*Review* **Current Status on Pulsed Laser Deposition of Coatings from Animal-Origin Calcium Phosphate Sources**

#### **Liviu Duta 1,\* and Andrei C. Popescu 2,\***


Received: 30 April 2019; Accepted: 22 May 2019; Published: 24 May 2019

**Abstract:** The aim of this paper is to present the current status on animal-origin hydroxyapatite (HA) coatings synthesized by Pulsed Laser Deposition (PLD) technique for medical implant applications. PLD as a thin film synthesis method, although limited in terms of surface covered area, still gathers interest among researchers due to its advantages such as stoichiometric transfer, thickness control, film adherence, and relatively simple experimental set-up. While animal-origin HA synthesized by bacteria or extracted from animal bones, eggshells, and clams was tested in the form of thin films or scaffolds as a bioactive agent before, the reported results on PLD coatings from HA materials extracted from natural sources were not gathered and compared until the present study. Since natural apatite contains trace elements and new functional groups, such as CO3 <sup>2</sup><sup>−</sup> and HPO4 <sup>2</sup><sup>−</sup> in its complex molecules, physical-chemical results on the transfer of animal-origin HA by PLD are extremely interesting due to the stoichiometric transfer possibilities of this technique. The points of interest of this paper are the origin of HA from various sustainable resources, the extraction methods employed, the supplemental functional groups, and ions present in animal-origin HA targets and coatings as compared to synthetic HA, the coatings' morphology function of the type of HA, and the structure and crystalline status after deposition (where properties were superior to synthetic HA), and the influence of various dopants on these properties. The most interesting studies published in the last decade in scientific literature were compared and morphological, elemental, structural, and mechanical data were compiled and interpreted. The biological response of different types of animal-origin apatites on a variety of cell types was qualitatively assessed by comparing MTS assay data of various studies, where the testing conditions were possible. Antibacterial and antifungal activity of some doped animal-origin HA coatings was also discussed.

**Keywords:** animal-origin calcium phosphate coatings; natural hydroxyapatite; doping; high adherence; pulsed laser deposition technique; biomimetic applications

#### **1. Introduction**

In the last few decades, the field of bone tissue engineering has been widely studied and expanded for addressing bone-related traumas. By combining biomaterials and cells for bone tissue ingrowth, an efficient and viable alternative to allografts or autografts could be delivered.

Bioactive materials represent a vast bioengineering research field with tremendous interest for the production of durable implants and bone substitutes able to bypass rejection difficulties.

Calcium phosphates (CaP) are bio-ceramic materials used especially for orthopedic and dental medical applications [1,2]. Synthetic hydroxyapatite (HA), with the complex chemical formula Ca10(PO4)6(OH)2, is the most well-known CaP material, and is frequently used in implantology due to its close chemical composition and crystallographic structure resemblance with the mineral phase of vertebrate bones (50% in mass and 70% in volume) [3–5]. We stress upon that the mineral constituent of vertebrate skeletal systems mainly consists of a calcium-deficient HA, doped with various ions [6]. According to the biomimetic approach, a material designed to repair the skeletal system must be similar to the biological one in terms of composition, stoichiometry, the crystallinity degree, morphology, and functionality. It was reported that HA can promote new bone in-growth through the osteoconduction mechanism without eliciting local or systemic toxicity, inflammation, or foreign body response [7–9]. When an HA-based ceramic is implanted, a fibrous tissue-free layer containing carbonated apatite forms on its surfaces and contributes to the implant bonding to the bone. This results in earlier implant stabilization and superior fixation of the implant to the surrounding tissues [8–10]. Moreover, the development of microbial biofilms onto the surface of medical devices or human tissues represents a worrying health problem, which can lead to a high diversity of biofilm-associated infections, with increasing incidence [11]. Therefore, the antimicrobial properties of HA are also of key importance [12,13]. Taking into account all these demands, one could, therefore, explain the large interest for synthesis and deposition of apatites enriched with biologically-active ions or molecules, as well as more resorbable (soluble) CaPs.

Despite its excellent bone regeneration properties, HA has also some important disadvantages: HA-based ceramics are very brittle in bulk [14] and are characterized by poor mechanical properties, especially in liquid media. Therefore, HA-based materials cannot be used in bulk for orthopedic devices, which must withstand the application of high loads during their lifetime [15]. To overcome these drawbacks, HA can be applied as a coating onto the surface of metallic or polymeric implants, which aim to significantly improve implants' overall performances, by successfully combining the excellent bioactivity of the ceramic with the mechanical advantages of the substrate implants [14,16].

Various techniques have been and are continuously developed to obtain HA. In this respect, they can be categorized in two main paths of producing HA: (*i*) the first one implies the use of chemical routes, and (*ii*) the second one involves extracting it naturally, from biogenic, mammalian, or fish bone sustainable, low-cost resources (further denoted as BioHA).

For depositing CaP coatings onto metallic implants, the industrial technique of choice is plasma-spraying, due to the synthesis speed, large area of deposition, and work in an ambient atmosphere [17]. However, HA coatings produced using this technique are prone to cracking and delamination and, because of high-processing temperatures, could contain residual decomposition phases. In this respect, current interests are quickly advancing toward two focused research directions: (*i*) increasing the biomimicry of HA-based coatings with respect to the composition and structure of bone apatite, and (*ii*) improving or even discovering alternative deposition techniques, which can allow for the achievement of novel HA (doped) coatings with increased mechanical and biological characteristics.

When compared to other physical vapor deposition techniques, i.e. thermal evaporation or sputtering, the pulsed laser deposition (PLD) technique stands as a simple, versatile, rapid, and cost-effective method, which can enable precise control of thickness and morphology for the fabrication of high-quality films [18,19]. The main advantage of the PLD technique applied for HA-based bio-ceramics is represented by its capacity to grow stoichiometric films with a controlled degree of crystallinity and thickness. By this method, one can assure the flexibility to control the morphology, phase, crystallinity, and chemical composition of obtained CaPs. These characteristics have a special influence over bio-resorption or dissolution, which are directly involved in the process of films' osseointegration.

Taking into account all these aspects, the aim of the current review is to emphasize the advantages of using animal-origin HA coatings as viable substitutes of synthetic HA ones, which are fabricated by the PLD technique for implantable applications. Conclusions will be drawn, future perspectives will be advanced, and a series of recommendations will be highlighted.

#### **2. Review of Literature**

A digital search was performed usingWeb of Science (http://apps.webofknowledge.com), following the criteria described below.

#### *2.1. Inclusion Criteria*

1. Articles written in English. 2. Use of animal-origin HA materials for physical-chemical, in vitro and in vivo studies. 3. Pulsed laser deposition as the technique used for the synthesis of animal-origin HA coatings. 4. Articles written starting from the year 2001.

#### *2.2. Exclusion Criteria*

1. Non-English articles. 2. Nanoparticles. 3. Other deposition techniques (such as Plasma spraying, Magnetron sputtering, etc.) used for the synthesis of animal-origin HA coatings.

After the review process, a total of 49 articles met the inclusion criteria and were further assessed in detail and parts of the reported results were considered and discussed in this review.

## **3. BioHA** *vs***. Synthetic HA**

Various methods have been used to obtain HA due to its attractive biological properties and resemblance to the mineral part of the bone. There are two main categories of methods used to produce HA.

(*i*) The first one, which is generally utilized due its reliability, implies the use of inorganic synthesis by different methods, such as hydrothermal [20,21], co-precipitation [22,23], or sol-gel [24]. However, these approaches are, on one hand, polluting and time-consuming, and, on the other hand, quite expensive [25,26]. Moreover, the resulting synthetic HA, which is a stoichiometric material with Ca/P ratio of 1.67, does not completely match the chemical composition of bone [27].

(*ii*) Thus, researchers managed to find alternative, low-cost methods to produce HA, such as obtaining it from biogenic, mammalian, or natural fish bone (sustainable/renewable) sources. BioHA is a carbonated, non-stoichiometric Ca-deficient material with a reduced degree of crystallinity. Therefore, it differs from synthetic HA in terms of composition, stoichiometry, crystal size/morphology, crystallinity degree, degradation rate, and overall biological performance. One should also note that there might exist some concerns about the use of natural HA because of the potential risk of dangerous diseases transmission, when the material is not well-prepared [27]. However, when all health security issues are handled [28,29], then HA derived directly from sustainable, low production cost materials (such as animal bones) has a composition that closely matches the morphological and structural architecture of the inorganic components of the human bone. In fact, it has been reported that HA obtained from biowaste such as eggshells, bovine bones, fish-scales, and fish bones can lead to overall properties and behavior comparable or even better than synthetic ones due to the similarities of bone apatites [2,30,31]. However, in order to completely benefit from the previously mentioned advantages against synthetic HA, BioHA should be extracted in a controlled manner, i.e., from resources for which one knows the exact alimentation and way of life of animals, since this could directly influence the overall quality of the final product, HA.

Functional groups from biological apatites can be substituted with trace amounts of cations (Na<sup>+</sup>, Mg2<sup>+</sup>, and K+), anions (F−, Cl−, SiO4 <sup>4</sup><sup>−</sup> and CO3 <sup>2</sup>−), or, in some cases, by both [2,32,33], either adsorbed on the surface of the crystal or incorporated in the lattice structure [34–37]. Specifically, in the apatite structure, the carbonate ions can replace either the hydroxyl or the phosphate ion sites, which leads to a type-A or type-B structure [38]. If these substitutions take place simultaneously, a type-AB substitution occurs, as in the case of the bone mineral [32,39,40]. One should note that the mineral bone substitutions with various trace elements are considered directly responsible for the modifications in crystallinity, solubility, and biological response [41] and, therefore, they play a key role in the performance of hard tissues and the overall osseointegration process. As a consequence, alternative methods used to modify the HA structure by incorporating different ions to improve the osteoconductive properties of synthetic HA is currently of interest to the scientific community [2].

Apatite crystals produced in the biological system are different in many ways from the crystals obtained using synthetic precursors. The apatite crystals grown in the living system bear smaller crystallite size. Therefore, they have a large surface area, which further allows them to absorb an extra number of ions. In short, biological minerals tend to attain an organized structure in a very short time [42].

Table 1 introduces the sample codes, which will be further used in the text.

**Table 1.** Sample acronyms related to different materials used throughout the review and their explanation.


The use of BioHA comes from the ambition and continuous efforts of researchers to attain biomimetism. While the chemically-synthesized HA is similar to the mineral part of the bone, it lacks trace elements and functional groups that modify the chemical formula of the natural HA in bone. Because of the fact that it is too laborious and probably impossible to chemically synthesize HA with the right amounts of trace elements and functional groups, an easy method is to isolate the biological HA from bones of various animals, to transform it into powder, and to press it into pellets that will be further used as PLD targets. Mammalian bones contain a higher source of ions and trace elements [43–46], with Mg2<sup>+</sup> and Na<sup>+</sup> being the most frequently found ones. One notes that the presence of Na<sup>+</sup> and Mg2<sup>+</sup> ions alongside HA play an important role in the development of teeth and bone, whereas their absence could cause bone loss and fragility [2]. Other trace elements such as K+, Sr2<sup>+</sup>, Zn2<sup>+</sup>, Si4<sup>+</sup>, Ba2<sup>+</sup>, F- , and Al- have been identified. However, the overall concentration of these elements can vary and this is due to differences in the diet of the animals [47].

BioHA mainly differentiates from chemically synthesized HA by the presence of carbonate ions. The Ca10(PO4)6(OH)2 complex formula of pure HA changes into:

$$\text{Ca}\_{10\cdot x} \text{(PO}\_4\text{)}\_{6\cdot x} \text{(CO}\_3\text{)}\_{x} \text{(OH)}\_{2\cdot x}$$

where 0 ≤ x ≤ 2 [48].

Another possibility can be that in (PO4) <sup>3</sup><sup>−</sup> sites to find hydrogen phosphate (HPO4) <sup>2</sup><sup>−</sup> ions [49]. Winand et al. [50] proposed a general formula for hydrogen phosphate-containing apatites:

$$\text{Ca}\_{10\text{-x}}\text{(PO}\_4\text{)}\_{6\text{-x}}\text{(HPO}\_4\text{)}\_{\text{x}}\text{(OH)}\_{2\text{-x}}$$

where 0 ≤ x ≤ 2.

However, it is highly possible that BioHA can concomitantly have substitutions with CO3 <sup>2</sup><sup>−</sup> and HPO4 <sup>2</sup><sup>−</sup> groups. Hence, Combes et al. [37] advanced the following formula:

$$\text{Ca}\_{10\cdot x} \text{(PO}\_4\text{)}\_{6\cdot x} \text{(CO}\_3 \text{ or } \text{HPO}\_4\text{)}\_{x} \text{(OH} \text{ or } 1/2\text{CO}\_3\text{)}^{2\cdot x}$$

where 0 ≤ x ≤ 2.

It should be mentioned that various studies showed a deviation of OH− groups in the composition of bone-extracted apatites as compared to the standard HA formula. A possible explanation could be that PO4 <sup>3</sup><sup>−</sup> groups are reactive with water molecules trapped in the lattice or from the surrounding medium, which results in the formation of hydrogen phosphate:

$$\mathrm{H\_2O} + \mathrm{PO\_4}^{3-} \rightarrow \mathrm{HPO\_4}^{2-} + \mathrm{OH^-}.$$

Trace elements in the composition of BioHA materials are commonly determined qualitatively by spectroscopic techniques. It should be noted that there are chemical methods that involve dissolution of the mineral part of the bone and identification of constituents, but they are susceptible to post-decomposition chemical reactions. For metallic ions such as Na, K, Al, and Sr, it is believed that they do not influence the apatite properties, as long as they remain in the normal concentrations known for the human body. However, by accumulation following diet intake, some elements' increased content might cause biological reactions such as bone weakening or induction of osteoporosis.

A trace element with a clear effect on bones is fluorine, which forms in bone apatites fluoride ions. In addition, 99% of fluorine in most organisms is accumulated in bones in the form of fluorides (as e.g. NaF) [51]. In the human body, the fluorine content is of 0.05% to 0.1%. Its presence favorizes an increase of bone density, but, in high content, causes a disease called fluorosis, in which bones become hard and brittle. Studies suggest that fluoride also has anti-plaque properties. It was reported that amine fluoride and stannous fluoride possess bactericidal properties against oral bacteria [52].

Dopants may have an important role in the structure and properties of BioHA coatings. For the crystallites size function of dopant materials in the case of BioHA coatings synthesized by PLD, the reader is guided to consult Table 2. Duta et al. [25] observed that a percentage of 1.5 wt.% Ti in composition of PLD targets from SHA causes synthesis of layers with a reduced degree of crystallinity. Deposited SHA coatings without Ti display a crystallinity degree of ~30%, while, for the ones doped with Ti, deposited in the same experimental conditions, the crystallinity degree dropped to 7% only.


**Table 2.** Crystalline coherence length in the case of BioHA coatings synthesized by PLD, function of doping materials. Determinations were performed along the *c*-axis [(002) crystal plane] and *a*-axis [(300) crystal plane] by applying the Scherrer Equation.

BHA was doped with various elements in order to identify their role on the crystallinity of the obtained coatings. Thus, it was shown that by inserting 1 wt.% lithium into the BHA target, a film with a higher degree of crystallinity could be obtained, as compared to undoped BHA structures. A mix of 10% commercial bioinert glass and 90% BHA could produce a film consisting mainly of HA with small crystallites (48.9 nm in length vs. 83.5 nm for BHA:Li and 71 nm for BHA coatings) and a high degree of anisotropy (D002/D300 = 2.1 vs. 1.49 in the case of BHA:Li and 1.34 in the case of BHA coatings) [53].

In FTIR investigations, bands characteristic to carbonate groups were identified both in targets and films made of BHA. The lines peaking at (1419, 1457, and 1544 cm<sup>−</sup>1, respectively) were characteristic to C–O asymmetric stretching and were present in both spectra of targets and films.

In the case of bovine HA doped with MgF2 (BHA:MgF) structures [54], the dopant indirectly induced the presence of MgO traces caused during sintering of the targets by partial oxidation of MgF2. The amount of MgF2 in targets was of 2% and the amount transformed in MgO was estimated to be of 0.4 wt.%. However, in films, peaks of MgF2 were not identified, but only supplemental weak MgO peaks were observed. This can hint toward the possible structural substitution of the Mg2<sup>+</sup> cations (in the Ca2<sup>+</sup> sites) and F<sup>−</sup> anions (in the OH<sup>−</sup> group sites) into the HA lattice during the deposition process. By incorporating MgF2 in the BHA structure, the HA crystallites' length was reduced from 152 to 100 nm, while the ratio between length and width, D300/D002, dropped from 1.47 to 1.11. Mihailescu et al. [54] also inserted 5 wt.% MgO in BHA targets and observed that, in this case, crystallites length increased from 152 to 170 nm. The D300/D002 dropped from 1.47 to 1.20.

One notes that, in the case of MgF2 doping, FTIR investigations revealed the disappearance of the (OH)− bands from the synthesized coatings. Moreover, supplemental vibration bands have been recorded at ~876 to 873 cm<sup>−</sup>1, which is ascribed to the overlapping of the characteristic peaks of (HPO4) <sup>2</sup><sup>−</sup> and (CO3)2<sup>−</sup> groups.

#### **4. Preparation of Materials**

#### *Powder Preparation*

It is important to mention that all the experimental procedures used for the fabrication of animal-origin HA materials are conducted in accordance with the European Regulations [28] and ISO 22442-1 standard [29].

#### 4.0.1. Extraction of HA from Mammalian Bones

BioHA, is usually extracted from bones and teeth of various animals. To obtain the raw materials, cortical bones originating from bovines or swine are the choices generally reported in various studies tackling the extraction of natural/biological HA [2,41]. Two possible explanations for this preference might be advanced: (*i*) the immediate availability of cow and sheep bones in commercial shops and abattoirs to the difference of more exotic sources such as marine animals, and (*ii*) the extracellular matrix of bovine bones is mainly composed of HA nanocrystals and collagen fibers. The bones represent 11% of pork, 15% of beef, and 16% of sheep carcasses, respectively. Due to human consumption of these animals' meat, billions of tons of such by-products are generated every year. Moreover, millions of tons of eggshells are thrown out as waste material yearly. Part of these by-products are repurposed as ingredients for pet food, fertilizers, or in pharmaceutic and cosmetic industries. Their applicability can be increased to the biomedical sector, where bones can be used as grafts, scaffolds, or as raw materials for the production of HA for coating of implants [2,55].

BioHA powders are usually obtained from the cortical part of the femoral bones of freshly slaughtered animals (received from slaughter houses, which use the other animal parts for general consumption) [11]. Bones are generally delivered on ice to laboratories and, before use, they are submitted to a veterinary control. HA derived from animal bones is generally prepared following a three-step procedure: (*i*) mechanical cleaning of soft tissue, (*ii*) deproteinization in the alkali media, and (*iii*) calcination at temperatures sufficiently high to remove any remnant organic and biological hazardous components. Additionally, all bones are thoroughly washed and/or boiled with distilled water. After this preliminary step of washing and drying, the bones are cut off, gently checked for any remnant ligaments and soft tissues and further processed. Next, bone marrows are extracted, and all other unwanted soft tissue residues or macroscopic adhering impurities and substances are removed from shafts. Then, the shafts are cut off into slices, cleaned, and washed with distilled water. Cleaned parts are deproteinized in either sodium hypochlorite, hydrogen peroxide, diammonium hydrogen phosphate, acetone, or NaOH solutions [56–59]. After washing and drying, for the calcination process, the bone pieces are heated in a furnace (at various temperatures ranging from 350 ◦C to 1400 ◦C, for 1 to 18 h in air (Table 3), in order to completely remove the organic matter from the bones. One should

emphasize that the as-prepared powders are biologically safe due to the high temperature fabrication route, which not only favors the crystallization of the material, but also completely prevents any risk of diseases transmission, since no pathogen can survive to such extreme conditions [25,53,54,60]. The resulting calcined bone specimens are cooled to room temperature (RT) by slow furnace cooling. Then, they are generally crushed inside an agate crucible, using a mortar and pestle, prior to undergoing ball-milling to fine powders (i.e., with particles of submicron size).

