Chemical studies of L chondrites—II. Shock-induced trace element mobilization

Previous studies of chondrites heated in the laboratory for extended periods under conditions approximating those in shock-heated collisional debris indicate that Au, Co, Se, Ga, Rb, Cs, Te, Bi, In, Ag, Zn, Tl and Cd progress in mobility. We report data for these 13 trace elements in 14 L4–6 chondrites of established shock history and discuss these and 13 additional chondrites studied earlier. Trace element contents vary with petrologic type, $\text{S}\text{Fe}$ sub-group and shock history, the last dominating strongly. Absolute abundances and interelement relationships for the 6 or 7 most mobile elements vary with degree of shock-loading (i.e. residual temperatures) established from mineralogic/petrologic study. A tertiary process, shock-heating, previously known to have affected radiogenic ⁴⁰Ar and/or ⁴He in meteorites but not other elements, apparently was at least as effective as other open-system processes (secondary [parent body] and primary [nebular and/or accretionary] episodes) in establishing mobile trace element contents of L chondrites and probably others. If conditions during early genetic episodes are to be deduced from compositional information, shocked meteorites should be avoided or effects of later processes should be compensated for.     

Trace elements in primitive meteorites—V. Abundance patterns of thirteen trace elements and interelement relationships in enstatite chondrites

Neutron activation analysis was used to determine As, Au, Bi, Cd, Co, Cu, Ga, In, Sb, Se, Te, Tl and Zn in 11 samples representing 9 chondrites of grades E4–6. These chondrites exhibit systematic intra- and inter-grade differences particularly for highly-variable elements, the differences being E4 ? E3 > E6 ? E5. The abundance pattern for these 13 and an additional 16 elements in E3-6 chondrites differs from those of other primitive meteorites—the carbonaceous and unequilibrated ordinary chondrites. A search for statistically-significant interelement relationships among the 13 elements (for grades E4–6) reveal that 40 elementpairs are linearly and/or exponentially correlated. Similar consideration of data for 37 elements in 12 chondrites (grades E3–6) reveals that 191 element-pairs exhibit such relationships, 170 involving linear and/or exponential correlations, the remainder involving anti-correlations. The patterns depicting these relationships—i.e. the correlation profiles—and elemental abundance patterns, factor analysis and two-element correlation diagrams are consistent with all enstatite chondrites representing a single evolutionary sequence. The primary process responsible for the chemical trends of these chondrites involved thermal fractionation accompanied by geochemical fractionation of sulfide-plus-metal from silicate, probably during condensation and accretion of solid material from the solar nebula. Chalcophile elements may have been fractionated during condensation or, after accretion, during thermal metamorphism in the parent body. No genetic model proposed thus far accounts for the detailed chemical trends, although the constrained equilibrium theory and two-component condensation theories qualitatively seem most satisfactory. The correlation profiles of enstatite, carbonaceous and unequilibrated ordinary chondrites are distinctly different, pointing to major differences in the formation conditions of these different sorts of primitive meteorites.     

Further studies of trace elements in C3 chondrites

Five carbonaceous chondrites (Renazzo C2V, Allende C3V, Omans C3O, Warrenton C3O, and Orgueil Cl) were analyzed by radiochemical neutron activation analysis for Ag, Au, Bi, Br, Cd, Cs. Ge, In, Ir, Ni, Os, Pd, Rb, Re, Sb, Se, Te, Tl, U and Zn. These data, together with earlier measurements on seven additional C3 s, are interpreted in the light of pétrographie studies by MCSWEEN (1977a, b) and revised condensation temperatures (WAI and Wasson, 1977). Elements condensing between ~ 700 and 420 K (Se, Zn, S, Te, Br, In, Bi, Tl) are systematically more depleted than those condensing between 1000 and 900 K (Ge, Ag, Rb), by factors of 1.3 to 2, and the depletion correlates inversely with matrix content and directly with degree of metamorphism. The most plausible explanation appears to be a gas-dust fractionation during condensation, by settling of dust to the median plane of the nebula. In this model, gas/dust ratios relative to the cosmic ratio ranged from 0.7 at 1000 K to 0.5 at 700 K for those C3O s that accreted first (Ornans, Warrenton) and from 1.3 to 0.6 for the last (Kainsaz). There appears to have been no further gas/dust fractionation below 700 K.Abundances of Sb, Au and Cd follow earlier trends. Depletion of Sb and Au correlates with abundance of Fe-poor olivine and seems to reflect greater volatilization upon more prolonged or intense heating during chondrule formation. The 50–100-fold depletion of Cd in C3Os compared to C3Vs suggests condensation in a region where enough Fe was present to buffer the H₂S pressure.     

