RSC Advances PAPER Hybrid approach to obtain high-quality BaMO3 Cite this: RSC Adv., 2020, 10, 28872 perovskite nanocrystals† Natalia Chamorro, ab Jordi Martı́nez-Esaı́n, ‡ab Teresa Puig,b Xavier Obradors,b Josep Ros,a Ramón Yáñez a and Susagna Ricart *b A novel hybrid solvothermal approach for perovskite nanocrystal formation via accurate control of the hydrolytic process is reported. This new synthetic methodology sets a whole general route to successfully tune the sizes of high-quality BaMO3 (M ¼ Ti4+, Zr4+, and Hf4+) perovskite nanocrystals. Purely cubic-phase nanocrystals (stable in alcohol media) were obtained using controlled water amounts, combining the well-known aqueous sol–gel process with the classic solvothermal method. Received 29th April 2020 Accepted 21st July 2020 Exhaustive optimizations revealed feasibility of a fast (1 hour) and reproducible synthesis with small variations in the crystal size or agglomeration parameters. The study also reveals water content as the DOI: 10.1039/d0ra03861j pivotal factor to achieve this wide range of sizes through a controlled hydrolytic step. Finally, the study rsc.li/rsc-advances of the hydrolytic process made it possible to shed some light on mechanistic insights of this synthetic route. is usually avoided because the fast hydrolysis promotes forma- Introduction tion of non-homogeneous systems; this requires a posterior Nowadays, the search for a fast, cheap, reproducible and robust calcination treatment, reducing the nal NC stability in solu- approach to synthesize oxide nanocrystals (NCs) still attracts tion.12,22 On the other hand, non-aqueous approaches23–27 have strong attention among researchers. This high interest derives risen up during last decades due to several advantages such as from their wide range of applications in electronics,1 medi- high crystallinity, controlled crystal growth and homogeneous cine,2,3 catalysis,4,5 energy storage6,7 and sensing.8 Among metal NC dispersions.28,29 However, this methodology shows two main oxide NCs, the perovskite family has gained attention due to its drawbacks: long reactions times and the need to use high stability under high-temperature treatments and its emergent temperatures during the process.12,17,18 We aimed to go further applications.9–11 Although, BaTiO3 NCs have been extensively by trying to control the hydrolytic step with different water studied,12–15 the achievement of BaZrO3 and BaHfO3 NCs proportions applying same conditions as in the “non-aqueous” continues to be an entangled process, in which researchers keep methodology. This approach aims to merge the advantages of nding difficulties to obtain small-size, crystalline NCs, stable both routes (i.e., sol–gel and solvothermal) of perovskite in colloidal suspensions without agglomeration.16–18 synthesis in one that combines short reaction times, high Although the applicability of these systems is clear and well- crystallinity and long stability of the NCs. known, the synthesis of this family of perovskites still entails Herein, we not only report a suitable synthetic methodology many limitations to obtain stable, homogeneous and high- for a group of perovskite NCs (BaMO3) but also a general route quality NCs. Our main goal is to nd alternative methods for that allows accurate size control of high-quality NC systems. these intricate perovskite NCs using the current background This new proposed synthetic procedure has been tested for researches in the eld. Two main strategies are highlighted: (i) three tetravalent cations (Ti4+, Zr4+ and Hf4+), being the rst non-aqueous solvothermal treatments and/or (ii) sol–gel time that the same methodology can be applied to this perov- aqueous processes. Although the hydrolytic mechanism of the skite family, obtaining comparable quality NCs. This novel aqueous sol–gel procedure is well-known,19–21 this methodology approach allows to ne-tune the NC size and an excellent uniformity of NCs dispersion, with promising applications in emergent energetic devices. The present work conducted by our a Departament de Quı́mica, Universitat Autònoma de Barcelona, 08193 Bellaterra, group showed the possibility to combine BaZrO3 and BaHfO3 Spain b NCs with YBa2Cu3O7 (YBCO) type II superconductors, to Institut de Ciència de Materials de Barcelona (ICMAB-CSIC), 08193 Bellaterra, Spain. E-mail: ricart@icmab.es enhance the superconducting properties, unravelling a novel † Electronic supplementary information (ESI) available. See DOI: pathway in electronic materials science and bestowing a crucial 10.1039/d0ra03861j breakthrough for the implementation of NCs to act as pinning ‡ Present address: Catalan Institute of Nanoscience and Nanotechnology (ICN2), centres.30–32 Thus, we present here a novel hybrid method (i.e., CSIC and The Barcelona Institute of Science and Technology, Campus UAB, Bellaterra, 08193 Barcelona, Spain. 