RESULTS AND DISCUSSION

Clay mineralogy: The clay fraction mineralogy of the concerned sandstones shows that the clay minerals are mainly represent by kaolinite as sole mineral or by mixture of kaolinite and rare illite. Kaolinite is identified by its basal reflections at d-spacing: 7.14, 3.56 and 2.35Å. These reflections were not affected by glycolation while removed out upon heating at 550°C for 2 h. Their sharpness and intensities reflect moderate crystallinity. Illite is identified by the most intense peaks at 10.04 and 4.95 Å which were neither affected by heating nor glycolation. The prevalence of kaolinite with minor or without illite indicates their sedimentary origin under continental weathering conditions in non-marine environment (Lonnie, 1982). The moderate crystallinity gained by most of kaolinites suggests that they are chiefly of detrital inheritance and are relate to fluvial environment (Keller, 1956; Abu-Zeid et al., 1991). This refers to their direct derivation from weathering horizons and/or soils on sedimentary and quartz-bearing plutonic rocks.

Geochemistry
Major elements: Bulk geochemistry of the major elements of 17 pre-Cenomanian samples from Gebel Ghazalani area are given in Table 1. Major element distribution patterns reflect the mineralogy of the studied samples. The sandstone samples have lower Al₂O₃ contents than the associated fine-grained mud enriched samples (samples 1, 4, 11 and 15). Most of these associated “mudstones” are enriched in Al₂O₃, K₂O, Fe₂O₃ and TiO₂ in comparison with the sandstones, reflecting their higher content in clay. In general, the pre-Cenomanian sandstones are rich in SiO₂ contents between 74.55 and 98.3% with an average 91.74% i.e. quartz-rich following the criteria of Crook (1974). They possess low contents of TiO₂ and Fe₂O₃, the alkali and alkaline oxides. Generally, low concentrations of Fe₂O₃ and TiO₂ in all sandstones reflect low abundances of Ti-bearing minerals (biotite, ilmenite, titanite and titaniferous magnetite) in the analyzed samples (Armstrong-Altrin et al., 2004). MnO and P₂O₅ are strongly depleted and there are no significant differences in the contents of MnO and P₂O₅ among the samples.

By comparison with the average Upper Continental Crust (UCC) and post Archean Australian Average Shale (PAAS), representative continentally derived sediments, values of Taylor and McLennan (1985), the pre-Cenomanian sandstones are enriched in SiO₂ but, are depleted in all other major elements (Fig. 3). The observed depletion in Na₂O, MnO and CaO is not only likely due to quartz dilution but also indicates that the studied sediments have been suffered from intense weathering and recycling (Joo et al., 2005; Jin et al., 2006). Generally, Ca, Na and K contents are controlled by feldspars and thus strong depletion in CaO, Na₂O further suggests destruction of plagioclase due to chemical weathering in the source or during transport. SiO₂ content has a strong negative correlation with Al₂O₃ content (r = -1; Fig. 4g) reflecting that much of SiO₂ is present as quartz grains. Low values of Al₂O₃/SiO₂ (range = 0.01-0.22; average = 0.06; Table 1) confirm the quartz enrichment in the studied sandstones. With the exception of SiO₂, the other oxides broadly follow the trend of Al₂O₃ (Fig. 4), indicating that they are associated with micaceous and/or clay minerals in the sediments. Plotting of Al₂O₃ (Fig. 4) vs. major oxides variation diagrams, Fe₂O₃, MnO, MgO, CaO, Na₂O and K₂O show positive correlation.

Table 1:	Major element concentrations (wt.%) of pre-Cenomanian sandstones in G. Ghazalani
CIA: Chemical index of alteration, PIA: Plagioclase index of alteration, CIW: Chemical index of weathering, ** Molar ratios. UCC: Average upper continental crust (Taylor and McLennan, 1985), PAAS: Post-Archean average Australian Shale (Taylor and McLennan, 1985)

This suggests that most oxides including CaO are resulting chiefly from aluminosilicates (Jin et al., 2006). Using the geochemical classification diagram of Herron (1988), G. Ghazalani sandstones are represented chiefly by quartz arenites with few litharenite and sublitharenite and rare wackes, indicating relatively mature sediments.

