Differential rates of feldspar weathering in granitic regoliths

Citation
Af. White et al., Differential rates of feldspar weathering in granitic regoliths, GEOCH COS A, 65(6), 2001, pp. 847-869
Citations number
90
Categorie Soggetti
Earth Sciences
Journal title
GEOCHIMICA ET COSMOCHIMICA ACTA
ISSN journal
00167037 → ACNP
Volume
65
Issue
6
Year of publication
2001
Pages
847 - 869
Database
ISI
SICI code
0016-7037(200103)65:6<847:DROFWI>2.0.ZU;2-N
Abstract
Differential rates of plagioclase and K-feldspar weathering commonly observ ed in bedrock and soil environments are examined in terms of chemical kinet ic and solubility controls and hydrologic permeability. For the Panola rego lith, in the Georgia Piedmont Province of southeastern United States, petro graphic observations. coupled with elemental balances and Sr-87/Sr-86 ratio s. indicate that plagioclase is bring converted to kaolinite at depths > 6 m in the granitic bedrock. K-feldspar remains pristine in the bedrock but s ubsequently weathers to kaolinite at the overlying saprolite. In contrast. both plagioclase and K-feldspar remain stable in granitic bedrocks elsewher e in Piedmont Province such as Davis Run, Virginia, where feldspars weather concurrently in an overlying thick saprolite sequence. Kinetic rate consta nts, mineral surface areas, and secondary hydraulic conductivities are fitt ed to feldspar losses with depth in the Panola and Davis Run regoliths usin g a time-depth computer spreadsheet model. The primary hydraulic conductivi ties, describing the rates of meteoric water penetration into the pristine granites, are assumed to be equal to the propagation rates of weathering fr onts, which, based on cosmogenic isotope dating. are 7 m/10(6) yr for the P anola regolith and 4 m/10(6) yr for the Davis Run regolith. Best fits in th e calculations indicate that the kinetic rate constants for plagioclase in both regoliths art: factors of two to three times faster than K-feldspar, w hich is in agreement with experimental findings. However. the range for pla gioclase and K-feldspar rates (k(r) = 1.5 X 10(-17) to 2.8 X 10(-16) mol m (2) s(-1)) is three to four orders of magnitude lower than for that for exp erimental feldspar dissolution rates and are among the slowest yet recorded for natural feldspar weathering. Such slow rates are attributed to the rel atively old geomorphic ages of the Panola and Davis Run regoliths. implying that mineral surface reactivity decreases significantly with time. Differential feldspar weathering in the low-permeability Panola bedrock env ironment is more dependent on relative feldspar solubilities than on differ ences in kinetic reaction rates. Such weathering is very sensitive to prima ry and secondary hydraulic conductivities (q(p) and q(s)). which control bo th the fluid volumes passing through the regolith and the thermodynamic sat uration of the feldspars. Bedrock permeability is primarily intragranular a nd is created by internal weathering of networks of interconnected plagiocl ase phenocrysts. Saprolite permeability is principally intergranular and is the result of dissolution of silicate phases during isovolumetric weatheri ng, A secondary to primary hydraulic conductivity ratio of q(s)/q(p) = 150 in the Panola bedrock results in kinetically controlled plagioclase dissolu tion but thermodynamically inhibited K-feldspar reaction. This result is in accord with calculated chemical saturation states for groundwater sampled in the Panola Granite. In contrast, greater secondary conductivities in the Davis Run saprolite, q(s)/q(p) = 800, produces both kinetically controlled plagioclase and K-feldspar dissolution. Faster plagioclase reaction, leadi ng to bedrock weathering in the Panola Granite but not at Davis Run. is att ributed to a higher anorthite component of the plagioclase and a wetter and warmer climate, In addition, the Panola Granite has an abnormally high con tent of disseminated calcite, the dissolution of which precedes the plagioc lase weathering front, thus creating additional secondary permeability. Cop yright (C) 2001 Elsevier Science Ltd.