**4.2 Preparation of single crystals**

Throughout the course of evolution of hydrothermal synthesis from the geoscientific applica‐ tions to modern technologies, the hydrothermal technique has captured the attention of scientists and technologists from different branches of science. The hydrothermal technique is popularly used by geologists, biologists, physicists, chemists, ceramists, hydro-metallurgists, materials scientists, engineers, etc. **Fig. 13** shows different branches of science either emerg‐ ing out from the hydrothermal technique or closely linked up with the hydrothermal techni‐ que. One could firmly say that this family tree will keep expanding its branches and roots in

The hydrothermal techniques for the preparation of compounds with the structure of apatite

The hydrothermal synthesis of all three normal apatite end-members was reported by BAUMER and ARGIOLAS [80]. They prepared crystallites of sizes from 50 to 500 μm. The synthesis of

The synthesis and the stability of carbonate-fluorapatite were examined by JAHNKE [81]. The carbonate-apatite phase is stable in solutions relatively rich in carbonate such as sea-water.

During the hydrothermal synthesis of HAP whisker, the acetamide was used by ZHANG and DARVELL [83] as an agent to drive homogeneous precipitation at temperatures below 100°C. Acetamide shows low hydrolysis rate in both acidic and basic conditions, releasing acetate and

which do no substitute in HAP lattice. The precipitation of hydroxylapatite from the solu‐ tion of Ca(NO3)2·4H2O and (NH4)2HPO4 in 0.05 mol·dm−3 (Ca:P = 1.67) treated to the tempera‐ ture of 180°C for 10–15 hours yielded to large rod-like and well-crystallized particles of

The stoichiometric single crystals of hydroxylapatite nanorods with mono-dispersion and narrow-size distribution in diameter were successfully synthesized by LIN et al [84] via the

the surfactant and *n*-pentanol as the cosurfactant. First, 0.5 M Ca(NO3)2 solutions and 0.3 M (NH4)2HPO4 solutions were obtained by dissolving Ca(NO3)2·4H2O and (NH4)2HPO4 in

40 The emulsification consists in dispersing of one fluid in another, non-miscible one, via the creation of interface [85].

When exposed to low-carbonate solutions, the carbonate-apatite should lose the CO3

CH CONH H O CH CO H NH 3 2 3 32 4

2 3 4 10 4 2 ( )<sup>6</sup> 10 CaCl 6 H PO Ca PO Cl 18 HCl +Û + (21)

+ + +® + (22)

The microemulsion was prepared using CTAB as

2− ion [82].

chlorapatite at 400°C and the pressure <3 kbar proceeds via the reaction:

the years to come [21].

should be divided to:

ammonia:

hydroxylapatite.

hydrothermal microemulsion method [85].40

**1.** low-temperature hydrothermal synthesis (LHS);

198 Apatites and their Synthetic Analogues - Synthesis, Structure, Properties and Applications

**2.** high-temperature hydrothermal synthesis.

The first technique that was used for the production of crystals **Fig. 14**(**a**) was described by VERNEUIL [90],[91],[92],[93] at the turn of the 20th century,42 but there is an evidence that the socalled Geneva ruby had been grown by a similar technique almost 20 years earlier. The second technique for single crystal growth was introduced by CZOCHRALSKY [94] few years later, who needed materials with small dimensions43 in order to study the growth kinetics of metal (**Fig. 14**(**b**)). This technique is based on pulling thin wires from the melt at various speeds and obtaining single crystals. Beginning in 1950s, it eventually developed into the complex technology required in order to obtain large-diameter prefect crystals which are a raw material for the electronics industry, but controlling the dimension of the crystal was very difficult. The idea of using a shaping device floating on the melt surface to stabilize the crystal growth was introduced by GOMPERZ [95], who used a drilled mica plate. Since that time, numerous types of shaping devices have been used to get crystals of various shapes [26],[87],[88].

Kyropoulos developed the melt growth techniques (**Fig. 15**) for growing large crystal from the melt using a cooled seed in 1926 [96],[97],[98]. The method was demonstrated via the produc‐ tion of large single crystals of alkali halides [99].

<sup>41</sup> The name surfactant is a contraction of the term: **surf**ace-**act**ive-**a**ge**nt**. It can be defined as the substance that, even if present at low concentration, has ability to be adsorbed onto the surface or interface of the system and significantly alter (usually decrease) its surface or interface free energy. While the term **surface** usually means the interface between condensed phase and gas, the interface is considered as a boundary between two immiscible phases. The molecular structure of surfactant contains lyophobic group (little attraction for solvent) and lyophilic group (strong attraction for solvent), i.e. amphipathic structure [86].

<sup>42</sup>VERNEUIL in fact wished to study the properties of ruby and other alumina-based crystals and was aware of very high melting temperatures of these materials, which prevented the use of any crucible material known in that time. This problem was solved- by melting alumina powder in a hydrogen-oxygen flame and solidifying the droplets on a colder seed. Nowadays this technique is used for the production of single crystals of sapphire (single crystal of Al2O3 in **Chapter 3** (**Fig. 14**) was prepared by this method) and spinel with only little changes [87]. The crystal grows from the melt film, which thickness is defined by the crystal diameter and the thermal conditions at the crystallization front [93].

The scheme of Verneuil's growth unit [87],[90],[93]: electromagnet (A, or camshaft) operating the hammer (M), supply chamber of fine Al2O3 powder (P), feeder (C, D), oxygen (O) and hydrogen (H) inlet, growing crystal (R), crystal holder (S) and device for the crystal adjustment (V).

<sup>43</sup> Small dimension is necessary to dissipate the latent heat of solidification efficiently and rapidly [87].

