Electrochemical deposition of nickel targets from aqueous electrolytes for medical radioisotope production in accelerators: a review

This paper reviews reported methods of the electrochemical deposition of nickel layers which are used as target materials for accelerator production of medical radioisotopes. The review focuses on the electrodeposition carried out from aqueous electrolytes. It describes the main challenges related to the preparation of suitable Ni target layers, such as work with limited amounts of expensive isotopically enriched nickel; electrodeposition of sufficiently thick, smooth and free of cracks layers; and recovery of unreacted Ni isotopes from the irradiated targets and from used electrolytic baths.


Introduction
Nuclear medicine is a fast-growing area of the medical sciences which has found numerous applications in modern laboratories and medical institutions. Its development strongly depends on availability of radioactive isotopes suitable for the medical applications. There are several methods of production of the latter, including nuclear reactions between target nuclei and high-energy charged particles acting as projectiles in medical accelerators [1][2][3][4][5]. Nickel isotopes (Table 1) are often applied as the target nuclei and, depending on the isotope and the projectile type and energy, their reactions produce radioisotopes of copper, cobalt and nickel which have medical importance. Some of these radionuclides are frequently applied in today's nuclear medicine while the others are considered for the future applications (Table 2). Thus, 64 Ni(p,n) reaction [2-4, 17-19, 25, 30-44, 62-64] is the main method of production of 64 Cu, an isotope frequently used in positron emission tomography (PET) studies [5,18,20,30,64,65] and also applied in internal radiotherapy [20,44,66,67]. Another method of 64 Cu production using Ni targets is based on 64 Ni(d,2n) reaction [16,20,34,40,58]. Other PET radioisotopes which can be produced by means of nuclear reactions of nickel nuclei include 60 Cu, 61 Cu, 62 Cu,55 Co and 57 Ni [5, 9-14, 16-22, 24-26, 28-30, 34, 35, 41, 45, 61, 65, 66, 68-70]. 64 Ni(α,p) is one of the reactions which can be used for production of therapeutic 67 Cu radioisotope [15,22,23,44,60]. Nuclear reactions of nickel can also produce 57 Co and 58 Co which are radiolabelling agents used in radiotracer medical studies (e.g. Schilling test [71,72]) [5,10,11,16,41,69]. Finally, 64 Ni(p,α) reaction produces 61 Co that was suggested for use in medical molecular imaging [17]. The reactions of Ni nuclei with protons, deuterons or α particles are the primary or secondary methods of production of the medical radioisotopes mentioned above. Even if not considered as the primary method of the radioisotope production, the nuclear reactions of Ni may constitute an attractive alternative to the main production route, depending on the particle accelerator capabilities (projectile type and energy) and the isotope (target) availability. Except for the 57 Ni case, the reactions listed in Table 2 allow production of carrier-free medical radioisotopes, i.e. free from other isotopes of the same element [23].
Preparation of Ni targets suitable for the use in the particle accelerators is one of the crucial steps in the process of production of the radioisotopes in question. Electrochemical deposition on a metallic substrate from an aqueous bath is the method most often applied for preparation of the nickel targets [4,73]. This procedure offers several advantages over other methods of the accelerator targets production. It is relatively simple and does not require sophisticated and expensive equipment. The method is highly efficient [74] and the process can be conveniently controlled by applying required current density or cathodic potential. Electrodeposited layers do not require further post-plating processing [75]. Electrochemical plating allows working with minute amounts of the material and produces deposits with a wide range of thicknesses, although the maximum available thickness is often limited by peeling off of the deposit [73,74]. Other than the latter disadvantages of the electrochemical deposition include possibility of contamination of the target with impurities from the electrolyte bath which are co-deposited or coprecipitated with the deposited metal [74]. Further on, the electrochemically plated deposits may have non-uniform thickness [31, 46,74] and when the deposition is carried out from a solution with low metal content, the process can be time consuming (e.g. [2,8,10,13,17,19,45,48,[59][60][61][62][63][64][65][66]).

Target requirements
The overall procedure of the production of the medical radioisotopes involves several steps, including the target preparation, irradiation of the target in an accelerator, separation of the reaction products and recovery of non-reacted target isotopes from the target material and purification of such obtained isotopes [12,13,17,31,33,34,46,[52][53][54]80]. This paper focuses on the first stage of the procedure, i.e. the target preparation.
The accelerator targets used for production of the medical isotopes must meet certain medical and technical requirements [34]. Therefore, the overall approach to the electrochemical deposition of such materials may differ from the one applied for, e.g. protective layers plating. The most obvious property of the targets used for production of the medical radioisotopes is their very high chemical purity [1,75,81,82]. The final product containing generated radioisotope must meet respective medical standards of purity [81,82] and this aspect must be addressed in composition of the electrodes and the electrolyte bath used for the target deposition. Thus, the bath should not contain elements that can be co-deposited or coprecipitated with nickel and must be prepared using chemicals of appropriate purity [34]. It should be also free from toxic reagents [75]. If required, the reagents can be subjected to additional purification. As an example, solutions containing Ni ions can be purified using, e.g. ion exchange columns prior to completing the bath preparation procedure [49,83]. The bath composition is simplified and usually includes only necessary components, such as supporting electrolyte (improves ionic conductivity of the electrolyte), pH regulators and hydrogen evolution inhibitors. Other components of typical plating baths, such as surface levellers, brighteners, accelerators and suppressors [76], are usually omitted. The components of the plating system, such as glassware and electrodes, should be subjected to a careful cleaning according to well-established procedures [17,34,49,55,83]. It is also recommended to work with a closed system where the electrolytic cell is emptied using a stream of a highpurity neutral gas [50]. Finally, the nickel deposits are also to be cleaned and dried [17,31,46,50,84].
Very important requirements in respect to the accelerator targets are imposed by parameters of the nuclear reaction of interest [3,17,36]. An example of the energy influence on the cross sections (σ) is shown in Fig. 1 for 64 Ni(p,n) reaction [40,85], which is commonly applied for production of 64 Cu for PET studies and radiotherapy. Data are shown for the beam normal to the target, i.e. the target-beam angle equals 0°. The most desired projectile energy is the one which provides an acceptably high rate of the desired process (sufficiently high σ value, e.g. maximum seen in Fig. 1) with simultaneous reduction of probability of unwanted nuclear reactions [16,17,33,35]. Figure 1 presents the case when all the incident protons hitting the target atoms have the same energy equal to the respective value given in the abscissa axis. However, the projectiles energy decreases continuously with the distance passed inside the target as a result of their interactions with the target's atoms [86,87]. Thus, sufficiently thick target is able to reduce high energy of the incident projectiles (descending section of the curve from Fig. 1) to the level corresponding to the highest σ values (maximum in Fig. 1). On the other hand, when the deposit is too thick, the energy of the projectile deep inside the layer drops to the value below the threshold level and the nuclear reaction is not observed. This leads to a poor utilisation of the expensive target material. Consequently, when the layer thickness is too small, the highenergy projectiles may reach the substrate beneath the Ni layer. As a result of the latter, undesired nuclear reactions of the substrate may take place making the latter highly radioactive. In this case, only a small fraction of the beam energy is used for production of the medical radioisotope. Finally, too thin target will not reduce high energy of the incident projectiles to the required level corresponding to the highest σ values.
