HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (6) Citing Articles via Google Scholar Google Scholar Articles by Essington, M. E. Articles by Roh, Y. Search for Related Content PubMed Articles by Essington, M. E. Articles by Roh, Y. GeoRef GeoRef Citation Agricola Articles by Essington, M. E. Articles by Roh, Y. Related Collections Heavy Metals Soil Chemistry

DIVISION S-9—SOIL MINERALOGY

The Soil Mineralogy of Lead at Horace's Villa

^a Biosystems Engineering and Environmental Science, University of Tennessee, 2506 E.J. Chapman Dr., Knoxville, TN 37996-4531
^b Soils International, Inc., P.O. Box 22026, Knoxville, TN 37933
^c Environmental Sciences Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831-6038

^* Corresponding author (messington{at}utk.edu).

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES

 
Lead pipe buried by calcareous alluvium at Horace's Villa near Licenza, Italy, for more than a millennium were excavated, along with soil from around and within the lead pipe, to identify the stable corrosion products and Pb-bearing soil minerals. The corrosion crust of the pipe, soil from inside the pipe, and soil from around the pipe were characterized by chemical means and by X-ray diffraction (XRD), scanning electron microscopy (SEM), and energy dispersive X-ray analysis (EDX). The chemical and thermal conditions necessary to produce the identified Pb mineral assemblage was investigated through the development of activity and predominance diagrams. Analysis of total soil Pb concentrations indicates that the lateral movement of Pb from the pipe in the pH >8 soil has exceeded 9 cm. However, background soil Pb concentrations are observed at a lateral distance of 50 cm. The corrosion crust of the lead pipe is composed of litharge [–PbO(s)], cerussite [PbCO₃(s)], and hydrocerussite [Pb₃(CO₃)₂(OH)₂(s)]. Similarly, soil inside the pipe and in the 0- to 1-cm zone around the pipe contains cerussite and hydrocerussite. A detailed thermochemical analysis indicates that the formation of hydrocerussite is favored, relative to cerussite, when the CO₂ partial pressure is approximately atmospheric (10^–3.5 atm or 10^–4.5 MPa) or less, or when the soil temperature is approximately 25°C or greater. The formation of stable lead phosphates has not occurred in these alkaline environments, even though apatite is present and thermochemical analysis predicts the formation of chloropyromorphite [Pb₅(PO₄)₃Cl(s)]. If chloropyromorphite is removed from consideration, hydroxypyromorphite [Pb₅(PO₄)₃OH(s)] is not predicted to form in the presence of apatite.

Abbreviations: EDX, energy dispersive X-ray analysis • SEM, scanning electron microscopy • XRD, X-ray diffraction

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES

 
QUINTUS HORATIUS FLACCUS (Horace) lived from 65 to 8 BP and was Rome's leading lyrical and satirical poet during the rein of Emperor Augustus. After Horace's death and on the site of his humble farmhouse near Licenza, Italy, the Roman Emperor Vespasian constructed an ornate villa, now known as Horace's Villa. As part of an archaeological investigation of the Villa (Frischer and De Simone, 2002), a study was conducted to determine the general stratigraphy of the site and to characterize the soils of the area. During the investigation, a number of lead pipes used to supply the Villa with fresh water were located and excavated from near the Villa's vivarium (pool for raising fish). Inscriptions on the lead pipes, as well as artifacts associated with the pipes, indicate that they were emplaced approximately 75 to 125 CE. Although it has not been established when the lead piping was buried by alluvium from the surrounding hills, there is archaeological evidence to suggest that portions of the Villa were occupied until sometime into the seventh to ninth centuries CE.

The discovery of metallic Pb that has been situated in a soil environment for more than a millennium offers the opportunity to evaluate the equilibrium mineralogy of Pb(s) weathering products. Metallic Pb and Pb-bearing alloys in surface soil are thermodynamically unstable, weathering to form a corrosion layer that encapsulates and stabilizes the Pb(s) (kinetically restricts alteration), such that corrosion rates reportedly range from 0.0004 to 0.3 mm yr^–1 in alkaline soils and 0.02 to 1 mm yr^–1 in acidic soils (Gilbert, 1946; Souvent and Pirc, 2001). Examinations of the corrosion layers of lead bullets, shotgun pellets, and other munitions fragments that have weathered in soils from 2 yr to several decades have routinely found massicot [orthorhombic ß-PbO(s)], cerussite [PbCO₃(s)], and hydrocerussite [Pb₃(CO₃)₂(OH)₂(s)], irrespective of soil pH (Souvent and Pirc, 2001; Lin et al., 1995; Lin, 1996; Cao et al., 2003a, 2003b). In addition, anglesite [PbSO₄(s)] and hydroxypyromorphite [Pb₅(PO₄)₃OH(s)] have been identified in the corrosion layers of bullets and pellets removed from acid sulfate soils (Lin et al., 1995) and P-rich acidic soils (Cao et al., 2003a, 2003b). A corrosion layer consisting of litharge [tetragonal -PbO(s)] and cerussite has been found to encase approximately 2500 yr-old lead artifacts buried in alkaline soils (Reich et al., 2003). An examination of soil surrounding the Pb artifacts has been somewhat less satisfying relative to the identification of weathering products, as soil Pb concentrations are generally too low to identify distinct Pb-bearing mineral phases by XRD.