28872 | RSC Adv., 2020, 10, 28872–28878 This journal is © The Royal Society of Chemistry 2020 Paper RSC Advances built from the classical approaches) to produce high-quality Transmission Electron Microscopy (HRTEM) micrographs ob- perovskite NCs and enhance the current applicability portfolio. tained on a 200 kV JEOL 2011 TEM. Dynamic (SPECS GmbH, Berlin, Germany) in ultra-high vacuum conditions (base pres- sure 4  1010 mbar) with a monochromatic aluminium Ka X- Experimental methods ray source (1486.74 eV). Light scattering (DLS) was done using Materials a Zetasizer Nano. XPS measurements were performed with The following chemicals were used without further purication. a Phoibos 150 analyser. Micro-Raman measurements were Zirconium(IV) n-butoxide solution 80 wt% in n-butanol, haf- performed in backscattering geometry at room temperature nium(IV) n-butoxide 99%, titanium(IV) n-butoxide, triethylene using the 5145 Å line of argon-ion laser with a XploRA PLUS glycol (TEG 99%), barium hydroxide octahydrate $98% and Raman spectrometer from Horiba, attached to an Olympus ammonium hydroxide solution 30% v/v were purchased from microscope and equipped with CCD detector. To perform Sigma Aldrich. Ethanol absolute was purchased from Scharlau. microwave reaction, a microwave advanced exible synthesis platform (exiwave) from milestone was used. Nanocrystal synthesis In a standard procedure, 1.24 mL (2.8 mmol) of Zr(OC4H9)4 or Results and discussion 2.8 mmol of the desired n-butoxide precursor (M ¼ Ti4+, Zr4+ or The hybrid solvothermal method Hf4+) was added to 10 mL of absolute ethanol at room According to our previous experience in metal oxide NC temperature under N2 atmosphere and vigorous stirring for synthesis, we aimed to use a modied polyol route33 and make it 5 min. Followed by the addition of barium hydroxide octahy- compatible with the BaMO3 (M ¼ Ti4+, Zr4+ and Hf4+) perovskite drate, 1.104 g (3.5 mmol) until forming a homogeneous solution oxide NCs. The optimised procedure requires mixing 10 mL of (10 min). Then, 6 mL of triethylene glycol (TEG, capping ligand) ethanol and 6 mL of TEG to prevent the agglomeration or and 1 mL of NH3 30% v/v solution (or instead the further precipitation. corresponding mmol of water detailed later in this work) were The versatility of this approach is reected by obtaining added under continuous stirring at room temperature resulting aggregation-free monodisperse BaMO3 (M ¼ Ti4+, Zr4+ or Hf4+) in a milky solution (5 min). The nal suspension was trans- NCs (Fig. 1a–c, respectively) with a uniform size and shape, as ferred into a steel autoclave equipped with a Teon-line and corroborated by the histogram in Fig. S1.† The transmission heated up at 180  C under constant stirring for 1 hour and electron microscopy (TEM) images and powder X-ray diffraction slowly cooled down to room temperature. The resulting orange- (XRD) reveals a square-like shape of the BaZrO3 with a 8 nm coloured suspension was washed as follows. Firstly 10 mL of size while the two other perovskites (Ti4+ and Hf4+) showed ethyl acetate was added, sonicated during 10 min and then spherical shape with sizes of 8 and 7 nm respectively. These centrifuged at 10 000 rpm for 10 min. Then, supernatant was results are in close concordance with those calculated using the discarded and 10 mL of ethyl acetate : ethanol (4 : 1) were Scherrer equation (Table 1). Moreover, the resulting perovskites added, the mixture was sonicated until total dispersion of the pellet and later centrifuged at 10 000 rpm for 10 min (this washing procedure was repeated twice). In the nal step, the pellet was dispersed into absolute ethanol. To ensure a homo- geneous size distribution of the NCs an extra centrifugation of 10 min at 7000 rpm was performed, discarding the pellet and keeping the supernatant, with yield loss of 15–20% from the initial. Microwave synthesis was performed following the same procedure as solvothermal reaction but replacing 1 mL of NH3 30% v/v solution with 0.5 mL H2O (added in the same