LEAD SPECIATION, BIOACCESSIBILITY, AND BIOASSAY ANALYSIS SUMMARY SOUTHWEST JEFFERSON COUNTY MINING SITE OU1, OU2, OU3 and OU5 JEFFERSON COUNTY, MISSOURI TO: Preston Law, EPA TOPO FROM: Chris Williams, HGL Task Order Manager THROUGH: Robert C. Overfelt, P.G., CHMM, HGL Program Manager DATE: November 15, 2011 SUBJECT: Lead Speciation, Bioaccessibility, and Bioassay Analysis Summary CONTRACT NO: EP-S7-05-05 TASK ORDER NOs: 0042 After field work for the Remedial Investigation at Southwest Jefferson County Mining Site, OU (operable unit) 1, OU2, OU3, and OU5 was complete, EPA requested that five of the fine sieve surface soil samples be submitted for lead speciation and lead Bioaccessibility analysis. Five samples were selected by EPA after the EPA laboratory had completed the originally requested analyses on the samples submitted. The samples were collected from two soil samples from OU1, one from OU2, and 2 from OU3 and submitted for lead speciation analysis. The samples were then submitted to the Laboratory for Environmental and Geological Studies (LEGS) at the University of Colorado in Boulder, Colorado for analysis. A brief summary of each sample is presented below. Lead Speciation and Bioaccessibility The LEGS conducted lead speciation and Bioaccessibility (Bioaccessability characterizes the potential for a toxin to be bioavailable) analyses on five surface soil samples collected during the RI at the Jefferson County Mining Site. The full report and backup data is included on the attached CD. The samples analyzed were fine sieve samples where the particles were less than 250 microns (μm) in size. From the analyses, the constituents and distribution of the lead- containing particles were determined. The resultant data was compiled into bar graphs for each sample and is discussed in detail in the following paragraphs. Three factors were analyzed in depth: • frequency of occurrence (the distribution of particles based on composition), • relative lead mass (the distribution of lead based solely on lead mass), and • bioaccessible lead mass (lead mass calculated by omitting any particle that is either greater than 250 μm in size or “included/enclosed” in another particle). JC-1277 C1 (Lab ID: 5044-2): The speciation results for this sample indicate that nearly 95 percent of the particles contained lead as part of a manganese oxyhydroxide (MnOOH) complex (approximately 80 percent) or iron oxide (FeOOH) complex (15 percent), leaving approximately 5 percent of the particles containing elemental lead (associated with brass or as
Transcript
LEAD SPECIATION, BIOACCESSIBILITY, AND BIOASSAY ANALYSIS SUMMARY
SOUTHWEST JEFFERSON COUNTY MINING SITE OU1, OU2, OU3 and OU5
JEFFERSON COUNTY, MISSOURI TO: Preston Law, EPA TOPO FROM: Chris Williams, HGL Task Order Manager THROUGH: Robert C. Overfelt, P.G., CHMM, HGL Program Manager DATE: November 15, 2011 SUBJECT: Lead Speciation, Bioaccessibility, and Bioassay Analysis Summary CONTRACT NO: EP-S7-05-05 TASK ORDER NOs: 0042 After field work for the Remedial Investigation at Southwest Jefferson County Mining Site, OU (operable unit) 1, OU2, OU3, and OU5 was complete, EPA requested that five of the fine sieve surface soil samples be submitted for lead speciation and lead Bioaccessibility analysis. Five samples were selected by EPA after the EPA laboratory had completed the originally requested analyses on the samples submitted. The samples were collected from two soil samples from OU1, one from OU2, and 2 from OU3 and submitted for lead speciation analysis. The samples were then submitted to the Laboratory for Environmental and Geological Studies (LEGS) at the University of Colorado in Boulder, Colorado for analysis. A brief summary of each sample is presented below. Lead Speciation and Bioaccessibility
The LEGS conducted lead speciation and Bioaccessibility (Bioaccessability characterizes the potential for a toxin to be bioavailable) analyses on five surface soil samples collected during the RI at the Jefferson County Mining Site. The full report and backup data is included on the attached CD. The samples analyzed were fine sieve samples where the particles were less than 250 microns (µm) in size. From the analyses, the constituents and distribution of the lead-containing particles were determined. The resultant data was compiled into bar graphs for each sample and is discussed in detail in the following paragraphs. Three factors were analyzed in depth:
• frequency of occurrence (the distribution of particles based on composition), • relative lead mass (the distribution of lead based solely on lead mass), and • bioaccessible lead mass (lead mass calculated by omitting any particle that is either
greater than 250 µm in size or “included/enclosed” in another particle). JC-1277 C1 (Lab ID: 5044-2): The speciation results for this sample indicate that nearly 95 percent of the particles contained lead as part of a manganese oxyhydroxide (MnOOH) complex (approximately 80 percent) or iron oxide (FeOOH) complex (15 percent), leaving approximately 5 percent of the particles containing elemental lead (associated with brass or as
HGL—Lead Speciation, Bioaccessibility, and Bioassay Analysis Summary, SW Jefferson Co. Mining Site—Jefferson Co., MO
U.S. EPA Region 7 2
native lead). The relative lead mass data indicates that 5 percent of the particles (as native lead and associated with brass) contain 35 percent of the lead by weight in the sample, and that the other 95 percent of the particles contain only 65 percent of the total lead content by weight. Based on the data, the native lead and lead contained in brass (or tin solder) is not bioaccessible.
