Geochemical Approach to Determine the Possible Precipitation Parameters of the Coniacian–Santonian Mazıdağı Phosphates, Mardin, Turkey

Abstract

The Tethyan phosphates were formed during the Upper Cretaceous and Eocene interval as a result of the collision of the African–Arabian and Eurasian plates and the closing of the Neo-Tethys Ocean. This study aimed to reveal the possible precipitation parameters of these phosphates by examining the main oxide, trace element, and rare earth element contents of the phosphates in the study region. The mean major oxide concentrations of the phosphates were found to be 51.6 wt.% CaO, 21.2 wt.% P₂O₅, 8.03 wt.% SiO₂, 18.1 wt.% CO₂, 0.51 wt.% K₂O, 0.12 wt.% Fe₂O₃, 0.05 wt.% Al₂O₃, 0.18 wt.% MgO, and 0.02 wt.% MnO. The average trace element concentrations were 79 ppm Ba, 1087 ppm Sr, 0.23 ppm Rb, 14.7 ppm Ni, 108 ppm Cr, 262 ppm Zn, 27 ppm Cd, 21.6 ppm Y, 58 ppm V, 6.43 ppm As, 30.3 ppm Cu, 1.36 ppm Pb, 6.32 ppm Zr, 39 ppm U, 0.21 ppm Th, and 1.33 ppm Co. The average trace element contents were 1742 ppm, with this indicating an enrichment assemblage of Sr, Cd, As, and Zn in comparison to PAAS (The Post-Archean Australian Shale). The total REE concentrations in the Mazıdağı phosphates varied from 3.30 to 43.1 ppm, with a mean of 22.1 ppm recorded. All phosphates showed heavy REE (HREE) enrichments and had similar REE patterns to PAAS (The Post-Archean Australian Shale). All samples had strongly negative Ce and positive Eu, Pr, and Y anomalies. These anomalies indicate the existence of oxic and suboxic marine conditions during the formation of the phosphates. According to the proposed genetic model, the phosphates mostly formed in the oxic and suboxic zones of the Tethys Ocean and were precipitated on slopes that depended on strong upwelling from an organic-rich basin in anoxic/suboxic conditions from deeper seawater. The Pb isotope data obtained also indicate the existence of a deep-sea hydrothermal contribution to this phosphate formation.

1. Introduction

Phosphate rock is a marine biochemical rock rich in phosphorus (P), with more than 18 wt.% P₂O₅ [1,2]. It is one of the most important raw materials, with it being applied, for instance, in agriculture for fertilizer production and in high-tech industries (e.g., REEs) due to the trace element concentrations extracted from it as a by-product [3,4]. Mazıdağı phosphates are part of the “Late Cretaceous–Eocene Giant Phosphate Belt” (Figure 1), which stretches across Jordan, Syria, and Turkey, and through North Africa from the Caribbean Sea in the west [5,6]. Phosphates are distributed widely in modern and ancient marine sediments. Today, phosphates are noted to form at continental margins and continental shelves in upwelling regions such as the Gulf of California, Namibia, Chile, and Peru [7]. Phosphates in this time period have been well studied in many palaeogeographic areas, including Australia [8], China [9,10,11], the West African craton [12,13,14], India [15,16], the Siberian Platform [17], and Mongolia [18]. Economic and subeconomic phosphate accumulations observed in the western margins of Gondwana and Baltica have been less studied [19]. Phosphates generally exhibit similar behavior to REE elements during precipitation. The REE composition of sedimentary phosphates is highly variable among different deposits. REE patterns can range from seawater patterns to shale and bell-shaped patterns [3]. The REE abundances and Ce and Eu anomalies in phosphates in different regions have been investigated by various researchers [4,20,21,22,23,24,25,26,27,28,29]. Phosphates generally have high REE contents and show negative Ce anomalies because of the redox conditions and composition of the depositional environment [13,30,31]. The Coniacian–Santonian Mazıdağı Sedimentary Basin is a region that can be studied in order to obtain the redox conditions and composition of the depositional environment. A number of studies have been conducted on the formation conditions and precipitation of phosphate in this basin [6,32,33,34,35,36,37,38]. The aim of this study was to investigate the major oxide, trace, and rare earth element contents of the studied phosphates and to reveal the physicochemical characteristics of the paleo environment where phosphates are formed.

2. Geological Setting

The studied phosphates are situated on the Arabian Platform, which is one of Turkey’s main tectonic units (Pontides, Anatolides, Taurides, Arabian Platform, and Border Folds) [39]. At the base of the region, the Precambrian Derik volcanics (Telbesmi Formation) crop out, which are composed of basaltic and andesitic pyroclastics, agglomerate, sandstones, and shale [32,33,34,35,36,38,39,40,41]. The Telbesmi Formation is represented around Derik by volcanogenic rocks in which red and grayish-burgundy colors are dominant in interlayers parallel to each other and inclined to the north. Volcanic rocks in the region consist of basalt and andesite lava products. Volcanogenic sandstones can be observed in the formation as interlayers with a thickness of 0.5–1 m. At some levels, andesitic rocks exhibit an agglomeratic/brecciated appearance and contain angular components with a diameter of 0.5–10 mm. Basalts are distinguished from red-colored andesite by their black-colored, fine-grained, and basically porous appearance. It has been noted that this formation was deposited in a shallow marine-coastal environment [39,42]. The predominance of volcanic lithology in the region indicates that volcanism is widespread, and the volcanogenic levels contain pyroclastic lithology concurrent with deposition, as well as lava-type products. The presence of clastic levels indicates that clastic sedimentation accompanies this volcanism. The clastic and carbonates with Cambrian, Ordovician, and Silurian ages in this region uncomfortably cover the Telbesmi Formation. Cretaceous carbonates with phosphate uncomfortably overlie both Precambrian series and Early Paleozoic passive-margin rocks in the study area (Figure 2). Cretaceous carbonates consist of three formations in the study area (Figure 2), the Sabunsuyu, Derdere, and Karababa formations from bottom to top [43]. The Sabunsuyu Formation mainly consists of coarse and thick-bedded crystalline dolomitic limestone. It is also overlain by the Derdere Formation, which consists of thick-bedded limestone, shale, and cherty limestone with Late Cenomanian macrofossils [6,37]. The Derdere Formation passes gradually to the Karababa Formation, which contains phosphate layers from the Late Coniacian to the Early Campanian [6,44]. The Karababa Formation includes a Tasit phosphate layer in its bottom section and has up to a 1.5 m thickness, including carbonates. Therefore, it also has secondary importance as a phosphate ore. The upper part of the Karababa Formation consists of cherty phosphate, limestone, and chert layers that are about 50 m in total thickness. Ghasemian et al. [38] reported that the Karababa Formation consists of limestone, white and gray chert, and reddish and cream phosphate layers.

