**The Influence of Ozonated Olive Oil-Loaded and Copper-Doped Nanohydroxyapatites on Planktonic Forms of Microorganisms**

**Wojciech Zakrzewski 1, Maciej Dobrzynski 2, Joanna Nowicka 3, Magdalena Pajaczkowska 3, Maria Szymonowicz 1, Sara Targonska 4, Paulina Sobierajska 4, Katarzyna Wiglusz 5, Wojciech Dobrzynski 6, Adam Lubojanski 1, Sebastian Fedorowicz 3, Zbigniew Rybak <sup>1</sup> and Rafal J. Wiglusz 4,\***


Received: 15 July 2020; Accepted: 29 September 2020; Published: 10 October 2020

**Abstract:** The research has been carried out with a focus on the assessment of the antimicrobial efficacy of pure nanohydroxyapatite, Cu2<sup>+</sup>-doped nanohydroxyapatite, ozonated olive oil-loaded nanohydroxyapatite, and Cu2+-doped nanohydroxyapatite, respectively. Their potential antimicrobial activity was investigated against *Streptococcus mutans, Lactobacillus rhamnosus*, and *Candida albicans*. Among all tested materials, the highest efficacy was observed in terms of ozonated olive oil. The studies were performed using an Ultraviolet–Visible spectrophotometry (UV-Vis), electron microscopy, and statistical methods, by determining the value of Colony-Forming Units (CFU/mL) and Minimal Inhibitory Concentration (MIC).

**Keywords:** Cu2<sup>+</sup> ions; ozonated olive oil; hydroxyapatite; antimicrobial activity; microorganisms

#### **1. Introduction**

Nowadays, the application of biomaterials is gaining popularity due to their high versatility. The development of modern medical science is based on the use of biomaterials, such as hydroxyapatite (HAp), to replace damaged hard tissue. Hydroxyapatite is the main inorganic component of bones and teeth, and it is related to the resorption and precipitation processes of calcium phosphates as well as the adsorption and formation of bones, dentine, and cementum [1]. Mainly, it crystallizes in the form of nanoplates or nanorods with an average size of approximately 50 nm × 25 nm × 2 nm [2,3]. The natural HAp is non-stoichiometric and poorly crystalline, and it contains numerous ionic substitutions,

e.g., Mg2+, Na+, K+, Sr2+, Zn2+, Mn2+, Cu2<sup>+</sup>, Co3 <sup>2</sup>−, F−, and Cl<sup>−</sup> [1,4]. Its synthetic form is isostructural and chemically similar to bone apatite and possesses a strong affinity for ion exchange, which causes its high bioactivity [5]. Its biological properties are determined by such parameters as Ca/P molar ratio, the type of ionic dopants in the crystal lattice, or particle size and morphology. Stoichiometric HAp has a typical lattice structure described as (A10(BO4)6C2), where A, B, and C are defined by Ca2<sup>+</sup>, PO4 3−, and OH−, respectively, with a calcium-to-phosphate ratio of 1.67 [6]. The HAp is non-immunogenic and non-toxic due to the outstanding bioactivity and biocompatibility. Moreover, synthetic hydroxyapatite has been widely applied as a bone substitute for the reconstruction of bone defects in maxillofacial surgery as well as orthopedics [1,7]. Furthermore, it can be used as a filler for repairing cavities on the enamel surface [8].

In orthopedics, bacterial adhesion on implant surfaces is the most predominant problem of post-surgical infections. Several studies have reported that the Ag+, Cu2<sup>+</sup>, and Zn2<sup>+</sup> ions are essential for preventing or minimizing initial microorganism adhesion [9–11]. Among them, the Cu2<sup>+</sup> ion occupies a prominent position as an antibacterial agent, because it reveals the highest inhibition of bacteria growth with simultaneous tolerable cytotoxicity for tissue cells, as was reported by Heidenau et al. [10]. The antimicrobial activity of cooper ions can be ascribed by several mechanisms. Under aerobic conditions, the Cu2<sup>+</sup> ion is proposed to be catalyzed producing hydroxyl radicals via the Fenton and Haber–Weiss reactions [12]. The possible mechanisms of action between the Cu2<sup>+</sup> ion-containing compounds and the microorganism are based on the structural damage of the cell membrane causing its permeability and finally cell death, the deactivation of proteins by binding metal ions, and the interaction with microbial nucleic acids preventing microbial replication [13,14]. In the presented study, it has been decided to choose 1 mol% Cu2<sup>+</sup> due to the fact that copper-doped nanohydroxyapatite (nHAp) retains an antimicrobial effect even at low Cu2<sup>+</sup> content, while its cytotoxicity against normal cells remains low [15]. Chui Ping Ooi et al. showed that the survival ratio of osteoblasts decreased as the Cu2<sup>+</sup> content increased [16], while Nam et al. [17] confirmed, that Cu2<sup>+</sup> concentration and contact time do not affect to the phase composition, but affect the crystal size and morphology. Moreover, from the physicochemical point of view, the crystal structure of the hydroxyapatite is stable at this (1 mol%) concentration of dopant. The presence of secondary phases could be observed with an increase of copper ions content in nHAp crystal lattice, as was presented by Sumathi Shanmugam et al. [13].

Cu2<sup>+</sup> ions have a strong activity against fungi and bacteria [18]. Moreover, the bactericidal effect of metal ions as well as nanoparticles has been attributed to their small size and a high surface-to-volume ratio, allowing close interaction with microbial membranes. Nanoparticles have coated surfaces and can be useful in various medical fields e.g., as cements or coatings in surgery, antimicrobial dressings, and actively targeted biomaterials [19,20].

The use of ozone in dentistry has increased in recent years due to its high oxidative power stimulating the immune response and blood circulation, together with its strong antimicrobial activity [21]. It has been demonstrated to be useful in controlling the physiology of microorganisms in dental plaque [22]. Ozone works synergistically—inducing the modification of intracellular contents and damaging the cytoplasmic membrane of cells [23]. Some medical products such as Ozonosept (see Section 2: Materials and Methods) contain ozone. It is fabricated during the process of ozonation of olive oil. Ozone is kept in the form of stable chemical compounds—ozonides. The ozonides show antibacterial, antifungal, and antiviral activity [24]. The antimicrobial activity of ozonated olive oil is related to the Criegee Mechanism i.e., a slow release of peroxides [25]. When an ozonide contacts with tissue, then carbonyl oxide reacts with water, and hydroxyhydroperoxide is produced. According to the Metrum Cryoflex leaflet, the Ozonosept has confirmed antimicrobial properties against *Staphylococcus aureus, Pseudomonas aeruginosa, Escherichia coli, Propionibacterium acnes*, and *Candida albicans*.

This study aimed to evaluate the selected materials against *Candida albicans, Streptococcus mutans,* and *Lactobacillus rhamnosus*. This set of microorganisms was used in our previous paper, Wiglusz et al. [26]. According to the literature, *C. albicans* has strong adhesive properties. Moreover, *S. mutans*

and *L. rhamnosus* are related to formation of a subgingival plaque. Moreover, these strains are referential in the case of in vitro studies for biomaterials.

The co-administration of nanoparticles and ozonated olive oil has not been extensively studied against microbial species isolated from persistent endodontic infections. Recent studies [27,28] have shown that the combination of ozonated olive oil and chitosan nanoparticles has a more significant killing effect—it prevents biofilm formation and eradicates resistant endodontic pathogens from root canals. The novelty of this work is its evaluation and comparison of the antimicrobial activity of the proposed materials on their own and in various compositions. Such a comparative study gives the opportunity to reveal the specificity of these materials toward various microorganisms including bacterial strains and pathogen yeast as well as ultimately leading to the better utilization of nanoparticles and ozonated olive oil.

#### **2. Materials and Methods**

#### *2.1. Synthesis of Nanocrystalline Hydroxyapatite*

The studies were carried out on the following materials: (i) nHAp, (ii) nHAp doped with Cu2<sup>+</sup> ions, (iii) nHAp with the addition of ozonated olive oil (Ozonosept, Metrum Cryoflex Sp. z o.o., Sp. K., Łomianki, Poland), and (iiii) nHAp doped with Cu2<sup>+</sup> ions and loaded with ozonated olive oil. The amount of ozone in olive oil was 100 mg/mL, as proven by Magnetic Resonance Spectroscopy (600 Mhz/16 tesla) (Bruker Corporation, Billerica, MA, USA) by the manufacturer.

The nHAp nanocrystals of Ca10(PO4)6(OH)2 and Ca9.9Cu0.1(PO4)6(OH)2 were synthesized by the wet chemistry method at the Institute of Low Temperature and Structure Research, Wroclaw, Poland. Analytical grade Ca (NO3)2·4H2O (99.0%, Alfa Aesar, Haverhill, MA, USA), NH4H2PO4 (99.0%, FlukaTM, Honeywell Specialty Chemicals Seelze GmbH., Seelze, Germany), and Cu (NO3)2·2.5H2O (98.0–102.0%, Alfa Aesar) were used as the starting materials. The pH was regulated by NH3·H2O (99%, Avantor Performance Materials Poland S.A., Gliwice, Poland). The concentration of Cu2<sup>+</sup> ions was 1 mol% to the overall molar content of calcium cations. All substrates were dissolved and mixed. The pH of the reaction mixture was adjusted to 10 with an ammonia solution. The reaction was performed at 100 ◦C for 60 min. The obtained product was washed several times with deionized water and dried at 70 ◦C for 24 h. The final product was heat-treated at 400 ◦C for 3 h.

#### *2.2. Characterisation*

The apatite crystal structure was confirmed by the X-ray diffraction technique (XRD). The XRD patterns were measured (five times for each samples) by using a PANalytical X'Pert Pro X-ray diffractometer (Malvern Panalytical Ltd., Malvern, UK) equipped with Ni-filtered Cu Kα1 radiation (Kα1 = 1.54060 Å, U = 40 kV, I = 30 mA). The measurements were done in the range of 3–70◦ (2θ). The Thermo Fisher Scientific Nicolet iS50 FT-IR spectrometer (Waltham, MA, USA) equipped with an Automated Beamsplitter exchange system (iS50 ABX containing DLaTGSKBr detector), which had a built-in all-reflective diamond Attenuated Total Reflectance (ATR) module (iS50 ATR), Thermo Scientific Polaris™ and HeNe laser, was used to record the FT-IR spectra (five times for each samples). The Fourier Transform Infrared (FT-IR) spectra in the mid-IR region (4000–400 cm<sup>−</sup>1) were measured using the standard KBr pellet method, while in the case of the far-IR region (400–100 cm<sup>−</sup>1), a Nujol suspension was used. Raman measurements (five times for each sample) were carried out with a Micro-Raman system Renishaw InVia Raman spectrometer equipped with a confocal DM 2500 Leica optical microscope (Wotton-under-Edge, Gloucestershire, UK), a thermoelectrically cooled Charge-Coupled Device CCD was used as a detector of the Raman spectra recorded. An argon laser operating at 831 nm was used. The chemical composition was performed by using an FEI Nova NanoSEM 230 scanning electron microscope (SEM, Hillsboro, OR, USA) with an energy-dispersive X-ray spectrometer (EDAX Genesis XM4). The Scanning Electron Microscopy coupled with Energy Dispersive X-ray Spectroscopy (SEM-EDS) was used for qualitative and quantitative analysis of

materials. The spectra were recorded three times for each sample, and the calculated value is an average result.

The tests were carried out on reference strains *Streptococcus mutans* (ATCC 25175), *Lactobacillus rhamnosus* (ATCC 9595), and *Candida albicans* (ATCC 90028).

The aim of the study was to conduct preliminary studies associated with the activity of the tested compounds against different microorganisms in their planktonic forms. The next stage will be related to an evaluation of the activity of the selected compounds against a mature structure of microbial biofilms.

#### *2.3. Spectrophotometric Examination*

A suspension of 0.5 McFarland density (1.5 <sup>×</sup> 108 CFU/mL in case of bacteria and 1.5 <sup>×</sup> <sup>10</sup><sup>6</sup> CFU/mL in case of fungi) in liquid medium Sabouraud Broth (Biomaxima) Brain Heart Infusion (BHI) Broth (Biomaxima) and De Man, Rogosa and Sharpe MRS Broth (Biomaxima) were prepared from fresh culture of the analyzed strains for *Candida albicans, Streptococcus mutans*, and *Lactobacillus rhamnosus*, respectively. First, 1 mL of the suspension prepared in this way was incubated with nHAp at a concentration of 0.1% and 1% (both pure and with an admixture of Cu2<sup>+</sup> and ozonated olive oil). According to previous studies including substituted hydroxyapatites with antibacterial properties it has been decided to choose two different concentrations of nHAp (0.1% and 1%) [15–17,28,29]. The samples were incubated at 37 ◦C (aerobic, anaerobic (GENbag anaer, Biomerieux), and microaerophilic (GENbag microaer, Biomerieux, Warsaw, Poland)) for 4 h and 24 h with shaking. After the incubation period, 100 μL of the suspension was transferred to the appropriate well of a 96-well plate according to the following scheme:


The reading was made on a Biochrom Asys UVM 340 spectrophotometer at 595 nm (Biochrom Ltd., Holliston, MA, USA).

#### *2.4. Determining the Value of the Colony-Forming Units, CFU*/*mL*

Suspensions of 0.5 McFarland density (1.5 <sup>×</sup> <sup>10</sup><sup>8</sup> CFU/mL in case of bacteria and 1.5 <sup>×</sup> 106 CFU/mL in case of fungi) in liquid Sabouraud, BHI, and MRS medium were prepared from fresh culture of the analyzed strains for *C. albicans*, *S. mutans*, and *L. rhamnosus*, respectively. In this way, 1 mL of the prepared suspension was incubated with nHAp at concentrations of 0.1% and 1% (pure, as well as doped with Cu2<sup>+</sup> ions and an addition of ozonated olive oil). The samples were incubated at 37 ◦C (aerobic, anaerobic (GENbag anaer, Biomerieux, Warsaw, Poland), and microaerophilic (GENbag microaer, Biomerieux)) for 24 h with shaking. After the incubation period, 100 μL of the suspension was withdrawn, and a series of dilutions were made in geometric progress (10<sup>−</sup>1–10−6). After plating on a solid medium (appropriate for the strain), the plates were incubated; then, the grown colonies were counted, and the value of the colony-forming units (CFU/mL) was determined. All tested samples were subjected to triplicate procedure.

Together with the test sample, a control test was done, which was a suspension of the analyzed strain. The antimicrobial properties of ozonated olive oil have also been evaluated. The proportions of the ozonated olive oil together with microbial strain were 1 mL of bacterial/fungal suspension and 1 mL of ozonated olive oil.

#### *2.5. Determining the Value of the Minimal Inhibitory Concentration, MIC*

First, 100 mL of the medium (appropriate for the strain) was applied to the wells of the 96-well plate. Then, 100 μL of the nHAp suspension (pure and doped with Cu2<sup>+</sup> ions and an addition of ozonated olive oil) was applied to the appropriate plate rows and diluted geometrically to a concentration range of 9.7–5000 μg/mL. After adding 20 μL of the diluted microorganism culture, the plate was incubated (37 ◦C; aerobic, anaerobic and microaerophilic). After the incubation period, the minimum inhibitory concentration value was read visually.

#### *2.6. Scanning Electron Microscopy*

Suspensions of 0.5 McFarland density (1.5 <sup>×</sup> <sup>10</sup><sup>8</sup> CFU/mL in case of bacteria and 1.5 <sup>×</sup> 106 CFU/mL in case of fungi) in liquid Sabouraud, BHI, and MRS medium were prepared from a fresh culture of the analyzed strains for *C. albicans*, *S.mutans*, and *L. rhamnosus*, respectively. First, 1 mL of the suspension prepared in this way was incubated with nHAp at a concentration of 0.1% and 1% (pure HAp as well as doped with Cu2<sup>+</sup> ions and the addition of ozonated olive oil). The samples were incubated at 37 ◦C (aerobic, anaerobic (GENbag anaer, Biomerieux), and microaerophilic (GENbag microaer, Biomerieux)) for 24 h with shaking. After the incubation period, 100 μL of the suspension was transferred to the appropriate well of a 12-well plate, fixed, sprayed with gold, and evaluated in a ZEISS scanning electron microscope model EVO LS15 (Carl Zeiss, Oberkochen, Germany).

#### *2.7. Statistical Methods*

For all quantitative features (number of colony-forming units, CFU/mL), their distribution was checked for compliance with the normal distribution. The conformity assessment was carried out with the Shapiro–Wilk test.

Qualitative variables (strains of microorganisms) are presented in the abundance tables (contingency) in the form of abundance (*n*) and proportion (%). The chi-squared test was used to assess the strength of the relationship between the two variables. In cases where the number expected in one of the tables (2 × 2) was less than 5, the Fisher's exact test was used. Mean values (±M) and standard deviations (±SD) were calculated for all measurable features. The homogeneity of variance was checked by the Bartlett and Levene test.

Analysis of variance (Anova) was used to compare the averages in several groups. Whether the analyzed feature in each of the examined groups had normal distribution and equal variances had been checked earlier. If the probability corresponding to the value of the Snedecor F distribution was lower than the assumed level of significance (*p* < 0.05), then multiple comparison tests (post hoc) were performed to determine which group significantly differs from the others. The Tukey test was used for this purpose.

The Statistica version 12.5 program (StatSoft, Tulsa, OK, USA) was applied for calculations and making charts.

#### **3. Results**

#### *3.1. Structural Analysis*

Structural analysis of the Ca10(PO4)6(OH)2 nanocrystals as well as nHAp doped with 1 mol% Cu2<sup>+</sup> ions was performed by using the XRD technique. The X-ray diffraction patterns of pure and Cu2+-doped nHAp are presented in Figure 1. As it can be seen, the observed XRD patterns are in good agreement with the reference hexagonal phase of nHAp (no. ICSD-26204) ascribed to the P63/m space group [30]. The successful replacement of calcium ions by cooper ions in the crystal structure was confirmed—no additional peaks originating from other phases were observed. The efficient substitution of Ca2<sup>+</sup> ions by Cu2<sup>+</sup> ions has been clearly confirmed by shifting the positions of the diffraction peaks. Hydroxyapatite doped with Cu2<sup>+</sup> ions exhibits a slight shift in the position of the (002) plane (c-plane) and (300) plane (a-plane), [13,28,31]. A shift toward higher 2θ angles was related to the

decrease in the cell parameters induced by the substitution of the bigger Ca2<sup>+</sup> cation (CN9 = 1.18 Å, Ca2<sup>+</sup> CN7 = 1.06 Å, where CN is coordination number) by the smaller Cu2<sup>+</sup> cation (CN7 = 0.73 Å) [28,32,33].

**Figure 1.** X-ray diffraction patterns of pure nanohydroxyapatite (nHAp) and nHAp doped with 1 mol% Cu2<sup>+</sup> after heat treatment at 400 ◦C with the indication of (002) and (300) planes shift induced by doping with Cu2<sup>+</sup> ions.

