Introduction

This article addresses the question: "what is the best dry suit inflation gas?" Consideration of the physical properties of various alternatives leads to some interesting candidates from a purely theoretical standpoint, however the choice is narrowed when we also consider safety, economics, and practicality. It turns out that while argon is not the most insulating gas, it is the most practical choice. Nonetheless, the reasoning leading to this conclusion is subtle, and the answer is not: "Argon is a good insulator because it is dense." If good insulation simply resulted from gas density, then on every deep dive, we would notice a large improvement in insulation as we add gas to our dry suits to offset the compression of depth. In fact, gas density and pressure have little effect on the conduction of heat from a diver, so what works on the surface, works just as well below.

Clearly, suit inflation gas is only one of many factors affecting a diver's thermal protection. Other important points include the choice of dry suit undergarments, the types of gloves and mask worn, and even eating an adequate meal before diving to ensure a supply of metabolic heat. I will not address these issues, which are covered by Aspacher's point-by-point analysis of heat loss mechanisms that affect divers (Reference 1), and a review of the field experience in thermal protection given at the tek95 conference (Reference 2).

Theory

In addition to maintaining a stable insulation space between a diver's body and dry suit, undergarments serve a number of other important functions such as reducing convective transport of heat by the inflation gas. For my purposes, these effects are ignored, and the only heat loss mechanism considered is through conduction by the composite insulator formed by the underwear and inflation gas. From this restricted view, the underwear maintains a physical space of thickness t  filled with gas of thermal conductivity K_GAS between the diver and suit. The underwear also has its own conductivity K_UNDERGARMENT, and conducts heat from the diver independently from the gas as a parallel loss mechanism.

The larger the ratio R, the less heat a diver will loose to surrounding cooler water, so our objective is to increase the resistance to heat loss. For a fixed K_UNDERGARMENT, R can be made larger by either increasing thickness t, or by decreasing the gas conductivity K_GAS. A diver can increase t  by wearing a combination of thicker underwear and more weight to compensate for the increased insulation volume. Nevertheless, there are comfort limits --the doughboy look is not conducive to efficient motion. For a particular set of undergarments and equipment weight, divers can best insulate themselves from heat loss by choosing a suit inflation gas with a small thermal conductivity.

Before looking up tables of gas conductivities, we can gain some idea of which gases should perform well by considering the microscopic origin of the numbers displayed in the tables. With this physical insight, we can predict the top candidates.

Simply stated, the thermal conductivity of a gas is a product of factors, which either aid or impede the flow of heat. Roughly, the thermal conductivity K_GAS increases with the specific heat C_V of the gas molecules, and decreases with the square root of the mass m and cross-sectional size s of the molecules. That is,

This equation will serve as a guide, with the squiggle implying mathematical form rather than exact equality. Our goal is minimize the conductivity K_GAS by finding a gas where we minimize C_V while simultaneously maximizing m and s. Details on how each of these three factors affect heat conduction are discussed on the page:  Molecules .

When the conductivities of different gases are compared,  the molecular specific heats and cross sections must be taken into account in addition to considering the masses of the molecules.  So, for example,  the effectiveness of argon as an insulator compared to air and helium mixtures is not simply due to argon's greater mass. If the conductivity of argon is compared with air (as in Table 1 below), the superior performance of argon is primarily due to its lower specific heat, rather than its greater molecular mass. On the other hand, argon's conductivity is much less than helium because of argon's greater mass and size --the two gases have the same specific heat.

You might think that gas conductivity K_GAS should increase with pressure because of the greater concentration of molecules available to carry heat energy in the dense gas. Although this seems reasonable, greater gas density also impedes the flow of heat by increasing the frequency of collisions between molecules. The random collisions scatter molecular motion away from the gradient of heat flow (from the warm diver to the cold water), canceling the density effects, and leaving only the residual proportionality constant as the factor of 1/s in K_GAS, related to the molecular cross section.

Candidate Gases

So, where does theory lead us? We found that the gas with the smallest K_GAS should simultaneously have low C_V along with large m and s . We can now see why hydrogen is the worst possible choice for an insulation gas: it has the lowest mass of all molecules, it is also a small, diatomic molecule giving it a large C_V and small s, adding up to three strikes. With the additional exploding-diver hazard, H₂ is definitely "out." Helium is only slightly better than hydrogen due to its smaller size, greater mass and lower specific heat. Moving to the other extreme, from a purely physical standpoint, the insulation gas of choice would be large, massive, monatomic radon (Rn). But radon also has the additional potential to warm the diver due to its "hot" radioactive nature so, we cannot rely on physics alone to guide the search --we need to be practical. Moving away from radon, the next two massive, large, monatomic gases are xenon and krypton, which have great thermal properties, but at $1000 per standard cubic foot (scf) cost too much. Argon (Ar) comes next in order of the massive monatomics and is obviously a reasonable choice, so we'll set it aside for further consideration. Another class of candidates is suggested by the large mass of uranium hexaflouride (UF₆), with K_GAS a close second to radon. Unfortunately, UF₆ shares radon's health disadvantages and raises certain state security issues. Reasoning in a similar vein to how we guessed Ar should have a low conductivity, an agreeable cousin to UF₆ is found in sulfur hexafluoride (SF₆), which has actually been used as a suit inflation gas by the US Navy. Under the category of miscellaneous candidates is carbon dioxide (CO₂), which has also been used in Navy tests, because of the ability of closed circuit UBA to scrub any residual CO₂ from the swimmer's breathing gas. From a physical standpoint, CO₂ is a reasonable choice because of its large size and mass, however, its C_V is large due to triatomic structure. For similar reasons, small, non-chlorinated freons such as Freon14 (CF₄) should perform well, so we can include them in a short-list of contenders for the optimal suit inflation gas: Ar, CO₂, SF₆, and CF₄. Sulfur hexafluoride and Freon 14 might be eliminated straight away based on cost (20 to 30 times the price of argon per scf), however, we'll keep them around for the sake of argument.

Table I displays the conductivities of some alternative gases as a percentage of the conductivity of air, with He and H₂ included as examples of poor choices. The best inflation gas choices all have ratios less than one (100%), representing lower conductivities, and an insulation improvement over air. The thermal conductivities of the candidate gases are all less than air --the inflation gas for which all dry suit divers have a "feel." The entry for all nitrox mixtures (from 0% to 100% O₂) is the same as air due to the near identical conductivities of N₂ and O₂. It should be noted that the thermal conductivity of trimix is not simply determined by the fractional conductivities of the individual gas components! The details are beyond the scope of this article, but mixed gas conductivity follows a nonlinear mixing rule.

Note from the equation for R above, we can trade off the thickness t with the total thermal conductivity (K_GAS + K_U.G.) while maintaining the same thermal resistance. If undergarment conductivity is neglected, then a diver can get the same amount of insulation from air as argon if they increase the thickness of their underwear by about 50% ( 1 / 0.67 = 1.48) to cancel the higher conductivity of air. Realistically, when undergarment and water vapor conductivity are considered, the difference in thickness is not this large. Other comparisons between each of the gases can also be made by taking the ratio of the tabulated conductivities, because the factors due to air will cancel out. As another example, argon has a small fraction of helium's conductivity as seen from the ratio: K_Ar / K_He = (67 / 583) = 12%.