Chemical and physical studies of type 3 chondrites—V: The enstatite chondrites

We report instrumental neutron activation analysis determinations of 19 major, minor and trace elements in three enstatite chondrites. Based on these, and literature data on the bulk and mineral composition of enstatite chondrites, we discuss the history of the type 3 or unequilibrated enstatite chondrites, and their relationship with the other enstatite chondrites. The type 3 enstatite chondrites have E chondrite lithophile element abundances and their siderophile element abundances place them with the EH chondrites, well resolved from the EL chondrites. Moderately volatile chalcophile elements are at the low end of the EH range and Cr appears to be intermediate between EH and EL. We suggest that the type 3 enstatite chondrites are EH chondrites which have suffered small depletions of certain chalcophile elements through the loss of shock-produced sulfurous liquids. The oxygen isotope differences between type 3 and other enstatite chondrites is consistent with equilibration with the nebula gas ~30° higher than the others, or with the loss of a plagioclase-rich liquid. The mineral chemistry of the type 3 chondrites is consistent with either low temperature equilibration, or, in some instances, with shock effects.     

Abundance patterns of thirteen trace elements in primitive carbonaceous and unequilibrated ordinary chondrites

A neutron activation analysis technique was used to determine Au, Re, Co, Mo, As, Sb, Ga, Se, Te, Hg, Zn, Bi and Tl in 11 carbonaceous chondrites, 12 unequilibrated ordinary chondrites (UOC), and 4 equilibrated ordinary chondrites. The first 6 elements are ‘undepleted’, the next 3 ‘normally-depleted’ and the last 4 ‘strongly-depleted’. Except for Hg, ‘depleted-element’ abundances in carbonaceous chondrites lead to mean relative ratios of C1:C2:C3 = 1.00:0.53:0.29, i.e. those predicted by a two-component (mixing of high-temperature and low-temperature fractions) model. The last 4 nominally ‘undepleted’ elements are somewhat depleted in ordinary chondrites, As and Sb showing partial depletion in C3 and the latter in C2 chondrites as well. This requires a modification of the two-component model to indicate that deposition of elements during condensation of high temperature material was not an all-or-nothing process.Apart from Bi and Tl, the elements studied have similar abundances in unequilibrated and equilibrated ordinary chondrites and only the former are unquestionably correlated with the degree of disequilibrium in silicate minerals. Only some ‘strongly-depleted’ elements exhibit at least one of the following—proportional depletion in UOC, progressive depletion in petrographic grades 3–6 ordinary chondrites and enrichment in the gas-containing dark portion of gas-rich, light-dark meteorites—indicating that such depletion does not ensure that an element will exhibit these trends. Partly or completely siderophile As, Au, Co, Ga, Mo, Re and Sb vary with chemical type in the same manner in both unequilibrated and equilibrated ordinary chondrites and doubtless reflect a process involving fractionation of metallic iron.     