This study examines the mineralogical composition of lead pipe corrosion material and surrounding soils to determine the extent of Pb migration and to identify Pb-bearing mineral phases that were generated during the oxidation of Pb(s). This paper also provides a detailed thermodynamic analysis of Pb mineral stability to assess the chemical and thermal conditions that have lead to the observed mineral assemblages.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES

 
Horace's Villa is located in the Sabine (Alpine) Hills in south central Italy approximately 50 km northeast of Rome and 1.5 km south of Licenza (42°4'28'' N lat., 12°54'2'' E long.). The Villa is situated in the deeply dissected limestone region of Italy with topographic changes from 360 to 410 m in the valleys to over 980 to 1059 m on the high peaks. The elevation of the Villa is approximately 420 to 430 m. The geologic materials at Horace's Villa are mainly sedimentary rocks dominated by limestone and shale. The soil in the vicinity of the Villa is alluvial, strongly calcareous, contains small limestone chips, and is generally dark brown (7.5YR 3/2) to very dark grayish brown (10YR 3/2) in color. During the archaeological excavation of the Villa bathhouse and gardens, lead pipes approximately 12 cm in diameter and 3 m in length were found at the interface of a buried A horizon and the alluvial material (at depths of approximately 0.5 m) (Fig. 1) . Fragments of the pipe and soil samples from within the pipe and surrounding the pipe were collected. The pH of a 1:5 soil/deionized water suspension of the soil from inside the pipe is 8.03. Soil samples exterior to the pipe were collected along a lateral transect in increments of 0 to 1, 1 to 3, 3 to 6, and 6 to 9 cm from the pipe surface. In addition, a soil sample from along the same lateral transect was obtained at a distance of 50 cm from the pipe surface. The mean pH of soil suspension (1:5 soil/ deionized water) of the samples collected from outside the pipe is 8.7.

View larger version (261K): [in this window] [in a new window]   Fig. 1. (top) Lead pipe exposed during the excavation of Horace's Villa showing the white corrosion crust; and (bottom) an excavated length of pipe.

Activity and predominance diagrams for the PbO-Al₂O₃–CaO-SiO₂–SO₄–Cl-PO₄–CO₂–H₂O system were constructed to examine the stability of Pb minerals as a function of temperature, pH, CO₂(g) partial pressure, and the activities of Pb²⁺(aq) and H₂PO^–₄. Both diagrams illustrate the relative stabilities of the minerals included; however, the predominance diagrams specifically identify the minerals that are predicted to control Pb²⁺(aq) activities as a function of the chemical environment (i.e., the independent activity variables). The Pb-bearing minerals considered in diagram construction were: cerussite [PbCO₃(s)], hydrocerussite [Pb₃(CO₃)₂(OH)₂(s)], hydroxypyromorphite [Pb₅(PO₄)₃OH(s)], chloropyromorphite [Pb₅(PO₄)₃Cl(s)], plumbogummite [PbAl₃(PO₄)₂(OH)₅•H₂O(s)], and alamosite [PbSiO₃(s)]. Litharge [tetragonal -PbO(s)] and anglesite [PbSO₄(s)] were also considered, but do not appear on the activity or predominance diagrams due to their relatively high solubilities. Ancillary phases used to control Al³⁺(aq) activity and to establish limits on H₂PO^–₄ activities were gibbsite [Al(OH)₃(s)], brushite (dicalcium phosphate dehydrate, DCPD) [CaHPO₄•2H₂O(s)], hydroxyapatite [Ca₅(PO₄)₃OH(s)], and calcite [CaCO₃(s)].

The axis variables for the activity diagrams are the activity ratio/products: (pH – 1/2pPb²⁺) and . The stability of any mineral may be expressed as a function of activity ratios/products. Consider the dissolution of hydroxypyromorphite:

[1]

This dissolution reaction may also be written:

[2]

where the equilibrium constant (K) is,

[3]

and pPb²⁺ and pH₂PO^–₄ represent the negative common logarithm of the Pb²⁺(aq) and H₂PO^–₄ activities. Equation [3] may also be written,

[4]

Rearranging Eq. [4] leads to the equation of a stability line for hydroxypyromorphite on a two-dimensional activity diagram where (pH – 1/2pPb²⁺) is the dependent activity variable and is the independent activity variable:

[5]

The dissolution reactions and corresponding stability functions for the Pb-bearing and ancillary minerals are presented in Table 1.

View larger version (35K): [in this window] [in a new window]   Fig. 7. Activity ratio/product diagrams showing the relative stability of lead-bearing minerals as a functions of pH, H2PO–4 activity, Pb2+(aq) activity, and CO2(g) partial pressure at 25°C and 1 atm (0.1 MPa) total pressure (other conditions specific to the generation of the diagrams are discussed in the text). ALA, alamosite; CER, cerussite; CPM, chloropyromorphite; DCDP, brushite; HCER, hydrocerussite; HAP, hydroxyapatite; HPM, hydroxypyromorphite; and PGM, plumbogummite.

View larger version (36K): [in this window] [in a new window]   Fig. 9. Activity ratio/product diagrams showing the relative stability of Pb-bearing minerals as a functions of pH, H2PO–4 activity, Pb2+(aq) activity, and CO2(g) partial pressure at 5°C and 1 atm (0.1 MPa) total pressure (other conditions specific to the generation of the diagrams are discussed in the text). ALA, alamosite; CER, cerussite; CPM, chloropyromorphite; DCDP, brushite; HCER, hydrocerussite; HAP, hydroxyapatite; HPM, hydroxypyromorphite; and PGM, plumbogummite.