Figueiredo et al. [61] performed a comparative study of the structural and chemical properties of both human and animal bone-derived HA, submitted to different calcination temperatures (600, 900, and 1200 ◦C, respectively). The obtained results indicated that the calcination temperature strongly influenced the properties of the bone samples. At higher temperatures, pure HA could be obtained, which presents higher crystallinity degrees, larger crystallite sizes, and a less porous structure. Figueiredo et al. concluded that the mammalian bone samples calcined at 600 ◦C exhibited the most promising combination of chemical composition and structure that could be exploited to provide good alternatives to synthetic apatite and/or allogeneic bone.

The viability of producing biogenic HA from bio-waste animal bones (bovine, caprine, and galline), by heat treatments in air atmosphere at different temperatures (600–1000 ◦C) was demonstrated by Ramesh et al. [62]. Among the three types of investigated animal bones, it was demonstrated that the bovine-derived HA was stable for all investigated temperatures, while those produced from both caprine and galline bones exhibited signs of decomposition, with the appearance of the β-TCP phase at temperatures beyond 700 ◦C. In addition, the bovine and caprine bones presented hardness values comparable with the ones of the human cortical bone, while, for galline bone samples, higher porosity levels and low hardness values were inferred.

Akyurt et al. [63] produced HA from sheep teeth, by calcination and sintering at different temperatures (1000–1300 ◦C). Compression strength and microhardness measurements were performed and the obtained results were the best in the case of samples sintered at 1300 ◦C. This behavior was attributed to the F-content presence in the enamel structure.

Gheisari et al. [64] prepared natural HA-Hardystonite (HT) ceramics with different percentages of HT (5, 10, and 15 wt.%, respectively). Their results showed that the mechanical properties of HA increased and the bioactivity behavior improved with the addition of HT to natural HA. The maximum value of the density was inferred for the 10 wt.% HT samples, which was attributed to the formation of HT silicate phases between the matrix particles and formation of glass bonds. Moreover, the 15 wt.% HT nanocomposite samples had the lower cold crushing strength as compared to the other ones investigated in the paper, which was due to the overlapping of glass bonds. Simulated body fluid test results indicated that the Si content increased in the samples with a higher amount of HT. Consequently, the number of Si-OH nucleation sites expands and the formation of apatite layers takes place along with an increase in the ceramics' bioactivity. In a similar study, Khandan et al. [65] reported on natural HA-diopside (Di) bio-ceramics, with different Di content (10, 20, and 30 wt.%, respectively). The obtained results indicated that an elevated content of Di in the samples determined an increase of both the density on the surface and the adhesion values. In order to maintain a surface free of cracks, Khandan et al. indicated the optimum values for voltage, time, and temperature as being 50 V, 10 min, and 850 ◦C, respectively.

Barakat et al. [30] proposed in their study three different methods to extract HA from bovine bones, which include thermal decomposition, subcritical water, and alkaline hydrothermal processes. One advantage of the first method was to produce an HA nano-rode rather than the nanoparticles obtained by using the other two proposed methods. However, the morphological investigations illustrated smaller-size particles in the case of subcritical and alkaline hydrothermal processes, as compared to thermal decomposition. Barakat et al. concluded that the proposed methods are simple and cheap, which are two important advantages that might advance their use for large scale applications.

Giraldo-Betancur et al. [66] reported on a physical-chemical comparison of synthetic and biological HA. The bio-HA samples originated from bovine bones and were prepared by using three different treatments (defat, alkaline, and calcination). It was indicated that the calcination and alkaline processes delivered crystalline HA, which presented a comparable quality to the synthetic one. In addition, it was illustrated that the calcination process was appropriate to obtain HA with the minor and major elements that appear in the natural bone tissue. After applying the alkaline process, a crystallinity degree greater than 62% was detected. Moreover, the surface of the alkaline sample presented a transition behavior between dense and porous morphology. In a similar study, Rincón-López et al. [67] performed a comparative study on the physical-chemical properties and biological behavior of sintered-bovine-derived HA and commercial HA. It was demonstrated that highly crystalline HA could be obtained from bovine bones. One should note that the Na<sup>+</sup> and Mg2<sup>+</sup> ions intrinsically presented in HA of bovine origin seemed to influence the sintering behavior evolving to ceramics with lower porosity and coarser microstructure compared to those obtained with synthetic HA. Both these studies indicated the possibility to use HA derived from bovine bones as a viable alternative to synthetic HA for biomedical applications.

In Table 3, the details for HA powder preparation using mammalian sources are summarized.

The colour of the resulting powders (Table 3) is either milky-white or green.

For more information on BioHA extraction methods from mammalian bones, the reader can consult additional information contained in references from Table 3.



#### *Coatings* **2019** , *9*, 335



#### *Coatings* **2019** , *9*, 335

#### 4.0.2. Extraction of HA from Fish Sources

Even though fish bones represent a rich source of calcium, phosphate, and carbonate [2], there are scarce reports on HA synthesis from these natural sources for biomedical applications [82]. In general, fish bones originate from fisheries, which capture fish and use it mainly for obtaining meat, oil, or some low-value fertilizers [2]. The fish bones are collected after gently removing the flesh parts from the entire fish. The bones are washed with a hot water jet or steam to remove all types of proteins and other organic impurities. The cleaned bony parts are further treated with reagents such as NaOH [83]. Similar to the case of mammalian bones, after washing and drying, the bone pieces are subjected to heating at various temperatures ranging from 100 to 1200 ◦C, for 1 to 12 h in air (see Table 4), to completely eliminate any organic matter from the bones. The resulting calcined fish bone specimens are cooled to RT by slow furnace cooling. Afterward, the specimens are either hand-crushed with a mortar and pestle or ball-milled to fine powders (Table 3). One notes that the color of the resulting powders (Table 4) is white.


**Table 4.** Preparation of HA powders derived from fish sources, presented in order of the calcination temperature.

Venkatesan et al. [83] isolated HA from fish sources by applying alkaline hydrolysis and thermal calcination methods. An increase in the particle size from 0.3 to 1.0 μm was observed when using the thermal calcination method, which is due to the particle agglomeration. HA obtained by the second method are of a nanorod shape, with 17–71 nm in length and 5–10 nm in width. Thermal calcination, in comparison to alkaline hydrolysis, produced HA with a higher crystallinity degree.

Kannan et al. [84] used hydrothermal transformation of aragonitic cuttlefish bones at 200 ◦C and calcination at temperatures up to 1200 ◦C to produce porous HA scaffolds with different levels of fluorine substitution (46% and 85%) on the OH sites. The F incorporation in the HA lattice determined a lowering of the unit cell volume because of the reduction of the *a*-axis length. The crystallites formed were close in size to bone-like apatite and were orientated along the *a*-axis rather than the *c*-axis. An type-AB carbonated apatite was also detected.

Boutinguiza et al. [85] reported on BioHA obtained from sword and tuna fish bones. The prepared powders consisted of a B-type carbonate HA with a Ca/P ratio higher than that of stoichiometric HA due to carbonate ions substituting phosphates. The presence of minor elements such as Na, K, Mg, and Sr substituting Ca was also indicated. The calcination treatment performed at 600 ◦C delivered a B-type carbonated HA. When reaching 950 ◦C, β-TCP was present in a minor amount and the carbonate content decreased. This indicated a partial decomposition of the material. It was concluded that these fish bone-derived materials originate from sustainable and cheap sources and could, therefore, represent a promising future alternative to synthetic HA for medical applications.

Huang et al. [86] compared the composition and biological properties of bio-waste HA derived from fish scale with those of synthetic HA. The experimental results indicated that the fish HA materials presented nano-sized particles with a high Ca/P ratio. Compared to synthetic HA, the sintering process increased porosity and surface roughness for fish-derived HA. The alkaline phosphate assay and von Kossa staining demonstrated that the fish-HA particles promoted osteogenic differentiation and mineralization of MG63 cells. Taking into account these results, one should recommend fish scales as a cost-effective and environmentally-friendly source for producing HA.

Piccirillo et al. [87] produced apatite-based and tricalcium phosphate-based materials from codfish bones, by annealing at temperatures between 900 ◦C and 1200 ◦C. Single phase HA, chlorapatite, and fluorapatite were obtained using CaCl2 and NaF solutions, respectively. One should note that this was the first study to report on compositional modifications of natural origin materials used to tailor the relative concentrations of elements. The obtained results indicated that, by using a simple and effective valorization technique, the conversion of this by-product into a viable compound for biomedical applications can be attained.

In Table 4, the details for HA powder preparation using fish resources are summarized.

For more information on BioHA extraction methods from fish sources, the reader can consult additional information contained in references from Table 4.

#### 4.0.3. Extraction of HA from Biogenic Sources

Every year, impressive quantities of biogenic resources such as eggshells, sea shells, and other calcite materials are disposed of as waste by restaurants, hatcheries, bakeries, or homes [2]. Due to the fact that they are very cheap and accessible, eggshells represent an important source of calcium precursor required for the synthesis of HA.

In general, the preparation procedure to extract CaO from shells consists first in washing them with boiling water or steam to completely remove all impurities. In the next step, the shells are crushed to fine powders, which is followed by heating at elevated temperatures. The resulting CaO reacts with phosphorous precursors to prepare HA by following a protocol detailed in Reference [93].

Starting from eggshells, Elizondo-Villarreal et al. [94] synthesized HA using a simple hydrothermal approach. The obtained product was a mixture of HA and CaHPO4 in a 3:1 ratio, with HA morphology in the form of whiskers. It was concluded that the obtained physical-chemical characteristics of HA should advance this type of material for dental prosthesis applications.

HA was prepared by Chaudhuri et al. [95] from eggshells and a solvent such as dipotassium phosphate. An appropriate amount of eggshells-derived CaO was immersed in K2HPO4 solution, for different soaking times, to obtain nanocrystalline HA. Both grain size and pH indicated a slight decreasing tendency with increasing soaking time. The lattice strain also showed similar behavior with increasing soaking time. The proposed procedure for the low-temperature synthesis of large-scale HA is rather simple, low cost, and eco-friendly.

Tamasan et al. [96] reported on the preparation and characterization of powders consisting of the different phases of CaPs obtained from marine-origin raw materials of sea-shells (*Sputnik sea urchins* and *Trochidae I. concavus*) in reaction with H3PO4. In the developing CaP powders, hot-plate, and ultrasound methods were used. In the Sputnik sea urchins samples, brushite was found to be predominant, while calcite was shown to exist as a small secondary phase. For the second analyzed material, monetite and HA phases were identified. A thermal treatment performed at 850 ◦C resulted in flat-plate whitlockite crystals (β-MgTCP) for both samples, regardless of the fabrication method.

Gunduz et al. prepared biphasic bioceramic nano-powders of HA and β-TCP [97] from shells of a sea snail, by using a novel mechano-chemical method. When compared to the conventional hydrothermal method, this chemical method is simple and economic, due to inexpensive and safe equipment. The characteristics of the as-produced powders, along with their biological origin, should advance these materials for further consideration and experimentation to fabricate nanoceramic biomaterials.

In Table 5, the details for HA powder preparation using biogenic resources are summarized.


**Table 5.** Preparation of HA powders derived from biogenic sources, presented in the order of the calcination temperature.

For more information on BioHA extraction methods from biogenic sources, the reader can consult additional information contained in references from Table 5.

#### **5. Pulsed Laser Deposition Method**

#### *5.1. Method Overview*

To the difference of wet synthesis methods, plasma-assisted techniques offer important advantages, such as: (*i*) a much faster process for surface deposition, (*ii*) industrial scaling, (*iii*) a stoichiometric transfer of the target composition in the synthesized structures, (*iv*) a better uniformity in terms of morphology and composition, (*v*) a lower porosity, and (*vi*) a decreased tendency of the deposited structures to crack or delaminate [100,101]. In the biomedical domain, for the fabrication of CaP coatings for bone implant applications, the most applied plasma-assisted techniques are radio-frequency magnetron sputtering and PLD [102].

PLD is a thin films synthesis technique consisting of the ablation by a focused, high power, pulsed laser beam of a solid target in a vacuum environment and the condensation of the resulting vapor phase on a deposition substrate placed parallel to the holder [19]. The most basic set-up consists of a vacuum chamber, a laser source, a focusing lens, a target holder rotated by a motor, and a substrate usually placed on a heater with the role to increase film adherence and promote crystallinity.

The wavelengths of choice for the laser sources used for ablation are in the UV range due to the higher penetration depth of this type of beam in the target material as compared to visible or IR laser sources and higher photon energy that allows for a more efficient vaporization of the target [103]. Popular laser sources used in PLD experiments are excimer lasers such as ArF [104], KrF [105], or XeCl [106] emitting at 193, 248, or 308 nm, respectively, or a solid-state laser such as a Nd:YAG [107], which emits at 266 nm.

To increase the amount of evaporated target material to the detriment of expulsed liquid or solid phases, lasers emitting in a pulsed regime, with pulse durations in the nanoseconds-picoseconds range, are generally used [108]. In these regimes, the absorption process occurs on a much shorter time-scale as compared to the thermal diffusion process. The surface layers of the target reach temperatures of tens of thousands of K and are instantly vaporized. A plasma plume consisting of atoms, ions, electrons, molecules, free radicals, and condensed particles expands at supersonic speed and is accelerated toward the substrate (on a direction perpendicular on the target), where the deposition takes place [109].

If a solid collector is placed in front of the target at a distance approximately equal to the plasma plume's length, the material condenses and, pulse-by-pulse, a thin film is grown [110]. If the placement of the substrate is closer to the target than the plasma plume's length, the material will be deposited and, subsequently, washed by the plasma plume, with the deposition being inefficient and uncontrollable. Similarly, if the distance is larger than the plume's length, the deposition will also be inefficient, since the film thickness is inversely proportional to the square of the target–to–substrate distance. Femtosecond laser sources are not considered optimal for thin film synthesis. Due to the high energy of the ablated species, the growing film is heavily bombarded by high speed particles, which makes the process inefficient and results in films with defects, vacancies, or made of small solid particles ripped from the target surface [111].

Due to the high temperatures generated on the target surface, any type of material can be ablated and, therefore, a wide variety of targets can be used for thin film deposition.

The technique is versatile in terms of process output. By independent control of the deposition temperature, background pressure, and substrate positioning, one can change the film's crystallinity, composition, adherence, surface roughness, and/or thickness [18]. The use of multiple targets irradiated subsequently allows for the deposition of multi-layered structures [112].

Moreover, the PLD technique can be easily adapted to increase the range of applications: by limiting the number of pulses, one can deposit nanoparticles instead of thin films [113]. Two targets of different materials can be irradiated simultaneously and the intermixing plasma plumes can create a compositional library on a substrate [114]. By irradiating a frozen target made from an organic compound dissolved into a matrix material, an organic thin film can be laser deposited without degradation of the target material in a variation of the method, denominated MAPLE, i.e. Matrix Assisted Pulsed Laser Evaporation [115].

The film growth by PLD also offers different types of advantages over other deposition techniques, such as:


Like all the other synthesis techniques, PLD presents some disadvantages too, such as a low deposition rate, an applicability domain limited to compounds that are not sensible to thermal decomposition and degradation that appears during processing by prolonged exposure to UV laser radiation, the presence of droplets (of a different composition and dimensions), and a low deposition area (no bigger than a few cm2). The last two drawbacks can be diminished or even eliminated by additional set-ups [18,116].

#### *5.2. BioHA Targets Preparation*

The powders obtained following the protocols mentioned in Section 4 are generally pressed and the resulting pellets are heat-treated in a furnace, to reach compactness by eliminating air bubbles and water vapors. The as-sintered pellets are compact and tough (Figure 1) and can be used as targets in the PLD experiments [11,25,53,54,60,80].

**Figure 1.** Photos of sintered PLD targets: (**a**) before and (**b**) after the action of the beam generated by a KrF\* excimer laser source.

#### The Importance of Thermal Treatments in the Case of Targets

Even though CaP bio-ceramics possess excellent biomedical characteristics, their poor mechanical behavior determined focused research to find viable solutions to this important drawback.

The properties, efficiency, phase purity, and size distribution of HA extracted from natural resources, especially from bones, depend upon factors like the extraction technique, the calcination temperature, and the nature of bones [2]. HA derived from natural resources (such as animal bones) is available in unlimited supply and can be produced by calcination methods [27,117]. Ozyegin et al. [117] reported that sintered BHA is safe from potentially transmitted diseases such as bovine spongiform encephalopathy ("mad cow disease"). The manufacturing of these powders is easy to carry out and economically viable [2,117]. Prions, which are the smallest proteinaceous infectious particles, can resist inactivation by procedures that modify nucleic acids and are able to transmit alone an infectious disease, which causes harm to the host tissue [118]. Thus, the temperature generally applied for calcination is around 850 ◦C and it was demonstrated that no disease transmission pathogens (including prions) can survive to such high temperatures [2,27,117]. However, a special calcination regime has to be carefully selected in order to remove all organic matter and enhance the crystallinity of HA, while avoiding thermal decomposition of the final product, HA [2].

#### *5.3. Substrates Used as Pulsed Laser Deposition Collectors*

In general, for the fabrication of bio-implants and bio–devices, there are five classes of materials used: (i) metallic materials, (ii) polymers, (iii) ceramics, (iv) composites, and (v) natural materials.

The most commonly used metals and alloys for medical device applications include stainless steels, cobalt-based alloys, and commercially-pure titanium and its alloys [119]. The last ones are generally preferred for bone-replacement applications mainly due to their intrinsic resistance. These materials have been widely used as endosseous implant materials in dentistry and orthopedics due to their high strength, good corrosion resistance, no allergic problems, low density, and excellent biocompatibility [120]. Moreover, the thin oxide layer, which is naturally formed on their surface, is acting as a protective barrier, which confers its well-known resistance among metals to corrosion in physiological conditions. However, it is far from being an ideal material from a medical point of view because of its insufficient bioactivity when implanted in the human body [121].

Polymers represent the largest class of biomaterials, which have gained greater attention due to some important advantages such as (i) ease of fabrication to various complex shapes and structures, (ii) wide range of bulk compositions and physical properties, and (iii) easy tailoring of surface properties [119].

Ceramics are inorganic compounds of metallic or non-metallic materials. When used for skeletal or hard tissue repairing, they are referred to as bio-ceramics. These bio-ceramics can be classified as bioinert (e.g. alumina, zirconia), bioresorbable (e.g. TCP), bioactive (e.g. HA, bioactive glasses, glass ceramics), and porous (e.g. HA coatings on metallic substrates for tissue in-growth).

A composite consists of two or more materials, characterized by a combination of the best physical and chemical characteristics of each component materials. In the biomedical field, composite materials are generally designed for superior mechanical and biological properties. They can be classified as a function of the (i) matrix material and (ii) bioactivity of the composite. In the first case, there are three types of composite materials: (i) polymer matrix composites, (ii) metal matrix composites, and (iii) ceramic matrix composites. For the second case, there are: (i) bioinert composites, (ii) bioactive composites, and (iii) bioresorbable composites.

Collagen and glycosaminoglycans are the most commonly used natural materials for clinical applications [122] due to a series of important advantages, such as: (i) easy recognition by the biological environment due to their similarity with the macromolecular substances, (ii) possibility to by-pass toxicity-related issues, chronic inflammation, or lack of recognition by cells (which generally occurs when dealing with synthetic materials), and (iii) biodegradable nature of the materials, which permits their use for focused applications, where a specific function is needed for a limited period of time [123].

