Chapter 1.  Methods.

Samples.

Soil and sediment samples were taken from various locations on Long Island to a depth of 12 m below the surface. Sediments vary in grain size from gravel to silt. Soil samples were taken from Fox Pond in Manorvile. Sediments were taken from construction sites in South Setauket and Port Jefferson, and from a beach cliff at David Weld Preserve in Nissequogue. In addition, a continuous Geoprobe core was collected at the Cathedral Pines County Park, in Yaphank.

Samples from Port Jefferson construction site (lat. 40056’3"N., long.73003’36"W).

Samples 1 and 5 were taken in Port Jefferson east of Sheep Pasture road near the extension of Willis Avenue north of the Harbor Hill moraine.

Sample 1 (100-120cm below the surface) consists of bright brown, coarse sand. The grains are heavily coated.

Sample 5 (160-180cm below the surface) consists of light brown to beige, medium sand to gravel.

Samples from South Setauket construction site, (lat.40054’54"N., long.7306’13"W.).

Samples 6 and 7 were taken from South Setauket. The site is located across the street from #6 Poet lane in a glacial outwash fan.

Sample 6 (140-150cm below the surface) consists of medium-coarse, orange-brown sand with layers of finer, light colored sand.

Sample 7 (200-220cm below the surface) consists of coarse, dark brown sand. Some grains are coated and some are clear.

Core from Cathedral Pines County Park (lat.40052’39"N., long.72056’39"W), Yaphank.

A continuous core of sediments from 80cm to 6m below the surface was collected by Geoprobe. Near surface samples were taken from a pit in the woods nearby. The geologic environment is a glacial stream channel. The water table was about 6m below the surface.

Sample 11 (0-30cm below the surface) - surface soil horizon. Sample is dark brown, dominantly clay and organic matter, a lot of plant material.

Sample 12 (30-60cm below the surface) - brown silt, rich in organic material

Sample 13 (60-80cm below the surface) - light brown, fine sand-silt contains pieces of partly decomposed plant material.

Sample 14 (90-120cm below the surface) - bright, brown-yellow, medium sand with some gravel and clay.

Sample 15 (150-180cm below the surface) - light brown, homogeneous, well sorted medium sand

Sample 16 (210-240cm below the surface) - light brown very heterogeneous medium sand to gravel.

Sample 17 (270-300cm below the surface) - gray-brown coarse sand contains some coated grains and some clear grains.

Sample 18 (340-360cm below the surface) - gray-beige coarse sand to gravel compacted (hard to remove sample from the tube).

Sample 19 (390-420cm below the surface) - yellow-brown, coarse sand, less gravel compared to 340-360cm sample.

Sample 20 (450-480cm below the surface) - light brown, medium-coarse sand.

Samples near Fox Pond (lat. 40053’28"N., long.72048’46"W.), Manorvile.

Soil samples were taken from near Fox Pond on the glacial outwash plain. The samples were collected from a pit. Soil is very sandy.

A1 (5-20cm)- Below the organic O horizon, contains a lot of partly decomposed organic material (leaves, needles, roots), gray-brown, sandy.

A2 (20-30cm)- depleted, eluvial horizon, gray-light gray, very sandy, containing significantly less amount of organic matter than sample A1.

B (30-50cm)- illuvial horizon, bright brown-orange color, heavily coated sand.

Samples from beach cliff at David Weld Preserve (lat.40057’58"N., long.7309’4"W), Nissequogue.

Samples were taken from a beach cliff face (lat.40057’58"N., long.7309’4"W.). These samples are from a glacial lacustrian environment.

8-10m below the surface – yellow-red colored fine, well sorted sand characterized with climbing ripples.

10-12m below the surface - pale, fine sand without distinctive coatings on the grains surfaces.

Analytical methods

Cation Exchange Capacity

0.1 M BaCl2 solution was used to displace exchangeable cations from sediment samples following the procedure of (Hendershot, 1986). Solutions were prepared from 99.999% pure BaCl2 powder from Alfa Aesar. This table shows concentration of elements in the BaCl2 powder.

elements

Concentration (ppb)