View larger version (35K): [in this window] [in a new window]   Fig. 11. Activity ratio/product diagrams showing the relative stability of Pb-bearing minerals as a functions of pH, H2PO–4 activity, Pb2+(aq) activity, and CO2(g) partial pressure at 35°C and 1 atm (0.1 MPa) total pressure (other conditions specific to the generation of the diagrams are discussed in the text). ALA, alamosite; CER, cerussite; CPM, chloropyromorphite; DCDP, brushite; HCER, hydrocerussite; HAP, hydroxyapatite; HPM, hydroxypyromorphite; and PGM, plumbogummite.

[6]

If G^°_r is expressed in units of kJ mol^–1 and T = 298.15 K, Eq. [6] reduces to:

[7]

This expression has great utility in the soil and environmental sciences. However, chemical reactions that occur in the environment do not often occur at 298.15 K (25°C). For a reaction in standard state, the standard free energy change for the reaction is:

[8]

where H^°_r is the standard enthalpy change for the reaction (H^°_r = _ii H^°_f,i and H^°_f,i is the standard enthalpy of formation and _i is the stoichiometry of substance i, where i is positive for reaction products and negative for reactants) and S^°_r is the standard entropy change for the reaction (S^°_r = _ii S^°_f,i and S^°_f,i is the partial molal entropy). Combining Eq. [6] and [8]:

[9]

The temperature dependence of the equilibrium constant can be determined by differentiating Eq. [9] with respect to T:

[10]

For a relatively narrow temperature range about the standard state temperature, the first and third terms on the right side of Eq. [10] are often negligible, as it may be assumed that H^°_r and S^°_r are approximately independent of temperature. The resulting expression is Van't Hoff's equation:

[11]

Transforming to the common logarithm and integrating Eq. [11] leads to an expression that computes log K as a function of T:

[12]

[13]

where K_T is the equilibrium constant at any temperature T, and H^°_r and K° are the enthalpy and equilibrium constant values for the reaction in a reference system at T° (usually the standard state at 298.15 K).

For wide temperature ranges, or when H^°_r and S^°_r are not independent of temperature, additional evaluation of Eq. [11] is required. By definition:

[14]

[15]

where C^°_P is the heat capacity of a reaction at fixed pressure. Substituting Eq. [14] and [15] into [10] yields:

[16]

Rearranging and integrating,

[17]

Utilizing Eq. [9] and rearranging, with standard state as the reference temperature:

[18]

This expression may be employed to determine the equilibrium constant of any reaction at any temperature. The solution to Eq. [18] requires values for the equilibrium constant and entropy of reaction at a reference temperature (and pressure), and the heat capacity of the reaction as a function of temperature. Typically, the reference system is standard state, as compilations of thermodynamic data are quite extensive for this condition. The heat capacity of a solid or gas at fixed pressure and as a function of temperature is typically expressed by the Maier–Kelley equation:

[19]

The Kelley coefficients, a, b, and c are constant for any given species and can also be found in thermodynamic data compilations. For a chemical reaction, C^°_P is computed using:

[20]

Substituting Eq. [20] into Eq. [18] and integrating (from Mattigod and Kittrick, 1980):

[21]

where

[22]

[23]

Over a relatively narrow temperature range, like that observed in the terrestrial environment, it is often assumed that H^°_r and S^°_r are independent of temperature. To test the validity of this assumption, the predicted influence of T on K_T was evaluated for the congruent dissolution of cerussite,

[24]

and the incongruent dissolution of hydroxyapatite to form calcite,

[25]

using both the Van't Hoff equation (Eq. [13]) (and the H^°_f values in Table 2) and Eq. [21] (Fig. 2a,b) . For both cerussite and hydroxyapatite, the maximum disparity between K_T values predicted by Eq. [13] and [21] occurs at 310 K (in the 280–310 K range). For cerussite dissolution, the difference between the ln K₃₁₀ values predicted by Eq. [13] and [21] (ln K₃₁₀) is 0.06, or a 0.22% error. Similarly, ln K₃₁₀ for hydroxyapatite is 0.24, or a 0.30% error. Therefore, the Van't Hoff equation is a satisfactory mechanism for determining the influence of temperature on the stability of minerals in the soil environment. In addition, the Kelley coefficients and entropy values required to use Eq. [21] are not available for a large number of trace minerals of environmental significance, including many of the lead phases considered here (excepting cerussite), further necessitating the use of the Van't Hoff equation. Using the H^°_f values in Table 2 and log K° values for the reactions that describe the dissolution of Pb-bearing and ancillary minerals, log K_T values were computed for 5 (288 K) and 35°C (308 K) conditions (Table 3).

View this table: [in this window] [in a new window]   Table 2. Enthalpy of formation values (H°f at 25°C and 1 atm or 0.1 MPa) used to determine the temperature-dependence of Pb-bearing mineral dissolution reactions. Values were obtained from Robie et al. (1978) unless noted otherwise.

View larger version (16K): [in this window] [in a new window]   Fig. 2. The equilibrium constant as a function of temperature for (a) cerussite dissolution (Eq. [24]) and (b) hydroxyapatite to calcite transformation (Eq. [25]) computed using the Van't Hoff equation (Eq. [13] and constant Hr°) or Eq. [21], [22], and [23] (Hr° is a function of temperature).