Fig. 3:	Spider plot of average major element contents for G. Ghazalani sandstones normalized against PAAS and UCC, data are after Taylor and McLennan (1985)

This interpretation agrees with petrographic data from correlative (pre-Cenomanian) strata in Central Sinai (Salem et al., 2001) that were regarded as mineralogically mature sandstones.

Trace elements: The trace element contents of 10 selected sandstone samples from G. Ghazalani are given in Table 2. The normalized trace element averages of the studied sediments with respect to the average UCC and PAAS (Taylor and McLennan, 1985) are shown in the spider diagram Fig. 5; in this diagram, the normalized trace elements are arranged in order of decreasing compatibility in typical igneous differentiation series. Most compatible elements are shown to the left followed by incompatible elements towards the right along the X-axis (Ghosh and Sarkar, 2010).

The main feature of the trace element distributions is the generally positive correlation with Al₂O₃, reflecting association of most elements with the clay fraction. The studied sandstones have low concentrations of the transition metals Co, Ni and V.

Table 2:	Trace-element concentrations (ppm) for sandstones of G. Ghazalani
PAAS: Post-archean average Australian shale (Taylor and Mclennan, 1985) and UCC: Average upper continental crust (Talyor and Mclennan, 1985)

Fig. 4(a-g):	Co-variation of Al2O3 vs. major elements for G. Ghazalani sandstones. Notice the positive correlation of Al2O3 with the major elements, SiO2 shows negative correlation

These elements are strongly depleted in respect to UCC and PAAS. Cu is slightly enriched relative to UCC and depleted to PAAS. Depletion of these elements may indicate the negligible role of the basic rocks as a source rocks. These elements are mainly concentrated in the clays or metal oxides (Turekian and Michael, 1960). Vanadium is positively correlated with Fe₂O₃ (r = 0.49) and TiO₂ (r = 0.97).

Fig. 5:	PAAS, UCC-normalized averages trace element concentrations for G. Ghazalani sandstones. PAAS and UCC data are after Taylor and McLennan (1985)

It is generally adsorbed on kaolinite and possibly associated with iron oxide minerals (Hirst, 1962). Positive correlations of Co, Ni, Zn, Cu and V with both Fe₂O₃ (r = 0.29, 0.53, 0.59, 0.42 and 0.49, respectively, Fig. 6) and Al₂O₃ (r = 0.63, 0.63, 0.77, 0.91 and 0.72, respectively, Fig. 7 ) also indicate that these elements are linked with iron oxides and clay minerals.

In general, the studied sandstones have significantly Low Large-Ion Lithophile Elements (LILE), Ba, Cs, Rb, Pb and Sr concentrations. These elements are highly depleted comparable to those of Upper Continental Crust (UCC) and PAAS (Fig. 5). Depletions of LILE may reflect intense weathering and recycling. Low Sr content is generally related to low CaO content, probably due to the lack of calcic plagioclase.

Rare Earth Elements (REE): The results of REE analyses are listed in Table 3 and are shown as chondrite-normalized patterns in Fig. 8. Total REE (ΣREE) abundances are variable throughout the succession (Table 3). Pre-Cenomanian sandstones have a wide range of total REE (ΣREE) contents of 17.4-419.47 ppm with an average 133.43 ppm (Table 3) which is relatively close to that of the UCC (Taylor and McLennan, 1985).

Fig. 6(a-e):	Plots of Fe2O3 against (a) Ni, (b) Co, (c) Zn, (d) Cu and (e) V for G. Ghazalani sandstones

Fig. 7(a-e):	Plots of Al2O3 against (a) Co, (b) Ni, (c) Zn, (d) Cu and (e) V for G. Ghazalani sandstones

Table 3:	Rare earth element concentrations (ppm) of pre-Cenomanian sandstones in G. Ghazalani
Eu/Eu* = EuN/(SmN xGdN)0.5 and Ce/Ce* = CeN/(LaNxPrN)0.5, where subscript n indicates normalization to chondrite. PAAS: Post-archean average Australian Shale (Taylor and Mclennan, 1985) and UCC: Average upper continental crust (Talyor and Mclennan, 1985)

The associated mud rich beds have greater REE abundances (ΣREE contents of 142-419 ppm with an average about 261 ppm) than sandstones, suggesting that clays control the REE distributions (Hossain et al., 2010). The good positive correlation between Al₂O₃ and the REE contents confirms that the REE are mostly concentrated in the fine fractions (clay). Generally, the ΣREE contents are less than 200 ppm that may reflect quartz dilution.