Calculation of the optimal thickness of the target layer to be deposited requires determination of the stopping power of the projectiles inside the layer in question. This parameter represents energy loss of a charged projectile inside the matter [88] and depends both on the penetrated material and on the projectile (type and actual energy). Accurate calculations of the optimal layer thickness require application of a dedicated software (e.g. [89,90]), as it was done in, e.g. [36,91]. Figure 2 shows how the energy of a charged projectile decreases with the distance travelled inside a solid nickel layer. Data were Fig. 1 Cross section, σ, vs. the proton energy plot (excitation curve) for 64 Ni(p,n) 64 Cu reaction. σ given in millibarn units (1 b = 10 −28 m 2 ). In the data from [40,85], the dotted line is meant as a guide to the eye. calculated using SRIM application [92] for incident 11 MeV and 15 MeV protons (beam normal to the target). This figure also shows cross section values for 64 Ni(p,n) 64 Cu reaction taken from Fig. 1 and plotted as a function of the average energy reached by the proton at a specified depth inside the nickel target. It follows that the effective Ni thickness, which reduces incident projectile energy by 90%, is equal to ca. 260 μm and ca. 450 μm for 11 and 15 MeV protons, respectively (Fig. 2). When the target-beam angle increases, the target penetration depth decreases accordingly [34,36,93,94]. According to several authors, the range of the incident protons energies, optimal for the 64 Ni(p,n) 64 Cu reaction, is from 8 or 9 to 12 MeV [33,35,40]. This corresponds to the highest cross section values seen in Fig. 1. It follows from Fig. 2 that the nickel thickness required to reduce energy of 11 MeV photons to 8 and to 9 MeV is equal to ca.77 and ca. 112 μm, respectively. A further increase in the layer thickness leads to a further decrease in the proton energy which eventually falls to the magnitude corresponding to considerably low cross section values (Fig. 1). Under such conditions the 64 Cu yield still increases with the Ni thickness but with a slower rate [17]. Nevertheless, higher thicknesses of Ni layers, which allow beam energy reduction down to, e.g. 6 MeV, are also used in practice for production of 64 Cu [55]. Some authors suggest that the optimal layer thickness should be equal to the maximum range of the projectiles in the selected material [36,94]. This approach can be analysed using threshold energy for the specified nuclear reaction. The threshold proton energy for 64 Ni(p,n) 64 Cu reaction is equal to ca. 2.5 MeV [36] and this value is reached at the depth of ca. 249 and ca. 439 μm for the incident 11 and 15 MeV projectiles, respectively (Fig. 2). Typical thickness of the Ni layers reported for the use in the medical accelerators vary from few to 200-300 μm, depending on the projectile type and energy and on the target-beam angle [17,36,41,95].
Preferred are smooth and homogenous deposits with uniform thickness which are free of cracks, craters, etc. [34,75]. Polycrystalline coatings, formed in a layer-bylayer manner or containing randomly oriented small size grains, meet these requirements [55,62,74,95,96]. Computer simulations reported in [97] show that application of targets, composed with crystals properly oriented in respect to the beam, may enhance the reaction yield through antichanneling phenomenon. Electrodeposition of Ni monocrystals, with the thickness even up to 100 μm or more, can be accomplished using well-oriented single crystal substrates, e.g. Cu [98][99][100][101]. This approach, however, seems to be hardly applicable for production of the accelerator targets in question. When exposed to air and moisture, the electrodeposited Ni undergoes surface oxidation which results in formation of surface nickel oxides and/or hydroxides [102]. This layer, however, does not create a significant barrier for energetic charged particles which are able to penetrate bulk of the sample.
Most of the discussed nuclear reactions use natural nickel isotopes of low abundance ( Table 1). The probability of a nuclear reaction is directly proportional to the amount of the target nuclei in the irradiated material [88]. It increases when the latter is enriched with the isotope of interest. As an example, the target used in 64 Ni(p,n) 64 Cu reaction, enriched with 64 Ni up to the level of 100%, provides probability of the reaction almost 100 times greater than the target made with natural nickel (0.93% of 64 Ni). Clearly, nickel used for preparation of the targets used for 64 Cu production should be enriched as much as possible with the required isotope. Isotopically enriched Ni can be very expensive and usually only a small amount of such material is accessible for the target preparation. Here, the electrochemical deposition is one of the methods most suitable for work with minute amounts of the material to be converted into the target. If possible, unreacted nickel isotopes should be recovered from the irradiated target and from depleted electrolytic baths.
Thermal stability is another one target property imposed by conditions of the nuclear reactions carried out in an accelerator. Under vacuum conditions, the target experiences a thermal shock due to a fast rise in the temperature when subjected to a beam of high-energy projectiles [95]. The temperature may rise up to 200-500°C and can be maintained for several hours [17,43,95]. After completing the irradiation, the target is quickly cooled down [95]. The nickel layers cannot crack or peel under such conditions. This requires high-quality nickel deposits which are free of cracks and which exhibit a very good adhesion to the substrate [34,75,95]. Separate tests of the target stability under conditions of fast heating and subsequent fast cooling are helpful here [34,55,95].

Bath preparation
The nickel can be purchased as a fresh reagent or can be recovered from the irradiated target or from a used electrolytic bath. Isotopically enriched nickel is expensive and usually accessible in limited amounts. As a result, its electrochemical deposition is typically carried out using a small volume (from few to 10 ml) of the electrolytic bath containing low or moderate concentration of the isotope (from few tens of mM to ca. 0.6 mM). The nickel concentration is selected in such a way that the deposition of all the nickel contained in the bath produces the metal layer with the required thickness.
