Introduction

Hydroxyapatite [Ca10(PO4)6(OH)2, HAp] is an essential biocompatible material used in orthopedics and dentistry, exhibiting excellent chemical similarity to the mineral components of bone and hard human tissue [1]. It has relevant applications in bone replacement technologies [2], bioactive coatings on metallic implants [3, 4] the design of drug and gene delivery agents [5], and as a dental material [6]. The crystal structure is hexagonal (space group P63/m) with an arrangement of Ca2+ and \({\text{P}}{\text{O}}_{4}^{3-}\) ions around \({\text{O}}{\text{H}}^{-}\) ion columns. The hexagonal unit cell contains ten calcium cations arranged in two non-equivalent positions. Four are located at the Ca(I) sites forming a channel surrounded by phosphates along the c-direction, and six are located at the Ca(II) sites, bound with \({\text{P}}{\text{O}}_{4}^{3-}\) ions and forming triangles staggered by 60° around the hydroxyl channel [7]. The relative abundance of these cations has been established as Ca(I)/Ca(II) = 2/3 [8, 9], making HAp a compound that lends itself to the incorporation of a range of cations by substituting into Ca2+ sites [10, 11]. These substitutions have emerged as an alternative approach to functionalize HAp, adding new luminescent and magnetic properties [12,13,14]. Some studies have focused on doping HAp with transition metal ions such as Mn2+, Co2+, Fe2+, and Fe3+, thus modifying the native diamagnetic behavior and obtaining a paramagnetic or superparamagnetic biomaterial [15,16,17].

Magnetic hydroxyapatite shows promising potential for applications in cancer theranostics, magnetic hyperthermia treatment, tissue engineering, bone regeneration, biosensors, and as contrast agents for magnetic resonance imaging [18]. Additionally, paramagnetic HAp is increasingly being explored for its utility as a green catalyst [19]. For instance, Campisi et al. [20] reported that HAp with Fe occupying Ca(II) sites could serve as an effective catalyst for reducing NOx emissions by ammonia. Similarly, Lu et al. [21] demonstrated that Fe-doped HAp exhibits outstanding performance in oxygen reduction reaction (ORR), suggesting its potential as a magnetic single-atom biological catalysts in biofuel cells and energy-related devices. Shi et al. [22] found that magnetic HAp enhances the catalytic detoxification of polycyclic aromatic hydrocarbons in fly ash. Furthermore, various studies have investigated how dopant percentages and HAp morphology influence catalytic activities across different reactions [23].

Recent advancements in catalysis and bone tissue engineering have focused on developing elongated one-dimensional (1D) HAp nanostructures such as tubes, wires, rods, ribbons, belts, and fibers owing to their large surface–area–to–volume ratios. The biological and catalytic activity of the HAp is highly influenced by its specific surface area [24], which enhances drug-loading capacities [25]. Notably, fibrous hydroxyapatite has garnered significant interest either as a self-reinforcement element or as a reinforcing component in biomedical composites, particularly in porous bone blocks and defect fillers [26]. Moreover, multifunctional HAp scaffolds based on paramagnetic HAp nanofibers have been engineered to influence bone growth through external magnetic fields [27]. Meng et al. [28] created a magnetic nanofibrous HAp scaffold incorporating γ-Fe2O3 superparamagnetic nanoparticles, which exhibits typical saturation magnetization values of 0.049 emu/g and promotes osteogenesis for bone repair in vivo. However, the safety of magnetic-responsive biomaterials is paramount for human applications. Thus, iron-based nanoparticles, known for their low cytotoxicity, are predominantly used in biomedical applications requiring magnetic stimulation [15, 29]. The selective incorporation of magnetic ions into the HAp lattice via ion exchange offers a promising approach to developing magnetic responsive HAp with minimal dopant concentrations while preserving its intrinsic biological properties [30].

Previously, we demonstrated the fabrication of luminescent and paramagnetic HAp through the selective incorporation of Fe3+ ions into HAp nanobelts, where the Fe3+ dopant substituted Ca(II) sites without inducing other oxidation states [31]. This selectivity in the oxidation state extends to other transition metals as well. For instance, Wakamura et al. [32] successfully incorporated Cr3+ into HAp nanostructures, noting morphology changes correlated with varying chromium concentrations. De Araujo et al. [33] also incorporated Cr3+ in HAp powders and confirmed the presence of dopant ions through characteristic absorption and emission bands. Moreover, Bangar et al. [34] conducted in vitro biological studies on Cr3+-doped HAp with HeLa and L929 cells, revealing a hemolytic ratio below 5% and affirming its biocompatibility with human blood for biomedical applications [35].

Cr3+ is recognized as an essential micronutrient vital for the human metabolism of proteins [36], characterized by low toxicity due to the poor solubility of its complexes at physiological pH levels, preventing cell membrane penetration [37]. In stark contrast, water-soluble Cr6+ ions pose significant health risks due to their structural similarity to phosphates, enabling their active transport into cells in place of anion \({\text{P}}{\text{O}}_{4}^{3-}\), leading to several outcomes including cancer and damage to the liver, kidney, respiratory, and immune systems [38, 39]. These health effects exclude the use of Cr6+ in biomedical applications and demand precise control in the chromium oxidation state that assures the exclusive presence of Cr3+ in a biomaterial. These adverse health effects necessitate stringent control over the chromium oxidation state in biomaterials, ensuring the exclusive presence of Cr3+ for safety.