Form Frequency (linear) %
Relative Pb Mass
% Bioaccessible Pb
Mass % MnOOH 80.3 61.63 94.34 FeOOH 15.27 3.67 5.62 Native Lead 2.04 34.67 0 Brass 2.4 0.02 0.04
JC-0851 C4 (Lab ID: 5044-27): The speciation results for this sample indicate that nearly 93 percent of the particles contained lead as part of a MnOOH complex (approximately 87 percent), FeOOH complex (8 percent) or phosphate complex (1.44 percent), leaving approximately 5 percent of the particles containing elemental lead (associated with brass or as native lead). The relative lead mass data indicates that 5 percent of the particles (as native lead or associated with brass) contain 20 percent of the lead in the sample by weight, and that the other 95 percent of the particles contain 80 percent of the total lead content by weight. In this sample nearly 100 percent of the bioaccessible lead is contained in the MnOOH, FeOOH, and phosphate components, and the elemental lead associated with brass (or tin solder) is not bioaccessible.
0 20 40 60 80 100
MnOOH
FeOOH
Native Lead
BrassSample 5044‐2
Bioaccessible Pb Mass Relative Pb Mass Frequency of Occurrence
HGL—Lead Speciation, Bioaccessibility, and Bioassay Analysis Summary, SW Jefferson Co. Mining Site—Jefferson Co., MO
JC-0471 C3 (Lab ID: 5045-1): The speciation results for this sample indicate that about 6 percent of the particles contained elemental lead as inclusions in brass; this 6 percent contains only about 1 percent of the total lead by weight. Additionally, another 6 percent of the particles are composed of a lead-metal-oxide (PbMO), which contains 36 percent of the total lead by weight. Eighty-six percent of the particles are either lead containing MnOOH or FeOOH complexes, with a further, roughly 2 percent of the media being composed of zinc oxide (ZnO), ferrous sulfate (FeSO4), and phosphates. Approximately 61.5 percent of the lead by weight contained in the sample is within the MnOOH and FeOOH components. In this sample, nearly all of the lead is bioaccessible.
Bioaccessible Pb Mass Relative Pb Mass Frequency of Occurrence
HGL—Lead Speciation, Bioaccessibility, and Bioassay Analysis Summary, SW Jefferson Co. Mining Site—Jefferson Co., MO
U.S. EPA Region 7 4
JC—1895 C3 (Lab ID: 5046-5): The speciation results for this sample indicate that nearly 99 percent of the particles are part of MnOOH (79 percent) or FeOOH (19.5 percent) complexes; the remaining constituents are combined with FeSO4 particles. Nearly 100 percent of the lead by weight is contained within the MnOOH and FeOOH fraction, and nearly all of the lead is considered bioaccessible.
Form Frequency (linear) %
Relative Pb Mass %
Bioaccessible Pb Mass %
FeOOH 19.48 6.92 6.92
MnOOH 79.01 92.81 92.81
FeSO4 1.52 0.27 0.27
0 10 20 30 40 50 60
FeOOH
MnOOH
Brass
ZnO
FeSO4
Phosphate
PbMOSample 5045‐1
Bioaccessable Pb Mass Relative Pb Mass Frequency of Occurrence
HGL—Lead Speciation, Bioaccessibility, and Bioassay Analysis Summary, SW Jefferson Co. Mining Site—Jefferson Co., MO
U.S. EPA Region 7 5
JC-0811 LS (Lab ID: 5046-15FD): The speciation results for this sample indicate that approximately 91 percent of the particles are related to MnOOH (68 percent) or FeOOH (23 percent) complexes; the remaining constituents are part of either an FeSO4 or phosphate particle, or associated with brass. Approximately 99 percent of the lead by weight is contained within the MnOOH and FeOOH fraction, and nearly all of the lead is considered bioaccessible.