In general, prior to the coating process and immediately after it, special attention needs to be paid to the surface preparation of various types of used substrates. In this respect, different substrate preparation techniques have been reported, from simple to complex ones, with their own peculiarities. Thus, simple substrate preparation procedures include cleaning or degreasing to remove any remnant surface contamination because of manufacturing conditions or improper storage. Solvents like acetone, ethylic alcohol, and/or distilled water might be used in this case, always keeping in mind the nature and properties of the used substrates. By contrast, more complex substrate preparation techniques include (i) mechanical modification of the surface, by sand-blasting [124], grit-blasting [125,126], polishing [127,128], or grinding [129], and (ii) chemical treatments, modifications, or functionalization, using chemical activation [130], alkaline treatments [131], anodization [124,125], acid-etching [124–126], oxidizing [132], and many other actions [124,133]. Additional physical treatments, resulting in various types of surface modifications, were briefly discussed in Reference [133]. After CaP synthesis, the applied post-deposition treatments are intended to (i) provide crystallization/recrystallization of various phases, (ii) improve their fixation to the substrate, and (iii) evaporate traces of solvent(s) that might remain trapped within the deposited structures.

#### *5.4. Pulsed Laser Deposition Experimental Set-Up*

A PLD experimental set-up is typically composed of a UV laser source, a vacuum stainless-steel deposition chamber (Figure 2), equipped with a rotating target, a fixed substrate holder, and pumping systems. Following target irradiation, a plasma cloud expands, either in vacuum or in different gas atmospheres, and the vapors are collected on substrates (generally placed parallel to the targets and heated up to a temperature in the range of 350–600 ◦C), in the form of thin, adherent films [18]. In this respect, amorphous or crystalline, extremely adherent, stoichiometric, and dense or porous structures from various complex materials can be synthesized, even at relatively low deposition temperatures, by varying the experimental parameters that are mainly related to the (i) laser (fluence, wavelength, pulse-duration, and repetition rate), and (ii) deposition conditions (target-to-substrate distance, substrate temperature, nature, and pressure of the environment) [19,134,135].

**Figure 2.** Schematic of an experimental set-up used for PLD synthesis of animal-origin HA coatings.

Due to the possibility to independently vary a large number of parameters (such as the wavelength, fluence, pulse repetition frequency, energy, target preparation, target–to–substrate distance, substrate temperature, area of the laser spot, deposition geometry, nature, and pressure of the gas in the deposition chamber), PLD represents a versatile technique to produce films with a high diversity of morphological and structural characteristics, which are superior to the ones appertaining to conventional deposition techniques (fast processing, safety in functioning, and low production costs) [18,19,116,136].

The substrate temperature during film synthesis by PLD is usually situated in the range of 350 to 600 ◦C. This way, one can assure the fabrication of highly crystalline and phase-pure films [137]. Moreover, lower or higher values for substrate temperature can be chosen to obtain layers with fine textures or different roughness, with improved adherence to the substrate, depending on the intended applications.

Various experimental conditions used for PLD synthesis of BioHA coatings have been reported in the literature. They are summarized in Table 6. The results from these papers will be thoroughly analyzed and discussed in Section 6.


**Table 6.** Different experimental conditions used for PLD synthesis of BioHA coatings.

#### *5.5. Thermal Treatments Applied to Pulsed Laser Deposited Coatings*

BioHA coatings are generally amorphous when deposited by PLD. HA loses (OH)− molecules and on the deposition substrate a mix of CaPs is formed. The most prominent of them are tricalcium phosphate (TCP) and octacalcium phosphate. The process is reversible and a post-deposition thermal treatment around 500 ◦C in water vapors leads to conversion of the deposited amorphous and non-apatite phases into more stable compounds, like HA. In addition, their crystallinity and corrosion resistance increase, and the residual stress is considerably reduced [138–141]. It is also important to note that, the presence of water during the post-deposition heat treatment also plays an important role in this conversion [142,143].

Dinda et al. [144] have reported that a post-deposition thermal treatment at 300 ◦C can produce pure, adherent, and crystalline HA coatings, which will not dissolve in simulated body fluids.

Post-deposition thermal treatments, generally performed in air or enriched water-vapor atmospheres, are of key importance. Their role is, on one hand, to restore the stoichiometry of the synthesized compounds and, on the other hand, to improve the overall crystalline status of the coatings [53].

The sintering temperature and atmosphere are considered as important parameters able to tailor the strength and toughness of HA [145]. For example, sintering at elevated temperatures has the tendency to eliminate the functional group OH in the HA matrix and this would result in the decomposition of the HA phase to form α-TCP, β-TCP, and tetra-calcium phosphate [145].

#### **6. Characterization of Pulsed Laser Deposited Coatings of Undoped and Doped Animal-Origin Hydroxyapatite**

The purpose of CaP coatings that cover metallic medical prostheses is to ensure biomimetism, i.e. to "mislead" the human body that the newly implanted device is very similar to its bones and, thus, start osteoblasts proliferation and osseous matrix synthesis. There is no perfect coating technique and the existence of shortcomings for each of them determined the implementation of others, among which PLD is worth mentioning due to its advantages such as stoichiometric transfer and precise thickness control.

For characterization of the PLD synthesized coatings, the most common investigation techniques are: Scanning Electron Microscopy (SEM) for surface morphology and cross-sectional investigations, Energy-dispersive X-ray spectroscopy (EDS) for Ca/P ratio determinations, profilometry or Atomic Force Microscopy (AFM) for surface parameters valuation, X-ray diffraction (XRD) for crystalline status, stress, crystallites size, and orientation investigations, Fourier Transform InfraRed (FTIR) spectroscopy for identification of chemical bonds, pull-out bonding strength or scratch tests for assessment of films' adherence to the substrate, and in vitro and in vivo tests. Therefore, we will concentrate in the following sections on these techniques for comparing the morphological, structural, and functional properties of PLD coatings synthesized from targets made of CaPs obtained from animal sources.

#### *6.1. Morphological and Compositional Analyses*

PLD coatings have general surfaces covered by round particles, which are further denominated as "droplets." Their presence on the surface can be explained like this: following target surface laser irradiation, the surface top layers are instantly vaporized. Part of the pulse energy is transformed into heat, which is concentrated in a subsurface layer. Therefore, due to heat accumulation, a mix of liquid and gaseous phases is generated. A bubble is formed, which expands until the top solid layer breaks and a plasma plume is released. The shockwave of the plume splashes the liquid phase on the target surface and liquid droplets are released with supersonic speeds reaching the substrate. Their number and diameter depend on the type of deposited material as well as pressure in the deposition chamber and on the laser parameters [18]. Droplets can be trapped using ingenious particle captors or off-axis geometries in order to obtain particle-free films. However, in the case of biological applications, a rough morphology could be beneficial for osteoblast proliferation and droplet-containing films are considered acceptable.

Sheep derived HA (SHA) coatings synthesized by PLD using a KrF\* excimer laser source [25] also had an irregular morphology composed of spherical particles whose distribution could be approximated by a Gaussian fitting centered around a mean diameter of 1 μm. The particles size varied between 0.35 to 3.5 μm. An interesting observation was made for SHA doped with Ti (SHA:Ti, 1.5 wt.%) coatings. The particle number on the surface almost doubled as compared to undoped material, which produced rougher surfaces. When depositing coatings with the same number of laser pulses from SHA and SHA:Ti targets, Duta et al. [25] obtained thicker SHA coatings as compared to SHA:Ti ones (1.8 vs. 1.2 μm, respectively).

In Reference [53], Duta et al. conducted a comparison between film morphology in the case of SHA and BHA doped with Li2O or a commercial inert glass. In all cases, the coatings displayed the morphology with round micronic droplets covering the surface. There were no big differences between particles diameter. However, a variation in the number of particles/cm<sup>2</sup> was statistically significant. Table 7 summarizes these differences.

**Table 7.** Droplets density in the case of synthetic HA and different BioHA coatings synthesized by PLD, using the same experimental conditions.


The coatings' thickness decreased in the following order: HAsyn, BHA:Li, SHA, BHA:CIG (3.71 ± 0.15 μm → 3.52 ± 0.16 μm → 2.95 ± 0.21 μm → 2.79 ± 0.02 μm). The roughness of the films decreased in the same order with values of 0.76 ± 0.1 for HAsyn, 0.71 ± 0.06 for BHA: Li, 0.61 ± 0.05 for SHA and 0.77 ± 0.03 μm for BHA:CIG, respectively [53].

To the difference, Mihailescu et al. [54] obtained smooth surfaces of BHA coatings on which a low number of spherical droplets appeared as sputtered. Their diameter ranged from 1.6 to 2.5 μm. The beginning of an organization of some droplets into clusters could be assumed from the SEM images. For BHA doped with MgF2 and MgO, deposited in the same experimental conditions, a clear increase of the droplets' density was observed. They tend to accumulate in a majority in clusters of particles, with only a few isolated particles being identified. BHA doped with MgO (BHA:MgO) displayed the largest area of film surface covered by particle clusters. Apparently, the mean size of the droplets was very similar between undoped and doped materials.

In the case of BHA doped with Li2CO3 and Li3PO4 structures (further denoted as BHA:LiC and BHA:LiP) [60], the films' surface was flat, with isolated round droplets, sputtered on the surface. Most droplets were more than half embedded in the film mass, which is an indicative for their high velocity at the moment of impact with the surface. The droplets' number quantification was performed on four randomly chosen SEM fields and the obtained values were of (2.2 <sup>±</sup> 0.2) <sup>×</sup> 104 particulates/mm<sup>2</sup> (for BHA), (2.8 <sup>±</sup> 0.2) <sup>×</sup> 104 particulates/mm2 (for BHA:LiC) and (1.5 <sup>±</sup> 0.1) <sup>×</sup> 10<sup>4</sup> particulates/mm2 (for BHA:LiP) [60].

From the cross-sectional SEM images, thicknesses of ~8.5, ~7.2, and ~6 μm, respectively, were inferred for BHA, BHA:LiC, and BHA:LiP coatings [60].

Comparative SEM micrographies showing different surface morphologies, which correspond to titanium and various types of BioHA coatings synthesized by PLD, using the same deposition conditions, are presented in Figure 3.

**Figure 3.** SEM micrographies corresponding to the surfaces of titanium and different types of BioHA coatings synthesized by PLD. Magnification bar: 50 μm.

Thickness of the PLD synthesized coatings can play an important role on their morphology, especially in the case of rough substrates. In Reference [89], films deposited in a water-vapor atmosphere onto Ti substrates were significantly thinner as compared to the ones deposited in vacuum. The calculated deposition rates were 3.5 nm/min for depositions conducted in water vapors, and 18.1 nm/min for depositions performed in vacuum. Therefore, films deposited in water vapors were thin and mimicked the Ti surface morphology. For deposition times under 30 min, surface features were not evident. For deposition times between 30 and 120 min, surface features became prominent. Aggregates of globular shape became visible at magnifications of 5000×, while at 20,000×, the droplets appeared to be made of elongated acicular crystalline structures of < 1 μm in length.

In general, the EDS quantitative measurements are performed during SEM investigations to determine the elemental distribution in the synthesized films and to estimate the Ca/P ratio.

A cross-sectional SEM image and its corresponding EDS spectra in the case of a BHA:CIG coating synthesized by PLD is presented in Figure 4. A compact structure of the coating can be observed. Moreover, the presence of trace elements (Na, Mg, Si, etc.) is also emphasized. A quasi-stoichiometric target-to-substrate transfer is reported in which the inferred Ca/P molar ratio is ~1.70. This value is typical to natural bone due to multiple doping with trace elements and functional groups.

**Figure 4.** Cross-sectional SEM image and the corresponding EDS spectrum of a bovine hydroxyapatite doped with commercial inert glass coating synthesized by PLD.

#### *6.2. Structural Investigations*

GIXRD patterns of post-deposition treated BioHA coatings synthesized by PLD should display the characteristic peaks of HA [hexagonal, P63/m (176) space group, ICDD: 00-009-0432] [25]. It is quite possible that the GIXRD patterns contain, besides the HA peaks, some pronounced humps (centered at 2θ ≈ 30◦), which are proof of an amorphous phase in the deposited films. Duta et al. [25] reported for SHA films deposited by PLD a crystallinity degree of ∼35%. The starting powders for PLD targets displayed in the XRD patterns some peaks that could be associated to MgCO3 (ICDD: 01-086-2345) and CaMg2 (ICDD: 00-013-0450) phases. However, after the deposition, these crystalline peaks were not present in the GIXRD patterns [25].

Some typical XRD patterns of HAsyn and BioHA coatings undoped or doped with Li2O (further denoted as BHA:Li) synthesized by PLD are presented comparatively in Figure 5. One can observe that the most intensive peaks of the patterns in Figure 5 appertain to the Ti peaks of the substrate (ICDD: 44-1294). The fabricated coatings consisted of an HA phase with different degrees of crystallinity, as observed from the diffracted intensity variation and the mean crystallite sizes estimated from the FWHM of (002) and (300) XRD lines [53]. It is visible in Figure 5 that a more pronounced crystallinity corresponds to BHA:Li films. In addition, several weak lines were also identified whose origin might be due to a small percent of oxygen-deficient titanium oxides. Their formation might be due to the substrate oxidation during the post-deposition treatments [53].

**Figure 5.** Comparative XRD patterns of synthetic and BioHA coatings synthesized by PLD onto titanium substrates (-, HA; Δ, TiO, and Ti2O sub-oxides). Reprinted from Reference [53], with permission from Elsevier.

Another way to identify BioHA is by means of characteristic functional groups that can be detected by FTIR. Besides the characteristic HA bands of ν3 asymmetric stretching mode of (PO4) <sup>3</sup><sup>−</sup> groups (1010–982, 1063–1023, and 1091–1090 cm<sup>−</sup>1, respectively), the ν1 symmetric stretching mode of (PO4) 3− groups (961–945 cm<sup>−</sup>1), ν4 bending mode of (PO4) <sup>3</sup><sup>−</sup> groups (558–553 and 600–594 cm<sup>−</sup>1, respectively), and the peak at <sup>∼</sup>630 cm−<sup>1</sup> characteristic to (OH)<sup>−</sup> groups, in the case of BioHA, which is a band centered at 874–870 cm<sup>−</sup>1, could be present in the spectra, assigned to (HPO4) <sup>2</sup><sup>−</sup> groups.

In the case of BHA targets and films, Duta et al. [53] identified small traces of hydrogen phosphorus fluoride hydrate (ICDD:77-0136) and brushite in the PLD targets [CaPO3(OH)·2H2O, ICDD:11-0293]. These traces were not present in the films obtained from irradiating these targets. Both targets and films displayed low amounts of fluorine besides the main HA phase. Popescu et al. [60] identified in BHA targets and films only FTIR bands corresponding to carbonate groups besides the characteristic orthophosphate bands of HA.

In the case of HA obtained from shark teeth, Hidalgo-Robatto et al. [89] reported on depositions conducted in water vapors. Depending on the pressure, different degrees of crystallinity were obtained. Under 0.25 mbar, spectra revealed a singular broad peak centered at ~31◦, while, for a pressure of 0.45 mbar, the characteristic peaks for apatites located at 29.4, 32.1, and 33.4◦, respectively, were present. In the FTIR spectra, besides the characteristic groups of CaPs, a peak associated with carbonate groups was present at 870 cm<sup>−</sup>1. Depending on the deposition conditions, the importance of this peak varied. As the vapor pressure increased, the intensity of carbonate bands also increased. The CO3 <sup>2</sup>−/PO4 3− ratio was 30% higher in the BioHA coating deposited at 0.45 mbar H2O vapors as compared to the case of depositions conducted without water vapors. F1s transition peaks were also shown by XPS in the PLD targets and in deposited films. Based on FTIR, XRD, and XPS results, one could conclude that films deposited by PLD from shark-derived HA contained a hydrated and fluorinated carbonated HA.

#### *6.3. Bonding Strength Tests*

The bonding strength at the bio-functional coating–substrate interface is considered a critical parameter for the fabrication, successful implantation, and long-term stability of implant-type structures, since it governs both the initial stability and long-term functioning of the medical devices [16,146].

The bounding strength values inferred for Ti, BHA, and doped BHA coatings synthesized by PLD are presented in Figure 6. The raw data of the obtained results reported in References [25,53,54,60,80] were plotted together for comparison reasons only.

**Figure 6.** Bonding strength values in the case of titanium and different BioHA coatings synthesized by PLD using the same experimental conditions.

First, in order to verify the quality of the bonding adhesive, control tests are carried out on bare Ti substrates. The adherence values determined at the stainless-steel dolly/Ti substrate interface were of ∼60 MPa. We note that this value is in accordance with the specifications provided by the manufacturer [147].

Duta et al. [53] recorded for the case of HAsyn coatings, adherence values generally similar to the ones reported in literature for HA films synthesized by PLD [148,149]. Significantly higher values were measured in the case of SHA and BHA:Li coatings. The decreased value of adherence registered for the BHA:CIG structure was attributed to the intrinsic friability of the glassy doping phase under mechanical stress [53].

Duta et al. [25] performed comparative pull-out adherence tests between SHA and SHA:Ti structures. The values recorded for SHA coatings were of ~55 MPa, while, for the doped ones, the values were of ~64 MPa. The increase in bonding strength of SHA:Ti coatings was due to the presence of the Ti-Ca segregation at the interface with the substrate. One notes that, during post-deposition annealing treatments, atomic inter-diffusion processes are known to take place at the coating–Ti substrate interface [150–152]. In general, this leads to the strengthening of the films' adherence by smoothing the interface between the two different materials. Therefore, the existence at the interfacial region of a segregated Ti-Ca inter-layer could accommodate an easier and more homogeneous inter-diffusion during the progress of the heat-treatments [25].

Mihailescu et al. [54] investigated the bonding strength for BHA and doped BHA structures. Compared to BHA coatings, the addition of MgF2 and MgO seemed to result in an improvement of the films' adherence to the Ti substrate with mean bonding strength values of ~49 and ~56 MPa, respectively, being measured [54].

Popescu et al. [60] reported on similar coating adherence values of (42–46) MPa, in the case of BHA and BHA:LiC coatings. These values are in agreement with the ones inferred by Duta et al. [80]. A decrease of the bonding strength down to ~33 MPa was recorded in the case of BHA:LiP structures, which was attributed to the existence of a larger content of the amorphous counterpart [60].

It is important to emphasize that the bonding strength response measured for all coatings synthesized by PLD (Figure 6) was superior to the minimum mandatory value of 15 MPa imposed by international standard 13779-2 [153] in the case of implant-type coatings used for load-bearing applications.

#### *6.4. In vitro Biological Observations*

#### 6.4.1. Bioactivity Effect

The most common method to qualitatively assess BioHA coatings' biocompatibility and cells' proliferation is the MTS assay, while cells' toxicity can be determined by using a lactate dehydrogenase (LDH) activity test. The MTS assay is based on the use of a tetrazolium compound (3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium, inner salt – MTS) that is chemically reduced by viable cells into formazan, which is soluble in tissue culture medium. Since the production of formazan is proportional to the number of living cells, the intensity of the produced color can be used as an indicator of cell proliferation.

LDH (or LD) is an enzyme involved in energy production that is found in almost all of the body's cells. Elevated levels of LD usually indicate some type of cell damage. LD levels will typically peak as the cellular destruction begins. Thus, if a substrate is cytotoxic, the detected LD levels will be high. In Reference [25], tests were conducted according to the ISO standard 10993-5 [154].

In literature, the biocompatibility of synthesized coatings was assessed by the MTS tests, by applying different protocols. In this respect, human mesenchymal stem cells (hMSC) were seeded on Ti and various BioHA surfaces and the obtained results are plotted in Figure 7a,b and reported in References [25,60].

