A New Preparation Method of Microfauna from Gypsum: Micropaleontological Association from the Middle Miocene Badenian Gypsum Deposits of Paratethys

Evaporitic gypsum deposits represent an important paleoenvironmental record of the Miocene Badenian of the Carpathian Mountains belt. In this study, we developed a nontoxic method to concentrate calcareous microfossils from gypsum (CaSO4·2H2O), by treating the sulfate with ammonium acetate. We applied the newly developed method to gypsum collected from the Evaporitic Formation outcropping northward of Slănic-Prahova in the Eastern Carpathians. For the first time for this formation, we describe a calcareous microfossil assemblage characterized by the presence of planktonic foraminifera as well as cysts and fragments of calcareous algae.


Introduction
In the Earth's history, sulfate rich marine waters are known for late Precambrian (Vendian), Pennsylvanian-Triassic, Miocene to Quaternary [1]. The formed evaporitic deposits represent important archives, their formation indicating particular depositional environment and climatic conditions. Microfossils in gypsum are important in order to investigate the ecology of organisms and their limit to tolerate high salinities. Microfossils are also excellent paleoenvironmental indicators, offering a unique opportunity in order to gain environmental information at the time of evaporite formation. Microfossils in marls and clay-rich rocks are concentrated by classical methods as mechanical crushing, sieving and hand picking. If the host rock is evaporitic and rich in sulfates as gypsum, the abundance of microfossils is low and it is challenging to gain substantial fossil material. Hydrogen peroxide H 2 O 2 dissolves organic material and disaggregates particles but it does not affect sulfates. Due to the solubility of Ca-carbonate at low pH, a preparation with acids is not suitable. In this study, we developed a non-toxic, cheap method for extraction of microfossils from gypsum CaSO 4 ·2H 2 O formed during evaporation when sea water salinity increases until saturation causes mineral deposition. We tested the newly developed method on massive gypsum boulders of the Middle Miocene (Badenian) Radiolarian Shales and Spirialis formations (red/white arrows indicates the position of the volcanic ash layer dated at 13.32 ± 0.07 Ma by [14]; C) Lithostratigraphy at the Green Stone Hill (Muntele Piatra Verde) after [2,3,11,16]. (1) Mediterranean regional stages [17]; (2) Central Paratethys regional stages [18]; (3) Substages and Forams biozonation after [5][6][7][8] (4) Calcareous Nannoplakton Zonation [19]; (5) Events: (a) Last Common Occurrence (LCO) Sphenolithus heteromorphus [11]; (b) absolute age of a white tuff layer 13.7 ± 0.26 Ma [2]; (c) absolute age of a tuff layer 13.32 ± 0.07 Ma [14]; (6) formation names for the post-tectonic cover deposits of the Tarcău nappe. Radiolarian Shales and Spirialis formations (red/white arrows indicates the position of the volcanic ash layer dated at 13.32 ± 0.07 Ma by [14]; (C) Lithostratigraphy at the Green Stone Hill (Muntele Piatra Verde) after [2,3,11,16]. (1) Mediterranean regional stages [17]; (2) Central Paratethys regional stages [18]; (3) Substages and Forams biozonation after [5][6][7][8] (4) Calcareous Nannoplakton Zonation [19]; (5) Events: (a) Last Common Occurrence (LCO) Sphenolithus heteromorphus [11]; (b) absolute age of a white tuff layer 13.7 ± 0.26 Ma [2]; (c) absolute age of a tuff layer 13.32 ± 0.07 Ma [14]; (6) formation names for the post-tectonic cover deposits of the Tarcău nappe.

