Effects of artificial diagenetic alteration on the microstructure, isotopic composition, and metal element concentrations in brachiopod shells

Abstract

In this study, we investigated the effect of burial diagenesis on the microstructure and isotopic (δ¹³C and δ¹⁸O) and chemical compositions of modern brachiopod (Terebratulina crossei) shells through controlled artificial diagenesis experiments. The shells were placed in sediment and artificial seawater mixtures and subjected to experimental conditions of 125°C and 75 MPa for 720 h. Statistically significant changes were observed in the isotopic and chemical compositions of the shells before and after the artificial experiments. Notable findings include decreases in δ¹⁸O values under all four experimental conditions; increases in Mn concentration in the carbonate powder-artificial seawater mixture, quartz powder-artificial seawater mixture, and artificial seawater; and decreases in the δ¹³C values in the carbonate powder-artificial seawater mixture and sandstone powder-artificial seawater mixture. The observed δ¹⁸O variations were predominantly influenced by temperature rather than by the isotopic and chemical compositions of the ambient sediments and fluids. The Mn concentration increased when the shells were placed in materials relatively poor in Mn (i.e., carbonate, quartz, and artificial seawater). This suggests that Mn originated from organic matter within the shells. The decrease in δ¹³C values is likely attributable to the thermal degradation of organic matter in the shells. Scanning electron microscopy (SEM) revealed minimal evidence of shell microstructure degradation and destruction due to the experiments. However, transmission electron microscopy (TEM) revealed traces of dissolution that were not discernible using conventional SEM. These findings underscore the importance of nanoscale analysis in future investigations of brachiopod shell-based paleoenvironmental reconstructions.

Introduction

Brachiopods taxonomically belong to the phylum Brachiopoda and possess two distinct shells: dorsal and ventral valves. Most brachiopods project a stalk-like pedicle through an opening (foramen) near the hinge of the ventral valve, allowing them to anchor to hard substrates. As filter feeders, brachiopods inhabit diverse environments ranging from equatorial to polar regions and coastal to deep-sea habitats. However, numerous species are sessile benthos and commonly found on relatively shallow seafloors, such as continental shelves (Rudwick, 1970). Brachiopods first appeared in the early Cambrian and have maintained a continuous and abundant fossil record throughout the Phanerozoic (Williams et al., 1996). The Paleozoic era witnessed the highest frequency and diversity of brachiopod occurrences among these periods. Despite the extinction of several brachiopod species during the Permian/Triassic boundary event, approximately 380 species across 112 genera persist in modern oceans (Eming, 1992). Approximately 90% of the extant brachiopods belong to the subphylum Rhynchonelliformea (Williams et al., 1996; hereafter called “brachiopods”).

Previous studies have demonstrated that brachiopod shells serve as exceptional indicators for reconstructing paleoceanographic conditions. The well-known Phanerozoic patterns in carbon and oxygen isotope compositions (δ¹³C and δ¹⁸O values, respectively) of marine calcifiers were defined using more than 4,000 measurements of biogenic carbonates, with brachiopod shells serving as the primary data source, particularly for the Paleozoic and Mesozoic eras (e.g., Veizer et al., 1997, 1999). The preferential use of brachiopod shells is attributed to their abundance, widespread distribution throughout the Phanerozoic, relatively low susceptibility to diagenetic alteration due to their primary mineralogical composition (low-magnesian calcite) and dense microstructures, and δ¹³C and δ¹⁸O values that are identical to those of calcite precipitated in isotopic equilibrium with ambient seawater (equilibrium calcite).

The following four screening criteria are commonly applied to evaluate the preservation of brachiopod shells (e.g., Ullmann and Korte, 2015; Yamamoto et al., 2017): (1) δ¹³C and δ¹⁸O values (Fujioka et al., 2019); (2) concentrations of minor elements (Mn, Fe, and Sr) that fluctuate due to diagenetic alteration (Brand and Veizer, 1980; Brand et al., 2012; Veizer et al., 1986); (3) luminescence/non-luminescence characteristics assessed using “cold cathode” techniques (Grossman et al., 1993; Mii et al., 2012; Popp et al., 1986a, 1986b); and (4) shell microstructures examined via scanning electron microscopy (SEM) (Bates and Brand, 1991; Brand et al., 2012; Popp et al., 1986b). Well-preserved shells or shell portions are non-luminescent and exhibit relatively high δ¹³C and δ¹⁸O values, high Sr and low Mn/Fe concentrations, and well-preserved microstructure. The contrary holds for altered shells.

Fossil brachiopod shells from geologically old successions (e.g., Mesozoic and Paleozoic) in tectonically active settings were commonly buried deep underground for extended periods. These shells have been subjected to burial diagenesis (which occurs deep underground under high-temperature and high-pressure conditions) and meteoric diagenesis (which occurs under near-surface conditions before or after deep burial). To better understand diagenetic processes in nature, separating and evaluating the effects of deep burial diagenesis from those of near-surface meteoric diagenesis is crucial, a task that presents considerable challenges.

In this study, we created artificial physical (temperature and pressure) conditions to simulate burial diagenesis and examined the alterations in the microstructure and isotopic and chemical compositions of modern brachiopod shells under these conditions. Only two previous studies have reported alterations in brachiopod shells under artificial diagenetic conditions (Riechelmann et al., 2016; Casella et al., 2018). In these studies, brachiopod shells were subjected to alterations induced by heat and hydrothermal fluids. The microstructure and isotopic and chemical compositions of the artificially altered shells were then compared with those of the initial (unaltered) shells. Finally, the effects of burial diagenesis on the microstructural and geochemical characteristics were evaluated. The heating method allows for the assessment of temperature effects but does not permit the evaluation of pressure effects (Casella et al., 2018). In their hydrothermal experiments, the pressure conditions corresponded to the water vapor pressure generated at the experimental temperatures, indicating that the pressure was not regulated (Riechelmann et al., 2016; Casella et al., 2018). Therefore, the physical and chemical conditions used in their experiments did not adequately represent the natural conditions of burial diagenesis. Furthermore, these studies did not consider the influence of host rock composition on burial diagenesis processes.

In this study, we aimed to quantitatively evaluate the changes in carbon and oxygen isotope compositions and metal element (Na, Mg, Mn, Fe, and Sr) concentrations in the shells of extant brachiopods. In addition, we examined the changes in shell microstructure under high-temperature and high-pressure conditions (125°C, 75 MPa; closed system) over an extended period (720 h). The conditions correspond to those at a depth of 3,500 m from ground level, utilizing a geothermal gradient of 2.5–3.0°C/100 m (with the average geothermal gradient in Japan being 3.0°C/100 m) and rock density of 2.1–2.2 g/cm³ (roughly equal to the density of consolidated but loosely compacted sedimentary rock). We selected these five elements because Fe and Mn concentrations increase while Na, Mg, and Sr concentrations decrease in diagenetically altered carbonates (e.g., Brand and Veizer, 1980).

Materials and Methods

Brachiopod shells

Modern brachiopod specimens were collected from the mouth of Otsuchi Bay (70–75 m water depth; 39°21.35'N, 142°00.14'E) on October 15, 2018, using a dredger operated by the R/V Yayoi of the International Coastal Research Center, The University of Tokyo (for oceanographic conditions, refer to Yamamoto et al., 2013; Takayanagi et al., 2015). From the collection of Terebratulina crossei, we selected the four most pristine, relatively large, and similarly sized shells (i.e., exhibiting minimal coverage by other sessile organisms and fewer shell scars) for this study: B-5, B-9, B-10, and B-16 (Fig. 1). After the hard and soft parts in the shells were removed, the shells were rinsed with distilled water and allowed to dry at 7–17°C temperature. The brachiopod specimens analyzed in this study were stored at room temperature at the Museum of Natural History, Tohoku University.

Fig. 1.

(a–d) Terebratulina crossei shells used for the artificial diagenesis experiments in this study. Specimens were collected alive at 70–75 m water depth from Otsuchi Bay, Iwate Prefecture, northern Japan [refer to Takayanagi et al. (2015) for locality]. Note that altered shells lost their original color and became whiter than the unaltered shells.

Artificial diagenesis experiments

Artificial diagenesis experiments (ADEs) were performed using a pressure vessel and high-temperature experimental oven provided by the Japan Agency for Marine-Earth Science and Technology (JAMSTEC). Sections of modern brachiopod shells were sealed in a silicone rubber tube (length: 12 cm, inner diameter: 12 mm, outer diameter: 16 mm, heat resistance: 200°C) containing artificial seawater and sediment equivalent materials (hereafter referred to as “sediment”). The artificial seawater was prepared by dissolving Tetra Marine Salt Pro (Tetra, Blacksburg, VA) in Milli-Q water. The tube was sealed using a silicone rubber stopper and silicone-based liquid gasket (Fig. S1). This assembly was then placed in a pressure vessel filled with Milli-Q water (Fig. 2).

Fig. 2.

Schematic diagram illustrating the methodology employed in the artificial diagenesis experiments.

In this study, we employed the indirect pressurization method with artificial seawater as the pressure medium for the pressure vessel. This ensured that the experiments remained unaffected in the event of a seal failure or pressure medium water entering the silicon tube. The pressure vessel was pressurized via a high-pressure pump and subsequently placed in a 125°C oven. At a temperature of 125°C and pressure of 75 MPa, the valve connected to the pump was closed, and the system was maintained under these conditions for 720 h to artificially alter the brachiopod shells.

Brachiopod shells occur in various types of rocks under natural conditions. We packed several brachiopod shells, namely B-5, B-10, and B-16, with quartz, carbonate, and sandstone powders, respectively, to replicate the variations in brachiopod-bearing rock. The quartz powder was produced from quartz sand with a grain size of 600–850 μm (Fujifilm Wako Pure Chemicals Corporation, Osaka, Japan). The carbonate powder was prepared from the aragonite shells of fossil bivalves collected from the Pliocene Tatsunokuchi Formation in Yakegawara, Sendai City, Miyagi Prefecture, northeastern Japan. The sandstone powder was produced from the Chibanian (Pleistocene) Yabu Formation sandstone in Semata, Ichihara City, Chiba Prefecture, Japan. These powders were sieved through a 500-mesh screen, and the sediment in the silicon rubber tubes was medium sand-sized or smaller. The weight of the artificial seawater was adjusted to achieve a water/rock ratio (W/R ratio; weight ratio) of ~0.2, given that the porosity of the consolidated clastic rocks, extending from the surface to a depth of ~4,000 m, was 10–30%. (Loucks et al., 1984). One shell (B-9) was packed exclusively with artificial seawater (~9.3 mL) to compare our experimental results with those reported by Riechelmann et al. (2016). Table 1 summarizes the brachiopod shell, sediment, and artificial seawater weights and water/rock ratio for each sample. The water/rock ratio represents the ratio of “the weight of seawater” to “the weight of the brachiopod shell plus sediment.” Rhynchonelliform brachiopod (Williams et al., 1996) shells are exclusively composed of low-Mg calcite. Using X-ray diffraction analysis, we confirmed that shell mineralogy did not change before and after ADEs, following the methodology outlined by Fujioka et al. (2019).

Table 1.

