POTENTIAL PAH RELEASE FROM CONTAMINATED SEDIMENT IN
GALVESTON BAY-HOUSTON SHIP CHANNEL
Chunlong Zhang, Gabriel Zheng, Gregory Holston, and George Lambert
University of Houston-Clear Lake, Houston, Texas 77058
Overview
Polycyclic aromatic hydrocarbons (PAHs) are a class of compounds consisting of two or
more fused aromatic rings. They represent the largest group of compounds that are
mutagenic, carcinogenic, and teratogenic and are included in the U.S. EPA priority
pollutants lists. In recent years, increasing attention has been drawn to PAH
contamination in aquatic sediments. High concentrations of PAHs have been reported in
various sediments from urbanized as well as pristine environment (Fernandez et al., 1999;
Ghosh et al., 2000; Li et al., 2001; van Metre et al., 2000; Yunker et al., 1996).
PAHs originate from both natural and anthropogenic sources such as terrestrial deposit of
coals, atmospheric input as a result of incomplete combustion (wood burning, forest fire,
fossil fuel, coke oven), oil seeps and spills, and highway dust associated with vehicle
exhaust (van Metre et al., 2000). Despite their widespread distributions, PAHs can be
ultimately deposited and persisted in bed sediment (as a sink) in the aquatic system. This
is due largely to the fact that most PAHs sorb strongly to the organic in sediments
because of their high hydrophobicity, and are resistant to bacterial degradation under
anoxic environment. When environmental conditions become favorable, however, PAHs
will be released to the overlying water as a long-term source and pose potential threat to
water quality and aquatic ecosystem via bioaccumulation in food chains.
The work presented herein describes our recent effort in investigating the potential
release of PAHs from sediments collected from Houston-Galveston area. Galveston Bay
is one of the largest estuaries on the U.S. coastlines. It is also the base for one of the
largest refineries in the world. Of particular interest to this study is the sediment sample
from Houston Ship Channel where various sources of PAHs are likely to be profound and
frequent waterborne transport may result in secondary pollution of overlying water due to
sediment re-suspension and enhanced sediment-to-water mass transfer processes such as
diffusion, dissolution, and desorption. Irvine et al. (1997) indicated that the passage of
large ships in harbors and navigable channels had significant impact on water quality due
to the disturbance and the resuspension of bed sediment.
Extensive work has been done to fingerprint PAHs source(s) and quantify sediment flux
in contaminated sediments of several harbors and waterways including Boston Harbor,
San Francisco Bay, and Milwaukee Harbor (McGroddy et al., 1996; Ghosh et al., 2000).
Unfortunately, the current database for PAHs in our region are scarce and our
understanding on the behavior of PAHs in sediment-water interface is very limited. A
recent study by Gill et al. (1999) indicated the importance of sediment in controlling and
mediating the concentration of several trace elements and nutrients in Galveston Bay. To
date no such information is available on the behavior of PAHs in aforementioned water
systems. This work was initiated in order to gain a better understanding of the mass
transfer of PAHs at the sediment-water interface. In particular, the emphases of this work
are to: (1) develop a thermodynamic equilibrium model that describes the extent of PAH
exchange between sediment and the overlying water, (2) quantify the kinetic rate of PAH
sorption at the water-sediment interfaces and estimate the aqueous flux rate of PAH, (3)
elucidate the key factors in contributing elevated PAH concentrations in the aqueous
phase.
Methods
Chemicals including naphthalene, phenanthrene and pyrene were selected as the probe
PAHs, as they represent a wide range of aqueous solubilities (naphthalene: 31.7 ppm;
phenanthrene, 1.29 ppm; pyrene: 0.135 ppm). These PAHs compounds and their
calibration standards were purchased from Aldrich Chemical Co. and Suppelco. Six
sediments were collected in Houston-Galveston area (Table 1). Samples were air-dried
and sieved for batch sorption studies, and subsamples were also stored at –70oC for PAH
characterization.