**Figure 7.** Proliferation of human mesenchymal stem cells assessed by MTS assay in the case of (**a**) bovine hydroxyapatite (BHA) and doped BHA coatings and (**b**) Ti, sheep hydroxyapatite (SHA), and doped SHA coatings synthesized by PLD. Reprinted from References [25,60], with permission from Elsevier. Slight modifications of the acronyms from the original figures were done to be in agreement with the sample codes used in this review.

In Figure 7a, one can observe that the cells' proliferation was better in the case of doped BHA coatings with respect to undoped ones. The proliferation increase was demonstrated to be in a direct relation with the augmentation of the polar component of the surface-free energy. Therefore, the beneficial effect of Li2CO3-doping could be emphasized [60].

After 24 and 48 h, the proliferation capacity of cells grown on the surface of SHA and SHA:Ti coatings was improved as compared to Ti surfaces for the same time intervals (Figure 7b). In addition, cell death at 24 h of growth was tested by using the LDH method [25].

Lithium is a trace metal in organisms. In pharmacy, it is used in composition of various drugs for the treatment of bipolar affective disorder and depression [155]. Lithium was also administered in drugs for treating osteoporosis in the last 20 years [156,157] due to its proven ability to enhance osteogenesis [158]. Despite its proven efficacy [159], there are reports when lithium administration determined side effects such as thirst and excessive urination, nausea, diarrhea, tremor, weight gain, sexual dysfunction, dermatological effects, and long-term effects on the thyroid gland, kidneys, and parathyroid glands [160]. However, recent studies challenge this negative perception of lithium administration, stating that intoxication occurs only when the concentration is higher than 1.5 mmol/L, and its main adverse effects can be properly monitored and managed [11,161–163].

Hidalgo-Robatto et al. [89] tested BioHA coatings (obtained from shark teeth) synthesized by PLD using a pre-osteoblastic cell line MC3T3-E1. The BioHA was compared to synthetic HA and a control of tissue culture polystyrene. Cell morphology was similar in the case of synthetic HA and BioHA. Both coatings were able to promote healthy morphology on the cells. MC3T3-E1 cells made contact with neighboring cells and covered the entire surface of their respective coatings. The MTT assay showed after seven days from seeding, an intense proliferation of cells on the biological apatite coatings, which is significantly higher than the proliferation on the synthetic HA coatings or the control by the MTS assay. After 21 days, all tested materials had higher values of cell proliferation than on day 7 and displayed similar absorbance values. Alkaline phosphatase (ALP) activity was tested as a measure of osteoblastic activity and bone formation. Statistically significant higher ALP activity was quantified on the BioHA coatings compared to both the synthetic HA and TCP, which presented lower values in the same range. After 21 days of incubation, both types of apatite coatings apparently manifested higher levels of ALP activity as compared to TCP values. Overall, BioHA seemed to accelerate osteogenic activity as compared to synthetic HA due to its fluoride content and its trace elements such as Na and Mg.

Doping of BioHA with various elements can drastically increase the cellular response toward PLD synthesized coatings. In such a study, the surfaces of medical-grade Ti, SHA, and SHA:Ti coatings (the precursor material for PLD targets was a mix of SHA powder and 1.5 wt.% Ti powder) were seeded with human mesenchymal stem cells (hMSC) [25]. On all types of samples, cells displayed the same typical morphology: small dimensions, well-spread, and having a polygonal shape. They have established liaisons with neighbouring cells by thin cellular extensions. The lowest proliferation value was reached for the SHA coatings, with respect to both SHA:Ti ones and controls (bare Ti and polycarbonate) [25]. The experiments also showed an important cell number growth (of ∼60%) in the case of SHA:Ti coated samples. The proliferation increase of SHA:Ti *vs*. SHA was found statistically significant (p < 0.05). The LDH results showed that both SHA and SHA:Ti coatings fostered a good cells' cytocompatibility. However, the amount of cells that underwent apoptosis and death was lower in the case of the polycarbonate control and SHA:Ti coatings. SHA structures presented almost double cell apoptosis death as compared to SHA:Ti ones. The dead cells' index (percent of dead cells from the total number of cells) was lower than 4.5% for all cases, which constitutes another proof of the good cytocompatibility of the PLD coatings. The beneficial effect of Ti doping in HA has not been yet completely elucidated. Duta et al. [25] advanced its lower degree of crystallinity as a possible explanation for the superior biocompatibility of SHA:Ti, which generated the premises for a better interaction with the organic component of the mesenchymal cell culture media. Ca2<sup>+</sup> ions released by dissolution may act as binders for organic moieties from culture media with roles in the healthy development of cells. A similar interpretation of enhanced protein binding by HA:Ti has been advanced by Kandori et al. [164] and Wei et al. [165], respectively.

In Reference [60], the influence of lithium doping of BHA on the biological response of the same type of cells was tested. The cells were stained with calcein/EthD-1 in order to evaluate the viable ones and the morphology of hMSC was observed three days after seeding on the surface of undoped and doped coatings. BHA:LiC and BHA:LiP samples produced a decrease in cell spreading without significantly modifying cell viability. Cells were adherent on the tested materials and exhibited typical fibroblast-like characteristics, with dendrites and prominent nuclei. There was no visible difference between lithium-doped BHA coatings and control materials (bare Ti and undoped BHA samples), as shown in Figure 8.

**Figure 8.** Cellular viability of human mesenchymal stem cells after 72 h of growth on the surface of Ti, bovine hydroxyapatite (BHA), and doped BHA coatings synthesized by PLD using the same experimental conditions. Reprinted from Reference [60], with permission from Elsevier. Slight modifications of the acronyms from the original figure were done for uniformization with the sample codes used in this review.

The in vitro cellular viability of hMSC grown for 72 h on Ti, BHA, and doped BHA coatings was assessed using the calcein AM/EthD-1 [60] method and the results are presented in Figure 8.

Based on relative intensity of calcein and EthD-1 signals on resulting scatter-grams, positive from negative cell populations could be distinguished. Compared to BHA samples, lithium-doped and MgF2-doped surfaces led to a slight improved growth of hMSC cells, whilst, for BHA:CIG and BHA:MgO surfaces, a reduced cellular viability was inferred (Figure 8).

The high biocompatibility of BHA:LiC coatings was assumed to be favored by the highest wettability and overall surface energy, as well as by an augmented contribution of the polar component of the surface energy. Other studies report in favor of this hypothesis that, in order to enhance the cellular response (including here adhesion, spreading, and cytoskeletal organization), an optimal window for the contact angle in the range of 60◦ to 80◦ between surface and water molecules should exist [166–168]. In this case, the cell proliferation results are in good agreement with this rationale, with the best results being obtained for coatings to elicit a water contact angle of ~72◦ and the highest polar component.

Dos Santos et al. [169] and Redey et al. [170] reported that the higher polar component of carbonated HA and natural calcium carbonated surfaces induced an increased osteoblast and osteoclast adhesion, respectively. Furthermore, Zhao et al. [171] indicated that a higher surface free energy is connected with the increased production of bone ALP and osteocalcin, in the case of MG63 osteoblastic cells grown on hydroxylated Ti surfaces.

#### 6.4.2. Antibacterial Effect

Mihailescu et al. [54] tested BHA and BHA:MgF or BHA:MgO against gram-positive *Micrococcus sp*. and gram-negative *Enterobacter sp.* for 72 h. The BHA coatings and Ti substrate used as controls had no antibacterial effect. When BHA was doped with MgF2, and pulsed laser deposited on a substrate, it exhibited a slight anti-biofilm activity (1 log reduction of viable cell counts as compared to bare Ti) toward *Enterobacter sp*. The highest anti-biofilm activity was observed for the BHA:MgO coatings, which inhibited the *Enterobacter sp*. biofilm in all development stages, *Micrococcus sp*. biofilm at 24 and 72 h, and *C. albicans* biofilm at 48 h.

In Reference [172], it was shown that MgO exhibits stronger bactericidal activity as compared to TiO2. Two hypotheses explaining the decrease of the number of biofilm-embedded cells could be considered, i.e. (*i*) the surface coatings prevent the microbial/bacteria adhesion and (*ii*) the coatings exhibit a microbicidal activity, which kills the cells before or after contact with the coated surfaces. The second hypothesis is supported by the results of the minimal inhibitory concentration assay showing an increased microbicidal activity of BHA:MgO and BHA:MgF coatings as compared to BHA against the tested strains. No cytotoxic activity of deposited coatings against human cells was revealed during tests on Hep2 cells. It has been reported in literature that pure Mg is highly effective against *Staphylococcus aureus (S. aureus)* [173]. Numerous studies tested the antibacterial efficiency of MgO against bacteria. Sawai et al. [174] demonstrated the efficiency of MgO powders against *S. aureus* and *Escherichia coli*, while Gayathri et al. [175] created a doped HA containing Mg in the form of nanoparticles that had antibacterial efficiency, but also insecticide action, which killed the larvae of three mosquito species. Photoluminescence studies conducted by Sawai et al. [174] showed that MgO slurries generate active oxygen species such as O2- and this seems to be the most probable cause for bacterial apoptosis. These oxygen species form in wet media hydroxyl radicals and hydrogen peroxide. Hydroxyl radical is the most reactive oxygen radical, able to interact with almost every type of molecule of the living cells of bacteria and fungi, which causes irreversible damage to cellular components and, eventually, apoptosis [176].

Mg is not toxic for the human organism in a dose of 200 to 400 mg per day. The Food and Nutrition Board of the Institute of Medicine in the United States has set an upper tolerable limit of 350 mg per day for supplemental magnesium. In larger doses, it can induce nausea and vomiting, muscle pain and weakness, confusion, and cardiac arrhythmias [177].

BHA and BHA doped with lithium compounds such as carbonate and phosphate (i.e., BHA:LiC or BHA:LiP coatings) proved antibacterial and antifungal activity against *S. aureus* and *C. albicans* strains [11], which was an unexpected discovery. After 24 h, all tested coatings manifested inhibitory activity for *S. aureus* colonies. After 48 and 72 h from inoculation, the antibacterial effect was much more visible. Therefore, a drastic decrease in the number of CFU/mL of 2 to 4 logs at these two-time intervals was recorded (Figure 9).

**Figure 9.** Graphic representation of the logarithmic values of *S. aureus* biofilm cells developed on the surfaces of Ti, bovine hydroxyapatite (BHA), and doped BHA coatings, at different time intervals (1–72 h) [11]. Slight modifications of the acronyms from the original figure were done for uniformization with the sample codes used in this review.

Representative SEM images of *S. aureus* biofilm developed on the surface of Ti and BHA:Li structures are given in Figure 10, at a magnification of 10,000×. One can observe that the density of *S. aureus* cells was significantly more reduced on BHA:Li surfaces as compared to Ti ones, which is in good agreement with the quantitative results presented in Figure 9.

**Figure 10.** SEM images corresponding to *S. aureus* biofilm after 72 h of development on the surfaces of Ti and bovine hydroxyapatite doped with Li2O coatings.

BHA:LiC coatings exhibited a slightly increased antimicrobial activity against *S. aureus* biofilm, as compared to BHA and BHA:LiP ones. In the case of *C. albicans*, the antifungal activity started to be pronounced after 12 h of incubation and with a drastic decrease of the fungal biofilm after 48 and 72 h, respectively (Figure 11).

**Figure 11.** Graphic representation of the logarithmic values of *C. albicans* biofilm cells developed on the surfaces of Ti, bovine hydroxyapatite (BHA), and doped BHA coatings, at different time intervals (1–72 h) [11]. Slight modifications of the acronyms from the original figure were done for uniformization with the sample codes used in this review.

Again, the order of efficiency in reducing colonies was BHA:LiC > BHA:LiP > BHA. SEM images revealed that, when grown on the surface of Ti, *C. albicans* produced biofilms, with interconnected cells through filaments, while, when seeded on the surface of the synthesized BHA or doped BHA coatings, the cells were ovoid and adhered in isolated unicellular forms, with completely absent filaments (Figure 12).

**Figure 12.** SEM images corresponding to *C. albicans* biofilm after 72 h of development on the surfaces of Ti and bovine hydroxyapatite doped with Li2O coatings.

Representative SEM images of *C. albicans* biofilm developed on the surfaces of Ti and BHA:Li structures are given in Figure 12, at a magnification of 10,000×. While bare Ti samples were completely covered by a continuous layer of cells, the number of yeast cells developed on the surface of BHA:Li coatings was significantly decreased. They adhered exclusively in isolated, unicellular form, without interconnecting filaments. Duta et al. [11] assessed the antibacterial and antifungal activity of BioHA coatings to one of the following factors: (*i*) difference in polarity between negative-coated surfaces (associated with the presence of hydroxyl hydrophilic functional groups) and *S. aureus*/*C.* *albicans* biofilms matrix, which is also negatively charged, due to its main component, represented by self-produced exopolysaccharides, (*ii*) coatings will facilitate the formation of a highly alkaline medium, which consequently leads to the destruction of lipids that are the main component of microbial cell membrane, or (*iii*) the smooth surfaces of the films, which reduce the contact area between material surface and bacterial cells, or (iv) metal accumulation inside bacterial cells, followed by the cellular wall disruption, and, eventually, cellular lysis.

#### *6.5. In Vivo Tests*

In vitro tests were demonstrated to play a key role in the biological assessment of biomaterials by offering in-sight information on their possible behavior in a biological environment. On the other hand, to understand the complex processes that might occur in a living system and to provide accurate data to completely validate the performances of a given biomaterial designed for clinical trials, in vivo tests are extremely important and should, therefore, follow in vitro ones.

In the literature, there are many in vivo reports on CaPs, but, to the best of our knowledge, there are few works only on HA coatings synthesized by PLD, and, unfortunately, no studies at all on BioHA coatings synthesized by PLD. However, the authors of this review are confident that future in vivo studies of BioHA coatings should deliver interesting results, since BioHA materials proved many times to possess important advantages over synthetic HA. In the following paragraphs, a selection of the papers dealing with PLD of synthetic HA will be considered and the obtained results will be reviewed.

The main purpose of CaP coatings is to stimulate and accelerate new bone formation around an implant [178], and most of the reported research studies revealed positive and favorable effects in this direction [179]. In general, the spectrum of animals used for testing CaP coatings synthesized by PLD is reduced, and this includes rats, rabbits, mini-pigs, dogs, goats, or sheep [14]. The advantages and disadvantages of different animal models used were discussed in Reference [139].

Chen et al. [180] synthesized using PLD fluoridated synthetic HA (HA:F) structures onto Ti implants and inserted them into rat femurs. By using microcomputed tomography, histology, and sequential polychromatic fluorescent investigations, their osteo-inductive and osseointegration activity in comparison with that of uncoated Ti implants was assessed. The results of their study demonstrated that HA:F-coated implants were characterized by superior osteogenesis and osseointegration properties (1.5 times more new bone at one week and more than four times more new bone at 4 and 8 weeks, respectively) as compared to bare (uncoated) Ti specimens, which can advance them as viable candidates for future applications in dentistry.

Mroz et al. [181] reported on Mg-doped HA coatings fabricated by PLD onto porous Ti implants, which were introduced in rabbit femoral bones for six months. The evaluation of bone–implant interaction and bone volume in the region of interest were performed. Histopathological investigations showed that all implants integrated well with the surrounding bone with ingrowth of newly formed bone into the pores of the implants. No inflammatory or foreign body reaction were noticed at the implant site. They note that, a significant increase in bone volume for Mg-doped HA implants, as compared to uncoated implants, was detected.

Peraire et al. [182] conducted a comparative study of the biological stability and the osteo-conductivity of HA coatings produced by PLD and plasma spraying. The functionalized grit-blasted titanium rods were implanted in the proximal tibia of mature New Zealand White rabbits. The obtained data suggested superior bone growth (percentage of bone contact) in the case of HA–PLD coatings (86%) as compared to the HA–PS coatings (67%) or the uncoated grit-blasted titanium (69%), after a period of 24 weeks implantation. Taking into consideration these results, the initiation of clinical evaluation of this coating method for medical applications was encouraged.

Mihailescu et al. [183] reported on synthetic HA, Mn2+-doped carbonated HA and octacalcium phosphate (OCP) coatings fabricated by PLD onto Ti implants for insertion in the tibia bones of New Zealand White female rabbits. The bone–implants bonding was assessed by a tensile (pull-out) test. The measured value of the pull-out force was found to be more than double in the case of

CaP-functionalized Ti implants, as compared to the control (uncoated) ones. In addition, the pull-out force increased by 25% for the Mn2+-doped carbonated HA and 10% in OCP-coated implants, in comparison with the synthetic HA-coated implants. Mihailescu et al. concluded that nanostructured CaP depositions manifested a significant potential for improving the performance of functionalized Ti implants in bone, and the composition and structure of the CaP coating have a significant influence on their biological effect.

Dostalova et al. [184] functionalized by PLD the surface of Ti6Al4V dental implants with HA coatings. Coated and uncoated implants were inserted into the lower jaw of mini-pigs. The evaluation of implants' osseointegration degree was assessed by polarized and fluorescent light microscopy with computer image processing. The results of the study proved that the osseointegration of the coated layer was superior for HA coatings (~77.2%) than the integration of uncoated Ti implants (~65.2%). With the owned advantage of delivering an inert and osseoconductive ceramic coating, the same team concluded that the PLD could be a viable method for coating metallic implants.

#### **7. Conclusions and Perspectives**

When animal-origin calcium phosphates (CaPs) are tackled, pulsed laser deposition (PLD) could be one of the most interesting choices for coating synthesis. If synthetic hydroxyapatite (HA) has a complex molecule with a large number of atoms and functional groups difficult to transfer in form of thin films by physical deposition techniques, natural apatites become even more complicated to transfer because of new functional groups and substitution ions, which further complicate the molecule. PLD is renowned for the ability to transfer stoichiometrically very complex molecules and this advantage should make it one of the premiere candidates for the transfer of such complex materials.

The most common sources for obtaining animal-origin HA for PLD targets are swine, ovine, and bovine teeth or bones. They are immersed in alkali solutions and further calcinated at ~800 ◦C to remove the organic parts. The powder resulting after grinding and milling can be pressed into targets or mixed with various doping agents to change mechanical or biocompatibility properties.

Coatings' morphology depends on the thickness, the origin of the natural apatite, and on the doping agents. In general, films are covered by round droplets that originate from the expulsion of the liquid phase during target irradiation. Their number and size can drastically differ depending on the experimental conditions. Usually, films in the range of hundreds of nanometers mimic the substrate morphology (especially if it is rough), with surface features that are not evident. Thickness seems to be dependent on the experimental conditions (number of pulses, laser fluence, and gas pressure), but also on the type of apatite. Studies on coatings synthesized in the same experimental conditions on sheep, cow, and cow-HA and bio-glass revealed different thicknesses for different types of materials.

Structural differences between synthetic HA and animal-origin apatites or between different types of animal origin-apatites cannot be shown by X-ray diffraction only. All these materials display the same crystalline peaks in the diffraction patterns. Differences appear in Fourier transform infrared spectroscopy that contain, in the case of animal-origin HA, bands corresponding to (HPO4) <sup>2</sup><sup>−</sup> or (CO3) <sup>2</sup><sup>−</sup> groups.

In terms of adherence to the substrate, PLD synthesized coatings from doped or undoped animal-origin HA could easily surpass the mandatory value of 15 MPa of bonding strength established by international standards for biocompatible coatings to be used in implantology.

Biocompatibility of animal-origin HA was consistently being reported as surpassing that of synthetic HA both in cells morphology, cells proliferation, and bioactivity. Moreover, when doped with Ti or Li2CO3, the effect was a boost in cell proliferation.

