Environ. Sci. Technol. 2009, 43, 8178–8183
Influence of Size and Aggregation on the Reactivity of an Environmentally and Industrially Relevant Nanomaterial (PbS) J U A N L I U , * ,† D E B O R A H M . A R U G U E T E , † MITSUHIRO MURAYAMA,‡ AND MICHAEL F. HOCHELLA, JR.† Center for NanoBioEarth, Department of Geosciences, Virginia Tech, Blacksburg, Virginia 24061, Institute for Critical Technology and Applied Science, 1991 Kraft Drive, Blacksburg, Virginia 24061
Received July 15, 2009. Revised manuscript received September 15, 2009. Accepted September 17, 2009.
Rarely observed nanoparticle dissolution rate data have been collected and explained for an environmentally and industrially relevant nanomaterial (PbS, the mineral galena) as a function of its particle size and aggregation state using highresolution transmission electron microscopy (HRTEM) and solution analysis. Under identical anoxic acidic conditions (pH 3 HCl), it has been determined that the dissolution rate of PbS galena varies by at least 1 order of magnitude simply as a function of particle size, and also due to the aggregation state of the particles (dissolution rates measured are 4.4 × 10-9 mol m-2 s-1 for dispersed 14 nm nanocrystals; 7.7 × 10-10 mol m-2 s-1 for dispersed 3.1 µm microcrystals; and 4.7 × 10-10 mol m-2 s-1 for aggregated 14 nm nanocrystals). The dissolution rate difference between galena microparticles and nanoparticles is due to differences in nanotopography and the crystallographic facespresent.Aggregatevs.disperseddissolutionratesarerelated to transport inhibition in the observed highly confined spaces between densely packed, aggregated nanocrystals, where selfdiffusion coefficients of water and ions decrease dramatically. This study shows that factors at the nanometer scale significantly influence the release rate of aqueous, highly toxic and bioavailable Pb in natural or industrial environments during galena dissolution.
Introduction A report from the National Research Council of the U.S. National Academy of Sciences (NAS), released in December 2008, finds “serious weaknesses” in the U.S. federal government’s plan to address the potential health and environmental risks posed by the manufactured nanomaterials that are being increasingly used in medical treatments, food additives, advanced electronics and batteries, skin-care products, and many other revolutionary materials, products, and processes (1). The report makes it very clear that there is a large gap between, for example, the benefits of nanotechnology in developing therapies for disease, and the very * Corresponding author tel: 509-371-6990; fax: 509-371-6354; e-mail:
[email protected]; current address: Chemical and Materials Sciences Division, Pacific Northwest National Laboratory, PO Box 999, MSIN K8-96, Richland, WA 99352. † Virginia Tech. ‡ Institute for Critical Technology and Applied Science. 8178
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 21, 2009
real potential risks of nanomaterial exposure to humans and the environment. At the same time, naturally occurring nanoparticles are ubiquitous in the hydrosphere, atmosphere, and soil environments as a result of mineral weathering, microbial activity, nucleation/crystallization processes, and anthropogenic activities, etc. (2, 3). Perhaps it is not surprising that there are a number of the same or very similar nanomaterials that are both (1) synthesized and studied for/ used in commercial applications, and (2) found naturally in the environment while playing important roles in geo- and biochemical reactions and cycles. These nanomaterials provide a valuable link between the study of the environmental implications of nanotechnology (3) and nanogeoscience (2) and make it apparent that these two fields are in fact highly complementary. One such nanomaterial relevant to several aspects of both nanotechnology and nature is nanoparticulate lead sulfide (PbS) which has been synthesized and utilized by material scientists (4), and which is naturally occurring as the mineral galena, the principal ore mineral of lead. Galena can also be synthesized with very small size and shape dispersion in the nanorange (4, 5), and is therefore an excellent material for this study. To better understand some of the nanomaterials that have environmental importance such as nano-PbS and several other nanophases, many efforts have been made to investigate their size-dependent properties as shown, for example, in references 5-10. These properties may be quite different relative to their bulk equivalents. Specifically, the understanding of size-dependent dissolution of nanoparticles is one of the keys to assessing potential impacts on the environment and human health, as well as being very important for developing industrial applications of nanomaterials (11, 12) and essential for the development of nanoparticles in the field of drug delivery (13). A modified Kelvin equation (eq 1) has long been used to describe the thermodynamic prediction that solubility is expected to exponentially increase as the grain size decreases into the nanorange according to:
