Cooling rates of hyaloclastites: applications of relaxation geospeedometry to undersea volcanic deposits

Abstract

Glass fragments from three different hyaloclastites have been used to evaluate the range of cooling rates experienced by undersea volcanic deposits. We found that the glass fragments retain structures with a range of apparent quench rates from 25 to 0.15 K min -1 . The most rapid cooling rates are interpreted to be those resulting from cooling of the lava near the water interface. Simple conductive cooling models produce a range of quench rates comparable to those of the more rapidly cooled samples. The very slow apparent quench rates are unlikely to result from simple linear cooling through the glass transition, because of the onset of crystallization; instead, they are indicators of a more complex thermal history that involves the annealing of glasses at temperatures within the glass transition interval for a dwell time sufficient to allow the relaxation of the glass to lower temperature structures. The thermal history recorded in these samples illustrates the complexity of eruptive processes and demonstrates that quench rates for natural glasses retain information relevant to more complex cooling models.

No depth limit to hydrovolcanic limu o Pele: analysis of limu from Lō`ihi Seamount, Hawai`i

“…Plastic growth of a limu bubble requires that the melt be kept above the glass transition long enough for the limu bubble to expand or buoyantly escape the melt. Differential scanning calorimetric (DSC) studies on the cooling rates (q) of submarine hyaloclastites have returned cooling rates as low as 0.003 K s −1 (Wilding et al 2000), and limu o Pele from Lō`ihi have been shown to be "hyperquenched," with extremely high cooling rates of up to 10 5.31 K/s (Potuzak et al 2008). While the slow end of the range would provide ample time (>> several seconds) for the expansion or escape of a limu bubble, the "hyperquenched" signature suggests that only~1.5×10 −3 s is available to stretch melt into a limu bubble, calculated by (T 2 -T 1 )/q, where T 2 and T 1 are the magmatic (~1,100°C) and approximate glass transition (~800°C) temperatures, respectively.…”

Cooling rates of basaltic hyaloclastites and pillow lava glasses from the HSDP2 drill core

“…However, as discussed above, basaltic melts are more fragile glass formers and are thus more susceptible to structural degradation above the glass transition (Angell, 1985). Wilding et al (2000) compared basaltic hyaloclastite glass before and after calorimetry experiments and found that the thermal treatment did not alter the composition of the glass. However, they found that for some samples the c p of the liquid field was difficult to resolve as a result of decreasing c p owing to the release of latent heat of crystallisation.…”

“…Most previous studies (Wilding et al, 1995(Wilding et al, , 1996(Wilding et al, , 2000(Wilding et al, , 2004Gottsmann and Dingwell, 2001a, 2001bGottsmann et al, 2004) that have calculated the cooling rates of volcanic glass from calorimetry data have used the T-N geospeedometer (Tool, 1946;Narayanaswamy, 1971). This relates c p , or the derivative of enthalpy (DH/DT), to changes in fictive temperature (DT f /DT).…”

“…Wilding et al (2000) determined rates of 0.003-0.4 K s À1 for submarine emplaced basaltic hyaloclastites from the East Pacific Rise, while Gottsmann et al (2004) calculated 0.1-2.3 K s À1 for the air-quenched crusts of pahoehoe lava flows generated during the Pu'u 'O' o-Kupaianaha eruption on Hawaii both with the T-N geospeedometer. Finally, most recently, Potuzak et al (2008) showed that for bubble wall fragments (limu o Pele) and glassy shards generated during submarine eruptions at Loihi seamount, Hawaii, the procedure developed by Yue et al (2002) to determine the cooling rates of hyperquenched synthetic glasses had to be applied to obtain cooling rates from the calorimetry data.…”

The thermal history of a spatter-fed lava flow: the 8-ka pantellerite flow of Mayor Island, New Zealand

“…). This technique involves the evaluation of the relaxation process of enthalpy as a geospeedometer using differential scanning calorimetry and has been recently applied to several volcanological/geological problems (Wilding et al 1995(Wilding et al , 1996a(Wilding et al , 1996b(Wilding et al , 2000Gottsmann and Dingwell 2001b) including the analysis of clastogenic phonolite flows on Tenerife, Canary Islands (Gottsmann and Dingwell 2001a).…”