step). Aer the mixing, the resulting milky solution was transferred to the MW Teon and heated with a ramp of 36  C min1 until 180  C for the BaMO3 (M ¼ Zr4+, and Hf4+) and 100  C for the BaTiO3, holding the corresponding nal temperature for 5 min Aerwards, the same cleaning process as the solvothermal procedure was performed and nally dispersed into absolute ethanol. Characterization Fig. 1 TEM images of BaTiO3 (a) BaZrO3 (b) and BaHfO3 (c) by using hybrid solvothermal reaction at 180  C and 1 hour, showing their X-ray powder diffraction (XRD) samples were recorded by corresponding HRTEM. The index of Miller of the principal peaks by a Phillips XPert diffractometer equipped with a two circle XRD (d) have been assigned into the diffractogram to confirm the diffractometers and Cu tube. Transmission electron microscopy cubic structure. On BaHfO3 appears an * BaCO3 peak suggesting a 5% (TEM) micrographs were obtained on a 120 kV JEOL 1210 TEM, of impurity on the pattern. This journal is © The Royal Society of Chemistry 2020 RSC Adv., 2020, 10, 28872–28878 | 28873 RSC Advances Paper Table 1 Summary of average sizes (Scherrer and TEM) and a powerful tool to predict the stability based only on the BaMO3 morphology of the obtained perovskites NCs perovskite compounds. This equation has a direct dependence on the ionic radii of each element (BaMO3), where rB is the ionic NCs Scherrer size (nm) TEM size(nm) Morphology radius of Ba, rM the ionic radius of (M ¼ Ti4+, Zr4+ or Hf4+) and rO BaTiO3 7.8 8.3  1.4 Spherical the ionic radius of oxygen. To corroborate the stability of a cubic BaZrO3 8.2 8.6  1.5 Square-like perovskite by tolerance factor the range of values should be BaHfO3 6.8 7.0  1.2 Spherical between 0.8 and 1.05. In the case of M ¼ Zr4+ and Hf4+, a tolerance factor of t ¼ 1 is obtained (Table 2), which is considered as the ideal cubic structure, where the ions are in show high crystallinity, without the need of thermal treatments a perfect cubic lattice. aer the synthesis (Fig. 1d) contrarily to standard sol–gel ðrB þ rO Þ procedures. t ¼ pffiffiffi (1) Many perovskite NCs from this family have been synthesised 2ðrM þ rO Þ using stoichiometric amounts of Ba : M (1 : 1), even though the resulting NC systems are not homogeneous enough nor well Applying eqn (1) to the BaTiO3, a value of t > 1.05 is obtained dispersed.34–36 Nonetheless, it has been demonstrated that the due to a smaller size of Ti4+ cations, this might cause a distor- addition/excess of hydroxide compounds could lead to an incre- tion on the lattice structure which could entail a transition from ment on homogeneity and NC sizes, as well as, a better dened cubic phase to tetragonal, therefore showing a ferroelectric NCs shape.37,38 We aimed to perform our hybrid solvothermal behaviour. method in both conditions (i.e. stoichiometric and non- To verify the possibility of tetragonal phase, XRD diffraction stochiometric Ba : Zr ratios) to study its effect to BaZrO3 NCs. As could show some limitation since both, cubic and tetragonal shown in Fig. S2a,† the use of stoichiometric Ba : Zr ratio phase, may coexist together. This would be observed by a split produced a non-homogeneous NC dispersion with agglomeration peak over 45 in 2q in the XRD as an evidence of tetragonality. observed via TEM and corroborated with DLS (i.e. NC size one Although Fig. 1d shows an individual broad peak, the assigna- magnitude order higher than non-stoichiometric ones). However, tion cannot be fully attributed to cubic phase as a consequence both NC dispersions show the same cubic phase in XRD of working at such reduced sizes (8 nm), which limits the XRD (Fig. S2c†), making us to postulate a surface effect on the stabili- analysis. zation of the NCs allowed with an excess of Ba(OH)2. To complete the study on the phase transition a comple- As these NCs appear as good candidates for a wide range of mentary analysis using Raman spectroscopy was performed material applications, assessing their stability and dispersion is (Fig. S3†). The inactive bands present on paraelectric cubic mandatory. The DLS analysis revealed that the NCs remain well phase (176, 270 and 520 cm1) become active on the transition dispersed and stable in alcohols for 6 months, at concentration to tetragonal phase13,14,42,43 at 306 cm1 (sharp peak) and up to 80 mM for all cases. DLS analysis of the BaTiO3, BaZrO3 713 cm1 with well-dened and high intensity peaks are the and BaHfO3 NCs in