[ ]
S 2γVj ) exp S0 RTr
(1)
where S is the solubility (in mol kg-1) of fine grains with inscribed radius r in m, and S0 is the solubility of the bulk material. Vj is molecular volume in m3 mol-1, γ is the surface free energy in mJ m-2, R is the gas constant in nJ mol-1 K-1, and T is the temperature in K. However, so far few data from nanoparticle dissolution experiments have been reported to support this theoretical prediction (14), and in some cases sparingly soluble phases show dissolution suppression as crystallites approach the nanosize range (15). Moreover, previous studies measured the dissolution rates of nanoparticles that were aggregated to some extent in solution (11, 12). The aggregation state was not characterized in these studies, which makes it difficult to interpret data and evaluate the size effect on dissolution alone. Beyond the size effect, adding the variable of aggregation state is very important. In natural aquatic systems, a large portion of colloids and nanoparticles exist in aggregates (16). Aggregates that are smaller than a critical size could be stable and highly mobile in natural waters (17). Further, engineered nanoparticles that are produced and used as dry powders usually remain in an aggregated state even after being exposed to an aqueous environment (18). Aggregation is expected to change the rates of surface reactions on nanocrystals (19, 20). Yet, no previous experimental studies have been conducted 10.1021/es902121r CCC: $40.75
2009 American Chemical Society
Published on Web 10/01/2009
to compare the dissolution reactivity of aggregated and nonaggregated nanoparticles. Our previous work (5) reported the acidic, nonoxidative dissolution of isolated galena (PbS) nanocrystals, as well as a few PbS nanocrytals in close proximity to one another. The evolution of the size, shape, and chemistry of the nanocrystals before and after dissolution experiments were studied in detail. The goal of this study was to determine, under the same experimental conditions, whether the observed dissolution rate of aggregated nanoparticles is dominated by an inherent particle size effect or by the aggregation state.
Experimental Section Galena Synthesis. Synthesis and purification of galena nanocrystals were described in detail in Liu et al. (5) and are briefly described here. Galena nanocrystals were synthesized by the reaction of PbCl2 and elemental sulfur in oleylamine (OLA) under inert atmosphere (4). Synthetic galena nanocrystals are totally or partially coated by OLA ligands, which stabilize nanocrystals in hexanes. These capping groups were removed before any dissolution experiments by repeatedly washing nanocrystals using isopropyl alcohol. Uncoated nanocrystals can quickly aggregate in aqueous solution, and the aggregates can be collected by centrifugation. Galena microcrystals were synthesized by a hydrothermal method (21). All reagents were purchased from Sigma-Aldrich and used without further purification. The procedure is as follows: 1 mmol (CH3COO)2Pb and 4 mmol Na2S2O3 were dissolved in 9.2 mL of deionized water, respectively. The solutions were transferred into a 23 mL Parr acid digestion bomb with an inserted PTFE sample cup. Then, the system was closed tightly and heated in an oven to 150 °C for 20 h. Black precipitates were collected, washed with deionized water several times, and dried in air at 50 °C. Particle Characterization. The crystal phases of the synthetic nanocrystals and microcrystals were identified by a Pan Analytical X’PERT-Pro X-ray powder diffractometer, equipped with a Co KR source. Samples were mounted on a zero-background sample holder. Diffraction patterns were collected in the 2θ range of 20-100° using a step size of 0.067°. The size and morphology of primary nanocrystals and aggregates were observed by a Philips EM420 transmission electron microscope (TEM) operated in bright field mode at 100 keV. For primary nanocrystal analysis, samples were prepared by dipping a carbon-coated 400-mesh copper TEM grid into the nanocrystal suspension and then drying the grid in air. For nanocrystal aggregate analysis, aggregates were sonicated in hexanes until the suspension became transparent. A drop of this suspension was then applied to a TEM grid and allowed to dry. Dynamic light scattering (DLS) was used to measure the size of nanocrystal aggregates in hydrochloric acid solutions. All measurements were made with a Malvern Zetasizer 3000 HS. Suspensions for the DLS analysis were prepared by dispersing dry nanoparticle aggregates in HCl solution at pH 3 using an ultrasonicator (Misonix Sonicator 3000 Homogenizer, power 10) for 1 min. In DLS measurements, three repeats for a single acquisition were chosen to ensure that the noise of the data was less than 0.1%. The hydrodynamic size of aggregates was measured until diluting the suspension did not change the size of aggregates. The lead concentration in the suspension for DLS was about 0.30 ppm. The change of aggregate size as a function of time was monitored by DLS over 3 h. To complement DLS results, the size and degree of aggregation of nanoparticle aggregates were observed by high-resolution field emission scanning electron microscopy (FESEM) (LEO 1550, FEG at 5 kV). About 5 mg of nanocrystal aggregates were dispersed in ∼5 mL of HCl solution (pH )
