A. Luebkea,
K. Lozaa,
R. Patnaikb,
J. Enaxa,
D. Raabec,
O. Prymaka,
H.-O. Fabritiusc,
P. Gaenglerd and
M. Epple*a
aInstitute of Inorganic Chemistry and Center for Nanointegration Duisburg-Essen (CeNIDE), University of Duisburg-Essen, Universitaetsstr. 5-7, 45117 Essen, Germany. E-mail: matthias.epple@uni-due.de
bCentre of Advanced Study in Geology, Panjab University, Chandigarh 160014, India
cMicrostructure Physics and Alloy Design, Max-Planck-Institut für Eisenforschung, Max-Planck-Str. 1, 40237 Düsseldorf, Germany
dORMED, Institute for Oral Medicine at the University of Witten/Herdecke, Alfred-Herrhausen-Straße 45, 58448 Witten, Germany
First published on 19th January 2017
The structure and composition of 13 fossilized tooth and bone samples aged between 3 and 70 million years were analysed. It was found that they all contained high amounts of fluoroapatite. This indicates that originally present hydroxyapatite had been converted to fluoroapatite during the diagenesis. Thus, the chemical analysis allows no conclusion with respect to the original composition of our fossil samples. Our results indicate that the diagenetic transformation of hydroxyapatite into fluoroapatite is at least partially dependent on microstructural characteristics of the original tissue such as the degree of porosity.
In an earlier publication we analysed the structure and the composition of six fossilized shark teeth.10 They contained high amounts of fluoroapatite in both dentin and enameloid, in some cases with a fluoride content close to stoichiometric fluoroapatite. In addition, we analysed the teeth of two plesiosaur and two dinosaur species and found a similar situation: high amounts of fluoroapatite within the tooth mineral. Our conclusion was that these species originally used fluoroapatite as tooth mineral and the change to hydroxyapatite during evolution was probably due to unconsidered environmental changes that caused unfavourable conditions for the use of fluoroapatite as tooth mineral.
However, it has been brought to our attention and commented in a recent publication,11 that the diagenetic conversion of hydroxyapatite to fluoroapatite is a common feature during fossilization (see ref. 12–14 and also the article by de Renzi et al.11). This means that no conclusion is possible for fossil samples with respect to their original fluoride content. Following this criticism, we have analysed further fossil samples of different origin with respect to their ultrastructure, their mineral composition, and their fluoride content. Our new results show that the fluorine content of fossil teeth and bones increases with age during the course of fossilization due to the diagenetic transformation of hydroxyapatite to the more stable fluoroapatite.
Fig. 1 Images of all investigated fossil specimens (see Table 1 for the individual species names and sample codes). The scale bar is 1 cm in all images. |
# | Taxa | Group | Stratigraphy | Location | Sample kind | Age/Ma |
---|---|---|---|---|---|---|
1* | Spinosaurus maroccanus | Dinosaur | Early Cretaceous | Oued-Zem, Morocco | Tooth | ∼110 |
2* | Carchadontosaurus saharicus | Dinosaur | Early Cretaceous | Taouz, Morocoo | Tooth | ∼100 |
3* | Plesiosaurus mauritanicus | Plesiosaur | Late Cretaceous | Oued-Zem, Morocco | Tooth | ∼70 |
4 | Most likely Sauropod dinosaur | Dinosaur | Late Cretaceous, Lameta Formation | Jabalpur, Madhya Pradesh | Bone | ∼70 |
5* | Squalicorax pristodontus | Shark | Late Cretaceous | Khouribga, Morocco | Tooth | ∼70 |
6* | Mosasaurus beaugei | Mosasaur | Late Cretaceous | Oued-Zem, Morocco | Tooth | ∼70 |
7* | Paleocarcharodon orientalis | Shark | Middle Palaeocene | Khouribga, Morocco | Tooth | ∼60 |
8* | Otodus obliquus | Shark | Middle Palaeocene | Khouribga, Morocco | Tooth | ∼60 |
