Platinum, palladium and rhodium release from powder particles exposed to simulated lung fluids
Claudia Colombo*, A. John Monhemius, Jane A. Plant
Department of Earth Science and Engineering, Imperial College London, London SW7 2AZ, UK
*Corresponding author. Tel.: +44 (0)207 759 41362; Fax: +44 (0)207 594 7444 ; E-mail address:
Abstract
Increases in platinum-group element (PGE) concentrations in ambient air and dust since the introduction of vehicle exhaust catalysts (VECs) is a cause of concerns. To assess the risks associated with the inhalation of VEC-emitted PGEs, extraction experiments on PGE-containing materials (i.e. road dust, auto catalyst and hydroxide samples) exposed to simulated lung fluids were performed. The aim of this study was to measure the release of PGEs in an inhalation scenario. Two synthetic lung fluids were used as surrogates for different areas of potential exposure in the human lungs: Gamble’s solution is representative of the interstitial fluid of the deep lung and artificial lysosomal fluid (ALF) is representative of the more acidic environment following phagocytosis by alveolar and interstitial macrophages within the lung. The results showed a higher PGE release for samples exposed to the acidic ALF compared to the neutral Gamble’s solution, implying that the inhaled particles would have to be phagocytized before a considerable solubilization of PGEs took place. The greatest percentage of released PGEs (up to 88%) was associated with road dust samples, most likely because of the transformations undertaken by these metals once deposited from VECs in the environment (i.e. formation of soluble and mobile species through complexation by inorganic and organic ligands common in the roadside environment). Pd and Rh showed higher % of metal release compared with Pt, probably due to differences in their mobilities and tendencies to form soluble complexes. The highest absolute bioavailability is observed for Pt however, because of its greater concentration in environmental samples. The elevated solubilization of Pt, Pd and Rh observed in the respiratory tract could involve the formation of PGE-chloride complexes. This observation merits particular attention considering the powerful toxic and allergenic effects of PGE-chloride salts on human beings and living organisms and the ever-increasing levels of PGEs in the environment due to VEC emissions.
Keywords: Platinum; palladium; rhodium; environment; lung fluids; health risk.
1. Introduction
Human exposure to particulate matter originating from traffic is finding increasing interest and the widespread use of platinum, palladium and rhodium in vehicle exhaust catalysts (VECs) has stimulated various scientists to study the concentration of these metals in airborne samples at different urban sites and working places {Gomez, 2001 #62}{Alt, 1993 #100}{Zereini, 2001 #113}{Bocca, 2004 #60}. Platinum group elements (PGEs) are emitted in the exhaust gas from VECs with emission rates in the ng/km range and their concentrations in urban air have increased by more than two orders of magnitude in the last 20 years (up to approximately 110 pg/m3 {Merget, 2001 #72}). PGE emissions are thought to be mainly as fine particulate material that originates from the abrasion and deterioration of the surface of the catalyst. The annual Pt emission from VECs has been estimated at 0.5–1.4 ton per year {Barbante, 2001 #102}. Pt, Pd and Rh are deposited along roadways, on adjacent vegetation and soil, may be dispersed in silt, water and plants. Although PGEs are mostly released in metallic and oxide form {Konig, 1992 #56;Artelt, 2000 #170;Schlogl, 1987 #188}, there is evidence that, following their deposition in the environment, they can be transformed into soluble species by complexation with ligands commonly found in nature, providing compounds able to enter the food chain {Alt, 1993 #100}{Freiesleben, 1993 #133}{Lustig, 1996 #68}{Lustig, 1998 #131}{Zimmermann, 2003 #85}{colombo, #246}{Wood, 2004 #138}.
