Al Fluxes Research Articles

Spatial variability of atmospheric fluxes of metals (Al, Fe, Cd, Zn and Pb) and phosphorus over the whole Mediterranean from a one-year monitoring experiment: Biogeochemical implications

During the ADIOS EU funded program, the total atmospheric deposition of metals (Al, Cd, Zn and Pb) and elements of biogeochemical interest (Fe, P) were collected monthly during one year (June 01/May 02) at nine coastal sampling sites around the Mediterranean Sea. In order to avoid bias in assessing the spatial and temporal variability, the sampling was performed using the same protocol and the same schedule, with a centralized location for preparation and samples analysis. The highest fluxes were found for Al and Fe (1004 and 665 kg km − 2 yr − 1 respectively), with these fluxes being two times higher in the western basin than in the eastern one, due to the occurrence of strong Saharan dust fall. The comparison of Al fluxes with previous data for the northwest Mediterranean over the past two decades underlines the very high variability, precluding any assessment of spatial trends from our study. The mean phosphorus flux was 33 kg km − 2 yr − 1 , with a possible slight decrease from the western/central basin to the eastern one. Zinc fluxes were slightly higher in the western basin than in the eastern one (12.1 and 7.7 kg km − 2 yr − 1 respectively). Lead fluxes were identical in both basins, at 1.46 kg km − 2 yr − 1 , but gradients within the basins indicated higher fluxes in the south of the western basin and in the north of the eastern one. A decreasing trend was observed over the past two decades for the northwestern basin, linked to reduction in the use of unleaded gasoline in the western European countries. For cadmium, the fluxes were of the same order of magnitude in both basins, with an average value of 0.037 kg km − 2 yr − 1 , and with no clearly-decreasing trend over the past 20 years. Using the element-to-Al ratio in Saharan dust, the non-crustal contribution to the total atmospheric deposition was on average 11, 70, 81, 88, and 96% for Fe, P, Pb, Zn and Cd respectively. Fe and Al were well correlated, with the Fe/Al ratio for western sites being typical of the Saharan end-member already defined in the western basin, but slightly higher in the eastern one, probably due to a variation in the composition of the source of the Saharan dust. Biogeochemical response to atmospheric inputs was considered at the scale of the whole stratification period and at the event scale. During the stratification period dissolved iron concentrations in the surface mixed layer could increase over the whole Mediterranean Sea by 1.8–5.3 nM in response to the accumulation of atmospheric iron from both anthropogenic and natural origins, confirming a strong impact of atmospheric inputs on the iron marine cycle in the Mediterranean Sea. Dissolved phosphate in surface waters was calculated to be significantly augmented by the atmospheric inputs during summer and we estimated that the atmospheric input of phosphate can sustain the new primary production during the stratification period at the scale of the whole open Mediterranean. At the event scale the response of the biota to a strong dust fall, which occurred during the Adios year in the Ligurian Sea, was registered by the water column biogeochemistry and chlorophyll content; this Saharan event did bring a considerable amount of dust and new nutrients, inducing a significant new production and a doubling of the chlorophyll concentration in the surface mixed layer. This event, accompanied by strong winds and heavy rains, also induced profound changes in the physical characteristics of the surface waters, triggering the beginning of the destratification of the surface waters that also induced an input of P from below.

MBE growth of cubic Al x In 1–x N and Al x Ga y In 1–x–y N lattice matched to GaN

Ternary and quaternary cubic c-AlxIn1–xN/GaN and c-AlxGayIn1–x–yN/GaN heterostructures lattice-matched to c-GaN on freestanding 3C-SiC substrates were grown by plasma-assisted molecular beam epitaxy. The c-AlxGayIn1–x–yN alloy permits the independent control of band gap and lattice parameter. The ternary and quaternary films were grown at 620 °C. Different alloy compositions were obtained by varying the Al and Ga fluxes. The alloy composition was measured by Energy Dispersive X-ray Spectroscopy (EDX) and Rutherford Backscattering Spectrometry (RBS). X-ray reciprocal space map of asymmetric (-1-13) reflex were used to measure the lattice parameters and to verify the lattice match between the alloy and the c-GaN buffer. (© 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)