Weights of brachiopod shells, sediments, and artificial seawater utilized in the artificial diagenesis experiments

Sample no.	Weight of brachiopod shell (g)	Sediment	Weight of sediment (g)	Weight of artificial seawater (g)	W/R (water/rock ratio)
B-5	0.396	Quartz	13.8	2.89	0.20
B-9	0.392	—	0	9.29	23.70
B-10	0.402	Carbonate (Aragonite)	14.1	2.99	0.21
B-16	0.374	Sandstone (Yabu Formation)	13.2	2.94	0.22

The chemical composition of the seawater in the silicon rubber tubes may have changed during the ADEs owing to their deterioration and subsequent leaching of chemicals. We conducted the ADEs under identical temperature and pressure conditions using tubes filled exclusively with artificial seawater. The chemical composition of the artificial seawater was checked before and after the experiments (= unaltered and altered artificial seawater, respectively).

Sampling for isotopic and chemical analyses

Shells (ventral valves) were cut along the maximum growth axis using a diamond saw. Half of the shells were used for the ADEs (Fig. 3). Because this study aimed to elucidate how diagenesis altered the primary isotopic and chemical compositions of the shells, those used in the experiments were not chemically treated, as in two previous studies (Riechelmann et al., 2016; Casella et al., 2018).

Fig. 3.

Methodology employed to collect samples for analyzing isotopic compositions and metal element concentrations in the examined shells. A = anterior shell edge; P = posterior shell edge.

The altered and unaltered shells were reinforced with epoxy resin and cut parallel to the maximum growth axis to yield ~1 mm wide slabs. We collected samples for carbon and oxygen isotope analysis from these slabs and samples for chemical analysis from cross-sections parallel to and <2 mm away from the maximum growth axis (i.e., the opposite side of the slabs from the unaltered and altered shells). Powder samples for isotope and chemical analyses were extracted from the unaltered and altered shells at 2 mm intervals along the inner shell surface using a microdrill with 0.12 mm diameter. The locations from which the powder samples were collected were sequentially numbered, starting with 1 for the most posterior aspect and increasing toward the anterior aspect. These locations are depicted in Fig. 4.

Fig. 4.

Profiles of the isotopic compositions (δ¹³C and δ¹⁸O values), metal element (Mn, Fe, Sr, Mg, and Na) concentrations, and mean CL intensity (MCLI) of the examined shells before and after the artificial diagenesis experiments (indicated as “unaltered” by solid lines and “altered” by broken lines, respectively). The numbers below the shells denote locations where powder samples for isotopic and chemical analyses were collected and MCLI values were measured. The scanning electron microscopy (SEM) observations were conducted on 4–5 “perpendicular lines” for each unaltered shell and altered shell (B-10, t1–t5; B-5, t1–t5; B-16, t1–t4; B-9, t1–t5). Refer to the “Carbon and oxygen isotope compositions” subsection in the “Results” section for the definitions of intervals 1 and 2.

The inner shelf surface was selected because of its reduced susceptibility to metabolic and kinetic isotope fractionation, thereby providing a time-averaged record of isotopic and chemical compositions (e.g., Yamamoto et al., 2013). Notably, inter-shell variations in the carbon and oxygen isotope compositions were minor (e.g., Fujioka et al., 2019), except near the posterior shell edge (interval 1 in Fig. 4; for the definition, refer to the “Carbon and oxygen isotope compositions” subsection in the “Results” section). An equal number of powder samples was obtained from corresponding portions of the unaltered and altered shells. In other words, the sampling location was symmetrical with respect to the maximum growth axis (Fig. 3). This sampling strategy enabled a comparison of the isotopic and chemical compositions between unaltered and altered shell portions and facilitated a quantitative evaluation of the alterations caused by ADEs.

Scanning electron microscopy (SEM)

After collecting the powder samples for chemical analysis, the shells were cut perpendicular to the maximum growth axis. The microstructures of the lower shell secondary layer on the perpendicular slabs were observed via SEM. Observations were performed in close proximity to the sampling locations for chemical analysis. The SEM observations were conducted on 4–5 “perpendicular lines” per one unaltered shell and per one altered shell to investigate the differences in their microstructures (B-5, t1–t5; B-9, t1–t5; B-10, t1–t5; B-16, t1–t4). The positions of the perpendicular lines were adjusted to examine the microstructures of the corresponding portions of the unaltered and altered shells.

A Keyence UE-8800 scanning electron microscope (Tokyo) was used at the Department of Earth Science, Graduate School of Science, Tohoku University, operating at an accelerating voltage of 5 kV. The shells were polished, etched with 0.2 mol/L hydrochloric acid (HCl) for 40 s, and coated with carbon.

Transmission electron microscopy (TEM)

To compare the microstructures of the shells before and after ADEs at a finer scale, thin foils were cut out from the polished sections using a focused ion beam (FIB; Thermo Fisher Scientific, Scios 2) installed at the Geodynamics Research Center of Ehime University. Foils of 12–17 μm in width and 5–9 in height were cut perpendicular to the growth direction from both primary and secondary layers of the brachiopod shells before and after the ADEs. The foils were mounted on a Cu lift-out grid with Pt deposition and observed using transmission electron microscopy (TEM; JEOL, JEM-2100F) at Ehime University. The TEM observations were performed at an accelerating voltage of 200 kV for bright-field imaging and selected-area electron diffraction.

Isotopic analysis

The carbon and oxygen isotope compositions (δ¹³C and δ¹⁸Ο values, respectively) of brachiopod (calcite) powder samples were analyzed using a Thermo Fisher Delta V Advantage isotope ratio mass spectrometer coupled to a ThermoQuest Kiel-III automated carbonate device, at the Department of Earth Science, Graduate School of Science, Tohoku University. The carbon and oxygen isotope compositions of carbonate (aragonite) powder packed with a brachiopod shell (B-10) were also measured. The samples (~0.1 mg) were reacted with 100% phosphoric acid at ~72°C. The isotope ratios were expressed in conventional (δ‰) notation and calibrated to the NBS-19 international standard relative to Vienna Pee Dee Belemnite (VPDB). The external precision (1σ) based on replicate measurements (n = 41) of the laboratory reference material (JCt-1; Okai et al., 2004) was 0.03‰ for carbon isotope analysis and 0.05‰ for oxygen isotope analysis. In accordance with Fujioka et al. (2019), we did not consider the Mg effect on the δ¹⁸O values (e.g., Tarutani et al., 1969) of brachiopod shells (Brand et al., 2013) in this study.

The carbon isotope composition of dissolved inorganic carbon (δ¹³C_DIC) and oxygen isotope composition (δ¹⁸O_SW) of the artificial seawater samples were analyzed using cavity ring-down spectroscopy (L2120-I Analyzer; Picarro, USA) at the Atmosphere and Ocean Research Institute, University of Tokyo, without chemical preprocessing. The measured carbon isotope ratios were expressed in conventional (δ‰) notation and calibrated against the NBS-19 international standard relative to Vienna Pee Dee Belemnite (VPDB). The measured oxygen isotope ratios were calibrated against an in-house standard and converted to the Vienna Standard Mean Ocean Water (VSMOW) scale. Resultant δ¹⁸O_SW VSMOW values were converted to δ¹⁸O_SW PDB values using the equation provided by Brand et al. (2014; δ¹⁸O (VPDB) = 0.97001 × δ¹⁸O (VSMOW) – 29.99), with two values indicated to facilitate comparability. The external precision (1σ) values for these analyses, derived from replicate measurements of the in-house standard, were 0.09‰ for δ¹³C_DIC (n = 26) and 0.23‰VSMOW (0.22‰VPDB) for δ¹⁸O_SW (n = 47).

Chemical analysis

Powdered brachiopod shell samples were dissolved in ~5.0–10.0 mL of volume-specific 2% (v/v) nitric acid (HNO₃). The acid volume was adjusted to achieve a Ca concentration of ~8–10 ppm in the solutions for most analyses. The concentrations of five metal elements (Na, Mg, Mn, Fe, and Sr) were quantified using an Agilent 7700x inductively coupled plasma mass spectrometer at the Department of Earth and Environmental Sciences, Nagoya University. The concentrations were expressed as ppm (= μg/g). The concentrations of the carbonate powder packed with the B-10 shell were also measured using the same procedure as that for the brachiopod shells. Metal element concentrations were measured in both unaltered Milli-Q water and unaltered artificial seawater. We also measured the altered Milli-Q water in the pressure vessel and altered artificial seawater packed within the B-9 shell without any additional material. The precision of the analytical method was expressed by the standard deviation (σ) of repeated analyses of the laboratory reference materials; JCt-1 from a Tridacna gigas shell, provided by the Geological Survey of Japan (Okai et al., 2004); and ECRM 752-1 from limestone, provided by the Bureau of Analysed Samples, England (https://webshop.bam.de/media/wysiwyg/Kategorien/Referenzmaterialien/Eisen_Stahl/Zuschlagstoffe_Feuerfestmaterialien/Zertifikate/b752_1e.pdf, accessed on June 22, 2024). The low concentrations of certain elements in the two standard materials (e.g., Mn and Fe in JCt-1 and Na in ECRM) necessitated the use of both to ensure good repeatability of the measurements. For the powder samples, the precision was 1.6% for Na and 0.8% for Sr (as determined by analyzing JCt-1) and 2.2% for Mg, 1.7% for Mn, and 3.1% for Fe (as determined by analyzing ECRM 752-1). For the water samples, the precision was 2.0% for Sr (as determined by analyzing JCt-1) and 19.0% for Na, 0.9% for Mg, 0.7% for Mn, and 0.2% for Fe (as determined by analyzing ECRM 752-1).

Cathode cathodoluminescence images

Thin sections were prepared for cathodoluminescence (CL) imaging using a Reliotron VII cold cathodoluminescence microscope (Relion Industries, Bedford, MA, USA) at the Department of Earth Science, Graduate School of Science, Tohoku University. The electron beam current was 1.0 mA, the voltage was 10.5–16.3 kV, and the vacuum was 29–36.5 mTorr (= 3.8–4.9 Pa). The spot diameter of the cathode beam measured approximately 1 cm. The analog gain was maximized, and the exposure time was set to 40 s when capturing the images. In this study, we measured the mean CL intensity (MCLI) near the sites where the powder samples were collected to quantify the luminescence intensity of the CL images (refer to Fujioka et al., 2019 for this method). The measured MCLI value was expressed as the mean red value (R) over the observed area of the CL image (200 μm × 1,900 μm). The mean red value was measured using ImageJ software (ver. 1.53k).

Results

Shell microstructure

SEM images revealed that the cross-sections of the unaltered shells consisted of calcite fibers, which were characteristic components of the secondary shell layer of Rhynchonelliformea brachiopods (Figs. 5 and S2). In the cross-sections, the calcite fibers exhibited a horizontally elongated pine needle-like shape near the shell edge (t1–t3 sections; Fig. 5g) and saddle shape characterized by an upper concave and lower convex fiber surface or trapezoidal form around the central part of the shells (t3–t5 sections; Fig. 5a, c, e). Calcite fibers were rarely and partly amalgamated in a few shells (Fig. 5g). Most parts of the altered shells exhibited a microstructure consistent with that of the unaltered shells. However, two or three calcite fibers were amalgamated in some areas (Fig. 5d, f). Considerable fiber fusion was observed in the t5 section of the B-10 shell, where ~50–100 fibers were amalgamated to form one large block (Fig. 5b).