Table 1. Test Sediment and Sampling Location
Location
Horsepen Bayou near UHCL (HPB)
Clear Lake near Hilton Hotel (CLK)
Galveston Bay near Kemah Post 4-6 (BAY)
Houston Ship Channel, Barge Holding Area (HSC1)
Houston Ship Channel, Lynchberg Ferry (HSC2)
Houston Ship Channel, New Fed Hartman Bridge (HSC3)
GPS Reading
29o34.789N/95o05.542W
29o19.256N/94o55.229W
29o33.096N/95o02.181W
29o47.251N/95o03.932W
29o45.840 N/95o04.833W
29o42.253 N/95o01.059W
Water Depth
(feet)
4
6
8
27
50
51
Three PAHs spiked in air-dried sediment samples for sorption study were extracted with
EPA SW 846 method 3550B – three successive ultrasonic extractions (1 g sediment in 5
mL 50:50 mixture of hexane and acetone, 15 s on and 15 s off pulsing for a total of 3
min). Aqueous PAHs were extracted using 3 mL methylene chloride in 10 mL sample.
Wet samples for sediment characterization were extracted using a more rigorous Soxhlet
extraction (3540C) without the use of a clean-up procedure (EPA method 3630C) to
characterize other potential contaminants. Extracts were analyzed on 5890 GC coupled
with 6972 MS detector with a 30 m x 0.32 mm x 1 µm HP-5 capillary column. The oven
was held at 35°C for 10 minutes and ramped to 280°C at 12°/min. The MSD was run in
Scan mode for sediment characterization and SIM mode for PAH quantification (128,
178 and 202 as the quantifier ions for naphthalene, phenanthrene and pyrene,
respectively).
Batch sorption studies for the acquisition of distribution coefficient (Kd) of each
individual PAH were performed according to procedures described previously by Chiou
et al. (1998) and Johnson et al. (2001). Into a series of 50-ml Teflon centrifuge tubes
containing 1 – 2 g sediment, 20 mL DI water and various volumes (~100 L) of PAH
stock solution in methanol were introduced to the sediment slurry. The sediment to water
ratio were adjusted to allow significant adsorption of the test compound as well as to
provide sufficient aqueous solution for quantitative measurement. Test concentrations of
PAHs were determined based on the aqueous solubility. The sediment slurry was
continuously mixed on a shaker at room temperature (22oC) for 48-72 h, and the solid
and aqueous phases were separated by centrifugation. Aqueous phase PAH
concentrations were analyzed after extraction into dichlormethane (Jee et al., 1998) and
corrected with % recovery with known standards. Sorbed phase concentrations were
determined by the difference of the total mass and the mass in solution. Batch desorption
study were conducted in 50-mL Teflon centrifugal tubes with screw cap to minimize
PAH volatilization. The same sediment/water ratio used in the previous sorption studies
were followed, but with the addition of HgCl2 (100 mg/L) to prevent any bacterial
degradation.
Results
1. Sediment PAH adsorption isotherms
Three of the six sediment samples have been tested for their adsorption isotherms of each
PAH compound. The isotherms for naphthalene, phenanthrene and pyrene are shown in
Figures 1, 2 and 3, respectively. Data were fitted to Freundlich and Langmuir isotherm
models. Sediment adsorption isotherms were found to be linear in the range of
concentration tested.
200
Cs(mg/kg)
150
100
50
0
Galveston Bay
Clear Lake
Horsepen Bayou
0
5
10
15
20
25
Caq(mg/L)
Figure 1. Adsorption isotherm of naphthalene in various sediments
250
Cs(mg/kg)
200
150
100
50
0
Galveston Bay
Clear Lake
Horsepen Bayou
0.00
0.25
0.50
0.75
1.00
Caq(mg/L)
Figure 2. Adsorption isotherm of phenanthrene in various sediments
8
350
7
300
Cs(mg/kg)
6
250
5
200
4
150
3
100
2
50
1
Galveston Bay
Clear Lake
Hosepen Bayou
0
0.00
0.03
0.06
0.09
Caq(mg/L)
0.12
0.0
0.5
1.0
1.5
2.0
2.5
0
3.0
Caq(mg/L)
Figure 3. Adsorption isotherm of pyrene in various sediments
The sediment-water partitioning coefficients Kd (L/kg) were determined according to the
formula: Kd = CS / Caq, where CS and Caq are sediment phase concentration (mg/kg) and
aqueous phase concentration (mg/L), respectively. A summary of Kd values is given in
Table 2. For instance, Kd values of phenanthrene for Clear Lake, Galveston Bay, and
Horsepen Bayou were 698, 298, and 52 L/kg respectively. The Kd value attributes to the
organic contents of the sediment samples. The Horsepen Bayou is a clayed sediment with
low organic content, while two others are silt with Clear Lake sediment being the most
heavily contaminated.