Ca

<1

Mg

<1

Mn

<1

Al

<1

Fe

<1

Na

<1

K

<100

Pure BaCl2 was used to reduce concentration of the elements in blank solution. Ba2+ ion was chosen instead of NH4+ because of its higher exchange affinity. Samples were prepared following the procedure outlined in Hendershot (1986). Three to four grams of sediment were placed in 45 ml centrifuge tube with 40 ml of BaCl2 solution so that the ratio of sediment: solution was approximately 1: 10. The samples in the BaCl2 solution were put in ultrasonic bath for 60 min then filtered. The supernatant was collected and exchangeable Ca, Mg, Na, K, Al, Fe, Mn were measured by a Beckman Instruments Direct Current Argon Plasma Emission Spectrometry (DCP-AES). The DCP analyses were done following a procedure described by McDaniel (1992). Different elements have different optimum conditions under which they may be analyzed. The major elements (Mg, Mn, K, Na, Al, Fe) were run together on a multi-element cassette. The signal was peaked on Mg. Ca was run separately with peaking on Ca. The four standards and the blank were prepared with the same 0.1 M BaCl2 solution as used for the sample preparation. All standards were prepared from a master solution by diluting certain amount of master solution with 0.1M BaCl2. The concentration of the highest standard was at least 10 times higher than the concentration of the highest sample.

Concentration of the elements in the master solution.

element

(ppm)

Ca

397

Mg

93.7

Fe

43.3

Mn

26.5

Na

161.6

Al

93.5

K

49.2

Spex Plasma Standards were used for preparation of the master solution. The calculated amount of 0.2 BaCl2 was added in the master solution so that the concentration of the Ba2+ ion in the master solution was identical to the concentration of Ba2+ ion in the solution used for sample preparation (0.1 M BaCl2). Each sample solution was analyzed 2 times. The analytical uncertainty for most samples is less than 5%, for concentrations well above detection limit. For K and Ca analytical uncertainty is approximately 10%.

Separation of the coatings

Coatings on the samples 1, 5, 6, 7 and B were separated from sand using an ultrasonic probe. Sediments were placed in 100ml plastic beaker with 40-50ml of distilled water and sonicated for 15-20 min. The supernatant with coatings was transferred into 45ml centrifuge tube. Another 40-50ml of water was added to the beaker with sediments and sonicated for 15-20 min and supernatant was collected again. This procedure was repeated up to 4 times that grains were clear (lost color). The supernatant was centrifuged for 1hr at 2000rpm. Most of the coatings settled on the bottom of the centrifuge tube. Supernatant was removed by pipette. Coatings were dried in the oven at 500C.

pH measurement

Water and BaCl2 pH was measured in the core from the Cathedral Pine County Park and samples from the South Setauket and Port Jefferson, following the procedure described in Jackson (1974). Three to four grams of sediment were placed in 45ml centrifuge tube with 40 ml of 0.1M BaCl2 solution or distilled water. The ratio of sediment: solution was approximately 1: 10. BaCl2 or distilled water suspension was put in ultrasonic bath for 60 min then filtered. The supernatant was collected and pH was measured using Titra-Line alpha titration unit.

X-ray powder diffraction (XRD)

The fine coated sand from beach cliff at David Weld Preserve (sample 10) was chosen for clay identification. The < 2m m clay fraction was separated from both the whole sand and the coatings. Clay minerals were separated from carbonates and iron oxides. Carbonates were removed by heating samples (< 2m m) in a sodium acetate-acetic acid buffer at pH 5 (Jackson, 1974). The buffered solution was prepared by dissolving 82g of sodium acetate in about 900ml of distilled water adding 27ml of glacial acid and adjusting pH to 5 by adding sodium hydroxide or acetic acid and then diluting with distilled water to 1L. Ten grams of sample were added to a beaker with 250 ml of buffered solution. The solution was stirred on a hot plate at low enough temperature to avoid boiling. The evidence of complete reaction is absence of bubbles. Sample was then washed several times with distilled water by centrifuging. We did not observe any evidence of carbonate in these samples.

It was necessary to remove iron oxides, because we used CuKa radiation. Fluorescent X-rays from Fe produce a high background that masks peaks. Also iron oxides cement the clay particles together inhibiting dispersion. Iron oxide was removed chemically with citrate-bicarbonate-dithionite (Jackson, 1974). The carbonate free samples were placed in a beaker in which 40-ml of 0.3N Na-citrate solution and 5 ml of 0.1 N NaHCO3 were added. The suspension was warmed to 800C on a hotplate, one gram of solid Na2S2O4 was then added into the solution. The suspension was stirred constantly for the first minute and occasionally for 15 min. then 10-ml of a saturated NaCl was added. The suspension was centrifuged with distilled water at least 5 times. The resulting material was then ultrasonically disaggregated.