Trace elements in primitive meteorites—VI. Abundance patterns of thirteen trace elements and interelement relationships in unequilibrated ordinary chondrites

Neutron activation analysis was used to determine As, Au, Bi, Cd, Co, Cu, Ga, In, Sb, Se, Te, Tl and Zn in 13 different unequilibrated ordinary chondrites (UOC), i.e. those having chemicallyinhomogeneous silicates. This study together with prior data completes our coverage of this group of 23 primitive chondrites. Four elements are quite variable in UOC (Cd—20 x, In—30 x, Bi—300 x and Tl—1300 x), the others varying by 2–8 x. Three highly-depleted elements—Bi, In and Tl—are richer by 5–35 x in unequilibrated chondrites than in their equilibrated congeners. All 3 elements vary directly in characteristic fashion with disequilibrium parameters for olivine and pyroxene in UOC and generally with petrologic type 3 > 4 > 5 > 6. The data do not provide unambiguous evidence for nebular fractionation of siderophile elements. Examination of statistically-significant interelement relationships among various ordinary chondrite populations involving 34 elements reveals patterns distinct from those of other chondritic groups. These patterns reflect nebular metal-silicate fractionation which preceded or accompanied thermal fractionation. The results point to significant differences in the formation of primitive carbonaceous, enstatite and ordinary chondrites.     

Spinel group minerals in LL3.00-6 chondrites: Indicators of nebular and parent body processes

We carried out a systematic study of spinel group minerals in LL3.00-3.9 and LL4-6 chondrites. With increasing petrologic type, the size and abundance of spinel increase. The compositions of spinel group minerals in type 3 chondrites depend on the occurrence; Mg-Al-rich spinel occurs mainly in chondrules. Some chromite occurs in chondrules and matrix, and nearly pure chromite is exclusively encountered in the matrix. The occurrence of nearly pure chromite and the wide compositional variations distinguish spinel group minerals in types 3.00-3.3 from those in the other types. Spinel group minerals in types 3.5-3.9 show a narrower range of compositions, and those in types 4-6 are homogeneous. The changes in composition and abundance of spinel in type 3 chondrites are most likely due to thermal metamorphism. Therefore, the chemistry of spinel group minerals could be used as a sensitive indicator of metamorphic conditions, not only for type 3-6, but also 3.00-3.9. They can be applied to identify the most primitive (least metamorphosed) chondrites. The bulk compositions of spinel-bearing chondrules and the textural setting of the spinel indicate that most spinel group minerals crystallized directly from chondrule melts. However, some spinel grains, especially those enclosed in olivine phenocrysts, can not be explained by in situ crystallization in the chondrule. We interpret these spinel grains to be relic phases that survived chondrule melting. This is supported by the oxygen isotopic composition of a spinel grain, which has significantly lighter oxygen than the coexisting olivine. The oxygen isotopic composition of this spinel is similar to those of Al-rich chondrules. Our discovery of relic spinel in chondrules is an indication of the complexities in the early solar nebular processes that ranged from formation of refractory inclusion, through Al-rich chondrule, to ferromagnesian chondrules, and attests to the recycling of earlier formed materials into the precursors of later formed materials. The characteristic features of spinel group minerals are not only sensitive to thermal metamorphism, but also shed light on chondrule formation processes.     

Chemical and physical studies of type 3 chondrites—VI: Siderophile elements in ordinary chondrites

The abundances of Fe, Ni, Co, Au, Ir, Ga, As and Mg have been determined by instrumental neutron activation analysis in 38 type 3 ordinary chondrites (10 of which may be paired) and 15 equilibrated chondrites. Classification of type 3 ordinary chondrites into the H, L and LL classes using oxygen isotopes and parameters which reflect oxidation state (Fa and Fs in the olivine and pyroxene and Co in kamacite) is difficult or impossible. Bulk compositional parameters, based on the equilibrated chondrites, have therefore been used to classify the type 3 chondrites. The distribution of the type 3 ordinary chondrites over the classes is very different from that of the equilibrated chondrites, the LL chondrites being more heavily represented. The type 3 ordinary chondrites contain 5 to 15 percent lower abundances of siderophile elements and a compilation of the present data and literature data indicates a small, systematic decrease in siderophile element concentration with decreasing petrologic type. The type 3 ordinary chondrites have, like the equilibrated ordinary chondrites, suffered a fractionation of their siderophile elements, but the loss of Ni in comparison with Au and Ir is greater for the type 3 chondrites. These siderophile element trends were established at the nebula phase of chondritic history and the co-variation with petrologic type implies onion-shell structures for the ordinary chondrite parent bodies. It is also clear that the relationship between the type 3 and the equilibrated ordinary chondrites involves more than simple, closed-system metamorphism.     