3) using ultrasonication for about 1 min. A drop of this suspension was deposited onto a piece of boron-doped silicon substrate that was attached onto a SEM sample stub, and then dried in air at 40 °C. The nanocrystal aggregate samples were not coated before SEM measurements. In addition, SEM was also used to observe the size and morphology of galena microcrystals. Microcrystal samples were coated with a 5 nm Au film to increase sample conductivity. The focused ion beam (FIB) in-situ lift-out technique was employed to prepare the cross-sectional specimen of nanoparticle aggregates. The FIB work was performed using a Ga+ beam in a dual-beam workstation (FEI, Helios 600 NanoLab). Prior to the FIB lift-out, nanoparticle aggregates were dispersed in deionized water using ultrasonication. A drop of the suspension was applied onto a Si substrate, and then dried in air. A ∼3 µm thick rectangular platinum stripe was deposited over a randomly selected area of the aggregates using an electron ion beam with current density of 5 pA µm-2. A slice with dimensions of ∼20 µm × 10 µm × 1 µm was cut free from the site using a 30 kV, 100 pA ion beam, and then welded onto a TEM grid by depositing Pt. After that, the slice was thinned further using the ion beam at lower energies. The final FIB thinning was performed with a 5 kV, 80 pA ion beam ((7° with respective to the plane of the specimen surface) to remove amorphous layers on the surface caused by beam damage. These samples were observed using a FEI Titan 300 TEM, operating in highresolution mode at 200 kV. The particle size was determined by analyzing microscopic images via the image processing and analysis program Image J. Particle diameter (size) of nanoparticles measured by analyzing TEM images is defined as the diameter of a circle having the same projected area as the particles. To determine the size, or effective diameter (de), of a given microcrystal, the edge length (a) of crystals in SEM images was measured, and then converted to de by the equation de ) 2
a2 π
(2)
Because SEM images show microcrystals in three dimensions, only faces nearly parallel to the image were measured. The specific surface area (SSA) of nanoparticle aggregates and microcrystals was determined via the BrunauerEmmett-Teller (BET) method. Measurements were carried out on a TriStar II 3020 system (Micrometrics Analytical Services) by determining the nitrogen adsorption-desorption isotherms at 77 K. Before analysis, the samples were heated to 60 °C for 16 h at a rate of 10 °C/min under vacuum. The BET SSA of microcrystals is 0.285 ( 0.002 m2/g, and the value for nanoparticle aggregates is 11.71 ( 0.03 m2/g. The geometric SSA, Ageo, of microcrystals was calculated by assuming that particles are smooth cubes according to Ageo )
6 F · de
(3)
where F is the density of galena. Dissolution Experiments. Dissolution experiments were performed in a glass reactor (∼750 mL capacity) under constant mechanical stirring at 25 °C. Hydrochloric acid solution (495 mL; pH 3) was added to the chamber and purged with nitrogen for 30 min to remove dissolved O2. Nanoparticle aggregates were sonicated in 5 mL of HCl solution (pH 3). This suspension was then added to a batch reactor with deoxygenated HCl solution (pH 3). Nitrogen purging was maintained throughout the experiment and the pH was monitored constantly with a pH meter. Samples of 6 mL of solutions were taken after certain intervals and immediately filtered. For the dissolution of microcrystals, syringe filters VOL. 43, NO. 21, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
8179
FIGURE 1. Comparison of nanoscale and microscale galena particles. TEM image of galena nanocrystals and their size distribution (a, b, respectively), and SEM image of galena microcrystals and their size distribution (c, d, respectively). with a pore size of 0.45 µm were used. In the dissolution of nanocrystal aggregates, filters with two different pore sizes were used to evaluate the effect of filtration on the measured dissolution rate: a syringe filter with a pore size of 100 nm, and a centrifugal ultrafilter (Amicon Ultra-4, Millipore) with a cutoff of 100k NMWL (nominal molecular weight limit), approximately equivalent to a pore size of ∼6.2 nm. Pb concentrations of the sampled solutions were measured using an inductively coupled plasma atomic emission spectrometer (ICP-AES). The detection limit for Pb is 0.016 ppm.