The fraction < 2m m was separated for XRD by centrifuging sample in 45 ml tube at 750 rpm for 3.3 min. The supernatant liquid was then decanted into another centrifuge tube. The centrifuge tube containing the clays fraction < 2m m was then centrifuged at 2000rpm for 20 min. The supernatant was removed. The residue containing the clay minerals was then pipetted onto a glass slide and left to dry at room temperature.

X-ray patterns were recorded from 3 to 50 2q degrees using CuKa radiation using a Scintag powder diffractometer operated at 45kV and 25mA.

Transmission Electron Microscopy (TEM)

Transmission electron microscopy analysis was made on the separated coating from one heavily coated sand (sample1) from Port Jefferson. This sample was sonicated in deionized water for 20 min. One drop of the dispersed coatings was then pippeted onto a porous carbon film grid and left covered to dry at room temperature. The TEM analysis follows that outlined by Barber, Reeder, and Smith (1984). The JEOL 200-CX TEM instrument fitted, with an EDS systems for micro-chemical analysis, was operated at an accelerating voltage of 200 keV. The sample was placed in a low-background graphite sample holder tilted at 35 degrees. Counting times were 200 sec. Integrated intensities of peaks were determined after background subtraction. Integrated peak intensity ratios were converted to atomic concentration ratio using the Cliff and Lorimer relation (1975). The Cliff and Lorimer give the equation:

CA/ CB = kAB * IA/ IB                                                                           (eq. 1)

Where CA and CB are the atomic concentrations of elements and A and B in a given compound, IA and IB are the intensities of emission lines of those elements and kAB is a proportionality factor. The kAB is experimentally determined for a standard compound of known composition. Si is element B, because of its ubiquitous occurrence in these samples.

Scanning Electron Microscopy (SEM)

Untreated coated sand (samples1 and 5 from Port Jefferson and sample 10 from the beach cliff at David Weld Preserve) were characterized using JEOL scanning microscope 5300 in the University Microscopy center at SUNY Stony Brook. The dry loose sand sample was placed on a holder covered with a carbon film and coated with gold prior to SEM analysis.

Sorption of hydrophobic organic compounds

The sorption partition coefficient between solids and water for hydrophobic organic compounds was determined on whole sand, separated coatings, grains with coatings removed, whole sand treated with ultrasonic probe but with removing the coatings and sand with iron compounds removed were studied. Hot HCl extraction was used to remove iron compounds. Extraction was done following the procedure described in Coston et al, (1995). The samples were leached with 4M HCl (3ml of acid per gram of sediment) for one hour at 1000C in Teflon beakers. Then the sediments were placed in centrifuge tubes and washed by centrifuging with distilled water at least 6 times. Then the sediments were air-dried.

The method used for determining sorption of hydrophobic organic compounds onto the sediments following the principals of a common headspace partitioning designed for determining partition coefficients of organic compounds and hydrophobic interactions in proteins (Wishnia and Pinder, 1966).

Hexachlorobenzene (HCB) was chosen as the hydrophobic organic compound, because HCB is highly sorbed on solids and is not degraded by microorganisms. 14C - labeled HCB (15.9 mCi/mmol) was used as sorbate.

A vessel was designed for the experiment. An inner space containing pure water is isolated from an outer space containing the sediment suspension. The two compartments share the common headspace through which volatile organic compounds can partition. At the equilibrium, dissolved phases of HCB are equal in an inner and an outer compartments. The amount of sorbed HCB can be determined by comparing the concentration of dissolved compounds and the total concentration (dissolved and partitioning) of the outside solution in the vessel with sediments; or by the difference in concentrations of the inner compartment (dissolved only) in the vessel with sediments and control vessel (see calculations below). The top of the vessel was sealed with an aluminum foil under a screw-capped lid. Controled mass-balance studies showed no significant leakage of HCB out of the vessels over the first 48 hours. However, slow leakage was apparent for longer periods. Therefore, sorption was limited to 48 hours. The vessels were treated with a solution of K2Cr2O7 in concentrated H2SO4 to oxidize and remove any organic material in the bottles between uses. The vessels were then washed with distilled water, acetone, hexane and dichloromethane (DCM). The vessels were air-dried for 2 hours and a known mass of sediment was then placed in an outer part of the bottle. Ten ml of distilled water was added into both an inner and an outer parts. The vessels were placed into an ultrasonic bath for 10 min. Then 7.5 microliters of 14C labeled HCB (159 mCi/ml) in methanol were added into an outer part of the vessel. The vessels were closed tightly and put on the shaker at the constant temperature (250C). Duplicates for the control vessels and the vessels with samples were prepared each time.