Monsoon control on trace metal fluxes in the deep Arabian Sea

Particulate fluxes of aluminium, iron, magnesium and titanium were measured using six time-series sediment traps deployed in the eastern, central and western Arabian Sea. Annual Al fluxes at shallow and deep trap depths were 0.47 and 0.46 g m-2 in the western Arabian Sea, and 0.33 and 0.47 g m-2 in the eastern Arabian Sea. There is a difference of about 0.9–1.8 g m-2y-1 in the lithogenic fluxes determined analytically (residue remaining after leaching out all biogenic particles) and estimated from the Al fluxes in the western Arabian Sea. This arises due to higher fluxes of Mg (as dolomite) in the western Arabian Sea (6–11 times higher than the eastern Arabian Sea). The estimated dolomite fluxes at the western Arabian Sea site range from 0.9 to 1.35gm-2y-1. Fe fluxes in the Arabian Sea were less than that of the reported atmospheric fluxes without any evidence for the presence of labile fraction/excess of Fe in the settling particles. More than 75% of Al, Fe, Ti and Mg fluxes occurred during the southwest (SW) monsoon in the western Arabian Sea. In the eastern Arabian Sea, peak Al, Fe, Mg and Ti fluxes were recorded during both the northeast (NE) and SW monsoons. During the SW monsoon, there exists a time lag of around one month between the increases in lithogenic and dolomite fluxes. Total lithogenic fluxes increase when the southern branch of dust bearing northwesterlies is dragged by the SW monsoon winds to the trap locations. However, the dolomite fluxes increase only when the northern branch of the northwesterlies (which carries a huge amount of dolomite accounting 60% of the total dust load) is dragged, from further north, by SW monsoon winds. The potential for the use of Mg/Fe ratio as a paleo-monsoonal proxy is examined.

Photochemical Source of Metals for Sediments

A mass budget study of major in-lake Al fluxes, palaeolimnological data on a >10,000 year old sediment record, and in situ photochemical experiments performed at Plesné Lake (Czech Republic) suggest that photochemical liberation of organically bound aluminum (Al) and iron (Fe) by solar radiation is a significant natural source of their ionic species for lakes and subsequent oxyhydroxides for sediments. The results show that photochemically induced transformation of dissolved Al and Fe to solid oxyhydroxides deposited to Plesné Lake sediment dominated (91 and 73%, respectively) their sedimentary flux throughout the preindustrial era, since soil formation initiated in the catchment. The following sequence of processes occurs: (i) soil organic acids dissolve and bind metals and export them from terrestrial to aquatic systems. (ii) Photochemical decomposition of organic-metal complexes liberates a significant portion (approximately 50% in Pleseé Lake) of organically bound Al and Fe as inorganic ions. (iii) The liberated ionic Al and Fe hydrolyze, precipitate as oxyhydroxide particles, and settle. We hypothesise that the same Al and Fe transporting process occurs in other lakes and coastal marine areas and is ecologically important because Al and Fe oxyhydroxides can bind trace metals and phosphorus.

Self-organized superlattices along the [001] growth direction in In 0.52Al 0.48As layers grown on nominally (001) InP substrates by molecular beam epitaxy

Horizontal self-organized superlattice structures consisting of alternating In-rich and Al-rich layers formed naturally during solid-source molecular beam epitaxy (MBE) growth of In 0.52Al 0.48As on exactly (001) InP substrates, with In and Al fluxes unchanged. The growth temperatures were changed from 490 to 510 ∘C, the most commonly used growth temperature for In 0.52Al 0.48As alloy. No self-organized superlattices (SLs) were observed at the growth temperature 490 ∘C, and self-organized SLs were observed in InAlAs layers at growth temperatures ranging from 498 to 510 ∘C. The results show that the period of the SLs is very highly regular, with the value of ∼6 nm, and the composition of In or Al varies approximately sinusoidally along the [001] growth direction. The theoretical simulation results confirm that the In composition modulation amplitude is less than 0.02 relative the In composition of the In 0.52Al 0.48As lattice matched with the InP substrate. The influence of InAs self-organized quantum wires on the spontaneously formed In x Al 1− x As/In y Al 1− y As SLs was also studied and the formation of self-organized In x Al 1− x As/In y Al 1− y As SLs was attributed to the strain-mediated surface segregation process during MBE growth of In 0.52Al 0.48As alloy.