Results and Discussion Nonoxidative dissolution of dispersed PbS galena nanoparticles in acidic solution has been reported in our previous paper on this system (5). The synthetic galena nanocrystals, as shown in Figure 1, are nearly monodisperse with an average diameter of 14.4 ( 1 nm (1σ) and a truncated-cubic shape (5). To eliminate aggregation as a factor in these dissolution experiments, widely spaced nanoparticles were attached on the carbon film of a TEM sample grid, and the dissolution rates were directly measured via monitoring the size change of nanoparticles using TEM. The geometric surface area (Ageo) normalized rate of dispersed galena nanoparticles in deoxygenated HCl solution (pH 3) was measured to be 4.4 × 10-9 mol m-2 s-1 (5). Size Effects. To assess the size effect on dissolution, galena cubic microcrystals were synthesized via a hydrothermal method (21). The representative scanning electron microscopy (SEM) image of the microcrystals (Figure 1c) shows that they are primarily cubic in shape with an average diameter of 3.1 ( 0.8 µm (1σ). XRD indicated that only the galena phase was present. Galena microcrystals were dissolved under the same conditions as the nanocrystals. Figure 2a shows the release rate of Pb2+ ions in deoxygenated HCl (pH 3) solution at 25 °C over 16 h. All data can be fit well using a linear function. The increasing rate of Pb2+ concentration, ∆CPb/∆t, determined from the slope of the line, is 0.01865 mg L-1 h-1. Assuming the 8180
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 21, 2009
FIGURE 2. Lead concentration vs. time in the dissolution of galena particles in HCl solution at pH 3 and 25 °C. (a) Galena microcrystals. The line is the linear fit of raw data. Its slope is (∆CPb/∆t) ) 0.01865 mg L-1 h-1. (b) Nanocrystals aggregates using 100 nm filter (9) and 6 nm filter (O), respectively. The slope of the solid line is (∆CPb/∆t) ) 0.4926 mg L-1 h-1; the slope of the dashed line is (∆CPb/∆t) ) 0.458 mg L-1 h-1. microcrystals are uniform smooth cubes, the geometric specific surface area (SSA) of microcrystals is 0.26 m2 g-1. The geometric surface area (Ageo) normalized dissolution rate can be calculated by
r)
( )
V ∆CPb S ∆t
(4)
where V is the volume of the acid solution, and S is the total surface area of galena particles. For 0.05 g of galena microcrystals dissolved in 0.4 L of deoxygenated HCl (pH 3) solution, the Ageo normalized dissolution rate is 7.7 × 10-10 mol m-2 s-1, which is 1 order of magnitude slower than the dissolution rate of dispersed nanocrystals (Table 1). Although the dissolution rates of dispersed nanoparticles and microcrystals were measured with two different techniques (monitoring size reduction or measuring the release of ions), these two measurements are directly correlated, and should not lead to any enormous difference in rates. The faster dissolution of galena nanoparticles agrees with the prediction of the modified Kevin equation (eq 1), in that the dissolution is more thermodynamically favored on nanocrystals relative to bulk materials. While in this study the galena nanoparticles are cubic and not spherical as generally assumed in the modified Kelvin equation, the nanoparticle size measured was the projected area diameter, defined as the diameter of a circle having the same projected area as the particles in TEM or SEM images. Therefore, the morphology does not affect the application of the Kelvin equation in this study. More specifically, the faster dissolution of galena nanoparticles can be attributed to a difference in surface reactivity as a function of size. Even with the same total surface area, and considering the roughness of typical galena crystal growth and cleavage surfaces as measured by scanning tunneling microscopy (STM) (22), nanocrystals have a much larger fraction of atoms at edges (or steps) and corners, and not on flat terraces, than larger particles. Such surface nanotopographic features are well-known as preferred detachment/ dissolution sites. Also, besides {100} faces which dominate the morphology of microcrystals, nanocrystals also present {110} and {111} faces which are more reactive to dissolution under these experimental conditions as determined by HRTEM analysis (5). Aggregation Effects. To assess the effect of aggregation on dissolution rate, we first formed and characterized the aggregates themselves. After removing the capping groups present during and after synthesis (5), the galena nanoparticles quickly aggregated in solution, and the resulting aggregates were collected by centrifugation. The size of aggregates in the suspension was measured by dynamic light scattering (DLS). The mean hydrodynamic diameter of aggregates in HCl at pH 3 is 240 nm, and this result was quite stable over the 3-h measurement. This measurement is consistent with the SEM images which show tightly packed aggregates in this size range, although these aggregates apparently clump together upon drying (Figure 3a). Clearly, it would not be possible to use a simple model to calculate the geometric SSA of these nanoparticle aggregates. BET measurements of the aggregates gave 11.71 ( 0.03 m2 g-1. To “look inside” the dense aggregates, TEM sections were prepared using the FIB in situ lift-out technique (Figure 3b). A representative HRTEM image of a cross-section (Figure 3c) shows how nanoparticles are arranged inside dense aggregates. Due to alignment of the cubic crystals, the interparticle spacing is exceptionally small; grain to grain contact is common, as well as interparticle spacings of 1 nm or less. The dissolution rate of aggregated galena nanoparticles was measured under the same conditions as dispersed microand nanocrystals. To investigate the effect of filtration on the measured dissolution rates, filters with two different pore sizes were separately used for sampling: a syringe filter with a pore size of 100 nm, and a centrifugal tube filter (Amicon