The Mazıdağı phosphates from bottom to top occur on the gently dipping northern flank of the Derik anticline [6,32]. (1) Tasit phosphate (Turonian), (2) Kasrik (East Kasrik) phosphate, and (3) Şemikan (West Kasrık) phosphate (both Santonian–Coniacian) are all phosphorus-rich members of the Karababa Formation (Figure 3). Furthermore, (4) Akras phosphate (Maastrichtian), a phosphorus-rich member of the Kermav Formation, is located in the eastern part of the study area. The Şemikan (West Kasrik) member is the richest in phosphates, reaching 25%–32% P₂O₅ in content [32].

The Şemikan phosphate area is one of the areas with the highest economic potential among the Mazıdağı phosphates deposits. It consists of four different sectors. The average phosphate thickness in these sectors is around 0.75 cm. The phosphate level in the region is characterized by interlayers, alternating with chert and carbonate veins (Figure 4). While the oolitic phosphate level in this level is sometimes 20–30 cm thick, in some regions the total cherty, calcareous, and oolitic phosphate level can reach up to 4 m in places (Figure 4). Between these levels, white-cream-colored limestone and chert layers are frequently observed. Their persistence varies across the region, and their general slope is 3o-5o, which dips in the N-NE direction.

Light gray-white, high-grade phosphates and red clay phosphates show transitions in the horizontal direction. These transitions are usually gradual from one color to another. The cover layers on the phosphate in four different areas in the Şemikan region are generally composed of white and cream-colored limestone with layers, and their thickness is usually not less than 10 m. Increases in the thickness of the cover layers on the phosphate negatively affect ore production due to increased costs.

3. Samples and Analytical Methods

3.1. Samples

Thirteen oolitic phosphate ore, limestone, and chert samples with phosphate (nine phosphate ore samples, as well as two limestone with phosphate and two chert samples) were collected from different outcrops of the Şemikan phosphate deposits and subjected to several different analyses. In September–October 2020, different units were sampled from bottom to top, as seen in Figure 4. DY-14 and DY-21 were the chert samples, and DY-02 and DY-13 were the white-colored limestone samples. The other samples belonged to different phosphate levels of the Şemikan phosphate area.

3.2. Analytical Methods

First, in our geochemistry laboratory, the thirteen samples were crushed and ground to 200 mesh. Then, in Bureau Veritas Minerals, a 0.25 g split was heated in HNO₃, HClO₄, and HF to fuming and then dried. The residue was then dissolved in HCl. Finally, trace and rare earth elements in these samples were analyzed by ICP-MS, and major oxides were analyzed by ICP-AES at the Bureau Veritas Minerals in Canada (www.bvlabs.com accessed on 5 May 2021). Repeated analyses showed better than 5% reproducibility.

Pb isotopes (²⁰⁸Pb, ²⁰⁷Pb, ²⁰⁶Pb, and ²⁰⁴Pb) for phosphate ores were determined by ICP-MS at Bureau Veritas Minerals (Vancouver, Canada). The samples were dissolved with aqua regia (1 HNO₃ + 3HCl mixture) and analyzed for ²⁰⁶Pb/²⁰⁴Pb, ²⁰⁷Pb/²⁰⁴Pb, and ²⁰⁸Pb/²⁰⁴Pb ratios. The NIST-SRM 981 standard was used, with the following results obtained: ²⁰⁶Pb/²⁰⁴Pb 16.9374; ²⁰⁷Pb/²⁰⁴Pb 15.4916; ²⁰⁸Pb/²⁰⁴Pb 36.7219 (www.bvlabs.com accessed on 5 May 2021).

The geochemical data were statistically subjected to the SNK (Student–Newman–Keul’s Procedure) using variance analysis (ANOVA) and SPSS 15.0 software. To find Ce, Eu, Y, and Pr anomalies in each sample, REE values were first divided by Post-Archean Australian Shale (PAAS) values (Taylor and McLennan, 1985) and then Eu, Ce, Pr, and Y anomalies were calculated with the following formulas: Ce/Ce* = Ce_n/√[La_n × Pr_n], Eu/Eu* = Eu_n/√[Sm_n × Gd_n], Y/Y* = Y_n/√[Dy_n × Ho_n], Pr/Pr* = Pr_n/√[Ce_n × Nd_n].

3.3. Quality Assurance

The Quality Assurance and ISO accreditations of all of the facilities were performed by recognized organizations. These accreditations and registrations satisfy the ISO standards. All Bureau Veritas Minerals are registered to ISO 9001 (www.bvlabs.com accessed on 5 May 2021).

4. Results and Discussion

4.1. Major Oxide Geochemistry

Among the major oxides in the phosphate ore samples, CaO was the most abundant major oxide, with an average of 51.6 ± 2.15 wt.%. The average P₂O₅ concentration in all of the the studied samples was 21.2 ± 1.88 wt.%, with this being the second most abundant major oxide in the studied phosphate samples, with high average contents varying from 2.22 to 31.7 wt.%. The average P₂O₅ content of the phosphate level in the study area was 26.3 ± 3.34 wt.%, except for the limestone (DY-02 and DY-13) and chert samples (DY-14 and DY-21). The P₂O₅ contents of the limestone samples (DY-02 and DY-13) in the Şemikan phosphate deposits were 4.29 and 10.1 wt.%, respectively (

Table 1. Major oxide contents (%) of the phosphate samples.