Fig. 8:	Chondrite-normalized averages REE patterns of G. Ghazalani sandstones. The PAAS and UCC patterns are given as a reference; data are after Taylor and McLennan (1985)

Values are lower in the sandstones having higher SiO₂ contents. In general, chondrite-normalized REE patterns for the studied sandstone from G. Ghazalani (Fig. 8) are characterized by high light REE/heavy REE (LREE/HREE) ratio (5.28-10.29, Table 3), flat HREE pattern and pronounced negative Eu anomaly. LREEs are fractionated, (La/Sm)_n = 3.00 and the HREE patterns are almost flat, (Gd/Yb)_n = 1.54. This chondrite-normalized pattern is typical of sediments and sedimentary rocks enrichment in light REE (LREE), flat heavy REE (HREE) and negative Eu anomaly (Borges et al., 2008). The chondrite-normalized REE patterns for Gazalani sandstones (Fig. 8) are similar to that displayed by upper continental crust and PAAS (Taylor and McLennan, 1985) but the LREEs are slightly similar relative to UCC and slightly depleted relative to PAAS. (HREE) are nearly typical to PAAS and enriched relative to UCC, this may be due to higher concentration of heavy minerals especially zircon which has a high concentration of HREE (Taylor and McLennan, 1985; Lee, 2002). The europium anomaly of Gazalani sandstones is always negative, ranging from 0.26 to 0.76 (average 0.64) which is quite typical to that of UCC (0.65) and PAAS (0.66).

TECTONIC SETTING

Bhatia (1983) and Roser and Korsch (1986) used the chemical composition of sandstones to investigate provenance and tectonic setting of sedimentary rocks. Roser and Korsch (1986) introduced a discrimination diagram utilizing log (K₂O/Na₂O) versus SiO₂ to determine the tectonic setting of terrigenous sedimentary rocks.

Fig. 9:	K2O/Na2O-SiO2 relations for G. Ghazalani sandstones, data recalculated to 100% volatile free, tectonic setting fields are after Roser and Korsch (1986)

Fig. 10:	Discrimination function plot of G. Ghazalani sandstones, data recalculated to 100% volatile free, tectonic setting fields are after Bhatia (1983)

Both SiO₂ and log (K₂O/Na₂O) values increase from volcanic arc to active-continental-margin to passive margin settings that signify continentally derived sediments distinct by high SiO₂ and low Na₂O. The plot K₂O/Na₂O versus SiO₂ for the study samples (Fig. 9) implies that they are typical of sediments deposited in passive margin setting. According to Roser and Korsch (1986), passive margin sediments are largely quartz-rich sediments derived from plate interiors or stable continental areas and deposited in intracratonic basins or on passive continental margins. Discrimination of tectonic settings based on major elements also was proposed by Bhatia (1983); it includes oceanic island arc, continental island arc, active continental margin and passive margin. Plotting the studied sandstones on this discriminant function, all samples fall deeply in the area of passive setting (Fig. 10).

Table 4:	Range of elemental ratios of Gebel Ghazalani sediments compared to the ratios derived from silicic and basic sources
1Cullers (1994), Cullers (2000), Cullers and podkovyrov (2000), Cullers et al. (1988). 2Bhatia and Crook (1986)

According to Asiedu et al. (2005), clastic rocks deposited in cratonic basins and passive margins, as in the case of the present study, are basically mature sediments, mainly quartz arenites and shales indicating deep weathering, low relief in source areas and prolonged transport across quiet continental surfaces.

According to the Th-Co-Zr/10 plot (Bhatia and Crook, 1986), the Pre- Cenomanian sediments lie in or close to the passive margin field; plot of samples close to Zr/10 apex reflects the enrichment of Zr.