EtOH, show that they are stable in solution footprints to determine the tetragonally. BaTiO3 NCs powder at at least 6 months, exhibiting constant sizes of 7–10 nm, in close room temperature was analysed, observing the absence of agreement with the results obtained by TEM and XRD tech- a sharp peak at 306 cm1, and a subtle broad peak on 713 cm1 niques (Fig. 2). without good denition. The information extracted from this To study the crystal structure stability, the Goldschmidt data suggests a non-dominant tetragonal phase on the NCs tolerance factor (t) equation (eqn (1))39,40 was reported as which leads to cubic perovskite postulation. Time and temperature screening Special attention was paid to the tuning of the NC synthesis. In particular, temperature and time are claimed to have a direct effect on the size, shape and crystallinity of the nal colloidal dispersions.44 By using solvothermal methodology, we are inducing a temperature and pressure (depending on the Table 2 Overview of the radii from each element of the different perovskites.41 Ionic radius of O2 is 1.40 Å Ionic radius [Å] Perovskites Tolerance BMO3 Cation X2+ Cation M4+ factor Fig. 2 A stability study by using DLS for monitoring the stability of the BaZrO3 Ba2+ 1.61 Zr4+ 0.72 1.00 nanocrystals (BaTiO3, BaZrO3 and BaHfO3 synthesised at 180  C for 1 BaHfO3 Ba2+ 1.61 Hf4+ 0.71 1.00 hour) from the moment of the synthesis to 6 month, all of them in BaTiO3 Ba2+ 1.61 Ti4+ 0.61 1.06 ethanol media. 28874 | RSC Adv., 2020, 10, 28872–28878 This journal is © The Royal Society of Chemistry 2020 Paper RSC Advances precursors and solvent) which will trigger the nucleation of the agglomeration parameters by the previous aqueous sol–gel step, crystals and a posterior growth of these nuclei to mature crys- thus a kinetic control over the NCs. tals. For this purpose, we performed a study modifying the temperature and time to observe the correlation of these Tuning the NC size parameters with the nal NCs (i.e., sizes, shape, crystallinity). Firstly, the temperature was decreased from 180 to 100  C for Using the standard method for the synthesis of BaZrO3, 1 mL of 1 hour, as it is observed in Fig. S4.† The lack of crystallinity in aqueous NH3 at 30% v/v is necessary to obtain the nal NCs the cases of BaZrO3 and BaHfO3 at 100  C is shown in the XRD solution. The role of ammonia in the reaction was not clear, due pattern (Fig. S5†) with a non-dened cubic shape, which entails to its ability to act as a base or as a cationic stabiliser on the NC not enough temperature to induce the nal crystals. However, surface (i.e., as ammonium cation stabilizing the system).47 In BaTiO3 shows a similar cubic pattern synthesised at both order to understand the role of the ammonia and its inuence temperatures 100 and 180  C (Fig. S4 and Table S1†). From this in the nal nanoscale system, X-ray photoelectron spectroscopy standpoint, Ti4+ precursor needs low temperatures (i.e., 100  C) (XPS) was performed. In Fig. S9† can be demonstrated the suggesting a different growth rate compared with the other two, absence of nitrogen compounds on the NC surface, meaning BaZrO3 and BaHfO3 (i.e., 180  C). This difference in growth that ammonia or ammonium cations are not playing a stabili- conditions could be driven by a faster supersaturation point and zation role. different surface energy by using titanium instead of zirconium To evaluate the role of NH3, we compared the NCs obtained or hafnium precursors.44 Although each metal has a different using 0.7 mL of water or 1 mL NH3 at 30% v/v (being 0.7 mL of chemical and physical proprieties, it should be noted the close water) as shown in Fig. S10.† Comparing TEM images and XRD similarities between hafnium and zirconium, showing almost results, we claim that there is not difference between the use of identical radii compared with titanium.45,46 1 mL ammonia 30% v/v or directly using its equivalent amount Then, we held the temperature and modied the reaction of water (0.7 mL of water). NCs form homogeneous dispersion time from 1 to 24 hours (Fig. S6†). It is observed that 1 hour is as observed in the TEM images with square-like NCs, narrow more than enough to synthesise the NCs, reducing the current size distributions and cubic phase (Fig. S10†). This evidence, in growth times for BaZrO3 and BaHfO3 on solvothermal proce- addition with the XPS analysis, suggest that ammonia did not dure from hours to just 1 hour (Fig. S7†). XRD