Ultra-4, Millipore) with a cutoff of 100k NMWL (nominal molecular weight limit) approximating a pore size of 6.2 nm. Figure 2b shows the concentration changes of dissolved Pb2+ ions as a function of time using these two kinds of filters. The slopes of the linear fits for the two sets of data are very similar, so there is not a significant number of primary nanoparticles or small aggregates that are released as a result of dissolution. Using ∆CPb/∆t ) 0.4926 mg L-1 h-1 and ABET ) 11.71 m2 g-1, the surface area-normalized dissolution rate is calculated to be 4.7 × 10-10 mol m-2 s-1. This rate is on the same order of magnitude as the rate of galena microcrystals, but 1 order of magnitude smaller than that of dispersed nanocrystals (Table 1). It is worth mentioning that BET measurement requires sample drying, and it is not necessarily the case that the aggregation state in solution is preserved. BET measurement is more likely to underestimate rather than overestimate the actual surface area in solution. As a result, the dissolution rate based on the BET results could be faster than the actual rate for the aggregated nanoparticles. Therefore, the difference between the dissolution rates of dispersed and aggregated nanoparticles could be even larger than what is reported in this study. The slow dissolution of nanoparticle aggregates is likely to be related to the dissolution inhibition in the highly confined space between the densely packed nanocrystals. Our previous HRTEM study of dispersed nanocrystalline galena (5) included our assessment of the dissolution of nanocrystals in very close proximity or in contact with one another. It was directly observed that surfaces adjacent to other nanoparticles show obvious dissolution retardation relative to the surfaces exposed to bulk solution. The dissolution inhibition in confined space is likely due to the structural characteristics and resulting chemical behavior of aqueous solutions in close proximity to surfaces. The general structural perturbations of water adjacent to mineral surfaces, as characterized by techniques such as nuclear magnetic resonance, optical sum-frequency generation, and highresolution X-ray scattering have been studied for some time (23, 24). The data show a general ordering of water molecules, the exact nature of which depends on the specific surface in question, extending out a few to several monolayers from the surface. It has even been shown experimentally that aqueous solution viscosities are larger in tightly confined space than in unconfined space (25-27). This alone, according to the Stokes-Einstein relation, results in a decrease in diffusion coefficients. Recent molecular dynamics simulations also revealed a 2.0-2.5 nm interfacial region (this for a feldspar surface) within which the self-diffusion coefficients of water and that of the electrolyte ions decrease dramatically as the diffusing species approach the surface (28). As shown in Figure 3c, the interparticle spacing in the nanoparticle aggregates is generally well within this range of a few nanometers. Thus, diffusion coefficients of water, as well as electrolyte and dissolving ions in the confined space inside nanoparticle aggregates, are likely to be considerably smaller than that in bulk solution. The dissolution rate constant, k, depends on the diffusion coefficient (D) of solute molecules as shown: k)
A×D V×h
(5)
where A is the surface area of solutes, V is the volume of solution, and h is the thickness of the diffusion layer (19). This relation indicates, as expected, that reduced D will lead to smaller k, i.e., slower dissolution. In addition, the transport of water and ions within tightly confined spaces is dominated by diffusion, not hydraulically driven water moving over surfaces and between grains (24). Therefore, the reduction of D is especially important VOL. 43, NO. 21, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
8181
TABLE 1. Size, Geometric SSA (Ageo), BET SSA (ABET), Geometric Surface Area Normalized Dissolution Rate (Rgeo), and BET Surface Area Normalized Dissolution Rate (RBET) of Primary Nanocrystals, Microcrystals, and Nanocrystal Aggregates
nanocrystals nanocrystal aggregate microcrystals
average diameter
Ageo (m2 g-1)
ABET (m2 g-1)
Rgeo (mol m-2 s-1)
RBET (mol m-2 s-1)
14.4 ( 1 nm 240 nm 3.1 ( 0.8 µm
55.0 s 0.26
s 11.71 ( 0.03 0.285 ( 0.002
4.4 × 10-9 s 7.7 × 10-10
s 4.7 × 10-10 7.0 × 10-10