Form Frequency (linear) %
Relative Pb Mass %
Bioaccessible Pb Mass %
FeOOH 22.89 9.08 9.08
MnOOH 67.9 89.79 89.79
Brass 7.3 0.12 0.12
FeSO4 1.69 0.32 0.32
Phosphate 0.23 0.69 0.69
0 10 20 30 40 50 60 70 80 90 100
FeOOH
MnOOH
FeSO4
Sample 5046‐5
Bioaccessable Pb Mass Relative Pb Mass Frequency of Occurrence
HGL—Lead Speciation, Bioaccessibility, and Bioassay Analysis Summary, SW Jefferson Co. Mining Site—Jefferson Co., MO
U.S. EPA Region 7 6
In Vitro Lead Bioassay Analysis Summary Five surface soil samples collected during the RI were analyzed by LEGS. From the analyses the in vitro bioaccessibility (IVBA) of lead in the samples was determined using the method that was developed at LEGS. Using the results of the five fine sieve samples analyzed, EPA Region VIII Swine Model relative bioavailability (RBA) values for lead were calculated (Bioavailability indicates the degree (percentage) of toxin absorption into a target organ, urine or blood). Approximately one gram of the sieved sample was used for the analyses. In accordance with the method, after digestion, the digestate was analyzed for lead content. The lead found in the digestate represents the IVBA, after applying a correlation factor; the laboratory generated the RBA. The IVBA and RBA lead are expressed as percentages of the total lead that was found in the sample. These values are represented in the table below.
0 10 20 30 40 50 60 70 80 90 100
FeOOH
MnOOH
Brass
FeSO4
PhosphateSample 5046‐15‐FD
Bioaccessable Pb Mass Relative Pb Mass Frequency of Occurrence
HGL—Lead Speciation, Bioaccessibility, and Bioassay Analysis Summary, SW Jefferson Co. Mining Site—Jefferson Co., MO
*Predicted based on Drexler and Brattin, 2007 The results of the LEGS analysis of lead bioavailability are higher than those found in the original investigation, this is most likely due to the variability of the soil itself and does not necessarily reflect an error in the method or analyses of the samples.
Summary and Conclusions The results of the lead speciation and bioaccessible lead analyses run on the above five samples indicate that the majority of lead-containing particles have lead in a form that is bioaccessible. Native lead, and lead included in brass (or tin solder) represent the fraction of particle that have no, or very low bioaccessibility. It is of note that the brass constituents of the samples are most likely a result of contamination from the sampling equipment used to process the samples. In particular the brass may have entered the samples during the sieving process. Brass sieves were utilized in the field to separate the soil samples into a bulk (using a No.10-mesh sieve) and fine (using a No. 60-mesh 250 µm sieve) fraction prior to analysis. It is also of note that the brass (or tin solder) was the only anthropogenic species found in the samples—all of the other phases are typical soil-forming minerals with sorbed lead. The results of the speciation report are distributions, and therefore are not immediately translatable into bioaccessibility values. The “bioaccessibility Pb Mass” title is not actually the percentage of bioaccessible lead, but a percent distribution of where the bioaccessible particles can be found. A better value for bioaccessibility, based on speciation results, would be generated by measuring the relative mass percent values for species other than native lead,
HGL—Lead Speciation, Bioaccessibility, and Bioassay Analysis Summary, SW Jefferson Co. Mining Site—Jefferson Co., MO
U.S. EPA Region 7 8
brass, and ZnO. While the values that result from the speciation analysis suggest that the majority of lead in the samples is bioaccessible, the bioassay results indicate that only 30-60% of that lead was actually bioaccessible. The difference in bioaccessibility values found in the speciation report and the bioassay results stem from theoretical calculations vs. experimental results. Additionally, there is some variability between the original EPA bioassay results and the bioassay results from the LEGS analysis; this is not necessarily an error. As the lead speciation report indicates, a small number of high percentage lead particles can drastically alter the bioaccessibility and bioavailability of a sample; 1 percent native lead particles can contribute up to 20 percent of the total lead contained in a sample (as found in sample 5044-27). Due to the variable nature of soil samples, a small difference in the particle distribution could lead to large changes in bioaccessibility and bioavailability. Abbreviations and Table Notes % percent MnOOH manganese oxyhydroxide FeOOH iron oxide Pb lead FeSO4 ferrous sulfate RBA Relative Bioavailability ID identification RI Remedial Investigation IVBA In Vitro Bioaccessibility RPD Relative percent difference Lab laboratory ZnO zinc oxide
Attachment: CD with LEGS Report
Laboratory Data Package
November 01, 2011
Prepared for:
HGL
Prepared by: Laboratory for Environmental and Geological Studies (LEGS)
University of Colorado Benson Earth Science 2200 Colorado Ave. Boulder, CO 80309
Table of Contents
Statement of Work.......................................................................... 3
CD ROM All raw data files, backscatter photomicrographs, and spectra.