Sediments of passive margin tectonic setting are characterized by low Ti/Zr ratio (generally less than 10), compared to the other tectonic settings (Bhatia and Crook, 1986). Ghazalani sandstones have Ti/Zr ratio (average ~7; Table 2) which is close to the Ti/Zr ratio (average = 6.74) estimated by Bhatia and Crook (1986) for the sediments deposited in passive margin setting (see latter Table 4). Negative Sr anomaly as shown by the studied sandstone (Fig. 5) is an indication of old recycled environments passive continental margin settings (Mader and Neubauer, 2004; Ghosh and Sarkar, 2010).

G. Ghazalani sediments deposited in the passive margin setting are resemble to those of the Atlantic type sediments and can be categorized by higher Zr contents, higher Zr/Nb, Zr/Th ratios and lower Rb, V and Ti/Zr ratio. These features mirror the recycling of these sediments, the depletion of feldspar and labile fragments and the increased abundance of heavy minerals, mainly zircon, through sedimentary processes.

PROVENANCE

Major elements, trace elements and the REE were used to infer the provenance of sandstones. Several trace and REE were expected to be more useful in discriminating source rock compositions and tectonic setting due to their relatively low mobility during sedimentary processes and their short residence times in seawater (Taylor and McLennan, 1985). These elements are probably transferred quantitatively into clastic sediments during weathering and transportation, reflecting the signature of the parent materials (Bhatia and Crook, 1986; McLennan, 1989).

Fig. 11:	Discriminant function diagram using major elements, Provenance fields are after Roser and Korsch (1988)

Roser and Korsch (1988) used major element discriminant functions to discriminate four provenances, namely mafic (P1), intermediate (P2), felsic (P3) and quartzose recycled (P4). Plotting the present data in this diagram, all the samples fall deeper within the P4 field (Fig. 11). This supports the interpretation that they were derived from granitic to gneissic or from a sedimentary source area (Roser and Korsch, 1988). Position of the studied samples deep into the P4 field is similar to that observed for several recycled suites elsewhere (Roser et al., 1996).This was interpreted as the effect of recycling, with progressive loss of feldspar and relative increase in quartz. The deeper plot position of G. Ghazalani samples may therefore reflect more advanced weathering and maturation in these fluvial sediments. This observation clearly indicates the less possibility of the mafic rocks as source rocks for the studied samples of the pre-Cenomanian (Nagarajani et al., 2007).

Transition metal elements (Co, Cu, Ni and V) are well-suited in magmatic processes and therefore they are highly concentrated in mafic and ultramafic rocks than felsic igneous source rocks and their weathered products. In addition, these elements are relatively immobile through weathering. They are supposed to be transported completely in the terrigenous component of sediment and thus mirror the chemistry of their source rocks (McLennan et al., 1980; Armstrong-Altrin et al., 2004). The significant low concentration of Co, Cu, Ni and V in the studied sandstones and their strong depletion relative to PAAS (Fig. 5) can be interpreted as the result of sediment derivation from a slightly more silicic and fractionated source than the PAAS. Also, relative contribution of mafic/felsic (granitoids) sources can be reflected by the Ba/Co ratio, as these elements are largely contained in K-bearing and ferromagnesian minerals, respectively (Taylor and McLennan, 1985; Cullers et al., 1988). The pre-Cenomanian sediments have elevated Ba contents but are depleted in Co. The higher Ba/Co ratio (Table 2) is an indication of sediments derived from weathered felsic-granitic sources (Cullers et al., 1988; Akarish and El-Gohary, 2008).

The High Field Strength Elements (HFSE) Zr, Nb, Hf, Y, Th and U are prefer initially partitioned into melts through crystallization (Feng and Kerrich, 1990) and therefore, these elements are enriched in felsic rather than mafic rocks. Along with the REE, HFSE are thought to specify compositions of provenance. In sandstones from G. Ghazalani, Nb and U have average concentration values similar to PAAS whereas Zr, Y and Hf, are slightly to highly enriched (Fig. 5). Compared to PAAS, enrichment of HFSE in G. Gazalani sandstones indicate that source rocks were more granitic in composition than those supplied detritus to PAAS (Lee, 2002).