Animal-origin HA could be, therefore, a reliable, promising, safe, and, most importantly, low-cost source for coatings that reach biomimetism. For PLD, the field is still in his incipient development, with numerous possibilities to expand in terms of natural-HA sources, doping agents, or morphology and structural control of the obtained coatings. The transfer from PLD, which is essentially a laboratory technique to an industrial coating technique also belongs to the future.

Besides their often-increased bioactivity as compared to synthetic HA, BioHA coatings can display antibacterial and antifungal properties, which are not present in the case of synthetic HA. Undoped or BioHA doped with MgO, or carbonates and phosphates of lithium, were active against various microorganisms. Possible explanations for this property could be the presence of dopants, which have antibacterial and antifungal properties by themselves, the smoother surfaces of PLD deposited BioHA films, which reduce the contact area between coating and bacteria, and the metal ions accumulation inside microorganisms followed by the membrane disruption and cell apoptosis.

The authors of this review express their confidence that the future of biomimetic coatings belongs to the natural-origin CaPs source materials and their perfect transfer and delicate tuning in terms of thickness, crystallinity, and functional groups can be optimally studied using PLD as deposition method.

**Author Contributions:** Conceptualization, L.D. and A.C.P. Methodology, L.D. and A.C.P. Software, L.D. Validation, L.D. and A.C.P. Formal analysis, L.D. and A.C.P. Investigation, L.D. and A.C.P. Resources, L.D. and A.C.P. Data curation, L.D. and A.C.P. Writing–original draft preparation, L.D. and A.C.P. Writing–review and editing, L.D. and A.C.P. Visualization, L.D. and A.C.P. Supervision, L.D. and A.C.P. Project administration, L.D. and A.C.P.

**Funding:** Two grants from the Ministry of Research and Innovation, CNCS-UEFISCDI, No. PN-III-P1-1.1-PD-2016-1568 (PD 6/2018) and PN-III-P1-1.1-TE-2016-2015 (TE136/2018), within PNCDI III supported this work. The authors acknowledge and thank the partial support from the Core Programme – Contract 16N/2019.

**Acknowledgments:** The authors acknowledge G.-P. Pelin for the SEM images that were used in this review for exemplification purposes and for the schematic with the PLD experimental set-up, C. Chifiriuc for data on biofilm development on bovine hydroxyapatite doped with Li2O coatings, and F.N Oktar for providing animal-origin powders that were used for numerous studies of bioactive coating synthesis.

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

#### **References**


© 2019 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* **Shellac Thin Films Obtained by Matrix-Assisted Pulsed Laser Evaporation (MAPLE)**

**Simona Brajnicov 1,2, Adrian Bercea 1,3, Valentina Marascu 1, Andreea Matei 1,\* and Bogdana Mitu <sup>1</sup>**


Received: 14 June 2018; Accepted: 3 August 2018; Published: 7 August 2018

**Abstract:** We report on the fabrication of shellac thin films on silicon substrates by matrix-assisted pulsed laser evaporation (MAPLE) using methanol as matrix. Very adherent, dense, and smooth films were obtained by MAPLE with optimized deposition parameters, such as laser wavelength and laser fluence. Films with a root mean square (RMS) roughness of 11 nm measured on 40 × <sup>40</sup> <sup>μ</sup>m2 were obtained for a 2000-nm-thick shellac film deposited with 0.6 J/cm<sup>2</sup> fluence at a laser wavelength of 266 nm. The MAPLE films were tested in simulated gastric fluid in order to assess their capabilities as an enteric coating. The chemical, morphological, and optical properties of shellac samples were investigated by Fourier transform infrared spectroscopy (FTIR), X-ray photoelectron spectroscopy (XPS), scanning electron microscopy (SEM), and atomic force microscopy (AFM).

**Keywords:** thin films; matrix-assisted pulsed laser evaporation; shellac; enteric coatings

#### **1. Introduction**

Shellac is a biomaterial, a resin secreted by the female lac bug *Kerria lacca* on the trees in India and Thailand. It consists of esters and polyesters of polyhydroxy acids. The main components of shellac are aleuritic and shellolic acid [1]. Other components can be butolic and jalaric acid [2]. The color can vary between light yellow and dark red depending on the origin of the raw material.

Shellac possesses very interesting properties, such as being UV resistant, and it is a barrier for water and a very good dielectric [3]. As a thin film, shellac can be easily deposited by drop casting from a solution with ethanol or methanol. Shellac thin films have a dielectric constant between 3 and 5 [3] and very low surface roughness. They have been used as a substrate for organic field-effect transistor (OFET) and as a gate dielectric [3]. When used as a substrate, a concentrated solution is prepared. The solution has to evaporate for about half an hour at 50 ◦C and then it has to be heated for an hour at 70 ◦C in order to obtain roughness lower than 1 nm [3].

Shellac is used in the food industry as a coating for fruits as protection against dehydration and for the glossy and attractive aspect [4]. Shellac can be added to the composition of food packaging materials for protection against bacteria and as a water repellent element [5]. Nail polish [6], furniture and wood varnish, and the protection of art objects [7] are some other applications shellac can be used for.

Shellac biopolymer can be used as a new carbon source for growing bubble-decorated graphene oxide films with high magneto-resistance as reported by Singhbabu et al. in [8].

Improved properties for wood protection can be achieved for shellac films as a coating by dispersing inorganic nanoparticles (1–2%) in alcoholic shellac solution as demonstrated by Weththimuni et al. in [9]. Increased hardness and increased UV resistance can be obtained using ZrO2

and ZnO nanoparticles. The low cost and low processing temperature make shellac a good candidate for microfluidics applications via hot embossing in the same way standard polydimethylsiloxane (PDMS) is used [10].

In pharmaceutics, shellac can be used for encapsulation and micro encapsulation [11] as an enteric coating due to its acidic character, delivering the drug to the colon for specific applications [12–14]. Kumpugdee-Vollrath et al. obtained controlled drug release using coatings of shellac and pectine in [15]. Coaxial electrospinning has been reported for the obtaining of shellac nanofibers for colon-targeted drug delivery by Wang et al. in [16].

In this work, we used a laser deposition technique to grow thin films of shellac for targeted drug release. The employed laser technique was Matrix-Assisted Pulsed Laser Evaporation (MAPLE), a variation of the pulsed laser deposition (PLD) procedure. Pulsed laser deposition is especially used in the deposition of inorganic thin films [17–19] and it cannot be used for all types of delicate materials. The PLD method can lead to their photochemical interaction when using laser emitting in UV. For polymers, organic materials, and materials with a complicated structure, the MAPLE system is preferred [20–22]. Important advantages arising from our approach using laser-based techniques are the good adherence, the high density of the films, and control over their thickness. Another important issue is the fact that MAPLE is a "dry technique". During the coating procedure with MAPLE, the drug is not in contact with the matrix (solvent) and drug–solvent chemical interactions are therefore avoided.

MAPLE-grown films have been tested in simulated gastric fluid, where the resin coating offers the needed resistance to the stomach and small intestine environment [23].

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

Shellac wax-free was purchased from Sigma-Aldrich (St. Louis, MO, USA) and used as received. The MAPLE technique was employed using a low concentration of shellac (1–2 wt % dissolved in methanol). The solutions were poured into a copper target holder and they were frozen with liquid nitrogen. During the experiments, the targets were kept in a solid state. The influence of various laser wavelengths (266, 355, 532, and 1064 nm) on the deposition process was investigated. The laser fluence was in the range 0.6–2 J/cm2 for the experiments using the UV wavelength and up to 4 J/cm<sup>2</sup> for the visible and IR wavelengths. We will present the results just for UV laser beam irradiation because green and infrared light did not lead to continuous films, most probably due to the fact that the methanol matrix does not absorb light at these wavelengths [24] and the target evaporation took place mainly on defects. The laser spot was set in the range 0.9–1 mm2 for all the experiments. The substrates of <sup>10</sup> × 10 mm2, i.e., two side polished Si (100) wafer and quartz, were kept at room temperature and positioned at a distance of 4 cm from the target. The base pressure in the deposition chamber was ~8 × <sup>10</sup>−<sup>6</sup> mbar and it increased during laser irradiation to 10−<sup>4</sup> mbar due to solvent evaporation. The number of pulses varied between 18,000 and 90,000 for a pulse repetition rate of 10 Hz. Before the deposition, the substrates were cleaned in an ultrasonic bath, using a standard procedure implying soap solution, water, acetone, and ethanol for 15 min each, and dried in a nitrogen gas flow.

For measuring the deposition rates of the films, contact profilometry studies were performed by means of a KLA-Tencor P-7 (Milpitas, CA, USA) step profiler equipped with a 2-μm radius and a 60◦ stylus working in contact with 2 mg applied force and scanning on a 3-mm length of the samples with a scan speed of 100 μm/s. Surface morphology was investigated by atomic force microscopy (AFM) and scanning electron microscopy (SEM) with an "XE-100" AFM produced by Park Systems (Suwon, South Korea) and a Quanta ESEM FEG 450 SEM (FEI, Hillsboro, OR, USA). Fourier transform infrared absorption spectra were acquired with a Jasco 6300 FTIR system (Oklahoma, OK, USA), in the range 4000–400 cm<sup>−</sup>1, with 4 cm−<sup>1</sup> resolution.

The wettability characteristics of the films were evaluated by contact angle measurements. The water contact angle was measured using an optical microscope KSV CAM101 (KSV Instruments Ltd., Helsinki, Finland) with water drops of 0.5–1 μL and by mediating the processed angles obtained upon five drops for each sample.

The simulated gastric fluid (SGF) dissolution media that is described in United States Pharmacopeia 33-28NF (2010) and European Pharmacopeia 7.0 (2010) was prepared as follows: 2.0 g NaCl, 3.2 g pepsin, 7.0 mL HCl, and 1000 mL milli q water. The solution used had a pH of 1.2. Shellac films deposited by MAPLE were immersed in SGF for 15, 30, 60, 120, and 240 min.

X-ray photoelectron spectroscopy (XPS) survey spectra and high-resolution XPS scan spectra were acquired for shellac dropcast, the initial film, and films immersed in SGF for different time durations using an Escalab Xi<sup>+</sup> system, Thermo Scientific (Waltham, MA, USA). The survey scans were acquired using an Al Kα gun with a spot size of 900 μm, a pass energy of 100.0 eV, and an energy step size of 1.00 eV. For the high-resolution XPS spectra, the pass energy was set to 10.0 eV, the energy step size was 0.10 eV, and 10 scans were accumulated.

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

The MAPLE-deposited films using a 266 nm irradiating wavelength are very smooth and the surface is droplet-free. The deposited films have low roughness: less than 1 nm on 5 × <sup>5</sup> <sup>μ</sup>m<sup>2</sup> for films with a thickness in the range 400–2000 nm; however, at micrometric scale, the surface has a wave-like aspect (Figure 1).

The films present a slight hydrophobic tendency, with contact angles in the range 92–105◦ on samples prepared by MAPLE at a 266-nm wavelength.

**Figure 1.** AFM and SEM images on a shellac film deposited from a 1 wt % shellac in a methanol frozen target on an Si substrate as a result of 72,000 pulses at a 266-nm wavelength with a fluence of 0.5 J/cm2. RMS roughness is 23 nm on 40 <sup>×</sup> <sup>40</sup> <sup>μ</sup>m2. The water contact angle is 97.2◦.

Profilometric studies on the shellac films deposited using UV-irradiated light revealed deposition rates between 0.01 and 0.3 Å/pulse for using fluences in the range 0.5–2 J/cm<sup>2</sup> (Figure 2). A significant increase was obtained when using a higher amount of shellac as a guest molecule in the target. The deposition rate as a function of fluence has very interesting behavior; there is no evident increase with fluence. More than that, a decrease of the deposition rate is encountered upon increasing the laser fluence when the target contained 2 wt % shellac. It has also been reported in Ref. [25] the case of depositing lysozyme and myoglobin proteins where the deposition rate increases up to a maximum, after which it has a clear decrease. This may be due to a low sticking coefficient on the already deposited film or to the ejection of the fragmented shellac scattered from the normal direction onto the target [25]. Additionally, at high fluences and higher shellac concentrations, the amount of the evaporated solvent molecules and fragments are higher, leading to a pressure increase in the deposition chamber, which blocks shellac to reach the substrate.

**Figure 2.** Dependence of shellac deposition rate on the laser fluence as resulting from profilometry on samples prepared by matrix-assisted pulsed laser evaporation (MAPLE). Inset AFM image on a shellac film deposited on an Si substrate using 355 nm laser wavelength as a result of 72,000 pulses at 1 J/cm2 from a 2 wt % shellac in a methanol frozen target.

The very low deposition rate obtained for the experiments performed by using the 355 nm laser wavelength (the red circle on Figure 2) corroborated with the high roughness of the surface covered with droplets and elongated structures (inset Figure 2) preventing the utilization of this wavelength for shellac deposition in applications where a continuous and smooth film is needed. As already mentioned in the experimental part, the use of the 532 nm and 1064 nm irradiation wavelengths did not even lead to the deposition of continuous films. The few aggregates that could be seen on the substrate are related to the absorption of laser light by defects, seed electrons, and impurities in the frozen target, which lead to material transfer onto the substrate. Taking into consideration all these aspects, the following experiments involved the use of films deposited using the 266 nm laser wavelength.

The surface analysis revealed smooth films with low roughness grown at low fluences, unlike the case for higher fluences, where the surface is covered with wrinkle-like structures (Figure 3). Such features were previously explained upon molecular dynamics simulations of the MAPLE process [26,27]. They originate from large clusters of target material, including the solvent, which are ejected from the frozen target and deposited onto the substrate, followed in a second stage by solvent evaporation from the substrate and film formation.

As a conclusion, for experiments on the film's behavior upon immersion in SGF, the low range of fluences with values up to 0.6 J/cm<sup>2</sup> were chosen.

**Figure 3.** (**a**) The RMS roughness determined for areas of 40 <sup>×</sup> <sup>40</sup> <sup>μ</sup>m2 for shellac films deposited by MAPLE. The line between the points is for eye-guidance; (**b**) From left to right: AFM images for 0.6, 1.3, and 1.6 J/cm2.

Fourier transform infrared spectroscopy (FTIR) investigations were performed on films deposited by MAPLE and the results were compared to the pristine material. The laser fluence and the amount of shellac in the target solution were the investigated parameters (Figure 4).

**Figure 4.** FTIR spectra of shellac films.

The large band in the range 3100–3700 cm−<sup>1</sup> with a maximum at about 3420 cm−<sup>1</sup> is attributed to the O–H stretching vibration, while the O–H bending vibration is identified at 1250 cm<sup>−</sup>1. The strong symmetric and asymmetric stretching vibrations of CH2 can be easily identified at 2855 cm−<sup>1</sup> and 2932 cm−1, respectively. The other two strong vibrations peaked at 1712 cm−<sup>1</sup> and 1733 cm−<sup>1</sup> are attributed to the C=O stretching vibration of esters [23] and acids, respectively [28,29]. The other medium bands of the characteristic fingerprint are: 1636 cm−<sup>1</sup> corresponding to the C=C stretching vibration of vinyl [30]; and 1463 cm−<sup>1</sup> and 1376 cm−<sup>1</sup> are attributed to CH3 asymmetric and symmetric bending, respectively [31]. The weak signal at 1412 cm−<sup>1</sup> was associated to the CH2 group attached to the ester chain as reported in Zumbühl et al. in [32]. The absorption bands at 1175 and 1050 cm−<sup>1</sup> may be due to stretching vibrations of C–O and C–C bonds [31]. Weak signals can be seen at 944 and 882 cm−<sup>1</sup> for pristine shellac, dropcast, and the low fluence used of 0.6 J/cm<sup>−</sup>2. They can be attributed to the O–H out-of-plane deformation in carboxylic acids, to the C–H out-of-plane deformation in aldehydes, or to their combination [33].

The FTIR spectra show that the general aspect of the spectra of shellac deposited using laser fluences up to 2 J/cm<sup>2</sup> is not changed, leading to the conclusion that a significant part of the film material remains unaffected by the laser transfer process. The best concordance with the pristine shellac of the intensities ratio was obtained for low fluences, where all shellac characteristic absorption bands can be found. However, at laser fluences of at least 1.33 J/cm2, one can notice the appearance of two additional bands in the region 1400–1600 cm−1, namely at 1515 and 1550 cm−1, related to the symmetric and antisymmetric stretching modes of carboxylate moieties (COO–) formed upon deprotonation of the carboxylic acids of shellac [34]. This is accompanied by a decrease of C–O and C–C stretching bands at 1175 and 1050 cm−1, suggesting a tendency for material crosslinking upon laser transfer, which was confirmed by the broadening of the bands and the loosening of the fine structure noticed at large fluences.

#### **4. Acidic Solution Resistance on Samples Prepared by MAPLE Deposition**

As mentioned in the experimental part, we have investigated the enteric capacity of shellac to be used as a coating for colon-targeted drugs. The chemical evolution of shellac in simulated gastric fluid (SGF) was analyzed at different time intervals. The time intervals were selected in accordance with [35]. Figure 5 presents the FTIR spectrum for a film immersed for 240 min in SGF. We have calculated and compared the intensities of the strong absorption bands from 3423 cm−<sup>1</sup> (O–H) and 1711 cm−<sup>1</sup> (C=O) normalized to the strongest band in the spectra appearing at 2933 cm−<sup>1</sup> for the films immersed in SGF to the as-deposited films.

**Figure 5.** FTIR spectra of shellac film as-deposited and film immersed in simulated gastric fluid (SGF) for 240 min. Relative intensities of the absorption bands of O–H and C=O (3423 and 1711 cm−1, respectively) normalized to the CH2 band (the strongest band of at 2927 cm<sup>−</sup>1).

It is clear that no significant modification in the material composition occurs even upon immersion of the shellac film into SGF for 240 min. Upon the first 30 min of exposure to SGF, a low decrease of the populations of the C=O stretching vibration of esters and the O–H stretching vibration relative to stretching vibrations of CH2 identified at 2927 cm−<sup>1</sup> was noticed. Nevertheless, upon a longer time of exposure in SGF, the shellac films were stable in the acidic gastric medium, no significant transformation could be identified, and the analyzed populations remained approximately constant up to the maximum investigated time of 240 min, which corresponds to the typical passage time through the digestive system.

The modification of surface chemical composition upon immersion in SGF, as revealed by the XPS survey spectra, is presented in Figure 6. It is shown that a slightly higher amount of oxygen is present in the MAPLE-deposited film as compared to the dropcasted shellac. Also, an incorporation of nitrogen onto the shellac surface appears and its concentration increases from 2.63% at 15 min of immersion up to 5.02% at 240 min of immersion. This is associated with pepsin interaction with shellac films, leading to the covalent attachment of nitrogen to the shellac structure.

**Figure 6.** Dependence of the surface elemental concentration on the SGF immersion time.

The high-resolution XPS spectra in the carbon binding energy region were deconvoluted both for the shellac material and for those kept in SGF for various times; the results for the 240 min immersion are presented in Figure 7.

**Figure 7.** High resolution C 1*s* and fitting for non-irradiated shellac (**a**) and for a sample immersed in gastric fluid for 240 min (**b**).

The spectrum of the initial shellac film is dominated by the C–C/C–H peak at 284.8 eV, while larger peaks at 285.7 eV and 286.7 eV are assigned to \*C–O–C\* hydroxyl ether and O–C\*=O/C\*–O carbonyl peaks, respectively. At higher binding energies, around 289.1 eV the presence of a COOH carboxyl peak is noticed. Upon the interaction of the shellac film with the SGF, one can see the diminution of C–C/C–H-related bonds accompanied by a significant increase and widening of the peak at 285.7 eV. This counts as C–N bonding, which superimposes on the \*C–O–C\* from the shellac, and it appears on the film surface upon the interaction with the pepsin. Looking into the N 1*s* high-resolution spectra, presented in the inset of Figure 7, one can see that the main nitrogen-related bonds are those related to the amide formation, peaked at 400.0 eV, as well as those associated with the protonated nitrogen obtained upon interaction with COOH functionalities [36]. The results are consistent with the increase of nitrogen atomic concentration with the time of immersion in SGF.