9 | Raoellid tooth fragment | Mammal | Middle Eocene, Subathu Formation | Kalakot, Jammu and Kashmir, India | Tooth | ∼40 |
10* | Carcharocles angustidens | Shark | Oligocene | South Carolina, USA | Tooth | ∼28 |
11 | Tomistoma sp. | Crocodile | Early Miocene, Khari Nadi Formation | Kutch, Gujarat, India | Tooth | ∼16 |
12 | Megaselachus chubutensis | Shark | Early Miocene, Khari Nadi Formation | Kutch, Gujarat, India | Tooth | ∼16 |
13* | Carcharocles megalodon | Shark | Early Miocene | South Carolina, USA | Tooth | 3–20 |
14 | Gavialis sp. | Crocodile | Late Pliocene, Tatrot formation | Devni Khadri, India | Tooth | ∼2.6 |
15 | Carcharinus sp. indet | Shark | Late Miocene, Baripada Beds | Baripada, Orissa, India | Tooth | 8–10 |
16 | Lamna sp. | Shark | Late Miocene, Baripada Beds | Baripada, Orissa, India | Tooth | 8–10 |
17 | Carcharinus sp. | Shark | Late Miocene, Baripada Beds | Baripada, Orissa, India | Tooth | 8–10 |
18 | Myliobatis sp. | Ray | Late Miocene, Baripada Beds | Baripada, Orissa, India | Tooth | 8–10 |
19 | Carcharinus perseus | Shark | Late Miocene, Baripada Beds | Baripada, Orissa, India | Tooth | 8–10 |
20 | Rhinoptera sp. | Ray | Late Miocene, Baripada Beds | Baripada, Orissa, India | Tooth | 8–10 |
21* | Isurus hastalis | Shark | Early Pliocene | Copiapo, Chile | Tooth | ∼5 |
22 | Crocodylus sp. | Crocodile | Late Pliocene, Tatrot Formation | Devni Khadri, India | Tooth | ∼2.6–3 |
23 | Crocodylus sp. | Crocodile | Late Pliocene, Tatrot Formation | Devni Khadri, India | Bone | ∼2.6–3 |
Scanning electron microscopy (SEM) was used to visualize the internal structure of enameloid, enamel, dentin, and the dentin–enameloid-junction.
For axial freeze fracture, the teeth were immersed into liquid nitrogen for 2 min and mechanically broken into two pieces. Secondary electron (SE) microscopy and qualitative energy-dispersive X-ray spectroscopy (EDX) were carried out with an ESEM Quanta 400 FEG instrument after sputtering with gold and palladium (80:20).
To determine the chemical composition and the differences between enameloid, enamel, and dentin with respect to crystallinity and crystallite size, X-ray powder diffraction (XRD) was used. For this, the samples were ground into a fine powder as follows: the teeth were transversely cut using a Proxxon FBS 230/E fine drilling and polishing tool equipped with a diamond-coated cutting disk. Fine powders of enameloid, enamel, or dentin, respectively (a few mg per sample), were obtained from corresponding areas of the cut teeth with the same instrument using a diamond-coated drill. This powder was used for X-ray diffraction and elemental analysis.
X-ray powder diffraction was carried out with a Bruker D8 Advance diffractometer (Cu Kα radiation, λ = 1.54 Å), using a silicon single crystal sample holder to minimize background scattering. Rietveld refinement for the calculation of the lattice parameters and the crystallite sizes was performed with the Bruker software TOPAS 4.2. For each Rietveld refinement, the instrumental correction was included as determined with a standard powder sample LaB6 (from NIST, National Institute of Standards and Technology, as standard reference material, SRM 660b). The size of the crystallites was calculated with the Scherrer equation after correction for instrumental peak broadening.16
The a-axis of the enameloid and of the geological fluoroapatite single crystal was shorter than in synthetic hydroxyapatite. We have estimated the fluoride content using the correlation function given by LeGeros and Suga.5 As this method assumes that the content of foreign ions (e.g. Na, Mg, carbonate) is negligible, these fluoride contents are associated with a considerable error. In particular, the presence of carbonate in the apatite lattice also leads to a shortening of the a-axis, which results in systematically higher apparent fluoride-contents using the LeGeros/Suga method.