View this table: [in this window] [in a new window]   Table 3. Enthalpy of reaction and equilibrium constants (log K values) as a function of temperature for the mineral dissolution reactions shown in Table 1 (data obtained from Nriagu, 1974; Lindsay, 1979; Smith and Martell, 1976; Robie et al., 1978; Taylor and Lopata, 1984).

[26]

The stability line that differentiates between the cerussite and hydroxypyromorphite predominance regions as a function of CO₂(g) partial pressure, pH, and H₂PO^–₄ activity is

[27]

Thus, on a pP_CO2 versus predominance diagram, cerussite will predominate in environments characterized by relatively high CO₂(g) partial pressures (low pP_CO2 values), low H₂PO^–₄ activities, and high pH values [high (pH + pH₂PO^–₄) values]; whereas, hydroxypyromorphite predominates in high pP_CO2 and low environments, with the transition from the cerussite predominance region to the hydroxypyromorphite region demarcated by Eq. [27]. The pertinent transitions and equilibrium constants used to construct predominance diagrams for the 5, 25, and 35°C systems are given in Table 4. The relationships that describe Pb mineral transitions on pP_CO2 versus predominance diagrams are given in Table 5. In addition, the predominance diagrams assume that (pH – 1/2pPb²⁺) is 4.5 (Pb²⁺ activity is 10^–7 at pH 8). All activity and predominance diagrams are constructed by assuming that the activities of the solid phases and of liquid water are unity and the total pressure is 1 atm (0.101 MPa). Other controlling conditions are discussed below.

View larger version (21K): [in this window] [in a new window]   Fig. 8. Predominance diagrams showing the relative stability of Pb-bearing minerals as a functions of pH, H2PO–4 activity, and CO2(g) partial pressure at 25°C and 1 atm (0.1 MPa) total pressure (other conditions specific to the generation of the diagrams are discussed in the text). (a) Including chloropyromorphite and (b) excluding chloropyromorphite. ALA, alamosite; CER, cerussite; CPM, chloropyromorphite; DCDP, brushite; HCER, hydrocerussite; HAP, hydroxyapatite; HPM, hydroxypyromorphite; and PGM, plumbogummite.

View larger version (20K): [in this window] [in a new window]   Fig. 10. Predominance diagrams showing the relative stability of Pb-bearing minerals as a functions of pH, H2PO–4 activity, and CO2(g) partial pressure at 5°C and 1 atm (0.1 MPa) total pressure (other conditions specific to the generation of the diagrams are discussed in the text). (a) Including chloropyromorphite and (b) excluding chloropyromorphite. ALA, alamosite; CER, cerussite; CPM, chloropyromorphite; DCDP, brushite; HCER, hydrocerussite; HAP, hydroxyapatite; HPM, hydroxypyromorphite; and PGM, plumbogummite.

View larger version (20K): [in this window] [in a new window]   Fig. 12. Predominance diagrams showing the relative stability of Pb-bearing minerals as a functions of pH, H2PO–4 activity, and CO2(g) partial pressure at 35°C and 1 atm (0.1 MPa) total pressure (other conditions specific to the generation of the diagrams are discussed in the text). (a) Including chloropyromorphite and (b) excluding chloropyromorphite. ALA, alamosite; CER, cerussite; CPM, chloropyromorphite; DCDP, brushite; HCER, hydrocerussite; HAP, hydroxyapatite; HPM, hydroxypyromorphite; and PGM, plumbogummite.

	RESULTS AND DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES

Other than Pb, the elemental content of the soil surrounding the lead pipe is similar to off-site and background levels, although somewhat elevated concentrations of Co, Cu, Ni, and Zn are observed in the soil that is inside the pipe compared with that surrounding the pipe (Table 6). It is possible that the source of the soil inside the pipe is different from the alluvium that surrounds the pipe, as evidence by higher concentrations of P, Ti, and Zr. On average, the P content of the soil surrounding the pipe (0–9 cm) is 906 mg kg^–1; while the soil P content inside the pipe is 2920 mg kg^–1. Similarly, average Ti and Zr concentrations external to the pipe are 2.34 g Ti kg^–1 and 273 mg Zr kg^–1, while soil concentrations inside the pipe are 4.46 g Ti kg^–1 and 774 mg Zr kg^–1. Elevated soil Cu concentrations inside the pipe may also be derived from Cu impurities in the pipe. Energy dispersive X-ray data (described below) illustrate the occurrence of Cu in association with Pb in pipe corrosion materials. Copper soil concentrations inside pipes from Hadrian's Villa and the Forum are also elevated (ranging from 198 to 292 mg kg^–1 Cu); however, concentrations of Co, Ni, and Zn were not elevated (relative to background levels).