To the best of our knowledge, the effect of chromium incorporation on the magnetic properties of hydroxyapatite, as well as the synthesis of elongated HAp:Cr nanostructures, have not been previously explored. This study presents the first evidence of paramagnetism in Cr3+-doped HAp nanofibers, demonstrating a magnetization saturation of 0.0460 emu/g at 10 kOe for Cr3+ concentrations exceeding 2.9 at.%. Our synthesis method achieves selective incorporation of Cr3+ into Ca(II) sites within the HAp lattice, leading to the formation of oxygen vacancies in \({\text{P}}{\text{O}}_{4}^{3-}\) groups and \({\text{O}}{\text{H}}^{-}\) groups. These vacancies become singly ionized, capturing electrons from Cr3+ ions within the HAp:Cr matrix. We suggest that these singly ionized oxygen vacancies act as spin ½ systems, thereby enhancing the paramagnetic response of the Cr3+ ions. Our findings position HAp:Cr as a promising candidate for bone regeneration treatments, offering potential for bone growth stimulation by magnetic fields and improved contrast in magnetic resonance imaging (MRI), which could significantly enhance diagnostic accuracy. Additionally, its versatility for theranostic applications in bone and tissue regeneration underscores the broad scope and potential utility of this biomaterial in the medical field.

Experimental Methodology

Hydroxyapatite nanofibers in powder form were prepared using a hydrothermal method described in our previous work [31, 40]. In brief, an aqueous solution (0.1 M) of di-ammonium phosphate [(NH4)2HPO4, Alfa Aesar, 11597-A1, CAS Number: 7783-28-0] was added drop-wise (1 drop per second) to a 0.1 M aqueous solution of calcium nitrate tetrahydrate [Ca(NO3)2⋅4H2O, 99%, Alfa Aesar, A16645-36, CAS Number: 13477-34-4] at a concentration of 0.2 M, previously heated under magnetic stirring in an electric heating mantle (Vakmo Tech, HMSC 250 mL) at a temperature of 100 °C. The calcium to phosphate (Ca/P) volume ratio of the precursor solutions was maintained at 1.667. After 48 h of stirring at 100 °C, the solution was cooled to room temperature. The HAp precipitate was collected by filtration and washed thoroughly with deionized water until the aqueous phase reached a pH of 7. Finally, the product was dried at 50 °C in an electric heating plate (Termolyne nuova HP18325) for 1 h under an air atmosphere.

For incorporation of the Cr3+ dopant ions, we used an ion-exchange procedure based by the following reaction:

$${\text{Ca}}_{10} \left( {{\text{PO}}_{4} } \right)_{6} \left( {{\text{OH}}} \right)_{2} + 2x {\text{Cr}}^{3 + } \left( {{\text{aq}}} \right) \to {\text{Ca}}_{{\left( {10 - 2x} \right)}} {\text{Cr}}_{2x} \left( {{\text{PO}}_{{4}} } \right)_{6} \left( {{\text{OH}}} \right)_{2} + 2x {\text{Ca}}^{2 + } \left( {{\text{aq}}} \right) + x V_{{{\text{Ca}}}}$$
(1)

in which three Ca2+ ions from the HAp lattice are substituted with two Cr3+ ions, resulting in one Ca2+ vacancy (\({\text{V}}_{\text{Ca}}\)). The chromium-exchange solution was prepared with chromium (III) nitrate nonahydrate [Cr(NO3)3⋅9H2O, 99.999%, Sigma Aldrich, 379972, CAS Number: 7789-02-8] at a concentration of 0.2 M. The previously synthesized HAp nanofibers were immersed in the chromium-exchange solution by mixing 0.5 g/dL and maintained under magnetic stirring at room temperature. The ion-exchange time was varied between 1, 12, 24, 36, and 48 h, to obtain five samples labeled HAp:Cr-1, HAp:Cr-12, HAp:Cr-24, HAp:Cr-36, and HAp:Cr-48, respectively. The powder samples were collected by filtration, washed until a pH of 7 was obtained, and dried at 50 °C for 1 h.

The crystalline structure of the powders was determined from X-ray powder diffraction (XRD) using a Philips X’pert MPD diffractometer operating at 30 kV and 10 mA with monochromatic CuKα radiation (λ = 0.15405 nm). The XRD spectra were recorded with Bragg- Brentano geometry in the range of 2θ from 20 to 70° with a step size of 0.02° and a second dwell per point. The vibrational modes of HAp were analyzed by Raman spectroscopy (Horiba Jobin–Yvon, LabRam HR800 micro-Raman system, Villeneuve d’Ascq, France), using a 633 nm He–Ne laser (Melles Griot, NY, USA) for excitation. An air-cooled charge-coupled device (CCD) detector was used to collect the scattered light dispersed by a 1600 lines/mm grating. The spectral resolution of the system used was 1.5 cm−1. For a systematic analysis, 0.5 g of each powder sample was deposited onto a silicon < 100 > wafer and flattened with a cover glass to obtain a homogeneous flat area, all Raman spectra were obtained with the 10 × Olympus objective (0.25 NA), which allows the measurements with a working distance of 10 mm to avoid the intense fluorescence of the hydroxyapatite. Raman spectra were collected and analyzed using the Labspec 5.0 software (Horiba, France, Labspec version 5.0) of the confocal microscope. To verify the correct operation of the spectrometer, the Raman spectra of well-known samples such as polypropylene and silicon were obtained and compared with reference ones. Furthermore, the LabRam HR800 micro-Raman system was used to acquire photoluminescence spectra from the samples in the 300–1100 nm spectral range, the same Raman measurements setup was used but changing the 325 nm line of a He-Cd laser and a 40 × UV objective (0.47 NA, Thorlabs, Newton, NJ, USA). An Oxford Instruments energy dispersive X-Max spectrometer (Abingdon, UK) with a detector size of 20 mm2, coupled to a JEOL JIB-4500 (Tokyo, Japan) scanning electron microscope (SEM), and the INCA X-Stream 2 (Oxford Instruments, Concord, MA, USA) were used to determine the elemental composition of the samples and to verify the presence of chromium in the HAp:Cr samples, as well as to quantify the Ca/P ratio. The powders were placed on double-sided conductive carbon tape mounted on the SEM holder. For each sample 10 EDS spectra were acquired at 15 keV with the No. 2 aperture (density current 38 A/m2) and 15 mm of work distance. Atomic percentage uncertainty of the detected elements corresponded to the standard deviation (sigma) of the INCA software calculation operated with a standards-based sequence. Hydroxyapatite (Ca10(PO4)6(OH)2) commercial powder sample (Sigma Aldrich, 900204, CAS Number: 1306-06-5) was used as a reference during the measurements to confirm the correct calibration of the EDS system, Ca, P and O concentrations were 43.5 (± 0.4), 26.1 (± 0.4) and 30.4 (± 0.7) atomic percent, respectively, and a Ca/P atomic relation of 1.67.

