The imprint of thermally induced devolatilization phenomena on radon signal: implications for the geochemical survey in volcanic areas

Abstract

S U M M A R YThermal gradients due to magma dynamics in active volcanic areas may affect the emanating power of the substrate and the background level of radon signal. This is particularly effective in subvolcanic substrates where intense hydrothermal alteration and/or weathering processes generally form hydrous minerals, such as zeolites able to store and release great amounts of H 2 O (up to ∼25 wt.%) at relative low temperatures. To better understand the role played by thermally induced devolatilization reactions on the radon signal, a new experimental setup has been developed for measuring in real time the radon emission from a zeolitized volcanic tuff. Progressive dehydration phenomena with increasing temperature produce radon emissions two orders of magnitude higher than those measured during rock deformation, microfracturing and failure. In this framework, mineral devolatilization reactions can contribute significantly to produce radon emissions spatially heterogeneous and non-stationary in time, resulting in a transient state dictated by temperature gradients and the carrier effects of subsurface gases. Results from these experiments can be extrapolated to the temporal and spatial scales of magmatic processes, where the ascent of small magma batches from depth causes volatile release due to dehydration phenomena that increase the radon signal from the degassing host rock material.

Transient to stationary radon ( 220 Rn) emissions from a phonolitic rock exposed to subvolcanic temperatures

“…Coherently, EXP1 shows that 220 Rn monotonically increases by about two orders of magnitude as the temperature increases from 100 to 900°C (figure 4 b ). This behaviour corroborates the strong dependence of Rn diffusion on temperature, leading to transient Rn emissions due to a new value of equilibrium activity concentration [29,59,60]. The change in Rn emission rate ( E in arbitrary units) is positively correlated with the temperature-dependent diffusion coefficient of Rn by a nonlinear Arrhenius equation (where T is expressed in Kelvin; [61–63].…”

“…Rn gas emitted from the rock samples was measured through an Rn monitor (RAD7, Durridge Company Inc., USA) connected to the furnace in a closed-loop configuration and equipped with a solid-state silicon alpha detector [29]. The most important characteristics of this novel set-up can be summarized as follows: the cylindrical rock sample is suspended in a gas-impermeable alumina tube of the vertical furnace;the rock sample is located in the isothermal hot zone of the furnace and the temperature is monitored on the rock surface by a Type S thermocouple with an uncertainty of ±3°C;gas leakage is prevented by upper and lower gas-tight flanges sealing the alumina tube;a Teflon tube inserted into the upper flange connects, in a closed-loop configuration, the alumina tube to a gas-drying unit filled with a desiccant (CaSO 4 ) to the RAD7;a recirculating pump moves the gas from the alumina tube towards the RAD7 solid-state detector for alpha counting;the gas is recirculated continuously through the system and ambient air is excluded from the experimental set-up during the run;the electrostatic detector collects the charged ions and discriminates the electrical pulses generated by the alpha particles of 216 Po for rapid determination of 220 Rn;the activity concentration of 220 Rn is measured continuously during experiments with an acquisition time (i.e.…”

“…The change in Rn emission rate ( E in arbitrary units) is positively correlated with the temperature-dependent diffusion coefficient of Rn by a nonlinear Arrhenius equation (where T is expressed in Kelvin; [61–63]. From an empirical point of view, the plot ln E versus 1/ T (figure 4 c ) provides a quantitative description of the temperature-dependent migration of Rn atoms through the microcracks of the rock sample (see [29] for a detailed mathematical treatment). The regression fit of experimental data from this study (figure 4 c ) yields a high correlation coefficient ( R 2 = 0.97), attesting to the close relationship between the enhancing effect of temperature and the escape of Rn atoms from the interconnected fractures of the rock sample.…”

“…The rock sample was first kept at 100°C for 48 h. Then, through a heating ramp (Δ t ) of 5 min, the temperature (Δ T ) was increased by 100°C and maintained constant for 22 h. This incremental step was replicated eight times (5 min Δ t and 100°C Δ T ) up to the maximum experimental temperature of 900°C. See Mollo et al [29] for a detailed mathematical treatment of the nonlinear Arrhenius equation describing the dependence of Rn emission rate on temperature.…”

Continuous radon monitoring during seven years of volcanic unrest at Campi Flegrei caldera (Italy)

“…Also, for this reason, we think that a general increase of the temperature and the pressure of the Campi Flegrei hydrothermal system 50,59 is the central parameter explaining the correlation among the different independent variables. The temperature increase may induce thermal devolatilization phenomena on radon 60 and the fluid pressure increase favours deformation, earthquakes and the flux of the hydrothermal gas from depth to the surface. For example, it was measured by various authors a strong increase in the hydrothermal-volcanic CO 2 both diffusively emitted at Solfatara (from 900-1000 t/d in 2010-2011 to 1500-3000 t/d in 2015-2016 43 ) and emitted by Pisciarelli vent (from 150-200 t/d in 2012-2013 to 400-600 t/d in 2017-2019 44 , Fig.…”