Dissolved Al distribution, particulate Al fluxes and coupling to atmospheric Al and dust deposition in the Arabian Sea

Abstract Concentrations of dissolved aluminium (Al) were measured in the Arabian Sea during three different seasons. Atmospheric Al concentrations and particulate Al fluxes at 100 m depth were measured during the northeast monsoon. Particulate Al fluxes in the deep Arabian Sea were determined from a 16 months sediment trap mooring. Surface water Al concentrations were higher in the more oligotrophic southern Arabian Sea (10–15 nM) than in the north-western Arabian Sea (3–9 nM). Seasonal variability was observed in the central and northern Arabian Sea with elevated concentrations during spring/summer. Mixed-layer residence times of 0.4–4 years for dissolved Al estimated with respect to both atmospheric input and particle flux exports indicate rapid removal from surface layers. Scavenging removal exerts major control on Al cycling. There are indications of a sediment source for dissolved Al. Atmospheric Al concentration during the northeast monsoon were 10–30 nmol m−3 and up to 81 nmol m−3 during dust pulses. The Al content in aerosol particles was variable with an average of 950 μmol g−1 (or 2.6% by weight), which is about threefold lower than in average continental crust (8.04%). Using available aerosol Al concentrations combined with satellite data, the annual atmospheric deposition rate of Al is estimated to ∼16000 μmol m−2 yr−1. Depending on the Al content in aerosol material (2.6–8.04%), this converts to a dust deposition rate of 5.4–16.7 g m−2 yr−1 that agrees with sediment trap results and sediment accumulation data. A simple model based on dissolved Al concentrations in surface waters is used to estimate dust deposition rates. We demonstrate that a relatively short residence time of ⩽3 yr and an aerosol Al content of only 2.6% provide improved constraints under steady-state conditions. The strong seasonality in particulate Al flux in the deep central Arabian Sea was dominated by the southwest monsoon sedimentation signal; about 46% of the annual Al flux is provided within only ca. 2 1 2 months. The estimated total atmospheric Al input and the observed deep-water flux of particulate Al of about 14000 μmol m−2 yr−1 agree closely. This suggests that the atmosphere is the major Al source to the Arabian Sea.

Release of potassium, calcium, iron and aluminium from Norway spruce, Scots pine and silver birch logging residues

Potassium (K), calcium (Ca), iron (Fe) and aluminium (Al) release from Norway spruce (Picea abies Karsten), Scots pine (Pinus sylvestris L.) and silver birch (Betula pendula Roth.) logging residues (fine roots, foliage and small branches) were studied by means of litterbags over a period of three years in clear-cut area and adjacent uncut Norway spruce dominated mixed boreal forest in eastern Finland (63°51′ N, 28°58′ E, 220 m a.s.l) to determine the amounts and rates of release for these elements and to evaluate whether clear-cutting accelerates mineralization. Almost all K was released from logging residues already during the first year. Calcium was released from foliage and roots but accumulated in branches. Most of the roots Fe and Al content were released during three years while the absolute amounts of Fe and Al in branches and foliage generally increased with decomposition. The results indicate that mineralization is slightly accelerated as a result of clear-cutting since K from foliage and branches of all studied tree species and Ca from pine and spruce roots was released significantly faster at the clear-cut plot than at the forest plot. In three years the initial K pool in the logging residues declined by 90%, Ca by 8%, Fe by 55% and Al by 61% in the clear-cut area. These results indicate that Ca is retained a long time; but Fe, Al and in particular, K are soon released from logging residues. Fine roots of the logged trees release large amounts of Fe and Al and can significantly affect Fe and Al fluxes.