pattern (Fig. S7 play any additional role in our methodology, being just the and Table S2†) did not reveal any evidence of difference on amount of water coming from ammonia solution who govern crystallinity or size (i.e., via Scherrer equation) between the the NC formation. different reaction times. Consequently, it seems that the To demonstrate the specic role of the water, we used several nucleation rate of the NCs is almost instantaneous once they are quantities (from 0 mmol to 139 mmol) to test which is the subjected to an adequate combination of pressure and pivotal role of the water in the system (Table 3). The most temperature in the autoclave, without changing the morphology striking result emerging from this data is how an incrementa- of the NCs. Therefore, 1 hour is enough for the NC formation, tion of water contain (0, 28, 55, 83, 111 and 139 mmol) produces and keeping the reaction 24 hours at the same temperature did a continuous increment in size from 4 to 31 nm (Fig. 3 and not affect the agglomeration of the NCs and even refute the Table 3) comparing the sizes by TEM average and Scherrer possibility of the NCs to suffer grain growth caused by an Ost- equation. wald ripening. Firstly, a test without adding water in the media using the It is well known that microwave is claimed to reduce reaction standard protocol as before (180  C and 1 hour in autoclave) time in the synthesis of oxide NCs compared with the sol- evidences initial seed formation of NCs (Fig. 3a). XRD pattern vothermal ones.16,24 However, in our case we were able to ach- showed a high presence of BaCO3 as the main product ieve shorter times in the solvothermal reactions without the (Fig. S12†), suggesting that the water amount in the Ba(OH)2- need of microwave procedure. Even still, the reaction was $8H2O is not enough to initiate the nucleation process or the checked by microwave methodology (Fig. S8 and Table S3†) at conditions are not still suitable to obtain desired nal NCs. 180  C (BaZrO3 and BaHfO3) and 100  C (BaTiO3) for 5 minutes, Considering the other cases (i.e., from 28 to 139 mmol of giving comparable NCs as in the autoclave. Moreover, a micro- water), DLS analysis corroborates by TEM sizes showing an wave prole of BaZrO3 has been exposed in Fig. S8b,† the graph increment of hydrodynamic radii while the amount of water shows the variation of pressure and temperature as functions of time, showing a pressure incrementation over 15 bars and Table 3 Summary of the NC sizes by using TEM average and Scherrer a reduction in reaction time from 1 hour in solvothermal route equation obtained at different amounts of water to 5 minutes by applying a microwave route. In consequence, H2O amount Scherrer size TEM size both heating procedures are compatible with the hybrid sol- vothermal method without signicant difference in crystal- 0 mmol — — linity, homogeneity or sizes. These results seem to evidence that 28 mmol 7.4 nm 7.0  1.2 nm the solvothermal reaction is only required to raise the system to 55 mmol 11.8 nm 11.9  1.8 nm 83 mmol 14.1 nm 14.7  3.5 nm a desired temperature and pressure for minutes to achieve the 111 mmol 25.0 nm 25.8  7.5 nm nal crystallisation, leaving the possibility to govern size and 139 mmol 29.0 nm 32.6  9.8 nm This journal is © The Royal Society of Chemistry 2020 RSC Adv., 2020, 10, 28872–28878 | 28875 RSC Advances Paper Fig. 3 (a) TEM images of BaZrO3 NCs using different additional water amounts during the synthetic process (from 0 to 139 mmol). (b) DLS measurements of the different BaZrO3 NCs at 28, 55, 83, 111 and 139 mmol of water (avoiding the ones at 0 mmol due to not fully formation NCs) and (c) the standard deviations of TEM histograms are showed to note the polydispersity when increasing water amounts. increases, as well as, the stability of the NCs on the ethanol stabilization and homogenisation of the alkoxide in the solvent is media (Fig. 3b). At this point, we realised that the effect of water considered the rst step to form the intermediate M(OBu)4x(- is not only related to the increase of size but also in the poly- OEt)x (eqn (2)). Alkoxides are extremely sensitive to water, so the dispersity of nal colloidal systems (Fig. 3c and S11†). Addi- addition of Ba(OH)2 octahydrate plays a double role; (i) the tionally, the standard deviation in TEM histograms increases addition of the binary cation for the perovskite structure and (ii) with the size of the NCs. From a kinetic