Fig. 5.

SEM images of (a, c, e, g) unaltered and (b, d, f, h) altered fibers from the examined shells. The observed points (t1–t5) are shown in Fig. 4. Black arrows indicate areas where fiber amalgamation is observed. pn = punctae.

To examine the microstructure of the brachiopod shells before and after the ADEs, two foils were cut out from the primary and secondary layers of the cross-sections of one sample (B-5) using the FIB (Fig. 6a, b). Figure 6c and d present backscattered electron images of foils from the primary layers. In the unaltered shell, the boundaries between individual fiber units were tightly sealed with virtually no gaps (Fig. 6c). Conversely, the altered shell revealed small gaps between each fiber unit (red arrows in Fig. 6d). Figure 6e–h show the bright-field TEM images of the foils, which clearly show small openings between individual fiber units in the altered shell (as indicated by the red arrows in Fig. 6f). Furthermore, at magnified views, the interface of the fiber units in the unaltered shell appeared smooth (Fig. 6g), whereas the unit boundary in the altered shell was slightly wavy (Fig. 6h). This suggests that a slight dissolution of the brachiopod shell occurred at the unit interface during the experiments and resulted in the elution of small amounts of calcium and carbonate ions into the solution.

Fig. 6.

Comparison of the microstructure of (a, c, e, g) unaltered and (b, d, f, h) altered shells. (a, b) SEM (back-scattered electron) images illustrate the trench sites from which (c, d) thin foils were cut out using a focused ion beam (FIB). (c) and (d) represent lateral views (back-scattered electron images) of the foils, with the areas from which transmission electron microscopy (TEM) images (e) and (f) were taken indicated by dotted lines. The red arrows in the SEM image (d) and TEM image (f) denote small openings between individual fiber units, while the blue arrows in (e) and (f) indicate the beam damage inflicted by the ion beam during FIB processing (the ion beam was irradiated from the direction parallel to the arrows). (g, h) Magnified TEM images were taken from the square areas shown in (e) and (f), with insets displaying electron diffraction patterns collected from the crystals (units) in the lower areas shown in black.

Carbon and oxygen isotope compositions

We used a paired sample t-test to assess the statistical significance of the changes in δ¹³C and δ¹⁸O values and Mn concentration in the brachiopod shells, which exhibited systematic changes due to the ADEs (Figs. 7, 8, and S3). Statistical tests were not performed on the concentrations of the other metal elements (Na, Mg, Fe, and Sr) because of large within-shell variations observed before and after the ADEs, which indicated no specific changes.

Fig. 7.

Statistically significant changes in the geochemical compositions of the examined shells and artificial seawater resulting from the ADEs. The upward and downward arrows indicate increases and decreases, respectively. SW = artificial seawater.

Fig. 8.

Box plots showing the changes in isotopic compositions (δ¹³C and δ¹⁸O values) and metal element (Mn, Fe, Sr, Mg, and Na) concentrations of the examined shells before and after the ADEs.

The carbon and oxygen isotope compositions of brachiopod shells exhibit substantial variation across different shell portions (Auclair et al., 2003; Yamamoto et al., 2010a, 2010b, 2011, 2013). The inner shell surface along the maximum growth axis, referred to as the inner series, in certain brachiopod species has been shown to reliably record δ¹³C_DIC and temperature/δ¹⁸O of the ambient seawater (Yamamoto et al., 2010a, 2010b, 2013; Takayanagi et al., 2015) because of the minimal fractionation effect in this shell region, which results from the very slow growth (shell thickening) rate. However, the posterior half of the inner shell surface, containing specialized shell portions (e.g., muscle scars and foramen), exhibited lower δ¹³C and δ¹⁸O values than those in calcite precipitated in isotopic equilibrium with ambient seawater at brachiopod growth sites. This difference is attributed to metabolic and kinetic isotope fractionation (e.g., Carpenter and Lohmann, 1995; Yamamoto et al., 2013; Takayanagi et al., 2015). For these reasons, Takizawa et al. (2017) divided the inner shell surface of Kikaithyris hanzawaii along the maximum growth axis into three intervals based on shell morphology and isotopic and geochemical data. They also employed data from interval 3, the anterior interval, for paleoenvironmental analysis. Fujioka et al. (2019) applied this division to Terebratalia coreanica and Laqueus rubellus shells.

However, T. crossei examined in this study showed no distinct abrupt changes in shell morphology (shape and thickness) along the maximum growth axis. Therefore, we could not divide the shell into multiple portions according to its morphology. Instead, two intervals were established based on the δ¹³C profile (Fig. 4). The maximum-value point was located 17–23 mm from the posterior shell edge in the four examined shells. Consequently, the section was divided into two intervals: interval 1 extends from the posterior shell edge to the maximum-value point, and interval 2 extends from the maximum-value point to the anterior shell edge. The initial isotopic composition of interval 1 of the T. crossei shells was likely affected by metabolic and kinetic isotope fractionation.

The δ¹³C and δ¹⁸O values of the four unaltered T. crossei shells (n = 86) ranged from –0.22–1.75‰ and 0.71–1.85‰, respectively, in interval 1 and 0.17–1.73‰ and 0.60–1.84‰, respectively, in interval 2 (Fig. 4, Table S1). The amplitudes of the four δ¹³C profiles were larger in interval 1 (1.02–1.35‰) than those in interval 2 (0.18–1.03‰) due to local minima in the former. In contrast, the amplitude of the δ¹⁸O profiles was similar in intervals 1 and 2 (interval 1: 0.34–0.71‰, interval 2: 0.50–0.78‰). The δ¹³C and δ¹⁸O profiles of the inner surface of T. crossei were previously reported only by Yamamoto et al. (2013), who analyzed a single shell collected from the same location and depth as this study. Their analyzed portion corresponds to interval 2 in this study. Their δ¹³C values ranged from 1.05–1.59‰, with an average of 1.3‰, which were identical to those of the four examined shells. In contrast, their δ¹⁸O values ranged from 1.63–1.93‰ with a mean of 1.82‰, which overlapped with those of the B-9 shell and were slightly greater than those of the B-5, B-10, and B-16 shells.

The δ¹³C and δ¹⁸O values of the four altered T. crossei shells (n = 85) ranged from –0.07–1.37‰ and 0.50–1.31‰ in interval 1 and 0.44–1.55‰ and –0.14–1.40‰ in interval 2, respectively (Figs. 4, 7, and Table S2). The amplitude of the δ¹³C profiles was larger in interval 1 (0.44–1.08‰) than that in interval 2 (0.38–0.80‰), similar to the unaltered shells (1.02–1.35‰ in interval 1 versus [vs.] 0.18–1.03‰ in interval 2). In the B-5 and B-16 shells, the amplitude of δ¹⁸O profiles was larger in interval 1 than that in interval 2. In comparison, the amplitude of these intervals was almost identical for the B-9 and B-10 shells.

Both δ¹³C and δ¹⁸O values tended to be lower in the altered shells than in the unaltered shells. This trend was particularly evident for the δ¹⁸O values (Figs. 4 and 7). The altered δ¹³C values were ~0.1–0.2‰ lower than those of the unaltered shells (Fig. 7). The results from two of the four experimental conditions, specifically the carbonate powder-artificial seawater mixture and sandstone powder-artificial seawater mixture, were statistically significant (p < 0.05). The δ¹⁸O values of the altered shells were ~0.4–0.6‰ lower than those of the unaltered shells. These decreases were significant (p < 0.05) under all the experimental conditions. The δ¹³C and δ¹⁸O values of the carbonate (aragonite) powder used to pack the B-9 shell decreased by 0.24‰ after the ADE (Table 2).

Table 2.

Carbon and oxygen isotope compositions and metal element concentrations in the carbonate powders and water (artificial seawater and hydraulic water) used in the artificial diagenesis experiments

Material	Condition	Mineralogy	Mn (ppm)	Fe (ppm)	Sr (ppm)	Mg (ppm)	Na (ppm)	δ13C (‰VPDB)	δ18O (‰VPDB)	δ18O (‰VSMOW)
Carbonate powder	Unaltered	Aragonite	3.97	32.7	2,200	81.2	2,910	0.38	0.84	—
Carbonate poweder altered with B-10 shell	Altered	Aragonite	3.90	54.1	2,040	167	2,710	0.23	0.60	—
Artificial seawater	Unaltered	—	0.144	0.043	4.71	908	3,540	–9.94	–38.9	–9.21
Artificial seawater	Unaltered	—	0.193	0.027	4.90	958	3,870	–10.1	–38.9	–9.21
Mili-Q water	Unaltered	—	0.042	0.029	0.015	5.10	85.0	NA	–39.0	–9.29
Mili-Q water	Unaltered	—	0.036	0.018	0.015	3.44	80.9	NA	–39.0	–9.33
Hydraulic fluid (Mili-Q water)	Altered	—	0.234	2.36	0.014	6.12	82.2	–6.82	–46.4	–16.9
Hydraulic fluid (Mili-Q water)	Altered	—	0.244	2.31	0.003	4.82	81.5	–6.87	–46.4	–16.9
Artificial seawater	Altered	—	0.149	0.058	4.22	822	3,820	1.6	–46.4	–16.9
Artificial seawater	Altered	—	0.133	0.054	4.19	807	3,820	1.3	–46.3	–16.9
Artificial seawater altered with B-9 shell	Altered	—	1.06	0.041	4.79	753	3,860	6.23	–46.2	–16.8

Numerous studies (e.g., Swart, 2015) have demonstrated that the δ¹³C and δ¹⁸O values of carbonate shells and sediments are modified during diagenesis through dissolution, recrystallization, and chemical precipitation from interstitial water (cement). SEM and TEM analyses in this study confirmed the absence of cements in the shells post-ADEs, indicating that the changes in δ¹³C and δ¹⁸O post-ADEs are attributable to the former.

The δ¹³C_DIC values of the artificial seawater exhibited anomalous increases before and after the ADEs (Fig. 7). The δ¹³C_DIC values of the artificial seawater packed with the B-9 shell and one devoid of any additional material increased from –10‰ to 6.23‰ (increase of 16.23‰) and –10‰ to 1.45‰ (by 11.45‰), respectively. The δ¹⁸O_SW values of the artificial seawater decreased from –9.2‰VSMOW (–38.9‰PDB) before the ADEs to –16.8‰VSMOW (–46.2‰PDB) for the artificial seawater packed with the B-9 shell and –16.9‰VSMOW (–46.4‰PDB) for one without any additional material post-ADEs. These decreases of 7.6 and 7.7‰VSMOW (7.3 and 7.5‰VPDB) were similar to those of the hydraulic water (Mili-Q water), which decreased from –9.3‰VSMOW (–39.0‰PDB) to –16.9‰VSMOW (–46.4‰PDB; decrease = 7.6‰VSMOW, 7.4‰PDB). These decreases indicate the potential leaching of carbon and oxygen from the silicon tube or gasket in these experiments; however, their isotopic compositions remain unknown (see the “Changes in carbon and oxygen isotope compositions” subsection in the “Discussion” section).