	SiO2	Al2O3	Fe2O3	MgO	CaO	Na2O	K2O	TiO2	P2O5	MnO	Cr2O3	CO2	Sum	CaO/P2O5
DL	0.01	0.01	0.04	0.01	0.01	0.01	0.01	0.01	0.01	0.01	0.002	−5.1
DY-01	2.05	0.04	0.11	0.22	54.9	0.71	0.01	<0.01	27.5	0.02	0.02	14.2	99.9	2.00
DY-02	0.91	0.01	<0.04	0.21	55.3	0.11	0.01	<0.01	4.29	<0.01	0.01	39.1	100	12.88
DY-03	1.45	0.01	0,07	0.21	55.4	0.43	0.01	<0.01	16.5	<0.01	0.02	25.7	99.9	3.35
DY-11	4.63	0,11	0.25	0.17	53.7	0.71	0.02	0.01	31.3	<0.01	0.03	8.8	99.9	1.72
DY-12	1.05	0.05	0.10	0.20	55.5	0.74	0.01	<0.01	30.2	<0.01	0.03	11.8	99.8	1.84
DY-13	12.80	0.02	0.10	0.17	48.8	0.26	0.01	<0.01	10.1	<0.01	0.01	27.5	99.9	4.81
DY-14	33.50	0.14	0.22	0.13	37.0	0.46	0.02	<0.01	22.3	<0.01	0.01	5.9	99.9	1.66
DY-20	0.53	0.01	0.01	0.17	56.3	0.74	0.01	<0.01	31.7	<0.01	0.02	10.2	99.9	1.77
DY-21	36.1	0.03	0.16	0.16	35.6	0.06	0.01	<0.01	2.22	<0.01	0.02	25.6	100	16.05
DY-22	1.25	0.04	0.06	0.20	55.5	0.73	0.01	<0.01	29.5	<0.01	0.03	12.3	99.9	1.88
DY-23	6.92	0.03	0.09	0.19	52.0	0.33	0.01	<0.01	13.5	<0.01	0.02	26.7	99.9	3.86
DY-24	0.73	0.02	0.01	0.17	56.1	0.69	0.01	<0.01	30.6	<0.01	0.02	11.2	99.9	1.83
DY-25	2.41	0.01	0.08	0.20	55.0	0.60	0.01	<0.01	25.6	<0.01	0.02	15.8	99.9	2.15
Avrg	8.03	0.04	0.11	0.18	51.6	0.51	0.01	0.01	21.2	0.02	0.02	18.1	99.9	4.29

). The P₂O₅ contents of the chert samples (DY-14 and DY-21) were 2.22 and 22.3 wt.%, respectively. The average concentrations of SiO₂ and Fe₂O₃ in the samples were 8.03 ± 0.23 wt.% and 0.11 ± 0.01 wt.%, respectively (Table 1). The average contents of Al₂O₃ and CO₂ were 0.04 ± 0.01 and 18.1 ± 0.66 wt.%, respectively. The averages of the other major oxides (MnO: 0.02 wt.%, K₂O: 0.01wt.%, MgO: 0.18 wt.%, Na₂O: 0.51wt.%, TiO₂: 0.01 wt.% and Cr₂O₃: 0.02 wt.%) were all lower than 1 wt.% (Table 1). The major oxide concentrations of the phosphate ores are listed as follows: CaO > P₂O₅ > CO₂ > SiO₂ > Na₂O > MgO > Fe₂O₃ > Al₂O₃ > K₂O > MnO > TiO₂. The CaO/P₂O₅ ratios for the phosphate ores varied from 1.72 to 3.35, with a mean of 2.27 recorded. The average CaO/P₂O₅ ratio for the limestone and chert samples was 8.85. These phosphate ore ratios show the economic importance of these phosphate raw materials, as when these materials have CaO/P₂O₅ ratios lower than 1, they are considered more suitable and are therefore highly used in the industry, while CaO/P₂O₅ ratios higher than 1 increase sulfuric acid consumption during H₃PO₄ manufacturing, thus reducing their use [45]. The CaO/P₂O₅ ratio in the phosphate rocks was higher than 1.31, and this may have been related to either the occurrence of calcite or dolomite in whole rocks or the substitution of PO₄ by CO₃ [46,47]. P₂O₅ was found to have high positive correlation coefficients with Na₂O (0.99), CaO (0.50), V (0.75), Cr (0.79), Sr (0.99), Y (0.91), Zr (0.79), Nb (0.69), Ba (0.93), Hf (0.69), U (0.95), Pb (0.81), As (0.89), and Zn (0.81). Total REEs (0.92) was found to show weak positive correlations with Cu (0.50), Rb (0.52), and Au (0.89), whereas P₂O₅ was found to show strong negative correlations with SiO₂ (−0.48) and CO₂ (−0.88) and weak negative correlations with Ni (−0.49 (

Table 2. Correlation relationships with major oxides and trace elements of P₂O₅ in phosphate samples (Negative correlations are green color, positive correlations are yellow color).

	SiO2	Al2O3	Fe2O3	MgO	CaO	Na2O	K2O	TiO2	MnO	CO2	V	Cr	Co	Ni	Cu
P2O5	−0.48	0.27	−0.04	0.03	0.50	0.99	0.23	0.06	−0.05	−0.88	0.75	0.79	0.22	−0.49	0.50
	Rb	Sr	Y	Zr	Nb	Ba	Hf	As	Cd	U	Th	Pb	Zn	Au	∑REE
P2O5	0.52	0.99	0.91	0.79	0.69	0.93	0.69	0.89	0.37	0.95	0.40	0.87	0.81	0.49	0.92

; Figure 5).