The thickness of the films remained intact for all the investigated samples. The interaction of the shellac film with the SGF was evaluated also by means of AFM measurements. We performed analysis of 2000-nm-thick films with an RMS roughness of 0.7–1 nm over a scanned area of 5 × <sup>5</sup> <sup>μ</sup>m2. The step profile measurements show that the film's thickness does not modify significantly even upon 240 min immersion. Nevertheless, the roughness increases by 1 order of magnitude even for an immersion time as low as 15 min and reaches 12.3 nm upon 120 min of immersion. Small pores, both regarding their diameter and depth, had appeared on the shellac surface already upon 15 min of immersion (Figure 8). The pores' density and their depth increases when the exposure time to SGF increases; however, the largest pore identified after 4 h of immersion had a depth of 85 nm, which represents less than 5% of the film's thickness.

**Figure 8.** AFM images on 5 <sup>×</sup> <sup>5</sup> <sup>μ</sup>m2 films of MAPLE shellac films immersed in SGF for different time durations. RMS Roughness on 5 <sup>×</sup> <sup>5</sup> <sup>μ</sup>m2 and number of pores counted on 5 <sup>μ</sup>m2. The MAPLE films have a thickness of 2000 nm and were deposited using 2% shellac dissolved in methanol with a fluence of 0.6 J/cm2 and 72,000 pulses.

The behavior of MAPLE-deposited shellac films upon immersion in SGF for 240 min was compared to that of films deposited by the dropcast method and submitted to the same environment. The AFM images are presented in Figure 9. The 5 × <sup>5</sup> <sup>μ</sup>m<sup>2</sup> AFM image for the dropcast film, before SGF immersion, presents a smooth film, with an RMS roughness of 0.6 nm. After SGF immersion, the surface of the film is strongly changed; the film is very rough and holes similar to those noticed in the case of MAPLE films can be identified. Nevertheless, the higher roughness of the dropcast films after immersion (88 nm measured on 5 × <sup>5</sup> <sup>μ</sup>m2) compared to the MAPLE films also supports the hypothesis that the MAPLE technique produces more dense and compact films. This can be associated with the higher impinging energy of the material reaching the substrate in the case of MAPLE processing, which leads to a more compact deposit with respect to the softer dropcast process.

**Figure 9.** AFM image on 5 <sup>×</sup> <sup>5</sup> <sup>μ</sup>m<sup>2</sup> on a dropcast film (**a**); AFM image on 40 <sup>×</sup> <sup>40</sup> <sup>μ</sup>m2 after 240 min immersion in SGF (**b**).

#### **5. Conclusions and Outlook**

The deposition of shellac thin films by MAPLE has proven to be successful by utilization of a UV laser. The obtained samples have low roughness and are droplet-free. We have obtained films with a thickness of 2000 nm and an RMS roughness less than 1% of the thickness for the optimal set of deposition parameters. It can be seen that laser fluence is an important factor in preserving the chemical structure of shellac. For low fluences, all the characteristic absorption bands were found and the best concordance to the pristine shellac was obtained. Upon immersion in Simulated Gastric Fluid, the thickness of the films remained intact even upon 240 min of immersion. The surface of the films changed even for an immersion time as low as 15 min, and the roughness increased as a result of the appearance of holes. The pores' density and size increased when the exposure time to SGF increased, but the biggest hole measured after 240 min in SGF had a depth of just 85 nm. This behavior, corroborated with the UV resistance [28] we have demonstrated previously, makes MAPLE shellac films promising for use as an enteric coating and opens the path for a variety of applications.

**Author Contributions:** Investigation, S.B.; Formal Analysis, A.B. and V.M.; Conceptualization, A.M.; Supervision, B.M.; Writing—Review & Editing, A.M., B.M., S.B., A.B. and V.M.

**Funding:** This research was funded by the Romanian National Authority for Scientific Research and Innovation, CNCS—UEFISCDI (PN-III-P2-2.1-PED-2016-0221 (contract PED 94/2017)—"IPOD").

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

#### **References**


© 2018 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/).

## *Review* **In Vivo Assessment of Synthetic and Biological-Derived Calcium Phosphate-Based Coatings Fabricated by Pulsed Laser Deposition: A Review**

**Liviu Duta**

**Abstract:** The aim of this review is to present the state-of-the art achievements reported in the last two decades in the field of pulsed laser deposition (PLD) of biocompatible calcium phosphate (CaP) based coatings for medical implants, with an emphasis on their in vivo biological performances. There are studies in the dedicated literature on the in vivo testing of CaP-based coatings (especially hydroxyapatite, HA) synthesized by many physical vapor deposition methods, but only a few of them addressed the PLD technique. Therefore, a brief description of the PLD technique, along with some information on the currently used substrates for the synthesis of CaP-based structures, and a short presentation of the advantages of using various animal and human implant models will be provided. For an in-depth in vivo assessment of both synthetic and biological-derived CaP-based PLD coatings, a special attention will be dedicated to the results obtained by standardized and microradiographies, (micro) computed tomography and histomorphometry, tomodensitometry, histology, scanning and transmission electron microscopies, and mechanical testing. One main specific result of the in vivo analyzed studies is related to the demonstrated superior osseointegration characteristics of the metallic (generally Ti) implants functionalized with CaP-based coatings when compared to simple (control) Ti ones, which are considered as the "gold standard" for implantological applications. Thus, all such important in vivo outcomes were gathered, compiled and thoroughly discussed both to clearly understand the current status of this research domain, and to be able to advance perspectives of these synthetic and biological-derived CaP coatings for future clinical applications.

**Keywords:** calcium phosphate-based coatings; synthetic and natural hydroxyapatite; pulsed laser deposition; in vivo testing; biomedical applications

#### **1. Introduction**

The biomedical domain has witnessed over the last decades a significant development due to an extensive demand for a wide variety of implants, grafts, and/or scaffolds. The bone tissue engineering field has therefore expanded to be able to address a wide spectrum of bone-related injuries and offer viable and efficient solutions. This is mainly achieved by combining the properties of bioactive materials and cells for an improved and faster bone tissue ingrowth. Implants' surface functionalization and modification with performant bioactive materials is of high interest as it provides various possibilities to modify the surface properties of biomaterials to make them suitable for specific medical applications. This technology is currently applied both for the prevention of failure and the prolongation of the bone implants' life [1]. The fabrication of resistant implants able to bypass the difficulties related to their rejection from the living bodies is therefore of huge research interest. It is important to note that the global market for implantable medical devices was valued at \$72,265 million in 2015 and foreseen to attain \$116,300 million in 2022 [2].

Calcium phosphates (CaP) represent the main inorganic component of bone tissues [3]. They are the most utilized bioceramics in the medical field (i.e., orthopedics and

**Citation:** Duta, L. In Vivo Assessment of Synthetic and Biological-Derived Calcium Phosphate-Based Coatings Fabricated by Pulsed Laser Deposition: A Review. *Coatings* **2021**, *11*, 99. https://doi.org/10.3390/ coatings11010099

Received: 11 December 2020 Accepted: 12 January 2021 Published: 18 January 2021

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

**Copyright:** © 2021 by the author. 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/).

dentistry [4–7]), as coatings for a wide range of metallic implants [8]. In the last few decades, research emphasis was put on hydroxyapatite (HA), which is the most frequently used CaP due to some interesting characteristics, such as its role as scaffold for osteogenic differentiation [9] and its capacity to stimulate and accelerate new bone formation around implants [10–12]. One should note that there are currently two ways to obtain HA: the first one uses inorganic synthesis by different chemical routes (i.e., hydrothermal [13], co-precipitation [14], or sol–gel [15]). In the case of hydrothermal method, the synthesis temperature is relatively low, and the reaction conditions are moderate. The obtained products have high crystallinity, high purity, and controllable shape and size [16]. The chemical precipitation of nano-sized powders from salt solutions allows for the rapid synthesis of large amounts of material in a controlled manner [17]. The sol–gel process is a wet chemical method that involves atomic level molecular mixing, and provides good control over the composition and chemical homogeneity [18]. It should be emphasized that some of these approaches imply the use of complex processes, which could generate pollutant chemical wastes. Moreover, the composition of synthetic CaP is more complex than the one of biological-derived apatites due to both multiple lattice substitutions and the presence of ion vacancies [19]. As a consequence, a good alternative to classical chemical routes (i.e., extraction from renewable CaP resources) was introduced. One should note that the most important primary natural reservoirs of CaP are either bones (of mammalian [20,21] or fish [22,23] origin) or biogenic sources (egg-shells [24], mussel shells [25], or marine shells [26]). Unfortunately, these are generally treated as food industry wastes only [27]. Besides its low production cost, another great advantage of this fabrication route is the preservation of compositional and structural properties of the source material [28]. Furthermore, these as-fabricated biological-derived HA (BioHA) materials are well-suited to achieve a perfect synergy with the biological media due to their content in trace elements [4,8]. These elements have a determining role in the proper adjustment of biological properties (i.e., solubility, surface chemistry, and morphology), to keep compatible with the natural human bone [5,29]. As compared to chemically-synthesized HA, BioHA has different composition, stoichiometry, degree of crystallinity, degradation rate, and overall biological performance. From the biomimetic point of view, BioHA materials are more appropriate than synthetic HA to repair the skeletal system. It was reported that HA obtained from renewable, low-cost resources (i.e., animal or fish bones and egg-shells) can lead to physical-chemical characteristics and biological response comparable or even improved than those obtained in the case of synthetic ones, due to their resemblance with bone apatite [30,31]. The conversion of these by-products into HA is envisaged both as a strategy for wastes management and an economically feasible approach to reduce the overall costs for HA production. As compared to synthetic HA, the osseoconduction speed of BioHA takes place more rapid due to its higher solubility and content of Mg2+ and Na<sup>+</sup> ions. It is very important to note that these elements are generally implicated in the process of bone remodeling [32]. A detailed comparison between BioHA and synthetic HA materials can be found in Ref. [33].

In contrast to their excellent bone regeneration properties, CaP-based materials are brittle in bulk [34] (the brittle nature being in relation with their primary ionic bonds [3]), and characterized by poor mechanical properties. Their low impact resistance and tensile strength [35] represent critical drawbacks, which limit their wider clinical applications. However, the compressive strength value is rather good, exceeding that of the bone [35]. In general, an implant is fabricated from Ti or its medical-grade alloys. Despite their improved mechanical characteristics, Ti implants are characterized by low osseointegration rates. To overcome these shortcomings, HA can be used as a coating for Ti implants. This way, a combination between the ceramic's bioactivity and the metallic substrate's mechanical performances is attained [34,36]. To even accelerate coatings' osteoinduction and biomechanical fixation to the metallic substrate, some additions or HA doping with different ion concentrations were also reported [7,8]. Besides orthopedic prostheses and dental implants, HA coatings are currently being deposited also on (macro) porous scaffolds, thus expanding their applications towards bone regeneration therapies. Next to CaP, another important class of bioactive materials is represented by bioglasses (BG) [37,38], that demonstrated excellent osteointegrative characteristics in bulk form. Metallic implants' functionalization with BG coatings can be considered a good alternative for rapid bone reparation and regeneration [39].

In the field of thin film growth, the PLD method proved to be a simple, versatile, fastprocessing, and cost-effective technique, which allows for a precise control over the growth rate and morphology to fabricate high-quality coatings [40,41]. In the case of CaP-based biomaterials, one of the most important and enabling characteristics of this technique is its capability to grow stoichiometric films [41]. This is mainly due to a high ablation rate that causes the evaporation of all elements at the same time [42]. The deposition temperature, substrate position, or background pressure can be independently controlled for the ease tailoring of the crystallinity, chemical composition, thickness and/or surface roughness of the fabricated structures [40]. One should emphasize upon that both film thickness and composition nonuniformity over high surface areas are two important drawbacks of the PLD technique, which can be fortunately overcome using special experimental set-ups (i.e., laser beam rastering over large diameter targets) [43]. In this respect, excimer lasers represent one economical choice for large-area commercial scale-up of the PLD process, advancing this technique as a viable alternative for future industrial applications [41].

There are studies in the dedicated literature on the in vivo testing of CaP-based coatings (especially HA) synthesized by many physical vapor deposition methods as alternatives to plasma spraying (PS), but only a few of them addressed the PLD technique. Moreover, to the best of my knowledge, there are no any review papers to summarize all the efforts dedicated in this direction, and the main objective of this review is to fill in this gap of knowledge. Therefore, the current review is focused on the gathering, compilation, and thoroughly discussion of the in vivo results pertaining to the studies reported in the last two decades on either synthetic or biological-derived CaP coatings only, synthesized by PLD technique, for various medical applications.

#### **2. Search Strategy**

#### *2.1. Study Selection*

A comprehensive literature search in the Web of Science (https://apps.webofknowledge. com/) database up to 11 December 2020, was carried out. The search included both animaland human-related studies, and only the publications written in English were considered. The applied search strategy was: ("in vivo" [Topic], "Ti implants" [Topic], AND "Pulsed laser deposition" [Topic]), OR ("hydroxyapatite coatings" [Topic], "in vivo" [Topic] AND "Pulsed laser deposition" [Topic]), OR ("hydroxyapatite" [Topic], "Pulsed laser deposition" [Topic], AND "patient" [Topic]), OR ("hydroxyapatite" [Topic], "in vivo" [Topic] AND "Laser" [Topic]), OR ("hydroxylapatite" [Topic], "osseointegration" [Topic], AND "KrF laser" [Topic]), OR ("hydroxyapatite" [Topic], "osteoconduction" [Topic], AND "laser" [Topic]), OR ("calcium phosphate" [Topic], "in vivo" [Topic], AND "Pulsed laser deposition" [Topic]), OR ("calcium phosphate" [Topic], "in vivo" [Topic], AND "osseointegration" [Topic]).

#### *2.2. Inclusion of Studies*

These searches resulted in the identification of more than 700 studies. These titles were initially screened for possible inclusion, resulting in further consideration of approximatively 100 publications. A careful screening of the abstracts led to 40 full-text articles. Considering the aim of this review, less than half of these articles (i.e., 15) met the inclusion criteria and were chosen and assessed in detail, and parts of the reported results were included and discussed in the present review.

#### **3. Pulsed Laser Deposition Technique**

Bioceramic coatings are currently used for various biomedical applications [9] to modify the implant surface (by increasing its surface roughness), to promote osseointegration. Nowadays, PS is the only industrial technology used for coating dental and orthopedic implants with CaP materials. In this respect, in the dedicated literature there are numerous and very interesting research works that point out and critically evaluate the physical-chemical and biological characteristics of the structures fabricated by this technique, along with their clinical performances [44,45]. The CaP-based coatings fabricated by conventional thermal PS onto medical implants function as an intermediate layer between the tissues and the metallic implants [46]. Despite its wide commercial availability, this technique still has some drawbacks, such as: (i) the synthesized coatings generally consist of several phases (i.e., β-TCP forms at 1200 ◦C, and it transforms into TTCP at T > 1400 ◦C); (ii) at higher synthesis temperatures, the negative influence of the mismatch between the thermal expansion coefficients of HA and TCP and the ones corresponding to Ti-based alloys (11–15 × <sup>10</sup>−<sup>6</sup> cm/(cm·K) vs. 8–10 × <sup>10</sup>−<sup>6</sup> cm/(cm·K), respectively), limit the obtaining of good CaP coatings onto metallic substrates [47]; (iii) it supplies very thick structures with low adherence to the substrate (because the coatings' tensile stresses have a greater tendency to initiate cracks and cause film delamination [47]); (iv) surface morphology, phase composition, or uniformity of crystallization [32,48] are difficult to be controlled; and (v) it is a line-of-sight method [3]. Therefore, no coating technique can be considered perfect and all these drawbacks gradually supported research efforts focused on the introduction of various, alternative coating techniques to PS (i.e., radio frequency magnetron sputtering, PLD, electrochemical deposition, etc.) [1,3,7,8,11]. Among these, PLD technique is worth mentioning due to some important advantages over PS method, such as: (i) a much faster surface deposition process; (ii) a stoichiometric transfer of the material's composition from the target in the synthesized coating; (iii) a better morphological and compositional uniformity; (iv) a lower porosity; (v) precise thickness control; (vi) effectiveness for coating small implants, with complex shapes; (vii) a decreased tendency of the synthesized structures to crack or delaminate; and, very important, (viii) a high adherence to the metallic substrate [1,49]. In the biomedical field, for the fabrication of CaP-based coatings for bone implant applications, one of the most applied plasma-assisted methods is PLD [7]. This technique is able to fabricate dense and extremely adherent films. In this respect, synthetic HA and BioHA coatings obtained by PLD previously demonstrated high adherence to metallic substrates [50]. Moreover, the composition of these coatings is consistent with the one corresponding to the raw (base) materials, along with an improved crystallinity [50]. One should note that in PLD, after the ablation of the target by laser pulses, a plasma plume is generated. When the species existing in the plume reach the surface of the substrate, they may deposit onto the surface and form a film [41]. The number of the deposited species depends both on their density in the plume and their energy. In the case of low energies, the species may not deposit on the substrate surface even if they arrive at the surface [51]. If the substrate temperature is high, the energy of the species can be compensated. Consequently, the number of deposited species onto the substrate surface will increase, along with the density of the droplets. High substrate temperatures also contribute to the atomic diffusion, which, in turn, can determine the appearance of two phenomena: the first one is the atomic rearrangements and crystallization of the film and the second one, the improvement of the film−substrate bonding state [52].

The laser sources appropriate for the ablation of a wide range of materials use wavelengths in the UV domain due to some important advantages over IR and/or visible laser sources, such as (i) a higher penetration depth of the laser beam in the target material and (ii) a higher energy of the photons that allows for a more efficient vaporization of the target [53]. In this respect, the laser sources used for PLD experiments are either excimer lasers (i.e., ArF [54], KrF [55], or XeCl [56], emitting at wavelengths of 193, 248, or 308 nm, respectively) or solid-state lasers (i.e., Nd:YAG [57], emitting at 266 nm). It should be also emphasized that to increase the amount of evaporated material from the ablated target to

the detriment of expulsed liquid or solid phases, lasers emitting in a pulsed regime with pulse durations in either the nanoseconds or picoseconds range are generally used [58]. In these regimes, the absorption process takes place more quickly than in the case of thermal diffusion processes. More insights on the PLD and PS techniques are well described elsewhere [33,41,49].

It is important to mention that post-deposition treatments are generally applied to transform CaP phases with lower Ca/P ratio to crystalline HA. Thus, there are two commercially used post-treatments: sintering [33,59,60] and soaking in alkaline solutions [61,62]. These treatments are generally applied for several hours, in the range of 600–800 ◦C. Their aim is to transform the water trapped in the film during the synthesis process in OH− ions, to stabilize the crystalline structure [34].

There are reports in the dedicated literature on the in vivo testing of CaP-based coatings (especially HA) fabricated by different physical vapor deposition methods, but, to the best of my knowledge, only a few of them addressed the PLD technique. Therefore, for an easy access to information, Table 1 introduces the PLD experimental details given in the papers considered in this review.

**Table 1.** Pulsed laser deposition (PLD) experimental details given in the papers considered in this review, in the chronological order in which they were reported in the literature.


