Polyol-Based Synthesis of Praseodymium Oxide Nanoparticles

A series of well-crystallized praseodymium oxide (Pr6O11) nanoparticles are synthesized via a highly scalable process from praseodymium nitrate. Praseodymium hydroxide is synthesized through a modified polyol process and subsequent calcination in air at 600°C for 2 h to transform the precursor material into Pr6O11 nanoparticles. A complete characterization of a specific sample by SEM, HRTEM, and X-ray diffraction (XRD) demonstrates that Pr6O11 nanoparticles form elementary blocks with a calculated size of 118 nm, containing slightly agglomerated nanocrystalline particles with a polycrystalline face-centred cubic phase and an individual size of around 10 nm. Particles can be easily dispersed in water through ultrasonification. Particles display a BET Specific Surface Area (SSA; 8.75 m2 g−1) with a nitrogen adsorption-desorption isotherm of type II and a helium density of powders (d = 5.66 g cm−3).


Introduction
Over the past few years, praseodymium oxide (Pr6O11) has been applied in different application domains such as sensors [1], high-temperature pigments [2], catalysts [3,4] and oxygen storage components of three-way automotive catalysts [5]. For such applications, nanosized materials are attractive because of the possibility to tune their physico-chemical properties in terms of shape, size and surface-to-volume ratio. Pr6O11 is conventionally prepared by solid-state reactions [6-8], a molten salt method [9], template-free precipitation [10], or sol-gel coupled with a shaping process such as electro-spinning [11]. A relatively recent paper has illustrated the possibility of preparing nanostructured Pr6O11 via two traditional methods (calcination of the nitrate and a sol-gel method) and two more sophisticated, modern techniques (citric method and polymerized precursor route/modified Pechini method) [12]. The praseodymium precursor used as a starting material for the synthesis in all cases was praseodymium nitrate Pr(NO3)3.6H2O. As far as we know, there is a very limited number of methods that have been reported in the literature for the synthesis of Pr6O11 nanopowders, although such nanomaterials are expected to provide access to a wider range of applications in set-up devices such as sensors or dispersed in porous ceramic supports to act as a catalyst. A major reason for this is that a reliable shape-, size-and phase-controlled synthesis process leading to the synthesis of nanosized Pr6O11 powders is still not available. Usually, direct heattreatment of praseodymium nitrate hexahydrate Pr(NO3)3·6H2O leads to micron-sized Pr6O11 powders. In addition, both direct precipitation of Pr(NO3)3·6H2O into praseodymium hydroxide Pr(OH)3 nanoparticles and synthesis of Pr(OH)3 nanoparticles in reverse micelles followed by heat-treatments lead to loss of nanoparticle cohesion, i.e., formation of nanorods. Within this context, developing a new and simple method for synthesis of Pr6O11 particles on the nanoscale could thus open up new avenues for uncovering their remarkable properties and for achieving useful applications.
Inorganic nanomaterials can, in principle, be synthesized using both top-down and bottom-up techniques. The bottom-up strategy uses the techniques of molecular synthesis [13], colloid chemistry [14], polymer science [15], and related areas to make structures with nanometre dimensions. The top-down strategy [16] uses miniaturizing techniques such as machining, templating or lithography to pattern materials. Self-assembly [17][18][19] bridges these two techniques and allows materials to be designed with hierarchical order and complexity. The use of organic surfactants and polymers in self-assembled nanoparticle synthesis has been a popular method of achieving morphological control, where the polymer or foreign ions typically work to regulate and stabilize the surface of nanoparticles. One of the most common and easy approaches that utilize this strategy is the polyol method [20].
Polyol synthesis is a chemical method involving forced hydrolysis of transition metal salts, which allows metal or metal oxide nanoparticles to be prepared with controlled size, texture and shaping [21][22][23][24][25][26]. Initial work on elemental metals or alloys exploited the reducing properties of a high-boiling alcohol (glycerol, glycol) towards a suitable metal precursor. Owing to their high dielectric constants, the alcohol acted as a solvent able to dissolve inorganic compounds, a growth medium and a complexing agent in some cases. Furthermore, owing to relatively high boiling points, such alcohol offered a wide operating-temperature range (from 25°C to the boiling point) for preparing inorganic compounds. It also often acted as a stabilizing agent that effectively limited particle growth and hindered inter-particle aggregation.
The polyol method holds promise as a low-cost, highyield technique with a wide range of scientific and technological applications, which is ideal for potential industrial scale-up [27,28]. In the case of oxide synthesis, the polyol method can be considered as a sol-gel process carried out at relatively elevated temperature with accurate control of particle growth. It has been studied and reported for the synthesis of a wide range of oxide sub-micrometre particles, including Y2O3, ZnO, Mn3O4, CoTiO3, VxOy, TiO2, PbO and SnO2 [29][30][31][32][33][34][35][36].
Ethylene glycol is among the most widely used solvents for the polyol synthesis of metal oxide nanoparticles due to its strong reducing capability, relatively high boiling point and high dielectric constant, which increases the solubility of inorganic salts [37]. Ethylene glycol is commonly used as a crosslinking reagent because of its propensity to coordinate with the central metal ion and form a metal glycolate leading to subsequent oligomerization [38]. It has also been demonstrated that as-synthesized glycolate precursors can later be converted to their more common metal oxide derivatives when calcined in air, while maintaining the original precursor morphology [35][36][37][38][39].
In this paper, we describe a fast, convenient and green synthetic route for the large-scale synthesis of praseodymium oxide nanoparticles with an individual size of approximately of 10 nm by polyol synthesis from praseodymium nitrate in diethylene glycol (DEG), and a subsequent addition of sodium hydroxide followed by calcination in air at 600°C. Thermogravimetric analysis and X-ray diffraction are combined to discuss the effect of both the nature of the Pr6O11 precursor and the quantity of sodium hydroxide added to the praseodymium nitrate-DEG mixture; electron microscopy, EDS, and BET studies were used to characterize the structure, crystallinity, morphology, chemical composition and texture of the Pr6O11 nanoparticles. Also, we note that our method provides nanoparticles that are stable and easily dispersed in water, and could be adapted to the preparation of coatings or porous materials.