4.2. Trace Element Geochemistry

The trace element analyses of the phosphate samples determined by the ICP-MS are shown in

Table 3. Trace element concentrations (ppm) of the phosphate samples.

	V	Cr	Co	Ni	Cu	Rb	Sr	Zr	Nb	Ba	Hf	As	Cd	U	Th	Pb	Zn	Au	Sc	∑TE	Th/U	Ni/Co	V/Sc	V/Cr
DY-01	63	123	1.8	7.2	28.0	0.30	1418	6.4	0.11	112	0.11	5.8	13.6	48.6	0.7	1.91	237	0.089	1.7	2062	0.014	4.00	37	0.51
DY-02	12	30	0.5	11.1	4.6	0.10	418	1.4	0.02	20	0.02	1.9	3.2	8.2	0.1	0.39	62	0.085	0.2	574	0.012	22.2	60	0.40
DY-03	46	86	0.6	10.0	16.6	0.20	951	6.5	0.05	77	0.05	5.5	32.9	31.8	0.2	1.16	209	0.065	1.1	1475	0.006	16.7	42	0.53
DY-11	76	189	0.2	23.6	43.9	0.50	1443	15.3	0.14	95	0.14	13.8	23.4	51.6	0.4	2.58	407	0.056	3.5	2396	0.008	118	22	0.40
DY-12	92	155	0.6	15.3	94.4	0.30	1531	9.8	0.08	112	0.08	7.5	46.6	60.2	0.2	1.78	557	0.098	1.7	2684	0.003	25.5	54	0.59
DY-13	24	77	1.2	18.8	17.9	0.10	567	3.6	0.02	38	0.02	2.6	33.9	17.4	0.1	0.65	196	0.047	0.7	998	0.006	15.7	34	0.31
DY-14	121	72	0.7	10.5	29.2	0.50	1064	7.0	0.06	113	0.06	6.7	57.3	57.3	0.2	2.20	206	0.076	1.9	1748	0.003	15.0	64	1.68
DY-20	72	123	0.6	7.8	19.1	0.20	1535	7.4	0.06	98	0.06	9.0	6.3	51.7	0.1	1.56	296	0.066	1.2	2228	0.002	13.0	60	0.58
DY-21	11	76	0.6	32.3	2.7	0.10	198	0.8	0.02	10	0.02	1.3	5.41	4.0	0.1	0.21	49	0.023	0.2	392	0.025	53.8	55	0.14
DY-22	91	158	8.8	10.6	59.7	0.20	1517	7.7	0.08	96	0.08	9.6	68.3	51.0	0.3	1.56	249	0.061	1.6	2327	0.006	1.20	57	0.58
DY-23	44	101	0.8	20.4	48.1	0.20	778	4.6	0.04	56	0.04	4.0	19.2	25.7	0.1	0.77	243	0.037	1.0	1346	0.004	25.5	44	0.44
DY-24	64	114	0.5	8.1	15.4	0.20	1414	7.6	0.05	103	0.05	7.9	25.9	47.5	0.1	1.69	379	0.144	1.5	2190	0.002	16.2	43	0.56
DY-25	33	96	0.4	15.9	13.9	0.10	1294	4.1	0.02	101	0.02	8.0	19.6	46.6	0.1	1.27	313	0.148	1.6	1946	0.002	39.8	21	0.35
Avrg	58	108	1.33	14.7	30.3	0.23	1087	6.32	0.06	79.2	0.06	6.43	27.4	39	0.21	1.36	262	0.08	1.38	1720	0.007	28.2	45.5	0.54

. The total trace element concentrations (∑TE) of the phosphate samples varied from 392 to 2684 ppm. The mean trace elements contents were 58 ppm for V, 108 ppm for Cr, 1.33 ppm for Co, 14.7 ppm for Ni, 30.3 ppm for Cu, 0.23 ppm for Rb, 1087 ppm for Sr, 6.32 ppm for Zr, 0.06 ppm for Nb, 79.2 ppm for Ba, 0.06 ppm for Hf, 6.43 ppm for As, 27 ppm for Cd, 39 ppm for U, 0.21 ppm for Th, 1.36 ppm for Pb, 262 ppm for Zn, and 0.04 ppm for Au (Table 3). The phosphates contained higher Cr, Sr, As, Cd, U, and Zn and lower V, Co, Ni, Rb, Zr, Hf, Nb, Ba, Th, and Pb in comparison to PAAS (Taylor and McLennan, 1985) (Figure 6). While the Sr from the large-ion lithophile elements were highly enriched, Rb and Ba were depleted compared to PAAS. Among the high field strength elements (HFSE; Nb, Hf, U, Th, and Zr), Hf, Zr, Nb and Th were significantly lower compared to PAAS, except for U (Figure 6).

The Th/U ratios in the studied phosphates varied from 0.002 to 0.025. The Ni/Co ratios changed from 1.20 to 118, with a mean of 28.2 recorded. The V/Sc and V/Cr ratios were 0.05 (0.03 to 0.11) and 0.54 (0.31 to 1.68), respectively (

Table 4. Redox classification of the depositional environment. Data from [48,49].

Indicator	Oxic	Suboxic	Anoxic	Euxinic	The Studied Phoshates
H2S		No free H2S in the water column		Free H2S present in the water column
O2 concentration in bottom waters	O2 > 2	0.2 < O2 < 2	O2 < 0.2	O2 = 0
Th/U	>7	2–7	0–2	-	0.88
V/Cr	<2	2–4.25	>4.25	>4.25	0.49
Ni/Co	<5	5–7	>7	-	3.69
V/Sc	-	-	-	>24	45.5

). These Ni/Co and V/Cr ratios indicate the oxic/suboxic association between redox conditions and the precipitation of phosphates, except for the Th/U and V/Sc values (Table 4). Furthermore, Algeo and Liu [48] have suggested that the use of these bimetallic indicators is not very reliable for the determination of past seawater conditions. The solubility of certain elements is influenced by the oxidation state of the water column. Marine sediments may therefore be enriched or depleted in these elements in accordance with the redox state of the water column. Suboxic environments are those with low levels of dissolved oxygen and intermediate H₂S levels. Anoxic environments are characterized by the presence of both H₂S and dissolved oxygen [49]. Redox-sensitive substitutes (Ni/Co, Th/U, V/Cr, V/Sc, and REEs) can show global or regional redox conditions, which can indicate the existence of a paleo-ocean redox environment [50].