Metal element concentrations

The metal element concentrations in the four unaltered shells (n = 86; Fig. 4 and Table S1) were 1,660–3,420 ppm for Na (median of the single shell: 2,300–2,360 ppm), 1,520–4,120 ppm for Mg (median: 2,410–3,230 ppm), 0.50–19.6 ppm for Mn (median: 1.33–1.65 ppm), 3.72–894 ppm for Fe (median: 11–28 ppm), and 598–1,040 ppm for Sr (median: 869–935 ppm). The Na concentration exhibited a nearly flat profile between 2,000 and 3,000 ppm, whereas the Mg concentration fluctuated irregularly throughout intervals 1 and 2. The Mn concentration showed flat profiles, although it increased toward the anterior shell edge in interval 2 of two shells (B-9 and B-16). The Fe concentration was low (mostly 15–30 ppm) and associated with 100–900 ppm local peaks. The Sr concentration exhibited a flat profile between 800 and 1,000 ppm. The ranges of the metal element concentrations described above and their profiles depicted in Fig. 4 are similar to those in brachiopod shells reported in previous studies (e.g., Brand et al., 2003; Fujioka et al., 2019).

The concentrations in the altered shells (n = 85; Fig. 4 and Table S2), excluding one outlier data set, were 1,910–3,990 ppm for Na (median: 2,160–2,430 ppm), 1,590–4,100 ppm for Mg (median: 209–3,130 ppm), 3.04–25.6 ppm for Mn (median: 5–20 ppm), 2.96–552 ppm (median: 12–80 ppm) for Fe, and 676–1,090 ppm (median: 901–960 ppm) for Sr. The Na, Mg, Fe, and Sr concentration profiles were similar to those of the unaltered shells. The Mn concentration in the altered shells fell within narrow ranges, with flat profiles similar to those of the unaltered shells. The former was higher, with a wider range of medians than the latter (Fig. 7). These increases were statistically significant (p < 0.05) under the three experimental conditions: carbonate powder-artificial seawater mixture, quartz powder-seawater mixture, and artificial seawater.

Similar to the isotopic compositions mentioned above, we assumed that the extent of diagenetic alteration was represented by differences in the mean metal element concentrations before and after the ADEs (Fig. 4). No clear trends were observed for the Na, Mg, Fe, and Sr concentrations. In contrast, the Mn concentration increased post-ADEs in all shells. The increase was relatively large (10–20 ppm) when the shells were packed with seawater (B-9) or carbonate powder (B-10) and small (4–5 ppm) when the shells were packed with quartz powder (B-5) or sandstone powder (B-16) (Fig. 8).

The metal element concentrations in the carbonate powder packed within the B-10 shell changed before and after the ADEs, except for the Mn concentration (Table 2). The Mg and Fe concentrations increased by 86 and 21 ppm, respectively, whereas the Na and Sr concentrations decreased by 200 and 160 ppm, respectively.

Table 2 lists the metal element concentrations in the waters used for the ADEs. The Mg and Sr concentrations in the artificial seawater packed with no additional material decreased by 100–150 ppm and 0.6 ppm, whereas the Na, Mn, and Fe concentrations did not show apparent changes. (Figs. 7 and 8). The Mg and Mn concentrations in the artificial seawater packed with the B-9 shell decreased by 150–200 ppm and increased by 0.9 ppm, respectively, whereas the Na, Fe, and Sr concentrations exhibited minimal variation. The Mn and Fe concentrations in the Milli-Q water in the pressure vessel increased by 0.2 ppm and 2 ppm, respectively. We assessed the significance of these changes in metal element concentrations before and after the ADEs by analyzing their relative standard deviations (%) and confirmed that they are significant.

Cathode cathodoluminescence images

The unaltered T. crossei shells (Figs. 9a, c, e, g, and S4) exhibited weak blue luminescence. This blue luminescence results from crystal defects in calcite, indicating the absence of an activating element, Mn (Sippel and Glover, 1965; Machel et al., 1991; Habermann et al., 1998). In this study, we consider this blue luminescence as “non-luminescence.” Weak red luminescence was detected in the stripes parallel or inclined at an angle of approximately 20° to the inner surface of the shell. This stripe luminescence has been observed in the CL images of the carbonate skeletons of marine invertebrates, including brachiopods, and is attributed to periodic fluctuations in the concentration of dissolved Mn in ambient seawater at their growth sites (Barbin and Gaspard, 1995; Barbin, 2013). The MCLI values of the unaltered shells ranged from 43 to 89 (Fig. 4). The mean MCLI values for the unaltered shells were 67 for B-5, 74 for B-9, 55 for B-10, and 60 for B-16.

Fig. 9.

Cathodoluminescence images of (a, c, e, g) unaltered and (b, d, f, h) altered shells. Refer to Fig. S2 for additional shell portions.

The B-9 shell altered exclusively with artificial seawater (Figs. 9h and S4-4b) exhibited no significant changes compared to that before alteration (Figs. 9g and S4-4a). The MCLI values of the altered B-9 shell ranged from 44 to 96, with an average value of 70 (Fig. 4d). Conversely, shells altered with sediment-artificial seawater mixtures (B-5, B-10, and B-16) showed stronger red luminescence in some or all shells (Figs. 9b, d, f, S4-1b, 2b and 3b) than in unaltered shells (Figs. 9a, c, e, S4-1a, 2a and 3a). The MCLI values of the altered B-9 shells ranged from 74 to 159 (Fig. 4). The mean MCLI values were 114 for B-5, 110 for B-10, and 85 for B-16, which were 1.7, 2.0, and 1.4 times higher than those of the unaltered shells, respectively.

Discussion

Changes in carbon and oxygen isotope compositions

Some material leached from the silicon tube or gasket into the artificial seawater may have altered the δ¹³C and δ¹⁸O values of the brachiopod shells. However, the changes in the δ¹³C and δ¹⁸O values of the brachiopod shells due to the ADEs exhibited similar trends, devoid of anomalous values, regardless of whether the shells were altered in artificial or a mixture of artificial seawater and additional materials. Therefore, we conclude that the influence of the leached material is limited, if present, and proceed to the following discussion.

Carbon isotope composition

Although the amplitude of δ¹³C profiles in the altered shells was comparable to that of the unaltered shells, the δ¹³C box plots showed decreases of ~0.1–0.2‰ in all samples post-ADEs (Fig. 8). Two of the four decreases were statistically significant (Fig. 7). Riechelmann et al. (2016) conducted ADEs on extant brachiopod shells using artificial meteoric fluid, artificial burial fluid, and natural seawater at 100°C and 175°C. The δ¹³C values of the unaltered shells did not significantly differ from those altered at 100°C. In contrast, the δ¹³C values significantly decreased post-ADEs at 175°C (Table S3). The observation that δ¹³C values did not decrease significantly at 100°C in the experiments conducted by Riechelmann et al. (2016) could be attributed to their sampling method. Riechelmann et al. (2016) did not compare the δ¹³C values between the altered and unaltered shells from the same individual. They collected powder samples (bulk samples) for isotope analysis from the altered dorsal and unaltered ventral valves. Therefore, their reported changes in the δ¹³C values partly represent inter- and intra-shell variations in the shell δ¹³C values. In contrast, we collected powder samples from the same shell portions of the same valves/individuals before and after the ADEs using a micro-drill to eliminate vital effects and minimize intra-shell variations in δ¹³C values. Consequently, our method is more appropriate for capturing the changes in brachiopod shells caused by ADEs.

The results of this study and those reported by Riechelmann et al. (2016) at 175°C show decreasing trends in δ¹³C values across various samples and experimental conditions (Table S3). The observed decreases suggest that a common source of light carbon (¹²C) existed under all experimental conditions. The three potential sources of ¹²C were (1) sediments, (2) a diagenetic fluid (artificial seawater), and (3) organic matter contained in brachiopod shells. If carbon in the sediments was the source of ¹²C, the findings of Riechelmann et al. (2016), who did not use sediments, as well as those of the present study, in which the shells were altered in quartz powder (B-5) and seawater (B-9) without carbon, could not be adequately explained. In the ADE using the B-10 shell, which was packed with carbonate powder (aragonite) containing ¹²C, the δ¹³C value of the powder decreased after the experiment (Fig. 7). Consequently, it is unlikely that the incorporation of ¹²C from the powder is responsible for the decreased δ¹³C value in the B-10 shell. If the source of ¹²C was diagenetic fluid (artificial seawater), we could not account for the observation that the δ¹³C values of the B-9 shell altered with fluid decreased less than those of the B-10 and B-16 shells altered with sediment (Fig. 7). In addition, the diagenetic fluids used in the ADEs by Riechelmann et al. (2016) consisted of NaCl- and MgCl₂-water solutions, which contained minimal dissolved inorganic carbon (DIC), a potential source of ¹²C.

Potential sources of ¹²C may include organic matter within brachiopod shells. Organic matter is present not only on the upper surface of the shells (periostracum; Williams et al., 1997) and in the puncta (Pérez-Huerta et al., 2009; Williams, 1997) but also in the spaces between the calcite fibers constituting the secondary shell layer (Casella et al., 2018; Immel et al., 2015; Simonet Roda et al., 2019a, 2019b), which account for approximately 3% of the dry weight of the shells (Peck et al., 1987). The organic matter may have been decomposed through heating. Thermal degradation of organic matter can begin at temperatures below 100°C (Gaffey et al., 1991). Therefore, thermal degradation is expected to occur to some degree under the experimental conditions of 125°C and 175°C (Casella et al., 2017; Pederson et al., 2019; Ritter et al., 2017, this study).

Post-ADEs, the altered shells lost their original coloration and appeared whiter than the unaltered shells (Fig. 1). This suggests thermal degradation of the shell pigments. Three families of pigments were identified using micro-Raman spectroscopy in extant and extinct brachiopod species, with carotenoids and melanin being the most prevalent (Gaspard et al., 2019). Therefore, the carbon derived from the shell pigments may decrease the shell δ¹³C values in the ADEs.

When carbon from the organic matter was incorporated into shell calcite, the shell δ¹³C values decreased. However, another process was conceivable. If, during the ADEs, certain organic matter transformed into a compound that interacted with phosphoric acid, a reagent used to dissolve carbonates in isotope analysis, to release CO₂, the decreased δ¹³C values can be explained.

Oxygen isotope composition

The δ¹⁸O values of the altered shells were significantly lower than those of the unaltered shells under all experimental conditions (Fig. 7). Riechelmann et al. (2016) reported that the δ¹⁸O values of the shells altered with diagenetic fluids at 100°C and 175°C were 0.1–0.5‰ and 0.8–1.2‰ lower than the primary values, respectively (Table S3). They used diagenetic fluids with very low δ¹⁸O values (artificial marine fluid: –20.6‰VSMOW, –50.0‰VPDB; artificial meteoric and formation water: –47‰VSMOW, –75.6‰VPDB) to facilitate the detection of changes in shell δ¹⁸O values. However, the decreases in the shell δ¹⁸O values at 100°C in the experiments conducted by Riechelmann et al. (2016) are comparable to those observed at 125°C in this study. In addition, no large differences were observed in the decreases in the δ¹⁸O values between shells altered in the artificial seawater (B-9) and those altered in the sediment-artificial seawater mixtures (B-5, B-10, and B-16). These results suggest that when the fluids involved in diagenetic alteration are sufficiently provided, the temperature at the time of alteration is likely a more critical factor of shell δ¹⁸O changes than the composition of the surrounding diagenetic fluid and sediment. Conversely, the fluids were sufficient in the ADEs of Riechelmann et al. (2016) and this study.