Experimental procedure
All chemicals were purchase from Sigma Aldrich and used as received. The polyol method involves the reduction of a metal salt precursor by a polyol, a compound containing multiple hydroxyl groups. The polyol used in this synthesis, diethylene glycol ((HOCH2CH2)2O-DEG, 99%), served as both reducing agent and solvent. 2.5 g of praseodymium nitrate (5.75 mmol, Pr(NO3)3.xH2O, 99 %) was stirred in 50 mL of diethylene glycol for 30 min in a Schlenk flask (in air). The as-prepared solution was then treated at 140°C (in air). Sodium hydroxide (NaOH, with a controlled concentration in water) was added after a dwelling time of 1 h, and then the mixture was directly heated up to 180°C and kept at this temperature for 4 h, after which it was cooled naturally.

B-OH/B-Ox 10 9
C-OH/C-Ox 20 9 Following centrifugation, precipitation isolation, washing several times with ethanol, acetone then distilled water, and drying, as-obtained samples labelled X-OH (with X = A, B, C, D and E) were annealed at 600°C for 2 h in air (ThermoConcept, Germany) to form the corresponding oxide labelled X-Ox. Table 1 reports the quantities and concentrations of NaOH solution added during the synthesis of the different hydroxide and oxide samples.

E-OH/E-Ox
We compared the effect of the praseodymium oxide precursor on the (nano)structure and morphology of the final materials. Following the procedure described above, we investigated the synthesis of praseodymium oxide nanoparticles from praseodymium chloride (PrCl3, 99 %) leading to the pre-(A'-OH) and post-calcined (A'-Ox) samples.
Thermogravimetric analysis (Hi-Res TGA 2950, TA Instruments,) of Pr(OH)3 was performed with the sample E-OH under the following conditions: sample mass 14 mg, platinum crucible of 100 µl, heating rate 5°C min -1 , range 25-600°C, and an atmosphere of air (60 ml min -1 ). The phase and crystallite size of the Pr6O11 nanoparticles were studied using X-ray diffraction (XRD, PANalytical X'pert Pro MPD powder diffractometer with a detector X'celerator; CuKα radiation (λ1 = 1.5406 Å, λ2 =1.5444); the sample was scanned with CuKα radiation within the range of 10°-60° (2θ), with a step size of 0.017° and an exposure time of 50 s. The accelerating voltage and current were 40 kV and 30 mA, respectively. Phase identification was achieved by matching the patterns against stick patterns from the JCPDS database. A semiquantitative estimation of the extent of crystallization in samples was achieved by calculating the integrated intensity of the (111) diffraction peak. The (111) diffraction peak was chosen for this purpose, based on two considerations: freedom from peak overlap and sufficient relative intensity. Special sample holders were used to irradiate a constant volume of sample powder to enable comparison of the integrated intensities of different samples. Crystallite sizes of Pr6O11 were calculated from the FWHM of the above diffraction peak using the Scherrer formula. FWHM values were corrected for instrumental line broadening. The surface morphology of the nanoparticles was examined using scanning electron microscopy (SEM, Hitachi S4800) equipped with energy-dispersive spectroscopy analysis (EDS) to obtain information on the praseodymium and oxygen contents of the material. Transmission electron microscopy (TEM) was carried out on the sample E-Ox using a JEOL 1200 EXII operating at 200 kV. A solution of ethanol containing Pr6011 particles was prepared via ultrasonic bath. Samples for TEM were prepared by spreading a drop of this solution on copper grids coated with a carbon film followed by evaporation under ambient conditions. The nitrogen adsorption/desorption isotherms of the sample E-Ox were measured on a Sorptomatic 1900 analyser (Fisons). Before adsorption measurements, all samples were out-gassed for 4 h at 150°C in the degas port of the adsorption analyser. The Brunauer-Emmett-Teller (BET) method was used to calculate the specific surface area.