X-ray photoelectron (XPS) spectroscopy measurements were performed using a SPECS system with a PHOIBOS WAL analyzer and an aluminum anode. A 150 eV pass energy and 1 eV step size were used to acquire survey spectra. To determine the oxidation state of chromium, high-resolution XPS spectra of the Cr 2p peak were acquired using a pass energy of 50 eV and a step size of 0.1 eV. Narrow spectra were obtained using 300 scans during the measurement. The XPS spectra were calibrated by setting the C 1 s peak at 284.6 eV as a reference for the binding energy scale [41]. The sample morphology was analyzed using a field-emission scanning electron microscope (FE-SEM, FEI, Nova NanoSEM 200). Furthermore, the morphology was studied by a transmission electron microscope (TEM, JEOL, JEM 2200FS + CS), equipped with a field emission gun, working at an acceleration voltage of 200 kV and designed for high-resolution (HRTEM) imaging with a CEOS aberration corrector for the objective lens. Its ultimate point resolution in HRTEM mode is 1.0 Å. Additionally, the elements (Ca, P, O, and Cr) distribution was observed by TEM-EDS maps. For the TEM measurements 0.01 g of powder samples were dispersed into 10 ml of ethanol (JT Baker solution, Phillipsburg, NJ, USA) and systematically diluted until a transparent solution was obtained, then a drop of the sample was added onto a carbon-coated copper mesh grid. A Gatan MonoCL4 system (Pleasanton, CA, USA) equipped on the JEOL JIB-4500 SEM CL was used to obtain cathodoluminescence (CL) spectra in the visible range (200–800 nm). In addition, a vibrating sample magnetometer (VSM) attached to an Evercool-Physical Properties Measurement System (PPMS) from Quantum Design (San Diego, CA) was employed to determine the magnetic behavior of the HAp:Cr samples [42]. In brief, VSM works by placing a sample in a constant and uniform external magnetic field at a controlled temperature. As the magnetization in the vibrating sample increases, the induced current by the detector coils also increases. Therefore, the differences in current are translated by a computer as variations in the magnetization of the sample. Axial and radial sample positioning error introduces inaccuracy of the reported magnetic moment. Hence, this work was carried out considering the correct centering of the samples. First, the HAp:Cr powders were homogenized in an agate mortar, and then 0.005 g of sample mass was deposited inside the VSM Quantum Design powder sample holder (P125E,4096-388), which is an injection molded polypropylene (Formolene 4100N) plastic part that snaps into a brass half-tube Quantum Design sample holder (C130B) 4500-608, for a secure press fit. Then, the total length of the holder and the exact location of the sample were defined by using the PPMS-VSM Quantum Design mounting station. For all magnetic measurements, the sample was set in the middle of the polystyrene holder keeping the sample length less than 3 mm to maintain the accuracy of the point source dipole approximation. A 0.25 g cylindrical shape palladium standard sample with an expected moment of 0.01313 emu was used to calibrate the magnetometer at 1 Tesla and 298 K. The magnetization per unit mass (M, emu/g) measurements as a function of the applied magnetic field (H, Tesla) were performed for all the samples at 298 K, and the magnetic field was swept from 0 to 1 T, from 1 T to − 1 T, and again from − 1 T to 1 T.

Results and Discussion

The elemental composition of the as-synthesized HAp and HAp:Cr powders, obtained from energy dispersive spectroscopy (EDS) and listed in Table 1, show (Ca + Cr)/P ratios below 1.67 (stoichiometric value), suggesting that the HAp powders are calcium deficient. The Cr/Ca atomic ratio increases with increasing additions of chromium, confirming that chromium is incorporated into the HAp powders. Figure 1a–f illustrate the XRD patterns of the powders, where the diffraction peaks correspond to the hexagonal P63/m HAp structure (JCPDS card number 09-0432). The powders do not contain crystalline chromium oxide or metallic chromium impurity phases. Since EDS measurements reveal a calcium deficiency in all samples (see Table 1), we assume that the calcium-deficient (Ca-D) phase is present. However, identifying this phase is difficult because the HAp and Ca-D HAp structures display diffraction peaks at similar Bragg angles [43]. The XRD patterns also show that incorporating higher amounts of chromium decreases the crystalline quality of the HAp:Cr powders, as indicated by the broadening in the main diffraction peaks located at diffraction angles between 30 and 35° 2θ, most evident in the HAp:Cr-48 sample with a chromium concentration of 3.4 at.%.