The δ¹⁸O_SW values of the artificial seawater in the silicon tube and hydraulic water (Mili-Q water) were almost the same as the original values. They showed identical decreases after the ADEs despite the absence of contact between them. The only experimental condition common to this case was the temperature change during the ADEs, which supported the probability that temperature was a critical factor for the decrease, although the detailed mechanism was uncertain.

Changes in metal element concentrations

In our experiments, the Mn concentration in the altered shells exceeded that in the unaltered shells by 1.6–17 ppm. The increases were statistically significant under three experimental conditions: carbonate powder-artificial seawater mixture, quartz powder-artificial seawater mixture, and artificial seawater. The B-16 shell packed with sandstone powder exhibited the greatest increase in Mn concentration after ADEs (17 ppm). However, because the B-10 shell packed with carbonate powder also had an increased Mn concentration (13 ppm), the contribution of Mn in the sandstone to the increase was limited. The systematic increases in the Mn concentration contrast well with those in the Na, Mg, Fe, and Sr concentrations in the altered shells. These may or may not be higher than those in the unaltered shells, with no clear trend (Figs. 4 and 7). Similarly, in the experiments by Riechelmann et al. (2016), the Mn concentration in the altered shells were 5–10 ppm higher than that in the unaltered shells, with some exceptions. The Mn concentration in the artificial seawater packed with the B-9 shell was 0.9 ppm higher after ADEs and increased by 1.0 ppm relative to that in the artificial seawater after blank experiments (in which only artificial seawater was contained in silicon tubes and subsequently altered; Fig. 7). The results of this study and those of Riechelmann et al. (2016) indicate that a Mn source was common in experiments in which the shells were solely packed with diagenetic fluid and those with sediments. Given that Mn, Cu, Zn, and Cd were associated with the organic matter in molluscan shells (Takesue et al., 2008), the organic matter might serve as a potential Mn source in the examined brachiopod shells. The organic matter is also a potential source of ¹²C, lowering the shell δ¹³C due to the experiments. However, because few studies have examined the elements contained in the organic matter of brachiopod shells (e.g., Cusack et al., 2008; England et al., 2007), the form and amount of Mn contained in the within-shell organic matter is unknown.

Given that Mn in the within-shell organic matter was likely decomposed through thermal degradation and subsequently released into the diagenetic fluid, two scenarios may account for the increased Mn concentration in the altered shells: the substitution of Ca ions in the calcite with Mn ions via diagenetic fluids, or the precipitation of calcite microcrystals from the diagenetic fluids on the calcite shells. In addition, the adsorption of Mn (likely in the form of manganese oxide) onto the shell surface (surface of the calcite shell) is a potential mechanism for accommodating the presumed Mn derived from organic matter.

The Na and Mg concentrations tended to decrease in the shells altered with sediment-artificial water mixtures (B-5, B-10, and B-16 shells), whereas the Fe and Sr concentrations increased (Fig. 8d–g). However, the opposite trend could be observed in the shell altered only with artificial seawater (B-9 shell). This contrast occurred because the artificial seawater was relatively rich in Na and Mg and poor in Fe and Sr compared to the sediment. Therefore, the increase/decrease in metal element concentrations in the altered shells is contingent upon the water/rock ratio of the material surrounding the shells in the ADEs and possibly in the actual diagenesis.

In summary, the statistically significant changes observed in the isotopic and chemical compositions of the shells before and after the ADEs include decreases in δ¹⁸O values across all four experimental conditions, increases in Mn concentration in the carbonate powder-artificial seawater mixture, quartz powder-artificial seawater mixture, and artificial seawater, and decreases in δ¹³C values in the carbonate powder-artificial seawater mixture and sandstone powder-artificial seawater mixture. Based on these results, the following considerations can be made:

a) The carbonate powder-artificial seawater mixture is the sole matrix where all three compositions are overprinted.

b) The compositions most likely to change are as follows (in order): δ¹⁸O, Mn concentration, and δ¹³C.

c) Given the analytical accuracy of each composition, it appears more feasible to detect changes in these compositions before and after the ADEs in this order.

Cathodoluminescence image analysis

No apparent change in luminescence was observed in the shell altered exclusively with artificial seawater (B-9 shell) when compared to the unaltered shell (Fig. 9g, h). The average MCLI values for the altered shell were almost the same (approximately 70) as those of the unaltered shell, thereby indicating no quantitative change. Riechelmann et al. (2016) altered shells exclusively with fluids at 100°C and 175°C without pressure control. As a result, the CL images of shells altered at 100°C were non-luminescent, similar to the unaltered shells, whereas those altered at 175°C displayed a reddish luminescence. Casella et al. (2018) performed artificial diagenesis experiments under conditions similar to those of Riechelmann et al. (2016), demonstrating that the shells altered under experimental conditions of 100°C and 175°C were non-luminescence like the unaltered ones. Consequently, the results of this study and those of two previous studies indicate that the shells altered exclusively with fluids at temperatures of 100°C and 125°C do not exhibit increased luminescence compared to unaltered shells (even if they become more luminescent, the change is too subtle to be detected). The effect of diagenetic alteration at a high temperature of 175°C on shell CL luminescence remains a subject of debate.

The shells altered with the sediment-artificial seawater mixtures exhibited varying degrees of red luminescence (B-5, Fig. 9d; B-10, Fig. 9b; B-16, Fig. 9f). The average MCLI values for these altered shells were higher than those of the unaltered shells, and the luminescence intensity was quantitatively higher (averages: 85–114 vs. 55–67). A factor contributing to the higher luminescence intensity of carbonate CL images is an increase in the number of Mn-substituting Ca sites in the calcite lattice (Machel et al., 1991). This study indicated that the Mn concentration in the altered shells is higher than that in the unaltered shells (Figs. 4 and 7). However, some results are inconsistent with the general relationship between Mn concentration and CL luminescence intensity (Fig. 4). The Mn concentration in the shell altered exclusively with artificial seawater (B-9 shell; Fig. 8c) increased more than that in the shells altered with sediment-artificial seawater mixtures (B-5, B-10, and B-16 shells), although the former was non-luminescent. These indicate that the observed increase is likely attributed to Mn adsorption on the shell surface as well as Mn substitution for Ca sites within the calcite lattice.

Alteration vectors

In this study, we presented trends and quantitative measures of artificial diagenetic alteration (i.e., changes in δ¹³C and δ¹⁸O values and Mn concentration) in the examined shells. These alterations are represented as vectors on the cross-plots of δ¹⁸O values vs. δ¹³C, Mn concentration vs. δ¹³C values, and Mn concentration vs. δ¹⁸O values, referred to as alteration vectors. Figure 10 shows the averaged alteration vectors (calculated as the arithmetic means of these changes under each experimental condition) for this study and that of Riechelmann et al. (2016). However, substantial differences exist between the two studies (e.g., variations in brachiopod taxa examined, sampling methods for isotopic and chemical analyses, and fluids used for the ADEs). The vectors of individual samples are shown in Fig. S5.

Fig. 10.

Averaged alteration vectors of the examined shells. The findings of Riechelmann et al. (2016) are presented for comparison. Horizontal and vertical bars indicate standard deviations. (a) δ¹⁸O values vs. δ¹³C values; (b) Mn concentration vs. δ¹⁸O values; (c) Mn concentration vs. δ¹³C values. SW = artificial seawater.

δ¹⁸O values vs. δ¹³C values

The averaged alteration vectors on the δ¹³C and δ¹⁸O cross-plots (Fig. 10a) indicate that the vectors show decreasing trends, with one notable exception. The slopes of the B-10 (altered with the carbonate powder-artificial seawater mixture) and B-16 (altered with the sandstone powder-artificial seawater mixture) shells were steeper than those of the B-5 (altered with the quartz powder-artificial seawater mixture) and B-9 (altered in artificial seawater) shells. The δ¹³C decreases of the B-10 and B-16 shells (~ –0.2‰) were approximately twice as large as those of the B-5 and B-9 shells (~ –0.1‰). The vectors of the B-5 and B-9 shells were similar in direction to those of the shells altered in artificial meteoric fluids by Riechelmann et al. (2016). The vector directions of shells B-10 and B-16 were similar to those altered in their artificial burial fluids. The Na and Mg concentrations in the burial fluids were ~10 times higher than those in the meteoric fluids. Similarly, the B-10 and B-16 shells were altered with carbonate and sandstone powders containing minerals (e.g., calcite, pyroxene, feldspar, mica, and clay), from which metal ions leached relatively easily during the ADEs. Based on these results, the ionic strength of the diagenetic fluids for the B-10 and B-16 shells increased during the experiments and resulted in diagenetic environments (e.g., isotopic and chemical compositions of the fluids) that became similar to those of the burial fluids of Riechelmann et al. (2016). Therefore, although the diagenetic fluids for the B-10 and B-16 shells were not recovered and the metal element concentrations were not measured, we considered that the δ¹³C and δ¹⁸O variations by burial diagenesis likely depended on the ionic strength of the diagenetic fluid. Casella et al. (2018) reported that the recrystallization and fusion of calcite fibers in brachiopod shells altered in artificial burial fluids at 175°C were more significant than those altered in artificial meteoric fluids. They concluded that the Mg concentration in these diagenetic fluids affected the diagenetic alteration process in their experiments.

The directions and lengths of the individual alteration vectors (thin arrows in Fig. S5-1a–d) are radially dispersed from the origin toward the second and third quadrants. If the variations in the δ¹³C and δ¹⁸O values of the examined shells resulted from microcrystalline calcite (cement) precipitation from diagenetic fluids, then these values were determined by the isotopic and abundance ratios of the shell calcites and cement; therefore, the alteration vectors should be aligned in their directions. However, in this study, the changes (decreases) in the δ¹³C values varied considerably among the samples, even though the δ¹⁸O values were reduced similarly. This resulted in the vector directions not aligning under all experimental conditions. This suggests that the microcrystalline calcite (cement) precipitation is not the leading cause of the changes (decreases) in the shell δ¹³C and δ¹⁸O values.