La/Yb and Th/Sc in settling particles: Vertical and horizontal transport of lithogenic material in the western North Pacific

In order to understand the origins and transport processes of lithogenic aluminosilicate in the western North Pacific, time-series sediment traps were deployed at Stn. KNOT at 44°N 155°E (water depth: 5320 m). Annual mean content of aluminum (Al) which is a proxy of lithogenic material in settling particles was 0.38% at 1 km depth, 0.78% at 3 km depth and 1.14% at 5 km depth, respectively. Mass fluxes were corrected by 230Th and 231Pa activities in particles and seawater. Annual mean corrected Al fluxes at 1, 3 and 5 km depths were calculated to 0.45, 0.46 and 0.33 g/m2/year, respectively. Lanthanum/ytterbium (La/Yb) and thorium/scandium (Th/Sc) ratios were used as proxies for estimating the origin of lithogenic aluminosilicate in settling particles. Both La/Yb and Th/Sc ratios in settling particles increased with depth, and the seasonal variation was the smallest at 5 km depth. It was considered that variations of La/Yb and Th/Sc ratios in settling particles reflected the change of source material of lithogenic aluminosilicates. Origins of lithogenic materials in the settling particles were summarized using two end-members, loess from central Asian continent (AL) and basalts from the Kurile-Kamchatka Islands (KK). Contributions of end-members in settling particles were estimated from the La/Yb and Th/Sc ratios. The contribution of KK particles in lithogenic materials was calculated to be 84% at 1 km depth, and 61% at 3 and 5 km depths, respectively. Although the KK material was the dominant constituent of the terrigenous component in settling particles at Stn. KNOT, the KK flux decreased with depth. On the other hand, the AL flux increased with depth and the significant increase of the AL flux was presumably due to the lateral import of AL particles to the deep layer at Stn. KNOT. From these results we conclude that the distribution of lithogenic materials recorded in sediments in the western North Pacific are influenced by the northward transport of AL within the deep layer.

Estimation of particulate organic carbon flux in relation to photosynthetic production in a shallow coastal area in the Seto Inland Sea

Sediment trap experiments were carried out three times from 1999 to 2000, in the western part of the Seto Inland Sea (Suo-Sound), Japan. We investigated both the particulate flux and the composition of chemical substances in the sediment trap samples. Based on the results, we discuss the origin of particulate organic carbon (POC) collected by the sediment traps in a coastal area. Moreover, we purposed to estimate the flux of the portion of the POC that is derived from phytoplankton photosynthesis. The fluxes of POC varied between 677 and 3424 mgCm−2d−1. Significant positive correlations between POC and aluminum (Al) fluxes suggested that these components show almost the same behaviour. The mean value of the Al flux was about eight times higher than that of Al burial rates on the sediment surface. Therefore, it seems that the POC flux observed with the sediment traps was considerably overestimated. Moreover, judging from the fact that Al is a typical terriginous element, it seems that most of the POC collected in the sediment traps derived from the re-suspended surface sediment or sediment transported laterally from shallow flanks such as intertidal mudflats. The fluxes of chlorophyll a (Chl a) were independent of the POC fluxes, and a relatively consistent correlation was found between Chl a abundance in the water column and the Chl a flux. Moreover, surface sediment Chl a content was approximately 100 times lower than that of suspended matter. Therefore, resuspension and terriginous contributions to Chl a collected in sediment traps are likely to be negligible. The POC content in the trap samples varied between 22.4 and 70.7 mgg−1 dry weight. The variations of POC contents were positively correlated with the Chl a contents: POC(mgg−1)=76.5×Chla(mgg−1)+26.0 (r=0.95, p<0.01, n=9). This result shows that POC contents strongly corresponded with phytoplankton and their debris. It was also considered that the fraction of POC derived from phytoplankton primary production could be estimated as Chl a content times a certain factor. In this study, we estimated the flux of the portion of the POC originating from phytoplankton production by multiplying the Chl a fluxes by 76.5 (the mean POC:Chl a ratio in the trap samples). These values varied between 308 and 758 mgCm−2d−1, and accounted for 35.1±21.2% of total POC flux. Although the amount of POC that originates from phytoplankton photosynthesis was a small portion of total POC flux, it seems to be a large portion of potential primary production in the water column.