for dissolution in confined spaces. Another factor to consider is the overlap of the diffusion layers on the surface of two adjacent nanoparticles. In any dissolution process, dissolved species must move from the particle surface to the bulk solution through a diffusion layer. The concentration gradient from the mineral surface to bulk solution is the driving force for dissolution. When two particles are close enough that their diffusion layers overlap, the concentration gradient between the mineral surfaces and the solution in the interparticle spacing becomes small, and the flux of solute away from the dissolving surface is reduced (24). An additional factor that may reduce effective diffusion rates is increased tortuosity within the aggregates (29). In other words, the nanoparticles act as geometric obstacles to the diffusion path of molecules and ions within the aggregate, which requires them to travel longer distances than they would through pure solution. Another way to look at the data presented in Table 1 is that the order of magnitude increase in surface area normalized dissolution rate for the dispersed nanocrystals relative to the microcrystals is lost as soon as the nanocrystals are allowed to aggregate. Roughly speaking, the 240 nm (average) aggregates each contain on the order of a few thousand galena nanocrystals, assuming a dense packing as observed with HRTEM. It is also important to note that the BET-measured SSA of the nanocrystal aggregates is only about 20% of the dispersed nanocrystals (Table 1), clearly supporting
the HRTEM images that show considerable grain-to-grain contact in the aggregates with very low porosity. It is expected that nanocrystals in the interior of the aggregates will not participate in the dissolution reactions over 10-15 h (the experiment duration in this study) due to the transport retardation of aqueous species in highly confined spaces as described above. Therefore, it is important to estimate the surface area of the galena nanocrystals that line the surface of a typical aggregate. One can think of this as the top monolayer of particles on the surface, about 14 nm thick, the average size of the nanoparticles. For aggregates in the size range observed, it can be geometrically estimated that roughly 30% of the total surface area available in a dispersed system with this number of particles would be in the first particle layer, but crystals could also be packed tightly at the surface. The surface area available on the aggregate surface must be less than 20% as implied by the overall surface areas reported in Table 1. Also, as discussed above, the dissolution rate is only 10% of the rate of dispersed nanoparticles (Table 1), probably because surfaces in close proximity to other surfaces on the surface of the aggregates have very sluggish dissolution kinetics, a conclusion suggested by studying the dissolution behavior of two or three closely adjacent nanocrystals ((5); Figure 3d). Implications for Predicting Environmental Impacts. This case study shows how critical subtle details are in determining the reactive properties of small particles as a function of size
FIGURE 3. (a) Representative SEM of galena nanoparticle aggregates averaging 240 nm in diameter that then clump together upon drying. (b) A cross-section of the aggregates being prepared by the FIB lift-out technique. (c) HRTEM image of a cross-section of a nanoparticle aggregate. White dashed lines define the edges of some grains. Observed lattice fringe spacings match galena crystallographic repeats. The spotty texture to the left and top is from the Pt coating deposited in the FIB. (d) While most of the nanocrystals in our previous study (5) were completely isolated, some were adjacent to each other, allowing for observations of the impact of proximity upon dissolution. This is a HRTEM image of a galena nanocrystal (P1) with adjacent crystals to the right (P2, P3) and open space to the left (5). {110} faces, denoted by white lines, dissolve much more rapidly (and therefore are larger) on the unrestricted side of the crystal. 8182
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 21, 2009
and aggregation state. It has been suggested before, for example in the case of the photochemical production of reactive oxygen species (ROS) by fullerene suspensions in water (30), that the exact nature of the aggregated arrangement can play the key role in the reactivity of the nanoparticles. The aggregation characteristics could affect the production and transportation of contamination or reactive species, like ROS by fullerenes, in aggregated nanomaterials. Therefore, aggregation state should be considered as an important factor when considering the environmental impact of nanomaterials. In the present study, HRTEM combined with dissolution rate measurements gives detailed answers as to the exact relationship among particle size, shape, and aggregation state in the dissolution process. More specifically in this case, these details dictate the release rate of aqueous, highly toxic and bioavailable Pb in natural or industrial environments during galena dissolution. Industrial nanotechnology, with thousands of nanomaterials already here or on the horizon, will soon represent trillions of dollars in investments and sales within the global economy (3). At the same time, the Earth has evolved, and will continue to evolve in the presence of a strikingly large number of naturally occurring nanomaterials present in massive amounts (31). The latter case tells us that their environmental impact has already been fundamentally defining in how the Earth naturally functions (2). In the former case, their environmental impact is imminent, both with positive and negative, as well as intended and unintended, consequences. Understanding reaction characteristics and trends, as shown in this study, will play a key role in predicting environmental impacts in the future.