point of view, this effect the source of –OH groups and water which will start the hydro- could be explained by the hydrolysis step, if the nucleation lysis reaction (eqn (3)) and the formation of intermediate species process is increased without modications on the thermody- based on a metal hydroxide bond M(OH)x(OBu)nx. Poly- namics, the system becomes polydisperse and with poor control condensations are promoted as it is described in eqn (4) and (5), on the stabilisation (i.e., we are increasing the amount of water forming a M–O–M framework, being a critical point for the with the same amount of capping agent, TEG). Although NC size further formation of nal BaMO3. The increase of –M–OH species is tuned with the addition of different amounts of water, other in our reaction media is controlled with the addition of water, properties are nearly similar in terms of shape, homogeneity allowing the formation of large –M–O–M– chains (poly- and crystallinity. Concerning their crystalline structure, all NCs condensation), producing bigger NCs. present a cubic structure with a Pm3 m space group (Fig. S12†), in which Scherrer sizes match with the obtained via TEM and M(OBu)4 + xEtOH / M(OBu)4x(OEt)x + xBuOH (2) DLS (Table 3). M(OBu)n + xH2O / M(OH)x(OBu)nx + xBuOH (3) Mechanistic insights in the hybrid solvothermal method –M–OH + H–O–M / –M–O–M– + H2O (4) To shed more light on this growing process we need to focus on the mechanistic insights of the reaction. If the NC size relies on –M–OH + Bu–O–M / –M–O–M– + BuOH (5) the amount of water, the hydrolysis step should be the key factor in the control of the NC size, and hence, the behaviour of Simultaneously, this experiment was carried out with the the metal cation against hydrolysis. Our starting precursors are other two different metal precursors Hf4+ and Ti4+ alkoxides, alkoxides, and even though their chemical reactivity has been revealing important insights into the growth process of the NCs. widely studied,48,49 there is still noted a controversial aspect for The behaviour of BaHfO3 is similar to BaZrO3, the addition of knowing about the full mechanism reaction.23,29 water (28 and 111 mmols) increases the NC size from 4.6  As it is well known, the strategy of sol–gel is based on the 0.7 nm to 14.8  3.4 nm respectively (Fig. 4a and b). However, hydrolysis and polycondensation steps (eqn (3)–(5)) and it has the ndings with BaTiO3 do not support the previous research, been extensible as the main step to synthesise oxide nano- showing a quite different behaviour not being the amount of particles.21,23,50,51 Herein by following this mechanism a sol–gel water the one who governs the NCs tuning sizes. Aer adding 28 process is shown with a chemical reactivity tailored by the and 111 mmols of water, BaTiO3 NCs do not show signicant manipulation of the water amount. In an ethanol media, the differences in size from 8.7  2.6 and 9.2  2.1 nm, respectively 28876 | RSC Adv., 2020, 10, 28872–28878 This journal is © The Royal Society of Chemistry 2020 Paper RSC Advances methodology. This novel strategy combines both aqueous sol– gel route and solvothermal reaction in a two-step process. As far as we know, this approach allows for the rst time the use of the same methodology to obtain BaTiO3, BaZrO3 and BaHfO3 NCs with high control of size, shape, reproducibility and stability. In addition, by using this approximation, we evidenced the crucial role of water in the hydrolytic step, arising as a key parameter to tune the sizes of the nal NCs. This allowed us to tune BaZrO3 and BaHfO3 size from 4 to 40 nm, and 4 to 15 nm respectively without deciency of crystallinity and shape- dened appearance. In the other case, BaTiO3 has a completely different behaviour, giving rise to constant 8–9 nm size NCs, independently of the water proportion. Moreover, a complete study in a particular sized NCs show a stability of several months in polar solvents (i.e., ethanol), being remarkable candidates for a full range of applications without taking care of the solution storage. This new outcome will open an interesting window for tuning NC sizes by the implementation of the described methodology and open a wide Fig. 4 Image by TEM of the synthesis of BaHfO3 and BaTiO3. In the range of further applications. case of (a) and (b) by adding 28 mmol and 111 mmol of water for synthesized BaHfO3. On the case of (c) and (d) by adding 28 and 111 mmol of water for synthesized BaTiO3. Conflicts