Dry deposition and foliar leaching of mercury and selected trace elements in deciduous forest throughfall

The estimated annual throughfall deposition flux of Hg in a northern mixed-hardwood forest in the Lake Huron Watershed was 10.5±1.0 μg m −2 compared to an annual precipitation Hg flux of 8.7±0.5 μg m −2 (June 1996–June 1997). The source of this additional Hg in throughfall is often attributed to wash-off of dry deposition, but foliar leaching of Hg may also be important. To determine the influence of both dry deposition and foliar leaching of Hg and other elements in throughfall, we measured a suite of trace elements (Hg, Al, Mg, V, Mn, Cu, Zn, As, Rb, Sr, Cd, Ba, La, Ce, and Pb) in throughfall, precipitation, and ambient air samples from a northern mixed-hardwood forest. Based on a multiple linear regression model, dry deposition had the most important influence on Hg, Al, La, Ce, V, As, Cu, Zn, Cd, and Pb fluxes while foliar leaching strongly influenced Mg, Mn, Rb, Sr, and Ba fluxes in net throughfall. The Hg dry deposition flux was estimated using gaseous and aerosol Hg measurements and modeled deposition velocities. The calculated dry deposition flux (∼12–14 μg m −2) of Hg to the canopy indicated that atmospheric deposition of Hg could easily account for all of the Hg deposited in net throughfall (1.9±0.1 μg m −2). Although there is a large uncertainty associated with these techniques, the modeling estimates indicate that atmospheric Hg may account for all of the Hg deposited in litterfall (11.4±2.8 μg m −2).

Settling fluxes of U- and Th-series nuclides in the Bay of Bengal: results from time-series sediment trap studies

A two-year (1990–1991) collection of sediment trap samples from three locations in the Bay of Bengal was analyzed for Al, Ca and U- and Th-series nuclides to obtain their annual fluxes and pattern of deposition. The time-series traps were deployed at two depths: ∼(900–1100) and ∼(2100–3000) m at ∼15°N 89°E (NBBT); ∼13°N 84°E (CBBT) and ∼5°N 87°E (SBBT). The deeper traps were several hundred meters above the sea floor. Each time-series trap had 13 cups and each cup was programmed to collect settling particles for ∼25 days. The mass fluxes in the NBBT shallow and deep traps were nearly the same, ∼33 g m −2 yr −1. At the CBBT site the mass flux in the deeper trap was ∼65 g m −2 yr −1, about 40% more than that in the shallow trap. The maximum particle flux of ∼300 mg m −2 d −1 was observed in the deep trap of CBBT site and is attributable to extreme floods in the Godavari and Krishna rivers and lateral injection of particulate matter to the deep sea. Variations in Al deposition flux in the traps is strongly coupled to that of mass flux, attributable to the scavenging of aluminosilicates from the water column by sinking organic aggregates. The radionuclide fluxes ( 230 Th a , 228 Th and 210 Pb ) in the traps provide insight into the role of their removal by vertical particle flux or lateral transport. At the NBBT site, the observations that (i) the authigenic flux of 230 Th ( 230 Th a ) in the deep trap nearly balances its production in the water column above the trap depth, (ii) the differences in the 210 Pb fluxes between the shallow and deep traps matches its in-situ production in the water column between the traps and (iii) the mass, Al and 228 Th fluxes are similar in the shallow and deep traps, all suggest that at this site 230 Th and 210 Pb produced in the water column are removed predominantly by vertical particle flux. At the CBBT site, there is a consistent increase in the mass, Al and 228 Th fluxes in the deep trap relative to those in the shallow trap, and the deposition of 210 Pb and 230 Th a are considerably in excess over their production. Enhanced particle aggregation in the intermediate waters, resuspension of “nascent” sedimentary material and focussing of particles in the region of the deep traps can all account for these observations. The particulate flux in the deep trap at the SBBT site, farthest from the river mouths, based on nine months of data is ∼38 g m −2 yr −1 and is dominated by CaCO 3. The authigenic 230 Th flux in the SBBT deep trap matches its overhead production, whereas for 210 Pb the measured flux is ∼30% less than that expected from its supply. This indicates that at this site the bulk of 230 Th produced is removed by vertical particle flux, whereas about one-third of the 210 Pb supplied is carried off laterally and presumably removed at the ocean boundaries. Evidence for uranium removal from the thermocline by settling particles is seen in the higher (U/Th) ratio of several shallow trap samples relative to that in the deep traps. The 210 Pb and 228 Th fluxes in the shallow and deep traps show strong co-variance with mass flux, which in turn is coupled to sediment discharge and productivity in the Bay through monsoon cycles. These observations suggest that nuclide scavenging processes in surface waters and interior of the Bay of Bengal respond quickly to particle flux variations. Thus, the study of mass, Al, Ca and radionuclide fluxes at these three sites provide not only data on spatial and temporal variability of their fluxes over a two-year period but also insight into the processes regulating their depositional fluxes.