Moreover, the X-ray diffraction can be used to determine the presence of the OCP (octacalcium phosphate) phase. The OCP structure can be described as an alternative stacking of "apatite" layers and "hydrated" layers [34]. Most of the OCP reflections within the 2θ range of 10–60◦ overlapped with those belonging to the hydroxyapatite structure. However, three reflections are specific to OCP in the low 2θ range of 4–10◦, at 4.7◦, 9.5◦, and 9.8◦ with the relative intensities equal to 100%, 8%, and 8%, respectively [35]. In the case of the studied materials, Bragg peaks at very low 2θ angles, especially the most intense (100) line, were not observed. Meanwhile, the most characteristic diffraction peaks belonging to the hydroxyapatite structure were found at 2θ equal to 31.8◦, 32.2◦, 32.9◦, and 25.9◦.

#### *3.2. Infrared Spectra*

The infrared spectra of investigated materials are shown in Figure 2. The absorption bands have been ascribed based on literature data [36–38]. The peaks at 1045 cm−<sup>1</sup> and 1095 cm−<sup>1</sup> correspond to the antisymmetric triply degenerate stretching vibrations of phosphate groups (PO4 <sup>3</sup>−)ν3. The peak at 963 cm−<sup>1</sup> belongs to the symmetric non-degenerate stretching vibrations of phosphate groups (PO4 <sup>3</sup>−ν1), while the modes at 604 cm−<sup>1</sup> and 570 cm−<sup>1</sup> identify the triply degenerate vibration (PO4 <sup>3</sup>−)ν4. The presence of the absorption band at 634 cm−1, belonging to the librational mode of the –OH group, clearly indicates the nHAp structure. The peak observed at 3571 cm−<sup>1</sup> is related to the stretching mode of the –OH group. The broad absorption band with a maximum at 3430 cm−<sup>1</sup> corresponds to the typical vibrations of water molecules.

The ATR-IR absorption spectra (in the range of 4000–400 cm<sup>−</sup>1) of the ozonated olive oil and nHAp with the addition of ozonated olive oil as well as nHAp doped with Cu2<sup>+</sup> and loaded with ozonated olive oil are presented in Figure 3. The bands related to ozonated olive oil and hydroxyapatite have been marked with black and red dashed lines, respectively. According to the spectroscopic results, it has been proven that the ozonated olive oil was adsorbed on the obtained compounds. The most typical peak associated with ozonated olive oil, indicating the existence of an ozonide ion, is located at 1104 cm−<sup>1</sup> and is correlated with the ozonide CO stretching mode [39]. The peak at 3006 cm−<sup>1</sup> is associated with the C–H stretching vibration of the cis double bond. There are also two intense peaks at 2922 and 2853 cm<sup>−</sup>1, which correspond to the C–H asymmetric stretching vibrations of both –CH2 and –CH3 groups. The peak at 1742 cm−<sup>1</sup> demonstrates C=O vibrations. The modes around 1460 and

723 cm−<sup>1</sup> are assigned to the bending C–H vibration, while the peak at 1160 cm−<sup>1</sup> is attributed to the C–O bands [40].

**Figure 2.** FT-IR spectra of nHAp and nHAp doped with 1 mol% Cu2<sup>+</sup> with an indication of typical active vibrational bands.

**Figure 3.** FT-IR spectra of ozonated olive oil as well as nHAp and nHAp doped with 1 mol% Cu2<sup>+</sup> and loaded with ozonated olive oil.

#### *3.3. Micro-Raman Spectra*

The micro-Raman spectra of pure and Cu2+-doped nHAp were recorded and presented in Figure 4. The spectra of the Cu2<sup>+</sup>-doped nHAp contain four characteristic vibrational transitions of phosphate groups. The maximum of the most intense peak is located at 961 cm−<sup>1</sup> and is correlated with the symmetric stretching mode of the phosphate groups (PO4 <sup>3</sup>−)ν1. The three overlapping vibration modes at 1075, 1046, and 1028 cm−<sup>1</sup> are attributed to the asymmetric stretching of (PO4 <sup>3</sup>−)ν3. In the region of the (PO4 <sup>3</sup>−)ν<sup>2</sup> bending mode, there are two peaks at 592 and 580 cm−1. The positions of 452 and 430 cm−<sup>1</sup> are associated with (PO4 <sup>3</sup>−)ν<sup>4</sup> bending modes. The analysis of the Raman spectrum related to pure nHAp revealed one distinguishing peak at 961 cm−<sup>1</sup> associated with (PO4 <sup>3</sup>−)ν<sup>1</sup> vibration. Moreover, the peaks correlated with the ν2, ν3, and ν<sup>4</sup> vibrational transitions of phosphate groups were not clearly detected [16,19]. The spectra of the samples loaded with ozonated olive oil reveal additional bands at 1660 cm−<sup>1</sup> derived from aliphatic unsaturation, at 1442 cm−<sup>1</sup> associated with the deformation of the CH2 group, and at 1302 cm−<sup>1</sup> correlated to the twisting of the CH2 group [41].

**Figure 4.** The micro-Raman spectra of nHAp and nHAp doped with 1 mol% Cu2<sup>+</sup> as well as both materials loaded with ozonated olive oil.

#### *3.4. EDS Analysis*

The EDS spectra (Figure 5) recorded for the samples was applied to identify and quantify the elements in the nHAp:Cu2+. The resulting contents of Ca, Cu, and P in the studied material were 28.2 at%, 0.39 at%, and 18.1 at%, respectively. The calculated value of the (nCa + nCu)/nP ratio was equal to 1.5—close to the theoretical nCa/nP ratio. The calculated concentration of Cu2<sup>+</sup> ions was equal to 0.1 mol%, which stays in agreement with the theoretical value.

**Figure 5.** EDS spectrum of nHAp doped with 1 mol% Cu2<sup>+</sup> after heat treatment at 400 ◦C.

#### *3.5. Spectrophotometric Indication*

The range of reduction of viable *C. albicans* cells after 4 h incubation under modified nHAp ranged from 46 to 87% for a concentration of 0.1% and 44 to 84% for a concentration of 1%. In the case of *S. mutans* and *L. rhamnosus*, it was respectively 47–97% and 77–100%, as well as 38–52% and 24–54% for concentrations of 0.1% and 1%.

After 24 h incubation, the degree of reduction was in the range of 21–89% and 50–52% in the case of *C. albicans*, and 41–75% and 19–81% for *S. mutans*. In the case of *L. rhamnosus*, these values ranged from 11 to 62% and 5 to 88% for 0.1% and 1%, respectively. The results are shown in Tables S1 and S2.

#### *3.6. Determining the Value of the Minimal Inhibitory Concentration*

The Minimum Inhibitory Concentration (MIC) value of undoped and Cu2<sup>+</sup>-doped nHAp loaded with ozonated olive oil against *C. albicans* was >5000 and 5000 μg/mL, respectively. In the case of *S. mutans* MIC, undoped and Cu2+-doped nHAp was >5000 μg/mL, and in the case of Cu2+-doped nHAp loaded with ozonated olive oil, it reached 5000 μg/mL. In the case of *L. rhamnosus*, the value of the minimum growth inhibitory concentration for all analyzed compounds exceeded the value of 5000 μg/mL. The results are shown in Table 1.

**Table 1.** Values of Minimal Inhibitory Concentration (MIC) of nHAp for analyzed strains.


I—nHAp; II—Cu2+-doped nHAp; III—nHAp with the addition of ozonated olive oil; IV—nHAp doped with Cu2<sup>+</sup> and loaded with ozonated olive oil.

#### *3.7. Determining the Value of Colony-Forming Units, CFU*/*mL*

In the case of *C. albicans*, the range of Colony-Forming Units was 1 <sup>×</sup> 105–3.20 <sup>×</sup> 108 for 0.1% and 1.4 <sup>×</sup> 105–1.79 <sup>×</sup> 108 for 1%. In the case of *L. rhamnosus* and *S. mutans*, the range was 6.07 <sup>×</sup> 108–1.67 <sup>×</sup> <sup>10</sup>9, 1 <sup>×</sup> 103–3.36 <sup>×</sup> 109, 0–8.10 <sup>×</sup> 107, and 0–6.13 <sup>×</sup> 107, respectively. The results are presented in Tables S3–S5 and Figures 6–12. It has been shown that the HAp doped with Cu2<sup>+</sup> ions and the ozonated olive oil-loaded HAp doped with Cu2<sup>+</sup> ions have caused a readable reduction of CFU/mL for *Candida albicans* and *Lactobacillus rhamnosus.* Moreover, the highest value of CFU/mL has been observed when pure nHAp was used against the *Candida albicans* as well as Cu2<sup>+</sup> ions-doped HAp against *Lactobacillus rhamnosus.* Furthermore, the highest CFU/mL value has also been observed in the case of pure nHAp, while the lowest occurred with both nHAp with the addition of ozonated olive oil and nHAp doped with Cu2<sup>+</sup> ions and loaded with ozonated olive oil against the *Streptococcus mutans* strain. On the other hand, the CFU/mL had been evaluated by the lowest values in contact with 1 mol% Cu2<sup>+</sup>-doped nHAp, 1 mol% Cu2<sup>+</sup>-doped nHAp with the addition of ozonated olive oil, and 1 mol% Cu2<sup>+</sup>-doped nHAp loaded with ozonated olive oil against *Streptococcus mutans* and *Candida albicans*.

**Figure 6.** Average Colony-Forming Units (CFU)/mL values and standard deviation (M ± SD) for *Candida albicans* after contact with tested materials (C—control group, I—nHAp, II—Cu2+-doped nHAp, III—nHAp with the addition of ozonated olive oil, IV—nHAp doped with Cu2<sup>+</sup> and loaded with ozonated olive oil, V—ozonated olive oil) for weights of 0.1% and results of the analysis of variance (ANOVA) and multiple comparisons by a post-hoc test. Incubation took 24 h.

**Figure 7.** Average CFU/mL values and standard deviation (M ± SD) for *Candida albicans* after contact with the tested materials (C—growth control, I—nHAp, II—Cu2+-doped nHAp, III—nHAp with the addition of ozonated olive oil, IV—nHAp doped with Cu2<sup>+</sup> and loaded with ozonated olive oil, V—ozonated olive oil) for weights of 1% and results of the analysis of variance (ANOVA) and multiple comparisons by a post-hoc test. Incubation took 24 h.

**Figure 8.** Average CFU/mL value and standard deviation (M ± SD) for *Candida albicans* after temporary contact with olive ozone in groups differing in weight and significance of test result. Incubation took 24 h.

**Figure 9.** Average CFU/mL values and standard deviation (M ± SD) for *Lactobacillus rhamnosus* after temporary contact with the tested materials (C—growth control, I—nHAp, II—Cu2+-doped nHAp, III—nHAp with the addition of ozonated olive oil, IV—nHAp doped with Cu2<sup>+</sup> and loaded with ozonated olive oil, V—ozonated olive oil) for weights of 0.1% and results of analysis of variance (ANOVA) and multiple comparisons by a post-hoc test. Incubation took 24 h.

**Figure 10.** Average CFU/mL values and standard deviation (M ± SD) for *Lactobacillus rhamnosus* after temporary contact with the tested materials (C—growth control, I—nHAp, II—Cu2+-doped nHAp, III—nHAp with the addition of ozonated olive oil, IV—nHAp doped with Cu2<sup>+</sup> and loaded with ozonated olive oil, V—ozonated olive oil) for weights of 1% and results of analysis of variance (ANOVA) and multiple comparisons by a post-hoc test. Incubation took 24 h.

**Figure 11.** Average CFU/mL values and standard deviation (M ± SD) for *Streptococcus mutans* after temporary contact with the tested materials (C—growth control, I—nHAp, II—Cu2+-doped nHAp, III—nHAp with the addition of ozonated olive oil, IV—nHAp doped with Cu2<sup>+</sup> ions and loaded with ozonated olive oil, V—ozonated olive oil) for weights of 0.1% and results of the analysis of variance (ANOVA) and multiple comparisons by a post-hoc test. Incubation took 24 h.

**Figure 12.** Average CFU/mL values and standard deviation (M ± SD) for *Streptococcus mutans* after temporary contact with the tested materials (C—growth control, I—nHAp, II—Cu2+-doped nHAp, III—nHAp with the addition of ozonated olive oil, IV—nHAp doped with Cu2<sup>+</sup> and loaded with ozonated olive oil, V—ozonated olive oil) for weights of 1% and results of the analysis of variance (ANOVA) and multiple comparisons by a post-hoc test. Incubation took 24 h.

In the case of a sample with 0.1% nHAp, we observed statistically significant differences in the growth of the *C. albicans* colony for the growth control and all other materials (*p* < 0.001), pure nHAp (material I) and other materials (*p* < 0.001), and nHAp doped with Cu2<sup>+</sup> ions and other materials (*p* < 0.001). The differences between materials containing ozonated olive oil (materials III, IV and V) turned out to be insignificant (*p* > 0.05); see Table S3 and Figure 6.

In the case of a sample weight of 1% nHAp, it was observed that there were significant statistical differences in the growth of *C. albicans* for pure nHAp (material I) and materials containing ozonated olive oil (materials III, IV, V); see Table S4 and Figure 7.

Ozonated olive oil turned out to be the best material for *C. albicans* (smallest colony growth). After two measurements, the sample weight did not have a significant statistical impact on the colony concentration (Tables S3–S5 and Figure 8).

In the case of sample weights of 0.1% nHAp, it was shown that there were statistically significant differences in the growth of *L. rhamnosus* colonies for all materials except pure nHAp and nHAp doped with Cu2<sup>+</sup> ions and loaded with ozonated olive oil (*p* = 0.900). It was proven that the ozone olive has been the best material (Table S3 and Figure 9).

In the case of sample weights of 1% nHAp, it was observed that there were statistically significant differences in the growth of *L. rhamnosus* colonies for all materials except those containing ozonated olive oil (III, IV, and V), confirming that the ozonated olive oil has been the best material (Table S4 and Figure 10).

In the case of a weight sample of 0.1% nHAp, the slight increase of *S. mutans* colonies under the influence of Cu2+-doped nHAp, nHAp with the addition of ozonated olive oil and pure ozonated olive oil was observed, and the differences between them were insignificant (*p* = 1.000); see Table S3 and Figure 11.

In the case of a weight sample of 1% nHAp, the slight growth of *S. mutans* colonies under the influence of Cu2+-doped nHAp, nHAp with the addition of ozonated olive oil, as well as nHAp doped with Cu2<sup>+</sup> and loaded with the addition of ozonated olive oil and pure ozonated olive oil was observed, and differences between them were insignificant (*p* = 1.000); see Table S4 and Figure 12.

In all cases, the results of the variance analysis were statistically significant (for example, in Figure 12: *F* = 111.3 and *p* < 0.001. It means that the difference between at least one of the pairs of materials was significant). Moreover, the post-hoc tests were carried out Least Significant Difference (LSD) showing significant differences between the pairs of materials presented in Figure 12 (*p* < 0.001).

The difference between growth control and nHAp was significant (*p* < 0.001), and that was not the case in comparison with Cu2<sup>+</sup>-doped nHAp and nHAp with the addition of ozonated olive oil (*p* = 1.000).

In the case of a sample with a concentration of 0.1% nHAp, statistically significant differences were observed in the growth of *L. rhamnosus* colonies for all materials, except for nHAp and nHAp doped with Cu2<sup>+</sup> ions and loaded with ozonated olive oil (*p* = 0.900). Moreover, regarding the sample weights of 1.0% nHAp, it was observed that there were statistically significant differences in the growth of *L. rhamnosus* colonies for all materials, except for those containing ozonated olive oil (III, IV, and IV). Furthermore, for the weight sample of 0.1% nHAp, the slight increase of *S. mutans* colonies under influence of Cu2<sup>+</sup>-doped nHAp, nHAp with the addition of the ozonated olive oil, and the pure ozonated olive oil was shown, and the differences between those materials were insignificant (*p* = 1.000).

Additionally, for the sample with a concentration of 1% nHAp, the growth of *S. mutans* colonies under the influence of materials II, III, IV, and V was the smallest, and the differences between those materials were insignificant (*p* = 1.000).

Taking into account the above, the result of the analysis of variance was statistically significant in all cases (*F* = 111.2 and *p* < 0.001; meaning that the difference was significant at least between one of the pairs of materials). Therefore, post-hoc tests were carried out (LSD) with the purpose of showing between which pairs of materials the differences were significant. For example, the difference between growth control and nHAp was significant (*p* < 0.001), but that was not the case in comparison with Cu2<sup>+</sup>-doped nHAp and nHAp with the addition of ozonated olive oil (*p* = 1.000); see Tables S6–S8.

#### *3.8. Scanning Electron Microscopy*

In all cases, growth control is more abundant than when microorganisms are in contact with nHAp and Cu2+-doped nHAp (Figure 13). The action of nHAp doped with Cu2<sup>+</sup> ions definitely affects the reduction of the number of bacteria (*Lactobacillus rhamnosus*, *Streptococcus mutans*). Changes are visible on *Candida albicans* surface—it is not smooth and oval, but more angular instead. Additionally, it looks as if it was dehydrated.

**Figure 13.** Scanning electron microscopy: (**a**) Growth control of *Lactobacillus rhamnosus*; (**b**) *Lactobacillus rhamnosus* with nHAp; (**c**) *Lactobacillus rhamnosus* with Cu2<sup>+</sup>-doped nHAp; (**d**) Growth control of *Candida albicans*; (**e**) *Candida albicans* with nHAp; (**f**) *Candida albicans* with Cu2+-doped nHAp; (**g**) Growth control of *Streptococcus mutans*; (**h**) *Streptococcus mutans* with nHAp; (**i**) *Streptococcus mutans* with Cu2<sup>+</sup>-doped nHAp; Mag = 10,000×.

#### **4. Discussion**

Hydroxyapatite is an inorganic component of bones and teeth, acting as a scaffold and giving them mechanical properties. In addition, it stimulates bone development in small bone defects and can be used as a coating material for implants [42]. In a medium containing the nHAp particle, bacteria may adhere to the solid and co-aggregate. It has been reported that biofilms that could be formed at 15 min after inoculation on nHAp disks consist mainly of single, non-aggregated cells [43].

The relationship between the size of nHAp and bacterial adhesion is crucial because of an effect on slower plaque formation. The nHAp scale allows having enhanced physical and chemical properties, including increased wettability, roughness, and adsorption of proteins [44]. Non-aggregated and condensed nHAp particles adsorb on bacterial surfaces in vitro [45]. They interact with bacteria and thus reduce their adhesion. Severin AV et al. [46] investigated the interaction of the nHAp nanocrystals with *Staphylococcus aureus* bacteria. Moreover, the nHAp nanocrystallites adhere to the surface of bacteria, significantly reducing their ability to form colonies.

The activity of nHAp against the planktonic form of the chosne microorganisms has been evaluated in this study. The degree of reduction of viable cells by pure nHAp (% viable microorganisms) was minimal after using pure nHAp. However, it is worth noting that with increasing exposure time as well as a concentration of pure nHAp (material I), the reduction of *C. albicans* initially increased at a concentration of 0.1% nHAp from 1% after 4 h to 14% after 24 h and at a concentration of 1% nHAp from 9% after 4 h to 37% after 24 h. In the case of other microorganisms, no relationship was found. Whereas, after doping or mixing nHAp with other reagents, a significant difference in the reduction of cells viability was observed in groups of nHAp doped with Cu2<sup>+</sup> ions, loaded with ozonated olive oil, or both. After 4 hours incubation, the reduction range of viable *C. albicans* cells for pure nHAp was 1% at a concentration of 0.1% and 9% for a concentration of 1%.

