Chemical weathering, that is, dissolution of Ag minerals is one of major biogeochemical processes that control [Ag(I)] in soil solutions and pore waters. In the natural system, the dissolution process is generally slow and is controlled by (1) mass transfer process of reactants/products and/or (2) surface processes associated with the detachment of reaction products (Stumm and Morgan, 1996). If the rate-determining step is controlled by transport processes through layers of minerals, the dissolution process is often defined by the parabolic rate law. Alternatively, when the surface reactions are much slower than the mass transfer of reactants and reaction products, we often assume that the reaction is controlled at the mineral surfaces. Because of the slow steps, there will be a uniform concentration gradient of solutes at the mineral–water interface. The dissolution kinetics often follows a zero-order rate law when we assume the system has reached at steady state.
In the case of Ag minerals, including emerging AgNPs, the dissolution process is facilitated or retarded by the following mechanisms: (1) the mass transport of protons or ligands from the bulk solution to mineral surfaces, (2) surface complexation of ligands, and (3) electron transfer reactions (e.g., oxidative dissolution of Ag(0) nanoparticles).
Several researchers have investigated the proton-promoted dissolution of AgNPs. In all studies, pH-dependent dissolution of AgNPs was reported.
Elzey and Grassian (2010) studied the pH-dependent dissolution of uncapped 10-nm manufactured AgNPs (Nanostructure and Amorphous Materials, Inc., Houston, Texas, USA). They reported that nearly 95% of total Ag was dissolved in nitric acid solutions at pH 0.5. The dissolution, however, was largely reduced from 95% to 1.2% with increasing pH from 0.5 to 3.5. This suggests that AgNPs are insoluble at environmentally relevant pH values (e.g., 5–7.5) (Elzey and Grassian, 2010). Stebounova et al. (2011)showed additional evidence of AgNP solubility. The dissolu- tion of these AgNPs in artificially created intestinal (Gamble’s solution, pH 7.4) and lysosomal fluids (pH 4.5) was minimum. Less than 0.1% of total Ag was dissolved in both artificial fluids after 24 h at 38C (Stebounovaet al.,
2011). While the pH-dependent dissolution process is clear, some research- ers have pointed out the effects of particle size on the rate of AgNP dissolution.Liuet al. (2010)reported that the dissolution of citrate-capped AgNPs (mass ratio of citrate:Ag ẳ3:1). The batch experiments were con- ducted in acetate buffer solutions at pH 4 under air-equilibrated condition.
The rate of dissolution was proportional to particle size. The first-order dissolution rate constant increased from 0.78 to 4.1 day1with decreasing particle size from 60 to 4.8 nm (Liuet al., 2010).
The dissolution of AgNPs is not only promoted by the activity of hydrogen ions but also by the presence of soft base ligands such as cyanide.
The dissolution of a silver sulfide mineral, Ag2S, is described in the follow- ing reaction (Xie and Dreisinger, 2007).
Ag2Sỵ4CN!2AgðCNị2ỵS2 DGo ẳ52:5 kJ mol1 ð6ị However, under aerobic conditions, sulfide can readily oxidize to thio- cyanate (SCN), making the dissolution of Ag2S less effective. The rate of dissolution in the aerated cyanide solution is reported to be 0.38mmol h1. The modified reaction can be expressed as follows.
2Ag2Sð ịs ỵ10CNỵ2H2OỵO2!4AgðCNị2ỵ2SCNỵ4OH DGoẳ 388:42 kJ mol1 ð7ị
The kinetic experiments indicated that both surface- and diffusion- controlled dissolution reactions were occurring during the reaction (Lun˜a- Sanchezet al., 2003). In another Ag2S dissolution study using a mixture of ferricyanide–cyanide ligand, an activation energy of 6.7 kJ mol1 was reported, suggesting a diffusion-controlled reaction (Xie and Dreisinger, 2007). Although the cyanide-promoted dissolution reaction is thermody- namically favorable, other soft ligands are known to suppress the dissolution of Ag minerals as well. Humic acids can suppress the dissolution of Ag2S as much as 75%. Cysteine and thiosulfate almost completely inhibited the dissolution of Ag minerals during long-term (22 days) dissolution experi- ment (pH 3.5–5) (Jacobsonet al., 2005b). Similar results have been reported by Liu et al. (2010). The dissolution of 4.8 nm AgNPs, which were pre- treated with 0.4 mmol l1Na2S and 4 mmol l111-mercaptoundecanoic acid, was nearly negligible in the pH 5.6 acetate buffer solution.
Oxidative dissolution of AgNPs was recently studied by several researchers. In air-equilibrated distilled water at pH 5.68, dissolution of citrate-capped AgNPs (4.8 nm) was as high as 0.3 mg l1after 1 day, while there was negligible dissolution of AgNPs in deoxygenated solution. This suggests the following oxidative dissolution reaction of AgNPs.