3
Statement of Work HGL provided LEGS with 5 samples on 10/03/11 (see attached COC). From these, 5 samples splis were
taken for EMPA lead speciation and lead bioaccessability using the bioassay of Drexler and Brattin,
2007. All data is provided on enclosed CD along with a brief review of the methodology used.
Speciation Methodology The Laboratory for Environmental and Geological Studies (LEGS) at the University of Colorado,
Department of Geological Sciences contains the following equipment was used for this project:
A JOEL 8600 electron microprobe, with four wavelength dispersive detectors (TAP, LIF, PET, LdB, LdC and Ld1 crystals) and an energy dispersive detector. The system includes backscatter and secondary detectors for imaging and can
produce both x-ray spectra and photomicrographs in TIF format. Certified mineral standards for all elements of concern are available for EMPA standardization. SOP for metal speciation is available at our website:
Representative backscatter photomicrographs (BSPM) illustrating sample characteristics were acquired
and EDS spectra acquired and it is recommended the client review these images. Data from EMPA will
be summarized using two methods as illustrated below.
The first method is the determination of FREQUENCY OF OCCURRENCE. This is calculated by
summing the longest dimension of all the lead-bearing phases observed and then dividing each phase by
the total.
Equation 1.0 will serve as an example to the calculation for a lead-bearing compound. Other metals
4
follow a similar calculation.
FPb - Frequency of occurrence of lead in a single phase. PLD - An individual particles longest dimension 3 (PLD) phase-1 FPb in phase-1 = ______________________________ 3 (PLD)phase-1 + 3 (PLD)phase-2 + 3 (PLD)phase-n %FPb in phase-1 = FPb in phase-1 * 100 This data thus illustrates which lead-bearing phase(s) are the most commonly observed in the sample or
relative volume percent.
The second calculation used in this report is the determination of RELATIVE LEAD MASS of a
metal-bearing phase. These data are calculated (using lead as an example) by substituting the PLD term
in the equation above with the value of MPb. This term is calculated as defined below.
MPb - Mass of lead in a phase SG - Specific Gravity of a phase
5
ppm Pb - Concentration in ppm of lead in phase MPb = FPb * SG * ppm Pb The advantage in reviewing the RELATIVE LEAD MASS determinations is that it gives one
information as to which metal-bearing phase(s) in a sample are likely to control the total bulk
concentration for lead. As an example, PHASE-1 may by relative volume comprise 98% of the sample,
however it has a low specific gravity and contains only 1000 ppm lead, while PHASE-2 comprises 2%
of the sample, has a high specific gravity and contains 850000 ppm of lead. In this example it is
PHASE-2 that is the dominant source of lead to the sample. A third calculation is made based on the
relative lead mass calculation. The bioaccessible lead mass is calculated by omitting any particle that is
either greater than 250 microns in size or “included/enclosed” in another particle.
Sample Preparation 1) Logging the samples of which polished mounts will be prepared 2) Inspection of all plastic cups, making sure each is clean and dry 3) Labeling each "mold" with its corresponding sample number. 4) All samples will be split to produce a homogeneous 1-4 gram sample. 5) Mixing epoxy resin and hardener according to manufacturer's directions. 6) Pour 1 gram of sample into mold. Double checking to make sure sample numbers on mold and sample match. Pouring epoxy into mold to just cover sample grains. 7) Using a new wood stirring stick with each sample, carefully blend epoxy and grains so as to coat all grains with epoxy. 8) Setting molds to cure at ROOM TEMPERATURE in a clean restricted area. Adding labels with sample numbers and covering with more epoxy resin. Leaving to cure completely at room temperature.