of interest (Fig. 4c and d). The experimental data taken from XRD There are no conicts to declare. (Fig. S13†) corroborates the TEM results. A possible explanation for the difference of the titanium, can start from the behaviour of the alkoxide precursors on solution, Acknowledgements as well as their tendency to polymerise. The formation of coor- This work was supported by Spanish Ministry of Economy and dination polymers [M(OR)x]y (where “y” is the degree of poly- Competitiveness through the “Severo Ochoa” Programme for merisation) determines the nal crystal formation. The degree of Centres of Excellence in R&D (SEV-2015-0496), CONSOLIDER polymerisation has a direct dependency from the physical prop- Excellence Network (MAT2015-68994-REDC), COACH- erties of the metals (i) the radius of the metal (Ti < Zr, Hf), (ii) the SUPENERGY project (MAT2014-51778-C2-1-R, conanced by the coordination number and (iii) the size of group R on the alkoxide. European Regional Development Fund) and SUPERINKS project In our case, it seems that Hf and Zr are favourable to NC growth (RTC-2015-3640-3, conanced by the European Regional being the hydrolysis process that triggers the formation of –M–O– Development Fund). We also thank support from the European M– chains. While titanium cations exposed on the surface are Union for EUROTAPES project (FP7-NMP-Large-2011-280432) less favourable to the NC growth, resulting on invariable NC size and ULTRASUPERTAPE project (H2020 ERC-2014-ADG- aer water additions. Here, we need to consider the analogy 669504) and from the Catalan Government with 2014-SGR-753 between coordination chemistry and NC surface as the well- and Xarmae. The authors acknowledge the technical support known covalent bond classication (CBC).52,53 CBC is mainly of Servei de Microscópia, and Servei de Difracció de Raigs- X, all based in the direct comparison between coordination chemistry at the UAB, and Dr Judith Oró (Microscope Service) at the and the surface of NCs, in which each metallic cation centre ICMAB-CSIC, Dr Guillaume Sauthier for his support in XPS could be considered as a coordination complex. This hypothesis analysis. NCG acknowledge the PIF predoctoral fellowship from is supported by previous publications related to studies in coor- the Universitat Autònoma de Barcelona. dination chemistry in which the authors found signicant differences between the reactivity of titanium and zirconium alkoxides against hydrolysis and/or coordination.45,50,54,55 Conse- References quently, by controlling the hydrolysis step, different sizes of 1 M. Segev-Bar and H. Haick, ACS Nano, 2013, 7, 8366–8378. BaZrO3 and BaHfO3 NCs are produced. However, BaTiO3 NCs 2 L. E. Woodard, C. L. Dennis, J. A. Borchers, A. Attaluri, show same sizes corroborating the different behaviour of these E. Velarde, C. Dawidczyk, P. C. Searson, M. G. Pomper and three elements from the same group of the periodic table. R. Ivkov, Sci. Rep., 2018, 8, 1–13. 3 M. Mahmoudi, M. Yu, V. Serpooshan, J. C. Wu, R. Langer, Conclusions R. T. Lee, J. M. Karp and O. C. Farokhzad, Nat. Nanotechnol., 2017, 12, 845–855. In summary, we presented a dual-based synthetic approxima- 4 S. A. Bonke, A. M. Bond, L. Spiccia and A. N. Simonov, J. Am. tion to obtain perovskite BaMO3 NCs using a hybrid Chem. Soc., 2016, 138, 16095–16104. This journal is © The Royal Society of Chemistry 2020 RSC Adv., 2020, 10, 28872–28878 | 28877 RSC Advances Paper 5 Q. Jia, S. Ghoshal, J. Li, W. Liang, G. Meng, H. Che, S. Zhang, 31 X. Obradors, T. Puig, Z. Li, C. Pop, B. Mundet, N. Chamorro, Z. F. Ma and S. Mukerjee, J. Am. Chem. Soc., 2017, 139, 7893–7903. F. Vallés, M. Coll, S. Ricart, B. Vallejo, F. Pino, A. Palau, 6 A. Noori, M. F. El-Kady, M. S. Rahmanifar, R. B. Kaner and J. Gázquez, J. Ros and A. Usoskin, Supercond. Sci. Technol., M. F. Mousavi, Chem. Soc. Rev., 2019, 48, 1272–1341. 2018, 31, 044001. 7 R. R. Salunkhe, Y. V. Kaneti and Y. Yamauchi, ACS Nano, 32 L. Soler, J. Jareño, J. Banchewski, S. Rasi, N. Chamorro, 2017, 11, 5293–5308. R. Guzman, R. Yáñez, C. Mocuta, S. Ricart, J. Farjas, 8 M. E. Franke, T. J. Koplin and U. Simon, Small, 2006, 2, 36– P. Roura-Grabulosa, X. Obradors and T. Puig, Nat. 50. Commun., 2020, 11, 344. 9 S. Han, X. Liu, Y. Liu, Z. Xu, Y. Li, M. Hong, J. Luo and Z. Sun, 33 E. Solano, L. Perez-Mirabet, F. Martinez-Julian, R. Guzmán, J. Am. Chem. Soc., 2019, 141, 12470–12474. J. Arbiol, T. Puig, X. Obradors, R. Yañez, A. Pomar, 10 B.-W. Li, M. Osada, Y.