2Agð ịs ỵ0:5O2 aqð ịỵ2Hỵ!2AgỵỵH2Oð ịaq DGoẳ 91:3 kJ mol1 ð8ị Ho et al. (2010) investigated the oxidative dissolution of AgNPs (5–
10 nm synthetic, citrate capped) by H2O2. The rate of dissolution (pH 7.4, I ẳ0.1 mol l1) was proportional to the concentration of H2O2, with a maximum rate of 0.139 s1in an acetate buffer solution at pH 7.4. The rate was positively linearly correlated with both particle size and temperature.
The chemically controlled rate-limiting step was supported by an activation energy of 35.1 kJ mol1. Interestingly, they showed that the rate of dissolution was not dependent on ionic strength, suggesting that it is predominantly uncharged particles involved in the rate-limiting step. Dis- solution experiments in the presence of ligands (i.e., 0–10 mmol l1 PVP and 0.05–5 mmol l1chloride) show a decrease in the rate, suggesting that sorption of ligands on AgNP surfaces perturbed electron transfer reactions.
We also conducted the dissolution of three manufactured AgNPs in oxic and reduced buffer solutions (0.005 mol l1 Na2SO4 plus 0.02 mol l1 sodium acetate at pH ẳ5 0.05 and [Ag]total500 mg l1). Nanoparticles include uncapped 50-nm particles (Inframat Advanced Materials), PVP- capped 20-nm particles (0.3% PVP by weight; Nanostructured & Amor- phous Materials, Inc.), and PVP-capped 15-nm particles (90% PVP by weight; Nanostructured & Amorphous Materials, Inc.). These AgNPs are abbreviated as Ag50, Ag20, and pAg15, respectively, and their physico- chemical properties are summarized inTable 4.
The results of the dissolution experiments are summarized inFig. 5. The pAg15 sample displayed the highest degree of Ag(I) release/dissolution of all the AgNP samples (Fig. 5C). It reached a maximum [Agþ] of 5.3% of AgTotal(26 mg l1) after 10 days under aerobic conditions, and a maximum [Agþ] of 6.3% of AgTotal(32 mg l1) after 1 day under anaerobic conditions.
This is substantially higher than the maximum [Agþ] release of Ag20: 0.32%
of AgTotal (1.5 mg l1) after 30 days (aerobic) and 0.50% of AgTotal (2.5 mg l1) after 1 day (anaerobic) (Fig. 5B) or the maximum [Agþ] release of Ag50: 0.60% of AgTotal(3.0 mg l1) after 15 days under aerobic condi- tions and 0.35% of AgTotal(1.7 mg l1) after 1 day under anaerobic condi- tions (Fig. 5C). Overall, all samples had an initial dissolution of at least 1 mg Agþl1after day 1. While dissolution under aerobic conditions remained fairly constant over time (filled black squares in Fig. 5A–C), anaerobic dissolution did not follow the same trend. All AgNPs had the highest Agþ release within the first 24 h. As time went on, the AgNPs showed a decreased release of Agþ. For both Ag50 and Ag20, <1% of AgTotal was released as Ag(I) under both aerobic and anaerobic conditions; pAg15, however, showed as much as 6.3% release of AgTotal as Ag(I). The large amount of PVP coating in pAg15 (90% by weight) increases the AgNP
affinity for water, which could lead to an enhanced oxidative dissolution.
One should expect enhanced oxidative dissolution under aerobic environ- ments, compared to anaerobic environments. However, this was not
0 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0
5 10
Time (days) Percentage of AgTotal
15 20 25 30
0 0.0 0.1 0.2 0.3 0.4 0.5 0.6
5 10 Percentage of AgTotal
15 20 25 30
0 0.20 0.25 0.30 0.35 0.40 0.45 0.50 0.55 0.60
5 10 Percentage of AgTotal
15 20 25 30
Aerobic Anaerobic
B A
C
Figure 5 Batch dissolution data of three manufactured nanoparticles under aerobic and anaerobic environmental conditions: (A) Ag50, (B) Ag20, and (C) pAg15. Total Agẳ500 mg l1. Dissolution solution composition is 0.005 mol l1 Na2SO4, pH buffered using 0.02 mol l1sodium acetate at 4.50.2. Descriptions of AgNPs are found inTable 4.
observed. This could be due to air oxidation of the powder AgNPs over time. Initially, high [Agþ], followed by a decrease over time in anaerobic samples (especially Ag20 and pAg15), could be due to re-adsorption of Ag(I) onto the AgNPs as they persist in an anaerobic environment.