The nHAp appeared to enhance biofilm formation by increasing glucosyltransferase transcription, which resulted in an increase in the production of insoluble glucans. Since the demineralization of nHAp in enamel caused by acids, including those produced by bacteria in the plaque, is important in the development of dental caries, nHAp is used in toothpastes for the remineralization of enamel [47]. In the current study, we have examined the effect of nHAp on the growth of *S. mutans* in two different media and a nutrient-rich environment. While exploring the extinction value measured as the percentage of cell reduction, it is observed that the antimicrobial activity of nHAp combined with Cu2<sup>+</sup> ions is higher than that of the pure nHAp, although it shows less efficacy than nHAp with the addition of ozonated olive oil.

The nHAp doped with Cu2<sup>+</sup> ions is distinct from other materials such as nHAp, nHAp with ozonated olive oil, as well as nHAp with Cu2<sup>+</sup> ions with ozonated olive oil and ozone olive alone, for the reason that it is more efficient after 4 hours than after 24 h in both concentrations of nHAp (0.1% and 1%), unlike the other materials. It significantly reduced microorganism growth.

With regard to nHAp containing Cu2<sup>+</sup> ions and ozonated olive oil, which is nHAp doped with Cu2<sup>+</sup> ions and loaded with olive oil, it is apparently more effective than other materials in reducing the number of microorganisms. The only exception is *L. rhamnosus* after 4 h incubation, compared to nHAp with ozonated olive oil. nHAp at a concentration of 0.1% is the most efficient in the reduction of *S. mutans* in nHAp doped with Cu2<sup>+</sup> ions, similarly to nHAp with ozonated olive oil. A similar effect is obtained regarding nHAp doped with Cu2<sup>+</sup> ions, where the lack of growth can be observed. Once again, the highest efficiency was detected in the case of *S. mutans*, which is the same in materials with ozonated olive oil. nHAp doped with Cu2<sup>+</sup> ions caused a statistically insignificant reduction of growth in the *L. rhamnosus* colony and *C. albicans.* Other studies also prove the antimicrobial properties of Cu2<sup>+</sup> ions [48,49].

Studies prove a wide spectrum of activity that increases the field of application of nHAp doped with Cu2<sup>+</sup> ions. It is suggested that it would have a wide spectrum of future use, for instance, in orthopaedics and bone prosthesis or dentistry and teeth implants [13,50].

Virgin olive oil has an abundance of unsaturated fatty acids due to an especially high content of oleic acid, which is a monounsaturated omega−9 (*n*−9) fatty acid. The process of ozonization oxides unsaturated bonds with the simultaneous formation of peroxidic substances, which results in the higher antifungal and antibacterial potential of ozonated substances [51].

The ozonated olive oil oil and its antimicrobial properties had been tested in some clinical trials concerning different diseases, for instance, in the gynecological or dentistry field. In terms of vulvovaginal candidiasis—that is an inflammation of vagina often caused by *Candida* species such as *Candida albicans* or NCAC (*non-Candida albicans Candida* species) such as *Candida glabrata, Candida krusei, Candida parapsilosis, Candida tropicalis*—it was found that the ozonated olive oil is equally effective as clotrimazole in a significant reduction of symptoms. Moreover, it also led to a negative culture growth, and there were no significant differences in terms of reducing itching and leukorrhea, although the burning sensation was diminished more effectively by clotrimazole [52].

Other studies provide an evaluation of ozonated olive oil in the treatment of chronic periodontitis, which is a disease caused mainly by bacteria, resulting in an inflammatory process of gums and tissues of the oral cavity. Although the clinical study was limited in terms of patient numbers, it was found that ozonated olive oil was effective not only as an adjunct to scaling and root planning but also in monotherapy. Nevertheless, when used as an adjunctive therapy, the patients' dentinal hypersensitivity had significantly risen [53]. Those findings suggest a high level of efficacy of ozonated olive oil as an antimicrobial agent, which coincides with the presented results. Its antifungal and antibacterial potential had been tested in laboratory conditions as well as in the clinical trials [54].

According to the obtained results on the efficacy of the acquired materials containing ozonated olive oil (nHAp mixed with ozonated olive oil, mixed with olive oil and doped with Cu2<sup>+</sup> ions, pure ozonated olive oil, respectively), their addition to microbiological samples causes the highest percentage reduction when compared to materials without ozonated olive oil (I and II). When analyzing the given data, there is an obvious conclusion that the highest efficacy of olive is presented in samples of *S. mutans*, especially in nHAp of a concentration of 0.1% for nHAp with the addition of ozonated olive oil, and also in 1% nHAp concentration for nHAp doped with Cu2<sup>+</sup> ions and ozonated olive oil. The total growth reduction of *S. mutans* strains is visible in nHAp doped with Cu2<sup>+</sup> ions and ozonated olive oil 4 h after application, as nearly 100% efficacy can be observed; however, in nHAp with ozonated olive oil, it occurs after 24 h. A complete reduction of *S. mutans* cells was observed after incubating this bacterial strain in the presence of ozonated olive oil. This study allows deducing that the *S. mutans* percentage reduction can reach the highest value among the measured microorganisms when it is mixed with nHAp doped with Cu2<sup>+</sup> ions and loaded with ozonated olive oil. In the case of ozonated olive oil, the lowest percentage reduction value of *S. mutans* is present 24 h after the use of 0.1% nHAp with the addition of ozonated olive. Additionally, olive efficacy was also measured by the number of colonies concentration (thousand CFU/mL) concerning *S. mutans*. The value was the same—0.018 ± 0.008 when the species were mixed with selected materials at a weight of 1% (nHAp with ozonated olive oil, nHAp with Cu2<sup>+</sup> and ozonated olive oil, ozonated olive oil). Interestingly, when the CFU/mL of *S. mutans* was compared after contact with selected materials at a weight of 0.1%, each material concerning olive lacked growth except for nHAp doped with Cu2<sup>+</sup> ions and ozonated olive oil, which had 2 <sup>×</sup> 106 <sup>±</sup> 0.01.

The lowest percentage reduction value of ozonated olive oil inflicts *L. rhamnosus* when it is mixed with 0.1% nHAp with the addition of ozonated olive oil. Even when mixed with pure ozonated olive oil, *L. rhamnosus* represents the lowest percentage reduction among every tested microorganism strain. The CFU/mL of *L. rhamnosus* had the same value as *S. mutans* in most cases (0.018 ± 0.008), but this was only after contact with pure ozonated olive oil at 0.1% sample weight. On the other hand, for nHAp with ozonated olive oil as well as nHAp doped with Cu2<sup>+</sup> ions and ozonated olive oil, the values were higher, ranging from 6.07 <sup>×</sup> 108 to 1.12 <sup>×</sup> <sup>10</sup>9.

The colony-forming unit of *L. rhamnosus* is less optimistic after contact with ozonated olive oil materials (nHAp with ozonated olive oil, nHAp with Cu2<sup>+</sup> with ozonated olive oil, and ozonated olive oil alone). At a sample weight of 1%, values ranged between 1 <sup>×</sup> 103 and 1 <sup>×</sup> 106.

By analyzing data acquired by the authors in the study, it can be concluded that the percentage reduction of the measured microorganisms and CFU/mL. have the most positive results for materials containing ozonated olive oil (nHAp with the addition of ozonated olive oil, nHAp doped with Cu2<sup>+</sup> ions and loaded with ozonated olive oil, ozonated olive oil). The percentage reduction in value after contact with pure ozonated olive oil is constantly high, ranging between 71 and 99%, and the highest results are obtained after 24 h of cultivation, ranging between 94 and 99% (see Tables S9 and S10). Similar results concerning a high efficacy of ozonated olive oil against these microorganisms were obtained by several other authors [54,55]. The CFU/mL of olive itself, with materials at a weight of 0.1%, has the best efficacy for each measured microorganism. On the other hand, at a sample weight of 1% nHAp, the same result is acquired only in the case of *S. mutans*, while the worst result is obtained for *L. rhamnosus*—1 <sup>×</sup> 106 <sup>±</sup> 10.14.

Nanohydroxyapatite doped with Cu2<sup>+</sup> ions or ozonated olive oil may limit the oral microbial activity. Moreover, it is successfully applied in dentine hypersensitivity treatment and maxillofacial bones regeneration. Doping it with Cu2<sup>+</sup> ions or ozonated olive oil may enhance its antimicrobial characteristics and limit the postoperative complications.

#### **5. Conclusions**

Calcium hydroxyapatite and calcium hydroxyapatite doped with Cu2<sup>+</sup> ions have been successfully synthetized by using the wet chemistry method. The obtained nanocrystals have been functionalized with ozonated olive oil, which resulted in the formation of a novel medical composition. In vitro screening of microorganism strains according to their activity in various experimental conditions may be a valuable method that could precede clinical efficacy treatments. The highest efficacy was observed for ozonated olive oil and nanocrystalline Cu2<sup>+</sup>-doped nHAp followed by undoped nHAp. The obtained results indicate that 10 times higher concentrations of pure as well as doped nHAp have better antimicrobial activity. *Streptococcus mutans* had the highest sensitivity, while *Lactobacillus rhamnosus* had the lowest. Pure ozonated olive oil had the highest antimicrobial efficacy.

In the next stage of the study, the authors plan to evaluate the activity of the nHAp against the biofilms in order to complete a whole view of the bacteria cell properties as well as the effectiveness of the antimicrobial compounds.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/2079-4991/10/10/1997/s1. Table S1: The effect of pure and doped nHAp on the analysed strains (optical density (OD) 595 nm); Table S2: Inhibition of the growth of microbial cells, %; Table S3: Average CFU/mL and standard deviation (M ± SD) for the tested strains with selected materials at a concentration of 0.1% nHAp; Table S4: Average CFU/mL and standard deviation (M ± SD) for the tested strains with selected materials with concentration of 1% nHAp; Table S5: Evaluation of the antimicrobial properties of ozonated olive oil (Inhibition % equals 99.99% for each strain); Table S6: Results of comparisons of *Candida albicans* colonies after contact with tested materials (C—growth control, I—nHAp, II—Cu2<sup>+</sup>-doped nHAp, III—nHAp with the addition of ozonated olive oil, IV—nHAp doped with Cu2<sup>+</sup> ions and loaded with ozonated olive oil, V—ozonated olive oil) for weight samples of 0.1 and 1% (M—average value); Table S7: Results of comparisons of *Lactobacillus rhamnosus* colonies after contact with tested materials (C—growth control, I—nHAp, II—Cu2<sup>+</sup>-doped nHAp, III—nHAp with the addition of ozonated olive oil, IV—nHAp doped with Cu2<sup>+</sup> ions and loaded with ozonated olive oil, V—ozonated olive oil) for weight samples of 0.1 and 1% (M—average value); Table S8: Results of comparisons of *Streptococcus mutans* colonies after contact with tested materials (C—growth control, I—nHAp, II—Cu2<sup>+</sup>-doped nHAp, III—nHAp with the addition of ozonated olive oil, IV—nHAp doped with Cu2<sup>+</sup> ions and loaded with ozonated olive oil, V—ozonated olive oil) for weight samples of 0.1 and 1% (M—average value); Table S9: Influence of 0.1% hydroxyapatite on microbial strains: *C. albicans*, *L. rhamnosus* i *S. mutans* (Inhibition growth%); Table S10: Influence of 1% hydroxyapatite on microbial strains: *C. albicans*, *L. rhamnosus* i *S. mutans* (Inhibition growth%).

**Author Contributions:** R.J.W. and M.D. conceived and designed the experiments in addition to analyzing all data; W.Z., S.T., P.S. and K.W. designed the experiments in addition to analyzing data; Z.R., W.D., A.L., S.F. and M.S. contributed reagents/materials/analysis tools and analyzing data; R.J.W., M.D., M.S. and Z.R. participated in funding acquisition; J.N. and M.P. designed and performed the microbiological experiments and analyzed data, and all authors contributed to the writing of the paper. All authors have read and agreed to the published version of the manuscript.

**Funding:** Financial support of the National Science Centre in the course of realization of the Projects "Preparation and characterization of biocomposites based on nanoapatites for theranostics" (no. UMO-2015/19/B/ST5/01330); "Elaboration and characteristics of biocomposites with anti-virulent and anti-bacterial properties against *Pseudomonas aeruginosa*" (no. UMO-2016/21/B/NZ6/01157) and "Preparation and investigation of multifunctional biomaterials based on nanoapatites for possible application in bone tumor treatment" (no. UMO-2017/27/N/ST5/02976).

**Acknowledgments:** The work was created as part of the cooperation of the Student Scientific Association of Microbiologists and the Scientific Association of Experimental Surgery and Biomaterials Research in Wroclaw Medical University, Wroclaw, Poland.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Article* **Investigation of Physicochemical Properties of the Structurally Modified Nanosized Silicate-Substituted Hydroxyapatite Co-Doped with Eu3+ and Sr2+ Ions**

**Sara Targonska \* and Rafal J. Wiglusz \***

Institute of Low Temperature and Structure Research, Polish Academy of Sciences, Okolna 2, 50-422 Wroclaw, Poland

**\*** Correspondence: s.targonska@intibs.pl (S.T.); r.wiglusz@intibs.pl (R.J.W.); Tel.: +48-071-3954-159 (R.J.W.)

**Abstract:** In this paper, a series of structurally modified silicate-substituted apatite co-doped with Sr2+ and Eu3+ ions were synthesized by a microwave-assisted hydrothermal method. The concentration of Sr2+ ions was set at 2 mol% and Eu3+ ions were established in the range of 0.5–2 mol% in a molar ratio of calcium ion amount. The XRD (X-ray powder diffraction) technique and infrared (FT-IR) spectroscopy were used to characterize the obtained materials. The Kröger–Vink notation was used to explain the possible charge compensation mechanism. Moreover, the study of the spectroscopic properties (emission, emission excitation and emission kinetics) of the obtained materials as a function of optically active ions and annealing temperature was carried out. The luminescence behavior of Eu3+ ions in the apatite matrix was verified by the Judd–Ofelt (J-O) theory and discussed in detail. The temperature-dependent emission spectra were recorded for the representative materials. Furthermore, the International Commission on Illumination (CIE) chromaticity coordinates and correlated color temperature were determined by the obtained results.

**Keywords:** spectroscopy; nanocrystallites; silicate-substituted hydroxyapatite; Eu3+ and Sr2+ ion co-doping; microwave-assisted hydrothermal method

#### **1. Introduction**

Nanotechnology has exerted a considerable impact on a vast number of scientific fields in the last decade. Studies on the preparation and characterization of apatite materials have been influenced by this. Nanoapatites are the focus of great research interest due to their biocompatible and nontoxic properties to encourage bone and tissue filling. The structure of the apatite allows for various modifications, providing the opportunity to create a material with intentional and targeted properties. Moreover, two unequal calcium positions can be substituted by ions with +1, +2 or +3 charges, such as Sr2+, Ba2+, K+, Na+, Mn2+, Li+, Mg2+ as well as lanthanide ions, etc. [1–3].

There is particular interest in the characterization of nanoapatite materials doped with optically active ions. Consequently, it is possible to successfully replace Ca2+ ion by Eu3+ [1,4,5], Ce3+ [6], Tb3+ [7], Dy3+ [8,9], Nd3+ [9,10], Sm3+ [8,9] ions to obtain materials with characteristic emission in red [1,4], green [11,12], violet [6] as well as blue [6] spectral regions. Recently, the apatite matrix has been the focus of attention and investigated in terms of white light emission. Promising materials are co-doped with Dy3+, Li+ and Eu3+ [13] or La3+, Dy3+ and Sr3+ ions [8]. The vast number of possibilities and favorable results encourage the detailed investigation of a variety of apatite modifications.

Apatite-based structures can tolerate numerous ionic substitutions in order to improve their properties for medical application. Recent research has shown that the combination of silica and strontium co-doping may improve their properties for medical application [14]. In vivo observations as well as in vitro studies have exposed the beneficial effects of using silica-based materials for bone treatment. It has been shown that silica promotes prolyl

**Citation:** Targonska, S.; Wiglusz, R.J. Investigation of Physicochemical Properties of the Structurally Modified Nanosized Silicate-Substituted Hydroxyapatite Co-Doped with Eu3+ and Sr2+ Ions. *Nanomaterials* **2021**, *11*, 27. https:// doi.org/10.3390/nano11010027

Received: 18 November 2020 Accepted: 16 December 2020 Published: 24 December 2020

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional clai-ms in published maps and institutio-nal affiliations.

**Copyright:** © 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

hydroxylase, stimulates the enzyme involved in collagen synthesis and participates in the proliferation and differentiation of bone mesenchymal stem cells and osteoblasts [15,16]. The synergy of great luminescence properties with bioactivity may result in obtaining a new group of specific materials dedicated to bioimaging and regeneration.

In this paper, we show the physicochemical characterization of silicate-substituted hydroxyapatite co-doped with Sr2+ and Eu3+ ions. Considerable attention has been paid to the luminescence properties, including emission and excitation spectra depending on Eu3+ ion concentration and heat-treating temperature as well as the influence of ambient temperature.

#### **2. Materials and Methods**

#### *2.1. Synthesis of the Co-Doped Materials*

Synthesizing of silicate-substituted hydroxyapatite co-doped with Eu3+ and Sr2+ ions involved a hydrothermal process. As substrates, the following were used: Ca(NO3)2·4H2O (99.0–103.0% Alfa Aesar, Haverhill, MA, USA), (NH4)2HPO4 (>99.0% Acros Organics, Schwerte, Germany), Eu2O3 (99.99% Alfa Aesar, Haverhill, MA, USA), Sr(NO3)2 (99.0% min Alfa Aesar, Haverhill, MA, USA) and tetraethyl orthosilicate TEOS (>99% Alfa Aesar, Haverhill, MA, USA). The concentration of strontium ions for all obtained materials was fixed to 2 mol% in a ratio of calcium ion molar content. Moreover, the concentration of optical active Eu3+ ions was set to 0.5; 1.0 and 2.0 mol% in a ratio to the calcium ion molar content. A stoichiometric number of substrates were dissolved separately in deionization water (see Table S1). A stoichiometric amount of Eu2O3 was digested in an excess of HNO3 (65% suprapure Merck KGaA, Darmstadt, Germany) to generate water-soluble Eu(NO3)3·xH2O. Afterwards, all starting substrates were added and mixed into a Teflon vessel. The ammonia solution (NH3·H2O 25% Avantor, Poland) was used to obtain a pH level of around 10. The hydrothermal process was conducted in a microwave reactor (ERTEC MV 02-02, Wrocław, Poland) for 90 min at elevated temperature (250 ◦C) and under autogenous pressure (42–45 bar). The achieved materials were centrifuged, cleaned by deionization water several times and dried for 24 h. Then, the obtained materials were heat-treated in the temperature range of 400–600 ◦C for 3 h, increased in steps of 3.3 ◦C/min.

#### *2.2. Physical–Chemical Characterization*

X-ray powder diffraction studies were carried out using a PANalytical X'Pert Pro X-ray diffractometer (Malvern Panalytical Ltd., Malvern, UK) equipped with Ni-filtered Cu *Kα*<sup>1</sup> radiation (*Kα*<sup>1</sup> = 1.54060 Å, *U* = 40 kV, *I* = 30 mA) in the *2*θ range of 10◦–70◦. XRD patterns were analyzed by Match! software version 3.7.0.124.