The electrodeposition depends on numerous factors and by varying respective of them one may modify the process. These factors affect the process at its various stages and it is worth to summarise briefly proposed mechanisms of the electrochemical formation of coherent electrodeposits. Detailed discussion of this problem is out of scope of this paper and the readers are referred elsewhere, e.g. [103][104][105][106][107][108][109]. The electrodeposition process starts with reduction of the metal cations and formation of neutral adatoms at the substrate surface. The adatoms migrate across the surface until they reach energetically favourable sites where they form 2D or 3D nuclei (islands) together with other adatoms. The nucleation sites can be either located at the substrate terraces or formed by the substrate surface imperfections, such as kinks or edges [109,110]. After reaching the critical size, the nuclei become thermodynamically stable and grow with the deposition time. Their growth can be accompanied by continuous formation of new nuclei (progressive nucleation with low nucleation rate) or the number of the nuclei may remain almost unchanged since the beginning of the process (instantaneous case with a high nucleation rate). The nuclei grow along the surface imperfection or at the terraces, depending on the substrate surface structure and the electrolyte composition. They may grow through a direct discharge of the metal cations at the nuclei surface and incorporation of such generated atoms into the nuclei lattice [111]. An alternative model assumes reduction of the metal cations at the surface of the bare substrate following surface diffusion of such formed adatoms towards the nuclei [111]. When the plating time passes, the nuclei coalesce to form continuous deposit: 2D nuclei produce a single monolayer of the metal atoms, while the 3D entities generate a few-atomic layer thick coating. Further formation of thicker, coherent deposits proceeds according to two mechanisms [112,113]. Layer-by-layer grow includes incorporation of newly formed adatoms into kinks, steps and edges. The metal cations may be discharged directly at these surface imperfections or may diffuse from the discharge sites located at the terraces towards edges or steps [112]. New monoatomic layer grows in direction along and outside of the step. This process takes place when the substrate surface is stepwise, i.e. contains relatively smooth terraces with steps, and when the mismatch between crystal lattices of the substrate and the deposited metal is not too high [112]. Each new layer reproduces general arrangement of the original steps and edges [113]. The other mechanism (normal to the surface growth) includes formation and growth of subsequent 3D nuclei on top of the previously formed deposit layer [112]. It is followed by coalescence of the nuclei to form next continuous layer and subsequent formation of the new nuclei on the top of the latter [112]. This process takes place at surfaces of higher roughness where numerous nucleation sites can be formed in a stochastic manner [112]. It has been reported that various types of the nucleation and various mechanisms of the layer growth may occur simultaneously and the final structure of the nickel electrodeposit is a result of competition between various growth mechanisms [113].
The course of the abovementioned processes taking place during the nickel electrodeposition depends on numerous parameters. Thus, the electrode potential not only controls the charge transfer kinetics, including transition from the kinetic to the diffusional control of the process [103,114,115], but also influences number of the nucleation sites and the nucleation rate (e.g. [106,116]). Further on, occurrence of parallel to the electrodeposition electrochemical reactions, such as hydrogen evolution, also depends on the electrode potential [103,104,117,118]. As a consequence, the deposit properties (texture, microstructure) can be strongly affected by the deposition potential [119,120].
The bath composition also has a strong influence on the deposition process. The bath may contain components additional to the deposited metal cations, supporting electrolyte and pH regulators. These additives affect the plating process through interaction with the substrate surface (adsorption-desorption), interaction with the nickel cations (complexation) and incorporation into the deposit [103,114,115,121]. Thus, wetting agents diminish surface tension of the electrolyte, reduce undesired effects due to formation of H 2 bubbles and improve homogeneity and microstructure of the Ni coating [75,115,[122][123][124][125]. Other surfactants act as levellers and affect the deposit morphology and crystalline structure (size and preferred orientation), improve uniformity of the deposit thickness and reduce its roughness [107,114,115,124,[126][127][128][129].
Organic complexing agents affect several aspects of the electrodeposition process [117,118,130,131]. Formation of complexes with Ni cations improves microstructure and surface morphology of the deposit [120,[132][133][134], promotes specific orientation of the crystallites [132], changes nickel deposition rate [130,132,133] and increases microhardness of the coating [134], although it may decrease the overall faradaic efficiency [120,[132][133][134]. Properly selected complexes should exhibit respectively low stability; otherwise, electrodeposition may require higher cathodic polarisation as compared to the bath free from the complexing agent [75,117]. In summary, the bath composition and pH affects the nucleation rate and number of the nuclei [106,116,135,136], the deposit microstructure and texture [120,127,135,[137][138][139] and the layer growth rate [140].
The first step in the bath preparation procedure depends on the chemical form of the raw nickel material. When the nickel is in the form of salts, e.g. chlorides or sulphates [34,39,52,55,91,95], the bath preparation usually starts with dissolution of the salt in water or in an electrolyte solution followed by addition of other soluble bath components, such as pH regulators or hydrogen evolution inhibitors. The salt can be subjected to a purification procedure which involves several steps of dissolution and precipitation of the nickel compounds [34]. However, isotopically enriched nickel is most often available in the form of a metallic powder [13,31,34,38,55,68] and the bath preparation procedure requires selection of a proper method of the powder dissolution. Concentrated nitric acid dissolves metallic Ni powder easily and is usually considered as the solvent of the first choice [12, 17, 19, 31, 34, 38, 41, 43, 46-51, 56, 62, 68, 73, 96]. Such prepared electrolytic baths contain nitrates which may seriously affect the plating process. Although presence of a small amount of these anions can be beneficial for the deposit quality, too high concentration of the nitrates reduces faradaic efficiency of the deposition [141]. Nitrate ions may undergo cathodic reduction during the electrodeposition process leading to unwanted precipitation of Ni(OH) 2 [142,143]. Finally, presence of the nitric acid in the bath may result in dissolution of the nickel deposit when the latter is left under open circuit conditions or accidentally subjected to anodic polarisation [144,145]. Therefore, various procedures aimed at removal or reduction of concentration of nitrates/HNO 3 in the solutions obtained from the metal dissolution have been proposed (Table 3) [12, 17, 19, 31, 34, 38, 41, 46-49, 51, 56, 62, 68, 96, 149]. They include several steps, such as dissolution, precipitation, filtration and evaporation. Synthesis of a very high-purity Ni salt requires application of several such steps aimed to remove traces of the acids used during the procedure [62]. When the solid product is to be obtained, the final step may include precipitation of a high-purity Ni salt using solvent displacement crystallisation with an organic solvent, such as acetone [150,151]. These procedures are, however, time consuming and require additional experimental set-ups dedicated to, e.g. safe acid evaporation. As an example, vacuum evaporation of the acid is sometimes carried out for as long as 24 h [62]. Solvents less often applied for dissolution of the metallic Ni include HCl and H 2 SO 4 [10,33,34,37,55,57,59,84]. The dissolution can be facilitated by using oxidising agents, such as H 2 O 2 [10,57,59]. After completing the process, H 2 O 2 can be catalytically decomposed using metallic Pt [57]. Table 3 also includes procedures of removal of excess chlorides/HCl in the course of the electrolytic bath preparation [10,13,55].