Acknowledgments Grants from the U.S. Department of Energy (DE-FG0206ER15786) and the Institute for Critical Technology and Applied Science at Virginia Tech provided major financial support for this study. We are also appreciative of the support from the National Science Foundation and the Environmental Protection Agency under NSF Cooperative Agreement EF0830093, Center for the Environmental Implications of Nanotechnology. D.M.A. acknowledges support from the NSF under a Minority Postdoctoral Research Fellowship, award 0610373. Important assistance from John McIntosh at the Nanoscale Characterization and Fabrication Laboratory at Virginia Tech is gratefully acknowledged.
Literature Cited (1) Review of Federal Strategy for Nanotechnology-Related Environmental, Health, and Safety Research; The National Academies Press: Washington, DC, 2008. (2) Hochella, M. F., Jr.; Lower, S. K.; Maurice, P. A.; Penn, R. L.; Sahai, N.; Sparks, D. L.; Twining, B. S. Nanominerals, mineral nanoparticles, and Earth systems. Science 2008, 319 (5870), 1631– 1635. (3) Wiesner, M. R.; Gregory, V. L.; Jones, K.; Hochella, M. F.; Di Giulio, R. T.; Casman, E.; Bernhardt, E. Decreasing uncertainties in assessing environmental exposure, risk and ecological implications of nanomaterials. Environ. Sci. Technol. 2009, 43 (17), 6458–6462. (4) Joo, J.; Na, H. B.; Yu, T.; Yu, J. H.; Kim, Y. W.; Wu, F. X.; Zhang, J. Z.; Hyeon, T. Generalized and facile synthesis of semiconducting metal sulfide nanocrystals. J. Am. Chem. Soc. 2003, 125 (36), 11100–11105. (5) Liu, J.; Aruguete, D. M.; Jinschek, J. R.; Rimstidt, J. D.; Hochella, M. F., Jr. The non-oxidative dissolution of galena nanocrystals: Insights into mineral dissolution rates as a function of grain size, shape, and aggregation state. Geochim. Cosmochim. Acta 2008, 72 (24), 5984–5996. (6) Madden, A. S.; Hochella, M. F., Jr. A test of geochemical reactivity as a function of mineral size: Manganese oxidation promoted by hematite nanoparticles. Geochim. Cosmochim. Acta 2005, 69 (2), 389–398.
(7) Chernyshova, I. V.; Hochella, M. F., Jr.; Madden, A. S. Size-dependent structural transformations of hematite nanoparticles. 1. Phase transition. Phys. Chem. Chem. Phys. 2007, 9 (14), 1736–1750. (8) Vikesland, P. J.; Heathcock, A. M.; Rebodos, R. L.; Makus, K. E. Particle size and aggregation effects on magnetite reactivity toward carbon tetrachloride. Environ. Sci. Technol. 2007, 41 (15), 5277–5283. (9) Zhang, H. Z.; Gilbert, B.; Huang, F.; Banfield, J. F. Water-driven structure transformation in nanoparticles at room temperature. Nature 2003, 424 (6952), 1025–1029. (10) Waychunas, G. A.; Zhang, H. Z. Structure, Chemistry, and Properties of Mineral Nanoparticles. Elements 2008, 4 (6), 381–387. (11) Vogelsberger, W.; Schmidt, J.; Roelofs, F. Dissolution kinetics of oxidic nanoparticles: The observation of an unusual behaviour. Colloids Surf., A 2008, 324 (1-3), 51–57. (12) Meulenkamp, E. A. Size dependence of the dissolution of ZnO nanoparticles. J. Phys. Chem. B 1998, 102 (40), 7764–7769. (13) Muller, R. H.; Jacobs, C.; Kayser, O. Nanosuspensions as particulate drug formulations in therapy. Rationale for development and what we can expect for the future. Adv. Drug Delivery Rev. 2001, 47 (1), 3–19. (14) Erbs, J. J.; Gilbert, B.; Penn, R. L. Influence of size on reductive dissolution