As shown in Figure 7, the Nb/Ta and Zr/Hf ratios were similar with North and South Atlantic deep water and Pacific Ocean deep water [51,52,53,54,55]. The HFSE enrichments in phosphate ores result from their absorption in seawater [55,56,57,58]. This also indicates that the low Zr/Hf and Nb/Ta ratios in the phosphates had low HFSE values in seawater during the precipitation of the phosphate ores.

4.3. Rare Earth Element Geochemistry

The rare earth element contents of the phosphate samples are summarized in

Table 5. Rare earth element contents of the phosphate samples (ppm).

Sample No	Dedection Limit	DY-01	DY-02	DY-03	DY-11	DY-12	DY-13	DY-14	DY-20	DY-21	DY-22	DY-23	DY-24	DY-25
La	0.5	10.1	1.30	6.60	15.4	9.10	4.10	5.90	11.7	1.00	10.5	4.40	12.8	7.60
Ce	0.1	3.10	0.40	1.90	5.10	2.80	1.30	2.20	3.60	0.30	3.30	1.40	3.90	2.40
Pr	0.02	1.17	0.13	0.67	1.74	0.98	0.46	0.61	1.29	0.11	1.09	0.49	1.39	0.78
Nd	0.02	4.16	0.51	2.57	7.63	4.01	1.85	2.37	5.31	0.45	4.81	1.92	5.67	3.02
Sm	0.02	0.80	0.07	0.59	1.39	0.74	0.34	0.48	0.84	0.09	0.86	0.42	1.09	0.62
Eu	0.02	0.27	0.03	0.13	0.44	0.22	0.11	0.15	0.30	0.02	0.24	0.10	0.34	0.21
Gd	0.02	1.50	0.22	0.82	2.07	1.37	0.49	0.75	1.60	0.13	1.40	0.58	1.99	0.94
Tb	0.02	0.25	0.03	0.13	0.38	0.21	0.09	0.12	0.25	0.03	0.20	0.09	0.27	0.16
Dy	0.02	1.58	0.17	0.97	2.64	1.47	0.68	0.87	1.75	0.14	1.56	0.69	1.78	1.30
Y	0.02	27.4	4.12	18.7	46.8	27.9	12.0	16.5	29.9	3.00	27.8	13.1	31.6	22.1
Ho	0.02	0.42	0.05	0.29	0.77	0.40	0.17	0.24	0.50	0.04	0.44	0.17	0.53	0.33
Er	0.02	1.33	0.18	0.90	2.45	1.39	0.60	0.88	1.60	0.14	1.33	0.65	1.77	1.11
Tm	0.02	0.20	0.02	0.14	0.37	0.23	0.09	0.12	0.24	0.02	0.25	0.08	0.25	0.16
Lu	0.02	0.24	0.03	0.14	0.42	0.22	0.10	0.15	0.25	0.02	0.24	0.10	0.26	0.19
Yb	0.02	1.24	0.14	0.85	2.20	1.23	0.55	0.74	1.32	0.13	1.19	0.53	1.51	1.02
∑REE	-	53.8	7.42	35.4	89.9	57.3	22.9	32.1	60.5	5.62	55.2	24.7	35.1	41.9
Y/Ho	-	65	82	65	61	70	71	69	60	74	63	77	60	67
(Ce/Ce*)n	-	0.21	0.16	0.18	0.20	0.20	0.18	0.21	0.22	0.39	0.19	0.17	0.22	0.19
(Eu/Eu*)n	-	1.90	2.23	1.22	1.85	1.67	1.72	1.69	2.01	1.25	1.58	1.29	1.81	1.84
(Y/Y*)n	-	3.22	4.20	3.49	3.46	3.49	3.14	3.41	3.22	3.70	3.31	3.37	3.34	3.11
(Pr/Pr)n	-	3.46	2.94	3.22	3.19	3.21	3.22	2.53	3.31	1.25	3.05	3.18	3.30	2.97
(La/Yb)n	-	0.60	0.69	0.57	0.51	0.55	0.55	0.59	0.65	0.57	0.76	0.64	0.63	0.55
((Gd/Yb)n	-	0.73	0.95	0.58	0.57	0.67	0.54	0.61	0.73	0.61	0.71	0.66	0.77	0.68
(Sm/Pr)n	-	1.00	0.06	0.06	0.05	0.06	0.05	0.06	0.06	0.06	0.04	0.05	0.05	0.06
(Sm/Yb)n	-	0.33	0.25	0.35	0.32	0.31	0.31	0.33	0.32	0.35	0.37	0.40	0.37	0.31