Separation
Ammonium acetate NH 4 CH 3 CO 2 is a white crystalline, hygroscopic, deliquescent substance with slight vinegar like odor and can be used with acetic acid to create a buffer solution [20]. It can be formed either by the reaction of ammonia and acetic acids or of acetic acid and ammonia carbonate. It is a frequent commercial and cheap, available substance of widespread use. In addition, as a component of buffer solutions, it is used as a food additive, a precursor of drugs, in agro industries, textile industries and in various chemical processes [21]. Ammonium acetate dissolves readily in water and ethanol [20] and is easy and safe to handle. At 4 • C, the solubility is~1480 g/L [22] and the solubility increases to~5300 g/L at 80 • C. As the dissociation constants for ammonium hydroxide and acetic acid are of identical values, ammonium acetate solutions have a pH-value of~7, independent from concentration [23]. Therefore, carbonates are insoluble in such a solution.
In contrast, gypsum is only moderately soluble in water (2.36 g/L at 25 • C), and it has a retrograde solubility, i.e., less soluble at higher temperatures (2.09 g/L at 75 • C) [24].
Lead and the alkaline-earth-sulfates are readily soluble in ammonium acetate solutions. The sequence of solubility of these sulfates is: lead-calcium-strontium and barium as the lowest soluble one. Calcium sulfate shows increasing solubility with temperature and concentration. A total of 1000 g of a 3N ammonium acetate solution at a temperature of 25 • C (corresponding to a concentration of 231 g/L) dissolves~26 g calcium sulfate [25]. As the solubility of ammonium acetate in water at 80 • C is~5300 g/L and the solubility of calcium sulfate in ammonium acetate solution is increasing with the concentration, the maximum of dissolved calcium sulfate is by potencies higher as 26 g/kg solution for such concentrations.
The sample prepared for this study belongs to the upper part of the Evaporitic Formation ( Figure 1B). To gain microfossil remains of gypsum rich rocks of the Muntele Piatra Verde sequence, Slănic Prahova, Romania, a representative gypsum rich sample (MV19) of 100 g was crushed in a jaw crusher and wet sieved for the fractions 63, 150, 250, 425 and 850 µm. To avoid precipitation of large amounts of salt during cooling down of a supersaturated solution, the experiment was started at room temperature and the associated maximum solubility of ammonium acetate. As the solubility of ammonium acetate is~1500 g/L at 20 • C, 750 g of ammonium acetate was dissolved in 500 mL deionized water at room temperature in a glass beaker by mixing with a magnetic stirrer. The resulting solution with a volume of 1.2 L, i.e., has a concentration of~8N. It is not recommendable to heat this solution directly on a heating plate as, within a short time, a precipitate is formed. Stirring the beaker is also not recommendable because of the fragility of the fossils. Therefore, the solution was filled in a wide neck bottle, a sieved fraction was added and the bottle was closed with a ground glass stopper. The bottle was placed in a compartment dryer at 80 • C for about 12 h. Possible gelatinous masses on the bottom of the bottle are a mixture of disaggregated fine-grained particles of the sample and precipitate from the solution. Both precipitate and undissolved sample were checked by powder XRD. No remaining gypsum was detected neither in the undissolved fraction of the sample nor in the precipitate. Main phases of the precipitate are Ca-acetate and a non-identified phase. Luckily, both phases are readily water-soluble. The remaining ammonium acetate solution was decanted and the bottle was refilled with deionized water and placed again at 80 • C in the compartment dryer for~1 h. Decanting and refilling was repeated once. The residue was spilled in a small beaker and again filled up with deionized water and treated with an ultrasonic cleaner to get rid of remaining fine particles. The cleaned sample was finally dried in the compartment dryer at 40 • C and afterward checked under the microscope and handpicked for final concentration and determination.

XRD Measurements
X-ray diffraction patterns of the sample before and after the treatment with acetate were measured on a Bruker AXS D8 diffractometer equipped with a one-dimensional strip-detector (lynx-eye) using a Geosciences 2020, 10, 158

of 14
Cu-kα (40 kV, 40 mA, sample rotation). The measuring conditions were: step-size 0.02 • , 2 sec/step (equivalent to 304 sec/steps with a conventional point-detector). Mineralogical compositions of samples before and after the treatment with ammonium acetate are given in Figure 2A,B. corrections.

Foraminifer Images
The determined foraminifers are displayed in Figures 3, 4, 5 and 6. For Figures 3, 4 and 6, scanning electron microscope (SEM) secondary electron images were done with a tungsten filament Jeol 6610LV microscope in the high vacuum mode. Fossil remains were placed on an aluminum pin mount with a carbon adhesive tab, and the samples were gold sputtered. Operating conditions were 10 kV with a working distance of 10 mm.
For Figure 5, we also used digital pictures taken with a Zeiss Stemi SV11 microscope with a Nikon digital camera adapted to it. 15 000 . 20 000 .

Carbonate Content
Before dissolution with ammonium acetate, we determined the carbonate content of the sample with a Scheibler calcimeter. A part of the sample was crushed and ca. 1 g was placed in the Scheibler apparatus. The sample was treated with a 10% HCl solution and the resulting CO 2 was volumetric quantified. The calculation of the carbonate content includes air pressure and temperature corrections.