Because of the toxic, cytotoxic and allergenic effects of some PGE species (especially the chloride ones) and the increasing concentrations of VEC-emitted Pt, Pd and Rh in the environment, there is growing concern over human health and environmental risks resulting from PGE emissions. One of the major concerns for exposure to VEC-emitted PGEs is sensitization of the airways caused by soluble PGE compounds {Merget, 2000 #212}{Cristaudo, 2005 #162}. Many PGE compounds are known as potent respiratory allergens leading to rhinitis, conjunctivitis, asthma and urticaria {WHO, 1991 #189}{WHO, 2002 #190}. Signs of acute toxicity of several PGE salts (such as H2[PtCl6], RhCl3, and PdCl2) in rats or rabbits included deaths, decrease in feed and water uptake, emaciation, cardiovascular effects, peritonitis and biochemical changes. There were also hemorrhages of lungs and the small intestine. Moreover interactions with DNA were observed in vitro and an inhibition of DNA synthesis were demonstrated both in vitro and in vivo {WHO, 2002 #190}. A study on cell viability and oxidative stress in human bronchial epithelial cells {schmid, 2007 #292} showed that Pt and Pd had considerable toxic effects that are comparable if not even exceeding the damage induced by heavy metal species such as Cd and Cr. The mode of action of PGE ions and of elemental PGEs as dispersed dust in biological systems is not fully clear. Similar to other transition metals, PGE ions follow a few basic principles in their mode of action. Due to their ability to form strong complexes with both organic and inorganic ligands, they have the potential not only to disturb cellular equilibria or replace other essential ions but also to interact with functional groups of macromolecules, such as proteins, enzymes and DNA/RNA, thereby disrupting a variety of cellular processes. From occupational studies conducted in VEC production, a no-effect level “critical range” of 15-150 ng/m3 has been suggested {Merget, 2001 #72}. Higher concentrations could potentially induce hypersensitivity reactions if halogenated PGE salts are present in VEC emissions or are formed through transformation processes in the environment.
Very little information is available on the bioavailability of VEC-emitted PGEs after inhalation. A study by Artelt et al. {Artelt, 1999 #54} provided bioavailability measurements of a model platinum substance, similar to that emitted by VECs. This substance was applied to Lewis rats during a 90 day inhalation study. The study showed that the percentage of bioavailable platinum was up to 30%. The authors concluded that the elevated in vivo solubility was most likely due to the ultrafine structure of the platinum particles.
Data on PGE release into lung fluids relevant for human exposure are non-existent in the literature. The inhaled particles may interact with cells and tissues in the lung compartment in many different fashions. The mechanisms of these interactions include the release of PGE species that may have a particular effect on the human body. Reliable information on PGE-release from the inhaled particles in the respiratory tract is crucial when assessing the potential risk for adverse effects arising from exposure to these substances. The solubility of metal compounds in simulated body fluids is commonly used for determining human health risk from exposure to specific substances of concerns {midander, 2007 #277}{herting, 2006 #275}{herting, 2007 #276}{shi, 2006 #274}{stopford, 2003 #273}. Therefore, in this study, PGE-containing materials, such as road dust and milled recycled catalyst have been exposed to various simulated lung fluids to provide quantitative estimates of Pt, Pd and Rh release in an inhalation scenario representative of a human being. Synthetic hydroxide species have also been included serving as reference materials for comparison of data. Artificial lysosomal fluid (ALF) and Gamble’s solution have been used to simulate different interstitial conditions in the lung. ALF is analogous to the fluid with which inhaled particles would come into contact after phagocytosis by alveolar and interstitial macrophages within the lung. Gamble’s solution represents the interstitial fluid deep within the lung. These fluids have previously been used to investigate the bioaccessibility of cobalt compounds {stopford, 2003 #273}, the copper release from powder particles {midander, 2007 #277} and metal release from stainless steel {herting, 2007 #276}.