Chemical studies of H chondrites. II: Weathering effects in the Victoria Land,Antarctic population and comparison of two Antarctic populations with non-Antarctic falls

We report RNAA data for 14 siderophile, lithophile and chalcophile volatile/mobile trace elements (Ag, Au, Bi, Cd, Co, Cs, Ga, In, Sb, Se, Te, Tl, U, Zn) in interior portions of 45 different H4–6 chondrites (49 samples) from Victoria Land, Antarctica and 5 H5 chondrites from the Yamato Mts., Antarctica.Relative to H5 chondrites of weathering types A and B, all elements are depleted (10 at statistically significant levels) in extensively weathered (types B/C and C) samples, probably by leaching on the ice sheet surface. Contents of 8 elements in extensively weathered samples may provide a qualitative ranking for terrestrial surface residence. Chondrites of weathering types A and B seem compositionally uncompromised and as useful as contemporary falls for trace-element studies. When data distributions for these 14 trace elements in non-Antarctic H chondrite falls and unpaired samples from Victoria Land and from the Yamato Mts. (Queen Maud Land) are compared statistically, numerous significant differences are apparent. Concentrations of 8 elements differ significantly in the Victoria Land and non-Antarctic H5 chondrite populations. Essentially the same compositional differences and ⁵³Mn content and shock history differences appear when H4–6 populations are compared. Contents of 8 elements differ when Queen Maud Land and Victoria Land populations are compared and 5 for the Queen Maud Land/non-Antarctic comparison.These and other differences give ample cause to doubt that the various sample populations derive from the same parent population. The observed differences do not reflect weathering, chance or other trivial causes: a preterrestrial source must be responsible. The least unlikely of these involves a temporal variation in the source regions from which meteoroids derive.     

Kinetically induced compositional zoning in titanite: implications for accessory-phase/melt partitioning of trace elements

Usually it is assumed that the partitioning of trace elements into titanite in metaluminous granitoid plutonic environments takes place under equilibrium conditions and that compositional zoning is due solely to progressive changes in melt chemistry and/or mineral/melt partition coefficients. Examination of titanites from a variety of Caledonian metaluminous granitoids and related rocks has revealed that sector zoning is present, indicating disequilibrium partitioning. The sector zoning in titanites is defined principally by the distribution of the rare earth elements (REE), Y, Nb, Al and Fe. The REE, Y and Nb preferentially occur within the minor (100) sectors relative to the morphologically important (111) sectors. The reverse is true of Al and Fe which preferentially occur within the (111) sectors relative to the (100) sectors. The patterns of sector zoning are complicated by the fact that the relative growth rates of the various crystal faces fluctuated during growth. Sector zoning indicates that crystal-interface kinetics are responsible for the observed patterns of element partitioning. It is concluded that differences in the lateral-layerspreading rates of crystal faces bring about the sector zoning. The results have implications for the use of trace element partition coefficients in the modelling of fractionation processes.     