Table 1 Atomic compositions of the HAp and HAp:Cr powders determined from energy dispersive spectroscopy
Full size table
Fig. 1
figure 1

X-ray diffraction patterns of the a HAp, b HAp:Cr-1, c HAp:Cr-12, d HAp:Cr-24, e HAp:Cr-36, and f HAp:Cr-48 powders. Scanning electron micrographs of the g HAp:Cr-36 and h HAp:Cr-48 powders

Full size image

Scanning electron micrographs obtained for the HAp:Cr-36 (Fig. 1g) and HAp:Cr-48 (Fig. 1h) powders show the formation of elongated nanostructures of varying lengths, and the transmission electron micrographs (Fig. 2) show widths between 50 and 100 nm. Elemental X-ray maps obtained for the HAp:Cr-48 powder show homogeneous signals for P (Fig. 3b), Cr (Fig. 3c), Ca (Fig. 3d), and O (Fig. 3e) along the nanofibers, confirming the uniform incorporation of chromium. Figure 4a, b illustrates high-resolution transmission electron micrographs along the [221] zone axis acquired for the HAp:Cr-36 and HAp:Cr-48 powders, respectively. The nanofibers exhibit a long axis in the [\(\stackrel{\text{-}}{1}\stackrel{\text{-}}{1}{\text{4}}\)] direction, which represents a tilt of only 7° from the [001] direction previously reported as the preferential growth direction of HAp nanofibers synthesized by the same method [31, 40]. Figure 4c, d are magnified images of the red squares in Fig. 4a, b, respectively, revealing a hexagonal array of atoms aligned along the [\(\stackrel{\text{-}}{1}\stackrel{\text{-}}{1}{\text{4}}\)], [\(\stackrel{\text{-}}{2}{\text{23}}\)], and [\(\stackrel{\text{-}}{2}{\text{1}}\stackrel{\text{-}}{2}\)] directions, with interatomic distances of 0.37 and 0.32 nm. These values agree with the a and b parameters of the ideal hexagonal HAp unit cell, modeled in Fig. 4e.

Fig. 2
figure 2

Transmission electron micrographs of the a HAp:Cr-1, b HAp:Cr-12, c HAp:Cr-24, d HAp:Cr-36, and e HAp:Cr-48 powders

Full size image
Fig. 3
figure 3

a Transmission electron micrograph of the HAp:Cr-48 powder with corresponding EDS maps for b phosphorus, c chromium, d calcium, and e oxygen

Full size image
Fig. 4
figure 4

High-resolution transmission electron micrographs of the a HAp:Cr-36 and b HAp:Cr-48 powders. c, d Insets (red squares) of the a and b micrographs, respectively. e Crystal structure of the HAp structure projected along the [221] zone axis

Full size image

The chemical states of elements on the surfaces of the powders were analyzed using X-ray photoelectron spectroscopy (XPS). Figure 5 displays the survey XPS spectra, showing peaks for O 1 s, Ca 2p, and P 2p signals at approximately 532, 347, and 133 eV, respectively. For the HAp:Cr powders, Cr 2p doublet signals at 576 and 586 eV were also identified. Figure 6 presents the high-resolution spectra of the Cr 2p1/2 and 2p3/2 signals in HAp:Cr powders, which are centered at the binding energy values listed in Table 2. A notable energy difference of about 9.7 eV between these signals corresponds exclusively to the + 3 oxidation state of chromium [44]. This figure also illustrates that the intensity of the Cr 2p1/2 and 2p3/2 signals increases with the Cr-exchange time used during sample preparation, corresponding with the chromium concentration determined by EDS (Table 1). Table 2 summarizes the elemental compositions of the sample surfaces as calculated from the XPS results, showing slightly higher values than those obtained from EDS measurements (Table 1), a discrepancy due to the ion exchange process primarily occurring at the HAp surfaces.

Fig. 5
figure 5

X-ray photoelectron spectra of a HAp, b HAp:Cr-1, c HAp:Cr-12, d HAp:Cr-24, e HAp:Cr-36, and f HAp:Cr-48 powders

Full size image
Fig. 6
figure 6

X-ray photoelectron spectra of Cr 2p1/2 and Cr 2p3/2 signals exhibiting an energy difference corresponding to the + 3 oxidation state of chromium

Full size image
Table 2 Atomic compositions and Cr 2p3/2 and Cr 2p1/2 signals of the HAp and HAp:Cr powders determined from X-ray photoelectron spectroscopy
Full size table

