doi:10.1016/j.chemgeo.2008.04.008
Copyright © 2008 Elsevier B.V. All rights reserved.
Experimental solidification of anhydrous latitic and trachytic melts at different cooling rates: The role of nucleation kinetics
References and further reading may be available for this article. To view references and further reading you must
purchase this article.
Gianluca Iezzia, b,
,
, Silvio Mollob, c, Guido Venturab, Andrea Cavallob and Claudia Romanoc
aDipartimento Scienze della Terra, Università, G.d'Annunzio, Chieti/Pescara, Italy
bInstituto Nazionale di Geofisica e Vulcanologia, Roma, Italy
cDipartimento di Scienze Geologiche, Università Roma Tre, Roma, Italy
Received 18 December 2007;
revised 7 April 2008;
accepted 9 April 2008.
Editor: D.B. Dingwell.
Available online 24 April 2008.
Abstract
Two sets of cooling experiments were run at atmospheric conditions for two anhydrous starting latitic and trachytic melts: 1) five cooling rates (25, 12.5, 3, 0.5, and 0.125 °C/min) between 1300° and 800 °C, and 2) a 0.5 °C/min cooling rate from 1300 °C with quench temperatures at 1200°, 1100°, 1000° and 900 °C. Trachytic run-products are invariably glassy. Nucleation is also suppressed in the latitic run-products at the three highest cooling rates. Conversely, in the 0.5 and 0.125 °C/min runs, latites have a crystal content of
90 vol.%. The phases are: plagioclase, clinopyroxene, glass and iron-bearing oxide (in order of abundance). The variable quench temperatures, investigated by coupling experiments with Pt wire and Pt capsule sample containers in set 2, again did not produce crystallization of trachyte, whereas latitic samples are characterized by
10 vol.% of oxides, pyroxenes and plagioclase (in order of appearance), at temperature < 1000 °C. Effects of (preferential) heterogeneous nucleation on sample holders, of superheating degree, and chemical species loss during cooling are absent for both melt compositions. The difference of solidification paths between these two silicate melts can be ascribed only to their small chemical differences. In comparison with calculated equilibrium conditions all the experimental latitic and trachytic run-products revealed strong kinetic effects, interpretable in the light of the nucleation theory. The glass-forming ability (GFA) of trachyte is higher, whereas their critical cooling rate (Rc) is lower (< 0.125 °C/min), in comparison to latitic melts (Rc > 0.5 °C/min). The experimental results carried out in this study can be applied to lava flows and domes; trachytic lavas are able to flow for longer period with respect to latitic ones in a metastable condition. Glass-rich terrestrial lavas, i.e. obsidians, can be the result of sluggish nucleation kinetics due to the relative high polymerisation of evolved silicate melts.
Keywords: Dynamic cooling experiments; Latite; Trachyte; Kinetic of nucleation; Viscosity
Fig. 1. Back-scattered SEM images of the first set of experiments, run as a function of the cooling rates (the applied cooling rate is reported on the SEM images); these experiments were run all using Pt wire. On the left column the five trachtic run-products, whereas on the right column the five latitic run-products. The lengths of the white scale bars are equal always to 50 μm.
 |
Fig. 2. a. Texture of sample AW-LAT-0.5. The original BSE-SEM images are reported on the top. The brightness and contrast were varied such to improve the different particular of crystal phase and glass. The major interpenetrated crystal phases correspond to plagioclases, whereas the prismatic to acicular crystal phases with the lighter grey are pyroxenes. The darker grey zones, mainly concentrated among the pyroxene crystals are interpreted as the residual glass melt. The black and white images correspond to the distribution of glass (centre) and pyroxenes (bottom), respectively. b. Texture of sample AW-LAT-0.125. Two original BSE-SEM images are reported on the top and centre. The original brightness and contrast parameters of these two BSE-SEM images were modified such to improve the textural features and particulars. The longer crystal phases correspond to plagioclases, whereas the (two) acicular crystal phases with the lighter grey colours are pyroxenes. Bottom image corresponds to the distribution of plagioclase crystals respect to both glass and pyroxene, respectively. All the scale bars are equal to 100 μm.
Fig. 3. Glass compositions of crystal-free trachyte and latite run-products solidified in the 1) set experiments. The full star and cross symbols refer to trachyte and latite samples, respectively. The same symbols, but empty, refer to the starting compositions. The dashed lines are eye-guide, joining starting and 0.125 °C/min produced glass compositions.
Fig. 4. Textural features of the run-products FW-LAT-0.5. See text for a complete description.
Fig. 5. Textures of the small- and plagioclase-free patch of FW-LAT-0.5 run-product. These two different images are particulars at higher magnifications of the SEM data reported in Fig. 4. See text for a complete description.
Fig. 6. Back-scattered electron images (top) and selected oxide distribution by EPMA map (see text) of the two Latitic run-products FW-LAT-0.5 (left column) and FC-LAT-0.5 (right column) EPMA maps.
Table 1.
Chemical composition of the run-products by EPMA analysis

The number reported in parentheses refers to the standard deviation calculated on at least 15–20 EPMA point analyses, for which the total mass percent were above 99%.
a calculated all as ferric iron.
b calculated all as ferrous iron.
c These EPMA analyses were obtained only on the larger feldspar crystals; feldspars in the groundmass, pyroxenes and matrix glasses were too small to be analyzed by EPMA.
Table 2a.
Experiments of set 1) in the thermal range 1300–800 °C

Table 2b.
Experiments of set 2) with fixed cooling rate at .5 °C/min


Corresponding author. Now at Dipartimento di Geotecnologie per l'Ambiente ed il Territorio (DiGAT), Università, G.d'Annunzio, Chieti/Pescara, Via DEI VESTINI 30, 66013 Chieti-Scalo, Italy. Tel.: +39 0871 3556147; fax: +39 0871 3556047.