Elemental analysis was carried out to determine the overall chemical composition of the samples. For the determination of calcium with atomic absorption spectroscopy (AAS), fluoride with ion-selective potentiometry, and phosphate with ultraviolet (UV) spectroscopy, the powders were dissolved in concentrated ultrapure hydrochloric acid. Calcium was determined with a ThermoElectron, M-Series atomic absorption spectrometer. Phosphate was determined with a Varian Cary 300 UV-Vis spectrophotometer as phosphate–molybdenum blue complex. For fluoride analysis, ion-selective potentiometry was used (ion-selective electrode, ISE; pH/ION 735, WTW; measurement performed by Analytische Laboratorien GmbH, Lindlar, Germany).
Thermogravimetry (TG) was used to determine the contents of water, organic matrix and carbonated apatite in the teeth. For TG analysis, the teeth were transversely cut. To obtain pure enamel or enameloid, we cut off the tip of the tooth for analysis. To obtain almost pure dentin, we used the lower part of the tooth where it met the root. Thermogravimetry was carried out with a Netzsch STA 449 F3 Jupiter instrument in dynamic oxygen atmosphere (heating rate 2 K min−1 from 25 to 1200 °C; open alumina crucibles).
The full results of the chemical and crystallographic composition of the samples are given in Table 2. All samples consist of apatite as calcium phosphate mineral and contain variable amounts of calcium carbonate and quartz due to diagenesis during fossilization. With respect to the fluoride content, all samples contain a considerable fraction of fluoride. The fluoride content in enameloid and partially also in enamel and dentin is almost stoichiometric to fluoroapatite. Where possible, we have separated enamel (ectodermal origin), enameloid (dentin-like, mesectodermal origin) and dentin (mesectodermal origin), and separately analysed these phases.
No. according to Fig. 1 and Table 1 | Species | Tissue | a-axis/Å | Crystallite size of fluoroapatite from XRD/nm | Composition (XRD) in wt% | F-content (XRD)/wt% | F-content (EA)/wt% | Ca-content (EA)/wt% | PO43−-content (EA)/wt% | Molar Ca/P ratio |
---|---|---|---|---|---|---|---|---|---|---|
4 | Sauropod bone | Compacta | 9.371(2) | 19(1) | 95% FAP; 4% calcite; 0.4% quartz | 4.12 | 2.97 | — | — | — |
Spongiosa | 9.387(2) | 22(1) | 65% FAP; 1.2% calcite; 34% quartz | 3.18 | 1.19 | — | — | — | ||
9 | Raoellid mammal | Enamel | 9.374(1) | 42(1) | 98% FAP; 1.6% quartz | 3.9 | 1.88 | 37.9 | 46.3 | 1.94:1 |
11 | Tomistoma | Enamel | 9.388(2) | 28(1) | 100% FAP | 3.12 | 3.23 | 33.55 | 44.3 | |
Dentin | 1.05 | — | — | — | ||||||
12 | Megaselachus chubutensis | Enameloid | 9.398(1) | 72(3) | 36% FAP; 7% quartz, 57% calcite | 2.53 | 0.87 | 34.65 | 46.35 | 1.77:1 |
Dentin | 9.396(1) | 19(1) | 77% FAP; 7% quartz, 16% calcite | 2.65 | 1.9 | — | — | — | ||
14 | Gavialis sp. | Enamel | 9.374(2) | 17(1) | 88% FAP; 10% calcite; 1.8% quartz | 3.9 | 2.68 | 38.25 | 45.25 | 2.00:1 |
15 | Carcharinus sp. | Enameloid | 9.383(2) | 26(1) | 100% FAP | 3.44 | 2.81 | 39.4 | 46.35 | 2.01:1 |
16 | Lamna sp. | Enameloid | 9.371(1) | 38(1) | 100% FAP | 4.12 | 3.32 | 37.0 | 50.2 | 1.75:1 |
17 | Carcharhinus sp. | 9.377(1) | 31(1) | 100% FAP | 3.76 | 34.9 | 48.2 | 1.72:1 | ||
18 | Myliobatis sp. | Enameloid | 9.376(9) | 22(1) | 100% FAP | 3.82 | 3.04 | 37.3 | 46.9 | 1.88:1 |
Dentin | 9.370(1) | 22(1) | 96% FAP, 4% quartz | 4.18 | 2.56 | 27.8 | 35.85 | 1.84:1 | ||
19 | Carcharhinus perseus | Enameloid | 9.379(1) | 54(1) | 100% FAP | 3.65 | 3.0 | 37.0 | 47.8 | 1.83:1 |
20 | Rhinoptera sp. | Enameloid | 9.377(3) | 19(1) | 99.6% FAP; 0.4% quartz | 3.76 | 3.25 | 33.9 | 45.2 | 1.78:1 |
22 | Crocodylus sp. | Enamel | 9.371(3) | 32(1) | 100% FAP | 4.12 | 2.7 | 37.25 | 44.9 | 1.97:1 |