Mn concentration vs. δ¹³C values

The changes in the Mn concentration and δ¹³C values are likely attributable to the decomposition of organic matter containing Mn within the shells. If this is the case, the co-increase of C and Mn leaching into diagenetic fluids results in similar directions of averaged alteration vectors on the Mn concentration and δ¹³C cross-plots among the shells. However, the slope of the B-16 shell was steeper than that of the B-5, B-9, and B-10 shells (Figs. 10 and S5-2a–d). This is attributed to the relatively smaller increase in the Mn concentration in the B-16 shell than that in the other shells, potentially due to Mn leaching into diagenetic fluids being retained by sandstone grains/minerals. When the B-5 and B-10 alteration vectors were compared, the directions were almost identical. However, the lengths were significantly different. The B-5 vector was shorter than the B-10 vector. This observed difference is likely attributable to the buffering effect of the carbonate powder on the decrease in pH during organic matter decomposition. When organic matter decomposes, CO₂ is released, which subsequently lowers the pH of diagenetic fluids (Kuma et al., 2019; Yoshida et al., 2015). However, organic matter tends to be less decomposed and more preserved in low-pH environments. For example, Paraguassu (1976) conducted experiments on the silicification of extant bivalve shells and reported that organic matter in the shells was preserved in low-pH environments. The organic tissue of wood is preserved when petrified (silicified) in highly acidic (pH = 3) silica-rich hot spring water (Akahane and Furuno, 1993; Akahane et al., 1999). In the case of the B-5 shell altered with the quartz powder-artificial seawater mixture, the pH continuously decreased, possibly resulting in the cessation of organic matter decomposition at some point during the ADE. In the B-10 shell experiments, the carbonate powder suppressed the pH decrease, while the organic matter in the shells exhibited greater decomposition than that in the B-5 shell experiments, thereby resulting in greater changes in the δ¹³C values and Mn concentration.

Mn concentration vs. δ¹⁸O values

The averaged and individual alteration vectors on the Mn concentration and δ¹⁸O cross-plots (Figs. 10c and S5-3) are oriented in the directions from the origin toward the fourth quadrant (representing decreases in δ¹⁸O values and increases in Mn concentration), except for some vectors of the B-16 shell. The directions of the individual vectors are well aligned regardless of the experimental conditions, in contrast to the averaged and individual alteration vectors observed in the Mn concentration and δ¹³C cross-plots (Figs. 10b and S5-2). Therefore, the Mn concentration in brachiopod shells can be used to determine whether the primary δ¹⁸O values are retained in the brachiopod shells or modified by burial diagenesis.

The relationships between the preservation state of shell microstructure and changes in isotopic and chemical compositions

In almost all parts of the artificially altered shells, the shell microstructure was well preserved, similar to that of the unaltered shells. However, a few fibers were amalgamated in certain locations within the observation area, and extensive fiber fusions rarely occurred (Fig. 5). By exposing brachiopod shells to high temperatures in an autoclave with artificial burial or meteoric fluids for long periods (1 week to 1 month), Casella et al. (2018) observed alterations in the microstructure of brachiopod shells and the crystallographic orientation of the calcite fibers. They observed the distortion and fusion of calcite fibers in the altered shells. However, unaltered portions of well-preserved fibers surrounded the altered portions, indicating that microstructural degradation and destruction exhibited a mosaic pattern rather than uniformity throughout the shell. The partial and mosaic-like occurrence of altered portions (where the shell microstructure was broken) was common in the altered shells in this study. Casella et al. (2018) reported that calcite fibers were preserved in the ADEs when using artificial meteoric fluids. In contrast, calcite fibers were found to be distorted or amalgamated in experiments using artificial burial fluids regardless of the alteration temperature. They concluded that abundant Mg in the artificial burial fluids contributed to microstructural degradation and destruction. Their finding is consistent with the occurrence of amalgamated calcite fibers in the shells altered with artificial seawater abundant in Mg (approximately 800–900 ppm, Table 2) in this study.

Almost no correlation existed between the degradation and destruction of the shell microstructure and changes in the isotopic and chemical compositions. For example, calcite fibers were amalgamated extensively the t5 section of the B-10 shell compared to the other shells (Fig. 5b). A significant correlation between these variables would lead to a decrease in δ¹³C and δ¹⁸O values and increase in Mn concentrations in areas where extensive fiber amalgamation was observed. However, the alterations in chemical compositions at t5 were not necessarily larger than those observed in other shell portions (Fig. 4); the δ¹³C values exhibited a slight decrease along the growth axis, while the δ¹⁸O values and Mn concentration showed roughly uniformly changes.

Conversely, this study indicates that the isotopic and chemical compositions of brachiopod shells may have been altered when they underwent alteration by burial diagenesis, even if the shell microstructure was well preserved. However, the maximum temperature and period for the diagenetic alteration (at 125°C for one month) were lower and shorter than those of Riechelmann et al. (2016)’s experiments (at 175°C for six months). Therefore, diagenetic alterations may not have progressed sufficiently in the present study. The relationship between the degradation and destruction of shell microstructure and the changes in isotopic and chemical compositions resulting from burial diagenesis shown in this study should not be regarded as definitive.

Previous geochemical studies on fossil brachiopod shells have used SEM imaging to check the preservation state of their microstructures, facilitating the screening of samples and analytical data for reliable palaeoenvironmental reconstructions (e.g., Rush and Chafetz, 1990; Veizer et al., 1997; Cummins et al., 2014; Jelby et al., 2014; Ullmann et al., 2014; Garbelli et al., 2016; Bergmann et al., 2018). This study indicates that the conventional approach to differentiating unaltered from altered shells requires revision because of the nonlinear relationship between the preservation state of their microstructures and their retained isotopic and chemical compositions.

Using field emission-SEM and atomic force microscopy (AFM), Casella et al. (2018) found that nanometer-sized particulate calcite microcrystals (nanocomposite mesocrystal biocarbonates; NMB) that constitute calcite fibers were replaced with inorganic rhombohedral calcite (IRC) via diagenetic alteration. In the present study, no significant changes were found in the shell microstructure before and after the ADEs, as observed through SEM. This lack of change may be attributed to potential experimental problem(s) (e.g., short experimental period, low temperature, or low pressure). However, our TEM observations clearly show moderate dissolution of calcite fibers at the grain boundaries, suggesting that nanoscale analysis is necessary to identify diagenetic alterations and evaluate their extent.

Prospects of the artificial diagenesis experiments

The extent of changes in shell δ¹⁸O values during the ADEs is primarily influenced by the alteration temperature when sufficient diagenetic fluids are provided. Consequently, additional experiments at various water/rock ratios are required to adequately compare the results. The extent of the changes in shell δ¹³C values is much less than that in the δ¹⁸O values and likely influenced by the decomposition of organic matter and associated changes in the pH of diagenetic fluids. Therefore, the extent likely varies based on the ambient sediment type. The extent of the ADE using the carbonate powder-artificial water mixture was considerable. These results suggest that shell δ¹³C values are more modified when shells are packed with a carbonate powder containing high organic matter content. Further experiments are necessary to accurately understand this process of diagenetic alteration. If verified, it should be acknowledged that brachiopod shells in organic-rich deposits may have undergone alterations due to burial diagenesis, leading to δ¹³C values that do not accurately reflect the seawater δ¹³C values of dissolved inorganic carbon during their lifespan and may have been modified (reduced). This knowledge is valuable for identifying specific brachiopod shells suitable for palaeoenvironmental reconstruction based on their δ¹³C values.

Experiments should be conducted at elevated temperatures and pressures (175°C and 90 MPa), representing the maximum conditions that fossil samples may have experienced. The data from these experiments are helpful for comparison with the results of previous studies. However, the silicon rubber tubes and gaskets used in this study were not resistant to these high temperatures and pressures in the long term (e.g., approximately one month). Therefore, more durable materials need to be identified that can withstand elevated temperatures and pressures in future experiments. Measuring the pH and isotopic composition of diagenetic fluids is essential for enhancing our understanding of alterations caused by burial diagenesis. Additionally, observations of shell structure and diagenetic products at micro- to nanoscales using various techniques, including SEM, TEM, and AFM, are necessary.

Conclusions

We quantitatively assessed the extent and mechanisms by which the isotopic (δ¹³C and δ¹⁸O) and chemical (Na, Mg, Mn, Fe, and Sr) compositions of brachiopod shells are altered via burial diagenesis. This was conducted through ADEs at 125°C and 75 MPa using shells of extant T. crossei from Otsuchi Bay. We altered the shells exclusively with artificial seawater or three types of artificial seawater and sediment mixtures (quartz, carbonate, and sandstone powders) to assess the effects of sediment on alteration.

Statistically significant changes were observed in the isotopic and chemical compositions of the shells before and after the ADEs: decreases in the δ¹⁸O values across all four experimental conditions, increases in Mn concentration in the carbonate powder-artificial seawater mixture, quartz powder-artificial seawater mixture, and artificial seawater, and decreases in the δ¹³C values in the carbonate powder-artificial seawater mixture and sandstone powder-artificial seawater mixture. The other metal concentrations did not show significant changes under any of the experimental conditions. Our results show that alteration temperature significantly impacts the δ¹⁸O values of brachiopod shells during burial diagenesis than the isotopic and chemical compositions of the ambient sediments and fluids. The Mn concentration increased when the shells were altered in materials that were relatively poor in Mn (i.e., carbonate, quartz, and artificial seawater). The directions of the individual alteration vectors are well aligned regardless of the experimental conditions on the Mn concentration and δ¹⁸O cross-plots. This suggests that the main source of Mn may be the organic matter in the shells. The δ¹³C values of the altered shells exhibited decreasing trends across most sampling locations under all experimental conditions, indicating a common source of ¹²C. Probable candidates for this decrease include the thermal degradation of organic matter in the shells. The increase in Mn concentration is common in all altered shells, although its extent varies greatly. This increase correlates with decreases in δ¹³C and δ¹⁸O values, suggesting that the Mn concentration may serve as an indicator of whether the primary isotopic and chemical compositions have been retained or modified by burial diagenesis. This finding is consistent with that of cathodoluminescence imaging, which shows that shells altered with sediment-artificial seawater mixtures exhibit luminescence. SEM observations showed limited degradation and destruction of the shell microstructure due to ADEs and no linear relationship with diagenetic alteration. This discrepancy can be explained by the dissolution observed exclusively at the nanometer scale using TEM; however, further investigation is required.

This study is significant as the ADEs of brachiopod shells were conducted at 125°C and 75 MPa, conditions presumed to reflect natural environments where actual burial diagenesis occurs. Consequently, we were able to reproduce certain known diagenetic alterations in the natural shells (e.g., decreases in δ¹³C/δ¹⁸O values and increases in Mn concentration in shells), while failing to reproduce others (e.g., extensive amalgamation of the shell fibers). However, the changes in δ¹³C_DIC and δ¹⁸O_SW observed in certain ADEs (e.g., blank test) remain unexplained, suggesting that our experiments did not achieve complete success. The potential cause of this failure is the leaching of substances from the silicon tubes or gaskets. Therefore, it is recommended that a more durable material suitable for higher temperatures and pressures (e.g., gold) be identified and utilized in future experiments.

Acknowledgments

J. Muto from the Department of Earth Sciences, Graduate School of Science, Tohoku University, provided suggestions regarding the artificial diagenesis experiments. Y. Ito, H. Kawanobe, and M. Abe from the same department supported the SEM observations and prepared thin sections for CL image observation. M. Nagasawa and T. Kuribayashi from Takeda Rika Kogyo Co., Ltd. assisted in preparing the materials used in the experiments. Financial support was provided by the Fukada Geological Institute and Fujiwara Natural History Foundation to HF and by the Japan Society for the Promotion of Science (Grants in Aid for Scientific Research, 21K18642 and 23K25959) to YI. This study was supported by the Cooperative Program of the Atmosphere and Ocean Research Institute at the University of Tokyo. The authors gratefully acknowledge the support from the World Premier International Research Center Initiative (WPI), MEXT, Japan. The manuscript has been significantly improved based on the comments and suggestions of Y. Yokoyama (editor) and two anonymous reviewers. We would like to thank Editage (www.editage.jp) for English language editing.

