Validation of the effect of helox on thermal conductance in homeotherms using heated models

Abstract

It is commonly stated that mixtures of 80% helium and 20% oxygen (helox) increase thermal conductance, and hence heat flux, from homeotherms by roughly a factor of two. However, because helox affects heat loss by conduction and convection differently, its effect on heat flux should vary according to the relative contributions of these two routes to the overall heat flux. We used heated models made of three sizes of copper tubing and covered with various grades of synthetic fur to measure heat flux under air and helox atmospheres. Thermal conductance in helox (C_helox; W•m⁻² •°C⁻¹) was highly correlated with thermal conductance in air (C_air) and could be predicted by C_helox=1.704+1.726 C_air. The relative increase in heat flux under a helox atmosphere compared with that in air (conductance ratio; C_helox/C_air) varied from a maximum of 2.5 when thermal conductance is infinitely small and heat flux is predominantly determined by conduction to a minimum of 1.85 when animals are naked and heat flux is dominated by convection. These values follow the relation: C_helox/C_air=2.503−0.081 C_air. Body size has no significant effect on the conductance ratio.

Introduction

For over 40 years, mixtures of 80% helium and 20% oxygen (hereafter termed helox) have been used to increase the rate of heat loss in homeotherms, usually with the objective of determining the maximum rate of oxygen consumption and metabolism (e.g. Cook et al., 1951; Rosenman and Morrison, 1974; Smith and Dawson, 1985; Chappell et al., 1995; but also see Grapé et al., 1960). Because helox has a coefficient of thermal transfer by conduction roughly four times greater than that of air (Lindsay and Bromley, 1950), replacing the air trapped in the fur or feather matrix with a helox atmosphere acts to increase the rate that heat is conducted through this insulation layer. Helox also has a coefficient of thermal transfer by convection that is approximately 2.1 times that of air (Epperson et al., 1966), so the rate of bulk heat transfer away from the outer boundary of the fur or feather layer is also increased. Although it is commonly stated that helox increases heat loss by a factor of two (e.g. Smith and Dawson, 1985; Geiser et al., 1996), thus implying that the effect of helox is constant, this is technically not correct. Because helox affects conduction more than convection, the effect of helox should not be constant, but rather should vary according to the relative importance of these two modes in limiting the overall heat flux. Thus, Rosenman and Morrison (1974)noted that the ratio of thermal conductance in helox to that in air was positively correlated with insulation (the inverse of thermal conductance in air).

Despite our understanding of the fundamental processes affected by helox, two problems remain. First, if conduction and convection were the only two routes of heat flux in animals then one would expect helox to increase thermal conductance by a factor ranging from roughly 2 to 4 times that in air. The fact that all reported values lie below 2.6 (Rosenman and Morrison, 1974) indicates that either animals invoke additional compensatory mechanisms to reduce heat loss when challenged by helox atmospheres or that additional routes are involved, notably losses through radiation and evaporation. However, because evaporative heat loss is rarely measured or reported, it is currently impossible to predict the effect of helox on dry thermal conductance. Second, the data presented in Figure 8 of Rosenman and Morrison (1974)indicate considerable inter-specific variation in the effect of helox for a given insulation value. Although they arbitrarily divided their data into two groups representing small and large species and fitted two separate curves to the data, there is no a priori reason to separate the data. The scatter in their data could be attributed to a true size effect as they suggest, whereby large animals offer greater tissue resistance allowing them to depress surface temperatures. Alternatively, the scatter may be attributed to inter-specific variation in evaporative heat loss or simply to intra- or inter-individual variation inherent in biological systems.

Both of the problems stated above indicate the need for a ‘null model’ whereby the effect of helox on the physics of heat transfer through an insulating barrier is calibrated independently of the voluntary and involuntary physiological responses available to living biological systems and affecting evaporative heat loss, tissue resistance and surface temperature. Such a ‘null model’ would allow us to partition the heat flux of live animals between processes determined by the physical effects of helox and those determined by physiological adjustments. Thus, the purpose of this study was to quantify the effect of helox on thermal conductance as a function of insulation using non-biological heated copper models. This allowed us to eliminate the potentially confounding effects of evaporation and tissue resistance.