The surface morphology and the element mapping were assessed by a FEI Nova NanoSEM 230 scanning electron microscope (SEM, Hillsboro, OR, USA) equipped with EDS spectrometer (EDAX GenesisXM4) and operating at an acceleration voltage in the range 3.0–15.0 kV and spots at 2.5–3.0 were observed. EDX analysis was carried out to confirm the chemical formula.

The Thermo Scientific Nicolet iS50 FT-IR spectrometer (Waltham, MA, USA) equipped with an Automated Beamsplitter exchange system (iS50 ABX containing DLaTGS KBr detector), built-in all-reflective diamond ATR module (iS50 ATR), Thermo Scientific Polaris™, was used to record the Fourier-transformed infrared spectra. As an infrared radiation source, we used the HeNe laser. FT-IR spectra of the powders were recorded in KBr pellets at 295 K temperature in the middle infrared range, from 4000 to 500 cm<sup>−</sup>1, with a spectral resolution of 2 cm<sup>−</sup>1.

#### *2.3. Spectroscopy Properties*

The emission, excitation emission spectra and luminescence kinetics were recorded by an FLS980 Fluorescence Spectrometer (Edinburgh Instruments, Kirkton Campus, UK) from Edinburgh Instruments equipped with a 450 W Xenon lamp and a Hamamatsu

R928P photomultiplier. The hydroxyapatite powders were placed into a quartz tube. The excitation of 300 mm focal length monochromator was in Czerny–Turner configuration. All spectra were corrected during measurement according to the characteristics of the intensity of the excitation source. All spectra were recorded at room temperature. The spectral resolution of the excitation and emission spectra was 0.1 nm. Excitation spectra were recorded, monitoring the maximum of the emission at 618 nm that relates to the 5D0 → 7F2 transition, and emission spectra were recorded upon excitation wavelength at 394 nm. Before analysis, the emission spectra were normalized to the 5D0 → 7F1 magnetic transition. The luminescence kinetics profiles were recorded according to the 5D0 → 7F2 electric dipole transition.

Temperature-dependent emission spectra were recorded using the laser diode (*λexc* = 375 nm), and as an optical detector, we used the Hamamatsu PMA-12 photonic multichannel analyzer (Hamamatsu Photonics K.K., Hamamatsu City, Japan). The presented emission spectra are the average result of 15 measurements with an exposure time of 500 ms.

#### **3. Results and Discussion**

#### *3.1. X-ray Diffraction*

The formation of the silicate-substituted hydroxyapatite crystalline nanopowders was investigated by the XRD measurements and is shown in Figure 1 as a function of the concentration of optically active ions as well as heat-treating process. All samples prepared via the hydrothermal method have shown detectable crystallinity for the entire range of sintering temperatures (400–600 ◦C for 3 h). The presence of the single phase of the final products was confirmed by the reference standard of hexagonal strontium-substituted hydroxyapatite ICSD-75518 [17]. No other phase was detected in the studied powders, indicating that dopant ions were completely dissolved in the silicate-substituted host lattice.

Structural refinement was carried out using the Maud software version 2.93 [18,19] and was based on apatite crystals with a hexagonal structure using better approximations, as well as indexing of the crystallographic information file (CIF). The formation of the hexagonal phase, as well as the successful incorporation of Eu3+ and Sr2+ ions into the apatite lattice, was confirmed by the results. The average grain sizes of silicate-substituted hydroxyapatite nanopowders were in the range of 16 to 56 nm (see Figure S2 and Table S2). The representative SEM image is presented in Figure S1a. The effect of dopant ion substitution is confirmed by EDS measurements (see Figure S1b).

The most intense diffraction peaks corresponding to hydroxyapatite structures are located at 25.9◦ (002), 31.7◦ (211), 32.2◦ (112), 32.9◦ (300) and 34.0◦ (202), assigned to the crystallographic planes in brackets. In the obtained materials, Eu3+ and Sr2+ ions replaced Ca2+ ions. In the apatite host lattice, Ca2+ ions are located in two different sites with various chemical and structural environments, Ca(1) and Ca(2) sites with C3 and CS symmetry, respectively. The Ca(1) site is surrounded by nine oxygen atoms coming from PO4 3− groups, which formed a tricapped trigonal prism with formula CaO9. The Ca(2) site is an irregular polyhedron with formula CaO6OH formed by six oxygen atoms from PO4 <sup>3</sup><sup>−</sup> and one hydroxyl group [3,20]. The difference between the ionic radii of trivalent europium and divalent strontium ions permit the occupation of two possible crystallographic positions of Ca2<sup>+</sup> ions in the apatite host lattice (Ca2+ (CN9) = 1.18 Å, Eu3+ (CN9) = 1.12 Å, Sr2+ (CN9)−1.31 Å (Ca(1) site); Ca2+ (CN7) = 1.06 Å and Eu3+ (CN7) = 1.01 Å, Sr2+ (CN7)−1.21 Å (Ca(2) site)) [5,21].

**Figure 1.** X-ray diffraction pattern of silicate-substituted hydroxyapatite co-doped with 2 mol% Sr2+ and x mol% Eu3+ ions.

#### *3.2. Kröger–Vink Notation*

The cationic vacancies formed by replacing Ca2+ ions with Eu3+ ions with higher charge could be balanced by SiO4 <sup>4</sup><sup>−</sup> ions substituted into the PO4 <sup>3</sup><sup>−</sup> position in the silicatesubstituted apatite matrix. The occupation of divalent calcium ion sites by trivalent europium ions could be described according to the Kröger–Vink notation by the charge compensation phenomenon. According to this theory, the total charge in the material should be compensated by the creation of relatively positive or negative charge. The following processes may be observed:

A double negative vacancy on the Ca2+ position (V"Ca) is created by the substitution of divalent calcium ions by trivalent europium ions (Equation (1)):

$$\mathrm{Eu\_2O\_3 + 3Ca^{'}\_{Ca} \to 2Eu\_{Ca} + 3CaO + V''\_{Ca}} \tag{1}$$

The substitution of divalent calcium ions by trivalent rare earth ions could be explained by the creation of interstitial oxygen O" <sup>i</sup> with double relative negative charge. The mechanism could be described as follows (Equation (2)):

$$\text{Eu}\_2\text{O}\_3 + 2\text{Ca}^\*\text{Ca}^\* \rightarrow 2\text{Eu}\cdot\_{\text{Ca}} + 2\text{CaO} + \text{O}^''\text{i} \tag{2}$$

Eu3+ ions first replace into the Ca(1) site and this preference changes with increasing Eu3+ concentration in favor of the Ca(2) site. In case of substitution into the Ca(2) site, where calcium(II) ions are surrounded with one hydroxyl group and six oxygen atoms from PO4 <sup>3</sup>−, these hydroxyl groups could participate in the charge compensation mechanism, expressed as (Equation (3)):

$$\text{Eu}\_2\text{O}\_3 + 2\text{Ca}^\*\_{\text{Ca(2)}} + 2\text{OH}^\*\_{\text{OH}} \rightarrow 2\text{Eu}\_{\text{Ca(2)}} + 2\text{O}^\prime\_{\text{OH}} + 2\text{CaO} + \text{H}\_2\text{O} \tag{3}$$

In the case of the obtained materials, the substitution of the PO4 <sup>3</sup><sup>−</sup> group by the more negative SiO4 <sup>4</sup><sup>−</sup> group could create a negative charge on the PO4 <sup>3</sup><sup>−</sup> position and two positive charge vacancies (V−Ca) on the hydroxyl group position. The mechanism can be expressed as follows (Equation (4)):

$$2\text{SiO}\_2 + 2\text{PO}\_4^\* \overset{\*}{\text{PO}4} + 2\text{OH}^\* \overset{\*}{\text{OH}} \rightarrow 2\text{SiO}\_{4'\text{PO}4} + 2\text{V} \cdot \text{OH} + \text{P}\_2\text{O}\_5 + \text{H}\_2\text{O} \tag{4}$$

In the present study, the charge compensation mechanism could be described as a combination of Equations (1)–(4). Equation (5) combines the creation of negative and positive vacancy, because of Ca2+ substitution by Eu3+ and PO4 <sup>3</sup><sup>−</sup> substitution by SiO4 <sup>4</sup>−, respectively.

$$2\text{SiO}\_2 + \text{Eu}\_2\text{O}\_3 + 2\text{Ca}^\*\text{Ca} + 2\text{PO}\_4^\*\text{PO4} \rightarrow 2\text{Eu}\cdot\_\text{Ca} + 2\text{SiO}\_4\text{P04} + 2\text{ CaO} + \text{P}\_2\text{O}\_5\tag{5}$$

#### *3.3. Infrared Spectra*

To confirm the presence of phosphate, silicate and hydroxyl groups, the infrared spectra were measured and are presented in Figure 2. According to previous reports, characteristic peaks are ascribed to the compound of hydroxyapatite [3,22,23]. The triply degenerated antisymmetric stretching vibration ν3(PO4 <sup>3</sup>−) of phosphate groups is observed at 1101.1 and 1049.1 cm−1. At 966.4 cm−1, lines are detected which can be described as non-degenerated symmetric stretching bands ν1(PO4 <sup>3</sup>−) vibrations. Strong absorption bands associated with the ν4(PO4 <sup>3</sup>−) triply degenerated vibrations are located at 566.2 and 604.8 cm−1. Two bands related to the stretching and bending modes of OH<sup>−</sup> groups are observed at 3571.3 and at 604.5 cm−1, respectively. These bands clearly confirm the presence of hydroxyl groups in the crystal structure. The broad peak between 3600 and 3200 cm−<sup>1</sup> belongs to H2O vibration. Peaks assigned to (SiO4) <sup>4</sup><sup>−</sup> vibrational modes and Si–O–Si stretch modes are observed at 890 and 478 cm<sup>−</sup>1, respectively. It should be pointed out that there are similarly located vibrational modes of the (PO4) <sup>3</sup><sup>−</sup> and the silicate groups in the hydroxyapatite matrix, causing some interpretation problems. The Si–O symmetric stretching mode is located at 945 cm−<sup>1</sup> and the weak peak corresponding to the P–O symmetric stretching mode is located at 962 cm−<sup>1</sup> [23].

**Figure 2.** FT-IR spectra of the silicate-substituted hydroxyapatite co-doped with Sr2+ (2.0 mol%) and Eu3+ (0.5, 1.0 and 2.0 mol%) ions. <sup>175</sup>

#### *3.4. Spectroscopy Properties*

The emission excitation spectra of the silicate-substituted appetites were recorded as a function of the europium ion concentration and heat-treated temperature in Figure 3a,b, respectively. The spectra were recorded at 300 K at an observation wavelength of 616 nm (16,233 cm−1). The presented spectra were normalized to the most intensity bands. In relation to the most intense band of the 5D0 → 7F2 transition, the spectra were recorded at an observation wavelength of 618 nm. In the UV range was observed a broad, intense band ascribed to the O2<sup>−</sup> → Eu3+ charge transfer (CT) transition with a maximum located around 205 nm (48,780 cm<sup>−</sup>1). Increasing the dopant concentration and heat-treating temperature does not have an influence on the CT maximum position. In the composition of the excitation spectra were recorded sharp, narrow bands, attributed to the 4f–4f transitions of Eu3+ ions, at: 7F0 → 5F(4,1,3,2), 3P0 at 299.3 nm (33,411 cm<sup>−</sup>1), 7F0 → 5H(6,5,4,7,3) at 319.8 nm (31,269 cm−1), 7F0 → 5D4, 5L8 at 363.7 nm (27,495 cm−1), 7F0 → 5G2, 5L7, 5G3 at 383.7 nm (26,062 cm−1), 7F0 → 5L6 at 394.4 nm (25,354 cm−1), 7F0 → 5D3 at 413.8 nm (24,166 cm−1) and 7F0 → 5D2 at 465.1 nm (21,500 cm−1). In lanthanide ions, strongly isolating the 4f orbitals by the external 5s, 5p and 5d shells causes only slight changes in the positions of the electronic transition bands.

Figure 4a represents the emission spectra for the 2 mol% Eu3+-doped sample as a function of sintering temperature. Figure 4b shows the emission spectra as a function of optically active ion concentration for samples sintered at 600 ◦C. The emission spectra were detected at an excitation wavelength of 394 nm to directly excite the f electrons of Eu3+ ions. All spectra were normalized according to the 5D0 → 7F1 magnetic transition. As should have been expected, the emission transitions of the Eu3+ ions are forbidden by selection rules but do not consider the subtle influence of atom vibrations which consequently change the dipole moment, causing the occurrence of forbidden transitions on the spectrum. In the spectra are presented emission lines due to the 5D0 → 7FJ transitions for J = 0, 1, 2, 3 and 4, typical of Eu3+ ions, which are ascribed in reference to previous reports [5,24–27]. The bands located at the listed wavelength were attributed to the following transition: 5D0 → 7F0 at 577 nm (17,331 cm−1); 5D0 → 7F1 at 588 nm (17,006 cm−1); 5D0 → 7F2 at 616 nm (16,233 cm−1); 5D0 → 7F3 at 653 nm (15,313 cm−1) and 5D0 → 7F0 at 700 nm (14,285 cm−1). In general, the spectroscopic properties can provide information about the local chemical environment of Eu3+ ions. The essential change in emission spectra can be dependent on the quantity of possible crystallographic positions and therefore the amount of potential sites of substitution, as well as the presence of defects, additional phases or impurities [3,27].

The typical emission spectrum of 4f–4f electrons of Eu3+ ions is recorded in the red range of the electromagnetic radiation spectrum. In this range, the 5D0 → 7F0,1,2 transitions are observed and the shape of the lines in this region is correlated with the structural properties of the material. Due to these transitions, the trivalent europium ion is called an optical probe. If the Eu3+ ion is located in a centrosymmetric crystal lattice, the 5D0 → 7F0 transition is not observed. In the contrary situation, if the Eu3+ ion is placed in a non-centrosymmetric lattice, the 5D0 → 7F0 transition is detected on the emission spectra. Moreover, this is possible only in the low-symmetry crystal position as in Cn, Cnv as well as CS symmetry. In the hydroxyapatite matrix, calcium ions are located at two different crystal positions: Ca(1) and Ca(2) with local symmetry at C3 and CS, respectively. Both calcium positions could be occupied by Eu3+ ions. The additional crystallographic position appears as a result of reverse cis and trans symmetry of the Ca(2) site with the same point symmetry. In the environment of the Ca(1) site, nine oxygen atoms from phosphate groups are present. The calcium ion located on the Ca(2) site is in sevenfold coordination with six oxygen atoms from the phosphate group and one from the hydroxyl group [28]. The number and ratio of the intensity of the lines in the range between 570 and 580 nm give information about Eu3+ ion site-occupied preference. The coordination polyhedra of the Ca(1) and Ca(2) cations are presented in Figure 5a,b, respectively.

**Figure 3.** Excitation emission spectra of Ca9.8-xSr0.2Eux(PO4)2(SiO4)4(OH)2, where x = 0.5, 1.0, 2.0 mol%, sintered at 600 ◦C (**a**) as well as Ca9.6Sr0.2Eu0.2(PO4)2(SiO4)4(OH)2 as a function of sintering temperature (**b**).

**Figure 4.** Emission spectra of Ca9.6Sr0.2Eu0.2(PO4)2(SiO4)4(OH)2 as a function of sintering temperature (**a**) and Ca9.8-xSr0.2Eux(PO4)2(SiO4)4(OH)2, where x = 0.5, 1.0, 2.0 mol%, sintered at 600 ◦C (**b**).

Emission spectra recorded for the investigated samples in the range of 570 to 580 nm are presented in Figure 5. In the emission spectra of 2 mol% Sr2+ and 0.5 mol% Eu3+ codoped silicate-substituted hydroxyapatites, the transition attributed to 5D0 → 7F0 presents some interesting features due to abnormally strong intensity. This fact can be related to an important perturbation of the symmetry of the dopant induced by the introduction of silicate groups. Moreover, the abnormally strong intensity of the 5D0 → 7F0 transition has been reported in apatites such as oxyapatite Ca10(PO4)6O2, fluoroapatite Ca5(PO4)3F, hydroxyapatite Sr10(PO4)6(OH)2, or silicophosphate apatite Sr5(PO4)2SiO4, etc. This intense emission is attributed to the existence of the strong covalence of the Eu3+-O2<sup>−</sup> bond in the Ca(2) site in the apatite lattice [28–30].

In the other cases, the population of the Ca2+ position replaced by the Eu3+ ion could be related to the thermal diffusion process of dopant ions in the apatite structure. It is commonly known that in the case of as-prepared apatite materials, only the emission associated with one type of site with C3 symmetry was observed, whereas with an increase in the calcination temperature, additional 0–0 peaks appeared [1,31,32]. The emission intensity ratio Ca(1)/Ca(2) has been calculated and results are presented in Table 1, showing values from 2.4 for as-prepared samples to 3.1 for sintered samples at 600 ◦C. Taking

into account all results, one can note that in the silicate-substituted strontium-doped hydroxyapatite matrix, the Ca(1) site is more occupied by Eu3+ ions.

**Figure 5.** The projection of the coordination polyhedra of (**a**) Ca(1) and (**b**) Ca(2) cations, respectively. Emission spectra of (**c**) Ca9.8-xSr0.2Eux(PO4)2(SiO4)4(OH)2, where x = 0.5, 1.0, 2.0 mol%, sintered at 600 ◦C and (**d**) Ca9.6Sr0.2Eu0.2(PO4)2(SiO4)4(OH)2 as a function of sintering temperature, for the 5D0 <sup>→</sup> 7F0 transition.

To obtain additional insight into the luminescence behavior of Eu3+ ions in silicatesubstituted strontium co-doped apatite, the Judd–Ofelt theory was applied [33,34]. The results of the calculation of radiative (Arad), non-radiative (Anrad) and total (Atot) processes, as well as intensity parameters (Ω2, Ω4), quantum efficiency (η) and asymmetry ratio (R), are presented in Table 1. These parameters were calculated based on emission spectra and decay profiles according to equations defined in previous reports [35,36].

A comparison of the Judd–Ofelt intensity parameters (Ω<sup>2</sup> and Ω4) is made between the samples with different Eu3+ ion concentrations and sintering temperatures. An increase in the Ω<sup>2</sup> value is noted with an increase in the heat-treating temperature. This result indicates the increasingly hypersensitive character of the 5D0 → 7F2 transition and the increasing polarization of the Eu3+ ion environment. Consequently, Eu3+ ions in the material heat-treated at the highest temperature are in a more polarizable environment. The influence of silicon group presence is analyzed in comparison with previous work by our group [37]. A decrease in the J-O <sup>Ω</sup><sup>2</sup> parameter is noted from 6.683 × <sup>10</sup>−<sup>20</sup> cm<sup>2</sup> (see [37]) to 4.506 × <sup>10</sup>−<sup>20</sup> cm<sup>2</sup> (see Table 1) for Ca9.7Sr0.2Eu0.1(PO4)6(OH)2 and for Ca9.7Sr0.2Eu0.1(PO4)2 (SiO4)4(OH)2, respectively. This observation can be related to the improvement of the Eu3+ cation polyhedral and to the decrease in the covalence character of the Eu3+–O2<sup>−</sup> bond in silicate-substituted apatite. In the case of the investigated samples, the Ω<sup>2</sup> > Ω<sup>4</sup> parameter suggests that Eu3+ ions are not located in the local symmetry of centrosymmetric character. The calculated results are in agreement with previous reports regarding apatite systems [1,24,35,37].