Results and discussion
The reduction ability of reagents is critical in the control of the size and morphology of nanostructures. Herein, a modified polyol process was designed to synthesize Pr6O11 nanoparticles. The Pr6O11 precursor (Pr(NO3)3 or PrCl3) was mixed with DEG ((HOCH2CH2)2O) to be heated at 140°C before addition of controlled quantities and concentrations of NaOH and subsequent heating to 180°C. After a dwelling time of 4 h at 180°C, the mixture was cooled down to be calcined at 600°C in air.  The as-synthesized suspension from Pr(NO3)3 can be used as-is for coating-based processes such as spray-/dip-or spin-coating, or to infiltrate inorganic templates such as ordered mesoporous powders or materials with hierarchical porosity such as those we prepared in our group [40][41][42][43][44][45][46]. In the present paper, we obtained precalcinated powders by centrifugation, precipitation isolation, washing several times with ethanol, acetone and distilled water and drying, then additional heating to 600°C to obtain the required oxide nanoparticles (Figure 1).
In our process, the polarity of polyol offers the ability to dissolve praseodymium nitrate. As a high-boiling-point solvent, polyol provides a high-temperature environment and improves the reactivity of various reactants. In this work, DEG acted both as the solvent, as a reducing agent and as a growth medium. NaOH provided the hydroxyl ion to form Pr(OH)3, as reported in Figure 2, which presents the XRD patterns of samples X-OH. Phase identification is achieved by locating the characteristic diffraction peaks of the respective phases in the XRD patterns. Prior to calcination, the phase was readily indexed for each sample to the hexagonal structure of praseodymium hydroxide consistent with the standard powder diffraction file of Pr(OH)3 (JCPDS 83-2304) (Figure 2) with lattice parameters a = 0.6505 nm and c = 0.3428 nm, with no effect of the NaOH concentration and quantity on the peak numbers and positions.
During the subsequent calcination step, the removal of gaseous by-products induced a weight loss of 75% after thermal decomposition at 600°C (see supporting information, Figure.  There are no impurity phases of hydroxides and other oxides. The transition from the hydroxide precursor to crystalline Pr6O11 was also noted, the sample changing colour from light green (pre-calcined) to dark brown (post-calcined). However, the XRD patterns of samples B-Ox and C-Ox showed a diffraction peak around 2θ = 31°. This peak can be assigned to sodium praseodymium oxide (NaPrO2). EDX of the samples confirmed the presence of sodium. As reported in Table 1, these samples were prepared with the highest quantity of NaOH. This indicates that the quantity of NaOH needs to be controlled to obtain pure Pr6O11. Comparing the XRD patterns of samples A-Ox, D-Ox, and E-Ox, it is clear that the increased concentration of NaOH added during the synthesis caused sharpening of the diffraction peaks, the result of an increase in the crystallite sizes. Having concluded that both the quantity and concentration of NaOH has a strong effect on crystallization, it would be of significant importance to study the effect of these parameters on crystallite size on the samples A-Ox, D-Ox, and E-Ox. The crystallite size was determined by measuring the full width at half the maximum in conjunction with the Debye-Scherer formula. The values for crystallite size are reported in Table 2.