Dentin | 9.361(3) | 21(1) | 96% FAP; 4% calcite | 4.77 | 2.92 | — | — | — | ||
23 | Crocodylus sp. jaw bone | Compacta | 9.371(2) | 58(1) | 15% FAP; 70% calcite; 15% quartz | 4.12 | 0.48 | — | — | — |
Spongiosa | 9.381(1) | 28(1) | 62% FAP; 37% calcite; 0.4% quartz | — | 1.7 | — | — | — | ||
Geological fluoroapatite single crystal8 | 9.375(3) | — | — | 3.64 | 38.45 | 53.25 | 1.71:1 | |||
Stoichiometric hydroxyapatite (computed) | 9.425 | — | — | 0 | 39.90 | 56.72 | 1.67:1 |
Diagenesis of bones and teeth is a common phenomenon that is influenced by the composition of the enclosing sedimentary rock. The presence of groundwater facilitates this process by providing a medium for the ion exchange. During diagenesis, pH and Eh conditions of the surrounding environment play a major role in the precipitation of matrix-derived calcium and iron carbonates in the bone and teeth cavities.17 Often, the biogenic apatite is replaced by francolite where PO43− is substituted by CO32− and OH− by F−.14,18–20 Fossil-bearing deposits that are exposed to ground water fluctuations undergo a diagenetic ion exchange leading to deposition of fluoride-bearing apatite.21 A recrystallization of francolite may introduce carbonate ions from groundwater in the bone cavities.22
Fig. 2 shows the fluoride content in all analysed species as a function of age. As expected, the enameloid of sharks and rays contains almost stoichiometric amounts of fluoride. The values for the fossil taxa correspond well to those known for recent species, with a slight trend towards elevated fluoride contents with increasing age. This trend is much more pronounced in the dentin samples from sharks and rays. As the dentin of recent Chondrichthyes consists of hydroxyapatite, the fluorine must have entered the teeth from the surrounding sediments. This is most probably facilitated by the higher natural porosity of dentin in comparison to the very dense enameloid. The frequently observed presence of small crystallites with various shapes within the dentin tubuli of the fossil samples supports this.
While all our tooth and bone samples from fossil dinosaurs, plesiosaurs, mosasaurs, crocodiles, bony fish and mammals contain elevated amounts of fluorine, there is no obvious correlation between fluorine content and sample age. Nevertheless, given the different age, taxa, and location of all samples it is highly unlikely that the fluoride was present at the time of the death of the organism. It must have entered the samples during fossilization by fluoride uptake from the surrounding sediments.
Besides the changed chemical composition, the overall structure of the samples was remarkably well preserved. However, at higher magnifications we frequently observed structures that can be interpreted as diagenesis artefacts, such as small, odd-shaped crystallites within shark tooth enameloid (Fig. 3B) or the small pointy crystallites in the sauropod bone (Fig. 9B–D) which are not present in samples from recent relatives. This is an observation that we have already briefly reported in our preceding publication.10 In the following, we show some representative examples. The complete results can be found in the ESI.†
Fig. 3 SEM-images of a tooth of Carcharinus, 10 million years old (sample 15, Table 1). The enameloid–dentin junction is well visible (A), together with the well-preserved enameloid ultrastructure (B). |
Fig. 3 shows SEM images of a fossilized shark tooth where enameloid, dentin and the enameloid–dentin junction have a well-preserved original morphology. The ultrastructure resembles that of recent shark teeth with the typical fluoroapatite crystallite bundles.8,9 However, larger crystallites with random shapes dispersed between the bundles indicate diagenetic alterations (Fig. 3B).