Recent terrestrial and carbonate fluxes in the pelagic eastern Mediterranean; a comparison between sediment trap and surface sediment

A sediment trap mooring was deployed in the central eastern Mediterranean from November 1991 to August 1994. At 3000 m water depth, total mass, Al, Ca, Mg, Sr and 230Th fluxes recovered by the sediment trap are highly seasonal, with highest fluxes during early spring in 1992 and 1993, and during late-spring/early-summer in 1994. Comparison of historic annual satellite-derived chlorophyll records (coastal zone colour scanner) with the trap flux time series indicates a lag of 4–6 months between maximum primary production in the surface ocean and maximum flux recorded by the trap. Only the flux of coccospheres to the trap is at a maximum ∼1 month after maximum pigment concentrations in surface waters, a value commonly found in other areas. Quantification of the inorganic (lithogenic) flux to the trap indicates that Saharan dust is likely to be the major contributor to the trap mass flux. The trapping efficiency of the sediment trap, as calculated from the intercepted 230Th flux, is only 23%, and the trap Al-flux is similarly ×4 lower than Al fluxes measured in nearby uppermost sediments. Compared with surface sediments, the trap-intercepted carbonate fluxes are even lower (×9) than the corresponding lithogenic fluxes. This is partly due to the very low abundance of large (>32 μm) foraminifera and pteropods found in the trap material compared to the surface sediment. We speculate that the period of our sediment trap deployment was insufficiently long to recover episodical large fluxes, such as may be triggered by North Atlantic Oscillation variations.

Atmospheric wet deposition of trace elements to Chesapeake and Delaware Bays

The atmospheric wet depositional fluxes of selected trace elements (Al, Fe, Mn, Pb, Cd, Zn, Cu, Ni and Cr) were examined at two mid-Atlantic sites, which were located on the shore of the Delaware (Lewes) and Chesapeake (Stillpond) Bays from September 1995 to August 1996. Sampling was carried out on a weekly integrated basis, using “clean” techniques developed especially for trace elements. Monthly integrated depositional fluxes showed large variability in space and time for most of the measured elements, with similar ranges of variation for both sites. A good correlation ( r 2=0.80 , Fe/Al=0.59) between Al and Fe fluxes at both sites suggests their major origin from the earth's crust (resuspended regional soils). Based on the enrichment factor calculation, it is suggested that at least half of the Cr and Mn and more than 90% of the Cd, Zn, Pb, Cu, and Ni are from non-crustal (presumably anthropogenic) sources. Seasonal variations of Al, Fe and Mn appear to be mainly controlled by local/regional agricultural activities, whereas Pb and Cd fluxes are largely dictated by precipitation amounts. Factors controlling fluxes of other elements (Zn, Cu, Ni and Cr) are not clearly distinguished due to large episodic inputs of local/regional contamination sources. A comparison of annual trace element fluxes during 1991–1996 reveals a factor of 2–4 higher fluxes of Zn, Cu and Ni during 1995–1996 at both sites, indicating regional (i.e., mid-Atlantic region) increase of the fluxes during the sampling period. The similar annual fluxes and trends at both sites suggest that these data can be employed to estimate atmospheric wet flux of trace elements to mid-Atlantic coastal waters (including Chesapeake/Delaware Bays and mid-Atlantic Bight).

Seasonal impact of mineral dust on deep-ocean particle flux in the eastern subtropical Atlantic Ocean