References

•  Akahane,  H. and  Furuno,  T. (1993) Recent silicified woods in the Tateyama Hot Spring (Shin-yu), Toyama Prefecture. J. Geol. Soc. Japan 99, 457–466. https://doi.org/10.5575/geosoc.99.457
•  Akahane,  H.,  Furuno,  T.,  Miyajima,  H.,  Goto,  M.,  Ota,  T. and  Yamamoto,  S. (1999) Experimental study on wood silicification in the stream of hot spring water. Estimation of rapidity and mechanism of silicification. J. Geol. Soc. Japan 105, 108–115. (in Japanese). https://doi.org/10.5575/geosoc.105.108
•  Auclair,  A.,  Joachimski,  M. M. and  Lécuyer,  C. (2003) Deciphering kinetic, metabolic and environmental controls on stable isotope fractionations between seawater and the shell of Terebratalia transversa (Brachiopoda). Chem. Geol. 202, 59–78. https://doi.org/10.1016/S0009-2541(03)00233-X
•  Barbin,  V. (2013) Application of cathodoluminescence microscopy to recent and past biological materials: A decade of progress. Mineral. Petrol. 107, 353–362. https://doi.org/10.1007/s00710-013-0266-6
•  Barbin,  V. and  Gaspard,  D. (1995) Cathodoluminescence of recent articulate brachiopod shells. Implication for growth stages and diagenesis evaluation. Geobios 28, 39–45. https://doi.org/10.1016/S0016-6995(95)80151-0
•  Bates,  N. R. and  Brand,  U. (1991) Environmental and physiological influences on isotopic and elemental compositions of brachiopod shell calcite: Implications for the isotopic evolution of Paleozoic oceans. Chem. Geol. 94, 67–78. https://doi.org/10.1016/0168-9622(91)90041-T
•  Bergmann,  K. D.,  Finnegan,  S.,  Creel,  R.,  Eiler,  J. M.,  Hughes,  N. C.,  Popov,  L. E. and  Fischer,  W. W. (2018) A paired apatite and calcite clumped isotope thermometry approach to estimating Cambro-Ordovician seawater temperatures and isotopic composition. Geochim. Cosmochim. Acta 224, 18–41. https://doi.org/10.1016/j.gca.2017.11.015
•  Brand,  U. and  Veizer,  J. (1980) Chemical diagenesis of a multicomponent carbonate system-1: Trace elements. J. Sediment. Res. 50, 1219–1236. https://doi.org/10.1306/212F7BB7-2B24-11D7-8648000102C1865D
•  Brand,  U.,  Logan,  A.,  Hiller,  N. and  Richardson,  J. (2003) Geochemistry of modern brachiopods: Applications and implications for oceanography and paleoceanography. Chem. Geol. 198, 305–334. https://doi.org/10.1016/S0009-2541(03)00032-9
•  Brand,  U.,  Jiang,  G.,  Azmy,  K.,  Bishop,  J. and  Montañez,  I. P. (2012) Diagenetic evaluation of a Pennsylvanian carbonate succession (Bird Spring Formation, Arrow Canyon, Nevada, USA)—1: Brachiopod and whole rock comparison. Chem. Geol. 308, 26–39. https://doi.org/10.1016/j.chemgeo.2012.03.017
•  Brand,  U.,  Azmy,  K.,  Bitner,  M. A.,  Logan,  A.,  Zuschin,  M.,  Came,  R. and  Ruggiero,  E. (2013) Oxygen isotopes and MgCO₃ in brachiopod calcite and a new paleotemperature equation. Chem. Geol. 359, 23–31. https://doi.org/10.1016/j.chemgeo.2013.09.014
•  Brand,  W. A.,  Coplen,  T. B.,  Vogl,  J.,  Rosner,  M. and  Prohaska,  T. (2014) Assessment of international reference materials for isotope-ratio analysis (IUPAC Technical Report). Pure Appl. Chem. 86, 425–467. https://doi.org/10.1515/pac-2013-1023
•  Carpenter,  S. J. and  Lohmann,  K. C. (1995) δ¹⁸O and δ¹³C values of modern brachiopod shells. Geochim. Cosmochim. Acta 59, 3749–3764. https://doi.org/10.1016/0016-7037(95)00291-7
•  Casella,  L. A.,  Griesshaber,  E.,  Yin,  X.,  Ziegler,  A.,  Mavromatis,  V.,  Müller,  D.,  Ritter,  A. C.,  Hippler,  D.,  Harper,  E. M.,  Dietzel,  M.,  Immenhauser,  A.,  Schöne,  B. R.,  Angiolini,  L. and  Schmahl,  W. W. (2017) Experimental diagenesis: Insights into aragonite to calcite transformation of Arctica islandica shells by hydrothermal treatment. Biogeosciences 14, 1461–1492. https://doi.org/10.5194/bg-14-1461-2017
•  Casella,  L. A.,  Griesshaber,  E.,  Simonet Roda,  M.,  Ziegler,  A.,  Mavromatis,  V.,  Henkel,  D.,  Laudien,  J.,  Häussermann,  V.,  Neuser,  R. D.,  Angiolini,  L.,  Dietzel,  M.,  Eisenhauer,  A.,  Immenhauser,  A.,  Brand,  U. and  Schmahl,  W. W. (2018) Micro- and nanostructures reflect the degree of diagenetic alteration in modern and fossil brachiopod shell calcite: A multi-analytical screening approach (CL, FE-SEM, AFM, EBSD). Palaeogeogr. Palaeoclimatol. Palaeoecol. 502, 13–30. https://doi.org/10.1016/j.palaeo.2018.03.011
•  Cummins,  R. C.,  Finnegan,  S.,  Fike,  D. A.,  Eiler,  J. M. and  Fischer,  W. W. (2014) Carbonate clumped isotope constraints on Silurian ocean temperature and seawater δ¹⁸O. Geochim. Cosmochim. Acta 140, 241–258. https://doi.org/10.1016/j.gca.2014.05.024
•  Cusack,  M.,  Dauphin,  Y.,  Cuif,  J. P.,  Salomé,  M.,  Freer,  A. and  Yin,  H. (2008) Micro-XANES mapping of sulphur and its association with magnesium and phosphorus in the shell of the brachiopod, Terebratulina retusa. Chem. Geol. 253, 172–179. https://doi.org/10.1016/j.chemgeo.2008.05.007
•  Eming,  C. C. (1992) Functional disposition of the lophophore in living Brachiopoda. Lethaia 25, 291–302.
•  England,  J.,  Cusack,  M. and  Lee,  M. R. (2007) Magnesium and sulphur in the calcite shells of two brachiopods, Terebratulina retusa and Novocrania anomala. Lethaia 40, 2–10. https://doi.org/10.1111/j.1502-3931.2006.00001.x
•  Fujioka,  H.,  Takayanagi,  H.,  Yamamoto,  K. and  Iryu,  Y. (2019) The effects of meteoric diagenesis on the geochemical composition and microstructure of Pliocene fossil Terebratalia coreanica and Laqueus rubellus brachiopod shells from northeastern Japan. Prog. Earth Planet. Sci. 6, 45. https://doi.org/10.1186/s40645-019-0289-7
•  Gaffey,  S. J.,  Kolak,  J. J. and  Bronnimann,  C. E. (1991) Effects of drying, heating, annealing, and roasting on carbonate skeletal material, with geochemical and diagenetic implications. Geochim. Cosmochim. Acta 55, 1627–1640.
•  Garbelli,  C.,  Angiolini,  L.,  Brand,  U.,  Shen,  S. Z.,  Jadoul,  F.,  Posenato,  R.,  Azmy,  K. and  Cao,  C. Q. (2016) Neotethys seawater chemistry and temperature at the dawn of the end Permian mass extinction. Gondwana Res. 35, 272–285. https://doi.org/10.1016/j.gr.2015.05.012
•  Gaspard,  D.,  Paris,  C.,  Loubry,  P. and  Luquet,  G. (2019) Raman investigation of the pigment families in recent and fossil brachiopod shells. Spectrochim. Acta A Mol. Biomol. Spectrosc. 208, 73–84. https://doi.org/10.1016/j.saa.2018.09.050
•  Grossman,  E. L.,  Mii,  H. S. and  Yancey,  T. E. (1993) Stable isotopes in Late Pennsylvanian brachiopods from the United States: Implications for Carboniferous paleoceanography. Geol. Soc. Am. Bull. 105, 1284–1296. https://doi.org/10.1130/0016-7606(1993)105%3C1284:SIILPB%3E2.3.CO;2
•  Habermann,  D.,  Neuser,  R. D. and  Richter,  D. K. (1998) Low limit of Mn²⁺-activated cathodoluminescence of calcite: State of the art. Sediment. Geol. 116, 13–24. https://doi.org/10.1016/S0037-0738(97)00118-8
•  Immel,  F.,  Gaspard,  D.,  Marie,  A.,  Guichard,  N.,  Cusack,  M. and  Marin,  F. (2015) Shell proteome of rhynchonelliform brachiopods. J. Struct. Biol. 190, 360–366. https://doi.org/10.1016/j.jsb.2015.04.001
•  Jelby,  M. E.,  Thibault,  N.,  Surlyk,  F.,  Ullmann,  C. V.,  Harlou,  R. and  Korte,  C. (2014) The lower Maastrichtian Hvidskud succession, Møns Klint, Denmark: Calcareous nannofossil biostratigraphy, carbon isotope stratigraphy, and bulk and brachiopod oxygen isotopes. Bull. Geol. Soc. Denmark 62, 89–104.
•  Kuma,  R.,  Hasegawa,  H.,  Yamamoto,  K.,  Yoshida,  H.,  Whiteside,  J. H.,  Katsuta,  N. and  Ikeda,  M. (2019) Biogenically induced bedded chert formation in the alkaline palaeo-lake of the Green River Formation. Sci. Rep. 9, 16448. https://doi.org/10.1038/s41598-019-52862-7
•  Loucks,  R. G.,  Dodge,  M. M. and  Galloway,  W. E. (1984) Regional controls on diagenesis and reservoir quality in lower Tertiary sandstones along the Texas Gulf Coast. Clastic Diagenesis (McDonald, D. A. and Surdam, R. C., eds.), 15–46, American Association of Petroleum Geologists.
•  Machel,  H. G.,  Mason,  R. G.,  Mariano,  A. N. and  Mucci,  A. (1991) Cause and emission of luminescence in calcite and dolomite. Luminescent Microscopy and Spectroscopy: Qualitative and Quantitative Applications (Barker, C. E. and Kopp, O. C., eds.), 9–25, Society for Sedimentary Geology (SEPM).
•  Mii,  H. S.,  Shi,  G. R.,  Cheng,  C. J. and  Chen,  Y. Y. (2012) Permian Gondwanaland paleoenvironment inferred from carbon and oxygen isotope records of brachiopod fossils from Sydney Basin, southeast Australia. Chem. Geol. 291, 87–103. https://doi.org/10.1016/j.chemgeo.2011.10.002