. The ΣREE concentrations of the samples ranged from 5.62 ppm to 89.9 ppm. The shale-normalized trend of the studied phosphates relative to PAAS [59] is shown in Figure 8. The REE concentrations of the phosphates were characterized by enriched heavy REEs and relatively depleted light and middle REEs. The mean rare earth element abundances (ppm) in the phosphates were as follows: heavy REEs (2.32) > middle REEs (1.01) > light REEs (0.44) (Figure 8). The La_n/Yb_n ratios showed enrichment patterns between heavy REEs and light REEs. The La_n/Yb_n values of the phosphates ranged between 0.51 and 0.76, with a mean of 0.60 ± 0.01 recorded (Table 5), which was comparable to the corresponding modern seawater values (0.2–0.5; [62]) of the phosphate samples, verifying the HREE enrichments [63]. The low (La/Yb)_n ratios in the phosphate samples indicate that the REE concentrations were associated both with the adsorption of REEs during the evolution of these phosphates and through a substitution mechanism by recrystallization [63]. The Lan/Hon ratios changed from 0.51 to 0.72, with a mean of 0.63 ± 0.02 recorded; this also indicates that the phosphate samples were more enriched in HREEs than in LREEs. Furthermore, the (Sm/Pr)_n and (Sm/Yb)_n values of the phosphates indicate that they were enriched in HREEs (Figure 9). As shown in Figure 9, phosphates from Mazıdağı, Alborz [22], Tebessa [20], Seamount [64], Yangtze [65], and Hazm Al-Jalamaid [66] have been found to be enriched in HREEs and depleted in LREEs and MREEs. Sonrai [27] and Gorgan [25] deposits have been found to be enriched in MREEs and depleted in LREEs and HREEs.

The Ce/Ce* values of the studied phosphate samples changed from 0.16 to 0.22 and had strong negative Ce anomalies (Table 5). The Eu/Eu* and Ce/Ce* ratios were used to determine the precipitation conditions of the depositional environment, such as the temperature, pH, and ƒO₂ [67,68]. Of all the rare earth elements, cerium is the most useful element for investigating the redox potentials of different sedimentary environments. Ce oxidizes readily and is irreversibly and continuously removed from seawater at the surface of oxide minerals; therefore, among all the REEs, it is best suited to being used to understand the genesis of Fe–Mn deposits [69]. The formation of tetravalent Ce is certainly not irreversible, and reduction does occur under anoxia, leading to positive Ce anomalies [70]. Negative Ce anomalies indicate both that there is excess oxygen in seawater and that the rapid precipitation of oxide minerals has occurred [71]. Ce is primarily Ce⁴⁺ under seawater pH and Eh conditions, and its solubility is very low. It precipitates as Ce⁴⁺ (CeO₄) because of its strong depletion in seawater. Ce in oxidation states can occur in both the +3 and +4, although this only occurs in oxic environments [72].

The Eu/Eu* anomalies of the phosphate samples changed from 1.22 to 2.33 (Table 5). In comparison to PAAS, all the phosphate samples had strong positive Eu anomalies. Eu is usually found in a trivalent form, like other REEs. However, Eu⁺³ can be reduced to Eu⁺² in reducing environments [74]. Therefore, according to Bau et al. [75], Eu can be used to investigate sedimentation conditions. Generally, a positive Eu anomaly is interpreted as a proxy for reducing conditions, high temperatures, and a hydrothermal REE supply [20,75]. An Eu anomaly can also be associated with other factors, such as a decrease in temperature or increases in ƒO₂ and pH [76]. While positive Eu anomalies indicate the existence of sub-reduced conditions during the deposition of the sub-layer of phosphates [20], negative Eu anomalies are not indicative of oxic conditions. Under oxic conditions, Eu is strictly trivalent and behaves similarly to other trivalent REEs, leading to Eu/Eu* values of ~1 [20,75,77,78,79]. In this study, the oxic water samples had a REE+Y trend with heavier REE enrichment and a more negative Y anomaly in comparison to the anoxic conditions. The Y concentrations of the phosphates in the study area change from 3.0 ppm to 46.8 ppm, and these have stronger positive Y anomalies (from 3.11 to 4.20) compared to PAAS (Table 5). Yttrium is considered a REE element mainly because of its similar behavior to Ho. According to its ionic radius, it is usually inserted between Ho and Dy in the REE series [68,80]. Y content values can be used to assess post-deposition changes during fluid circulation [81]. These values show positive anomalies in seawater and are not affected by redox conditions, but the magnitude of these anomalies may decrease under diagenetic effects [68,74,81,82]. Kechiched et al. [20] suggested that the southern Algerian phosphates have low Y/Y* ratios and, therefore, that they reflect slight early diagenesis. However, in this study, while the phosphate samples were mostly affected by possible weathering at the top of the series, they were not affected by strong late diagenetic processes. The anoxic hypersaline brine waters have a negative Y anomaly in comparison to overlying oxic seawater in the Tyro sub-basin [83].

4.4. Source of REE Enrichment in Phosphate Samples

Emsbo et al. [3] have reported that REE contents in phosphates are generally homogeneous within a specific geologic time but show variety among different geologic times. Modern phosphate rocks have low REE contents and similar patterns to modern seawater, showing that rare earth elements were fractionated minimally [84]. In this study, the Mazıdağı phosphates (Figure 10) were observed to have lower REE contents than phosphates of similar geologic ages, such as Tunisian, Moroccan, and Algerian phosphates [22]. Furthermore, the REE, Lan/Ybn, Ce/Ce*, Gdn/Ybn, and Eu/Eu* values of the Mazıdağı phosphates were lower than those of these Tunisian, Moroccan, and Algerian phosphates [85], except for their Eu/Eu* values. As seen in Figure 10, the Algerian, Tunisian, and Moroccan phosphates were found to have negative Eu anomalies of 0.70, 0.71, and 0.66, respectively. The REE contents of phosphates are intensely controlled by geologic processes, and many researchers have suggested that they resemble modern seawater [30,86]. Differences in REE abundances have been related to the depth and duration of precipitation, as well as facies and grain size [74,87]. REE concentrations in phosphate samples vary due to differences in ocean chemistry, the variability of clastic phases, basin configuration, and the degree of influence of hydrothermal fluids [3,30,88]. Alternatively, extreme REE concentrations in phosphates are considered to be caused by the redistribution of REEs between authigenic and detrital phase forms during the diagenetic equilibration of phosphate with pore water [87,89].

4.5. Pb Isotope Geochemistry

The average concentrations (ppm) of ²⁰⁴Pb, ²⁰⁶Pb, ²⁰⁷Pb, and ²⁰⁸Pb isotopes in the Mazıdağı phosphates were 0.028 ppm, 0,54 ppm, 0.447 ppm, and 1.102 ppm, respectively (

Table 6. Pb isotope ratios of the Mazıdağı phosphates.