The highest quantum efficient (η) is observed for the 1 mol% Eu3+-doped compound with the value of 40%. The quantum efficient is reduced by the increase as well as the decrease of optically active ion concentration to 37% and 34%, respectively. Samples heattreated at higher temperatures demonstrate higher quantum efficient values, and the η parameter increases by the maximum of 15%.

The luminescence intensity ratio (R) of the electric dipole transition 5D0 → 7F2 to the magnetic dipole transition 5D0 → 7F1 has been calculated and the results are presented in Table 1. As expected, the sintering process and concentration of optical active ions have a significant influence on the symmetry of the Eu3+ ion environment in the apatite matrix.

The R factor increases with an increase in the heat-treating temperature, which indicates lower symmetry around Eu3+ ions and suggests that the covalence of Eu3+-O2<sup>−</sup> is higher. The R parameter increases for 0.5 mol% Eu3+ to 1.0 mol% Eu3+ from 3.1 to 4.5, respectively. Then, the opposite trend is observed, and the R parameter decreases from 4.5 to 4.1 for 1.0 mol% Eu3+ and 2.0 mol% Eu3+.

**Table 1.** Decay rates of radiative (Arad), non-radiative (Anrad) and total (Atot) processes of 5D0 <sup>→</sup> 7FJ transitions, luminescence lifetimes (τ), intensity parameters (Ω2, Ω4), quantum efficiency (η) and asymmetry ratio (R) for investigated samples.


Figure 6 presents the energy level diagram of Eu3+ corresponding to the detected excitation and emission spectra. As seen, Eu3+ ions were pumped to upper excited levels. The emission bands from the 5D1, 5D2 and 5D3 levels are not observed at room temperature in the case of the investigated samples, which suggests a fast, non-radiative (NR), multiphonon relaxation from the excited state 5L6 to the 5D0 state.

#### *3.5. Temperature-Dependent Emission*

To further study the possible application under high temperature, the temperaturedependent emission spectra of the silicate-substituted hydroxyapatite co-doped with 2 mol% Sr2+ and 2 mol% Eu3+ were measured and are presented in Figure 7. Emissions were recorded in the range of 80 to 725 K and at an excitation wavelength of 375 nm into the 5G2 level. It is seen that the emission intensity decreases clearly with an increase in the ambient temperature, but the decrease is not linear in the whole range of measured temperatures. For the most intense line corresponding to the 5D0 → 7F2 transition, the linear relationship applies between 80 and 325 K (R<sup>2</sup> = 99.4%). Then, between 450 and 800 K, the decrease can be described by the exponential equation (see Figure S3). The line corresponding to the 5D0 → 7F0 transition of the Ca(2) calcium site (573 nm) is completely eliminated at 350 K.

**Figure 6.** The simplified energy level scheme for Eu3+ ion in silicate-substituted strontium-doped hydroxyapatite.

**Figure 7.** Temperature-dependent emission spectra of the Ca9.6Sr0.2Eu0.2(PO4)2(SiO4)4(OH)2, sintered at 600 ◦C.

In Figure 8 has been shown the Commission Internationale de l'Eclairage (CIE) 1931 chromaticity diagram for the Ca9.6Sr0.2Eu0.2(PO4)2(SiO4)4(OH)2 sample. The CIE color coordinates are listed in Table S3, which are calculated from the temperature-dependent emission spectra [38]. It has been reported that the emission color changed from reddishorange to orange with the increasing of ambient temperature, but the reddish-orange emission was stable until 750 K. These results show that the Eu3+-activated silicate-substituted

apatites have the potential to be color-stable materials capable of operating at a wide range of ambient temperatures.

**Figure 8.** CIE 1931 chromaticity diagram of the Ca9.6Sr0.2Eu0.2(PO4)2(SiO4)4(OH)2 as a function of ambient temperature.

#### *3.6. Decay Time*

The luminescence kinetics corresponding to the 5D0 → 7F2 transition were obtained at room temperature. As expected, all the recorded decays presented non-single exponential character. This phenomenon was consistent with the existence of two non-equivalent Eu3+ positions and, because of this fact, the effective emission lifetime was calculated by Equation (6). The recorded decays and calculated luminescence lifetimes (τ) are presented in Figure 9, as a function of Eu3+ concentration (a) and (b) as well as heat-treating temperature ((c) and (d)), respectively.

$$\tau\_{\rm mf} = \frac{\int\_0^\infty tI(t)dt}{\int\_0^\infty I(t)dt} \cong \frac{\int\_0^{t\_{\rm max}} tI(t)dt}{\int\_0^{t\_{\rm max}} I(t)dt} \tag{6}$$

**Figure 9.** Calculated average lifetimes (**a**) as well as luminescence decay profiles (**b**) of Ca9.8-xSr0.2Eux(PO4)2(SiO4)4(OH)2, sintered at 600 ◦C, as a function of Eu3+ ion concentration. Luminescence decay profiles (**c**) and calculated average lifetimes (**d**) of the Ca9.6Sr0.2Eu0.2(PO4)2(SiO4)4(OH)2 as a function of sintering temperature.

#### **4. Conclusions**

In this study, it has been shown for the first time that a series of silicate-substituted hydroxyapatite co-doped with 2 mol% Sr2+ and Eu3+ ions in the range of 0.5–2.0 mol % in a ratio to the entire Ca2+ ion content were successfully synthesized by the hydrothermal method assisted with microwave and heat-treated. The average crystal sizes of the studied materials were in the range of 16–56 nm as calculated by the Rietveld method.

Attention was paid to the structural and spectroscopic properties related to a variable amount of Eu3+ ion concentration. The spectroscopic properties have shown for the sintered samples that Eu3+ ions occupied three independent crystallographic sites: one Ca(1) site with *C*<sup>3</sup> local symmetry and two Ca(2) sites with *Cs* local symmetry with *cis* and *trans* symmetry. The 5D0 → 7F2 hypersensitive transition is the most intense for most obtained materials, excluding the sample co-doped with 0.5 mol% Eu3+ ion and sintered at 600 ◦C, where the most dominant is 5D0 → 7F0 transition. Moreover, the charge compensation mechanism in the materials induced by the substitution of Eu3+ and Sr2+ ions into the silicate-substituted hydroxyapatite host lattice was rendered in the Kröger–Vink notation.

The luminescence decay times corresponding to the most intense 5D0 → 7F2 transition were recorded. The luminescence kinetics was characterized by a non-exponential decay profile and was in the range of 2.15 ms (0.5 mol% Eu3+) and 19.4 ms (1 mol% Eu3+) to 1.9 ms (2 mol% Eu3+) for the samples sintered at 600 ◦C. On the other hand, the decay times for the samples doped with 2 mol% Eu3+ as a function of sintered temperature were in the range of 15.2 to 1.92 ms. The typical modes of the vibrations of the silicate-substituted hydroxyapatite ion group were detected in the FT-IR spectra, and these included the OH− group vibrations characteristic of the hydroxyapatite matrix. The simplified Judd–Ofelt theory was used for a detailed analysis of the luminescence spectra. The hydroxyapatite containing 1 mol% of Eu3+ ions was evaluated to be the most optically efficient material among all the studied silicate-substituted hydroxyapatites. The International Commission on Illumination (CIE) color coordinates showed that the emission color can be tuned by varying the ambient temperature. The emission color was changed from reddish-orange to orange.

**Supplementary Materials:** The following are available online at https://www.mdpi.com/2079-499 1/11/1/27/s1, Figure S1: The representative SEM image (a) and EDS spectra (b) of the Ca9.6Sr0.2Eu0.2 (PO4)2(SiO4)4(OH)2 nanopawders. Figure S2: Representative results of the of the Sr0.2Eu0.2Ca9.6(PO4)2 (SiO4)4(OH)2, obtained at 600 ◦C, Rietveld analysis (red—fitted diffraction; blue—differential pattern; column—reference phase peak position). Figure S3: Temperature-dependent emission intensity of the lines correspond to the listed transitions. Table S1. The amount of substrates used for synthesis of silicate-substituted hydroxyapatite co-doped wth Eu3+ and Sr2+. Table S2: Unit cell parameters (a,c), cell volume (V), grain size as well as refine factor (RW) for the Ca10(PO4)2(SiO4)4(OH)2 co-doped with 2 mol% Sr2+ and x mol/% Eu3+ ions (where x = 0.5–2). Table S3: The comparison of the CIE color coordinates (x,y) of Ca9.6Sr0.2Eu0.2(PO4)2(SiO4)4(OH)2 as a function of ambient temperature.

**Author Contributions:** R.J.W. conceived and designed the experiments and contributed reagents/materials/analysis tools and participated in funding acquisition in addition to analyzing all data; S.T. designed the experiments in addition to analyzing data. Both authors have read and agreed to the published version of the manuscript.

**Funding:** The authors would like to acknowledge financial support from the National Science Centre (NCN) within the project "Preparation and characterisation of biocomposites based on nanoapatites for theranostics" (No. UMO-2015/19/B/ST5/01330).

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Not applicable.

**Acknowledgments:** We are grateful to E. Bukowska for the help with XRD measurements and to D. Szymanski for SEM measurements.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


## *Article* **Mechanistic Insight of Sensing Hydrogen Phosphate in Aqueous Medium by Using Lanthanide(III)-Based Luminescent Probes**

**Jashobanta Sahoo 1,2,3, Santlal Jaiswar 4, Pabitra B. Chatterjee 2,5, Palani S. Subramanian 1,2,\* and Himanshu Sekhar Jena 6,\***


**Abstract:** The development of synthetic lanthanide luminescent probes for selective sensing or binding anions in aqueous medium requires an understanding of how these anions interact with synthetic lanthanide probes. Synthetic lanthanide probes designed to differentiate anions in aqueous medium could underpin exciting new sensing tools for biomedical research and drug discovery. In this direction, we present three mononuclear lanthanide-based complexes, EuLCl3 (**1**), SmLCl3 (**2**), and TbLCl3 (**3**), incorporating a hexadentate aminomethylpiperidine-based nitrogen-rich heterocyclic ligand **L** for sensing anion and establishing mechanistic insight on their binding activities in aqueous medium. All these complexes are meticulously studied for their preferential selectivities towards different anions such as HPO4 <sup>2</sup>−, SO4 <sup>2</sup>−, CH3COO−, I−, Br−, Cl−, F−, NO3 −, CO3 <sup>2</sup>−/HCO3 −, and HSO4 − at pH 7.4 in aqueous HEPES (2-[4-(2-hydroxyethyl)piperazin-1-yl]ethanesulfonic acid) buffer. Among the anions scanned, HPO4 <sup>2</sup><sup>−</sup> showed an excellent luminescence change with all three complexes. Job's plot and ESI-MS support the 1:2 association between the receptors and HPO4 <sup>2</sup>−. Systematic spectrophotometric titrations of **1**–**3** against HPO4 <sup>2</sup><sup>−</sup> demonstrates that the emission intensities of **1** and **2** were enhanced slightly upon the addition of HPO4 <sup>2</sup><sup>−</sup> in the range 0.01–1 equiv and 0.01–2 equiv., respectively. Among the three complexes, complex **3** showed a steady quenching of luminescence throughout the titration of hydrogen phosphate. The lower and higher detection limits of HPO4 <sup>2</sup><sup>−</sup> by complexes **1** and **2** were determined as 0.1–4 mM and 0.4–3.2 mM, respectively, while complex **3** covered 0.2–100 μM. This concludes that all complexes demonstrated a high degree of sensitivity and selectivity towards HPO4 <sup>2</sup>−.

**Keywords:** lanthanides; luminescence; nitrogen-rich ligand; phosphate sensing; quenching

#### **1. Introduction**

Inorganic phosphates, the charged anions of phosphoric acid such as [H2PO4] −, [HPO4] <sup>2</sup>−, and [PO4] <sup>3</sup>−, are essential components during the synthesis of DNA/RNA and phospholipid membrane [1]. Further, their influence in the metabolic process in human, plant, and animal cells are inevitable. Sensing of phosphate draws special attention [2–13] due to its biological role as polyphosphate, and hyper- and hypophosphatemia in Chronic Kidney Disease (CKD) patients [14]; energy source through dephosphorylation [15] of

**Citation:** Sahoo, J.; Jaiswar, S.; Chatterjee, P.B.; Subramanian, P.S.; Jena, H.S. Mechanistic Insight of Sensing Hydrogen Phosphate in Aqueous Medium by Using Lanthanide(III)-Based Luminescent Probes. *Nanomaterials* **2021**, *11*, 53. https://doi.org/10.3390/nano11010053

Received: 26 October 2020 Accepted: 23 December 2020 Published: 28 December 2020

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license ( https://creativecommons.org/ licenses/by/4.0/).

ATP, ADP, AMP, and PPi; and reverse polycondensation to form polyphosphates. Various methods were developed for the determination of phosphates in fertilizers, plants, natural waters, and other environmental samples [16–18]. Generally, serum phosphates are measured based on a photometric approach using ammonium phosphate, which forms a chromogenic complex with inorganic phosphates (Pi) [19]. However, the search for new receptors with selective response to phosphates remains active behind many challenges. Moreover, with phosphates being important bioanlytes [20–29], varieties of colorimetric sensors [30–35] and fluorosensors [36–38] were reported for their detection. Among these, luminescent lanthanide [20] complexes gained significant attention due to their potential applications in clinical diagnosis, biomarkers [39,40], MRI contrast agents [41–47], screening of drugs, etc. Parker et al. reported Eu(III) and Tb(III) tetra-azaphenylene complexes for the detection of phosphates in live cells [48,49]. It is important to note that the concentrations of phosphate vary significantly in inter- and intracellular environments of human cells, ranging from 0.15 to 1.3 mM [50–52]. Among the various lanthanide complexes reported so far in the literature, Eu-Tc [53,54] was recognized as an efficient probe for phosphates due to its lower detection limit (LOD = 3 μmolL<sup>−</sup>1). Moreover, there are many intracellular processes, where the concentrations of phosphate vary among different subcellular compartments present therein [55]. Therefore, a highly sensitive and selective probe which can detect phosphate at a considerably low concentration is very much required to investigate such intracellular processes. In this context, recently, we have reported a set of europium(III) and terbium(III) complexes, incorporating different hexadentate ligands which showed highly selective and efficient recognition of inorganic phosphates and nucleoside phosphates [56,57]. In this direction and as a part of our ongoing research, herein, we report a series of relatively simple, cheap, and water-soluble Ln(III) complexes **1**, **2**, and **3** (Scheme 1) (Ln = Eu, Sm, and Tb, respectively) using an aminomethylpiperidinefunctionalized 1,10-phenanthrolene-based nitrogenous heterocyclic ligand **L** as the metal chelator. The anion-sensing ability of these hydrophilic rare-earth complexes (**1**–**3**) was explored and found high selectivity and sensitivity for hydrogen phosphate ions in HEPES (2-[4-(2-hydroxyethyl)piperazin-1-yl]ethanesulfonic acid) buffer at pH 7.4. Moreover, a mechanistic insight into the anion binding behavior of complex **1** was also explored in this work.

**Scheme 1.** 2,9-Bis(aminomethylpiperidine)-1,10-phenanthrolene (**L**) and its Ln(III) complexes **1**, **2**, and **3**.

#### **2. Results and Discussion**

Schiff base ligand **L** was obtained by condensing 2,9-dialdehyde 1,10-phenanthroline and 2-(aminomethyl) piperidine. The characteristic azomethine peak at 8.25δ in 1H NMR (Figure S1) and 13C (Figure S2), DEPT 135◦ NMR (Figure S3) in combination with MS spectrum (Figure S4), CD spectrum (Figure S5) and IR spectra (Figure S6), confirms the formation of **L**. Treating **L** with the respective LnCl3 salt, the corresponding complexes

EuLCl3**(1**), TbLCl3(**2**), and SmLCl3(**3**) were isolated as per the procedure. The formulation of each complex was confirmed from the ESI-MS analysis (Figure S7–S9). The emission spectra of these complexes were studied at 25 ◦C in aqueous HEPES buffer at pH 7.4 (Figure S10). All the complexes showed a significant red-shift of the emission spectra with respect to the emission profile of the free ligand **L**. Being luminescent in nature, we sought to investigate the excited state photophysical properties of **1**–**3** in the presence of various important anions. Complexes **1** and **3** showed characteristic luminescent bands at 614 nm and 545 nm, respectively, while complex **2** displayed two sensitive bands at 595 and 644 nm attributable for their metal centered emission. Although water functions as a luminescence quencher [58], all these complexes showed an intense luminescence in aqueous HEPES buffer at physiological pH at 25 ◦C. The effects of the addition of a range of anions such as hydrogen phosphate, sulfate, acetate, iodide, bromide, chloride, fluoride, nitrate, carbonate/bicarbonate, and bisulfate on the emission spectra of **1**–**3** is showed in Figure 1a–c. The bar diagrams, as insets in Figure 1a,c, show the changes in emission intensities of the hypersensitive peaks at 614 and 545 nm of complexes **1** and **3**, while Figure 1b depicts the changes in the ratio of the 644/595 nm hypersensitive peaks of complex **2** with the addition of various anions.

**Figure 1.** Emission spectra of (**a**) **<sup>1</sup>** (1 <sup>×</sup> <sup>10</sup>−<sup>5</sup> M), (**b**) **<sup>2</sup>** (2 <sup>×</sup> <sup>10</sup>−<sup>5</sup> M), and (**c**) **<sup>3</sup>** (4 <sup>×</sup> <sup>10</sup>−<sup>5</sup> M) upon the addition of various anions (100 equiv. for **1** and 10 equiv. for **3** in aqueous HEPES (2-[4-(2-hydroxyethyl)piperazin-1-yl]ethanesulfonic acid) buffer at pH = 7.4, λexi = 276 nm): in the case of **2**, spectra obtained after the addition of both 100 (cyano) and 500 equiv. (light brown) of HPO4 <sup>2</sup><sup>−</sup> are overlaid. Insets: luminescence intensities of **1** and **3** at 614 and 545 nm, respectively, in the presence of different anions, while in the case of **2**, relative intensity ratios (644/595 nm) are plotted along the y-axis.

Among the anions scanned, complex **1** illustrated a significant emission change with hydrogen phosphate and bicarbonate ions. While the addition of HCO3 − showed 14.7% luminescence enhancement, phosphate in contrast leads to luminescence quenching by 29% of the emission intensity of **1** (Figure 1a). In Figure S11, the emission spectra of **1** against varying concentrations of phosphates (0–400 equiv.) are shown. The emission intensity was found to increase (2.6-fold) initially in the range 0.01–1 equiv. of HPO4 <sup>2</sup><sup>−</sup> (Figure 2a).