The La/Th vs. Th/Yb and Th/Co vs. Zr/Co plots have been used to differentiate between felsic and mafic nature of source rocks (McLennan et al., 1980; Bhatia and Crook, 1986; Borges et al., 2008). In these plots (Fig. 12a, b), the studied samples show felsic character of source rocks (high Th/Co and high La/Th).

REE and Th are fairly helpful to deduce crustal compositions because their distribution is not significantly affected by secondary processes such as diagenesis and metamorphism and is less affected by heavy mineral fractionation than that for elements such as Zr, Hf and Sn (Bhatia and Crook, 1986; Armstrong-Altrin et al., 2004). REE and Th contents are higher in felsic than mafic igneous source rocks and in their weathered products, In addition, Eu/Eu*, (La/Lu)_n, Th/Co and La/Co ratios are significantly unlike in mafic and felsic source rocks and may let constraints on the provenance of sedimentary rocks (Wronkiewicz and Condie, 1989; Cullers et al., 1988; Cullers, 1994; Cox et al., 1995; Armstrong-Altrin et al., 2004). The Eu/Eu*, (La/Lu)_n, Th/Co and La/Co ratios (Table 4) of the studied pre-Cenomanian of G. Gazalani are compared with those in sediments derived from felsic and mafic source rocks (Cullers et al., 1988, 2000; Cullers and Podkovyrov, 2000) (Table 4). Such comparison indicates that the trace elemental ratios of this study are comparable to the range of sediments derived from felsic source rocks rather than mafic source rocks.

The relative REE patterns and Eu anomaly size have also been utilized to deduce sources of sedimentary rocks (Taylor and McLennan, 1985; Wronkiewicz and Condie, 1989). Mafic rocks contain low LREE/HREE ratios and tend not to contain Eu anomalies, whereas more felsic rocks usually contain higher LREE/HREE ratios and negative Eu anomalies (Cullers and Graf, 1984).

Fig. 12 (a-b):	(a) La/Th vs. Th/Yb plot show felsic vs. mafic character (McLennan et al., 1980) and (b) Th/Co-Zr/Co plot show mafic-felsic character and recycling (McLennan et al., 1993). The dashed line indicates compositional difference, and deviation from this trend (solid line) represents sedimentary effects from sorting and recycling

The depletion of Eu may be interpreted as shallow, intracrustal differentiation which resulted in Eu-depletion in the upper continental crust, associated with the production of granitic rocks (McLennan, 1989). In the present study, all samples show higher LREE/HREE ratio (5.28-10.29; average 7.26; Table 3) and a significant negative Eu anomaly (0.64 Table 3; Fig. 8). Prominent negative Eu anomaly, high LREE/HREE and flat HREE pattern reflect sediment derivation from predominantly felsic rocks of the upper continental crust.

From the above mentioned trends the provenance of the studied sandstones can be explained by a sedimentary system dominated by detritus derived from recycled highly weathered silicic (granatic) materials. The trace element compositions and ratios are consistent with most sandstone being derived from a felsic (granitic) source.

SOURCE AREA WEATHERING

The intensity and duration of weathering in siliciclastic sediments can be designated through examining the relationships among alkali and alkaline earth elements (Nesbitt and Young, 1982). Since the upper crust is dominated by the presence of feldspars and volcanic glass (Nesbitt and Young, 1982, 1984), the dominant process during chemical weathering and soil formation is the degradation of labile feldspars from source rocks to secondary clay minerals. These chemical signature are ultimately transferred to sedimentary records and supply a useful means for monitoring the original composition and following weathering conditions.