Application of the procedures listed in the Table 3 leads to obtaining a solid nickel salt (NiSO 4 or NiCl 2 ) or an aqueous solution containing Ni 2+ (NiSO 4 or Ni(NO 3 ) 2 in H 2 SO 4(aq) ). These crude products are subjected to a further treatment in the course of the bath preparation. It includes dissolution in water (in the case of solid salts), dilution with H 2 O, pH settings by adding respective buffering agents and, sometimes, addition of H 3 BO 3 and other components of the supporting electrolyte. Two most often applied aqueous electrolytic baths used for deposition of the Ni targets are the alkaline one, usually with initial pH of 9-11 [12, 17, 19, 31-33, 43, 46, 47, 49-52, 56, 68, 83, 91, 96], and the slightly acidic one, typically with initial pH of 3-5 [10,39,41,50,51,55,62,91,95]. Application of the baths with other pH values, such as neutral [39,43] or strongly acidic [38], is rarely reported. A long-time nickel electrodeposition from acidic baths (initial pH of 3.0-4.6) is accompanied by a pH decrease to the values of ca. 1.3-1.7 [55,91]. These pH changes can be compensated by addition of pH regulators (e.g. NH 4 OH) during the deposition process [84].
Published compositions of the electrolytic baths used for electroplating of Ni targets used in medical accelerators are listed in Table 4 together with the deposition conditions and other parameters, such as efficiency and the deposit thickness. The alkaline baths are prepared by addition of NH 4 OH and NH 4 Cl or (NH 4 ) 2 SO 4 (or Na 2 SO 4 ) [12, 17, 19, 31-34, 43, 46, 47, 49-52, 56, 68, 91, 96], which act as pH regulators and supporting electrolyte. In some cases, also H 3 BO 3 is added [34,52,91]. In these solutions, the Ni 2+ cation forms a series of complexes with NH 3 with stoichiometry depending on the concentration of both components of the complex and on pH [91,116,[152][153][154]. Formation of such complexes increases solubility of Ni(OH) 2 which otherwise strongly decreases with the increasing pH [143,[155][156][157]. These complexes have a blue colour [143] which allows to track progress in the deposition process by means of colorimetric analysis of Ni content in the bath. This can be done optically or using a UV-vis spectrophotometer (main absorption maxima at ca. 612 and 373 nm [158]). Thus, when the whole Ni contained in the bath is to be deposited the plating process is continued until the electrolyte becomes colourless [17, 31-34, 38, 46, 49, 50]. It should be stressed that in the alkaline baths the NH 3 can be electrochemically oxidised in reaction taking place on the anode surface. These processes lead to formation of N 2 [159][160][161], nitrates [161,162] or chlorinated organic compounds when the chlorides are present in the electrolyte [163].
The slightly acidic baths are prepared by dissolution of Ni salts in water or by respective treatment of the acidic solutions obtained from the metal dissolution, e.g. dilution with water ( Table 4). The next steps may include addition of H 3 BO 3 [10,39,55,91] and inorganic salts containing NH 4 + [51] which also act as a supporting electrolyte and which may improve quality of the deposit [76]. Although both chlorides and sulphates which may be present in these baths can form complexes with nickel(II) cations [156], it was reported that at low Table 3 Procedures of conversion of the metallic nickel into soluble salts or their aqueous solutions which involve removal or reduction of concentration of nitrates/HNO 3 and chlorides/HCl in the crude product obtained by the metal dissolution. The products of these procedures (either solid Ni salt or its aqueous solution) are subjected to a further treatment, such as dissolution in water, dilution and pH setting, which lead to obtaining the electrolytic bath with the required composition. The final compositions of the baths are given in Table 4 Procedure Final product Reference pH values and for sufficiently low concentration of the anions, the Ni 2+ exists mainly as an aqueous complex in such electrolytes [152,[164][165][166]. These complexes are green (main absorption maxima at ca. 719 and 393 nm [152,158,167]), and, similarly to the alkaline bath case, the colour of the electrolyte can be used to track the progress in the deposition process. Ni 2+ complexes with chlorides and sulphates prevail at high concentration of the anions, especially in excess to the nickel cations [116,120,165]. Although sulphates/bisulphates apparently do not affect significantly the form of the Ni 2+ ions in the electrolyte [152], the anion identity influences the kinetics of the cation reduction [78,168]. This effect is attributed to formation of complexes between partially discharged Ni cations (Ni + ) and the anions. These complexes act as intermediates in the electrochemical reduction of Ni 2+ [168]. The anion identity (chloride or sulphate/bisulphate) also influences the density of the nucleation sites and the nickel nucleation rate [169]. Presence of chlorides in the acidic bath increases electrical conductivity of the electrolyte [152], affects hydrogen evolution [138,168], reduces stress inside the deposited nickel [170] and improves quality of the latter [91,168]. It is also convenient to use the chloride bath when the nickel is recovered from the irradiated target by means of dissolution in HCl [38]. Such obtained solution or solid nickel salt already contains chlorides and can be easily converted to the electrolytic bath [33,38,55]. The chlorides, however, are anodically oxidised on both the Pt [171] and graphite [172,173] anodes. Further on, the presence of the chlorides enhances anodic dissolution of platinum which may take place under strong anodic polarisation [174][175][176], i.e. under conditions often applied in the target deposition procedures. Such dissolved platinum can contaminate the nickel deposit.