of six-line ferrihydrite. J. Phys. Chem. C 2008, 112 (32), 12127–12133. (15) Tang, R. K.; Wang, L. J.; Nancollas, G. H. Size-effects in the dissolution of hydroxyapatite: an understanding of biological demineralization. J. Mater. Chem. 2004, 14 (14), 2341–2346. (16) Lead, J. R.; Wilkinson, K. J. Aquatic colloids and nanoparticles: Current knowledge and future trends. Environ. Chem. 2006, 3 (3), 159–171. (17) Gilbert, B.; Lu, G. P.; Kim, C. S. Stable cluster formation in aqueous suspensions of iron oxyhydroxide nanoparticles. J. Colloid Interface Sci. 2007, 313 (1), 152–159. (18) Zhang, Y.; Chen, Y. S.; Westerhoff, P.; Hristovski, K.; Crittenden, J. C. Stability of commercial metal oxide nanoparticles in water. Water Res. 2008, 42 (8-9), 2204–2212. (19) Borm, P.; Klaessig, F. C.; Landry, T. D.; Moudgil, B.; Pauluhn, J.; Thomas, K.; Trottier, R.; Wood, S. Research strategies for safety evaluation of nanomaterials, Part V: Role of dissolution in biological fate and effects of nanoscale particles. Toxicol. Sci. 2006, 90 (1), 23–32. (20) Cwiertny, D. M.; Handler, R. M.; Schaefer, M. V.; Grassian, V. H.; Scherer, M. M. Interpreting nanoscale size-effects in aggregated Fe-oxide suspensions: reaction of Fe(II) with goethite. Geochim. Cosmochim. Acta 2008, 72 (5), 1365–1380. (21) Ni, Y. H.; Wei, X. W.; Hong, H. M.; Ma, X. Hydrothermal preparation of PbS crystals and shape evolution. Mater. Res. Bull. 2007, 42 (1), 17–26. (22) Eggleston, C. M.; Hochella, M. F., Jr. Scanning tunneling microscopy of galena(100) surface oxidation and sorption of aqueous gold. Science 1991, 254 (5034), 983–986. (23) Fenter, P.; Sturchio, N. C. Mineral-water interfacial structures revealed by synchrotron X-ray scattering. Prog. Surf. Sci. 2004, 77 (5-8), 171–258. (24) Hochella, M. F., Jr.; Banfield, J. F. Chemical weathering of silicates in nature: a microscopic perspective with theoretical considerations. In Reviews in Mineralogy; White, A. F., Brantley S. L. Eds.; Mineralogical Society of America: Washington, DC, 1995; Vol. 31, pp 354-406. (25) Derjaguin, B. V.; Churaev, N. V. Properties of Water Layers Adjacent to Interfaces; Wiley: New York, 1986; pp 663-738. (26) Hibara, A.; Saito, T.; Kim, H. B.; Tokeshi, M.; Ooi, T.; Nakao, M.; Kitamori, T. Nanochannels on a fused-silica microchip and liquid properties investigation by time-resolved fluorescence measurements. Anal. Chem. 2002, 74 (24), 6170–6176. (27) Tsukahara, T.; Hibara, A.; Ikeda, Y.; Kitamori, T. NMR study of water molecules confined in extended nanospaces. Angew. Chem., Int. Ed. 2007, 46 (7), 1180–1183. (28) Kerisit, S.; Liu, C. X. Molecular simulations of water and ion diffusion in nanosized mineral fractures. Environ. Sci. Technol. 2009, 43 (3), 777–782. (29) Moldrup, P.; Olesen, T.; Komatsu, T.; Schjonning, P.; Rolston, D. E. Tortuosity, diffusivity, and permeability in the soil liquid and gaseous phases. Soil Sci. Soc. Am. J. 2001, 65 (3), 613–623. (30) Hotze, E. M.; Labille, J.; Alvarez, P.; Wiesner, M. R. Mechanisms of photochemistry and reactive oxygen production by fullerene suspensions in water. Environ. Sci. Technol. 2008, 42 (11), 4175– 4180. (31) Hochella, M. F., Jr. Nanogeoscience: From Origins to CuttingEdge Applications. Elements 2008, 4 (6), 373–379.
ES902121R
VOL. 43, NO. 21, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
8183