The weathering history of the sources of siliciclastic sediments can be deduce through Quantitative estimation of the chemical weathering of silicates such as the calculated values of Chemical Index of Alteration (CIA), Plagioclase Index of Alteration (PIA) and Chemical Index of Weathering (CIW) (Nesbitt and Young, 1982; Fedo et al., 1995). Where, CIA = Al₂O₃/ (Al₂O₃+CaO*+Na₂O+K₂O) x 100, PIA = (Al₂O₃-K₂O)/ (Al₂O₃+CaO*+Na₂O-K₂O) x100 and, CIW = Al₂O₃/ (Al₂O₃+CaO*+Na₂O) x100, in molecular proportions. CaO* is defined as CaO in silicates fraction only. However, in the present study there was no objective way to distinguish carbonate CaO from silicate CaO, so the detected CaO is used here. This is justified on the basis that there none of the samples is calcareous and CaO is very low (0.05-0.85, average = 0.16) (Nagarajani et al., 2007). Moreover, as mentioned before the detected CaO is positively correlated with Al₂O₃ implying that CaO is related only to silicates.

CIA, PIA and CIW values for G. Ghazalani sandstones show wide range with highly variable amount comparable with moderate to high chemical weathering (Fig. 13). The CIA values are in good agreement with those of PIA and CIW and show a good linear relationship (Fig. 13). The variations in CIA, PIA and CIW values may be due to the different concentrations of alumina in samples rather than variable degrees of source area weathering. Except for sample three, all the samples have CIA, PIA and CIW values more than 68 (Table 1). This reflects the moderate and high weathering conditions either in the original source terrane or during transportation before deposition. Moreover, high CIA values, suggest derivation from a stable cratonic source (Hossain et al., 2010).

Also, PIA values indicate that the plagioclases in the possible parent rock displayed high weathering condition and resulted in low CaO content, especially with increasing PIA values. This implies that with increasing chemical weathering the sediments are steadily depleted in plagioclases and enriched in secondary aluminous clay minerals (Roy et al., 2008).

Some elemental ratios, Al/Na, Al/K, Ti/Na, K/Na and Rb/K were used also as markers for intensities of chemical weathering (Roy et al., 2008). These ratios progressively increase with increasing chemical weathering. Such case may be referred to the solubility and ion potential where labile cations (Na, K, Ca, Rb) are leached in favorite to insoluble hydrolysates (Ti, Al) (Nesbitt and Young, 1982; Roy et al., 2008). However, it is indicated that both K and Rb are integrated into clay minerals by adsorption and cation exchange during initial weathering of fresh rocks. But, with increasing weathering, K is preferentially leached compared to Rb (Wronkiewicz and Condie, 1989). Following Roy et al. (2008), plots of the present CIA values against Al/Na, Al/K, Ti/Na, K/Na and Rb/K elemental ratios (molar proportion; Fig. 13) support the intermediate to extreme chemical weathering.

The CIA data are presented graphically on the Al₂O₃-(CaO*+Na₂O)-K₂O (A-CN-K) ternary diagram (Nesbitt and Young, 1984). This can helps to understand the trends of chemical weathering, infer bulk source composition and the extent of post-depositional K-metasomatism, if present. Additionally, the role of steady-state or non-steady state weathering can be evaluated (Nesbitt and Young, 1984; Fedo et al., 1995; Roy et al., 2008; Hossain et al., 2010). In the A-CN-K diagram (Fig. 14), the studied sandstones commonly form a weathering trend that is almost parallel to the A-CN line. All the studied samples plot away from the plagioclase K-feldspar line in a direction towards the A apex (i.e., near to the highly weathered minerals). The increases in CIA values possibly reflect increasing proportions of clay minerals with respect to feldspar.

Fig. 13:	Scatter plots of chemical index of alteration (CIA) vs. IA, CIW, Al/Na, Al/K, K/Na, Ti/Na and Rb/K showing the degrees of chemical weathering in G. Ghazalani sediments

Fig. 14:	The Al2O3-(CaO+Na2O)-K2O (A-CN-K) ternary diagram in molecular proportions for G. Ghazalani sandstones (after Nesbitt and Young, 1984). Also plotted is the average upper continental crust (Taylor and McLennan, (1985)), as well as some rock forming minerals important in silicate rock weathering

This will lead to increase proportion of Al₂O₃-rich clay minerals (e.g., kaolinite). In turn, it reflects that these samples experienced steady state weathering, supporting a stable and intensely weathered source interpretation (Nesbitt et al., 1997). This can be interpreted as progressive removal of labile cations as is commonly done for weathering profiles; also, indicates that more mature material is being supplied to the basin of deposition.