Boric acid is the only other than the supporting electrolyte or the pH regulators addition often present in electrolytic baths used for the Ni target electrodeposition [10,13,34,39,52,55,91]. Dissociation of this weak acid strongly depends on the acidity. Thus, for pH below ca. 8-9 and for the acid concentrations in the range of 10 −1 M, the undissociated H 3 BO 3 is the prevailing form of the reagent present in the electrolyte [177]. This reagent has a strong influence on the nickel electrodeposition, although the literature is not fully consistent about its role [77,78]. According to various reports, it prevents formation of Ni(OH) 2 [178][179][180][181][182], improves deposit morphology and adhesion to the substrate [55,76,168,183,184], reduces number of cracks in the deposit [168], catalytically enhances Ni electrodeposition [179], regulates pH near the electrode surface [76,180,184] and suppresses hydrogen evolution [184]. The mechanism of the boric acid influence on the nickel electroplating seems to be not fully explained; it might be related to pH changes near the electrode surface (buffering agent) [178,179,183], surface adsorption [183] or formation of soluble complexes with the nickel cations [137,177,185]. Results of studies on H 3 BO 3 influence on the electroplating of Ni targets for production of medical radioisotopes were reported in [91]. It was confirmed that under conditions of the target electrodeposition, the presence of the boric acid in the bath improves quality of the deposit (lower surface roughness and lack of pitting), although it has no evident and clear  [84] (1) The same procedure (starting from point 2) has been applied to purify a solution containing nickel recovered from the irradiated targets [49] (2) Also applied for Ni salt recovered from the target after previous evaporation of the recovered solution [17] (  Metal powder See Table 3 for details Pt (a) [13] Metal See Table 3 for details NiSO 4 (94 mM (2) ); );      (2) );    );

Metal
According to [47] According to [  r.t. room temperature Atomic weights of natural nickel and its isotopes taken from [6] (1) Calculated on the basis of the substrate/deposit dimensions (radius, diameter, etc.) reported in the reference (2) Approximated values calculated on the basis of data (mass, volume, etc.) reported in the reference. In the case of the enriched Ni, the actual concentration values may vary depending on the enrichment level (not always reported) Electrodeposition until the bath becomes colourless (4) Indicates whether the current (galvanostatic conditions) or the potential difference (potentiostatic conditions) was applied and controlled (5) Calculated on the basis of reported current and deposition time (6) Calculated on the basis of reported bath composition and the amount of the electrodeposited nickel (7) Calculated on the basis of reported mass of the deposit and the surface area or on the basis of reported surface density (assumed molar bulk density of the natural nickel 0.15 M cm −3 [146]) (8) Calculated on the basis of reported faradaic efficiency and nickel concentration (9) Anions from Ni salt which was dissolved in H 2 SO 4 (see Table 3 for details) (10) Chemicals added to the bath regardless on their actual form in the final electrolyte (see text for the discussion of Ni complexes present in the electrolyte), e.g. NH 4 OH was added to acidic baths as pH regulator (11) Involves the following steps: concentration of the solution by heating, dissolution of the residue in concentrated HNO 3 , evaporation of the liquid, dissolution of the salt in concentrated H 2 SO 4 , concentration of the latter solution (12) Involves the following steps: dissolution in HNO 3 , evaporation of the acid and dissolution of the residue in H 2 SO 4 (13) ca. 1 g of dissolved Ni, used for determination of the excitation function, not applied for production of medical radioisotopes (14) Measured vs. Hg|Hg 2 SO 4 reference electrode (15) Surfactant-dodecyl-poly-ethylene-oxide-ether [147] influence on the plating efficiency [55,91]. Auger electron spectroscopy analysis of the surface and layers next to the surface of Ni deposited from a bath containing H 3 BO 3 shows that boron is present only in the bottom layers of the deposit, i.e. in those being in contact with the substrate [91]. NiB alloys are not formed during nickel electrodeposition from acidic baths containing H 3 BO 3 [55,186]. The limited number of the chemical reagents present in the electrolytic baths listed in the Table 4 helps to reduce the risk of contamination of the deposited targets. Reported results of chemical analysis reveal that such prepared Ni targets are free of significant quantities of metallic impurities [91,96]. Before starting the plating procedure, the bath, especially the one intended to be mixed by stirring during the deposition process, can be subjected to filtration in order to remove undissolved solid matter [34].

Recovery of Ni isotopes
The irradiated target usually contains a significant amount of non-reacted Ni isotopes [4, 12, 16, 17, 19, 25, 31-33, 38, 41-43, 46, 50, 52, 54-56, 60, 80, 83, 93] that can be recovered and used for preparation of a new electrolytic bath. This can be accomplished providing that the purity of such obtained reagents is satisfying [54,80]. In the first step of the recovery process, the irradiated target is dissolved using aggressive reagents, such as concentrated HCl, HNO 3 , often at elevated temperatures. Such obtained solution is subjected to ion exchange chromatography aimed at separation of the medical radioisotope from other reagents, including unreacted nickel [16, 17, 25, 31-33, 38, 41, 43, 46, 48-50, 52, 54, 56, 60, 80, 83]. As an alternative, one may separate Ni and Cu using selective liquid-liquid extraction [93] or using spontaneous electrodeposition with, e.g. Ni and Pt electrodes in a two compartment electrochemical cell [84]. Conditions applied in the recovery step are imposed mainly by the medical purity requirements and discussion of details of this procedure is out of scope of this paper. Instead, we focus on the eluted or extracted fraction containing the unreacted nickel. Several methodologies have been proposed here. In a simplest approach, one may directly convert the nickel containing eluate into the plating solution by diluting or preconcentrating, setting pH, adding respective chemicals, such as H 3 BO 3 [33]. More complex approaches involve additional purification of the Ni containing fraction [47,49]. Thus, it can be passed through an additional ion exchange column [31,32,38,46,49,52,80] and/or subjected to evaporation of the liquid and dissolution of such formed solid Ni salt. The latter evaporation-dissolution procedure can be applied once [17,31,38,46,48,55] or repeated multiple times [12,41,47,49,50,54,56]. When the acid used for the dissolution of the nickel target and elution of the ion exchange columns contain the same type of anion as the one present in the intended bath, e.g. chlorides, one may easily convert the solution obtained in the recovery procedure into the plating solution by diluting or preconcentrating, setting pH, adding respective chemicals, such as H 3 BO 3 [31,33]. The final composition of the baths containing recovered Ni and the deposition parameters are the same as for the solutions prepared using the fresh reagents [12,17,31,34,41,46,47,49,50].
A nickel recovery from the electrolytic baths is another source of the valuable isotopes which can be used for preparation of new electrolytes. A method of the metal recovery from an acidic bath after a failed deposition attempt has been described in [187]. It is based on quantitative adsorption of Ni ions on an ion exchange resin. The resin is introduced into a chromatographic column followed by elution of nickel. The eluate can be further purified using another ion exchange column. When the resin and eluents are properly selected, one obtains a Ni 2+ solution with a very low content of metallic impurities and additives originally present in the bath, such as H 3 BO 3 . Nickel can be also recovered from used baths which are depleted with the metal due to its consumption in the plating process [32]. Several such baths can be jointly reprocessed to obtain one with a sufficiently high Ni concentration [32].