Table 2. Summary of sediment-water distribution coefficient (Kd in L/kg)
Horsepen Bayou
Galveston Bay
Clear Lake
Naphthalene
Kd = 4.85
Kd = 2.30
Kd = 6.41
2
2
(R = 0.975)
(R = 0.992)
(R2 = 0.963)
Phenanthrene
Kd = 52.0
Kd = 298
Kd = 698
2
2
(R = 0.880)
(R = 0.983)
(R2 = 0.983)
Kd = 52.7
Kd = 140
Kd = 477
Pyrene
(R2 = 0.965)
(R2 = 0.967)
(R2 = 0.956)
Assuming an average sediment organic carbon (SOM) of 5% (foc = 0.05), the log Koc
values calculated from SOM-normalized Kd compare favorably (with an exception of
pyrene) to those determined by Chio et al. (1998) who reported log Koc values are 2.613.07, 4.10-4.64, and 4.96-5.45 for naphthalene, phenanthrene and pyrene, respectively.
Analytical error may have occurred due to the very low concentration of pyrene tested in
the sorption study.
The experimentally determined Kd value can be used to estimate the equilibrium
concentration of PAHs in sediment pore water using historical data available for sediment
PAHs. Our linear model is consistent with that reported by Chio et al. (1998), however, it
may have limited utility as others suggested nonlinear Freundlish isotherm (Johnson and
Weber, 2001). Note also that the slowly desorbing fraction of PAHs results in lower Caq,
and the Kd thus obtained in short-duration study (hrs – days) will be higher than the actual
equilibrium Kd. Therefore, use of such Kd will probably underestimate aqueous phase
concentration (Caq) as well as PAH flux.
2. Factors affecting PAH sorption
The adsorption / desorption kinetics have been tested in a batch study over a 3-week
period and major factors that have potential effect on PAH partitioning have been
investigated. Several important factors that may contribute to elevated PAH
concentrations in the aqueous phase were tested at various simulated conditions:
synthetic seawater, actual water at the same location, and DI water at pH 3, pH 5, pH 7
(static vs. mixing), and pH 9. Preliminary results have shown that mixing significantly
enhanced the mass transfer and therefore the sorption rate, especially for a low organic
Horsepen Bayou sediment (Figure 4). The effects of pH and composition of overlying
water were minimal presumably due to the nonionic / hydrophobic adsorption of PAH
compounds. Under most conditions, the PAHs appeared to irreversibly bind to sediment
in a 3-week period and remained not detectable. PAH desorption was noted at low pHs
and for a treatment with seawater (data not shown).
16
Clear Lake / Static
Galveston Bay / Static
Horsepen Bayou / Static
Clear Lake / Mixing
Galeveston Bay / Mixing
Horsepen Bayou / Mixing
14
Naphthalene (mg/L)
12
10
8
6
4
2
0
0
5
10
15
20
25
Time (Days)
Figure 4. Sorption / desorption of naphthalene in various sediments: Static vs. mixing
conditions (Sediments were spiked with 15 ppm naphthalene)
3. Sediment PAH desorption rate
Desorption of PAHs from soil/sediment is often biphasic – a rapid desorption from hours
to days followed by a slow process from months to even years. The slowly desorbing
fraction is related to intraorganic matter diffusion and hindered pore diffusion (Johnson et
al., 2001), and can be expressed in the following three-parameter model:
q (t )
k t
  s e ks t  (1   s ) e r
qo
where q(t) – the sediment phase PAH concentration at a given time, qo – initial sedimentphase PAH concentration, s – slowly desorbing fraction, 1-s – rapidly desorbing
fraction, ks and kr – apparent first-order rate constants for the slowly and rapidly
desorbing fractions, respectively.
Table 3 lists the typical values of three parameters in the above kinetic model. The time
required to reach pseudo desorption equilibrium predicted by this model varies from 2
weeks (Cornelissen et al., 1998) to about 8 months. Figure 5 is a simulated desorption
profile based on the typical values chosen from Table 3, i.e., s = 0.7, ks = 1.5x10-3 day-1,
and kr = 0.2 day-1. The difference in the kinetic rates between slowly desorbing and
rapidly desorbing PAHs is evident. Figure 5 also implies that experiment duration
commonly used in desorption studies (days to weeks) are inappropriate for PAHs when
slowly desorbing fraction may be significant.