**Figure 2.** (**a**) Changes in emission maxima of **<sup>1</sup>** (1 <sup>×</sup> <sup>10</sup>−<sup>5</sup> M) upon gradual addition of HPO4 <sup>2</sup><sup>−</sup> in aqueous HEPES buffer at pH = 7.4; (**b**) nonlinear curve fitting of the titration data as a function of HPO4 <sup>2</sup><sup>−</sup> concentrations in the range 0.01–1 equiv. (luminescence enhancement part); (**c**) nonlinear curve fitting of the titration data as a function of HPO4 <sup>2</sup><sup>−</sup> concentrations in the range 1–400 equiv. (luminescence quenching part), in which a factor of 10 was multiplied with enhancement to maintain the same intesnsity for both quenching and enhancement; and (**d**) Job's plot analysis of mixtures of complex **1** with HPO4 <sup>2</sup>−(*C*complex **<sup>1</sup>** <sup>+</sup> *<sup>C</sup>*HPO42<sup>−</sup> = 1.0 <sup>μ</sup>M) in aqueous HEPES buffer pH 7.0, indicating 1:2 complex formation (λemi = 614 nm).

Upon further addition of HPO4 <sup>2</sup><sup>−</sup> to the reaction mixture, a gradual decrease in the luminescence, at 614 nm, of receptor **1** was observed (Figure 2a). The changes in the luminescence intensities, as displayed in Figure 2a, can be attributed to the two distinct behaviours of **1** against HPO4 <sup>2</sup>−. Therefore, the spectrometric titration (Figure S11) has offered two association constants. Figure 2a also displayed that the luminescence intensity of the emission maximum decreased to a constant level after the addition of 4 mM of phosphate. Therefore, it is evident from Figure 2a that **1** can be used to sense a wide range of phosphate concentrations. The analytical limit of detection (LOD) [59–62] of **1** for phosphate was calculated as 0.1 μM. Since the sensing of hydrogen phosphate showed nonlinear fitting in Figure 2b,c with one enhanced and the other quenching the luminescence intensity, we applied the nonlinear fit data point results by following Equations (1) and (2), [63–66] respectively, providing the binding constants *<sup>K</sup>*<sup>1</sup> = 2.0 × 104 <sup>M</sup>−<sup>1</sup> (R = 0.9869), 1st part, attributable to luminescence enhancement and *<sup>K</sup>*<sup>2</sup> = 0.94 × 104 <sup>M</sup>−<sup>1</sup> (R = 0.9875), 2nd part, associated to luminescence quenching.

$$F = F\_0 + \frac{F\_{\text{max}} - F\_0}{2} + \left\{ \left( 1 + \frac{[M]}{\mathbb{C}\_L} + \frac{1}{\mathbb{C}\_L K} \right) - \sqrt{\left( 1 + \frac{[M]}{\mathbb{C}\_L} + \frac{1}{\mathbb{C}\_L K} \right)^2 - 4 \frac{[M]}{\mathbb{C}\_L}} \right\} \tag{1}$$

$$F = F\_{\max} + \frac{F\_0 - F\_{\max}}{2} + \left\{ \left( 1 + \frac{[M]}{\mathbb{C}\_L} + \frac{1}{\mathbb{C}\_L K} \right) - \sqrt{\left( 1 + \frac{[M]}{\mathbb{C}\_L} + \frac{1}{\mathbb{C}\_L K} \right)^2 - 4 \frac{[M]}{\mathbb{C}\_L}} \right\} \tag{2}$$

where *F*<sup>0</sup> is the luminescence intensity in the absence of hydrogen phosphate and *Fmax* is the luminescent intensity in the presence of HPO4 <sup>2</sup>−, and *CL* and *K* are the concentration and binding constant of the complex, respectively. To find the association stoichiometry between complex **1** and HPO4 <sup>2</sup>−, Job's plot was performed (Figure 2d), which established 1:2 binding stoichiometry, i.e., [**1**:HPO4 <sup>2</sup><sup>−</sup> = 1:2]. Further, a final confirmation regarding the abovementioned 1:2 stoichiometry was provided by ESI-MS (Figure S12), where a peak at *<sup>m</sup>*/*<sup>z</sup>* = 855.23 with 100% abundance was attributed to [EuL(HPO4)2(H2O) + 2Na+]·H2O (calcd. *m*/*z* = 855.11).

Unlike hydrogen phosphate, HCO3 − showed little enhancement in the emission of **<sup>1</sup>** (Figure 1a). The binding constant (*K*) was determined to be 1.2 × 103 <sup>M</sup>−<sup>1</sup> from the spectrometric titrations of **1** against increasing concentrations of HCO3 − (10 to 600 equiv.), as shown in Figure S13. Luminescence enhancement of **1**, observed in the addition of HCO3 −, may occur due to chelate formation between the bicarbonate ion and europium (III) center by replacing the weakly bound inner sphere water molecules [67]. In Figure 1b, the luminescence response of **2** towards different anions is displayed. Among the four emission bands observed for **<sup>2</sup>**, the peaks at 595 nm (5G5/2→6H7/2) and 644 nm (5G5/2→6H9/2) were found to be hypersensitive [68–70].

Spectrophotometric titrations of **2** (Figure S14 and Figure 3a) against varying amounts of HPO4 <sup>2</sup><sup>−</sup> ranging from 0.01 to 800 equiv. illustrated similar patterns as observed earlier in case of **1**. Interestingly, the initial luminescent enhancement of **2** was found up to 18 equiv. of phosphate addition and further increases in phosphate concentrations quench the luminescence of the resulting solution. Applying nonlinear fitting of the data points (Figure 3b,c), the respective binding constants (*K*<sup>1</sup> and *<sup>K</sup>*2) were calculated to be 2.1 × 104 <sup>M</sup>−<sup>1</sup> (R = 0.9828), 1st part of luminescence enhancement and 2.9 × 103 <sup>M</sup>−<sup>1</sup> (R = 0.9858), 2nd part of luminescence quenching. The LOD was calculated to be 0.4 μM, a little higher than that observed for **1**. To derive the complex **2** to phosphate ratio, Job's plot was performed, which clearly indicated 1:2 stoichiometry (Figure 3d). The respective positive ion ESI-MS (Figure S15) also confirmed the proposed 1:2 composition by depicting a molecular ion peak at *m*/*z* = 862.27 attributable to the formation of [Sm(L-2H)(HPO4)2 + 4Na+] (calcd. *m*/*z* = 862.05).

Screening complex **3** towards various anions (10 equiv.), shown in Figure 1c, illustrates again an excellent luminescent probe for HPO4 <sup>2</sup><sup>−</sup> with superior selectivity and sensitivity. Upon the addition of HPO4 <sup>2</sup>−, the emission intensity of **3** was reduced to 8%. Systematic spectrophotometric titration with an increasing concentration of hydrogen phosphate ions in the range 0.01–5 equiv. was performed (Figure 4a and Figure S16). Unlike, **1** and **2**, the luminescence intensity of **3** demonstrated a steady quenching process. The luminescence intensity was quenched continuously from the very beginning of HPO4 2− addition and at the 100 μM HPO4 <sup>2</sup><sup>−</sup> concentration; the emission quenched completely and remained constant thereafter (Figure 4a). The LOD for complex **3** was derived as 0.2 μM. The binding constant was determined from nonlinear data fitting using the Equation (2), and the respective association constant (*K*) was found to be 7.0 × 104 <sup>M</sup>−<sup>1</sup> (R = 0.9870) (Figure 4b). A molecular ion peak at *m*/*z* = 803.56 (Figure S17) can be attributed to the generation of [TbL(HPO4)2 + Na+ + H+] (calcd. *m*/*z* = 803.45) in solution. Thus, all three complexes are established to bind phosphates in 1:2 stoichiometries.

**Figure 3.** (**a**) Changes in emission maxima of **<sup>2</sup>** (4 <sup>×</sup> <sup>10</sup>−<sup>5</sup> M) upon gradual addition of HPO4 <sup>2</sup><sup>−</sup> in aqueous HEPES buffer at pH = 7.4; (**b**) nonlinear curve fitting of the titration data as a function of HPO4 <sup>2</sup><sup>−</sup> concentrations in the range 0.01–18 equiv. (luminescence enhancement part); (**c**) nonlinear curve fitting of the titration data as a function of HPO4 <sup>2</sup><sup>−</sup> concentrations in the range 20–800 equiv. (luminescence quenching part); and (**d**) Job's plot of complex **2** with HPO4 <sup>2</sup>−(*C*complex **<sup>2</sup>** <sup>+</sup> *<sup>C</sup>*HPO42<sup>−</sup> = 1.0 <sup>μ</sup>M) in aqueous HEPES buffer pH 7.0 showing 1:2 complex formation (λemi = 644 nm).

**Figure 4.** (**a**) Change in the emission spectra of **<sup>3</sup>** (2 <sup>×</sup> <sup>10</sup>−<sup>5</sup> M) upon the addition of HPO4 <sup>2</sup><sup>−</sup> in aqueous HEPES buffer at pH = 7.4: the inset shows the Job's plot of mixtures of complex **3** with HPO4 <sup>2</sup>−(*C*complex **<sup>3</sup>** + *C*HPO42<sup>−</sup> = 1.0 <sup>μ</sup>M) in aqueous HEPES buffer pH 7.4 showing 1:2 complex formation. (**b**) Nonlinear curve fitting of luminescence intensities of **3** (2 <sup>×</sup> <sup>10</sup>−<sup>5</sup> M) as a function of HPO4 <sup>2</sup><sup>−</sup> concentrations (λemi = 545 nm).

τ τ

Ligand **L** with its hexadentate nature fulfils six coordination sites of Eu(III), Sm(III), and Tb(III) in their respective complexes **1**, **2**, and **3**. The remaining sites at the metal centers were calculated by measuring the hydration states [11,71] (denoted hereafter by "*q*") of **1**–**3**, adapting Equation (3). Based on the experimental results, the respective inner-sphere hydration numbers calculated for **1**–**3** are compiled in Table 1. Accordingly, complexes **1** and **2** are found to possess four coordinated water molecules while complex **3** accommodates only three water molecules, presumably due to the smaller ionic radius of Tb(III).

$$q\_{corr} = A' \Delta k\_{corr} \text{ [uvherre } \Delta k\_{corr} = (k\_{H2O} - k\_{D2O})] \tag{3}$$

whereas *kH*<sup>2</sup>*<sup>O</sup>* and *kD*<sup>2</sup>*<sup>O</sup>* are radiative rate constants in H2O and D2O solvent, *A'* is a proportionality constant signifying the sensitivity of the lanthanide ion to vibronic quenching by OH oscillators, and *qcorr* is the hydration state, i.e., number of solvent molecules attached to a metal center.


**Table 1.** Excited state lifetime measurements of **1**–**3** in H2O and D2O at pH 7.4 (HEPES buffer).

*<sup>a</sup> qcorr* values were determined by adapting *A'* = 1.2 ms (Eu3+) and 5 ms (Tb3+) and <sup>Δ</sup>*kcorr* <sup>=</sup> −0.25 ms−<sup>1</sup> (Eu3+) and −0.06 ms−<sup>1</sup> (Tb3+). \* For Sm(III), since <sup>Δ</sup>*kcorr* values are not available in the literature, the *qcorr* value for complex **2** is calculated without applying the correction.

To understand the underlying mechanism behind the titration profiles of **1** and **2** against the hydrogen phosphate ions (Figures 2a and 3a), we performed time-resolved luminescence decay studies (Figures S18–S21) and also calculated the quantum yield of each complexes (Table S1). As a representative example lifetime was determined for **1** in the absence and presence of phosphate ions at different stoichiometry and compared with lifetime of the complex **1,** a laser excitation source of 276 nm was used, and the decay luminescence pattern was monitored at 614 nm. Aqueous (H2O as well as D2O) HEPES buffer solutions of **1** were used for these studies. In the absence of HPO4 <sup>2</sup>−, luminescence profiles for **1** could be best fitted to single exponential decay traces with lifetime values τ = 0.22 ms (H2O) and 1.56 ms (D2O) (*κ*<sup>2</sup> = 1.16, 1.18) (Figure S19, Table 1). Upon the addition of one equivalent HPO4 <sup>2</sup><sup>−</sup> to these two solutions, the luminescence decay profiles could be best fitted to τ = 0.41 ms (H2O) and 1.72 ms (D2O) (*κ*<sup>2</sup> = 1.17, 1.13) (Table 2). Thus, the decrease in the hydration state of **1** from *q* = 4 to *q* = 2 upon the addition of 1 equivalent of HPO4 <sup>2</sup><sup>−</sup> (i.e., upon 1:1 association) was obvious to understand. After the addition of another equivalent of HPO4 <sup>2</sup><sup>−</sup> to this mixture, the lifetime values were found to be 0.60 ms (H2O) and 1.89 ms (D2O) (*κ*<sup>2</sup> = 1.20, 1.18) (Table 2), therefore revealing that hydration state *q* = 1, i.e., one coordinated water molecules present at 1:2 binding ratio between **1** and HPO4 <sup>2</sup>−. The gradual addition of excess HPO4 2− did not change the hydration state of the resulting species in solution further, i.e., *q* = 1 (Table 2). Based on the results summarized in Table 2, a plausible mechanism of phosphate's interaction with complex **1** is schematically represented in Figure 5. The initial little luminescence enhancement of **1** upon one equivalent HPO4 <sup>2</sup><sup>−</sup> addition possibly arose due to the replacement of two coordinated water molecules by the incoming phosphate group, which normally functions as a strong chelating species [24]. It is noteworthy to mention that such an effect has already been observed in the case of bicarbonate [72]. The addition of a second equivalent of HPO4 <sup>2</sup><sup>−</sup> to this 1:1 mixture resulted in the displacement of one more coordinated water molecule from the metal center (also supported by Figure S12) and shows quenching of luminescence. Possibly, steric crowding played an important role here by forcing the phosphate group to form a hydrogen bond with the piperidine NH moiety of ligand **L**, causing an energy mismatch between the lowest triplet state (T1) of **L**

and the excited state of the Eu(III). Therefore, the energy transfer process from the ligand L to europium(III) terminated and hence resulted in quenching of the luminescence process. However, at higher concentrations, the emission intensity reduces completely, which can be ascribed to the leaching of lanthanides from the complexes in the presence of more strongly coordinating phosphates ions.

**Table 2.** Summary of the changes in the lifetimes of complex **1** in the presence of different equivalent HPO4 <sup>2</sup><sup>−</sup> in aqueous media (H2O as well as D2O).


**Figure 5.** Proposed mechanistic pathway for successive HPO4 <sup>2</sup><sup>−</sup> binding with **1** in aqueous medium.

> Although a significant number of luminescent complexes are applied in bio-imaging with excitation range below 300 nm [73,74], the presented complexes with excitation at 276 nm (i.e., in the UV region) limit their usage in in vivo bio-imaging. However, for in vitro conditions, the complexes are expected to be significant.

#### **3. Conclusions**

A series of mononuclear Ln(III) complexes (**1**–**3**) based on aminomethylpiperidine functionalized 1,10-phenanthrolene-based nitrogen-rich hexadentate heterocyclic ligand **L** has been reported. All these rare-earth complexes showed red-shifted metal-centered luminescence. The excited state photophysical properties of these complexes were explored to find their specific recognition affinity towards various important anions. The selective sensing of **1**–**3** for hydrogen phosphates over other anions is remarkable. Systematic spectrophotometric analysis demonstrates that, in the case of **1** and **2**, the emission intensities were increased slightly at the very beginning of phosphate addition (up to 1 and 2 equivalents, respectively) and finally decreased to a plateau at high phosphate concentrations (at mM level). The limits of detection (LOD) fall in the range 0.1–0.4 μM. Luminescence decay studies revealed that successive replacement of weakly bound coordinated water molecules from Eu(III) and Sm(III) probably caused the initial emission enhancement of these two complexes upon hydrogen phosphate addition. However, the addition of excess hydrogen phosphate causing steric crowding at the metal site and possible hydrogen bond formation between the piperidine NH group and phosphate might have created an energy mismatch between the lowest triplet state (T1) of **L** and the excited state of the Eu(III), which in turn resulted in termination of the energy transfer between the *o*-phenanthrolene moiety and Ln(III).

#### **4. Experimental Section**

#### *4.1. Materials and Methods*

All chemicals were purchased from Aldrich. Sodium salts of all anions were used in this study. Elemental analyses of the complexes were carried out by using a vario Micro cube from Elementar. IR spectra were recorded from KBr pellets (1% *w*/*w*) on a Perkin–Elmer spectrum GX FTIR spectrophotometer. Electronic spectra were recorded on a Shimadzu UV 3600 spectrophotometer and scanned in the range 200–800 nm. The mass-spectrometric analysis was performed by using the positive ESI technique on a Waters Q-ToF Micromass spectrometer in CH3OH. NMR spectra were recorded on a Bruker *Avance* 500 MHz FT-NMR spectrometer. The chemical shifts (δ) for proton resonances are reported in ppm relative to the internal standard TMS (Tetramethylsilane). The CD spectra were recorded by using a JASCO 815 spectrometer. Milli-Q water was used as a solvent. pH measurements were carried out using an ORION VERSA STAR pH meter. Emission spectra were recorded using an Edinburgh Instruments model Xe-900, and all the spectra recorded are reported hereafter applying emission correction. The slit sizes for emission and excitation were adjusted as 3.0/3.0 nm. For Job plot analysis (continuous variations method), a series of samples were prepared with a constant sum of concentrations at 1.0 μM but with varying concentrations of complex and hydrogen phosphate. The luminescence spectra were recorded for each sample with λex = 276 nm for all these complexes. The maximum luminescence intensity was plotted versus the mole fraction of the corresponding hydrogen phosphate. For determination of the maximum, the ascending and descending segments of the curve were fitted to linear lines, respectively, and the intercept of both lines denotes the maximum and thus the stoichiometry of the complex.

**Synthesis of ligand L. 4(H2O)**: 1,9-Diformyl-1,10-phenanthroline (0.001 mmol, 0.200 g) (*17*) was dissolved in 50 mL of CH3OH. To this methanolic solution, 2-(aminomethyl) piperidine (0.002 mmol, 0.184 g) was added drop by drop. This reaction mixture was stirred continuously for 48 h at 50 ◦C. During this, the color of the reaction mixture changed to red-brown, indicating the formation of a Schiff base. The solvent was removed under vacuum, and the resultant orange-red powder was isolated. Yield. 70%. -1H NMR (CDCl3, 500 MHz): δ = 8.26, 8.24 (*dd*, *J* = 3 Hz, 2H), 7.86, 784, 7.82(*t*, *J* = 10 Hz, 4H), 7.77(*s*, 2H), 3.55(*brs*, 2H), 3.27(*m*, 2H), 3.09, 3.07, 3.05, 3.03 (*q*, *J* = 10 Hz,2H), 2.83, 2.81, 2.79 (*t*, *J* = 12 Hz, 2H), 2.50 (*m*, 2H), 2.24, 2.22, 2.20 (*t*, *J* = 11 Hz, 2H), 1.93–1.83(*dd*, *J* = 12 Hz, 4H), 1.60–1.50(*m*, 6H), 1.34–1.31 (*m*, 2H). 13C NMR, (CDCl3, 125. MHz) δ = 160.22, 145.27, 136.89, 128.66, 126.30, 122.75, 122.64, 83.14, 83.06, 64.20, 50.66, 48.99, 28.64, 24.91, 23.95. DEPT-135◦. 131.59, 121.00, 117.44, 117.34, 77.85, 58.93 (CH, UP), 45.40, 43.72, 23.37, 19.64, 18.68 (CH2, DOWN). IR (KBr): *υ* cm−<sup>1</sup> = 3418 (br), 1616 (s), 1598 (s), 1370 (s). UV vis (CH3OH, nm (ε, M−<sup>1</sup> cm−1)): λ = 275 (32170), 234(31530); ESI[MS]+ in methanol: *m*/*z* (calcd (found)) 429.28 (429.69 for **L**+H+; 100% abundance); 451.26 (451.69 for **L**+Na+; 90% abundance). Elemental data: Calc (found) for C26H32N6.4H2O: C 62.38 (62.44), H 8.05 (7.74), N 16.79 (16.22)%.