Fig. 4 shows an EDX line scan across the fossilized shark tooth shown in Fig. 3. There is no significant gradient in the fluoride content. The average EDX intensities are 106 ± 29 for dentin and 99 ± 46 for enameloid (average ± standard deviation). The correlation coefficients after linear regression are 0.197 for dentin and 0.019 for enameloid. Altogether, this indicates constant and identical fluoride concentrations in dentin and enameloid within the accuracy of the method and the statistical scatter within each tissue.
Fig. 4 EDX line-scan of fluorine across the enameloid (left site) into the dentin (right site) of a tooth of Carcharinus (sample 15, Table 1). The fluorine is homogeneously distributed across the tooth. |
The individual EDX spectra of this tooth are shown in Fig. 5 with no discernible difference between enameloid and dentin.
Fig. 5 EDX spectra of a tooth of Carcharinus (sample 15, Table 1). Fluorine is detectable in enameloid (top) and dentin (bottom). |
Fig. 6 shows a representative X-ray powder diffractogram of this tooth. The only crystalline phase is apatite. A further analysis shows the presence of fluoroapatite. The lattice parameters and calculated fluorine content according to LeGeros/Suga et al. are given in Table 2.
Fig. 6 Rietveld refinement of an X-ray powder diffraction measurement of a tooth of Carcharinus (sample 15, Table 1). |
Fig. 7 shows SEM images of the tooth of a fossil crocodile. The dentin–enamel junction is well preserved, and small individual crystallites oriented perpendicular to the tooth surface can be distinguished in the enamel, which is similar to the structure in recent crocodile teeth.23
Fig. 7 SEM images of a tooth of Crocodylus, 3 million years old (sample 22, Table 1), showing the enamel–dentin junction (A), and an enamel close-up (B). |
Fig. 8 shows SEM images of the jawbone of the same crocodile species. Partially, the fossil bone has preserved its original ultrastructure consisting of nanoscale platelet-like structures. However, the major part of the mineral phase is composed of fluoroapatite and calcite (see Table 2) according to X-ray powder diffraction and Rietveld refinement. The amount of fluoroapatite was higher in the substantia spongiosa, while the amount of calcite was higher in the substantia compacta. In both layers, high amounts of quartz were found, representing fossilisation artefacts.
Fig. 8 SEM-images of the jawbone of Crocodylus, 3 million years old (sample 23, Table 1), showing the platelet nanostructure. |
Fig. 9 shows SEM images of a fossil sauropod dinosaur bone. The transition from substantia compacta (small pores) to substantia spongiosa (large pores) is well visible (Fig. 9A). Those pores are large enough to be filled with fossilisation artefacts. The EDX mapping shown in Fig. 10 indicates that the material present in the pores is most likely quartz, SiO2.
Fig. 9 SEM images of a sauropod dinosaur bone, 70 million years old (sample 4, Table 1). (A) shows the intersection of spongy and compact bone, (B) and (C) show close-ups of compact bone, and (D) shows a close-up of porous spongy bone. |
Fig. 10 EDX mapping of sauropod dinosaur bone (sample 4, Table 1), indicating quartz-filled pores. |
Fig. 11 shows SEM images of a middle Miocene mammal and another fossilised shark tooth. In the case of the raoellid mammal, individual small crystallites are arranged in parallel and oriented with their long axes perpendicular to the tooth surface, which is comparable to the ultrastructure of recent crocodile enamel23 and to human enamel.24 The enameloid of teeth from the fossil Lamna species consists of thin and long crystallites arranged in parallel forming bundles that can also be found in recent shark tooth enameloid.9
Fig. 11 SEM images of mammal tooth (whale ancestor, 50 million years old; sample 9, Table 1) (A), and a tooth of the shark Lamna sp. (8–10 million years old; sample 16, Table 1) (B). Both images show individual bundles in parallel direction in enamel (A) as well as in enameloid (B). |
Footnote |
† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra27121a |
This journal is © The Royal Society of Chemistry 2017 |