Table 3. PAH desorption rate parameters for the three-parameter biphasic first-order
desorption model
Source:
ks (day-1)
kr (day-1)
s
Ghosh et al. (2001)*
0.06
5.01x10-3
0.80
¶
4
Cornelissen et al. (1998)
0.2-0.74
1.92-4.56x10
19.2-67.2
Johnson and Weber (2001)§
0.69-0.736
1.38-3.01x10-3
0.134-0.362
Johnson et al. (2001)§
0.657-0.736
1.38-2.63x10-3
0.134-0.332
¶
§
*Clay/silt (<63 µm); 2-4 ring PAHs; Phenanthrene in a sediment containing 8.27%
sediment organic matter.
PAH fraction remained: q(t)/q 0
Sum of slowly and rapidly
desorbing PAHs
(B)
1.0
1.0
0.9
0.9
0.8
0.8
0.7
0.7
0.6
0.6
0.5
0.5
0.4
0.4
0.3
0.3
0.2
0.2
0.1
0.1
0.0
0.0
0
100
200
300
Time (days)
400
500
0
100
200
300
400
PAH fraction remained: q(t)/q 0
Slowly desorbing PAHs
Rapidly desorbing PAHs
(A)
500
Time (days)
Figure 5. Simulated PAHs desorption profile from a contaminated sediment
4. Sediment PAH flux rate
Contaminant flux from sediment is important especially in shallow water in Texas
estuaries including bay waters (e.g., 6-12 feet in Galveston Bay). We have not been able
to give rigid estimate of PAH flux from contaminated sediment, our effort herein is to
provide an order of magnitude estimate on the flux rate based on the literature values
listed in Table 4.
If the flux rate in Lake Michigan is assumed, the mass flux rate of PAHs in Ship Channel
(length: 51 miles; width: 400 ft) can be estimated in the range of 0.9 ~ 10.5 kg/yr. This is
equivalent to the contribution of an elevated concentration in the range of 6.5 ~ 76.6 ppt
in the overlying water assuming the average depth of 45 ft and a hydraulic retention time
of one year. Note that the actual PAH concentration in the sediment pore water adjacent
to sediment bed may be significantly higher than the estimated concentration due to the
limitation of mass transfer to the overlying water. Elevated PAH concentrations in
sediment porewater could pose additional stress to benthic organisms in the Ship Channel
and bay area. Similarly, the total PAH flux in Galveston Bay can be estimated to be in the
range of 140 ~ 1,630 kg/yr based on a surface area of 600 square miles. This is
comparable to 371 kg/yr of PAHs contribution from the lower watershed nonpoint
sources (The Galveston Bay National Estuary Program, 1994).
Table 4. PAH sediment fluxes in urban/industrial lakes and coastal areas in the U.S.*
Location
Type of water
PAH flux (µg/m2/yr)
Lakes at Northern Great Lakes
Remote lakes
(105 ─ 550)
Lakes in Northern Florida
Remote lakes
(280 ─ 540)
Lakes in Northern New England Remote lakes
(130 ─ 680)
Lake Michigan
Urban/Industrial lake 90-1050
Basin of Puget Sound, WA
Coastal areas
(3100)
Narragansett Bay, RI
Coastal areas
400 ─ 3100
*Adapted from Fernandez et al. (1999). PAH value corresponds to the sum of all parent
compounds except perylene and retene. In parentheses, sum of phenanthrene,
fluoranthene, pyrene, benzo[a]anthracene, chrysene, benzofluranthenes, benzo[e]pyrene,
benzo[a]pyrene, indeno[1,2,3-cd]pyrene, and benzo[ghi]perylene.
Conclusion
This initiative work has raised some interesting and challenging aspects with regard to
PAHs exchange at the sediment-water interface. Low concentrations of PAHs
accompanied with complex sediment matrix require time consuming processes for
sample preparation / clean-up and costly analysis of PAHs. The slowly desorbing process
necessitates desorption study to last several months to one year or even a longer period to
reach the desorption plateau, which makes mass balance a difficult task. Further research
are needed to determine thermodynamic and kinetic parameters in order to better
understand the fate and transport of PAHs in water-sediment system. Studies are under
way to fully characterize the PAHs in the obtained sediments, to examine the controlling
factor(s), to quantify the rate of release from sediments, and to validate the desorption
model using experimental data. Data generated from this study are expected to increase
the awareness of water quality issue related to contaminated sediment in the HoustonGalveston area. The results obtained will also be of significance to pollution prevention,
risk assessment and water quality management for PAH-contaminated sediments
associated with ship passage and dredging operations in harbors and navigable channels
in our region.