	204Pb	206Pb	207Pb	208Pb	206Pb/204Pb	207Pb/204Pb	208Pb/204Pb
DL	0.01	0.01	0.01	0.01
DY-01	0.035	0.68	0.550	1.36	19.40	15.71	38.86
DY-02	0.020	0.39	0.315	0.78	19.53	15.75	39.00
DY-03	0.025	0.48	0.393	0.98	19.20	15.72	39.20
DY-11	0.033	0.62	0.516	1.28	18.80	15.64	38.80
DY-12	0.028	0.54	0.468	1.10	19.30	15.70	39.30
DY-13	0.016	0.31	0.266	0.62	19.10	15.63	38.75
DY-14	0.038	0.73	0.602	1.48	19.20	15.84	38.90
DY-20	0.036	0.68	0.564	1.41	18.90	15.67	39.17
DY-21	0.028	0.54	0.440	1.10	19.30	15.71	39.30
DY-22	0.020	0.39	0.315	0.79	19.25	15.75	39.35
DY-23	0.018	0.35	0.282	0.72	19.44	15.67	38.80
DY-24	0.040	0.76	0.628	1.55	18.90	15.70	38.75
DY-25	0.030	0.57	0.470	1.18	19.00	15.67	39.33
Avrg	0.028	0.54	0.447	1.102	19.18	15.70	39.04

). The average ²⁰⁶Pb/²⁰⁴Pb, ²⁰⁷Pb/²⁰⁴Pb, and ²⁰⁸Pb/²⁰⁴Pb ratios were 19.18, 15.70, and 39.04, respectively (Table 6). According to Wang et al. [90], these Pb isotope results show that the lead in the Mazıdağı phosphates was of a lower crust origin (Figure 11a,b). In particular, the ²⁰⁷Pb/²⁰⁴Pb isotope ratios and their potential sources show that the Mazıdağı phosphates were sourced mostly from fluids or metals with a lower crust origin [91]. These ratios indicate that Pb isotope data can be related to oceanic crust areas, such as Atlantic and Pacific crust areas and pelagic sediments (Figure 11c,d). Dissolved Pb in seawater mostly derives from terrestrial input to the oceans, young continental margins, Pb derived from mid-ocean ridge basalts (MORB), the dissolution of wind-blown debris, and island arcs [92]. Highly radiogenic Pb isotope results in the North Atlantic have been associated with intense weathering processes in the former cratonic regions of northern Canada and Greenland [92]. In the present study, the Pb isotope results observed in the Fe–Mn crusts in deep water show that the highest ²⁰⁶Pb/²⁰⁴Pb radiogenic compositions occurred in the NW Atlantic (>19.1) whereas the lowest ²⁰⁶Pb/²⁰⁴Pb ratios occurred in the Southern Ocean (<18.9) and central North Pacific (<18.7). The Fe–Mn crusts spanned the water depth range of ~700 and 4600 m, including within the present-day NE Atlantic deep water and Mediterranean outflow water [93]. At the same time, the studied Pb isotope data also covered the same area as terrigenous, pelagic, biogenic sediments, and Mn nodules [92]. The Pb isotope ratios obtained indicate that the potential source of Pb in Mazıdağı phosphate ores was seawater from the shallow marine basin of the Tethys Ocean during the Upper Cretaceous period. This appears to be consistent with the geological and geochemical evidence [94,95]. Further, the Pb isotope ratios obtained indicate that hydrothermal activity at mid-ocean ridges contributes to the formation of phosphate deposits, such as the early Oligocene Paratethyan manganese deposits in Nikopol (Ukraine) [96] and Chiatura (Georgia) [97].

4.6. Origin of the Phosphate and Its Genetic Model

Marine phosphates form as irregular masses, layers, sands, oolites, and pellets along the continental margins in the different periods [98]. While some scientists have proposed that phosphates formed directly as a result of the inorganic precipitation of the phosphates from seawater, Burnett [98] reported that the phosphate concentrations in the pore waters are higher than in the bottom waters. Therefore, it has been suggested that marine phosphates can be formed by precipitation of phosphate in the pore waters of anoxic sediments. The presence of phosphates in pore waters is necessary for phosphate formation. REE concentrations in phosphates are up to 50–100 times higher than both shale-normalized values and bulk REE content in modern seawater [99]. Therefore, it is unlikely that phosphates were precipitated directly from seawater. Felitsyn and Morad [100] have indicated that the most important source of REEs in phosphate ores is organic material, and they suggested that the most important mechanism to explain REE enrichment in authigenic phosphate is the direct ejection of REEs from marine pore waters. In addition, it has been suggested that hydrothermal fluids may be sources of phosphate and REEs [101]. Besides seawater and hydrothermal fluids, diagenetic pore water is considered the main source of REEs in sedimentary phosphates [3].

In order to explain the formation of Mazıdağı phosphates, a genetic model was developed, which is shown in Figure 12. According to this model, seawater can be separated into several different layers, such as the upper oxic layer (oxygen content: 2.0–8.0 mL/L), the dysoxic layer (oxygen content: 0.2–2.0 mL/L), the suboxic layer (oxygen content: 0.0–0.2 mL/L), and the bottom anoxic layer (oxygen content: 0.0 mL/L), based on redox conditions [102]. A previous study found that phosphates were precipitated on slopes during the strong upwelling stage in an organic-rich basin of deeper anoxic/suboxic conditions [103]. Most of the settled organic material in the sediment–water interface of the seafloor is, firstly, oxidized; then, phosphate moves to the water above, along with the sediment pore water. At the same time, phosphate is adsorbed by the iron/manganese oxides and moves up to the slopes together with benthic flux containing large amounts of trace elements and REEs [104]. The Fe–Mn oxides in benthic flux enter pore water together with REEs and PO₄³⁻. PO₄³⁻ combines with Ca²⁺ to form phosphates, and eventually, PO₄³⁻, trace elements, and many REE phases (with most of these elements likely originating from hydrothermal vents (Figure 1 and Figure 12) collect in ores [10,58,96,97].

