrations of around 10% for high speed welding of stainless steel tubing. Its solubility in many carbon steel alloys results in serious embrittlement, as trapped hydrogen atoms exert intermolecular forces in the metal after cooling.

Nitric oxide
The production of ozone can be reduced by the addition of nitric oxide and it also stabilises the arc when welding aluminium and stainless steel.

Commercial welding gas blends
There is a wide variety of welding gas blends available that are sold under a profusion of proprietary brand names and this can complicate a welding engineer’s task.

Argon and helium are the only inert components, with pure argon as the preferred option for low cost properties. Helium blends for use on high conductivity metals may vary from 25% to 75% depending on the degree of performance required. Hydrogen blends may be selected for stainless steel welding only.

A few guidelines are given below to aid in the selection of a suitable blend to give the required weld properties:

If weld appearance and low spatter are the first priority, then the mixture with the lowest carbon dioxide, helium or nitrogen blend yet consistently adequate penetration & sidewall fusion should be selected
If mechanical properties are first priority then the blend with the highest concentration of carbon dioxide, helium or nitrogen, that still produces acceptable weld appearance and spatter, is recommended. The carbon dioxide content should be limited when welding low carbon alloys
If a given shielding gas enables adequate fusion and weld integrity, but some undercutting occurs, then a blend with higher oxygen content may be ideal because of the improved wetting out and metal flow. Additional oxygen will also encourage spray transfer deposition and can reduce spatter
Welding thin material necessitates operating in short-circuit or dip transfer mode, to minimise heat input and control burn-through. High carbon dioxide blends are recommended, because they suppress the transition into spray transfer, making the process more controllable
Finally, the oxidation potential of the shielding gas should be checked by evaluating the amount of surface slag deposits, versus tolerance for impurities evidenced by porosity or fume. A useful calculation will enable the comparison of different blends by multiplying the percentage of oxygen by 10 and adding the result to the percentage of carbon dioxide. Hydrogen may be added to reduce the oxidising potential

Despite considerable technical improvements and innovative digital control systems incorporated into modern welding machines, there is no doubt that careful selection of the shielding gas based on understanding the properties and effect of each component is still vitally important for achieving the desired weld quality at acceptable cost.