References
1. Chiou, C.T., S.E. McGroddy, D.E. Kile (1998), Partitioning characteristics of
polycyclic aromatic hydrocarbons on soils and sediments, Environ. Sci. Technol.,
32(2):2769-2775.
2. Cornelissen, G, H. Rigterink, M.M.A. Ferdinandy, P.C.M. Van Noort (1998), Rapidly
desorbing fractions of PAHs in contaminated sediments as a predictor of the extent of
bioremediation, Environ. Sci. Technol., 32(7):966-970.
3. Fernandez, P., R.M. Vilanova, J.O. Grimalt (1999), Sediment fluxes of polycyclic
aromatic hydrocarbons in European high altitude mountain lakes, Environ. Sci.
Technol., 33(21):3716-3722.
4. Galveston Bay National Estuary Program (1994), The State of the Bay. A
Characterization of the Galveston Bay Ecosystem. Galveston Bay National Estuary
Program Publication GBNEP-44, 232pp.
5. Ghosh, U., J.S. Gillette, R.G. Luthy, R.N. Zare (2000), Microscale location,
characterization, and association of polycyclic aromatic hydrocarbons on harbor
sediment particles, Environ. Sci. Technol., 3499):1729-1736.
6. Ghosh, U., J.W. Talley, R.G. Luthy (2001), Particle-scale investigation of PAH
desorption kinetics and thermodynamics from sediment, Environ. Sci. Technol.,
33(17):3468-3475.
7. Gill, G.A., N.S. Bloom, S. Cappellino, C.T. Driscoll, C. Dobbs, L. McShea, R.
Mason, J.W.M. Rudd (1999), Sediment-water fluxs of mercury in Lavaca Bay, Texas,
Environ. Sci. Technol., 33(5):663-669.
8. Irvine, K.N., I.G. Droppo, T.P. Murphy, A. Lawson (1997), Sediment resuspension
and dissolved oxygen levels associated with ship traffic: Implications for habitat
remediation, Water Quality Research Journal of Canada, 32(2):421-437.
9. Jee, V., D.M. Beckles, C.H. Ward, J.B. Hughes (1998), Aerobic slurry reactor
treatment of phenanthrene contaminated sediment, Water Res., 32(4):1231-1239.
10. Johnson, M.D. and W.J. Webber, Jr (2001), Rapid prediction of long-term rates of
contaminant desorption from soils and sediments. Environ. Sci. Technol., 35(2):427433.
11. Johnson, M.D., T.M. Keinath II, W.J. Weber, Jr (2001), A distributed reactivity
model for sorption by soils and sediments. 14. Characterization and modeling of
phenanthrene desorption rates, Environ. Sci. Technol., 35(8):1688-1695.
12. Li, K., E.R. Christensen, R.P. Van Camp, I. Imamoglu (2001), PAHs in dated
sediments of Ashtabula River, Ohio, USA, Environ. Sci. Technol., 35(14):2896-2902.
13. McGroddy, S.E., J.W. Farrington, P.M. Gschwend (1996), Comparison of the in situ
and desorption sediment-water partitioning of polycyclic aromatic hydrocarbons and
polychlorinated biphenyls, Environ. Sci. Technol., 30(1):172-177.
14. U.S. Environmental Protection Agency, SW-846, Test Methods For Evaluating Solid
Waste, Physical/Chemical Methods, Aug. 2001.
15. van Metre, P. C., B.J. Mahler, E.T. Furlong (2000), Urban sprawl leaves its PAH
signature, Environ. Sci. Technol., 34(19):4064-4070.
16. Yunker, M.B., L.R. Snowdon, R.W. MacDonald, J.N. Smith, M.G. Fowler, D.N.
Skibo, F.A. McLaughlin, A.I. Danyushevskaya, V.I. Petrova, G.I. Ivanov (1996),
Polycyclic aromatic hydrocarbon composition and potential sources for sediment
samples from the Beaufort and Barents Seas, Environ. Sci. Technol., 30(4):13101320.