•  Okai,  T.,  Suzuki,  A.,  Terashima,  S.,  Inoue,  M.,  Nohara,  M.,  Kawahata,  H. and  Imai,  N. (2004) Collaborative analysis of GSJ/AIST geochemical reference materials JCp-1 (coral) and JCt-1 (giant clam). Chikyukagaku (Geochemistry) 38, 281–286. https://doi.org/10.14934/chikyukagaku.38.281
•  Paraguassu,  A. B. (1976) Experimental replacement of silica. Revista Brasiliera de Geociências 6, 89–94.
•  Peck,  L. S.,  Clarke,  A. and  Holmes,  L. J. (1987) Size, shape and the distribution of organic matter in the Recent Antarctic brachiopod Liothyrella uva. Lethaia 20, 33–40. https://doi.org/10.1111/j.1502-3931.1987.tb00757.x
•  Pederson,  C.,  Mavromatis,  V.,  Dietzel,  M.,  Rollion-Bard,  C.,  Nehrke,  G.,  Jöns,  N.,  Jochum,  K. P. and  Immenhauser,  A. (2019) Diagenesis of mollusc aragonite and the role of fluid reservoirs. Earth Planet. Sci. Lett. 514, 130–142. https://doi.org/10.1016/j.epsl.2019.02.038
•  Pérez-Huerta,  A.,  Cusack,  M.,  McDonald,  S.,  Marone,  F.,  Stampanoni,  M. and  MacKay,  S. (2009) Brachiopod punctae: A complexity in shell biomineralisation. J. Struct. Biol. 167, 62–67. https://doi.org/10.1016/j.jsb.2009.03.013
•  Popp,  B. N.,  Anderson,  T. F. and  Sandberg,  P. A. (1986a) Brachiopods as indicators of original isotopic compositions in some Paleozoic limestones. Geol. Soc. Am. Bull. 97, 1262–1269.
•  Popp,  B. N.,  Anderson,  T. F. and  Sandberg,  P. A. (1986b) Textural, elemental, and isotopic variations among constituents in Middle Devonian limestones, North America. J. Sediment. Petrol. 56, 715–727.
•  Riechelmann,  S.,  Mavromatis,  V.,  Buhl,  D.,  Dietzel,  M.,  Eisenhauer,  A. and  Immenhauser,  A. (2016) Impact of diagenetic alteration on brachiopod shell magnesium isotope (δ²⁶Mg) signatures: Experimental versus field data. Chem. Geol. 440, 191–206. https://doi.org/10.1016/j.chemgeo.2016.07.020
•  Ritter,  A. C.,  Mavromatis,  V.,  Dietzel,  M.,  Kwiecien,  O.,  Wiethoff,  F.,  Griesshaber,  E.,  Casella,  L. A.,  Schmahl,  W. W.,  Koelen,  J.,  Neuser,  R. D.,  Leis,  A.,  Buhl,  D.,  Niedermayr,  A.,  Breitenbach,  S. F. M.,  Bernasconi,  S. M. and  Immenhauser,  A. (2017) Exploring the impact of diagenesis on (isotope) geochemical and microstructural alteration features in biogenic aragonite. Sedimentology 64, 1354–1380. https://doi.org/10.1111/sed.12356
•  Rudwick,  M. J. S. (1970) Living and Fossil Brachiopods. Hutchinson Univ. Libr., 199 pp.
•  Rush,  P. F. and  Chafetz,  H. S. (1990) Fabric-retentive, non-luminescent brachiopods as indicators of initial δ¹³C and δ¹⁸O composition: A test. J. Sediment. Petrol. 60, 968–981.
•  Sippel,  R. F. and  Glover,  E. D. (1965) Structures in carbonate rocks made visible by luminescence petrography. Science 150, 1283–1287. https://doi.org/10.1126/science.150.3701.1283
•  Simonet Roda,  M.,  Griesshaber,  E.,  Ziegler,  A.,  Rupp,  U.,  Yin,  X.,  Henkel,  D.,  Häussermann,  V.,  Laudien,  J.,  Brand,  U.,  Eisenhauer,  A.,  Checa,  A. G. and  Schmahl,  W. W. (2019a) Calcite fibre formation in modern brachiopod shells. Sci. Rep. 9, 598. https://doi.org/10.1038/s41598-018-36959-z
•  Simonet Roda,  M.,  Ziegler,  A.,  Griesshaber,  E.,  Yin,  X.,  Rupp,  U.,  Greiner,  M.,  Henkel,  D.,  Häussermann,  V.,  Eisenhauer,  A.,  Laudien,  J. and  Schmahl,  W. W. (2019b) Terebratulide brachiopod shell biomineralization by mantle epithelial cells. J. Struct. Biol. 207, 136–157. https://doi.org/10.1016/j.jsb.2019.05.002
•  Swart,  P. K. (2015) The geochemistry of carbonate diagenesis: The past, present and future. Sedimentology 62, 1233–1304. https://doi.org/10.1111/sed.12205
•  Takayanagi,  H.,  Asami,  R.,  Otate,  T.,  Abe,  O.,  Miyajima,  T.,  Kitagawa,  H. and  Iryu,  Y. (2015) Quantitative analysis of intraspecific variations in the carbon and oxygen isotope compositions of the modern cool-temperate brachiopod Terebratulina crossei. Geochim. Cosmochim. Acta 170, 301–320. https://doi.org/10.1016/j.gca.2015.08.006
•  Takesue,  R. K.,  Bacon,  C. R. and  Thompson,  J. K. (2008) Influences of organic matter and calcification rate on trace elements in aragonitic estuarine bivalve shells. Geochim Cosmochim Acta 72, 5431–5445. https://doi.org/10.1016/j.gca.2008.09.003
•  Takizawa,  M.,  Takayanagi,  H.,  Yamamoto,  K.,  Abe,  O.,  Sasaki,  K. and  Iryu,  Y. (2017) Paleoceanographic conditions at approximately 20 and 70 ka recorded in Kikaithyris hanzawai (Brachiopoda) shells. Geochim. Cosmochim. Acta 215, 189–213. https://doi.org/10.1016/j.gca.2017.08.002
•  Tarutani,  T.,  Clayton,  R. N. and  Mayeda,  T. K. (1969) The effect of polymorphism and magnesium substitution on oxygen isotope fractionation between calcium carbonate and water. Geochim. Cosmochim. Acta 33, 987–996. https://doi.org/10.1016/0016-7037(69)90108-2
•  Ullmann,  C. V. and  Korte,  C. (2015) Diagenetic alteration in low-Mg calcite from macrofossils: A review. Geol. Q. 59, 3–20. https://doi.org/10.7306/gq.1217
•  Ullmann,  C. V.,  Campbell,  H. J.,  Frei,  R. and  Korte,  C. (2014) Geochemical signatures in Late Triassic brachiopods from New Caledonia. New Zealand J. Geol. Geophys. 57, 420–431. https://doi.org/10.1080/00288306.2014.958175
•  Veizer,  J.,  Fritz,  P. and  Jones,  B. (1986) Geochemistry of brachiopods: Oxygen and carbon isotopic records of Paleozoic oceans. Geochim. Cosmochim. Acta 50, 1679–1696. https://doi.org/10.1016/0016-7037(86)90130-4
•  Veizer,  J.,  Bruckschen,  P.,  Pawellek,  F.,  Diener,  A.,  Podlaha,  O. G.,  Carden,  G. A. F.,  Jasper,  T.,  Korte,  C.,  Strauss,  H.,  Azmy,  K. and  Ala,  D. (1997) Oxygen isotope evolution of Phanerozoic seawater. Palaeogeogr., Palaeoclimatol., Palaeoecol. 132, 159–172. https://doi.org/10.1016/S0031-0182(97)00052-7
•  Veizer,  J.,  Ala,  D.,  Azmy,  K.,  Bruckschen,  P.,  Buhl,  D.,  Bruhn,  F.,  Carden,  G. A. F.,  Diener,  A.,  Ebneth,  S.,  Godderis,  Y.,  Jasper,  T.,  Korte,  C.,  Pawellek,  F.,  Podlaha,  O. G. and  Strauss,  H. (1999) ⁸⁷Sr/⁸⁶Sr, δ¹³C and δ¹⁸O evolution of Phanerozoic seawater. Chem. Geol. 161, 59–88. https://doi.org/10.1016/S0009-2541(99)00081-9
•  Williams,  A. (1997) Shell structure. Treatise on Invertebrate Paleontology Part H Brachiopoda [Revised 1: Introduction] (Kaesler, R. L., ed.), 267–320, Geol. Soc. Am. and Univ. Kansas.
•  Williams,  A.,  Carlson,  S. J.,  Brunton,  C. H. C.,  Holmer,  L. E. and  Popov,  L. (1996) A supra-ordinal classification of the Brachiopoda. Phil. Trans. R. Soc. London, Ser. B 351, 1171–1193. https://doi.org/10.1098/rstb.1996.0101
•  Williams,  A.,  Brunton,  C. H. C. and  MacKinnon,  D. I. (1997) Morphology. Treatise on Invertebrate Paleontology Part H Brachiopoda [Revised 1: Introduction] (Kaesler, R. L., ed.), 321–422, Geol. Soc. Am. Univ. Kansas Press.
•  Yamamoto,  K.,  Asami,  R. and  Iryu,  Y. (2010a) Carbon and oxygen isotopic compositions of modern brachiopod shells from a warm-temperate shelf environment, Sagami Bay, central Japan. Palaeogeogr. Palaeoclimatol. Palaeoecol. 291, 348–359. https://doi.org/10.1016/j.palaeo.2010.03.006
•  Yamamoto,  K.,  Asami,  R. and  Iryu,  Y. (2010b) Within-shell variations in carbon and oxygen isotope compositions of two modern brachiopods from a subtropical shelf environment off Amami-o-shima, southwestern Japan. Geochem. Geophys. Geosyst. 11, Q10009. https://doi.org/10.1029/2010GC003190
•  Yamamoto,  K.,  Asami,  R. and  Iryu,  Y. (2011) Brachiopod taxa and shell portions reliably recording past ocean environments: Toward establishing a robust paleoceanographic proxy. Geophys. Res. Lett. 38, L13601. https://doi.org/10.1029/2011GL047134
•  Yamamoto,  K.,  Asami,  R. and  Iryu,  Y. (2013) Correlative relationships between carbon- and oxygen- isotope records in two cool-temperate brachiopod species off Otsuchi Bay, northeastern Japan. Paleontol. Res. 17, 12–26. https://doi.org/10.2517/1342-8144-17.1.12
•  Yamamoto,  K.,  Takizawa,  M.,  Takayanagi,  H.,  Asami,  R. and  Iryu,  Y. (2017) Nonlinear relationship between preservation of microstructure and geochemical composition of Pleistocene brachiopod shells from the Ryukyu Islands, southwestern Japan. Isl. Arc 26, e12217. https://doi.org/10.1111/iar.12217
•  Yoshida,  H.,  Ujihara,  A.,  Minami,  M.,  Asahara,  Y.,  Katsuta,  N.,  Yamamoto,  K.,  Sirono,  S.,  Maruyama,  I.,  Nishimoto,  S. and  Metcalfe,  R. (2015) Early post-mortem formation of carbonate concretions around tusk-shells over week-month timescales. Sci. Rep. 5, 14123. https://doi.org/10.1038/srep14123