**Fig. 14.** Scheme of Verneuil's method42 (a) [87],[90], Czochralsky growth apparatus (b) [88] and μ-PD apparatus (c) [89].

**Fig. 15.** Schematic illustration of Kyropoulos method [96] and three stages of Kyropoulos method [98].

After important growth processes based on capillarity, historically the next development was the BRIDGMAN method [100], aiming at increasing the crystal size and consisting in growing the crystal in crucible. The next method to be invented in 1952 by PHANN [101] was the floating zone (FZ) technique.44 This method is capillary-based technique that was originally devel‐ oped for the material purification [87]. The schematic representation of convection in the molten zone according to HIGUCHI et al [102] is shown in **Fig. 16**. The Marangoni convection45 in molten zone leads to the formation of tiny bubbles, which are not arranged randomly, but form a ring inside the crystal.

<sup>44</sup> The floating zone is generated by means of water-cooled induction coil fed by radio frequency power in the megahertz range [103].

<sup>45</sup> Marangoni convection, which is caused by the differences in the surface tension over the melt surface, flows along the interface from the surface to a central region of the melt. On the other hand, forced convection, which is caused by the crystal rotation, flows towards the periphery from the center [102].

**Fig. 16.** Schematic representation of convection in the molten zone [102].

Pulling wire Seed **(b)**

Crucible shaft Ar gas + SiO

**Fig. 14.** Scheme of Verneuil's method42 (a) [87],[90], Czochralsky growth apparatus (b) [88] and μ-PD apparatus (c) [89].

**I.**

After important growth processes based on capillarity, historically the next development was the BRIDGMAN method [100], aiming at increasing the crystal size and consisting in growing the crystal in crucible. The next method to be invented in 1952 by PHANN [101] was the floating

oped for the material purification [87]. The schematic representation of convection in the molten zone according to HIGUCHI et al [102] is shown in **Fig. 16**. The Marangoni convection45 in molten zone leads to the formation of tiny bubbles, which are not arranged randomly, but

44 The floating zone is generated by means of water-cooled induction coil fed by radio frequency power in the megahertz

<sup>45</sup> Marangoni convection, which is caused by the differences in the surface tension over the melt surface, flows along the interface from the surface to a central region of the melt. On the other hand, forced convection, which is caused by the

This method is capillary-based technique that was originally devel‐

**Fig. 15.** Schematic illustration of Kyropoulos method [96] and three stages of Kyropoulos method [98].

Magnetic field coil

Heater Silicon melt

Optical system

Gate valve Ar gas

Main heater Grown

crystal Melt Pt crucible

Afterheater Seed

**II. III.**

**III.**

Wire reeling and rotation

Neck View port Silicon crystal Crucible Crucible holder Shield

200 Apatites and their Synthetic Analogues - Synthesis, Structure, Properties and Applications

**(a)** Crystal

**(a) (b)**

zone (FZ) technique.44

range [103].

form a ring inside the crystal.

crystal rotation, flows towards the periphery from the center [102].

**(c)**

**(**

Afterhe

Micro nuzzle

cru nu

> Since then various modifications of these basic methods have been proposed, such as the pedestal growth, edge-defined film-fed growth (EFG) process, inverted EFG process, micropulling down (μ-PD, **Fig. 14**(**c**)), etc., all based on the use of capillary force in order to maintain and shape the liquid. **Fig. 17** show the classification of these methods based on the presence or absence of the crucible or shaping die in contact with molten material and on the direc‐ tion of pulling [87],[89],[103],[104].

> Whiskers can be described as long filamentary defect-free single crystals of great mechanical strength, which is attributed to their high structural perfection. The explanation of whisker growth is based on the screw dislocation theory. The dislocation appears only along the whisker axis, while in another two dimensions the faces will stay perfect. Consequently, no growth will occur at an appreciable rate on the side faces of whisker. Due to the presence of axial screw dislocation the whisker grows only along its axis [105]. Apatite whiskers are usually prepared by hydrothermal synthesis [83],[106],[107],[108], molten salt method [109] and also via the precipitation method [110].

> Dendrites46 are normally single crystals, and the branches follow definite crystallographic orientation. The branches are regularly oriented and the opposite sides of the primary stem show marked symmetry. The growth of dendritic crystal is controlled by the diffusion of latent heat from growing crystal-melt interface [110],[111]. The dendritic growth of apatite crystal is described in glass ceramics [111],[112] and the formation of comb-shaped acicular and dendritic apatite was also observed as a product of quenching of trapped phosphate melt inclusions [113].

<sup>46</sup> The name was derived from the Greek word "tree like".

**Fig. 17.** Classification of various crystal growth processes using capillary forces for maintaining or shaping the molten material [87].

#### **4.2.1 Fluorapatite**

Single crystals of fluorapatite up to 5 cm long and of 1 cm maximum diameter were first prepared via the Kyropoulos method (pulling the crystal from the melt) by JOHNSON [114]. The Czochralsky method was used by MAZELSKI et al [114],[115] to grow the crystals up to 30 cm long. The ratio of CaF2 to Ca3(PO4)2 as determined by chemical analysis of crystals depends upon the value of the same ration in the melt. Even if the melt had correct stoichiometric composition, grown fluorapatite would appear to have the deficiency of CaF2 of about 5%. Fluorapatite as well as chlorapatite crystal with the length from 5 to 6 mm were grown by PRENER [116] from the solutions of apatite in molten calcium fluoride and chloride, respective‐ ly. The analyses of these flux-grown crystals agreed with theoretical values within 0.1% [114], [117].