High-resolution, deconvoluted XPS spectra of O 1 s, Ca 2p, P 2p, and Cr 2p signals for the HAp:Cr-48 sample are shown in Fig. 7. Figure 7a identifies components of the O 1 s signal at 531.6, 532.7, and 534.6 eV, corresponding to O–C bonds from atmospheric CO2, O–P and O–H bonds from phosphates and hydroxyl groups, and O–H2 bonds from adsorbed water, respectively [45]. Figure 7b displays the Ca 2p3/2 and 2p1/2 spin–orbit coupling signals at 347.8 and 351.3 eV, respectively, linked to Ca–PO3 bonds in the HAp lattice [46, 47]. The 2p3/2 and 2p1/2 components of the P 2p signal are shown in Fig. 7c, centered at 133.0 and 133.8 eV, respectively, consistent with the binding energy for P–O bonds in the HAp structure [48]. Lastly, Fig. 7d presents the 2p3/2 and 2p1/2 components of the Cr 2p signals, where the 2p3/2 signal has been deconvoluted using the multiplet-splitting components for Cr3+ reported by Biesinger et al. [49], at 575.7, 576.7, 577.5, 578.5, and 578.9 eV. These multiplets, generated by the exchange interaction between unpaired valence and core electrons during photoionization, definitively indicate the + 3 oxidation state of chromium. Complementing this, Lima et al. [50] used X-ray absorption near-edge structure (XANES) measurements to demonstrate that the valence of chromium ions in HAp changes from 3 + to + 6 when the chromium concentration exceeds 18 mol.% (5.5 at.%), a finding consistent with our XPS results as all our Cr-doped samples contained less than 4.4 at.% chromium (Table 2).

Fig. 7
figure 7

High-resolution X-ray photoelectron spectra for a O 1 s, b Ca 2p, c P 2p, and d Cr 2p of the HAp:Cr-48 powders

Full size image

The Raman spectra of the HAp powders (Fig. 8) exhibit four characteristic vibrational signals centered at 430, 590, 962, and 1043 cm−1. The signals centered at 430 cm−1 (v2) and 590 cm−1 (v4) correspond to the O–P–O symmetric and asymmetric bending modes, respectively. The signals at 962 cm−1 (v1) and 1043 cm−1 (v3) correspond to the symmetric and asymmetric stretching vibration modes of P–O bonds, respectively [51, 52]. The weak signal centered at about 322 cm−1 corresponds to translational modes of the Ca2+ ions in the HAp lattice [53]. Raman spectra of the HAp:Cr powders reveal the same characteristic vibrational signals detected for HAp, but with the appearance of two signals at about 900 and 970 cm−1, which increase in intensity as the chromium concentration increases (arrows in Fig. 8). We propose that these signals correspond to vibrational modes related to the presence of Cr3+. Previously, Gomes et al. [54] reported that Zn incorporation in HAp induces a distortion of the phosphate tetrahedron in the vicinity of the incorporated Zn2+ ions, resulting in distortion of the v1 vibrational mode by the appearance of two signals around 958 and 968 cm−1, both of which increase in intensity as the Zn concentration in HAp is increased. Therefore, considering that the v1 vibrational mode represents the symmetrical stretching vibration of oxygen atoms around the phosphorus atoms in the PO4 groups, which are bound to Ca(II) ions in the HAp lattice, we postulate that the replacement of Ca(II) by Cr3+ ions generates the distortion of the v1 vibrational mode due to the charge imbalance generated. In addition, such charge imbalance produces calcium vacancies (\({\text{V}}_{\text{Ca}}\)) as a compensatory mechanism following Eq. (1), and as will be discussed later, also produces oxygen vacancies (\({\text{V}}_{\text{O}}\)) in the PO4 groups, which certainly modify the stretching vibration of the remaining oxygen atoms.

Fig. 8
figure 8

Raman spectra of the HAp and HAp:Cr powders. Arrows point to signals attributed to the distortion of P-O vibrational modes due to Cr3+ incorporation in the Ca(II) sites of the HAp lattice