Seasonal lithogenic particle and Al fluxes were obtained from a deep-ocean sediment trap deployment during 1992 and 1993 off NW Africa, and were compared concurrently with atmospheric Al concentrations and two-dimensional backward trajectories of windfields from two barometric levels in the lower and mid troposphere. Marine Al fluxes, lithogenic particle fluxes and grain size distributions in the area were found to be directly linked to airmass pathways and surface mineral aerosol concentrations. At 1000 m water depth, highest Al fluxes (10.77 mg m −2 day −1), lithogenic particle fluxes (99.25 mg m −2 day −1) and smallest mean grain sizes (11.9 μm) occurred during the winter and spring season, concurrent with highest atmospheric dust load and Al-concentrations (15 300 ng m −3) in the lower troposphere. A strong seasonal change of the main atmospheric dust transport from low altitude winds during winter/spring to higher altitudes during summer is clearly reflected at depth by a significant coarsening of mean grain sizes (18.6 μm) and lowest Al (0.81 mg m −2 day −1) and lithogenic particle fluxes (11.3 mg m −2 day −1) found in the sediment traps. The comparison of marine, atmospheric and model derived data used within this study highlights the close temporal coupling between atmospheric dust transport and the deep-ocean particle stock.

In situ monitoring of AlGaAs compositions and GaAs growth at high temperature by spectroscopic ellipsometry and desorption mass spectroscopy

In situ spectroscopic ellipsometry has been used in conjunction with desorption mass spectrometry to determine both the composition of Al and growth rate of AlxGa1−xAs and GaAs at high temperatures (600–720 °C). Ex situ cross sectional transmission electron microscopy verified the AlGaAs/GaAs layer thicknesses grown by molecular beam epitaxy at these elevated temperatures. The substrate temperature dependence on Ga desorption rates, given the As, Ga, and Al fluxes and V/III flux ratios, was examined by Monte Carlo simulation. These simulations showed good agreement with the experimental results.

Uptake of aluminium by plant cells

SUMMARYThe question of intracellular or extracellular primary lesion in the Al toxicity syndrome is unresolved. One of the crucial points for answering that question is quantification of Al fluxes across the plasma membrane within seconds or minutes of exposure of roots to Al, i.e. concurrent with or preceding the first symptoms of Al toxicity. A review of the available literature on Al uptake shows that there is abundant information on Al accumulation in root tissues only after the relatively prolonged uptake period, ranging from 30 min to over 24 h. Most of these reports either assume or claim explicitly that intracellular Al was measured, even though Al ions are bound strongly to negative charges in the apoplasm. Therefore, an effective and complete desorption of apoplasmic Al after the uptake period is crucial for measurements of intracellular Al. However, published studies do not seem to have desorbed cell‐wall‐bound Al appropriately. Recent studies with giant algal cells of Chara corallina, where physical separation of cell wall and cytoplasm after the uptake period can be achieved surgically, showed that desorption of Al from the apoplasm, even employing a wider variety of desorbents and more stringent conditions than in any of the previously published reports, cannot be achieved completely within 5 h. Consequently, measured rates of uptake of Al across the plasma membrane of intact Chara cells employing the cell wall/cytoplasm separation technique are up to several orders of magnitude lower than previously published values in which cell‐wall Al was attributed to the transmembrane uptake component. Although there is no doubt that Al does cross the plasma membrane, no information exists about which Al species or complexes take part in the transmembrane flux, mainly because of complexities inherent in Al speciation at the solution/ion exchanger (i.e. the cell wall and the membrane) interface. Similarly, it is not known which membrane transporters are involved in transport of Al across the plasma membrane. The Al‐resistant plant genotypes generally accumulate less Al in the root tips than do the Al‐sensitive genotypes. No direct relationship, however, appears to exist between increased organic acid extrusion as a mechanism of resistance to Al and decreased transmembrane flux of Al. Al‐accumulator plant species accumulate relatively large amounts of Al in their tissues without having a greater Al uptake rate than non‐accumulator genotypes. Progress in deciphering structural and functional aspects of transport of Al across the plasma membrane of intact plant cells will rely on using giant algal cells in which physical separation of the cell wall and the cytoplasm can be achieved, because, at present, there is no reliable quantitative method which can overcome problems presented by a relatively large apoplasmic Al pool remaining after desorption of intact root cells of higher plants. CONTENTS Summary 389 I. Introduction 390 II. Toxicity symptoms and uptake of aluminium 390 III. Ionic speciation of aluminium 392 IV. Analytical techniques for measuring aluminium 393 V. Uptake of aluminium 395 VI. Localization of aluminium in cells and tissues 398 VII. Genotypic differences 400 VIII. Aluminium accumulators 401 IX. Future research 401 Acknowledgements 402