Target electrodeposition
The nickel layers intended to be used as accelerator targets are deposited on a substrate which must possess good electrical and thermal conductivity [2,32,34,74,188]. The substrate must be relatively inexpensive, must be available at the required purity level and must provide a good adhesion of the deposited layer [32,34,188]. It must be also chemically inert when in contact with aggressive solvents, such as concentrated HCl, H 2 SO 4 , HNO 3 or aqua regia. These reagents are used for dissolution of the irradiated target layer containing generated medical isotopes and unreacted nickel [10, 12, 13, 17, 27, 31-33, 37-39, 41, 43, 46-50, 52, 53, 55, 56, 60, 62, 80, 83, 84, 91, 93, 96, 188]. Typical substrates used here include solid (disc, foil etc.) polycrystalline Au [13, 17, 19, 25, 32-34, 38, 42, 47-49, 51, 53, 55, 56, 60, 83, 188], Pt [34,46,50], Ag [12,37,39] and Cu, the latter covered with a thin protective layer of Au or Ag [34,41,52,93,95]. The latter are intended to prevent the copper from dissolution when the target is in contact with the acids. Such protective layers are usually obtained by electrodeposition [34,52,95]. It should be bearing in mind, however, that practical application of Cu substrates covered with Au or Ag can be complicated by intermixing of the atoms at the interface between the copper and the protective layer. Cu atoms which enter the protective layer diffuse towards the Ni deposit and may contaminate the latter [34,[189][190][191][192][193]. Such interface mixing (interdiffusion) and subsequent diffusion inside the solid state are accelerated at elevated temperatures [189,190] which can be reached by the target during the irradiation. Silver has the highest electrical and thermal conductivity amongst the most common substrate materials [32,146] but is less chemically stable than Au or Pt and may dissolve when in contact with HCl [32]. Further on, adhesion of Ni deposits to Ag substrate may be relatively weak [194]. Platinum has better mechanical properties than gold but exhibits a lower electrical and thermal conductivity [32,146]. It has been reported that thin gold foil is more suitable for use as the substrate for the accelerator targets than massive Au discs [32]. Less often reported substrates include Rh [84], graphite and Nb [34].
Apart from the chemical and physical properties of the substrate, one must also consider possible nuclear reactions taking place when the beam reaches the substrate material. The selection is based on a compromise between safety of the personnel and time required to deactivate the irradiated substrate. Preferred are substrates which do not produce significant amounts of radioisotopes and, if any, generate only short-lived radionuclides which do not emit very harmful radiation [32,84]. Short half-life (τ 1/2 ) increases the radionuclide-specific activity and dose delivered to the personnel and the chemical reagents, but, on the other hand, it allows reducing the time between the irradiation and the next use of the same substrate. As an alternative, one may allow generation of very long-lived radioisotopes which activity is very small due to a very long τ 1/2 . Here, we analyse 64 Ni(p,n) 64 Cu reaction as an example. Typical energy of the protons used for the 64 Cu production is not higher than 20 MeV. We assume that the nickel layer is thick enough, so the energy of the protons which hit the substrate is reduced to the value not exceeding 15 MeV [90]. Most important reactions of such protons with typical substrates are collected in Table 5. It follows that gold meets the abovementioned requirements better than Pt and Ag: it produces radioisotopes with shorter τ 1/2 (faster deactivation of the target) in reactions with lower cross sections (lower amount of produced radionuclides) as compared to Pt and Ag [32]. Rh is a very expensive metal but its irradiation with protons with energy up to 15 MeV produces almost exclusively short-lived 103 Pd which does not emit harmful radiation [84]. Niobium activation with protons produces short-lived 93m Mo which decays to very long-lived 93 Mo. The latter, however, emits only low-energy electrons and X-rays [7]. Apparently, application of niobium as the substrate has no other advantages [34].
The cathode (substrate) is flat (e.g. disc shaped) and usually located at the bottom of the electrolyte cell [10,12,46,47,50,55,62,68,96,208], although it can be arranged also as a component of the wall of the cell [39]. The exposed area of the target (typically 0.5-1 cm 2 ) can be limited using, e.g. dedicated o-rings, perforated plugs, masking plates or other components of the cell [10,12,13,46,49,55,83]. In automated systems, the substrate is arranged as a component of an element (a transport shuttle) which is automatically transferred between various sections of the system used for production of the radioisotopes [38,46,50]. The substrate is to be subjected to a proper cleaning/pretreatment procedure in order to provide good adhesion of the deposit [37,38,49,95]. These procedures include both chemical (dissolution of surface contaminants; etching) as well as mechanical (fine sanding) treatment [31,32,34,37,38,49,95,209]. In contrast to the substrates made with noble metals, less noble materials, such as Cu, may require removal of native surface oxides/hydroxides. The separation/recovery of the elements from the irradiated target can be accomplished using electrochemical anodic dissolution of the target layer [10,17,33,50] which also requires good electronic conductivity of the substrate. It is worth to stress that the electrochemical plating can be applied also for production of self-supporting nickel layers that can be used as targets in medical accelerators. A thick nickel layer is plated on an easily soluble substrate, e.g. Cu [29,149,209]. The Ni layer remaining after the dissolution of such substrate acts as the target, providing that its thickness and mechanical stability are sufficiently high [29,149,209]. This approach, however, has not been applied in production of medical isotopes.
The plating procedure usually requires deposition of all the nickel dissolved in the bath and this task is complicated by the fact that the metal concentration in the bath is usually low or moderate being limited by availability of the expensive nickel isotopes [51]. Under such conditions the deposition rate may quickly become limited by diffusion of nickel species towards the substrate (diffusion overpotential or concentration polarisation). Deposition of thick Ni layers from such baths is a time consuming process [2, 8, 10, 13, 17, 19 ,45, 59, 60-66]. It is, therefore, desired to develop the deposition procedure sufficiently fast as to ensure complete Ni deposition within a reasonable time. This requires an analysis of the factors which affect the plating efficiency. This includes the bath composition, the substrate surface pre-treatment, the applied current density or voltage, temperature and the electrolyte motion (stirring or recirculation). In some cases, the reports do not contain enough data required for complete scientific evaluation of the plating procedure. Nevertheless, data reported in Table 4 and supported by results of other works devoted to Ni electrodeposition for non-medical purposes allow drawing binding conclusions about selection of plating parameters suitable for preparation of the targets for the medical accelerators.

Influence of the nickel content in the bath
In general, the faradaic efficiency decreases with the decrease in the Ni concentration in the bath ( Table 4). The plating and faradaic efficiency decrease with the deposition time due to the decreasing Ni content in the bath [51,55,91] and due to influence of the deposition time on pH of the bath which, in turn, affects the hydrogen evolution [91,106]. Only at the beginning of the plating process the faradaic efficiency as high as 80% is observed [55]. Thus, even if as many as more than 85% of Ni is deposited during the first few hours of the process, the almost complete (quantitative) deposition of, say 97% of the nickel present in the bath requires up to 10-20 h [17]. As an example, deposition of ca. 100-130-μm-thick layers of Ni from baths containing 0.14-0.38 M of Ni requires from 6 to more than 20 h (estimated faradaic efficiency of 8-58%) ( Table 4). For comparison, when the nickel concentration in an acidic bath is as high as ca. 1.7 M, the deposition of a nickel layer with the thickness of 108 μm takes ca. 2.5 h with the faradaic efficiency of ca. 74% [10]. Therefore, the nickel concentration in the bath should be respectively high, what can be achieved by reducing volume of the electrolyte used, if possible.