Full size image

Cathodoluminescence (CL) measurements were used to evaluate the optical properties of the crystalline defects in the powders. Figure 9a illustrates the CL spectrum of the HAp nanofibers with seven components centered at 1.83, 2.15, 2.57, 2.95, 3.32, 3.67, and 4.05 eV, which we previously reported for HAp nanofibers, assigning an origin to native defects of hydroxyl vacancies (\({\text{V}}_{\text{OH}}\)) and oxygen vacancies (\({\text{V}}_{\text{O}}\)) in \({\text{O}}{\text{H}}^{-}\) and \({\text{P}}{\text{O}}_{4}^{3-}\) groups [31, 40]. Such components were calculated using Gaussian curves with a FWHM value of 0.4 eV to obtain values for the coefficient of determination, R2, between 0.9995 and 0.9999. Figure 9b–f display CL spectra obtained from the HAp:Cr powders, composed of the same seven components detected for HAp, revealing an increase in their panchromatic luminescence with chromium incorporation. This luminescence enhancement by the exclusive incorporation of Cr3+ ions in HAp demonstrates the feasibility of this biomaterial to be used as a luminescent marker in fluorescence microscopy and bioimaging during bone regeneration treatments, enabling the visualization of the implanted HAp interaction with the surrounding biological environment. CL measurements reveal additionally that the intensities of the 2.95 and 2.57 eV emissions increase more notably with Cr doping (Table 3). As illustrated in Fig. 10, the band at 2.57 eV corresponds to transitions between levels of hydroxyl vacancies next to an H+ interstitial (\({\text{V}}_{\text{OH}} \, - \, {\text{H}}_{\text{i}}\)) and \({\text{V}}_{\text{O}}\) in the \({\text{P}}{\text{O}}_{4}^{3-}\) groups [40, 55, 56], while the emission of 2.95 eV corresponds to transitions between energy levels generated by \({\text{V}}_{\text{OH}}\) and \({\text{V}}_{\text{O}}\) in the OH− groups [60]. The emission of 3.67 eV does not exhibit an increase in intensity with chromium doping (Table 3), even though this emission is generated by electronic transitions between the same energy levels of \({\text{V}}_{\text{OH}}\) that produce the 2.95 eV emission (Fig. 10). Thus, considering that the electronic capture cross-sectional area of point defects does not change by the incorporation of chromium in the HAp lattice, and that both emissions of 2.95 and 3.67 eV correspond to transitions between the same levels of \({\text{V}}_{\text{OH}}\), we infer that the increase of 2.95 eV emission can only be attributed to a rise in the density of electronic states associated with \({\text{V}}_{\text{O}}\) in the OH– groups. We estimated the proportion of oxygen vacancies in Cr-doped samples compared to the undoped HAp by calculating the intensity ratio of the 2.95 eV emission between the HAp:Cr and HAp samples (Table 3). Our findings show that the incorporation of Cr3+ ions into the HAp lattice increases the concentration of \({\text{V}}_{\text{O}}\) in the \({\text{O}}{\text{H}}^{-}\) groups by 1.2 and 2.5 times compared to undoped HAp. Previously, Bystrov et al. calculated the unit cell parameters and volume of HAp with one \({\text{V}}_{\text{O}}\) in the \({\text{O}}{\text{H}}^{-}\) group, reporting values of a = 9.3437 Å, c = 6.8463 Å, and V = 517.64 Å3 [57]. These were compared to stoichiometric HAp, which showed a = 9.3640 Å, c = 6.8621 Å, and V = 521.09 Å3. Given that one oxygen vacancy in a 44 atom HAp unit cell corresponds to a 2.2 at.% concentration (a relatively high concentration), the slight reduction in the a and c parameters and a similar unit cell volume, indicate the ability of the HAp structure to integrate oxygen vacancies with minimal distortion. This suggests a degree of flexibility withing the crystal structure, allowing for the incorporation of \({\text{V}}_{\text{O}}\) in the \({\text{O}}{\text{H}}^{-}\) group while maintaining overall geometry and ensuring structural stability. Furthermore, Bystrov et al. [58] also demonstrated that oxygen vacancies occur under equilibrium conditions in HAp as simple vacant oxygen sites with a neutral charge, and that the formation of \({\text{V}}_{\text{O}}\) in the \({\text{O}}{\text{H}}^{-}\) groups occurs by removal of neutral O from \({\text{O}}{\text{H}}^{-}\) leaving a hydride anion (\({\text{H}}^{-}\)). In addition, these authors report that oxygen vacancies generated in \({\text{O}}{\text{H}}^{-}\) groups exhibit lower formation energy than \({\text{V}}_{\text{O}}\) in \({\text{P}}{\text{O}}_{4}^{3-}\) groups. Therefore, to explain the generation of oxygen vacancies in the \({\text{O}}{\text{H}}^{-}\) groups of the HAp:Cr samples, we propose that during the ion-exchange process, pairs of extra electrons are generated by substituting Cr3+ with Ca2+ following the reaction:

$${\text{Ca}}_{10} \left( {{\text{PO}}_{4} } \right)_{6} \left( {{\text{OH}}} \right) + 2x{\text{ Cr}}^{3 + } \left( {{\text{aq}}} \right) \, \to \,{\text{Ca}}_{{\left( {10 - 2x} \right)}} {\text{Cr}}_{2x} \left( {{\text{PO}}_{4} } \right)_{6} \left( {{\text{OH}}} \right) + 2x {\text{Ca}}^{2 + } \left( {{\text{aq}}} \right) + 2x e^{ - }$$
(2)
Fig. 9
figure 9

Cathodoluminescence spectra of the a HAp, b HAp:Cr-1, c HAp:Cr-12, d HAp:Cr-24, e HAp:Cr-36, and f HAp:Cr-48 powders

Full size image
Table 3 Cathodoluminescence intensities of the 2.57, 2.95, and 3.67 eV emissions of the HAp and HAp:Cr powders
Full size table
Fig. 10
figure 10

Schematic representation of the electronic transitions for HAp [21]

Full size image

Afterward, the two free electrons can be captured by neutral oxygen atoms that have been removed from \({\text{O}}{\text{H}}^{-}\) groups to form O2− ions in the aqueous solution and \({\text{V}}_{\text{O}}\) in the HAp lattice, as follows:

$${\text{Ca}}_{{\left( {10 - 2x} \right)}} {\text{Cr}}_{2x} \left( {{\text{PO}}_{4} } \right)_{6} \left( {{\text{OH}}} \right) + 2x e^{ - } \, \to \,{\text{Ca}}_{{\left( {10 - 2x} \right)}} {\text{Cr}}_{2x} \left( {{\text{PO}}_{{4}} } \right)_{6} {\text{H}} + 2x {\text{O}}^{2 - } \left( {{\text{aq}}} \right) + x V_{{\text{O}}}$$
(3)

Here, the compound \({\text{Ca}}_{\left(\text{10-2}{\text{x}}\right)}{\text{Cr}}_{{2}{\text{x}}}{\left({\text{P}}{\text{O}}_{4}\right)}_{6}{\text{H}}\) represents the HAp with hydride anions (\({\text{H}}^{-}\)).