A 300 kyr high‐resolution aridity record of the North African continent

New oxygen isotope data and a high‐resolution carbonate record of a core from the northwest African margin are presented, in combination with geochemical (Zr/Rb ratio, Al, Ba, organic carbon, and biogenic opal fluxes) and sedimentological (median grain size) parameters, to show variations in African continental aridity, wind strength, and upwelling productivity. The carbonate record is diluted by the supply of aeolian dust from the arid Saharan region of north Africa and is seen to vary with a characteristic sawtooth pattern indicating rapid increases in dust flux followed by gradual declines. Spectral analysis of the carbonate profile and comparison with SPECMAP (stacked oxygen isotope) and ETP (solar radiation) profiles shows that the precessional (23 kyr) insolation‐induced African climate system is decoupled from global (ice volume) climate changes operating on a 100 kyr frequency. Detailed studies of glacial terminations I, II, and III confirm the presence of rapid, short timescale climatic variations (in aridity).

Substrate temperature dependence of SQW alloy and superlattice lasers grown by MBE using As 2

Until recently most semiconductor device structures grown by MBE have used As 4. In such structures during the growth of (Al,Ga)As alloys a characteristic roughness is observed under some growth conditions. The threshold current of lasers grown using As 4 is also strongly influenced by the choice of substrate temperature. It is possible therefore that these two factors are related. We have compared the morphology of thick films of Al 0.5Ga 0.5As grown using As 2 and As 4 over the temperature range 610 to 710°C. For films grown with As 4 the characteristic roughness in such structures is observed but over the whole temperature range films grown with As 2 have a specular appearance. We have grown, using As 2, a series of graded refractive index separate confinement heterostructure single quantum well (57 Å) lasers with and without pre-layers at temperatures from 610 to 710°C. In this series of samples the Al and Ga fluxes were kept constant, the nominal Al fractions for the outer confinement and barrier regions were 50% and 25%, respectively. A similar set of stepped separate confinement structures with prelayers was grown for comparison. In a fourth series of structures the separate confinement and part of the outer cladding regions were replaced by short period superlattices. Contrary to previous reports for structures grown using As 4, no strong dependence of threshold current on growth temperature was observed. Results for four sets of lasers will be discussed in detail. An additional benefit of using As 2 is the reduced loss of Ga at high temperatures due to a reduction of free metallic group III element present on the surface during growth.

Layer by layer growth of FeAl on InP(100) substrates

Thin films of FexAly have been grown by molecular-beam epitaxy on InP(100) substrates. Before growth of the epitaxial intermetallic alloy a 0.2-μm InAlAs buffer and a ten-layer AlAs diffusion barrier were deposited. Initial films were grown at 200 °C and were stable up to at least 600 °C. Upon the initiation of growth the reflection high-energy electron diffraction intensity was observed to decrease slowly and then exhibited oscillations corresponding to layer-by-layer growth. There was a short incubation period before the onset of the intensity oscillations that depended upon the ratio of the Fe and Al fluxes. The shape of the specular beam was measured versus incident angle and indicates that the surface is nearly as smooth as the starting AlAs surface. Upon heating to 550 °C the beams became sharper yet. Growth on the annealed surface at 200 °C gave intensity oscillations that were an order of magnitude stronger. Initial attempts to grow epitaxial AlAs on top of the FeAl layer showed three-dimensional island growth.

Mass-action control of AlGaAs and GaAs growth in molecular beam epitaxy

GaAs evaporation during molecular beam epitaxy (MBE) is measured using reflection high-energy electron diffraction (RHEED). On the (001) surface there is a first-order phase transition from a 2 × 4 to 1 × 1 reconstruction. Upon crossing the phase boundary into the 1 × 1 structure, layer-by-layer evaporation of GaAs is observed. This evaporation affects the rate of growth of GaAs and AlxGa1−xAs by MBE. The dependence of the growth rate on substrate temperature and on As, Al, and Ga fluxes is followed by measuring RHEED intensity oscillations. The results agree quantitatively with the mass-action analysis of Heckingbottom [R. Heckingbottom, J. Vac. Sci. Technol. B 3, 572 (1985)]. At substrate temperatures above 850 K no differences between As2 and As4 incident fluxes are observed.