#### **4. In Vivo Assessment of PLD CaP-Based Coatings**

In vitro tests play key-roles in the overall evaluation of a biomaterial by delivering important information on its potential behavior inside a living system. To better understand the complex processes occurring in a living environment and to deliver accurate data for the validation of the biomaterials' performances that target clinical trials, in vivo tests (using either animal or human models) are of key-importance and should, therefore, follow thorough in vitro acceptance [75].

Biocompatibility is an essential and required characteristic of the biomedical materials introduced inside the living systems, presenting a beneficial interaction with the surrounding bone tissues. Immediately after CaP-based functionalized implants are surgically inserted into the living bodies, there are tissue responses that take place at the implanted materials−soft/hard tissues interface. The formation of connective tissue fibers (2–3 weeks), which make a fiber mesh containing carbonated apatite, is one such response that occurs initially [76,77]. This tissue-free layer develops onto the ceramic surface and actively contributes both to a strong fixation of the implant to the surrounding bone tissues and to the acceleration of bone integration and healing at early implantation times [78–81]. These aspects are of key-importance for any implant system [82]. The amount and nature of osseointegration in metallic implants are determined by several factors such as surface

topography, which includes surface roughness [83], wettability [84,85], and surface morphology [86,87]. It was demonstrated that an initial stability and a further stronger interface with the bone tissues are more likely to be achieved in the case of implants with rough surfaces as compared to smooth ones [88], in both animal and human models, with bone ongrowth interface [89,90]. Furthermore, the bone–implant interfacial shear strength is directly related to the degree of surface roughness. However, for an optimal clinical performance of metallic implants in bone, both an ideal type/shape of the implant and the surface roughness degree still remain unknown. It is also important to mention that initial stability is also achieved by drilling the bone socket to a diameter slightly inferior to the implant's dimensions, but to the proper matching length of the implants. If the opening hole is drilled through the bone and enlarged progressively, and the implant is screwed in with low momentum, the surrounding bone is going to be compressed. This provides a firmer fit to the implant itself and it also increases the calcium concentration in the interface. This method decreases also the blood clot thickness around the implants. If growth factors are used, a boost in the healing process will be observed, while in their absence, the osteoconduction process only will occur.

A nowadays challenge in implantology is the capability of a surface to assist cells' colonization and differentiation. Cell migration, adhesion, and proliferation onto implant surfaces are key-processes for the initiation of tissue regeneration [91]. Osteoconduction and osseointegration represent two mechanisms that involve the adhesion or proliferation of cells and integration in the CaP [35,92]. The adhesion of cells is directly related to the CaP ability to adsorb the proteins from the extracellular matrix. For CaPs, this is strongly determined by important parameters, such as the surface roughness, crystallinity degree, Ca/P ratio, solubility, phase content, grain and particle sizes, and surface energy [92].

Osteoinduction is the property of a material to induce the differentiation of progenitor cells to osteoblasts [35,92]. It was demonstrated that CaPs in the absence of supplements are osteoinductive materials [92]. In turn, the osteoinduction ability depends on several CaP properties, i.e., surface charge, morphology, and chemistry, which can influence the adsorption of proteins [92].

The success rate of a medical implant is in relation to several factors such as the implant's design, the structure, and properties of the used material, the applied loads' magnitude, the employed surgical technique, health conditions of either animals, or human patients [82], and the existence of an inflammation-free environment. In the latter case, there exist two major problems: (i) the area intended for the implant insertion might be infected and (ii) the implant's surface might get contaminated during surgery. Both situations can be easily by-passed, by appropriate dental work and by a proper isolation during the surgical act, respectively. All aspects related to the biological interactions to improve the long-time stability and reliability of medical devices inside living systems have pushed forward the research of a wide range of surface modification techniques. The aim was to achieve rapid healing and an improved bone-implant interaction, for an early osseointegration. Therefore, in the last two decades, the surface functionalization of implants by increasing the bioactivity using various biological and chemical processes or by surface macro- and microtexturing was one of the major research topics in the biomedical domain [82].

#### *4.1. Used Substrates*

Currently, in comparison to polymers and ceramics, metallic biomaterials are extensively used in orthopaedics, dentistry, and oral and maxillofacial surgeries. This is due to several important advantages, such as the high mechanical strength, superior biocompatibility, high resistance to corrosion, and improved chemical stability under biological conditions [93–95]. Bioinert metallic implants (i.e., stainless steel (316L) and Ti, Ti-based, and cobalt-chromium alloys [96]) are generally utilized for a wide range of medical applications, but there is currently a new generation of biodegradable metals (i.e., Mg, Zn or Fe) that has been intensively employed for temporary applications [97,98]. Because they do

not chemically bond to bone tissue unless their surfaces are modified, the use of the latter ones is generally limited.

#### *4.2. Animal and Human Implant Models*

After the in vitro validation of biomaterials' surfaces, the use of laboratory in vivo models is a step forward biocompatibility assessment [99] and future clinical outcome of metallic implants introduced in bones. In this respect, in vivo testing on animal models is a key-parameter for both understanding and evaluation of the biological processes that occur in a living system.

The general animal spectrum used for in vivo testing of CaP-based coatings synthesized by the PLD technique only is limited to rats, rabbits, (mini) pigs, dogs, goats, and sheep [34], and each animal model having its own advantages and limitations [100]. Small animal models demonstrated some important advantages over larger ones, i.e., overall lower costs, osteogenic ability, the possibility to both shorten the implantation time-periods [101,102], and monitor bone formation by imaging methods (e.g., μCT). Thus, the possibility to carry out longer experimental times is worth considering. In this respect, rat or mice models are generally used for subcutaneous examination of implants [103,104], while rabbits represent the easiest way to investigate the interaction between the coating and the femoral bones [105–107]. It should be noted that the rabbit bone model was previously indicated [108,109] as a valuable screening tool able to select favorable implant surface characteristics/technologies before moving to the next step (i.e., human trials). It is currently used in various medical tests [110], due to both its size and ease of handling. In addition, it was demonstrated that its skeleton reaches maturity very fast (~24 weeks) [111]. In comparison, large animal models (i.e., dogs, sheep, or goats) are generally indicated for verifying the practicability of implants closer to real clinical situations. In such large animal models, the investigation of the osteoinduction process is easier to be attained than in the case of small ones. Large animals have the advantage of an immune system more similar to humans than in the case of small animals and can offer the possibility to host different types of test materials [112]. Despite this, such large animal models are not so intensively utilized for the evaluation of osteoinduction because of their lower metabolic rates, besides higher costs for management and maintenance [113].

Usually, implants are surgically inserted in either the femur, tibia, or mandible bones [114]. The rabbit's bones manifest faster changes as compared to the case of larger animal models [115]. Considering the difficulties met when extrapolating the results obtained on rabbit bones to the human case, various screenings for implant design and validation of the tested materials are still needed to be performed [75], before their testing on larger animal models. Dog models are generally used for testing dental implants [116]. One should emphasize upon that the focus of the most studies is directed to the biological response of the living bone to CaP-based materials.

For an easier follow-up of the text, Table 2 introduces information on the sample codes, which will be further used in the paper.

For an ease access to information presented in the papers considered in this review, Tables 3–6 introduce data on the investigated coating materials (along with the type and dimension of substrates, bone implantation sites, length of studies, and all performed analyses). All the information was grouped according to the applied small and large animal models, respectively (Table 3—rats/mice, Table 4—rabbits, Table 5—minipigs, and Table 6—dogs/sheep/humans).

It should be emphasized that because of the reduced number of in vivo studies on the PLD synthesis of CaP-based coatings, the use of a certain animal model as the most suitable one for the optimal assessment of metallic implants' osteoconduction has not been established yet. The reports on clinical trials (using human patients) are therefore scarce [71].


**Table 2.** Sample acronyms related to different materials used in the review and their explanation.

**Table 3.** Information related to the investigated coating materials, according to the applied small animal models (i.e., rats/mice).


**Table 4.** Information related to the investigated coating materials, according to the applied small animal models (i.e., rabbits).



**Table 5.** Information related to the investigated coating materials, according to the applied small animal models (i.e., mini-pigs).

**Table 6.** Information related to the investigated coating materials, according to the applied large animal models (i.e., dogs/sheep/human patients).


#### *4.3. Characterization Methods*

For the in vivo assessment of the CaP-based coatings considered in this review, the performed investigations, on which we will focus our attention, are standardized radiography, microradiography, (micro) computed tomography (μCT), tomodensitometry, histology (under polarized and fluorescent light), histomorphometry, planimetric analysis, fluorescent microscopy, scanning and transmission electron microscopies (SEM, TEM), and mechanical testing. While mechanical testing aims to determine the bonding strength between the newly formed bone tissue and the implant, histological investigations are used for a wide range of purposes, such as the measurement of the new bone area, bone apposition ratio, etc. [105–107,117]. It is important to note that light microscopy or optical microscopy is the most common laboratory technique used for biological investigations. It is a cheap, robust, and typically noninvasive method that uses visible light to detect and magnify small objects. The resolution limit is an intrinsic property due to the wavelength of visible light radiation [118]. Polarized light microscopy is a contrast-enhancing technique with a high degree of sensitivity that can be used for both qualitative and quantitative investigations. It can provide information on both absorption color and optical path boundaries and the structure and composition of materials that are invaluable for identification and diagnostic purposes. The technique of fluorescence microscopy has become an essential tool in biology and biomedical sciences, as well as in materials science, being capable of revealing the existence of single molecules. SEM and TEM microscopies enable the characterization of microstructures at many different length scales, from micro- to nanoscale, within an imaging session. They all generate a highly focused beam of electrons, which impacts the specimens inside a vacuum chamber. SEM microscopy is used to examine material surfaces, in comparison to TEM microscopy, which primarily focuses on investigation of the internal structure of the specimens [118].

#### 4.3.1. Standardized Radiography and Microradiography

Dostalova et al. [63] evaluated quantitatively and qualitatively the Ti implants by radiographical measurements. After 16 weeks of implantation, the osseointegration process was confirmed by the presence of newly formed bone around all implants, along with a strong bone–implant connection. When measured in the long-axis cross-section, the implant area

was in the range of ~25–28 mm for Ti implants and ~21–37 mm in the case of the HA-coated ones. This corresponded to an osseointegration area of ~18–22 mm for Ti implants and of ~15–30 mm for the HA-functionalized ones. The inferred percentage of osseointegration varied from 73% to 79% for Ti implants and from 70% to 86% for the HA-coated ones. No differences between various implants were found to be statistically important.

#### 4.3.2. Computed Tomography

Using μCT, Mroz et al. [72] performed qualitative and quantitative investigations on Mg:HA samples and Ti implants, and the obtained results revealed significantly higher differences between these groups (*p* = 0.0311).

In another study, Chen et al. [73] demonstrated by μCT that for both 4 and 8 weeks, the ratio of bone volume to total volume, mean trabecular number, and mean trabecular thickness were significantly higher (*p* < 0.05) in the case of FHA-coated implants, which was indicative for an accelerated osteogenesis process in the region of interest. Moreover, for both time periods, the mean trabecular separation was found to be significantly lower (*p* < 0.05).

Duta et al. [75] used CT scans, at 4 weeks after the surgical procedure, to observe the correct placement and the good integration of all implants into the surrounding bone. Thus, the presence of the peripheral osteosclerosis and no inflammatory process of the soft tissues were indicated. The obtained results pointed out to an increase in the osseous density, for both investigated time periods (4 and 9 weeks, respectively). Thus, the bone density values corresponding to the Li-C and Li-P coatings, measured at 9 weeks, were ~1.2 times higher than those inferred at 4 weeks after implantation. At 4 weeks, both Li-C and Li-P structures indicated bone density values ~1.3 times higher than those obtained for Ti implants. Moreover, at 9 weeks, the density values inferred in the case of functionalized 3D Ti implants were ~1.4 times higher as compared to Ti ones.

#### 4.3.3. Tomodensitometry

In a pioneering contribution, Duta et al. [71] used the tomodensitometry analysis to evidence the osteogenesis process of simple and functionalized Ti meshes implanted in human patients, at 3 and 6 months after surgery. After 3 months, no changes were detected for any of the investigated patients in terms of tissue density on Hounsfield (HU) scale. After 6 months, in the case of two patients with Ti mesh and for all with Ti meshes functionalized with bioactive surfaces, changes in measurements of osteoinductive and osseointegration phenomena were evidenced. In the first case, these changes were assessed only on the edge area, which was in direct contact with the bone. In the second case, four patients from the total of six, showed changes of tissue density, both on the edge and interjacent areas of the mesh. Up to 6 months, no patient evidenced those changes in the center region of the implants. Thus, the inferred values for the functionalized Ti meshes were of 583 and 412 HU (for the edge and interjacent region, respectively), in comparison to 78 HU, obtained in the case of Ti meshes.

#### 4.3.4. Histology

Antonov et al. [64] performed histological evaluations of their synthesized structures for three different time periods of 15, 30, and 60 days, respectively. It is important to note that no inflammation was observed around the sites of implantation for any of the investigated structures. However, in the case of Ti samples, a significant fibrous tissue formation was pointed out. For both types of structures (synthesized by either KrF or CO2 lasers), a thin fibrous layer could be observed after 15 days of implantation. At 30 days, a partially direct bone–implant contact was indicated, while very little fibrous tissue and the formation of new bone, which was in direct contact with the implant surface, were shown at 60 days. A very interesting observation was that the osteointegration rate was slightly superior in the case of annealed samples, rather than for the non-annealed

ones. Between the two types of synthesized coatings (using either KrF or CO2 lasers), no significant statistical differences in the osteogenesis process were inferred.

The histological evaluation under polarized and fluorescent light was used by Dostalova et al. [65] to indicate the presence of newly formed bone tissue around the investigated implants. No osteoclasts, macrophages, or any inflammatory reaction cells could be observed in the ground sections. In the case of synthesized coatings, a fibrous connective tissue occurred in ~23% of the implant surface, but without the formation of a continuous layer. In contrast, the fibrous connective tissue between the implant and the newly formed bone tissue occupied ~35% of the surface, especially in the middle part of the implant. However, it was indicated that these differences were not statistically significant (*p* = 0.05). Under fluorescent light, a uniform distribution of the fluorescent label in the whole bone, most probably due to a remodeling process of early formed bone, was observed. These results were supported by the higher magnification investigations, in which active bone cells were observed.

Using histological analyses, Peraire et al. [67] succeeded to demonstrate bone upgrowth in the endosteal areas, with a great amount of lacunae area in contact with them, while in the center, the bone apposition was indicated to be totally absent. The fiber mesh either disappeared in some implants or was very thin. In the case of the Ti rods functionalized with HA coatings by plasma spraying (HA–PS), a good bone regeneration at the ends of the implants (endosteal zones) was observed, but the newly formed bone tissue in contact with the implant surface evidenced a significant large amount of lacunae area. This active remodeling process might occur because of the coating degradation. HA–PS coatings partially disappeared in some areas of the implants, were delaminated or even detached, and HA particles could be therefore observed. In the central zone, bone apposition areas could be hardly seen, with a slightly thicker fiber mesh. Inflammatory cells (i.e., lymphocytes, macrophages, or neutrophils) were also present. In the case of HA coatings synthesized by PLD (HA–PLD), a good bone regeneration at both endosteal ends of the implant was indicated. All samples presented superior (grade 4) responses in bone reaction and interface analysis parameters. The newly formed bone tissue was similar to the cortical one, and the presence of mature osteocytes was detected. Under optical microscopy, HA–PLD coatings could not be evidenced because of their low thickness, while under polarized microscopy, the mineralized matrix in apposition to the implants could be observed for all investigated areas (cortical insertion area, opposite endosteal area, and bone marrow). The central area in contact with the bone marrow indicated bone apposition areas interspersed in a thin fiber mesh, with large trabeculae growing from the coating surface.

Hayami et al. [32] performed histological investigations of the interface between bone and the tip of each implant. After 4 weeks, large gaps at the Ti implant−bone interface could be observed, bone and connective tissues being intermingled. In the case of the sprayed implants, a thick adhering coating could be clearly seen, along the observable length of the interface. For the BHA/HA bilayered implants, no detectable gap was visible, and the bone and the implant closely adhering to each other along the full length of the interface. Thus, the process of biointegration was assumed to be complete after only 4 weeks of implantation. This strong bone−bilayered implants connection was also observed at 8 and 24 weeks post-operation. At 24 weeks after surgery, normal bony structures with osteocytes were indicated surrounding the implants.

Using histological analysis, Hontsu et al. [70] could observe that the HA-functionalized implants were surrounded by a mesh of thin fibrous tissue. Inside the implants, the mesh porosity was filled with fibrous tissue containing capillaries. Inside the HATW structure, a slight amount of new immature bone formation was indicated. This amount clearly increased from the second to the fourth week of implantation, for all types of scaffolds. At 4 and 8 weeks, more sites containing osteoblast-like cells and osteoid tissue expressing active bone formation could be observed, while at 12 weeks, the structure of the bone was further matured and contained larger areas of remodeled lamellar bone.

After observing the stained tissue sections, Wang et al. [52] indicated an obvious crack between the newly formed bone tissue and the 200 ◦C-synthesized film. This was mainly because of the preparation process of the tissue sections.

Although SEM results demonstrated good connections between the newly formed bone tissues and the surfaces of the two type of films (Figure 1a,b), the histological micrographs indicated a superior bone growth of the structures synthesized at 600 ◦C, in comparison to those prepared at 200 ◦C. The osteoblasts in the newly formed bone tissue were shown clear and homogenous, with a growth oriented along certain directions.

**Figure 1.** Morphologies of the implants after embedding for 1 month. (**a**,**b**) are the SEM morphologies of 200 and 600 ◦C films. (Reprinted with permission from [52]. Copyright 2013 Elsevier.)

It was therefore concluded that all tissue sections from the implants deposited at 600 ◦C were good, with a satisfactory connection between the film and newly formed bone tissue. In contrast, for the structures synthesized at 200 ◦C, a lower adherence between the newly formed bone tissue and the films was demonstrated. This was mainly influenced by the presence of a dark area between the film and the new bone tissue, which seemed to appear because of the high shearing stress during the cutting process.

Using the histopathological evaluation, Mroz et al. [72] showed that Ti implants had a tight adherence to the mineralized bone tissue, being in many cases within the fatty marrow. Small fields of connective tissue were also observed directly adjacent to the implant, along with a direct penetration of regular bone into the implants' pores. In the case of Mg:OCP samples, a direct integration of the bone and penetration into the pores and the structure of the implant were indicated. For the Mg:HA group, evidence of a direct bone−implant integration was shown, with no sign of osteoclastic or osteoblastic reaction around the implants.

Longitudinal sections were collected by Chen et al. [73] to assess the newly bone tissue formation around the implants. No adverse inflammatory reactions or gaps at the bone–implant interface were observed. At 8 weeks after surgery, the bone area ratio and bone–implant contact were significantly higher (~6 and ~1.5 times, respectively) around FHA-coated implants in comparison to Ti ones.

Wang et al. [74] performed histological investigations to assess the osteoconduction capacity of HA/BG composite coatings, with three different BG concentrations. Thus, the 90%HA + 10%BG structure was shown to connect well with the bone tissue. However, the bone matrix filled the implant−bone tissue interspace, and osteoblast cells (OB) were not observed in this region before 2 months after implantation. In contrast, the 80%HA + 20%BG film exhibited better osteoconduction behavior as new OB could be observed near the implant surface in the first month after surgery. It is important to note that the improved biological activity of the film was indicated to be mainly due to their *c*-axis orientation. Macroscopically, a good connection with the bone tissue was also observed in the case of the 20%HA + 80%BG structure. In this case, a crack was indicated between the film and the bone tissue, which corresponded to a poor bond state (mainly because of the shearing force coming from the inner circle saw, used to cut the specimen into slices).