Figure 11a illustrates the photoluminescence (PL) spectrum obtained for the HAp powder with two broad primary emissions centered at 2.40 and 2.98 eV, which were deconvoluted using Gaussian curves with a FWHM value of 0.38 eV, and centered at 2.15, 2.40, 2.58, 2.98, and 3.31 eV. These components correspond well with the CL components shown in Fig. 9, except for the 2.40 eV band that corresponds to electronic transitions between levels generated by \({\text{V}}_{\text{Ca}}\) and \({\text{V}}_{\text{O}}\) in \({\text{P}}{\text{O}}_{4}^{3-}\) [60]. Figures 11b–f show the PL spectra obtained from the HAp:Cr powders, which were deconvoluted with the same components calculated for the HAp sample, and including the component of 1.85 eV detected by CL, which corresponds to electronic transitions from the HAp conduction band to levels generated by \({\text{V}}_{\text{OH}}\) [60]. Figures 11b–f reveal that the 3.31 eV band gradually increases in relative intensity by increasing the chromium concentration in the samples, which correlates with that observed for the 3.32 eV CL emission (Fig. 9b–f). Considering that this band corresponds to electronic transitions between the HAp conduction band and energy levels of \({\text{V}}_{\text{O}}\) in \({\text{P}}{\text{O}}_{4}^{3-}\) groups [60], the increase of the 3.31 eV emission must be associated with an increase in the amount of \({\text{V}}_{\text{O}}\) in \({\text{P}}{\text{O}}_{4}^{3-}\) in the doped samples.

Fig. 11
figure 11

Photoluminescence spectra of the a HAp, b HAp:Cr-1, c HAp:Cr-12, d HAp:Cr-24, e HAp:Cr-36, and f HAp:Cr-48 powders

Full size image

The magnetic response of the materials was analyzed by applying magnetic fields of ± 10 kOe at room temperature. Figure 12 illustrates the magnetization (M) with respect to the applied magnetic field (H) for the powders. Line 1 corresponds to the magnetic response of the HAp powder, which exhibits a diamagnetic behavior in agreement with previous reports [17, 31, 59, 60], showing a value for the mass magnetic susceptibility (χ) of − 2.6 × 10–6 emu/Oe⋅g. In diamagnetic materials, atoms have an even number of electrons in the partially filled shells and do not exhibit a net magnetic moment without an external magnetic field. Under a magnetic field, the spins of these electrons experience a precession motion, producing a magnetization in the opposite direction to the external magnetic field, described by the Langevin theory [61]. We expect diamagnetism in undoped HAp because it contains even numbers of electrons in the atomic orbitals of Ca2+, \({\text{O}}{\text{H}}^{-}\), and \({\text{P}}{\text{O}}_{4}^{3-}\) ions. The HAp:Cr-1 sample also exhibits a diamagnetic behavior (line 2), although with a susceptibility of − 7.7 × 10–7 emu/Oe⋅g, representing a decrease compared to the value observed for the HAp powder. This effect is due to the generation of local paramagnetism at the atomic sites of Cr3+ ions, which partially inhibits the native diamagnetism of HAp. The origin of such local paramagnetism is the unpaired electrons of Cr3+ in the 3d orbitals. According to Hund’s rule, these electrons occupy separate orbitals with their spin parallel before pairing up, leading to three unpaired electrons in the 3d subshell. Each unpaired electron has a magnetic moment associated with its spin, and in the presence of a magnetic field, these magnetic moments can align with the field, minimizing the energy. Figure 12 also shows that the HAp:Cr-12 powder exhibits a positive susceptibility of 9.2 × 10–8 emu/Oe⋅g (line 3), corresponding to a weak paramagnetic behavior of the sample, due to the increase of unpaired electrons in the Cr3+ ions that entirely inhibit the native diamagnetism of undoped HAp. Additionally, Fig. 12 shows that the HAp:Cr-24, HAp:Cr-36, and HAp:Cr-48 powders (lines 4–6) exhibit paramagnetism with magnetic susceptibility values increasing with chromium doping (Table 4). Notably, these results do not reveal a magnetic transition to superparamagnetism, as other authors have reported for HAp nanoparticles doped with Fe at 0.7 at.%, which corresponds to a similar concentration of the magnetic impurity added to the HAp:Cr-24 sample [62], demonstrating the absence of magnetic domains within the HAp:Cr nanofibers. Previously, we had reported that undoped HAp nanofibers exhibit electron paramagnetic resonance (EPR) signals identifying the formation of singly ionized oxygen vacancies (\({\text{V}}_{\text{O}}{\prime}\)) acting as spin ½ systems due to the presence of one unpaired electron [40], with a magnetic moment, \({\mu }_{s}\), given by:

$$\mu_{{\text{s}}} = \frac{1}{2}g_{{\text{s}}} \mu_{{\text{B}}}$$
(4)

where gs is the Landé g-factor for the spin with a value of about 2.0023 for a free electron, and μB is the magneton of Bohr defined as \(e\hslash /2{m}_{e}\). Thus, intrinsic magnetic moments of \({\text{V}}_{\text{O}}{\prime}\) can align with an external magnetic field, resulting in a paramagnetic behavior by minimizing the energy. Since CL and PL results demonstrate that HAp:Cr samples contain the same point defects we reported for undoped HAp [21], we infer that singly ionized oxygen vacancies are also present in HAp:Cr powders. Furthermore, since the PL and CL results reveal a notable increase of oxygen vacancies in HAp with Cr3+ doping, we propose that the formation of \({\text{V}}_{\text{O}}{\prime}\) is highly probable in our samples. Previously, Bystrov et al. [58] demonstrated that the replacement of \({\text{PO}}_{4}^{{3}-}\) by \({\text{PO}}_{3}^{{2}-}\) anions produces stable singly charged oxygen vacancies, which they labeled as \({\text{V}}_{\text{O}}^{+}\) to highlight the + 1 charge state for substitution \({\left[{\text{P}}{\text{O}}_{3}^{{2}-}\right]}_{{\text{PO}}_{4}^{{3}-}}^{+}\). Considering the charge neutrality of the HAp, this chemical mechanism be described as follows:

$${\text{PO}}_{4}^{3 - } \, \to \,{\text{PO}}_{3}^{3 - } + V_{{\text{O}}} \, \to \, {\text{PO}}_{3}^{2 - } + V_{O}^{ - } ,$$
(5)

where \({\text{V}}_{\text{O}}^{-}\) corresponds to a singly ionized oxygen vacancy, or \({\text{V}}_{\text{O}}{\prime}\) using Kröger-Vink notation. Equation (5) corresponds to the formation of \({\text{V}}_{\text{O}}\) in the \({\text{P}}{\text{O}}_{4}^{{3}-}\) groups, which alternatively can be rewritten considering that free electrons generated during the Cr3+ incorporation in HAp (see Eq. 2) can be trapped by \({\text{V}}_{\text{O}}\) forming singly ionized oxygen vacancies as follow:

$${\text{Ca}}_{{\left( {10 - 2x} \right)}} {\text{Cr}}_{2x} \left( {{\text{PO}}_{{4}} } \right)_{6} \left( {{\text{OH}}} \right) + 2x e^{ - } \, \to \, {\text{Ca}}_{{\left( {10 - 2x} \right)}} {\text{Cr}}_{2x} \left( {{\text{PO}}_{{4}} } \right)_{5} \left( {{\text{PO}}_{3} } \right)\left( {{\text{OH}}} \right) + V_{O} + 2x e^{ - } \to \,{\text{Ca}}_{{\left( {10 - 2x} \right)}} {\text{Cr}}_{2x} \left( {{\text{PO}}_{4} } \right)_{5} ({\text{PO}}_{3} )\left( {{\text{OH}}} \right) + V_{O}^{ - } + (2x - 1) e^{ - } .$$
(6)
Fig. 12
figure 12

B-M curves obtained from the HAp (line 1), HAp:Cr-1 (line 2), HAp:Cr-12 (line 3), HAp:Cr-24 (line 4), HAp:Cr-36 (line 5), and HAp:Cr-48 (line 6) powders

Full size image
Table 4 Magnetic susceptibility (χ) and maximum magnetization (Mmax) of the HAp and HAp:Cr powders by applying a magnetic field of 10 kOe
Full size table

Thus, we propose that the paramagnetic behavior of HAp:Cr is produced both by the unpaired 3d electrons of the Cr3+ ions and by the formation of \({\text{V}}_{\text{O}}{\prime}\). Table 4 lists the maximum magnetization (Mmax) and magnetic susceptibilities (χ of the HAp:Cr samples obtained for a magnetic field of 10 kOe, revealing a saturation magnetization of 0.0460 and 0.0470 emu/g for the HAp:Cr-36 and HAp:Cr-48 powders, respectively.

In summary, this study provides the first evidence of paramagnetism in hydroxyapatite (HAp) nanofibers doped with Cr3+ ions. The incorporation of Cr3+ leads to significant oxygen vacancy formation in \({\text{P}}{\text{O}}_{4}^{3-}\) and \({\text{O}}{\text{H}}^{-}\) groups. These vacancies become singly ionized, capturing electrons from Cr3+ ions within the HAp:Cr matrix. This mechanism underlies the observed paramagnetism, with HAp:Cr samples achieving saturation magnetization of 0.0460 emu/g at 10 kOe for Cr3+ concentrations above 2.9 at.%. We suggest that these singly ionized oxygen vacancies act as spin ½ systems, significantly enhancing the paramagnetic response of the Cr3+-doped HAp. This paramagnetic biomaterial is an exceptional candidate for bone regeneration treatments, offering potential applications for bone growth stimulation by magnetic fields and improved contrast in magnetic resonance imaging. Biocompatibility studies of HAp:Cr are still pending and are planned for future investigations.

Conclusions

Chromium-doped hydroxyapatite (HAp:Cr) nanofibers, synthesized via a modified hydrothermal method, exhibited a calcium deficiency and a hexagonal P63/m structure, with amorphization occurring at Cr concentrations exceeding 3.4 at.%. X-ray photoelectron spectroscopy confirmed Cr3+ substitution for Ca(II) within the HAp lattice. Raman spectroscopy detected standard HAp vibrational modes alongside new modes at 900 and 970 cm−1, intensified by increasing Cr content due to a Cr3+-induced v1 P–O bonds vibrational mode distortion. Cathodoluminescence indicated oxygen vacancy-related emissions at 2.95 and 2.57 eV, increasing with Cr concentration, suggesting generation of oxygen vacancies in hydroxyl groups. Photoluminescence also showed enhanced 3.32 eV emission with rising Cr doping, attributed to increased oxygen vacancies in \({\text{P}}{\text{O}}_{4}^{3-}\) groups. Vibrating sample magnetometry confirmed diamagnetism from pure HAp and generation of paramagnetism in HAp:Cr nanofibers, with chromium concentrations above 2.9 at.% reaching a magnetization saturation of 0.0460 emu/g at 10 kOe, attributable to the alignment of Cr3+ magnetic moments and singly ionized oxygen vacancies (\({\text{V}}_{\text{O}}{\prime}\)) acting as spin ½ systems. This paramagnetism is likely enhanced by the substitution of \({\text{PO}}_{4}^{{3}-}\) with \({\text{PO}}_{3}^{{2}-}\), leading to increased oxygen vacancies in the PO4 groups.