The corrosion crust on the lead pipe is primarily composed of litharge [-PbO(s)], with smaller amounts of cerussite [PbCO₃(s)], hydrocerussite [Pb₃(CO₃)₂(OH)₂(s)], quartz [SiO₂(s)], and metallic Pb(s) (Fig. 3a) . Scanning electron microscopy and EDX analysis of the crust clearly illustrates the presence of Pb-bearing particles that are fibrous (needle-like) and platy (Fig. 4a,b) , reflecting the crystal habits and chemistry of cerussite (Reich et al., 2003) and hydrocerussite (Craig et al., 2001), respectively. The rather nondescript matrix that supports the needle-like and platy particles (Fig. 4c) is typical of litharge precipitates (Wilkinson et al., 2002). Also evident in the EDX spectra of the pipe crust is the presence of Cu, which may be found as an impurity in the lead pipe and appears to exist as a substituent in the Pb phases. Interestingly, qualitative examination of the EDX spectra (comparison of peak heights) indicates that Cu substitution in the platy hydrocerussite crystals is greater than in the fibrous cerussite structures. In the 0- to 1-cm soil sample, calcite is the dominate mineral, accompanied by small amounts of quartz, cerussite, and hydrocerussite (Fig. 3b). Similarly, the soil within the pipe is predominantly composed of cerussite and calcite, with a small amount of quartz and hydrocerussite (Fig. 3c). X-ray diffraction analysis of the particle-size separates of the soil inside the lead pipe indicates that cerussite and hydrocerussite are present in all size separates (Fig. 5 and 6) , perhaps indicating that these minerals exists as coatings on larger particles. The Pb-affected soil clay fraction is composed of calcite, smectite, mica, hydrocerussite, cerussite, and quartz (Fig. 6). This phyllosilicate mineral assemblage indicates that the soil is young in relative age, having received very little moisture to hasten weathering processes (mean annual precipitation is 72 cm).

View larger version (29K): [in this window] [in a new window]   Fig. 3. X-ray diffraction profiles of (a) lead pipe corrosion crust, (b) soil collected from a 0- to 1-cm distance from the pipe, and (c) soil collected from with the pipe. C, cerussite; Ca, calcite; H, hydrocerussite; L, litharge; Pb, metallic lead; and Q, quartz.

View larger version (286K): [in this window] [in a new window]   Fig. 4. Scanning electron micrographs and energy dispersive X-ray spectra of lead pipe corrosion crust. (a) Particles having fibrous morphology; (b) particles having platy morphology; and (c) appearance of crust at low magnification.

View larger version (35K): [in this window] [in a new window]   Fig. 5. X-ray diffraction profiles of sand- and silt-sized fractions of soil obtained from inside lead pipe. C, cerussite; Ca, calcite; H, hydrocerussite; and Q, quartz.

View larger version (46K): [in this window] [in a new window]   Fig. 6. X-ray diffraction profiles of treated clay separates of soil collected from inside the lead pipe. C, cerussite; Ca, calcite; H, hydrocerussite; M, mica; Q, quartz; and smectite.

The types of Pb minerals that result from metallic Pb(s) oxidation appear to depend on a number of factors. Metallic Pb(s) is a thermodynamically unstable phase in the presence of O₂(g). The oxidation of metallic Pb(s) to litharge (log K = 33.15) or massicot (log K = 33.04) is described by the reaction, Pb⁰(s) + 1/2O₂(g) PbO(s). Assuming the activities of metallic Pb(s) and litharge are unity, it follows that litharge should predominate relative to metallic Pb(s) and massicot when the partial pressure of O₂(g) > 10^–16.6, which is insignificant relative to the atmospheric O₂(g) level . Although litharge [log K = 12.67 for the reaction, PbO(s) + 2H⁺ Pb²⁺ + H₂O(l)] is thermodynamically stable relative to massicot (log K = 12.78), and is predicted to predominate in natural water systems, both massicot and litharge are observed in association with Pb(s) in soils (although never together). In reality, massicot and litharge can form concurrently during the oxidation of Pb(s) at terrestrial temperatures and pressure. Interestingly, it appears as though massicot is associated with lead bullets and pellets that have weathered for relatively short periods; whereas, litharge encases Pb(s) that was been weathering since antiquity. This correlation hints that massicot is a metastable fast-former that in time reverts to litharge. The inclusion of anglesite with massicot observed on the surfaces of shotgun pellets (Lin et al., 1995; Lin, 1996) was thought to result from the acidic conditions of the soil environment that supports low CO^2–₃ relative to SO^2–₄ activities, thus favoring the formation of the lead sulfate mineral. Similarly, high soluble phosphate activities (inferred from the high total soil P concentrations) favors hydroxypyromorphite formation (Cao et al., 2003a, 2003b).

The long-term resistance of Pb(s) to complete disintegration in soil is due to the formation of the stable Pb crusts composed of hydrocerussite and cerussite (forming in both acidic and alkaline environments). As indicated in the above examples, the environmental conditions that favor the formation of cerussite in Pb(s) crusts do not appear to be consistent. Cao et al. (2003a)(2003b) detected cerussite only in acidic systems, while Lin et al. (1995) and Lin (1996) detected cerussite only in alkaline soils (as did Reich et al., 2003). In the present study, both cerussite and hydrocerussite are identified in the pipe crust and in the alkaline soil surrounding the pipe. Attempts to predict the environmental conditions that lead to the formation of cerussite or hydrocerussite have been hampered by the ambiguous nature of hydrocerussite thermodynamic data, much of which can be traced to the original results of Randall and Spencer (1928) (Mercy et al., 1998). However, more recent solubility and electrochemical studies by Bilinski and Schindler (1982), Taylor and Lopata (1984), and Mercy et al. (1998) have produced comparable Gibbs energy of formation values for hydrocerussite, suggesting that thermochemical evaluations may be performed with a relatively high degree of confidence.