Foraminifer Images
The determined foraminifers are displayed in Figures 3-6.     For Figures 3, 4 and 6, scanning electron microscope (SEM) secondary electron images were done with a tungsten filament Jeol 6610LV microscope in the high vacuum mode. Fossil remains were placed on an aluminum pin mount with a carbon adhesive tab, and the samples were gold sputtered. Operating conditions were 10 kV with a working distance of 10 mm.
For Figure 5, we also used digital pictures taken with a Zeiss Stemi SV11 microscope with a Nikon digital camera adapted to it.

Results and Discussion
In Figure 2A, XRD-powder analysis of the bulk rock sample shows the presence of gypsum with minor amounts of quartz, dolomite and celestine. Calcite is diluted below the detection limit of the method. As determined with the Scheibler apparatus, the total carbonate content of the sample is 3.51 weight % (average of two analyses 3.48 and 3.54 wt.%). After gypsum dissolution with acetate, in the residuum, quartz, dolomite, siderite, celestine and calcite were detected in the fraction 63-150 micron, which hosted most of the fossil remnants ( Figure 2B). Moreover, using the experimental data, namely the Scheibler measurement on total carbonate before dissolution as well as the partitioning between carbonates obtained by XRD after dissolution, it is possible to calculate a rough estimation of the fraction of various carbonates in the bulk sample. Accordingly, the content of calcite, dolomite and siderite was performed from the main peak areas (at 3.03, 2.89 and 2.80 Å):~5% calcite,~80% dolomite and~15% siderite, which corresponds to calcite content in the bulk sample MV19 before dissolution of 0.2 weight %. These quantitative estimations show how efficient the present method is, allowing final concentration of foraminifera from a bulk sample with trace calcite content.
The planktonic foraminifera assemblage determined in washed residuum is characterized by small-sized individuals, as indicated in For some of the identified species, the interval of occurrence within the Badenian regional stage is given in Table 1. This microfossil association contains planktonic taxa with a wide stratigraphic range. Few species are stratigraphically more restricted, being very abundant in the Langhian (Moravian) deposits [8,26,27].
The identified foraminifera assemblage shows similarities with the planktonic foraminifera assemblages mentioned from Slănic Formation by [9,10]. A similar occurrence of small planktonic foraminifera has been mentioned just below the gypsum deposition in NW Transylvanian Basin, indicating clear paleoenvironmental changes toward more cold waters [27]. During the Messinian stage, increasing abundance of small planktonic foraminifers in clays, before deposition of evaporates, were described by [33]. Taking into consideration that the gypsum deposition corresponds to a saline episode, in which marine connections are progressively closed and the water level dropped, we suggest that the identified planktonic foraminiferal assemblage is reworked or represents inputs of fresh water from less saline environments surrounding the lagoon where gypsum deposited. In this context, the ecological factor responsible for the small forms may be a combination of factors, such as cold water at the time of gypsum deposition and or oligotrophic conditions.
In the microfossil assemblage, cysts of Halicoryne morelleti (Pokorny), a green algae, were also identified ( Figure 6). Most cysts have a spherical shape, sometimes they are ovoid. They have no contact traces, which could indicate that they were separated during life. This type of algae is usually associated with microfossils of the Sarmatian age [34,35]. In only one area from Romania (Costei locality), these algae are mentioned for Badenian deposits [36]. In Poland, green algae assigned to Halicoryne cf. morelleti (Pokorny) are found together with marine shallow-water ostracod fauna indicating the Badenian age [37]. These green algae lived in marine environments, mainly shallow (up to 30 m depth), with normal salinity and subtropical-tropical waters [37,38], but they could adapt to lower salinity conditions [34]. They lived in low turbulence environments typical for coastal lagoons.

Conclusions
Ammonium acetate is a substance with a low chemical hazard. Treatment of gypsum rich rocks for fossil preparation with ammonium acetate solution is therefore easy to handle, fast and a low-cost method. No additional devices are necessary besides the standard equipment of a chemical laboratory. Using this method, micropaleontological studies of sedimentary sequences may be extended to the gypsum rich layers.
For the first time, we determined a foraminiferal association exclusively made of small sized planktonic species. The microfossil assemblage is represented by a mixed association of warm to cold water species, presenting similarities with the planktonic foraminifera assemblages mentioned for Slănic Formation. Green algae cysts were identified as well. No benthonic foraminifers have been found.