Diffusion of trace elements in FeNi metal: Application to zoned metal grains in chondrites

We have measured diffusion coefficients for P, Cr, Co, Ni, Cu, Ga, Ge, Ru, Pd, Ir, and Au in Fe metal from 1150 to 1400°C and at 1 bar and 10 kbar. Diffusion couples were prepared from high-purity Fe metal and metal from the IIA iron meteorite Coahuila (single crystal kamacite) or the pallasite Springwater (polycrystalline kamacite) and held at run conditions for 3.5 to 123 h. Diffusion profiles were measured using laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS) or the electron microprobe. Many elements were measured from the same experimental runs so interelemental comparisons are improved over other data sets in which data for different elements come from different experiments. Some literature diffusion coefficients (D) for Ni and Co in taenite can be up to a factor of 3 higher for Ni than Co, yet our results show no difference (e.g., D_Ni and D_Co ∼ 2.2 × 10^-15 m²/s at 1150°C). Thus, diffusion of Ni and Co in single crystal taenite will not measurably fractionate the Ni/Co ratio. On the other hand, the large difference in D_Ni and D_Ir (D_Ir is ∼5 times lower) and the similarity of D_Ni and D_Ru at all temperatures investigated indicates that Ni/Ir and Ni/Ru ratios in zoned metal grains will be useful discriminators of processes controlled by diffusion vs. volatility. In zoned metal grains in primitive chondrites, deviations of the Ni/Ru and Ni/Ir ratios from a condensation curve are opposite to a diffusion-controlled process, but consistent with a volatility-controlled process. The new multielement diffusion coefficients will also be useful in evaluating a variety of other processes in planetary science.     

Post-shock annealing of Miller Range 99301 (LL6): Implications for impact heating of ordinary chondrites

MIL 99301 is an LL chondrite that has experienced successive episodes of thermal metamorphism, shock metamorphism and annealing. The first recognizable petrogenetic episode resulted in thermal metamorphism of the rock to petrologic type 6 (as indicated by homogeneous olivine compositions, significant textural recrystallization, and the presence of coarse grains of plagioclase, metallic Fe-Ni and troilite). The source of heat for this thermal episode is not identified. The rock also experienced shock metamorphism to shock stage ∼S4 as indicated by extensive silicate darkening (caused by the dispersion within silicate grains of thin chromite melt veins and trails of metallic Fe-Ni and troilite blebs), polycrystalline troilite, myrmekitic plessite, a relatively high occurrence abundance (OA) of metallic Cu (3.6), the presence of numerous chromite-plagioclase assemblages, and coarse grains of low-Ca clinopyroxene with polysynthetic twinning. The shock event responsible for these effects must have occurred after the epoch of thermal metamorphism to type-6 levels; otherwise the polycrystallinity of the troilite would have disappeared and the low-Ca clinopyroxene would have transformed into orthopyroxene. Despite abundant evidence of strong shock, olivine and plagioclase in MIL 99301 exhibit sharp optical extinction, consistent with shock stage S1 and characteristic of an unshocked rock. This implies that an episode of post-shock annealing healed the damaged olivine and plagioclase crystal lattices and thereby changed undulose extinction into sharp extinction. The rock was probably annealed to metamorphic levels approximating petrologic type 4; more significant heating would have transformed the low-Ca clinopyroxene into orthopyroxene. It is not plausible that an episode of annealing occurring after the epoch of thermal metamorphism could have been caused by the decay of ²⁶Al because this isotope would have decayed away by that time. Impact heating is a more plausible source of post-metamorphic annealing of rocks in the vicinity of impact craters on low-density, high-porosity asteroids with rubble-pile structures.     

The compositional classification of chondrites: IV. Ungrouped chondritic meteorites and clasts

Six specimens of unusual chondritic materials were analyzed by neutron activation for 30 elements in order to assess their degree of chondritic compositional pristinity and to search for evidence of genetic links to other chondrites. Five have highly recrystallized textures; the other, the Cumberland Falls chondrite, has suffered minor metamorphic recrystallization. Acapulco and Allan Hills A77081, are closely related and have subpristine compositions; they are more distantly related to Enon which has an altered composition. Udei Station appears to be a IAB meteorite even though its $\text{FeO}\text{(FeO + MgO)}$ ratio is slightly above the IAB field. The highly weathered meteorite Tierra Bianca is closely related to IAB but has a δ¹⁸O value 5 standard deviations higher than the IAB mean and is designated ungrouped. Udei Station and Tierra Bianca have altered compositions; rare earth element patterns indicate loss of a phosphate phase. The elemental composition of the Cumberland Falls chondrite is virtually identical to that of LL chondrites and its O-isotope composition is closely similar to those of some unequilibrated ordinary chondrites including LL Semarkona. The $\text{FeO}\text{(FeO + MgO)}$ ratios in its olivine are generally much lower than those in pyroxene, a relationship we show to be indicative of in situ reduction resulting from exchange with the aubritic host. The names winonaites and forsterite chondrites have no taxonomic utility.     