The activity ratio and predominance diagrams in Fig. 7 and 8 examine the relative stability of the Pb-bearing minerals; litharge, cerussite, hydrocerussite, hydroxypyromorphite, chloropyromporphite, plumbogummite, alamosite, and anglesite, as a function of solution pH, Pb²⁺(aq) activity, H₂PO^–₄ activity, and P_CO2 in a 25°C environment. The stabilities of the calcium phosphates, hydroxyapatite, and brushite (DCPD) are also included to establish the range of H₂PO^–₄ activities that is predicted to be controlled by mineral precipitation and dissolution reactions. In addition, analysis of sand-sized particles by SEM–EDX also indicates the presence of Ca- and P-bearing particles (data not shown), indicating the occurrence of apatite in the Pb-affected soil (from inside the lead pipe). It is a characteristic of activity ratio/product diagrams that three components, the activities of H⁺(aq), Pb²⁺(aq), and H₂PO^–₄, are incorporated into two axis variables: (pH – 1/2pPb²⁺) and . For the development of Fig. 7, additional components; P_CO2 and the activities of Cl^–(aq), Al³⁺(aq), Ca²⁺(aq), H₄SiO⁰₄, and SO^–2₄, are controlled by either a reservoir or a controlling mineral phase. The partial pressure of CO₂(g) is fixed at four levels, from 0.0001 to 0.1 atm (1 x 10^–5 to 0.01 MPa) in decade increments, resulting in four diagrams (one for each P_CO2 value), covering the range of P_CO2 commonly observed in soil environments. Figure 8 assumes (pH – 1/2pPb²⁺) = 4.5 (the activity of Pb²⁺ is 10^–7 at pH 8). The activity of Cl^–(aq) is fixed by assuming (pH + pCl^–) = 11 (the activity of Cl^– is 10^–3 at pH 8) and the activity of SO^–2₄ is fixed by assuming = 9.5 (the activity of SO^–2₄ is 10^–3 at pH 8). The activity of Ca²⁺(aq), required to establish the hydroxyapatite and brushite stability lines and regions, is controlled by calcite. The activity of Al³⁺, required to establish the plumbogummite stability line and region, is controlled by gibbsite. Finally, the activity of H₄SiO⁰₄ is controlled by amorphous SiO₂•2H₂O (required to establish alamosite stability).

The minerals predicted to control the solution activities of Pb²⁺(aq) in 25°C environments are alamosite, cerussite, chloropyromorphite, and plumbogummite, depending on P_CO2 (Fig. 7). Although the stabilities of litharge and anglesite were also considered in the development of these figures, these minerals are predicted to support (pH – 1/2pPb²⁺) levels that are >5, and thus do not appear on the diagrams. At equilibrium, hydroxypyromorphite (in addition to litharge and anglesite) is not predicted to control Pb²⁺ activity under the constraints used to construct the diagram. If the potential formation of alamosite is discounted, as this mineral is a high-temperature silicate phase, hydrocerussite is predicted to be more stable than cerussite when P_CO2 <10^–3.82 atm (10^–4.82 MPa) (Fig. 8), a value that is approximately 50% of atmospheric levels (10^–3.5 atm or 10^–4.5 MPa). Thus, the formation of hydrocerussite should be favored in systems that are alkaline and closed with respect to CO₂(g), or are open but depleted in CO₂(g) at a rate that is greater than the rate of diffusion through the soil pores. The Pb²⁺–controlling mineral is also determined by the activity of H₂PO^–₄ in the soil solution. Cerussite is the controlling phase when levels are below those controlled by apatite dissolution and P_CO2 levels are approximately atmospheric and greater. Plumbogummite is only predicted to control Pb²⁺(aq) activities and predominate when is controlled by a highly soluble calcium phosphate (e.g., brushite), when P_CO2 levels are between 0.01 and 0.1 atm (0.001 and 0.01 MPa), and when Cl^–(aq) activities are sufficiently low (<10^–12.9 at pH 8) to favor the stability of hydroxypyromorphite over chloropyromorphite. Interestingly, hydroxypyromorphite is found to be associated with lead bullets in high P, acidic soils (Cao et al., 2003a, 2003b), and both the hydroxy- and chloro-forms of pyromorphite have been identified in Pb-contaminated soils treated with phosphate amendments (Cao et al., 2002, 2003c; Yang et al., 2001). As illustrated in Fig. 8b, low Cl^–(aq) activities (<10^–12.9 at pH 8) will lead to the formation of lead carbonates, even when hydroxyapatite is present.

The chemical conditions in the soil surrounding the lead pipe at Horace's Villa must be consistent with the stability of cerussite and hydrocerussite, as these minerals are present. Thus, the local environment within and about the pipe must fluctuate relative to P_CO2. In addition, soluble phosphate levels must be below those controlled by apatite dissolution, as pyromorphites should readily form when apatite, or a more soluble phosphate source, is introduced into Pb-contaminated soils (Cotter-Howells, 1996; Traina and Laperche, 1999; Zhang and Ryan, 1999; Cao et al. 2003a, 2003b, 2003c; Melamed et al., 2003). The fact that chloropyromorphite is not identified in the soil environment near the soil–pipe interface may indicate that H₂PO^–₄ activity levels are depressed relative to those supported by apatite, even though the presence of this mineral is supported by SEM/EDX data and total soil P concentrations are 890 mg kg^–1 outside the pipe and 2920 mg kg^–1 inside the pipe.