2. Experimental procedure
2.1. Test materials
Substrates evaluated for PGE release in the lung fluid experiments included two
-Certified Reference Materials (CRM) representative of (i) road dust with a maximum particle size of 90 µm, from the Institute of Reference Materials and Measurements, and (ii) powdered auto catalyst with a maximum particle size of 74 µm, from the National Institute of Standards and Technology (Table 1),
- Synthetic PGE-hydroxide species Pt(OH)2, Pd(OH)2 and Rh(OH)3prepared from the appropriate nitrate complexes (Johnson-Matthey) by precipitation with NaOH, were also tested.
Test material / Pt / Pd / RhRoad dust (BCR-723) / 81.3 µg/kg / 6.0 µg/kg / 12.8 µg/kg
Recycled catalyst (SRM-2557) / 1131 mg/kg / 233.2 mg/kg / 135.1 mg/kg
Table 1: PGE content of Certified Reference Materials employed in the lung fluid experiments.
1g road dust, 1g catalyst and 0.1g hydroxide sample (composed of 0.033g Pt(OH)2, 0.033g Pd(OH)2 and 0.033g Rh(OH)3) were employed in the lung fluid experiments.
2.2. Composition of simulated lung fluids
The chemical composition of each ALF and Gamble’s solution is presented in Table 2.
Chemicals (g/L) / ALFpH = 4.5 / Gamble’s solution
pH = 7.4
MgCl2 / 0.050 / 0.095
NaCl / 3.21 / 6.019
KCl / — / 0.298
Na2HPO4 / 0.071 / 0.126
Na2SO4 / 0.039 / 0.063
CaCl2·2H2O / 0.128 / 0.368
C2H3O2Na / — / 0.574
NaHCO3 / — / 2.604
C6H5Na3O7·2H2O / 0.077 / 0.097
NaOH / 6.00 / —
C6H8O7 / 20.8 / —
H2NCH2COOH / 0.059 / —
C4H4O6Na2·2H2O / 0.090 / —
C3H5NaO3 / 0.085 / —
C3H3O3Na / 0.086 / —
Table 2: Composition (g/L) of extraction fluids {midander, 2007 #277}{stopford, 2003 #273}.
The components of Gamble’s solution were added following the order in Table 2 to avoid precipitation of salts. Citrate was used in place of proteins and acetate to replace organic acids. All chemicals were of analytical grades and both solutions were prepared using ultra-pure water. Except for differences in salt composition, the main difference between Gamble’s solution and ALF is the acidity: the former has pH = 7.4 whereas the latter has pH = 4.5. The ALF acidity may explain the higher level of solubility for metals that are phagocytized compared to those that remain extracellular {Costa, 1995 #278}.
2.3. Extraction method
The extraction experiments were performed by mixing the test material (1.0 g for road dust and recycled catalyst or 0.1 g for the hydroxide species) with 100mL of simulated lung fluid. The experiments were carried out in sealed containers to minimize interactions between the reaction fluid and atmospheric oxygen and the potential for cross contamination. Each experiment was conducted in duplicate and blanks (containing only simulated lung fluids) were used.Each sample was placed under dark conditions in a shaking incubator keeping the temperature at 37°C. Extracts were filtered through a 0.2 µm filter to avoid any solid carry over. Different extraction periods (up to 30 days) were investigated. The pH of each sample was controlled at every extraction time and no variation superior to 0.2 units was observed.
2.4. PGE analysis
All the extracts were analyzed for PGE content by inductively coupled plasma mass spectrometry (ICP-MS), using an ICP-MS Varian model 810. The operating conditions are given in Table 3.
PlasmaPlasma flow: 18.00 L/min
Auxiliary flow: 1.80 L/min
Sheath gas flow: 0.30 L/min
Nebulizer flow: 1.07 L/min
Sampling depth: 5.00 mm
Power: 1.40 kW
Pump rate: 5 rpm
Stabilization delay: 10 sec
Analysis Modes
Analysis type: Quantitative
Acquisition mode: Steady state
Scan mode: Peak hopping
Dwell time: 10000 µsec
Replicates/sample: 3
Scans/replicates: 30
Analytes: 63Cu, 65Cu, 87Rb, 87Sr, 88Sr, 89Y,
103Rh, 105Pd, 179Hf, 195Pt, 206Pb
Table 3: ICP-MS operating conditions.