6
9) One at a time, removing each sample from its mold and grinding flat the back side of the mount. 10) Using 600 grit wet abrasive paper stretched across a grinding wheel for removing the bottom layer and exposing as many mineral grains as possible. Follow with 1000 grit paper. 11) Start polishing with 15μ oil based diamond paste on a polishing paper fixed to a lap. Using paper instead of cloth minimizes relief. 12) Next use 6 μ diamond polish on a similar lap. 13) Finally polish the sample with 1μ oil based diamond paste on polishing paper. Followed by .05 μ alumina in water suspension. The quality should be checked after each step. Typical polishing times are 30 minutes for 15μ, 20 minutes for 6μ, 15 minutes for 1μ and 10 minutes for.05μ. NOTE: use low speed on the polishing laps to avoid "plucking" of sample grains. 14) Samples should be completely cleaned in an ultrasonic cleaner with isopropyl alcohol or similar solvent to remove oil and finger prints. 15) To insure that no particles of lead are being cross contaminated with sample preparation procedures, a blank epoxy only) mold will be made every 50th sample following all of the above procedures. This mold will then be speciated along with the other samples. 16) Each sample be carbon coated. Once coated the samples should be stored in a clean, dry environment with the carbon surface protected from scratches or handling. POINT COUNTING
Counts are made by traversing each sample from left-to-right and top-to-bottom. The amount of vertical
movement for each traverse would depend on magnification and CRT (cathode-ray tube) size. This
movement should be minimized so that NO portion of the sample is missed when the end of a traverse is
reached. Two magnification settings should be used. One ranging from 40-100X and a second from 300-
600X. The last setting will allow one to find the smallest identifiable (1-2 micron) phases.
7
The portion of the sample examined in the second pass, under the higher magnification, will depend on
the time available, the number of lead-bearing particles, and the complexity of metal mineralogy. A
maximum of 8 hours will be spent per sample.
Speciation Results
The majority of the identifiable lead in the studied samples is found associated with manganese,
approximately 51-93% of the relative mass. The manganese-rich phases range in size from 1 to 105
microns with an average of 15 microns (Figure 1) and are generally found cementing other non-lead
silicate grains. The lead concentration in the manganese-rich phases ranges from 5.0-18 wt. %. In a few
samples a significant proportion of lead is found in small native lead inclusions associated with brass or
a tin-solder (Figure 2) although these represent only a few 1-10 particles of lead in a sample, they
contain a significant concentration of lead and thus make an impact on the relative lead mass
calculation. Since these particles are generally found as inclusions, the bioaccessible lead mass
calculation may be more helpful, as this calculation omits these particles.
8
Figure 1. Backscatter photomicrograph and x-ray spectra of manganese oxide particle.
9
Figure 2. Native lead inclusions in a tin alloy (solder).
10
In Vitro Bioassay
Lead relative bioavailabilty, was determined using the method developed at the University of Colorado,
Boulder and calibrated to EPA’s Region VIII Swine Model Medlin and Drexler, 1996, Medlin, 1997,
Drexler and Brattin 2007. The method has a high level of correlation to the Swine Model for lead
(r=0.96) however, at present the correlation for arsenic is not as good (r=0.76). Based on these data it is
recommended that one interpret arsenic bioavailability results with greater caution. No animal
calibration is available for other metals.
The method follows a carefully designed laboratory SOP, which is available on request. The procedure
uses 1.0 grams of the <250µ size fraction, this material is placed in 125ml wide-mouth HDPE bottles
along with 100ml of 1.5 pH stomach solution. The mixture is rotated end-on-end at 37ºC in a water bath
for one hour. After one hour 10ml of sample is removed, filtered (0.45µ), and analyzed for lead and/or
arsenic following Methods 6010B, 6020, or 7061A. Results from this extraction procedure are then used
to calculated bioavailable lead and/or arsenic from the bulk <250µ concentrations.
Quality assurance and a more complete SOP can be obtained at our web site:
http://www.colorado.edu/geolsci/legs
In Vitro Results
11
TABLE 1 . Preliminary Summary Of In Vitro Bioassay Results