It is conceivable (and quite probable) that the 25°C stability diagrams may not represent the temperature condition of the soil environment at the archaeological site. Although the air temperatures at the site range from an annual average low of 10.6°C to an average high of 20.4°C, the mean air temperatures may not represent the range in soil temperatures. Using the enthalpy of formation data in Table 2, activity ratio/product and predominance diagrams representing potential low (5°C) and high (35°C) temperature conditions in the soil were constructed. The activity ratio and predominance diagrams in Fig. 9 and 10 for a 5°C environment are quite similar to those describing the 25°C system. However, hydrocerussite is not predicted to be a stable Pb-bearing phase (under the given conditions). In addition, plumbogummite is unstable relative to chloropyromorphite when P_CO2 levels are less than approximately 0.1 atm (0.01 MPa). The significant differences between the 25 and 35°C activity ratio diagrams (Fig. 11 and 12) is the displacement of cerussite (and alamosite) by hydrocerussite as the stable Pb-bearing carbonate phase when P_CO2 < 10^–3.40 (approximately 26% greater than atmospheric levels). The fact that both cerussite and hydrocerussite are identified in the Villa soil indicates that P_CO2 levels are approximately atmospheric or lower, and that soil temperatures are consistent with standard state or higher. It is also evident that a lead carbonate phase (cerussite or hydrocerussite) will control Pb²⁺(aq) activity in the calcareous environment, irrespective of temperature, given the restricted H₂PO^–₄ activities that are typical of calcareous soils, as well as the low total P levels in the soil.

A number of important points relative to the potential long-term behavior of Pb in alkaline soils can be obtained through the analysis of a soil system in contact with metallic Pb(s) for more than a century (Horace's Villa). Despite the apparent thermodynamic instability of the metallic Pb(s), the pipe remains in the Villa soil more than a millennium after burial, owing to the insulating effect of the stable corrosion layer. Conceivably, spent lead ammunition, a present-day environmental concern, may also remain in affected calcareous soils for extended periods. Efforts to stabilize soil Pb, as through the in situ incorporation of a phosphate source, will be of limited long-term effectiveness. It is also evident from the activity ratio diagrams that hydroxyapatite is an unstable phosphate mineral under the conditions necessary for lead carbonate precipitation. Since cerussite and hydrocerussite are directly observed, it follows that hydroxyapatite is not controlling soluble phosphate activity. The incorporation of apatite as a stabilization mechanism may result in the initial formation of a pyromorphite phase (as illustrated in the literature); however, without continued management to counter the absorption of CO₂(g), coupled with the nonlabile properties of the metallic Pb, the long-term stability of a pyromorphite will not be realized. Instead, it appears that Pb metal in a calcareous or alkaline soil will ultimately oxidize to litharge, which in turn will weather to cerussite and hydrocerussite, which are the stable Pb-bearing phases.

Received for publication July 16, 2003.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES

 