Influence of the bath pH
In general, the efficiency of the Ni electroplating increases with pH [48,50,51,91,95,106,134,138,158] and the target electrodeposition proceeds significantly faster and with a higher faradaic efficiency when an alkaline bath is used instead of an acidic one [50,51,91,95]. Nevertheless, even in the case of the alkaline baths, the complete deposition of few tens to 100 mg of Ni takes many hours and the total faradaic efficiency lower than 50% is often observed (Table 4). Further on, the quality and mechanical stability of the nickel deposits deteriorates at high pH [39,158] and poor reliability and reproducibility of the target deposition carried out in the alkaline baths is reported [48,51]. Finally, the internal stress in Ni coatings strongly increases with the increasing pH of the bath [76]. The maximum current densities are comparable for the acidic and the alkaline baths ( Table 4). The bath pH may affect the nickel electrodeposition in different ways. Thus, it has an influence on Ni 2+ complexation: at high pH values, stable ammonia complexes predominate, while at low pH values, Ni 2+ form stable complexes mainly with water or anions [116]. Identity of nickel complexes, which act as intermediates in the electroreduction process, is also pH dependent [116,218]. The course of various eletrodeposition steps, such as adsorption at the electrode surface, decomposition of the complexes and the Ni 2+ reduction (kinetics and thermodynamics), depend on composition of these Ni containing species, which, in turn, is pH governed [116,136,218,219]. pH also affects adsorption of ions and hydrogen evolution [116,136,220]. All these factors lead to pH-dependent nucleation rate and nuclei density [116] and this, in turn, results in pHdependent deposit microstructure and texture [215,220].
Apparently, the alkaline baths are more often applied when the whole nickel is to be deposited and are more suitable when the deposition time and the faradaic efficiency play important roles as compared to the acidic electrolytes. Nevertheless, due to drawbacks of the deposition carried out from the alkaline baths, such as reported poor reproducibility and deposit quality, there is room for development of the acidic plating solutions.

Influence of the cathodic polarisation
Application of a strong cathodic polarisation of the substrate, i.e. high voltage or high current density, is probably the most obvious method of increasing the plating rate. An analysis of Table 4 shows that the faradaic efficiency increases with the voltage applied to the substrate. This approach, however, is limited by the fact that under too strong cathodic polarisation of the substrate, some undesired effects occur. Thus, under such conditions, the metal electroreduction takes place simultaneously to hydrogen evolution reaction, HER. As a result, a significant amount of the electricity is consumed by HER [33,91] which lowers the faradaic efficiency of the target deposition, even down to a few percent (Table 4).
Further on, at too high current densities, the morphology of the deposit deteriorates [39,55,91,95,158,209] and a high internal stress is developed [76,184,221]. Poor-quality deposits are obtained also when the current density is too low indicating that the applied current density should vary in a limited range [95]. The decrease in the deposit quality due to the increasing cathodic current can be attributed to the effects caused by the hydrogen evolution taking place parallel to the Ni deposition. It is assumed here that the HER contribution to the total current rises when the electrode polarisation becomes more cathodic. Presence of bubbles of gaseous hydrogen attached to the electrode surface leads to formation of the deposits with a high surface roughness and containing pores or cauliflower-like structures [117,221]. Further on, the hydrogen bubbles can be entrapped inside the Ni deposit forming microvoids which may affect its mechanical properties [222]. The deposits formed under conditions of the excessive H 2 bubbling are brittle [223] and this, in connection with non-uniform thickness of the rough deposits, limits application of high current densities for depositions of the accelerator targets.
Hydrogen absorption (uptake) by nickel, which may take place during electrochemical plating of this metal, is another process which may affect structure of the deposit and which importance rises with the increasing cathodic polarisation [78,180,[223][224][225][226][227]. Atoms of the absorbed hydrogen occupy interstitial sites inside the nickel lattice and form a solid solution or a nickel hydride (bulk or subsurface), depending on the absorbed hydrogen concentration [78,130,[228][229][230][231]. Desorption of hydrogen which was absorbed in the nickel coatings during the metal deposition as well as absorptiondesorption of hydrogen in/from earlier prepared Ni layers lead to changes in the lattice constant of the metal [231,232]. As a result, undesired stresses inside the metal lattice are generated [223,226,227,233,234]. This, in turn, may lead to formation of cracks in the deposit [234,235] which make them unsuitable for the accelerator applications. The importance of the hydrogen absorption, which is so disadvantageous to the Ni deposit quality, increases with the cathodic polarisation [232] and becomes more important when certain absorption promoters are present in the electrolyte (e.g. thiourea, CN − , S 2− , depending on pH) [138,223,227,229,232]. It follows then that such species should not be present in significant amounts in the baths used for plating the nickel targets.

Influence of other parameters
An analysis of Table 4 shows that a higher faradaic efficiency is achieved when the galvanostatic deposition is applied instead of the potentiostatic mode and when the Au substrate is used instead of Pt (complete Ni deposition from alkaline baths). The latter effect is attributed to the fact that the activity of Pt towards the hydrogen evolution reaction in alkaline electrolytes is higher than for Au [236,237]. Studies carried out in the temperature range of 24-60°C show that the target deposition efficiency is improved when temperature is raised [55,95]. It should be stressed, however, that these tests were performed for relatively thin nickel target layers with the thickness of up to ca. 65 μm. The Ni targets obtained at 55 o C were found to have a better surface morphology [55,95]. On the other hand, decorative and protective Ni coatings deposited from Watts bath reveal a decrease in the tensile strength and in the Vickers hardness when the bath temperature is increased up to ca. 50-60 o C [76]. The target deposition at elevated temperature carried out from a small volume bath may require continuous compensation of evaporated water using a dedicated pumping system [41].
Several studies on the electrochemical deposition of Ni targets reveal that the plating efficiency increases when the electrolyte is subjected to a motion [46,50]. This can be done using a rotating anode [10,13,32,41,47,49,62], magnetic stirring rod [56], other rotating element [95] or using an electrolytic cell with recirculation of the electrolyte in a closed system [46,50]. These procedures not only restore Ni 2+ concentration near the electrode surface but also effectively remove H 2 from the electrolyte [50]. The electrolyte stirring may affect morphology, texture and physical properties of the nickel deposits [41,107,134,238,239]. It has been reported that the deposit quality is improved when direction of the stirrer rotation is reversed during the plating process [41].