Chemical and physical studies of type 3 chondrites—I: Metamorphism related studies of Antarctic and other type 3 ordinary chondrites

Thermoluminescence sensitivity measurements have been made on 18 unequilibrated ordinary chondrites; 12 finds from Antarctica, 5 non-Antarctic finds and 1 fall. The TL sensitivities of these meteorites, normalized to Dhajala, range from 0.034 (St. Mary's County) to 2.3 (Allan Hills A78084), and, based primarily on these data, petrologic type assignments range from 3.3 (St. Mary's County) to 3.9 (Allan Hills A78084). Although the very low levels of metamorphism experienced by types 3.0 to ～3.4 evidently cause large changes in TL sensitivity, the new data demonstrate that they are unable to cause any appreciable homogenization of silicate compositions. We have therefore slightly revised the silicate heterogeneity ranges corresponding to the lower petrologic types.We have discovered that the temperature of maximum TL emission and the broadness of the major TL peak, vary systematically with TL sensitivity; as TL increases these parameters first decrease and then increase. Several mechanisms which could account, partially or completely, for the relationship between TL sensitivity and metamorphism are discussed. Those which involve the formation of feldspar—the TL phosphor in equilibrated meteorites—seem to be consistent with the trends in peak temperature and peak width since experiments on terrestrial albite show that the TL peak broadens and moves to higher temperatures as the stable form changes from the low (ordered) state to the high (disordered) state. (The post-metamorphism equilibration temperature of type ～3.5 meteorites would then correspond to the transformation temperature for the high to low form of meteorite feldspar.) Other factors which may be involved are obscuration of the TL by carbonaceous material, changes in the composition of the phosphor and changes in the identity of the phosphor.     

Chemical mobilizations in laterites: evidence from trace elements and U-U-Th disequilibria

Geochemical and mineralogical investigations, including measurements of major and trace elements, Sr isotope ratios, and ²³⁸U-²³⁴U-²³⁰Th activity ratios, were made on an old African laterite to reconstruct its formation steps and assess recent chemical mobilization. The present data support a scenario of discontinuous formation for the laterite, with different bedrock weathering conditions during the formation of each unit, rather than a scenario of continuous formation. Absolute accumulation of Fe, U, and lanthanides in the uppermost ferruginous unit suggests an autochthonous origin of this iron cap by leaching of an older overlying profile. Present chemical distributions of lanthanides, as well as of Rb, K, Ba, and Sr, within the profile cannot be linked to the mineralogical distribution of both relictual primary and authigenic secondary phases. Complementary lanthanide patterns indicate that these elements were primarily accumulated in the uppermost ferruginous unit before further remobilization and accumulation in the underlying horizons. These redistribution processes may be related to the chemical instability of the ferruginous cap. The ²³⁸U-²³⁴U-²³⁰Th disequilibria indicate that recent U mobilization occurs in the whole profile and that, as for lanthanides, there is a vertical redistribution of U from the uppermost ferruginous unit to the underlying horizons. Moreover, these data show that both U losses and gains exist at each level of the profile. A simple modeling of this double U mobilization process is proposed to interpret the ²³⁸U-²³⁴U-²³⁰Th data. Differences in the mobilization and fractionation intensities of the U input and removal processes can account for the two evolution trends, which distinguish the ferruginous unit from the underlying ones. Furthermore, on the basis of this modeling, the profile appears to be in a transient state because of recent changes in the U mobilization conditions, which could correspond to major Pleistocene climatic variations.     