ICP-MS is often used for the determination of PGEs in environmental samples {Barefoot, 1997 #26;Ek, 2004 #33;Ravindra, 2004 #57} because of its excellent detection limits that allow accurate determination of PGEs in the ppt range. The determination of PGEs by ICP-MS is, however, hampered by spectral interferences from monatomic and polyatomic ions produced in the argon plasma by the matrix constituents: elements such as Cu, Y, Hf, Sr, Rb and Pb from road dust and catalyst samples cause interferences with the PGE signals (Table 4).
Analyte / Possible interfering species195Pt
105Pd
103Rh / 179Hf16O+
40Ar65Cu+ 89Y16O+ 88Sr17O+ 87Rb18O+
40Ar63Cu+ 87Sr16O+ 87Rb16O+ 206Pb2+
Table 4: Possible interferences in the ICP-MS determination of 195Pt, 105Pd and 103Rh.
All the experimental solutions were prepared daily in 2% HNO3 and matrix-matched using simulated lung solutions. PGE calibration standards were prepared by dilution of stock solutions of 1000 mg/L of Pt, Pd and Rh (all from Alfa Aesar, Johnson Matthey). The linearity of the calibration plots (correlation coefficient > 0.99) was checked for Pt, Pd and Rh in the concentration range 0.1 – 100 ng/L and 0.1 – 100 µg/L. To compensate for instrumental drifts, 10 µg/L of internal standards (ISs) were added to all samples. 115In for 105Pd and 103Rh and 191Ir for 195Pt were selected as ISs because of their low abundance in the sample matrix, similar atomic mass and chemical and physical properties to the PGEs. For the interference studies, single element stock solutions (1000 mg/L) of Cu, Rb, Sr, Y, Hf and Pb (all from Alfa Aesar, Johnson Matthey) were used.
3. Results
3.1. Interference correction
The isotopes 195Pt, 105Pd and 104Rh were used to determine the concentration of Pt, Pd and Rh, respectively, in the samples because of their high natural abundance and because they are subject to fewer interferences; possible interferences are reported in Table 4.
Interferences were corrected mathematically by estimating the contribution of interfering species to the PGE signal through the analysis of single element standard solutions of interferents {Rauch, 2001 #47;Moldovan, 1999 #25;Gomez, 2000 #63}.The concentrations of the interfering elements in the standard solutions were at similar levels as those in the sample solutions. Corrections were calculated using the following equation:
SA = SA,S – (SI,SRIO)
where SA is the corrected analyte (A) signal, SA,S is the interfered analyte signal measured for the sample solution, SI,S is the signal for the interfering element (I) in the sample solution, and RIO is the signal ratio IO+/I+ (ratio between the concentration of the ions produced in the plasma that interfere on the PGE signal and the concentration of the interfering element itself, i.e. 40Ar63Cu+/63Cu+). The signal ratio RIO for all the interfering species needed in the correction equations were measured daily. This mathematical correction method requires a linear dependence of the IO+ signal on the I+ concentration. In this study, a linear dependence was observed for all the interfering elements.
Considering all the interferences, the corrections for Pt, Pd and Rh were:
SPt = SA,S – (SHf,SRHfO)
SPd = SPd,S – (CCu,SRArCu + CY,SRYO + CSr,SRSrO + CRb,SRRbO)
SRh = SRh,S – (CCu,SRArCu + CSr,SRSrO + CRb,SRRbO + CPb,SRPb)
Unfortunately, the extremely low PGE concentrations in the extracted samples and the high content of interfering elements in the matrixes hampered the determination of Pd released by road dust exposed to ALF. Even lower PGE concentrations were observed in road dust samples exposed to Gamble’s solution, preventing the determination of all the three PGEs (i.e. Pt, Rd and Rh).