Kinetic sorption experiments were conducted. Pure water in both inside and outside compartments and the sediments samples were equilibrated for 2 hours, 8 hours, 24 hours, 72 hours, 10 days and 30 days. We found out that system needed just over 24 hours for the dissolved phase to reach equilibrium. When the system is at equilibrium the concentration of 14C labeled HCB is identical in the inner and in the outer part of the vessel (checked using the control vessels). There was a slight loss of HCB after 72 hours. This loss increased with time. Therefore, we used 48 hours equilibration in all of our studies.

The control vessel for each experiment consistently showed equilibration and good mass-balance at 48 hours. After 48 hours the replicated 2ml aliquots were taken from the inner part of the vessel with sediments and from both the inner and an outer part of the control vessel. Samples were placed into scintillation vials and 10 ml of ULTIMA GOLD LLT cocktail was added. C14 was counted on a LKB 1217 RACKBETA Liquid Scintillation counter.

The distribution coefficient (Kd) was calculated based on assumption that

M total = M water + M sorbed + M vapor                                       (eq. 2)

Where M total is the total mass of HCB added into the system, M water is the mass of HCB dissolved in the water, M sorbed is the mass of HCB sorbed onto the sediments and M vapor is the mass of HCB in the vapor phase of the vessel. Experiments were done with 14C labeled HCB and the measured disintegrations per minute (DPM) were proportional to mass.

DPM total = DPM water +DPM sorbed + DPM vapor + DPM wall              (eq. 3)

DPM total was measured. DPM water was measured in the vessel containing the sediments. DPM wall was found to be negligible.

For soluble compounds like HCB the vapor phase concentration is directly proportional to the dissolved phase through Henry’s Law (Schwarzcenbach et. al, 1993). DPM vapor in the treatment can be calculated by determining the ratio of HCB in the vapor to that in the water in the control vessels

(DPM total – DPM water) / DPM water and multiplying this ratio by the measured DPM water in the vessel with the sediments. This calculation assumes 100% mass-balance and that all of the 14C is in water, vapor or sorbed onto sediments. The calculated fraction in the vapor phase was in the reasonable agreement with the Henry Law constant of HCB. DPM water was measured in the vessel containing the sediments. DPM vapor in the vapor phase of the vessels was calculated as

DPM vapor = ((DPM total - DPM water control bottle)/ DPM water control bottle))*DPM water sample          (eq. 4)

DPM sorbed on the sediments was calculated as

DPM sorbed = DPM total – (DPM water sample + DPM vapor)           (eq. 5)

The fraction of the sorbed HCB (f sorbed) was calculated.

f sorbed = DPM sorbed onto the sediments/ (DPM sorbed onto sediment +½ DPM water sample)  (eq. 6)

Half the DPM water sample was used in the calculations because the sediments were placed only into the outer part of the bottles, which contains half the total volume of water. The fraction HCB dissolved in the water was calculated as one minus fraction HCB sorbed onto the sediments.

f dissolved = 1- f sorbed                                                            (eq. 7)

The distribution coefficient (Kd) was calculated as

Kd = fs / fd * (mass (of sediments) / volume (of water))                                  (eq. 8)

Where fs is the sorbed fraction of HCB, fd is the fraction of HCB dissolved in the water. The mass of the sediments used for experiments was expressed in grams and the volume of distilled water, added into the outer part of the bottle, expressed in ml (in our case it was 10ml).

Surface area measurements

The surface area was analyzed for sand samples 1, 5, 6 and 7 and for coatings separated from the same samples. Samples were measured using a multi-point BET (Brunauer-Emmett-Teller) method. Samples were analyzed at the Pennsylvania State University Material Characterization Laboratory by Tom Rusnak.

Total Organic Carbon (TOC)

Total organic carbon (TOC) was analyzed in whole sand and coatings for samples 1,5,6 and 7. David Hirschberg did the analyses at the Marine Science Research Center SUNY Stony Brook. Total organic carbon was measured on a CHN Elemental Analyzer. The whole sand samples were ground to a fine powder followed by 2M HCl treatment to remove carbonate. The samples were then washed with distilled water at least 6 times and dried at 500 C.

One portion of the coatings was untreated. Another portion was treated using the procedure of Mayer (1994). In this procedure the coatings were placed in a beaker and kept in a dessicator with concentrated HCl overnight. The coatings were then dried overnight at room temperature and then dried in an oven at 700C. This method avoids leaching organic carbon bonded to iron compounds.

Return to Table of Contents