#### *4.2. Synthesis of Complexes*

**General synthetic procedure for compounds 1**–**3**. The methanolic solution of the ligand **L** (0.001 mmol) and LnCl3 salt (0.001 mmol) was mixed together and allowed for constant stirring at room temperature for 4 h. After completion of the reaction, the solution was evaporated by rotary and the solid was further dried under vacuum.

**Complex 1**. Yield: 75%. IR (KBr): *υ* cm−<sup>1</sup> = 3398 (br), 1623 (s), 1458 (m), 1432 (m) 1400 (s), UV vis (HEPES Buffer, pH 7.4, nm (ε, M−<sup>1</sup> cm−1)): λ = 237 (21,392), 285 (20,212); ESI-[MS]<sup>+</sup> in methanol: *m*/*z* calcd(found) 687.10 (687.12) for ([Eu**L**(Cl)3+H+]; 65% abundance), 651.13(651.15) for ([Eu**L**(Cl)2] +; 100% abundance). Elemental data: Calc (found) for C26H70Cl3EuN6O19, C 30.34(30.54), H 6.86 (6.51), N 8.17 (8.24)%.

**Complex 2**. Yield: 68%. IR (KBr): *υ* cm−<sup>1</sup> = 3436 (br), 1629 (s), 1459 (m), 1431 (m) 1386 (s), -UV vis (HEPES Buffer, pH 7.4, nm (ε, M−<sup>1</sup> cm−1)): λ = 236 (16,825), 287 (14,175); ESI-[MS]+ in methanol: *m*/*z* calcd(found) 706.27 (706.21) for ([Sm(**L-2H).4**(CH3OH)]+; 100% abundance); 770.33 (770.27) for ([Sm**L-2H).6**(CH3OH)]+; 100% abundance). Elemental data: Calc (found) for C26H66Cl3N6O17Sm: C 31.49(31.11), H 6.71(6.65), N 8.48(8.54)%.

**Complex 3**. Yield 65%. IR (KBr): *υ* cm−<sup>1</sup> = 3432 (br), 1627(s), 1459 (m), 1432 (m) 1390 (s). -UV vis (HEPES Buffer, pH 7.4, nm (ε, M−<sup>1</sup> cm−1)): λ = 236 (215,725), 285 (13,917); ESI-[MS]+ in methanol: *m*/*z* calcd(found) 693.11.(693.09) for ([Tb**L**(Cl)3+H+]; 90% abundance); 657.13 (657.12) for ([Tb**L**(Cl)2] +; 100% abundance). Elemental data: Calc (found) for C26H70Cl3N6O19Tb: C 30.14(29.68), H 6.81(7.05), N 8.11(8.34)%.

Detection limit (DL) calculation:

$$\text{DL} = \text{CL} \times \text{ET} \tag{4}$$

where DL = detection limit, CL = Concentration of complex, and ET = Equivalent of Titrant at which change was observed. Here, the titrant is phosphate.

**Supplementary Materials:** The following are available online at https://www.mdpi.com/2079-4991/11 /1/53/s1, Figure S1: 1H NMR of **L** in CDCl3, Figure S2: 13C NMR of **L** in CDCl3, Figure S3: DEPT-135◦ NMR of **L** in CDCl3, Figure S4: ESI-MS Spectrum of **L**, Figure S5:CD spectra of Ligand **L** in CHCl3, Figure S6: IR-Spectra of **L**, **1**, **2**, and **3**, Figure S7: ESI-MS Spectrum of **1**, Figure S8: ESI-MS Spectrum of **2**, Figure S9: ESI-MS Spectrum of **3**, Figure S10: Normalization Spectra, Figure S11: Emission Curve of **1** against HPO4 <sup>2</sup>−, Figure S12: ESI-MS spectrum of [1]:2[HPO4 <sup>2</sup>−], Figure S13: Non-linear fit curve of **1** against HCO3 −, Figure S14: Emission curve of **2** against HPO4 <sup>2</sup>−, Figure S15: ESI-MS spectrum of [2]:2[HPO4 <sup>2</sup>−], Figure S16: Emission curve of **3** against HPO4 <sup>2</sup>−, Figure S17: ESI-MS spectra of [3]:2[HPO4 <sup>2</sup>−], Figure S18: (a) UV-vis spectra of ligand **L** and its complexes **1**, **2** and **3** and (b) possible energy transfer, Figure S19: Excited state lifetime of complex **1** with HPO4 <sup>2</sup><sup>−</sup> with 1:1 ratio, Figure S20: Excited state lifetime of complex **1** with HPO4 <sup>2</sup><sup>−</sup> with 1:2 ratio, Figure S21: Excited state lifetime of complex **1** with HPO4 <sup>2</sup><sup>−</sup> with 1:10 ratio, Table S1: Quantum yield calculation for complex **1**, **2** and **3**.

**Author Contributions:** Concept, J.S., P.S.S.; methodology, J.S., P.S.S.; experiments and data collections, J.S., S.J., P.B.C.; validation, J.S., P.S.S., H.S.J.; writing—original draft, J.S., P.S.S., H.S.J.; formatting, H.S.J.; funding acquisition, H.S.J. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Data used to support the findings of this study are included within the article and supporting material.

**Acknowledgments:** The manuscript has been assigned CSMCRI communication number CSIR-CSMCRI 193/2015. J.S. acknowledges CSIR, Govt. of India, New Delhi for the CSIR-SRF award. The members of Analytical Division and Centralized Instrument Facility (AD&CIF) of CSIR-CSMCRI are gratefully acknowledged for their analytical supports. H.S.J. thanks FWO [PEGASUS]<sup>2</sup> Marie Sklodowska-Curie grant agreement No. 665501 for the incoming Post-doctoral Fellowship.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


### *Article* **Quenching of the Eu3+ Luminescence by Cu2+ Ions in the Nanosized Hydroxyapatite Designed for Future Bio-Detection**

**Katarzyna Szyszka 1,\*, Sara Targo ´nska 1, Agnieszka Lewi ´nska 2, Adam Watras <sup>1</sup> and Rafal J. Wiglusz 1,3,\***


**Abstract:** The hydroxyapatite nanopowders of the Eu3+-doped, Cu2+-doped, and Eu3+/Cu2+-codoped Ca10(PO4)6(OH)2 were prepared by a microwave-assisted hydrothermal method. The structural and morphological properties of the products were investigated by X-ray powder diffraction (XRD), transmission electron microscopy techniques (TEM), and infrared spectroscopy (FT-IR). The average crystal size and the unit cell parameters were calculated by a Rietveld refinement tool. The absorption, emission excitation, emission, and luminescence decay time were recorded and studied in detail. The 5D0 <sup>→</sup> 7F2 transition is the most intense transition. The Eu3+ ions occupied two independent crystallographic sites in these materials exhibited in emission spectra: one Ca(1) site with C3 symmetry and one Ca(2) sites with Cs symmetry. The Eu3+ emission is strongly quenched by Cu2+ ions, and the luminescence decay time is much shorter in the case of Eu3+/Cu2+ co-doped materials than in Eu3+-doped materials. The luminescence quenching mechanism as well as the schematic energy level diagram showing the Eu3+ emission quenching mechanism using Cu2+ ions are proposed. The electron paramagnetic resonance (EPR) technique revealed the existence of at least two different coordination environments for copper(II) ion.

**Keywords:** apatite; europium ions; cooper ions; photoluminescence spectroscopy; EPR spectroscopy

#### **1. Introduction**

Apatite-type materials can be applied in many industrial fields, e.g., as sorbents, biocompatible and biodegradable materials for bone and teeth reconstruction, catalysts, materials for the wastewater treatment, fertilizers, and luminescent materials [1,2]. Hydroxyapatite (Ca10(PO4)6(OH)2—abbr. HAp) is used in medicine as a bone implant material due to its biocompatibility, bioactivity, and similarity to bone mineral [3,4]. However, it is still widely investigated in order to improve its properties by obtaining appropriate grain size, morphology, mechanical strength, and solubility and by adding some dopants that are, e.g., naturally built into bone apatite, ions possessing antibacterial properties, or ions enabling bio-imaging [5,6]. Infections after grafting of bone implant material are a serious problem in surgery. The idea is that doping with antibacterial ions into the grafted biomaterial will prevent bacterial biofilm formation and infection development. Inorganic antibacterial agents possess advantages such as stability and safety. The antibacterial agents include ions such as copper, silver, and zinc [7–9], and several studies have shown that they can play important roles in the prevention or minimization of initial bacterial adhesion [10,11]. Metal ions can react with microbial membrane, causing structural changes and permeability. Then, Ag+ and Cu2+ ions have the ability to complex anions such as –NH2, –S–S–, and –CONH– of the proteins or enzymes in the bacterial cells. It provides bacterial DNA and RNA damage and inhibits proliferation [7,11,12]. Copper is an essential microelement that

**Citation:** Szyszka, K.; Targo ´nska, S.; Lewi ´nska, A.; Watras, A.; Wiglusz, R.J. Quenching of the Eu3+ Luminescence by Cu2+ Ions in the Nanosized Hydroxyapatite Designed for Future Bio-Detection. *Nanomaterials* **2021**, *11*, 464. https://doi.org/10.3390/ nano11020464

Academic Editor: Jiangshan Chen

Received: 31 December 2020 Accepted: 8 February 2021 Published: 11 February 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

is involved in many metabolic processes that taken place in human bodies [7,11,13,14]. However, copper ions may have potentially toxic effects at higher amount in human beings due to their ability to generate ROSs (Reactive Oxygen Species). On the other hand, recent studies have shown that copper-doped apatite-type materials are very promising as a new kind of ow-toxic pigment that can be used in the paint and varnish industry [1,15–18].

Lanthanide(III)-doped nanomaterials are promising candidates for fluorescent biolabels due to their stable luminescence over time, high photochemical stability, sharp emission peak, low levels of photobleaching, and toxicity compared with organic fluorophores. Eu3+ ions are structural and luminescence probe-sensitive to changes in the local environment around the ion. Furthermore, the luminescence of Eu3+ ions are identified by a narrow emission band and long lifetimes of the excited state [19,20].

Apatite is a big family of compounds, and it is widely investigated due to its outstanding properties such as good biocompatibility or possibility to be doped with different ions in a broad concentration range for applications in the industry, in medicine, etc. There are a lot of papers focusing on apatite synthesis [21–24]; its doping with antibacterial ions [9,10,25]; as well as its doping with luminescence ions such as Eu3+ [5,19,26,27], Tb3+ [28,29], Eu3+/Tb3+ [30], Er3+/Yb3+, or Eu3+/Cu2+ [17]. Moreover, there is a lot of research focusing on anion-substituted apatite such as silicate [31,32], vanadate [33], borate [34], or carbonate [35].

In the presented work, the synthesis, structural, morphological, and luminescence properties of Eu3+/Cu2+ co-doped HAp were investigated attentively. To the best of our knowledge, this is the first time that the quenching mechanism in this system has been elucidated.

#### **2. Materials and Methods**

#### *2.1. Synthesis*

The Ca10(PO4)6(OH)2 nanopowders doped with Eu3+ and Cu2+ ions were synthesized by a microwave-assisted hydrothermal method. The starting materials used were CaCO3 (99.0%, Alfa Aesar, Karlsruhe, Germany), NH4H2PO4 (99.0%, Fluka, Bucharest, Romania), Eu2O3 (99.99%, Alfa Aesar, Karlsruhe, Germany), Cu(NO3)2·2.5H2O (98.0–102.0%, Alfa Aesar, Karlsruhe, Germany), and NH3·H2O (99%, Avantor, Gliwice, Poland) as a pH regulation reagent. The concentration of dopants was calculated based on inductively coupled plasma-optical emission spectrometer (ICP-OES) results. The concentrations of europium ions were 0.5 mol%, 1 mol%, 2 mol%, and 5 mol%, and the concentrations of the cooper ions were 2 mol% and 5 mol% to the overall molar content of calcium cations. First, the stoichiometric amounts of CaCO3 as well as Eu2O3 were separately digested in excess of HNO3 (suprapur Merck, Darmstadt, Germany) to obtain water-soluble nitrates. The obtained europium nitrate hydrate was recrystallized three times to remove excess HNO3. Then, the stoichiometric amount of Eu(NO3)3 was dissolved in deionized water, and then, the Cu(NO3)2 was added to the stoichiometric amount of calcium nitrate. After this, NH4H2PO4 was added to the abovementioned mixture and the pH value was adjusted to 9 by ammonia. The suspension was transferred to a Teflon vessel and was placed into the microwave reactor (ERTEC MV 02-02, Wrocław, Poland). The reaction system was heat-treated at 280 ◦C for 90 min under autogenous pressure of 60 atm. The obtained product was washed several times with deionized water and dried at 70 ◦C for 24 h.

#### *2.2. Powder Characterization*

The crystal structure and phase purity were studied using a PANalytical X'Pert Pro diffractometer (Malvern Panalytical Ltd., Malvern, UK) equipped with Ni-filtered Cu Kα radiation (V = 40 kV and I = 30 mA). The recorded X-ray powder diffraction patterns (XRD) were compared with the reference standard of hexagonal calcium hydroxyapatite (P63/m) from the Inorganic Crystal Structure Database (ICSD-2866) and analyzed. Rietveld structural refinement was performed with the aid of a Maud program (version 2.93) (University of Trento-Italy, Department of Industrial Engineering, Trento, Italy) [36,37]

based on the apatite hexagonal crystal structure with better approximation and indexing of the Crystallographic Information File (CIF). The quality of structural refinement was checked by R-values (Rw, Rwnb, Rall, Rnb, and σ), which were followed to get a structural refinement with better quality and reliability.

The morphology was investigated by high-resolution transmission electron microscopy (HRTEM) using a Philips CM-20 SuperTwin microscope (Eindhoven, The Netherlands), operating at 200 kV. The specimen for the HRTEM measurement was obtained by dispersing a small amount of powder in methanol and by putting a droplet of the suspension onto a copper microscope grid covered with carbon.

Fourier transform infrared spectra were measured using a Thermo Scientific Nicolet iS50 FT-IR spectrometer (Waltham, MA, USA) in the range of 4000–400 cm−<sup>1</sup> at 295 K. Absorption spectra were recorded with an Agilent Cary 5000 spectrophotometer, employing a spectral bandwidth (SBW) of 0.1 nm in the visible and ultraviolet range and of 0.7 nm in the infrared. The spectra were recorded at room temperature.

The excitation spectra were recorded with the aid of an FLS980 Fluorescence Spectrometer (Edinburgh Instruments, Kirkton Campus, UK) equipped with 450 W Xenon lamp. The excitation of 300 mm focal length monochromator was in Czerny–Turner configuration and the excitation arm was supplied with holographic grating of 1800 lines/mm grating blazed at 250 nm. The excitation spectra were corrected to the excitation source intensity. The emission spectra were measured by using a Hamamatsu PMA-12 photonic multichannel analyzer (Hamamatsu, Hamamatsu City, Japan) equipped with BT-CCD line (Hamamatsu, Hamamatsu City, Japan). As an excitation source, a pulsed 266 nm line of Nd:YAG laser (3rd harmonic; LOTIS TII, Minsk, Belarus) was chosen (ƒ = 10 Hz, t < 10 ns). The detection setup was calibrated and had a flat response for the whole working range (350–1100 nm). The measurements were carried out at 300 K.

The time-resolved luminescence spectrum was obtained by recording decay curves during changing observed wavelength and by creating a two-dimensional map (i.e., intensity vs. time and wavelength). It was recorded by an in-house developed software that controlled the equipment. A Dongwoo Optron DM711 monochromator (Hoean-Daero, Opo-Eup, Gyeonggi-Do, Korea) with a focal length of 750 mm was used to select the observed wavelength, while luminescence decay curves were acquired with a Hamamatsu R3896 photomultiplier (Hamamatsu, Hamamatsu City, Japan) connected to a digital Tektronix MDO 4054B oscilloscope (Bracknell, UK). An optical parametric oscillator Opotek Opolette 355 LD (Carlsbad, CA, USA) emitting 5 ns pulses was used as an excitation source.

The luminescence kinetics were measured by using a Jobin-Yvon THR1000 monochromator (HORIBA Jobin-Yvon, Palaiseu, France) equipped with a Hamamatsu R928 photomultiplier (Hamamatsu, Hamamatsu City, Japan) as a detector and a LeCroy WaveSurfer as a digital oscilloscope (Teledyne LeCroy, Chestnut Ridge, NY, USA). As an excitation source, a pulsed 266 nm line from an Nd:YAG laser was used. The luminescence kinetics were monitored at 618 nm according to the most intense electric dipole transition ( 5D0 → 7F2), and the effective emission lifetimes were calculated using the following equation:

$$\tau\_{\rm{fl}} = \frac{\int\_0^\infty tI(t)dt}{\int\_0^\infty I(t)dt} \cong \frac{\int\_0^{t\_{\rm{max}}} tI(t)dt}{\int\_0^{t\_{\rm{max}}} I(t)dt} \tag{1}$$

where *I*(*t*) is the luminescence intensity at time *t* corrected for the background and the integrals are calculated over the range of 0 < *t* < *t*max, where *t*max >> *τm*.

The effective content of elements was determined by using an Agilent 720 bench-top optical emission spectrometer with inductively coupled Ar plasma (Ar-ICP-OES) and was corrected to an effective value. The ICP standard solutions were used to record the calibration curves to determine the Ca2+, P5+, Cu2+, and Eu3+ ion content. The samples for elemental analysis were prepared by digesting in the pure HNO3 acid (65% suprapur Merck).

The electron paramagnetic resonance (EPR) spectra were measured at 295 K and 77 K using a Bruker Elexsys 500 CW-EPR (Bruker GmbH, Rheinstetten, Germany) spectrometer operating at the X-band frequency (≈9.7 GHz), equipped with frequency counter (E 41 FC) and NMR teslameter (ER 036TM). The spectra were measured with a modulation frequency of 100 kHz, microwave power of 10 mW, modulation amplitude of 10 G, time constant of 40 ms, and a conversion time of 160 ms. The first derivative of the absorption power was recorded as a function of the magnetic field value. An analysis of the EPR spectra was carried out using the WinEPR software package, version 1.26b (Bruker WinEPR GmbH, Rheinstetten, Germany).