Metamorphism of the H-group chondrites: implications from compositional and textural trends in chondrules

Major and minor element bulk compositions of 373 individual chondrules from 18 H3 to H6 chondrites were determined in polished thin sections by broad-beam electron probe analysis. Bulk chondrule FeO and Al₂O₃ increase and TiO₂ and Cr₂O₃ decrease with increasing petrologic type; normative fayalite, albite and plagioclase increase through the petrologic sequence. Chondrule diameters correlate with phenocryst sizes in porphyritic chondrules of type 3 chondrites, but this correlation is diminished in the higher petrologic types. Furthermore, for a given chondrule diameter, phenocryst sizes are larger in the higher petrologic types. We attribute most compositional trends in chondrules through the petrologic sequence to diffusion and equilibration among chondrules and between chondrules and matrix in response to increasing degrees of thermal metamorphism. Increased phenocryst sizes in the higher petrologic types are probably the result of grain growth during metamorphism.We suggest that H-group chondrites formed by accretion of high-temperature (chondrules) and low-temperature (matrix) materials. Parent materials of each of the petrologic types resembled type 3 chondrites, but had slight compositional differences (e.g. volatiles, rare gases, total iron) inherited during accretion. These differences were predominantly functions of decreasing temperature in the nebula as accretion progressed. Internal reheating of the parent materials to different temperatures and (probably) for different times, as a function of depth in the parent body, caused compositional equilibration, grain coarsening, and reduction of FeO to Fe° by carbon.     

Chemical studies of L chondrites. VI: variations with petrographic type and shock-loading among equilibrated falls

To study compositional trends associated with open and closed system metamorphism and/or shock-induced heating of the L4-6 chondrite parent(s), we used ICPMS and RNAA to quantify 51 trace elements in 48 chemically representative fall samples. With these data, we used graphic and two multivariate statistical methods for examining evidence for compositional differences with respect to petrographic type and degree of shock loading. Comparisons of mildly shocked (S1-S3) L5 and L6 suites (9 and 8 chondrites, respectively) yield no convincing statistical evidence for a difference in trace element content. Our multivariate comparisons show a difference on a model-dependent basis, but yield indeterminate results on a model-independent basis. Compositionally, suites of strongly shocked (S4-S6) and mildly shocked L4-6 chondrites (26 and 19 samples, respectively) can be distinguished at statistically significant levels on both model-dependent and -independent bases. In the strongly shocked suite, contents of refractory lithophiles are higher, and siderophiles and volatiles are lower than those of the mildly shocked suite at moderately (p ≤ 0.05) to highly significant (p ≤ 0.01) levels. Our studies suggest that chemical differences from vaporization and loss of volatiles along with metal/silicate partitioning are present from extended cooling of shock-heated bodies produced by intermittent impacts, especially the massive impact(s) that disrupted the L chondrite parent(s) ∼500 Ma ago.     

Chemical studies of L chondrites—III. Mobile trace elements and 40Ar39Ar ages

We report data for trace elements Ag, Au, Bi, Cd, Co, Cs, Ga, In, Rb, Se, Te, Tl and Zn determined by radiochemical neutron activation analysis in L4–6 chondrites with undisturbed ⁴⁰Ar release patterns or with patterns showing some disturbance in the 4.4–4.6 Gyr plateau indicating shock-induced loss. Mean concentrations are lower, many significantly so, in 16 chondrites with disturbed patterns than in 4 with undisturbed ones, consistent with shock-induced mobilization. Similar trends were noted earlier in L4–6 chondrites having mineralogically observable shock indicators: mean concentrations are lower in strongly shocked (i.e. > 22 GPa) than in mildly shocked (<22 GPa) samples. From trace element contents, L4–6 chondrites with undisturbed ⁴⁰Ar release patterns are mildly shocked but chondrites with disturbed patterns are more strongly heated, on average, than those of shock facies d-f (i.e. 22 to > 57 GPa). Pooling these populations, significantly lower mean concentrations of nearly all trace elements in 26 strongly shocked L4–6 chondrites than in 14 mildly shocked ones indicate loss in shock-formed FeS-Fe eutectic and/or by vaporization during cooling of shock-heated collisional debris. Two-element correlations and the pattern of them, i.e. correlation profiles, are also consistent with this picture. Trace elements can act as thermometers for collisional episodes in L4–6 chondrites but not for earlier thermal fractionations, unless compensation can be made for late shock heating.     