Table 2. Calculated crystallite size in samples A-Ox, D-Ox and E-Ox
This clearly indicates the increase in crystallite size along with the concentration of NaOH used during the synthesis.
There are no impurities detected by energy-dispersive X-ray spectroscopy (EDS) analysis (see supporting information, Figure 3SI) except for carbon elements arising from the carbon film in SEM observations.

OH/A-Ox, D-OH/D-Ox and E-OH/E-Ox.
Pre-calcined nanostructures are mainly in the form of nanorods and are transformed by calcination at 600°C into nanoparticles. This means that the structure of Pr(OH)3 collapsed during heat-treatment to 600°C, leading to the formation of spherical Pr6O11 nanostructures, most probably due to the fact that it represented the most thermodynamically stable structure. The collected SEM micrographs of samples A-Ox, D-Ox and E-Ox showed that the particle population is relatively homogeneous, including small round-shaped particles and an extremely low proportion of particles of large size. The sample E-Ox displays the smallest individual particle sizes. The high-magnification SEM image of the sample E-Ox (Figure 4) shows that the sample was composed of nanoparticles aggregated in clusters. SEM is not really suited to observing nanosized materials, but it provides a view of the general particle population. Therefore, it is difficult to estimate the size of the individual particles.

Figure 4. SEM micrographs of the sample E-Ox
We focused our study of the sample E-Ox. The TEM micrograph of the sample E-Ox was investigated to assess the morphology and the overall particle size of samples. A typical bright-field TEM image of the sample E-Ox is shown in Figure 5. The micrograph shows that the obtained sample formed elementary blocks composed of a large amount of spherical-shaped nanoparticles with an average diameter of 9.4 ± 0.9 nm, as shown on the statistical histogram deduced from the TEM image ( Figure 6). The corresponding selected area electron diffraction (SAED) pattern, inset in Figure 5, confirmed the identification of the face-centred cubic phase of Pr6O11. We investigated HRTEM to refine the structural information and to corroborate the SAED results. Figure 5 100 nm Caroline Quievryn, Samuel Bernard and Philippe Miele: Polyol-based Synthesis of Praseodymium Oxide Nanoparticles shows a high-resolution TEM (HRTEM) image used to gather information on powder nanostructure. The HRTEM image of the particle core demonstrates that the specimen consists of very fine nanocrystalline Pr6O11; the observed interplanar distance is d111 = 0.313 nm and the growth direction is along the <111>. The sample E-Ox displayed a helium density of 5.66 g cm -3 and a BET specific surface area of 8.75 m 2 g -1 . A similar value is commonly found for nanosized powders [46][47][48]. Figure 7 shows the nitrogen adsorption-desorption isotherm, which is of type II according to the IUPAC classification [50,51], with an evident hysteresis loop in the 0.9 to 1.0 range.

Conclusions
In conclusion, this is a highly scalable and reproducible process for the synthesis of a series of high-purity and well-crystallized Pr6O11 nanoparticles. We have described polyol synthesis from praseodymium nitrate in diethylene glycol (DEG) and the subsequent addition of sodium hydroxide, followed by calcination in air at 600°C to convert into praseodymium oxide (Pr6O11). The morphological and textural analyses of a specific sample showed that the sample formed elementary blocks with a calculated average aggregate size of 118 nm based on the BET Specific Surface Area (SSA; 8.75 m² g -1 ) and the helium density of powders (d = 5.66 g cm -3 ). These elementary blocks are composed of spherical-shaped nanoparticles connected to each other with individual sizes of around 10 nm. The Pr6O11 nanoparticles possess a polycrystalline face-centred cubic phase.