Low content of K₂O in G. Gazalani sandstones may be ascribed to intense chemical weathering (Cox et al., 1995) which enhanced leaching of soils and removal of alkalis in solution. Very low K₂O/Al₂O₃ ratios (0.01-0.11, average: 0.04), caused by the presence of high proportion of kaolinite, possibly indicate moderate to intense chemical weathering of source rocks (Potter et al., 1980). This is confirmed by the mineralogical analysis that revealed presence of kaolinite as sole clay mineral or kaolinite with minor illite. This compositionally mature sandstones possibly support deposition in a tectonically quiescent or cratonic setting where recycling and associated weathering were active during such sustained periods of tectonic quiescence (Taylor and McLennan, 1985). However, intense chemical weathering of first-cycle detritus can also produce sandstones as mature as those of the pre-Cenomanian sandstones (Barshad, 1966; Ghosh and Sarkar, 2010).

Lacks of epeirogenic movements during Early Cretaceous time corroborated such tectonic quiescence conditions and reflect that relief and rate of mechanical erosion were insignificant. Moreover, tectonic quiescence coincided in time with the warm and humid climate conditions that prevailed in the Egyptian territory during Paleozoic to pre-Cenomanian and mostly in Early Cretaceous times (Said et al., 1976). Such conditions let to increase the intensity of chemical weathering and the decomposition of the source area parent rocks.

High ratio of Th/U reflects intense weathering condition, ratio increases with kaolinite content, with the highest value in the more deeply weathered rocks (McLennan et al., 1990). However, the Th/U ratios can also change due to differences in oxidation state and therefore do not necessarily reflect source-area weathering conditions (Bauluz et al., 2000; Joo et al., 2005). The presence of correlation between Th/U ratios and CIA values (r = 0.41) suggests that the Th/U ratios are probably a reflection of chemical weathering rather than a result from changes in redox conditions during diagenesis.

Variation in REE contents in sediments may be related either to weathering or to variations in grain-size fractions and mineral contents due to hydraulic sorting (Cullers et al., 1975; McLennan, 1989). During weathering the REE are relatively immobile, so only minor enrichment or loss is expected. However, the LREE and HREE show different types of behavior and may become fractionated (Condie et al., 1995). Another possibility for the variations in ΣREE content among the studied sandstones can be related to an influence of a quartz dilution effect on abundance of heavy minerals and/or clay (Taylor and Mclennan, 1985; Armstrong-Altrin et al., 2004). Thus, we interpret the differences in ΣREE content of the pre-Cenomanian sandstones may be due to influence of a quartz dilution. This interpretation is supported by the significant good correlation between the LREEs, HREEs and ΣREE and Al₂O₃ (r = 72, 75 and 73, respectively). The average ΣREE contents in clay rich samples (contain higher Al₂O₃ concentration) is about 2.5 times higher than those contain less alumina; this strongly suggests that the REE are hosted mainly in the sample rich in clay content rather than those rich in quartz as reported by Taylor and McLennan (1985). On the other hand, clays are not the sole responsible for the variation in REE contents. Zircon (ZrSiO₄), as a heavy mineral, is enriched in HREE and Hf (Taylor and McLennan, 1985; Borges et al., 2008; Mahjoor et al., 2009). Presence of very good correlation between Zr and Hf (r = 1) and a good correlation between Zr and Yb as representative of HREE (r = 0.88) suggest that zircon controls the values of HREE and Hf in the studied pre-Cenomanian sandstones.

Ce fractionation can take place at some stage in weathering and sedimentary processes. In the early stages, negative Ce-anomalies are noticed in weathering products like secondary hydrous phosphates (Braun et al., 1998) and positive Ce-anomalies emerge in deeply weathered and lateritic profiles where soluble Ce³⁺ oxidizes to insoluble, stable Ce⁴⁺ and hosted in secondary cerianite, Ce(IV)O₂ (Braun et al., 1998; Borges et al., 2008). The relatively high positive Ce anomaly of G. Ghazalani sandstone samples (Ce/Ce* range from 1.03-1.25 average 1.07; Table 3) support that the source area had experienced intensive chemical weathering. This interpretation is consistent with the occurrence of kaolin deposits located at the base of many sandstone successions belong to the pre-Cenomanian (example, Early Cretaceous) in different localities in Sinai. This kaolin deposits were formed as a result of extensive chemical weathering (laterizing conditions) that led to the development of kaolinitic profiles on the parent source rocks, then, later on, were carried out to the basins of deposition (Yanni, 1987).