#### **3. Results and Discussion**

#### *3.1. Structural Analysis*

The structural characterization of the HAp nanocrystals doped with xEu3+ (where x = 0.5, 1, and 3 mol%) and co-doped with xEu3+ and yCu2+ (where x = 0.5, 1, and 4 mol% and y = 0.5 and 1 mol%) was carried out by powder X-Ray diffraction measurements as a function of doping ion concentration (see Figure 1). Detectable crystallinity and pure hexagonal phase corresponding to the reference standard (ICSD—180315 [38]) were observed. Only in the case of the 4 mol% Eu3+/0.5 mol% Cu2+:HAp, an extra peak at 29.5◦ of 2θ was observed (assigned as asterisk in Figure 1).

**Figure 1.** X-ray powder diffraction patterns of the Eu3+-doped and Eu3+/Cu2+ co-doped Ca10(PO4)6(OH)2.

Structural refinement was performed to obtain the unit cell parameters and the average grain sizes of synthesized materials. Hexagonal phase formation and the successful incorporation of Eu3+ and Cu2+ ions were verified. The theoretical fit with the observed XRD pattern was found to be in good agreement, which indicated the success of the Rietveld refinement method (see Figure 2). More details are displayed in Table 1. As can be seen, it was possible to observe an increase in the cell volume and a parameters with the increase in Eu3+ ion concentration in single-doped materials, which was caused by a

smaller ionic radii of the dopant (Ca2+ (coordination number—CN9), 1.18 Å; Eu3+ (CN9), 1.12 Å; Ca2+ (CN7), 1.06 Å; and Eu3+ (CN7), 1.01 Å) [39]. Moreover, shrinkage of the average grain size with an increase in the Eu3+ ion concentration in the host lattice was observed. In the case of co-doped materials, no straightforward dopant concentration dependence on cell parameters (a, c, and V) or average grain size was observed.

**Figure 2.** Representative results for the 0.5 mol% Eu3+/1 mol% Cu2+:Ca10(PO4)6(OH)2:, Rietveld analysis (red—fitted diffraction, green—differential pattern, and blue column—reference phase peak position).

**Table 1.** Unit cell parameters (a and c), cell volume (V), grain size, as well as refine factor (RW) for the Eu3+-doped and Eu3+/Cu2+ co-doped Ca10(PO4)6(OH)2.


The morphology of the calcium hydroxyapatite was investigated by HRTEM. Nanoparticles are crystalline in nature and elongated, as can be seen in Figure 3. The particle size distribution is relatively wide, and the mean grain sizes of particle is in the range between 60 and 120 nm in length and about 40 nm in width.

**Figure 3.** Representative TEM images (**a**–**c**) and particle size distribution (**d**) of the 1 mol% Eu3+:Ca10(PO4)6(OH)2.

The infrared spectra of the copper-doped, europium-doped, and co-doped hydroxyapatite materials are presented in Figure 4. The most intense peaks are the triply degenerated antisymmetric stretching bands of phosphate groups ν3(PO4 <sup>3</sup>−) located at 1044.5 cm−<sup>1</sup> and 1097.8 cm−1. The peaks observed at 566.0 cm−<sup>1</sup> and 603.1 cm−<sup>1</sup> correspond to the triply degenerated ν4(PO4 <sup>3</sup>−) vibrations. The peaks at 963.0 cm−<sup>1</sup> are assigned to the non-degenerated symmetric stretching ν1(PO4 <sup>3</sup>−) band. Two peaks corresponding to OH<sup>−</sup> group at 3571.5 cm−<sup>1</sup> and 633.5 cm−<sup>1</sup> are observed on the infrared spectra. The existence of these peaks clearly confirms the hydroxyapatite structure with a hydroxyl group in the host lattice. The broad bands between 3690 and 3290 cm−<sup>1</sup> were connected with H2O vibration.

**Figure 4.** Infrared spectra of the Eu3+-doped and Eu3+/Cu2+ co-doped Ca10(PO4)6(OH)2.

#### *3.2. Absorption, Excitation, and Emission Spectra*

The absorption spectra of the pure, copper-doped, europium-doped, and co-doped hydroxyapatite nanopowders were recorded in the visible range from 350 nm to 800 nm at room temperature (see Figure 5). The pure hydroxyapatite matrix is transparent for these wavelengths. The copper-doped materials absorbed the blue radiation in the range from 350 nm to 420 nm. All materials are relatively transparent for the radiation from 450 nm to 550 nm of the wavelength. The copper-doped materials absorbed the radiation from 550 nm to 800 nm, and the absorption coefficient increases with the increase in wavelength. The broad absorption band is attributed to the 2E → 2T2 intra-configurational (d-d) transition of the Cu2+ ions [40,41]. In the absorption spectra, the peaks related to the 4f-4f transitions of Eu3+ ions are observed. These peaks are attributed to the following transitions: the 7F0 → 5D4, 5L8 at 362 nm and 7F0 → 5G6, 5L7, 5G3 at 376 nm. The 7F0 → 5L6 transition with a maximum at 394 nm was observed in the case of europium and the copper co-doped materials. This transition is the most intense f-f transition of Eu3+ ions.

The excitation emission spectra, which were recorded at room temperature by monitoring the intense red emission at 618 nm (5D0 → 7F2), of investigated materials are presented in Figure 6. The representative excitation spectra of the 1 mol% Eu3+:HAp and 1 mol% Eu3+/1 mol% Cu2+:HAp are presented. As demonstrated, the excitation spectra consisted of visible intra-configurational 4f-4f transitions with sharp lines characteristic of Eu3+ ions. Particularly, these narrow bands located at around 320, 363, 383, 395, 416, and 466 nm originated from the 7F0 → 5HJ; 7F0 → 5D4, 5L8; 7F0 → G2, 5L7, 5G3; 7F0 → 5L6; 7F0 → 5D3 transitions of Eu3+ ions, respectively [31,35]. The absorption peak of the 7F0 → 5HJ transition at 320 nm indicates that the energy band-gap of HAp is considerably larger than that in, e.g., Eu2Ti2O7 oxide [42], in which the 5H3,6-related transition peak is completely masked by the charge transfer band due to its lower energy band-gap nature. The f-f

electron transitions are weakly affected by the crystal field; thus, their positions remain almost steady due to good isolations of lanthanide's f orbitals by an external shell [19,31,43]. As can be seen, the intensity of the emission excitation spectra is much lower in the case of the co-doped material than in that single doped with Eu3+ in the HAp host.

**Figure 5.** The absorption spectra of the Eu3+-doped and Eu3+/Cu2+-co-doped Ca10(PO4)6(OH)2.

**Figure 6.** The excitation spectra of the 1 mol% Eu3+ and 1 mol% Eu3+/1 mol% Cu2+-co-doped Ca10(PO4)6(OH)2.

The spectroscopic properties of Eu3+ ions allow us to receive vital information about the symmetry of the Eu3+ ions surrounding the crystal lattice; the amount of crystallographic positions; and therefore, potential sites of substitution, structural changes occurring in the matrix caused by external factors, etc. The emission spectra of the Eu3+ ions consist of characteristic bands present in the red region of the electromagnetic radiation assigned to the electron transitions developing in the 4f-4f shell of Eu3+ ions. The 5D0 → 7F0,1,2 transitions are the most important in analysis, particularly in correlation with the structural properties. The 5D0 → 7F0 transition can provide direct information about the number of crystallographic sites occupied by Eu3+ ions in the host lattice [19].

The emission spectra of y mol% Eu3+:HAp (where y—0.5, 1, and 3 mol%) and x mol% Eu3+/y mol% Cu2+:HAp (where x—0.5, 1, and 4 mol% and y—0.5 and 1 mol%) were measured by excitation wavelength at 266 nm at room temperature and are shown in Figure 7. The spectra were normalized to the 5D0 → 7F1 magnetic dipole transition. The emission spectra were dominated by an intense red emission band situated at about 618 nm corresponding to the 5D0 → 7F2 transition of Eu3+ ions. Meanwhile, four weaker emission bands peaking at around 578, 589, 652, and 698 nm were also detected and ascribed to the 5D0 → 7F0, 5D0 → 7F1, 5D0 → 7F3, and 5D0 → 7F4 transitions of Eu3+ ions, respectively [31,35]. The presence of a 5D0 → 7F0 transition confirms that europium ions are located in a low-symmetry environment. Furthermore, the number of lines directly indicate the number of occupied crystallographic positions in the investigated lattice by Eu3+ ions. In the apatite molecule, ten calcium atoms are found in two non-equal crystallographic positions, in agreement with the results showed in Figure 7.

**Figure 7.** The emission spectra of the Eu3+-doped and Eu3+/Cu2+ co-doped Ca10(PO4)6(OH)2.

In Figure 8, the time-resolved emission spectrum of the 1 mol% Eu3+/1 mol% Cu2+:HAp is presented.

**Figure 8.** Representative emission spectra map of the 1 mol% Eu3+/1 mol% Cu2+:Ca10(PO4)6(OH)2.

#### *3.3. Decay Profiles*

The luminescence decay curves were registered and analyzed for the synthesized materials to determine the comprehensive characteristics of the luminescence properties. The decay curves presented in Figure 9 are not single-exponential, which is compatible with the presence of nonequivalent crystallographic sites of Eu3+ ions accordingly. The lifetimes values were calculated as the effective emission decay time by using Equation (1). The average lifetimes obtained for the sample single-doped by Eu3+ are equal to 0.93; 0.82, and 0.91 for concentrations of optically active ions at 0.5, 1.0, and 3.0 mol%, respectively.

The emission kinetic of Eu3+ ions strongly depends on the presence of Cu2+ ions, which effectively quenched the 5D0 level. The average lifetime obtained for co-doped materials is much shorter than that for Eu3+-doped materials, and the decay time values are estimated at about 0.33, 0.22, and 0.23 ms for the 0.5 mol% Eu3+/1 mol% Cu2+, 1 mol% Eu3+/1 mol% Cu2+, and 4 mol% Eu3+/0.5 mol% Cu2+ co-doped materials, respectively. The emission quenching of the Eu3+ ions may be interpreted as nonradiative energy transfer between the Eu3+ and Cu2+ ions. The efficiency of energy transfer was estimated by Equation (2) for pairs of materials: single-doped with Eu3+ ions and co-doped with Eu3+ and Cu2+ ions with the same concentration of Eu3+ ions. The calculated lifetimes of Eu3+ ions (donor) in the absence and presence of Cu2+ ions (acceptor) are used [40,44]. The results of energy transfer efficiency are presented in Table 2.

$$\mathfrak{m}\_{\text{Eu}^{3+} \to \text{Cu}^{2+}} = 1 - \left(\frac{\mathsf{T}\_{\text{Eu}^{3+} \to \text{Cu}^{2+}}}{\mathsf{T}\_{\text{Eu}^{3+}}}\right) \tag{2}$$

**Table 2.** The average lifetime of Eu3+-doped (τEu), Eu3+/1 mol% Cu2+ (τEu→Cu) co-doped Ca10(PO4)6(OH)2 and energy transfer efficiency (ηEu→Cu).


**Figure 9.** Decay times of the Eu3+-doped and Eu3+/Cu2+ co-doped Ca10(PO4)6(OH)2.

The obtained efficiency is equal to 65 and 73% for 0.5 mol% Eu2+/1 mol% Cu2+ and 1 mol% Eu2+/1 mol% Cu3+ co-doped HAp, respectively. The observed luminescence properties of europium ions in apatite lattice in the presence of copper ions are dominated by emission quenching of Eu3+ by Cu2+ ions. With an increase in Eu3+ concentration, the efficiency of quenching grew. This would suggest that the relatively huge probability of Eu3+ → Cu2+ nonradiative energy transfer probably behaves by electric dipole interaction [40,44].

The simplified energy level diagram of Eu3+ and Cu2+ ions was proposed and is shown in Figure 10 in order to explain the quenching mechanism occurring in hydroxyapatite codoped with Eu3+ and Cu2+ ions. When the materials were excited by 266 nm wavelength, a charge transfer transition O2<sup>−</sup> → Eu3+ occurred. Then, nonradiative relaxation to the 5D0 first excited state was performed. From this state, the energy can be relaxed in two ways: by radiative transition to the ground state of Eu3+ ions (7F0–6) or by energy transfer to the 2T2g energy level of Cu2+ and then nonradiative relaxation to the ground state of Cu2+ ions. These two manners of energy relaxation compete among themselves, and doping with Cu2+ ions causes Eu3+ emission quenching.

**Figure 10.** Simplified energy level scheme of Eu3+ and Cu2+ explaining quenching of Eu3+ ion emission.

#### *3.4. The EPR Spectra Analysis*

The EPR spectroscopy is especially predisposed to identifying the structural properties of paramagnetic compounds. The unpaired electron interacts (couples) with the nuclear spin (I) to form a 2I + 1 line hyperfine structure centered on g and spaced with the distance quantified by the hyperfine coupling parameter A. The coupling between the nuclear and electron spins becomes stronger as the A parameter becomes larger. The combination of g and A parameters can be utilized to differentiate between electron environments of ion.

There are two distinct Ca coordination sites in the HAp unit cell, that is the Ca(1) site with the Ca2+ ion surrounded by 9 oxygen atoms from 6 PO4 <sup>3</sup><sup>−</sup> groups and the Ca(2) site with the Ca2+ ion surrounded by 7 oxygen atoms from the 5 PO4 <sup>3</sup><sup>−</sup> and 1 OH<sup>−</sup> anions. The Ca2+ ions in both coordination sites can be replaced by Eu3+ and Cu2+ ions [45]. The EPR properties of trivalent europium (Eu3+) is relatively little because it is a non-Kramer ion, and its EPR spectrum should be silent because of the short spin-lattice relaxation time [46]. Therefore, in the EPR spectra recorded for the samples, only signals due to Cu2+ ions are observed.

The spectra recorded at room temperature and at 77 K (Figure 11) are anisotropic as a consequence of the Jahn–Teller effect operating for the d<sup>9</sup> electron configuration of Cu2+ ions that leads to considerable departure from a regular symmetry of the coordination sphere. The spectra reveal a weakly resolved hyperfine interaction between the spins of unpaired electrons and copper nuclei (I = 3/2), which for powder spectrum suggest a large distance between paramagnetic centers (Cu2+ ions).

**Figure 11.** Experimental and simulated electron paramagnetic resonance (EPR) spectra of the 1 mol% Eu3+/1 mol% Cu2+:Ca10(PO4)6(OH)2.

Spectral analysis revealed the existence of at least two different coordination environments for copper(II) ions. The EPR spectrum of the 1 mol% Eu3+/1 mol% Cu2+ co-doped HAp can be decomposed into two superimposed resonance signals due to two different Cu(II) coordination sites, hereinafter referred as "a" and "b". This stays in line with the fact that there are two distinct Ca coordination sites in HAp in which Cu(II) can be doped. From the simulation of the spectrum recorded at 77 K, the estimated parameters are gz(a) = 2.41, gz(b) = 2.37, gy = 2.11, and gx = 2.08 with Az(a) = Az(b) = 110 G. However, the accuracy of these parameters is inevitably limited due to the fact that the Cu(II) signals are superimposed.

The observed EPR parameters contrast their counterparts determined for synthetic hydroxyapatite doped with Cu2+, in which also two different Cu(II) coordination sites were identified: gz = 2.485, gy = 2.17, gx = 2.08, and Az = 52 G for one coordination site and gz(a) = 2.420, gy = 2.17, gx = 2.08, and Az = 92 G for the second [47]. This difference in gz and Az parameters clearly stems from the structural divergence between the 1 mol% Eu3+/1 mol% Cu2+:HAp and SHA. At the same time, the gz and Az parameters for 1 mol% Eu3+/1 mol% Cu2+:HAp are similar to the ones reported for copper(II) ions bonded to lattice oxygens in montmorillonite((Cu(AlO)n(H2O)4-n)x): gz and Az in the ranges 2.37–2.41 and 100–140 G, respectively [48]. Therefore, the geometry of Cu(II) coordination sites in 1 mol% Eu3+/1 mol% Cu2+:HAp are expected to structurally resemble ((Cu(AlO)n(H2O)4-n)x).

Trends were found that enabled the Cu(II) EPR parameters to be correlated to the copper(II) ligands and the overall charge of the complexes [49–52]. According to these general trends, the gz and Az parameters for 1 mol% Eu3+/1 mol% Cu2+:HAp are characteristic of positively charged Cu–O complexes [49,50]. This fact indicates that, in the crystal lattice of the 1 mol% Eu3+/1 mol% Cu2+:HAp, the negative charge of the PO4 <sup>3</sup><sup>−</sup> and OH<sup>−</sup> anions are primarily neutralized by remained Ca2+ cations. Moreover, the observed difference between gz values found for two Cu coordination sites can be used to determine their possible assignment to Ca(1) and Ca(2). The increase in gz is associated with the rise in positive charges for the Cu–O complex. Hence, the higher value of gz(a) indicates that this Cu(II) ion is surrounded by a lower number of oxygen atoms, which are the primary

carriers of a negative charge, and therefore should be labeled as Cu(II) ion doped into the Ca(2) site.

#### **4. Conclusions**

The pure crystal hydroxyapatite powder doped co-doped with Eu3+ and Cu2+ ions was successfully synthesized by a microwave-assisted hydrothermal method that was confirmed by the X-ray powder diffraction method. The nanometric size of the obtained materials was confirmed by Rietveld refinement and TEM techniques. In the absorption spectra, the transitions occurring in Eu3+ as well as Cu2+ ions were observed. In the emission spectra, the typical transition of Eu3+ ions (5D0 <sup>→</sup> 7FJ) were observed and the 5D0 <sup>→</sup> 7F2 transition is the most intense. The 5D0 <sup>→</sup> 7F0 transition consists of two lines, which means that the Eu3+ ions are localized in two independent crystallographic sites: in Ca(1) with C3 point symmetry and in Ca(2) with Cs symmetry. The emission decay times of Eu3+/Cu2+:HAp are much shorter than the decay times of Eu3+:HAp, which indicates that the Eu3+ emission is quenched by the Cu2+ ions. The simplified energy level diagram was proposed, and the quenching mechanism was explained. Based on the EPR measurement, the existence of at least two different coordinations surrounding copper(II) ions was detected.

**Author Contributions:** Conceptualization, K.S. and R.J.W.; methodology, K.S.; investigation, K.S., S.T., and A.L.; data curation, K.S. and S.T.; writing—original draft preparation, K.S.; writing—review and editing, K.S., S.T., A.L., A.W., and R.J.W.; visualization, K.S.; supervision, K.S. and R.J.W.; project administration, R.J.W.; funding acquisition, R.J.W. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the National Science Centre (NCN) within the projects "Preparation and characterisation of biocomposites based on nanoapatites for theranostics" (No. UMO-2015/19/B/ST5/01330) and "Elaboration and characteristics of biocomposites with anti-virulent and anti-bacterial properties against Pseudomonas aeruginosa" (No. UMO-2016/21/B/NZ6/01157).

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Not applicable.

**Acknowledgments:** We are grateful to E. Bukowska for the XRD measurements and to J. Komar for help with the emission spectra map measurements.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


MDPI St. Alban-Anlage 66 4052 Basel Switzerland Tel. +41 61 683 77 34 Fax +41 61 302 89 18 www.mdpi.com

*Nanomaterials* Editorial Office E-mail: nanomaterials@mdpi.com www.mdpi.com/journal/nanomaterials

MDPI St. Alban-Anlage 66 4052 Basel Switzerland

Tel: +41 61 683 77 34 Fax: +41 61 302 89 18

www.mdpi.com

ISBN 978-3-0365-3458-9