-H. Kim, Y. Ebina, K. Akatsuka and S. Ricart and J. Ros, J. Nanopart. Res., 2012, 14, 1034. T. Sasaki, J. Am. Chem. Soc., 2017, 139, 10868–10874. 34 M. Han, Y. Rong, Q. Li, X. Xing and L. Kang, CrystEngComm, 11 T. Li, Y. Pan, Z. Wang, Y. Xia, Y. Chen and W. Huang, J. 2015, 17, 1944–1951. Mater. Chem. A, 2017, 5, 12602–12652. 35 T. Charoonsuk, W. Vittayakorn, N. Vittayakorn, P. Seeharaj 12 M. Niederberger, G. Garnweitner, N. Pinna and M. Antonietti, J. and S. Maensiri, Ceram. Int., 2015, 41, S87–S94. Am. Chem. Soc., 2004, 126, 9120–9126. 36 M. L. Moreira, J. Andrés, V. R. Mastelaro, J. A. Varela and 13 M. B. Smith, K. Page, T. Siegrist and A. Et, J. Am. Chem. Soc., E. Longo, CrystEngComm, 2011, 13, 5818–5824. 2008, 130, 6955–6963. 37 H. W. Lee, S. Moon, C. H. Choi and D. K. Kim, J. Am. Ceram. 14 D. Caruntu, T. Rostamzadeh, T. Costanzo, S. Salemizadeh Soc., 2012, 95, 2429–2434. Parizi and G. Caruntu, Nanoscale, 2015, 7, 12955–12969. 38 K. Kanie, Y. Seino, M. Matsubara, M. Nakaya and 15 S. O'Brien, L. Brus and C. B. Murray, J. Am. Chem. Soc., 2001, A. Muramatsu, New J. Chem., 2014, 38, 3548–3555. 123, 12085–12086. 39 R. Mouta, R. X. Silva and C. W. A. Paschoal, Acta Crystallogr., 16 K. De Keukeleere, J. Feys, M. Meire, J. De Roo, K. De Buysser, Sect. B: Struct. Sci., Cryst. Eng. Mater., 2013, 69, 439–445. P. Lommens and I. Van Driessche, J. Nanopart. Res., 2013, 15, 40 C. J. Bartel, C. Sutton, B. R. Goldsmith, R. Ouyang, 2074. C. B. Musgrave, L. M. Ghiringhelli and M. Scheffler, Sci. 17 J. De Roo, K. De Keukeleere, J. Feys, P. Lommens, Z. Hens Adv., 2019, 5, 1–10. and I. Van Driessche, J. Nanopart. Res., 2013, 15, 1778. 41 R. D. Shannon, Acta Crystallogr., Sect. B: Struct. Sci., Cryst. 18 Z. Lu, Y. Tang, L. Chen and Y. Li, J. Cryst. Growth, 2004, 266, Eng. Mater. A, 1976, 32, 751–767. 539–544. 42 H. Hayashi, T. Nakamura and T. Ebina, J. Phys. Chem. Solids, 19 J. Liu and J. C. Berg, J. Mater. Chem., 2007, 17, 4430–4435. 2013, 74, 957–962. 20 C. N. Chervin, B. J. Clapsaddle, H. W. Chiu, A. E. Gash, 43 G. Yang, Z. Yue, T. Sun, J. Zhao, Z. Yang and L. Li, Appl. Phys. J. H. Satcher and S. M. Kauzlarich, Chem. Mater., 2006, 18, A, 2008, 91, 119–125. 1928–1937. 44 J. Van Embden, A. S. R. Chesman and J. J. Jasieniak, Chem. 21 A. E. Danks, S. R. Hall and Z. Schnepp, Mater. Horiz., 2016, 3, Mater., 2015, 27, 2246–2285. 91–112. 45 U. Schubert and N. Hüsing, Synthesis of Inorganic Materials, 22 D. Gömpel, M. N. Tahir, M. Panthöfer, E. Mugnaioli, VCH-Wiley Verlag GmbH, Weinheim, 3rd edn, 2012. R. Brandscheid, U. Kolb and W. Tremel, J. Mater. Chem. A, 46 N. Y. Turova, E. P. Turevskaya, V. G. Kessler and 2014, 2, 8033–8040. M. I. Yanovskaya, The chemistry of Metal Alkoxide, Kluwe 23 M. Niederberger and G. Garnweitner, Chem. - Eur. J., 2006, Academi Publisher, 2002. 12, 7282–7302. 47 J. Martı́nez-Esaı́n, J. Faraudo, T. Puig, X. Obradors, J. Ros, 24 I. Bilecka and M. Niederberger, Nanoscale, 2010, 2, 1358– S. Ricart and R. Yáñez, J. Am. Chem. Soc., 2018, 140, 2127–2134. 1374. 48 A. Vioux, Chem. Mater., 1997, 9, 2292–2299. 25 I. Bilecka, L. Luo, I. Djerdj, M. D. Rossell, M. Jagodič, 49 D. C. Bradley, Chem. Rev., 2002, 89, 1317–1322. Z. Jagličić, Y. Masubuchi, S. Kikkawa and M. Niederberger, 50 N. Y. Turova, E. P. Turevskaya, V. G. Kessler and J. Phys. Chem. C, 2011, 115, 1484–1495. M. I. Yanovskaya, Hydrolysis of Metal Alkoxides and 26 K. De Keukeleere, J. De Roo, P. Lommens, J. C. Martins, Synthesis of Simple Oxides by The Sol-Gel Method, in The P. Van Der Voort and I. Van Driessche, Inorg. Chem., 2015, Chemistry of Metal Alkoxides, Springer, Boston, MA, 2002. 54, 3469–3476. 51 R. R. J. P. Corriu and D. Leclercq, Angew. Chem., Int. Ed. Engl., 27 M. Niederberger, Acc. Chem. Res., 2007, 40, 793–800. 1996, 35, 1420–1436. 28 N. Pinna, J. Mater. Chem., 2007, 17, 2769–2774. 52 J. Owen, Science, 2015, 347, 615–616. 29 N. Pinna, G. Garnweitner, M. Antonietti and M. Niederberger, J. 53 M. A. Boles, D. Ling, T. Hyeon and D. V. Talapin, Nat. Mater., Am. Chem. Soc., 2005, 127, 5608–5612. 2016, 15, 364. 30 Z. Li, M. Coll, B. Mundet, N. Chamorro, F. Vallès, A. Palau, 54 U. Schubert, E. Arpac, W. Glaubitt, A. Helmerich and J. Gazquez, S. Ricart, T. Puig and X. Obradors, Sci. Rep., C. Chau, Chem. Mater., 1992, 4, 291–295. 2019, 9, 1–14. 55 A. Kumar, S. De, A. G. Samuelson and E. D. Jemmis, Organometallics, 2008, 27, 955–960. 28878 | RSC Adv., 2020, 10, 28872–28878 This journal is © The Royal Society of Chemistry 2020