Temporal and spatial patterns of trace elements in the Patuxent River: A whole watershed approach

Trace element distributions, partitioning, and speciation were examined at 15 sites in the Patuxent River watershed from May 1995 through October 1997 to determine possible sources of trace elements to the river and estuary, to examine the relationship of the trace element discharges to freshwater discharges as well as to land use and geographic region, to validate previous estimates of loadings to the river, and to provide baseline data for trace elements in the Patuxent River watershed and estuary. Six freshwater sites were examined, representing different basins and geographic provinces, and nine sites along the estuarine salinity gradient. Subregions within the watershed varied considerably in concentrations and areal yields for some elements. Concentrations of As, Cd, Ni, Pb, and Zn were elevated in the Coastal Plain sites compared to the Piedmont sites, while Cu and Hg were more evenly distributed. Cadmium, Cu, Hg, Ni, Pb, and Zn showed overall positive correlations with river flow while As and methylHg (meHg) showed negative correlations with river flow. Concentrations of trace elements in the estuarine portion of the river were generally low, and consistent with mixing between Patuxent River water with elevated concentrations and the lower concentrations of the Chesapeake Bay. Interesting features included a local Cd maximum in the low salinity region of the estuary, probably caused by desorption from suspended sediments, and a significant input of water containing high As concentrations from the Chesapeake Bay and from As being released from bottom sediments in summer. Comparisons between the estimated annual flux of trace elements and the estimates of suspected source terms (atmospheric deposition, urban runoff, and known point sources) suggest that, except for Hg, direct atmospheric deposition is small compared to fluvial loads. Current estimates of trace element inputs from point sources or from urban runoff are inadequate for comparison with other sources, because of inappropriate techniques and/or unacceptably high detection limits. A complete examination of trace element dynamics in the Patuxent River (and in other coastal systems) will require better data for these potential sources.     

Chemical and physical studies of type 3 chondrites—III. Chondrules from the Dhajala H3.8 chondrite

Fifty-eight chondrules were separated from the Dhajala H3.8 chondrite and their thermoluminescence properties were measured. Chips from 30 of the chondrules were examined petrographically and with electron-microprobe techniques; the bulk compositions of 30 chondrules were determined by the fused bead technique. Porphyritic chondrules, especially 5 which have particularly high contents of mesostasis, tend to have higher TL (mass-normalized) than non-porphyritic chondrules. Significant correlations between log(TL) and the bulk CaO, Al₂O₃ and MnO content of the chondrules, and between log(TL) and the CaO, Al₂O₃, SiO₂ and normative anorthite content of the chondrule glass, indicate an association between TL and the abundance and composition of mesostasis. Unequilibrated chondrules ( i.e. those whose olivine is compositionally heterogeneous and high in Ca) have low TL, whereas equilibrated chondrules have a wide range of TL, depending on their chemical and petrographic properties.We suggest that the TL level in a given chondrule is governed by its bulk composition (which largely determined the abundance and composition of constituent glass) and by metamorphism (which devitrfied the glass in those chondrules with high Ca glass to produce the TL phosphor). We also suggest that one reason why certain chondrules in type 3 ordinary chondrites are unequilibrated, while others are equilibrated, is that the mesostasis of the unequilibrated chondrules resisted the devitrification. This devitrification is necessary for the diffusive communication between chondrule grains and matrix that enables equilibration.