The quality of the deposit is determined also by the electrolytic cell geometry. The local plating rate and the deposit thickness distribution depend on the location of the anode in respect to the centre of the substrate (cathode) and resulting homogeneity of the electric field [46,75]. Although the electrodeposition may lead to formation of cone-shaped Ni deposits with non-uniform thickness [31,46], such targets also can be used in medical accelerators.

Suggested future works
It follows from the presented review that the electrochemical deposition of Ni targets for medical purposes is a wellestablished process which is routinely used in practice today. Nevertheless, several of its aspects need to be improved. It is suggested that the future works on this topic should address the following points: -Improvement and simplification of the method of the nitrite removal from the solution obtained by the dissolution of the metallic Ni in HNO 3 . As an alternative, a novel method of nickel powder dissolution in a HNO 3 free solvent may be also developed.
-Reduction of the electrodeposition time. Suggested solutions to the problem include application of a proper motion of the electrolyte during plating (stirring, circulation etc.) and selection of optimal current density or voltage.
-Increase in the faradaic efficiency of the process and, consequently, reduction of the plating time. Addition of proper hydrogen evolution inhibitors or Ni deposition accelerators can be helpful here [76,78]. Special care must be taken in order to avoid introduction of substances which may contaminate the deposit.
-Addition of proper concentration of respective, non-toxic surfactants and complexing agents to the electrolyte bath in order to improve quality of the deposit (e.g. [57,59,75,115,124,137,[240][241][242][243][244]. Such an approach may facilitate high current fast electrodeposition and may reduce the plating time. These additives or its reduction products, however, can be incorporated into the deposit leading to presence of unwanted elements in the target layer [75,103,107,114,115,121,123,124,[245][246][247][248][249]. Therefore, separate studies must be performed in order to analyse the mechanism of the additives action (e.g. adsorption-desorption, reduction) and to evaluate its incorporation into the nickel deposit [103,107,115,250,251].
-Microstructure of the nickel deposits is significantly improved when complex current or voltage programmes, such as pulsed plating (e.g. [117,127,240,248,[252][253][254][255][256] or different initial and final deposition parameters (current or voltage) [221], are applied. In the "regular" pulsed electrodeposition procedure, the cathodic polarisation is periodically switched on and off [257,258]. During the "off" step, the concentration of the metal cations at the electrode surface undergoes relaxation, while hydrogen diffuses out of the electrode [252,254,257]. It is also suggested that Ni(OH) 2 particles, which can be formed on the deposit surface, are desorbed during the "off" cycle [252,254]. The reverse pulsed plating procedure comprises of alternatively applied cathodic and anodic polarisation steps [75,117,255,[257][258][259][260]. Apart from Ni 2+ concentration relaxation and hydrogen removal from the electrode, the application of the anodic step also leads to a partial dissolution of the nickel deposit [75,117,255,257,258]. When the anodic polarisation conditions (time and current or potential) are properly selected, only most active surface features, such as low coordination-number adatoms, which form roughness increasing features, like pyramids or peaks, are dissolved. This produces smoother surfaces as compared to the direct current electrodeposition and "regular" pulsed approach [117,245,260]. The faradaic efficiency can be also improved when the reverse pulse plating is applied [245]. This may increase the plating rate through improving quality of the deposits obtained under conditions of strong cathodic polarisation of the substrate (applied high voltage or high current density). Such procedure, however, requires proper selection of several parameters, such as duration of respective steps, pulse frequency and applied cathodic and anodic polarisation. Further on, the reversed pulsed plating requires electrolytic bath with pH sufficiently low for nickel dissolution. Otherwise, the anodic polarisation may lead to surface passivation with nickel oxides or hydroxides instead of the dissolution. Finally, the equipment used in this type of the experiments is more sophisticated than that used for single current or voltage deposition.
-Addition of anodic and cathodic depolarisers reduces polarisation (overvoltage) of the anode and cathode (substrate), respectively [75,118,139,248,261]. Thanks to lowering overpotentials on the respective electrodes, they diminish undesired evolution of gases on both electrodes and dissolution of the anode material [139,261]. Electrochemical reactions of such depolarisers, however, may result in unwanted effects. Thus, products of the electrochemical reaction of the depolariser may be incorporated into the plated layer or may cause additional chemical processes, such as those leading to precipitation of Ni(OH) 2 when nitrates are used as the cathodic depolariser.
-Texture and microstructure of relatively thin Ni can be affected by selection of the substrate with a certain crystalline structure [108]. Also proper pre-treatment of the substrate surface, such as mechanical or electrochemical polishing, annealing and etching, improves properties of the electrodeposited nickel, e.g. roughness, microstructure, texture and tensile strength [262][263][264]. These approaches can be easily applied when the substrate has a form of, e.g. regular metal plate, but their applicability becomes limited when the substrate constitutes integral section of a complex transport shuttle used in automated systems.
-Deposition of the Ni target layers on different than the described above materials suitable for use in accelerator systems, e.g. Al [32,68]. These substrates must be properly secured against dissolution during the isotopes recovery.

Conclusions
Reactions of nickel nuclei with high-energy charged particles (protons, deuterons, α particles) produce Cu, Co and Ni radioisotopes of medical importance. These reactions require a beam of high-energy charged particles, provided by an accelerator, and a respective nickel containing target to be bombarded by these projectiles. Typically applied are solid nickel targets prepared by electrochemical deposition on a conducting substrate. The targets must meet certain medical and technical conditions in order to be suitable for production of the medical isotopes. These requirements, in turn, must be addressed in the procedure of the electrochemical target plating. Preparation of the suitable Ni targets requires both electrolytic bath with proper composition and application of carefully selected deposition parameters. Several methods of the electrochemical deposition of the accelerator targets containing nickel layers are reviewed in this paper. They differ in respect to the electrolyte composition, including nickel concentration, and the deposition parameters. Different types of the raw nickel material, i.e. salts or metallic powder, are used for preparation of the bath. Further on, various methods of treatment of the isotopically enriched metallic Ni powder prior to its introduction to the bath are reported. It follows that the nickel layers with the thickness ranging from 10 0 to over 200 μm can be deposited from aqueous solutions containing low or moderate concentrations of the nickel ions. Properties of these layers, such as surface morphology, mechanical stability, cohesion and adhesion to the substrate, satisfy the requirements of the medical accelerators of the particles. Deposition of such targets, however, is time consuming and often requires at least several hours. It is suggested that certain aspects of the procedure of the electrochemical plating of the nickel targets can be improved in the future.