3.2. PGE release
Reproducible results were observed for every duplicate experiment and the blank analysis showed that neither of the simulated lung fluids interfered with the PGE signal. The results on Pt, Pd and Rh release from the various test materials in ALF and Gamble’s solution are illustrated in Figures 2 and 4 respectively. In Figures 3 and 5 results are represented in terms of % of PGE released to allow comparison between compounds and between extraction fluids.
a
b
c
Figure 2: Total Pt, Pd and Rh release from road dust (a), auto catalyst (b) and hydroxide samples (c)exposed to ALF.
Figure 3: % of Pt, Pd and Rh released from road dust, auto catalyst and hydroxide samplesexposed to ALF.
a
b
Figure 4: Pt, Pd and Rh release from auto catalyst (a) and hydroxide samples (b)exposed to Gamble’s solution.
Figure 5: % of Pt, Pd and Rh released from auto catalyst and hydroxide samplesexposed to Gamble’s solution.
The results of the extraction experiments for the three different test materials (road dust, auto catalyst and hydroxide) exposed to ALF showed that metal releases changed with substrates, with the road dust samples showing much higher percentages of released PGEs than either the catalyst or the hydroxide samples (Fig. 3). For road dust samples, the percentage of released Pt was about 36% and released Rh about 88%, whereas for the auto catalyst and hydroxide samples, the percentage of released PGEs was less than 8% and 10%, respectively (Fig. 3). The extraction experiments conducted employing Gamble’s solution showed a low percentage of release for PGEs: up to 0.45% for auto catalyst and 0.15% for hydroxide samples (Fig. 5).
The results also showed considerably lower metal release from samples exposed to Gamble’s solution (pH 7.4) compared to ALF (pH 4.5) (Fig. 3 and 5).
For road dust samples exposed to ALF, the percentage of released Rh was higher than the one for Pt in road dust samples exposed to ALF (Fig. 3). The auto catalyst and hydroxide samples exposed to ALF and Gamble’s solution, showed percentages of released PGEs in the following order: Pd > Rh > Pt (Fig. 3 and 5).
The results for the absolute PGE releases showed that for road dust samples the amount of released Pt was greater than the amount of released Rh (Fig. 2a) and for auto catalyst samples, the absolute release order was Pt > Pd > Rh (Fig. 2b and 4a). Because the hydroxide sample contained the same amount of Pt, Pd and Rh (see paragraph 2.1), the extraction experiments employing this material showed an absolute release order identical to the percentage of release order (Fig. 2c and 4b).
4. Discussion
The elevate percentages (36% for Pt and 88% for Rh) of released PGEs observed for road dust samples (Fig. 3 and 5) corroborate previous assumptions that PGEs in road dust have been transformed from a metallic form to much more soluble species, probably through complexation by inorganic ions {colombo, #246} and humic substances commonly found in the environment {Wood, 1996 #222} {Lustig, 1998 #132;Lustig, 1996 #68;Lustig, 1998 #131}. Besides, the % of released Pt from road dust observed in this study (36%) was in accordance with the results of Artelt et al. {Artelt, 1999 #54} that exposed Lewis rats to a material resembling VEC-emissions during a 90-day inhalation study and found that about 30% of the Pt deposited in the lung was bioavailable.
The higher metal release observed from samples exposed to ALF compared to Gamble’s solution (Fig. 3 and 5) is probably due to the acidity of the former compared to the neutrality of the latter (ALF pH = 4.5 and Gamble pH = 7.4). This suggestion is supported by various studies reporting increased PGE solubilities in acidic environments {Nachtigall, 1996 #129}{Fuchs, 1974 #130}{Zereini, 1997 #22}. The higher PGE release observed in ALF implies that inhaled PGE particles would have to be phagocytized before a significant solubilization occurred.