#### 4.3.5. Histomorphometry

Dostalova et al. [65] inferred the area of the bone–implant interface and found the values to vary from ~65%, for Ti implants, to ~78%, for HA-functionalized ones. It is important to note that between both types of surfaces, no significant statistical differences were inferred (Student's *t*-test, *p* = 0.05). As a consequence, a similar osseointegration trend for both Ti implants and HA-functionalized ones was indicated.

In the case of the implants inserted by Kim et al. [66] in the femur, a value of ~48% of the mineralized bone at the bone–coating interface for the synthesized HA coatings and of ~63% for the HA/TTCP biphasic ones, was inferred. For the case of tibia-implanted samples, values of ~51% and ~56% for the mineralized bone were indicated in the case of HA and HA/TTCP biphasic coatings, respectively. Very important, all samples were shown to exhibit a good integration, with no significant foreign body response.

Among the three materials evaluated by Peraire et al. [67], there were no statistically significant differences in bone in-growth around the drilling hole. However, the HA–PLD implants presented the highest value of the percentage of bone contact (86%), and the lowest value of percentage of lacunae contact (14%). This difference was statistically significant (Scheffe Test, *p* < 0.05,). When referring to the ratio between the total bone surface and length of the evaluated area, the HA–PLD implants showed a significant increase in comparison to the HA–PS group (Scheffe test, *p* < 0.05). This did not apply as well to the Ti group. In the case of the HA–PS group, even though the bone response was slightly different than the one corresponding to the Ti, no statistically significant differences were found, in any of the quantified variables.

Paz et al. [69] used in their study two methods to evaluate the percentage of bone– implant contact (% BIC): conventional light transmission microscopy and environmental scanning electron microscopy (ESEM). The obtained results demonstrated no statistically significant differences between these two methods, although ESEM is believed to offer greater precision when characterizing bone areas in close contact with implant surfaces. ESEM analysis showed a considerable improvement of the bioactivity in the case of HAcoated samples, the bone–implant interface being more homogeneous and continuous in comparison to Ti implants. Moreover, the bone was observed to enter deeper into the craters of the macrostructure for the HA-coated samples than for the Ti implants. In areas with low bone density, the HA-coated structures presented a superior behavior in contrast to Ti. ESEM analyses showed an improvement of the percent of total BIC for both HAsynthesized coatings (50 and 100 nm-thick samples, respectively) with regard to Ti implants. There were no significant differences between the two coated samples, although the percent apical BIC indicated better characteristics of the thicker coating over the thinner one. The conventional light transmission microscopy images showed similar results as ESEM, but with a modification in the percent total BIC; for this investigation, the thinner HA coating showed superior behavior in comparison to the thicker one, while for the percent apical BIC, the thicker coating presented a better behavior, but with a great increase in the standard deviation for the thinner one.

#### 4.3.6. Planimetric Analysis

The results of planimetric investigations performed by Mroz et al. [72] revealed the best bone integration in the case of Mg:HA samples, this group having the highest average length of the bone–implant interface and also the best reproducibility of the results. No significant differences were inferred between the Mg-doped (OCP and HA) implants and the Ti ones.

#### 4.3.7. Fluorescent Microscopy

When fluorescent microscopy was used by Dostalova et al. [63], the active bone formation could be observed, both in the neck and in the bottom of the implants, surrounding osteons and braiding the fibrous connective tissue on the implant.

Chen et al. [73] investigated by fluorescence microscopy the bone formation around the implanted structures. As determined by confocal laser scanning microscopy, the bone area between the implant surface and the boundaries observed at 1, 4, and 8 weeks, respectively, was significantly higher for FHA-coated implants as compared to Ti ones (*p* < 0.05). These implants indicated 1.5 and more than 4 times more new bone tissue formation at 1 week and 4 and 8 weeks, respectively, as compared with the case of Ti implants.

#### 4.3.8. Scanning Electron Microscopy

Antonov et al. [64] used SEM to investigate the in vivo behavior of the synthesized coatings. New bone formation was therefore observed, which surrounded implantation site of the HA-coated alloy samples.

SEM observations performed by Peraire et al. [67] allowed for the detection of the HA–PLD coatings on the grit-blasted Ti surfaces, at 24 weeks after surgery. The images obtained using backscattered electrons showed different contrasts for the bone and the coating, which allowed for their clear differentiation.

Hayami et al. [32] investigated by SEM the holes from the removed implants. Because the threads of the screw implants had a long pitch, the troughs left in the bone were trapezoidal in shape. For Ti implants, bone growth was slow and insufficient because of the imperfect trapezium formation, in comparison to the BHA/HA-bilayered implants, where the growth rate and quantity of newly formed bone tissue were greater due to the formation of regular trapezia. In the *V*-shaped groove of the thermal-sprayed HA coatings, a flaky piece was indicated, which appeared to be a fragment that peeled away from the substrate.

Wang et al. [52] investigated the morphologies of the implants at 1 month after implantation. These are shown in Figure 1a,b.

A layer of newly formed bone tissue was clearly seen, developing along both types of structures (synthesized at 200 and 600 ◦C, respectively). This layer intimately adhered to the implants' surface and had a thickness of ~5 μm. After 3 months of implantation, the newly formed bone tissue occupied the bone−implant interspace (Figure 2a,b).

**Figure 2.** Morphologies of the implants after embedding for 3 months. (**a**,**b**) are the SEM morphologies of 200 and 600 ◦C films. (Reprinted with permission from [52]. Copyright 2013 Elsevier.)

A good bioactivity of the films was therefore indicated, the two types of implants being capable of inducing new bone growth on their surfaces.

Using SEM (Figure 3), coupled with the analysis of backscattering electrons, Duta et al. [75] managed to infer the adherence ratio of the remaining bone fragments onto the surface of the extracted implants.

**Figure 3.** SEM micrographs indicating bone detachment on the surface of a control and functionalized (with Li-C and Li-P coatings) Ti implant, at 4 weeks after surgery.

Thus, adherence ratios up to ~38% higher in the case of functionalized 3D Ti implants (with Li-C and Li-P coatings) as compared to the Ti ones were indicated. This result, corroborated with the higher values of the detachment force obtained in the case of functionalized 3D Ti implants, in comparison to Ti ones, were indicative for an enhanced osseointegration process. Moreover, the presence of such osseous structures onto the surface of the implants suggested, besides the beginning of the implant integration process into the bone, the absence of any adverse reactions at the implantation site.

#### 4.3.9. Transmission Electron Microscopy

Using TEM, Dostalova et al. [63] have shown no marks of irritation and inflammation in the surrounded bone. OB could be seen both in the border of the implants and newly formed bone tissue. A low number of foreign body cells were indicated in the close vicinity of the implant cover. In the case of Ti implants, a fibrous connective tissue between the implants and the newly formed bone was seen, while in the case of the synthesized HA coatings, this layer could be observed only seldom.

#### 4.3.10. Mechanical Testing

The pullout (tensile) test was used by Mihailescu et al. [68] to compare the strength of the biological–chemical bonding between the bone and Ti implant surfaces functionalized with HA, Mn-CHA, and OCP coatings, respectively. The inferred values for all three tested groups demonstrated a significantly improved bone attachment strength value (*p* ≤ 0.05), which was about twice as high as the one associated to the Ti implants (~5 N). In comparison to the strength value corresponding to the synthetic HA (~8 N), up to 10% (in the case of OCP, ~9 N) and 25% (in the case of Mn-CHA, ~11 N) higher values were obtained.

Duta et al. [75] evaluated the quality of the implants' osseointegration by mechanical (tensile) tests. In none of the cases, alteration or disruption of the implants were present. The detachment force (Fmax) of implants under tensile pull-out testing, inferred for Ti and functionalized (with Li-C and Li-P coatings) 3D Ti implants, at 4 and 9 weeks after surgery, are represented in Figures 4 and 5, respectively.

**Figure 4.** Detachment force, Fmax, of implants (n = 10) under tensile pull-out testing, inferred in the case of control 3D Ti implants (marked in blue color) and of those functionalized with (**a**) Li-C (marked in green color) and (**b**) Li-P (marked in orange color) coatings, at 4 weeks after surgery. \*\*\*\* Represents highly significant differences (*p* ≤ 0.0001). \*\* Represents significant differences (*p* ≤ 0.01).

At 4 weeks after surgery (Figure 4), the obtained mean detachment force values demonstrated significant and highly significant differences between the Ti group and the Li-P one, and between the Ti group and the Li-C one, respectively. When referring to the extractions performed at longer periods of time, i.e., 9 weeks from implantation (Figure 5), the inferred mean detachment force values indicated highly significant differences for both investigated groups.

It is important to mention that the failure loads of the implants functionalized with both Li-C and Li-P coatings measured at 9 weeks were ~5 times higher in comparison to those inferred at 4 weeks after surgery, respectively. Moreover, for both time periods, the Li-C and Li-P functionalized implants demonstrated a bone attachment strength of ~2 times stronger than the one corresponding to Ti implants. One could therefore indicate that both the PLD surface functionalization of the implants and a longer implantation time

period could induce a positive influence on the overall bone bonding strength characteristics of the investigated medical devices. It should be emphasized that the fabrication of novel BioHA implant coatings derived from sustainable and inexpensive CaP resources, with improved mechanical properties, correlated with an increased bone fixation in vivo, could stand for a pioneering contribution to the progress of advanced medical devices.

**Figure 5.** Detachment force, Fmax, of implants (n = 3) under tensile pull-out testing, inferred in the case of control 3D Ti implants (marked in blue color) and of those functionalized with (**a**) Li-C (marked in green color) and (**b**) Li-P (marked in orange color) coatings, at 9 weeks after surgery. \*\*\*\*\* Represents highly significant differences (*p* ≤ 0.00001).

#### **5. Discussion**

Because the biomaterials' osteoinduction mechanism is not yet entirely understood, one could not precisely answer the question whether if the sole biomaterial or an interaction between the biomaterial and the relevant proteins present in the living system are responsible for the osteoinduction process. Because most of the implants do not possess the capability to induce bone growth, specific material properties are required to activate the osteoinduction process. To begin the differentiation of the undifferentiated inducible osteoprogenitor cells into bone-forming cells, it was suggested that both the chemistry and the geometry of the biomaterial in contact with these cells represent critical factors to be considered [119].

Metallic implants (including Ti) are generally used for various biomedical applications, mainly due to their resistance to corrosion and favorable mechanical characteristics [120]. Because of its bioinert nature, bulk Ti is not capable to form a biochemical bond with the bone, and this biological inactivity often generates a fibrous tissue that surrounds the implanted device [121]. To improve both osseointegration rates and longevity of Ti implants, the deposition of CaP-based coatings onto their surfaces is envisaged. It was therefore demonstrated that implants' surface functionalization with CaP-based coatings could promote the formation of real bonds with the surrounding bone, due to their proved chemical similarity with natural bone tissue and their high biocompatibility [122]. This process occurs rapidly along the entire surface of the coating, in comparison to the case of simple Ti implants (used as controls in the experiments) [47].

A nowadays growing research interest in the field of biomaterials is related to the use of biological-derived CaP materials as viable, safe, and low-cost alternatives to synthetic CaP-ones [33]. It should be emphasized that, unfortunately, the Earth's available mineral resources are threatened to become limited in the near future because of the rapid demographic increase and economic growth. The access to sustainable resources is therefore critical. Consequently, this will generate a beneficial economic and environmental impact over the society, allowing for an intelligent use of these renewable resources.

The mechanical properties of CaP-based coatings are responsible with the overall success rate of an implant [123]. Thus, the optimal functioning of an endosseous implant is directly influenced by the biomaterial's mechanical stability, which can be easily evaluated by extraction tests. To obtain information on the force that occurs between the bone tissue and implanted materials, various experimental study models have been developed, each of them with their own particularities [115,124,125]. In this respect, the investigation of the coating bond strength is typically performed by scratch [126–128], pull-off [129], tensile adhesion [130–132], or shear strength tests [133], respectively. It is important to emphasize upon that the ISO 13779-2:2008 standard requirement for tensile adhesion strength of CaP-based coatings, used for load-bearing applications, is of 15 MPa [134]. It was reported that, in general, CaP-based coatings synthesized by the PLD technique easily surpass this imposed value [33,59,135]. There are some studies in the literature concentrated on tensile strength measurements [136,137], which, very importantly, can provide a direct measurement of the attachment between the bone and the implant surface, being therefore influenced only by the chemical bonding between those two [138–140]. This way, the effects of friction and of mechanical forces introduced by surface roughness can be minimized [141]. In the case of animal trials, the implant's increased bone retention is considered a clinically relevant indicator of improved stability and capacity of the implants to carry loads without detaching. Unfortunately, this type of information cannot be acquired by histological or SEM investigations, which provide only limited information on the functional performance of an implant. One should note that in none of the studies included in this review, related to mechanical testing of CaP-based coatings, alteration or disruption of the implants were present. In general, the inferred values for all functionalized Ti implants demonstrated significantly improved bone attachment in comparison to Ti ones. Next to the PLD surface functionalization of metallic implants, a longer implantation time period was demonstrated to induce a positive influence on the overall bone bonding strength characteristics of the investigated medical devices. One should note that the fabrication by PLD of novel BioHA implant coatings derived from sustainable and inexpensive CaP-based resources, with improved mechanical properties, correlated with an increased bone fixation in vivo, could stand for a pioneering contribution to the progress of advanced medical devices.

In general, the results of the studies included in this review, obtained using standardized radiography and microradiography [63], computed tomography [72,73,75], histomorphometry [65–67,69], and tomodensitometry [71], have confirmed the osseointegration process (pointing to an increase in the osseous density), along with a strong bone–implant connection and no inflammatory process of the soft tissues. The histological investigations, performed with various microscopy techniques [32,52,63–65,67,70,72–75], have indicated

also new active bone formation and demonstrated no adverse inflammatory reactions or gaps around the sites of implantation or at the bone–implant interface, for any of the investigated structures. In addition, the bone and the implant were shown to tightly adhere to each other along the full length of the interface. One interesting observation was that the temperature applied during the deposition process seemed to play an important role in the bone growth of the synthesized structures, the osteoblasts in the newly formed bone tissue being shown clear and homogenous [52]. Moreover, the osteointegration rate was demonstrated to be slightly superior in the case of annealed samples, rather than for the non-annealed ones [64]. It seemed also that no matter what the laser source used (i.e., KrF or CO2), between the synthesized coatings, no significant statistical differences in the osteogenesis process were inferred [64]. Significantly higher values of the bone area between the implant surface and the boundaries and bone adherence ratios were inferred at various implantation time periods in the case of functionalized implants in comparison to control ones [73,75]. In this respect, in the case of control samples, a fibrous connective tissue between the metallic implants and the newly formed bone was shown, while in the case of the synthesized coatings, this layer could be observed only seldom [63].

Even though it is generally accepted that CaP-based coatings deposited by PLD improve bone strength and the initial osseointegration rate, the coatings' properties necessary to achieve an optimum bone response are yet to be determined. This is mainly because of the limited number of in vivo studies available in the dedicated literature. It should be emphasized upon that the in vivo testing should demonstrate stability in biological environment for up to 1 month, which corresponds to the initial healing phase of the implants [142]. The limitations on such experiments can be related to (i) the difficulty to select a suitable animal model in order to properly simulate the actual mechanical loading and unloading conditions in which an implant should function inside a living system; (ii) the need to sacrifice a large number of animals to reach a significant statistical relevance, able to validate the obtained results; (iii) the demands for high costs and long time frame in the case of clinical trials; (iv) the lack of coordination among material scientists and biologists and thus an insufficient understanding of this interdisciplinary subject; and (v) the serious ethical concerns related to the used animals (including also the choice of their correct number), as they might be sometimes subjected to painful procedures or toxic exposures during the experimental trials [143].

Taking into consideration all these aspects, a future important progress of CaP-based materials might be linked to a shift of the focus from osteoconduction to osteoinduction, e.g., by additive manufacturing of scaffolds with complex, controlled three-dimensional porous structures and development of novel ion-substituted CaPs with increased biological activity. Moreover, new strategies, possibly based on self-assembling and/or nanofabrication might be developed for the successful fabrication of load-bearing bone graft substitutes. In the future, the composition, microstructure, and molecular surface chemistry of various types of CaPs might be tailored in such a way to match the specific biological and metabolic requirements of tissues or disease states. The multilayer composite coating systems, fabricated by the PLD technique, should represent also a future trend, able to provide multifunctional properties for the biomedical implants.

#### **6. Conclusions**

This review summarizes a two decades achievements reported in the field of in vivo assessment of calcium phosphate (CaP)-based coatings deposited onto metallic implants by one of the most frequently used plasma-assisted techniques, i.e., pulsed laser deposition (PLD). Due to their proven biocompatibility, mechanical (high adherence), and osseointegration and osteoconduction properties, CaP-based bioceramics are widely used in the field of bone regeneration, both in orthopedics and dentistry. For an in-depth in vivo assessment of various CaP-based coatings synthesized by the PLD method only, the results of the studies included in this review were obtained using a wide range of investigation techniques, among which a special focus was put on standardized and (micro) radiographies, (micro) computed tomography and histomorphometry, tomodensitometry, histology, scanning and transmission electron microscopies, and mechanical testing. It is important to note that all these results indicate superior osseointegration characteristics of the metallic (generally Ti) implants functionalized with CaP-based coatings when compared to simple (uncoated) Ti ones, which are considered as the "gold standard" for implantological applications.

In the last two decades, research studies performed on CaP-coated metallic implants by PLD resulted in an interesting progress in vitro and in vivo, while not enough comparable clinical results were delivered so far for an easier assessment. This was mainly because of the lack of standardization of the coating properties and in vivo models. Therefore, additional testing is still needed in this direction, both to be able to advance a certain "recipe" to obtain optimum in vivo results, and to further reveal the relative influence of implant design, surgical procedure, and coating characteristics (thickness, structure, porosity, and surface morphology, which includes the wettability behavior), on either short-term or long-term clinical beneficial effects of the CaP-based coatings. In addition, one should emphasize upon the growing interest on the biological-derived CaP-materials as viable, safe, and cheap alternatives to the CaP synthetic ones, along with their improved biological properties and a greater resemblance to the mineral part of the human bones. The domain of PLD synthesis of natural-CaP sustainable coatings is in its first stages of development, and therefore, various possibilities to expand in the near future in terms of natural-CaP new resources, different concentrations of doping agents, or morphology and structural control of the obtained coatings are envisaged.

**Funding:** L.D. acknowledges the financial support of the Romanian Ministry of Education and Research under Romanian National Nucleu Program LAPLAS VI—contract No. 16N/2019.

**Institutional Review Board Statement:** The animal surgical protocols complied with the regulations and precautions of ISO-10993-Part 2 and Part 6, National Animal Care Guidelines and EU Council Directive of 22 September 2010, regarding the care and use of laboratory animals for scientific purposes (2010/63/EU), and were approved by the Institutional Animal-Care Committee, the Local Ethical Committee for the Affairs of Experiments on Animals in Lodz (22.12.2008, decree No. 56/ŁB 440/2008), the Osaka Dental University Ethics Committee, Japan (approval No. 16-08002, 2 August 2016), and the "Committee of Ethics and Academic and Scientific Deontology" at the UMF in Craiova, Romania (document No. 135/20.12.2019).

**Informed Consent Statement:** Informed consent was obtained from all subjects involved in the study.

**Data Availability Statement:** The data presented in this review are available in Refs. [32,52,63–75].

**Acknowledgments:** L.D. acknowledges F.N. Oktar (Marmara University, Istanbul) for providing animal-origin powders that were used for a part of the studies reported in this review. L.D. thanks George E. Stan for his useful comments and kind help.

**Conflicts of Interest:** The author declares 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

*Coatings* Editorial Office E-mail: coatings@mdpi.com www.mdpi.com/journal/coatings

MDPI St. Alban-Anlage 66 4052 Basel Switzerland

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