The recycling influence can also be recognized on the Th/Co vs. Zr/Co plot (Fig. 12b). Zr is the physically and chemically ultra-stable mineral zircon that can indicate the effect of recycling (McLennan et al., 1993; Borges et al., 2008). In that plot (Fig. 12b) most of studied samples demonstrate a linear compositional trend, with some spread toward recycling.

The mobile elements Na₂O, K₂O, MgO and P₂O₅ were depleted in respect to the UCC which is reliable with the loss through sedimentary processes. Depletion in Al₂O₃ and enrichment in SiO₂ relative to the UCC may be due to sedimentary sorting and loss of small grain-sized clays (Al₂O₃-rich and SiO₂-poor) and retention of sand-size composition (Al₂O₃-poor and SiO₂-rich) (Borges et al., 2008).

The SiO₂/Al₂O₃ is a commonly employed index of sedimentary maturation. Values increase because of increase of quartz at the expense of less resistant components such as feldspar and lithic fragments during sediment transport and recycling. The SiO₂/Al₂O₃ ratio is about 3 in basic rocks (basalts and Gabbros) and it is around 5 In the acidic end member (granites and rahyolites) (Le Maitre, 1976; Roser et al., 1996). Ratio more than 5 or 6 in sedimentary rocks provided evidence of sedimentary maturation (Roser et al., 1996). The average value of SiO₂/Al₂O₃ ratio for the studied sandstones is about 48 which is significantly high, indicating highly mature sediments. Moreover, if we omit the obvious higher values (samples 3, 8 and 12, Table 1), the

SiO₂/Al₂O₃ = 33.80 which is strongly comparable to the value of 30 reported in the modern sediment (Valloni and Maynard, 1981; McLennan et al., 1990) that interpreted as highly mature sediments. The dominant of quartz arenites as well as presence of the ultra-stable clay minerals kaolinite and illite, identified within the clay fractions of these sandstones, is an additional criterion for their high maturity.

CONCLUSIONS

The pre-Cenomanian sandstone sequence (~ 214 m thick) exposed at Gebel Ghazalani area, East Sinai starts at the base by Lower Paleozoic pre-Carboniferous sandstones that unconformably overlies the Katherina Volcanics and followed unconformably upwards by the Lower Cretaceous sediments of the Malha Formation which underlies the Cenomanian shales and marls. Geochemically, the Pre-Cenomanian sandstones are mainly classified as quartz arenite with few sublitharenite and rare litharenite and wacke. Major, trace elements and REE ratios and diagrams suggest that sediments were deposited in a passive margin and the sediments were derived from felsic (granitic) source rocks. The source area may have contained quartzose sedimentary rocks. The G. Ghazalani sediments deposited in the passive margin setting resemble to those of the Atlantic type sediments and can be categorized by higher Zr contents, higher Zr/Nb, Zr/Th ratios and lower Rb, V and Ti/Zr ratio. These features mirror the recycling of these sediments, the depletion of feldspar and labile fragments and the increased abundance of heavy minerals, mainly zircon, through sedimentary processes. The values of the weathering indices (CIA, PIA and CIW) along with the elemental molar ratios suggest moderate to extreme weathering conditions in the source area and/or during transportation. Most of the studied sandstones are inferred as highly mature sediments evidenced from their high SiO₂/Al₂O₃ ratio (47.94) and the presence of the ultra-stable clay minerals, Kaolinite and illite. Chondrite-normalized REE patterns for most of the sandstone samples display high LREE/HREE ratio, flat HREE pattern and pronounced negative Eu anomaly and in general correlate well with the UCC and PAAS average composition.