• Ammons, J.T., M.E. Essington, R.J. Lewis, A.O. Gallagher, and G.M. Lessman. 1995. An application of a modified microwave total dissolution technique for soils. Commun. Soil Sci. Plant Anal. 26:831–842.
• Bilinski, H., and P. Schindler. 1982. Solubility and equilibrium constants of lead in carbonate solutions (25°C, i = 0.3 mol dm^–3). Geochim. Cosmochim. Acta 46:921–928.
• Cao, X., L.Q. Ma, M. Chen, D.W. Hardison, and W.G. Harris. 2003a. Lead transformation and distribution in the soils of shooting ranges in Florida, USA. Sci. Total Environ. 307:179–189.[Medline]
• Cao, X., L.Q. Ma, M. Chen, D.W. Hardison, and W.G. Harris. 2003b. Weathering of lead bullets and their environmental effects at outdoor shooting ranges. J. Environ. Qual. 32:526–534.[Abstract/Free Full Text]
• Cao, R.X., L.Q. Ma, M. Chen, S.P. Singh, and W.G. Harris. 2003c. Phosphate-indiced metal immobilization in a contaminated site. Environ. Pollut. 122:19–28.[Medline]
• Cao, X., L.Q. Ma, M. Chen, S.P. Singh, and W.G. Harris. 2002. Impacts of phosphate amendments on lead biogeochemistry at a contaminated site. Environ. Sci. Technol. 36:5296–5304.[Medline]
• Cotter-Howells, J. 1996. Lead phosphate formation in soils. Environ. Pollut. 93:9–16.[Medline]
• Craig, J.R., J.E. Callahan, J.W. Miller, and W.R. Lusdardi. 2001. Preliminary studies of some base and precious metals from the Queen Anne's Revenge. Southeast. Geol. 40:41–48.
• Frischer, B., and M. De Simone. 2002. New excavations at Horace's Villa, 1997–2001 [Online]. Available at http://www.humnet.ucla.edu/horaces-villa/newex/Results.htm (posted 1 Jan. 2002; verified 4 Feb. 2004).
• Gilbert, P.T. 1946. Corrosion of copper, lead, and lead-alloy specimens after burial in a number of soils for periods up to 10 years. J. Inst. Met. 74:139–147.
• Hemingway, B.S., and R.A. Robie. 1977. Enthalpies of formation of low albite (NaAlSi₃O₈), gibbsite (Al(OH)₃) and NaAlO₂; revised values for H^°_{f, 298} and G^°_{f, 298} of some aluminosilicate minerals. J. Res. U.S. Geol. Surv. 5:413–429.
• Hemingway, B.S., R.R. Seal, II. and I-Ming Chou. 2002. Thermodynamic data for modeling acid mine drainage problems: Compilation and estimation of data for selected soluble iron-sulfate minerals [Online]. U.S. Geological Survey Open-File Report 02–161. Available at http://pubs.usgs.gov/of/2002/of02-161/OF02-161.htm (posted 8 May 2002; verified 19 Jan. 2004).
• Lin, Z. 1996. Secondary mineral phases of metallic lead in soils of shooting ranges from Örebro County, Sweden. Environ. Geol. 27:370–375.
• Lin, Z., B. Comet, U. Qvarfort, and R. Herbert. 1995. The chemical and mineralogical behavior of Pb in shooting range soils from central Sweden. Environ. Pollut. 89:303–309.[Medline]
• Lindsay, W.L. 1979. Chemical equilibria in soils. J. Wiley & Sons, New York.
• Mattigod, S.V., and J.A. Kittrick. 1980. Temperature and water activity as variables in soil mineral activity diagrams. Soil Sci. Soc. Am. J. 44:149–154.[Abstract/Free Full Text]
• Melamed, R., X. Cao, M. Chen, and L.Q. Ma. 2003. Field assessment of lead immobilization in a contaminated soil after phosphate application. Sci. Total Environ. 305:117–127.[Medline]
• Mercy, M.A., P.A. Rock, W.H. Casey, and M.M. Mokarram. 1998. Gibbs energies of formation for hydrocerussite [Pb(OH)₂•(PbCO₃)₂(s)] and hydrozincite {[Zn(OH)₂]₃•(ZnCO₃)₂(s)} at 298 K and 1 bar from electrochemical cell measurements. Am. Mineral. 83:739–745.[Abstract]
• Nriagu, J.O. 1974. Lead orthophosphates. IV. Formation and stability in the environment. Geochim. Cosmochim. Acta 38:887–898.
• Randall, M., and H.M. Spencer. 1928. Solubility of lead oxide and basic carbonate. J. Am. Chem. Soc. 50:1572–1583.
• Reich, S., G. Leitus, and S. Shalev. 2003. Measurement of corrosion content of archaeological lead artifacts by their Meissner response in the superconducting state; a new dating method [Online]. arXiv Codensed Matter:0304238. Available at http://xxx.arxiv.cornell.edu/ftp/cond-mat/papers/0304/0304238.pdf (posted Apr. 2003; verified 19 Jan. 2004).
• Robie, R.A., B.S. Hemingway, and J.R. Fisher. 1978. Thermodynamic properties of minerals and related substances at 298.15 K and 1 bar (10⁵ pascals) pressure and at higher temperature. USGS Bull. 1452. U.S. Gov. Print. Office, Washington, DC.
• Sangameshwar, S.R., and H.L. Barnes. 1983. Supergene processes in zinc-lead-silver sulfide ores in carbonates. Econ. Geol. 78:1379–1397.[ISI]
• Smith, R.M., and A.E. Martell. 1976. Critical stability constants. Volume 4: Inorganic complexes. Plenum Press, New York.
• Souvent, P., and S. Pirc. 2001. Pollution caused by metallic fragments introduced into soils because of World War I activities. Environ. Geol. 40:317–323.
• Taylor, P., and V.J. Lopata. 1984. Stability and solubility relationships between some solids in the system PbO-CO₂–H₂O. Can. J. Chem. 62:395–402.
• Traina, S.J., and V. Laperche. 1999. Contaminant bioavailability in soils, sediments, and aquatic environments. Proc. Natl. Acad. Sci. USA 96:3365–3371.[Abstract/Free Full Text]
• Wagman, D.D., W.H. Evans, V.B. Parker, R.H. Schumm, I. Harlow, S.M. Bailey, K.L. Churney, and R.L. Nutall. 1982. Selected values for inorganic and C₁ and C₂ organic substances in SI units. J. Phys. Chem. Ref. Data 11:(Suppl no. 2).
• Walther, J.V., and H.C. Helgeson. 1977. Calculation of the thermodynamic properties of aqueous silica and the solubility of quartz and its polymophs at high pressures and temperatures. Am. J. Sci. 277:1315–1351.[Abstract/Free Full Text]
• Wilkinson, T.J., D.L. Perry, E. Spiller, P. Berdahl, S.E. Derenzo, and M.J. Weber. 2002. A facile wet synthesis of nanoparticles of litharge, the tetragonal form of PbO. Proc. Mater. Res. Soc. 704:117–121.
• Yang, J., D.E. Mosby, S.W. Casteel, and R.W. Blanchar. 2001. Lead immobilization using phosphoric acid in a smelter-contaminated urban soil. Environ. Sci. Technol. 35:3553–3559.[Medline]
• Zhang, P., and J.A. Ryan. 1999. Transformation of Pb(II) from cerussite to chloropyromorphite in the presence of hydroxyapatite under varying conditions of pH. Environ. Sci. Technol. 33:625–630.

This article has been cited by other articles:

				R. A. Moseley, M. O. Barnett, M. A. Stewart, T. L. Mehlhorn, P. M. Jardine, M. Ginder-Vogel, and S. Fendorf Decreasing Lead Bioaccessibility in Industrial and Firing Range Soils with Phosphate-Based Amendments J. Environ. Qual., October 23, 2008; 37(6): 2116 - 2124. [Abstract] [Full Text] [PDF]

				M. Levonmaki, H. Hartikainen, and T. Kairesalo Effect of Organic Amendment and Plant Roots on the Solubility and Mobilization of Lead in Soils at a Shooting Range J. Environ. Qual., May 31, 2006; 35(4): 1026 - 1031. [Abstract] [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal