Identification Distribution Pattern and Origin of Petroleum Hydrocarbons in Core Sediment of Southwest Coast of the Caspian Sea (Kiashahr)

I. INTRODUCTION
Petroleum hydrocarbons are as the most ubiquitous organic contaminants worldwide in the marine area. The hydrocarbons in sediment cores have been used to identify source and to reconstruct the historical records of these hydrocarbon inputs for environmental impact studies. Of these, the n-alkanes are commonly used to characterize organic matter of water, suspended matter and sediments from various environments (Bixiong et al., 2007).
Similarly, a large number of terpane, hopane and sterane biomarker parameters are important tools for discriminating between biogenic and anthropogenic origins hydrocarbons deposited in sediments, confirming a petroleum contribution. (Ou et al., 2004; Gao et al., 2007).
Caspian Sea is the largest freshwater lake in the world. International oil and gas industry special attention always was paid to this lake especially since early 1900s. Oil fields in Azerbaijan covered the southern part of the Caspian Sea where the first data on oil extraction go back to the seventh century and at the beginning of the 17th century. Also there are oil resources in Kazakhstan and Turkmenistan that 30–40% of these are located in offshore (Silva et al., 2012). There was not exploration from Iran wells (Anzali wells) but since natural seeps should also be taken into consideration, these wells can be one of the possible sources of the hydrocarbon in sight. However, there has been no report on biomarkers (anthropogenic hopanes and steranes) in sediment cores to identify origin reconstruct a history of oil pollution.
In this respect, this study focuses on the determination of composition, concentration and origin of hydrocarbons based on the examination of the following geochemical markers: n-alkanes, isoprenoid alkanes, petroleum biomarkers in one core sediments from the southwest Coast of the Caspian Sea.

II. MATERIALS AND METHODS
A. Sample Collection
Core sediments were collected using a gravity corer at Kiashahr (CK) at the depth of 20 m on 25 October 2012. In the field, the cores were sectioned at into 1, 2 and 5 cm intervals immediately after sampling in the upper 10 cm, from 10 to 30 cm and higher than 30 cm to down of the core, respectively (a total of 35 samples)
Samples were stored in clean aluminum foil and transported to the laboratory in a cool box and then stored at 21˚C until further analysis.
B. Extraction and Fractionation
Sediment samples were freeze-dried and about 5 g of each freeze-dried sample were extracted by soxhlet apparatus using 100 ml of dichloromethane; over 8 h. The extracts were transferred onto the top of 5% H2O deactivated silica gel column. All Hydrocarbons were eluted with 20 mL of DCM/hexane (1:3, v/v) and transferred onto the fully activated silica gel column (0.47 cm i.d._18 cm) and eluted with 4 ml hexane to obtain the aliphatic hydrocarbon fraction (AH) fraction.
C. Analytical Methods
GC/MS analyses were carried out on an Agilent Technologies (Palo Alto, CA, USA) instrument with a gas chromatograph (GC), model 7890A coupled to a quadruple mass spectrometer (MS), 5975 C. Compound identification was based on individual mass spectra and GC retention times in comparison to literature, library data, and authentic standards. Detection of hopanes and steranes, was carried out using mass fragment ions at m/z = 191 and 217 respectively.
III. RESULTS AND DISCUSSION
A. Aliphatic Hydrocarbon
The total aliphatic hydrocarbon (TAH) concentrations varied from 5.11 to 643.30 μg g−1 in CK. According to results; TAHs had maximum concentrations at 3, 4, 8, and 55 cm that were concomitant to an increase of U/R and diagenic hopane, pointing out on a greater enrichment of petrogenic organic matter in these layers.
The distribution pattern of n-alkane in sediments of Ck presented bimodal pattern in vertical profile: First model that happened in most depths containing an n-alkane distribution with not obvious odd to even predominance with high Unresolved Complex Mixtures )UCM( indicating petrogenic input in these layers (Maioli et al., 2010). The second model that happened in some of the bottom depths of CK contains an n-alkane distribution with odd to even predominance and Cmax was at C27. These distributions indicated biogenic source for aliphatic hydrocarbons in these samples (Gao al., 2007).
The relative UCM abundance compared to the resolved alkanes (R) is served as diagnostic criteria of pollutant inputs that it is varying from 0.5 to 6.8 in CK. UCM/R values >2 confirm the widespread presence of petroleum-related residues (Silva et al., 2012). This is apparent in all layers of CK, except for layers (75, 85, 90, 95 and 105 cm) confirming that these sediments contain mainly contributions from perogenic sources.
The n-alkanes from petrogenic inputs show CPI ratios of approximately 1, while CPI values of 4 to 10 have been recorded for the terrestrial plants (Simoneit, 1986; Lipiatou and Saliot, 1991) Our results showed CPI values (0.83- 1.2) in most sediments of Ck. CPI values higher than 2 were observed only in layers (75, 85, 90, 95 and 105 cm) of CK indicating characteristic of biogenic n-alkane input.
The presence of pristane (Pr) and phytane (Ph) is noticed in all sections. Pri/Phy ratios for uncontaminated sediments are typically between 3 and 5, whereas values close to or lower than 1 suggest a petrogenic origin (Volkman et al., 1992). The Pr/Ph ratio varied from 0.01 to 0.9 in Ck. This low Pr/Ph values suggest predominant petrogenic sources rather than biogenic in this core.
B. Petroleum Biomarkers
Petroleum biomarkers pentacyclic triterpanes (C27 to C35 carbon atoms) were found at most of the samples. In deferent depths, hopanes ranged within 23 to 1841 ng g-1. Proportions of terpanes to the major hopanes [∑Ter/ (∑Ter + Tm+ C29αβ + C30αβ)] were found to be low for all the samples (0.16 to 0.33), indicating higher concentrations of these catagenetic hopanes. The high relative abundance of C29αβ and C30αβ hopanes is indicative of pollution from fossil fuel products (Hauser et al., 1999).
Several geochemical values elaborated from hopane biomarkers have been used to characterize petroleum residues. One of these ratios that is confirmation of petroleum contaminants is determined by values of the ratio 18α(H)-22,29,30-trisnor-hopane (Ts)/sum of Ts+Tm17α(H)-22,29,30-trisnor-hopane ranged from 0.33 to 0.8 in CK. Such a range confirms the presence of mature petroleum in sediment.
The ratio of 22S/(22S+22R) epimers of the homohopanes for C31αβ-hopanes was from 0.56 to 0.66 in, very close to the equilibrium value of full maturity at 0.6. This criterion has been used for characterizing the origin and degree of maturation of crude oils.
Bacterial (biogenic) hopanes (particularly neohop-13(18)-ene, 17β(H),21β(H)- hopane and 17β(H),21β(H)-hop-22(29)-ene) generally are absent in most of these samples. The highest values of bacterial hopanes with concentrations ranging from 0.08 to 0.50 ng g−1 were observed at 22, 24 and 60 which could reflect the presence of bacteria to organic matter in these layers (Philp, 1985; Harris et al., 2005).
Steranes present in fossil fuels are also useful biomarker indicators for petroleum pollution in urban coastal areas. The sterane patterns showed a prevalence of 5α,14β,17β and 5α,14α,17α configurations occurring as 20S and 20R epimers. The total sterane concentrations in the study area ranged from 2.65 to 768.84 ng g_1 in CK.
Another evidence of the contamination by crude oils in core sediment is steranes present. In general, C27 and C29 steranes are indicative of algae and higher plant source of organic matter respectively. So, the a dominance of C27 over C28 and C29 steranes indicates the predominance of organic matter input from marine algae, while a predominance of C29 steranes suggests a preferential higher plant input.
In these sediments, the ratios of C27/C29-steranes are in the range of 0.5–0.74, further indicating an abundance of organic matter with higher plant origin in the source rocks for these petroleum contaminants Harris et al., 2011).