The phosphate deposits of the eastern Mediterranean region are almost always associated with chert, porcelanite, and oyster-bearing limestone, in addition to minor marl, chalk, and sandstone [105,106,107,108,109]. In this study, within the Mazıdağı phosphate deposits, chert levels were observed with a thickness of 25–40 cm in certain places (Figure 4). The development of chert levels is thought to have come to an end during the precipitation of carbonates and phosphates, or otherwise to have been precipitated simultaneously with the carbonates. Chert levels have also been observed in Moroccan, Jordanian, Tunisian, and Syrian phosphate deposits on the same belt [20,106,109,110,111,112]. Tlili et al. [112] claim that chert levels may be related to the bacterial activity observed in chert beds, which induces the dissolution of silica, which, in turn, saturates the depositional environment through silicic acid. Another mechanism that enhances this process is oceanic upwelling, which brings P and Si from the deep ocean (which is rich in these and other elements) to the surface. Organic matter preserved in shallow marine sediments and containing P and Si is then released and enriched in the pore water of the sediments through microbial mediation and diagenesis, leading to the precipitation of phosphate and siliceous deposits [6,102,103,104,105,106,107,113,114]. In addition, the significant enrichment of some trace elements (As, Cd, Cu, U, and Zn) in phosphate ores, according to PAAS, indicates the presence of hydrothermal contributions (Figure 12).

5. Conclusions

The Mazıdağı phosphates were formed on a shallow marine platform located in the south of the Neo-Tethys Ocean between the African–Arabic and Eurasian plates during the Upper Cretaceous period. The Mazıdağı phosphate deposits from bottom to top occur on the gently dipping northern flank of the Derik anticline. (1) Tasit phosphate (Turonian), (2) Kasrik (East Kasrik) phosphate, and (3) Şemikan (West Kasrık) phosphate (Santonian–Coniacian) are all phosphorus-rich members of the Karababa Formation. (4) Akras phosphate (Maastrichtian) is a phosphorus-rich member of the Kermav Formation, which is located in the eastern part of the study area. The Şemikan (West Kasrik) member was found to be the richest in phosphate, reaching 25–32% P₂O₅ in content. This Şemikan phosphate is the phosphate with the highest economic potential among the Mazıdağı phosphates. It consists of four different sectors. The average phosphate thickness in these sectors was found to be around 0.75 cm. The phosphate levels in the region were characterized by interlayers, alternating with chert, clay, and carbonate veins. While the oolitic phosphate levels were sometimes 20–30 cm thick, in some regions the total cherty, calcareous, and oolitic phosphate levels reached up to 4 m in places. The major oxide concentrations of the phosphate ores were CaO > P₂O₅ > LOI > SiO₂ > Na₂O > MgO > Fe₂O₃ > Al₂O₃ > K₂O > MnO > TiO₂. The CaO/P₂O₅ ratios for phosphate ores varied from 1.72 to 3.35, with a mean of 2.27 recorded. The average CaO/P₂O₅ ratio for limestone and chert samples was 8.85. The total trace element contents in the studied phosphate nodules varied from 395 to 2712 ppm, with a mean of 1742 ppm obtained. The phosphates contained more Cr, Sr, As, Cd, U, and Zn and lower amounts of V, Co, Ni, Rb, Y, Zr, Hf, Nb, Ba, Th, and Pb in comparison to PAAS. While the Sr from the large-ion lithophile elements was highly enriched, Rb and Ba were depleted compared to PAAS. Among the high field strength elements (HFSE; Nb, Hf, U, Th, Zr, and Y) Hf, Zr, Nb, Y, and Th were significantly lower compared to PAAS, except for U. Samples DY-24 and DY-25, taken from the Şemikan 3 region, contained high Au concentrations (0.14 and 0.15 ppm). Some trace element ratios indicated that the precipitation of the phosphate ores took place mainly in oxic and suboxic zones. The phosphate ores had low total REE concentrations, ranging between 3.30 and 22.1 ppm, and were found to have been enriched with mostly heavy REEs; they had lower amounts of middle REEs and light REEs when compared to PAAS. All of the the phosphate samples had negative Ce and positive Eu, Y, and Pr anomalies. The studied Pb isotope data covered the same area as terrigenous, pelagic, biogenic sediments, and Mn nodules. These ratios indicate that Pb isotope data can be related to oceanic crust areas such as the Atlantic and Pacific crust areas and pelagic sediments. The Pb isotope ratios obtained are also consistent with the existence of a hydrothermal contribution to ore deposits.

The source of phosphorus in shallow marine environments is related to the transport of phosphorus to oceans as a result of the weathering of continental rocks or the uplift of phosphorus and silica carried by cold water from deep-sea environments. According to the proposed model, phosphorus and silica in shallow marine environments are derived from both elements being recycled between the deep oceanic reservoir and the surface water by upwelling currents, followed by sedimentation. Upwelling currents spread deep, cold marine water to sea surfaces rich in Si and P. These elements are the basic nutrients for phytoplankton, the lowest tier in the marine food chain, which inhabit the photic zone or the upper 100 to 200 m of the seawater column. Phytoplankton follows zooplankton and other higher marine life, and bioproductivity is thus greatly enhanced by this process. Most of the oxygen in the water column below the photic zone is depleted by aerobic bacteria feeding on the rain of dead organic matter, thus creating an oxygen minimum zone. In shallow-water epicontinental shelves, such as those of the Neo-Tethys Afro-Arabian margins, the oxygen minimum zone coincides with the sea floor, thus producing a favorable anaerobic environment for the preservation of organic matter. The continuation of this regime maintains a high rate of sedimentation and, consequently, a high rate of burial, as well as the formation of organic-rich sediments.