hydroxides in solution – An insight on the phase transformation mechanisms

Materials Chemistry and Physics xxx (2015) 1e11

Contents lists available at ScienceDirect

Materials Chemistry and Physics journal homepage: www.elsevier.com/locate/matchemphys

Controlled phase formation of nanocrystalline iron oxides/hydroxides in solution e An insight on the phase transformation mechanisms
nio Portugal a, Luisa Dura ~es a, * Jorge Pedrosa a, Benilde F.O. Costa b, Anto a b

CIEPQPF, Department of Chemical Engineering, Faculty of Sciences and Technology, University of Coimbra, Rua Sílvio Lima, 3030-790 Coimbra, Portugal CFisUC, Department of Physics, Faculty of Sciences and Technology, University of Coimbra, Rua Larga, P-3004-516 Coimbra, Portugal

h i g h l i g h t s

g r a p h i c a l a b s t r a c t


Study of the effect of pH and Fe(II)/ Fe(III) ratio in iron oxides/hydroxides.
Iron(II) proves to have a catalytic effect.
pH of 4 inhibits the catalytic effect of iron(II).
pH of 7 and 12 promote phase transformation.
The predominant mechanism is defined by the amount of Fe(II).

a r t i c l e i n f o

a b s t r a c t

Article history: Received 24 October 2014 Received in revised form 8 May 2015 Accepted 5 July 2015 Available online xxx

In this work, the effect of pH in the formation of iron oxides/hydroxides obtained by chemical precipitation, as well as the influence of the presence of Fe(II) in the phases formation process were studied. Iron(III) chloride nonahydrated and iron(II) chloride tetrahydrated were used as precursors and sodium hydroxide as a base to adjust the pH and to promote the hydrolysis and condensation reactions. Syntheses were performed at three pH values e 4, 7 and 12 e and two Fe(II)/Fe(III) molar ratios e 0.02 and 0.5. The obtained phases were observed by SEM, showing agglomerates of nanometric crystallites ranging from ~1 to ~15 nm in size. The iron oxides/hydroxides identification/quantification was per€ ssbauer spectroscopy. It was concluded that at low pH formed by the combined use of FTIR, XRD and Mo the catalytic effect of Fe(II) in the transformation of ferrihydrite is inhibited independently of the Fe(II) amount. In alkaline medium and at low concentration of Fe(II), goethite was formed. A Fe(II)/Fe(III) ratio of 0.5, for pH of 7 or 12, led to the simultaneous formation of magnetite and goethite. Thus, the extent of the transformation mechanisms e topotactic and reconstructive e is strongly influenced by the pH and the Fe(II)/Fe(III) ratio. In alkaline conditions, the predominant mechanism is defined by the ratio of Fe(II)/ Fe(III) e a higher ratio favors the topotatic transformation of ferrihydrite in magnetite, while a low ratio leads to a dominant reconstructive mechanism. © 2015 Elsevier B.V. All rights reserved.

Keywords: Magnetic materials Nanostructures Chemical synthesis Phase transition Crystal growth Mossbauer spectroscopy

1. Introduction Iron oxides have been used by cavemen as coloring agents.

* Corresponding author. ~es). E-mail address: [email protected] (L. Dura

Today, after thousands years of progress and technology development, multiple advanced applications have emerged, due to the unique catalytic, magnetic, sorption, and optical properties, among other characteristics, exhibited by the iron oxide based materials. Overviews of the broad applicability given to iron oxides, which can range from industrial, environmental, biological and medical fields, are given by Cornell and Schwertmann [1], Machala et al. [2], Faraji

http://dx.doi.org/10.1016/j.matchemphys.2015.07.018 0254-0584/© 2015 Elsevier B.V. All rights reserved.

Please cite this article in press as: J. Pedrosa, et al., Controlled phase formation of nanocrystalline iron oxides/hydroxides in solution e An insight on the phase transformation mechanisms, Materials Chemistry and Physics (2015), http://dx.doi.org/10.1016/j.matchemphys.2015.07.018

2

J. Pedrosa et al. / Materials Chemistry and Physics xxx (2015) 1e11

et al. [3] and Mohapatra and Anand [4]. Iron oxides/hydroxides are an important group of minerals. In the majority of these compounds, iron is in a trivalent state of oxidation. Exceptions are FeO and Fe(OH)2, in which iron is present exclusively in a divalent state. A mixture of both states of oxidation (Fe(III) e Fe(II)), appears in other compounds namely Fe3O4 [1]. Iron oxides/hydroxides are usually crystalline, being schwertmanite and ferrihydrite exceptions with a low degree of crystallinity [4]. The most unstable form of iron oxide is ferrihydrite. Due to its metastable nature, it can be transformed into more thermodynamically stable species [5], being thus an important precursor for iron oxides of higher crystallinity. Ferrihydrite particles can be generally prepared by adding alkali to a Fe(III) salt solution. Nevertheless, the synthesis method is quite sensitive to the process conditions, such as the type and concentration of the iron salt, pH, temperature, and aging time. Consequently, ferrihydrite with different degrees of crystallinity is obtained. These different degrees can be interpretated by XRD patterns, and can vary from two to six lines, corresponding the two lines to the less ordered structure [6]. When freshly precipitated, ferrihydrite exhibits a very small size, typically a few nanometers in diameter. Accordingly, it is highly reactive due to its high surface area [7]. Synthesis of iron oxides tailored for various applications has been a challenging area. The ceramic method, based on the direct reaction of powder mixtures, has been one of the most widely used techniques for the production of iron oxides. This type of synthesis requires high temperatures and small particle sizes of the raw materials. However, this method is a rather crude approach, compared to other new methods, like wet chemistry methods [8]. The wet approach allows a better size and composition control of the final products. However, even with a careful choice/control of conditions (pH, concentrations, temperature, mixing method, etc.), the obtained phases morphology still depends strongly on the competition between nucleation, growth, aggregation and adsorption of impurities. In many cases, the synthesis is carried out by the transformation of other iron oxide precursor particles, due to the impossibility to directly precipitate a specific iron phase [4]. In these last few years, some studies have been published about the transformation of ferrihydrite in more stable phases, especially from Liu and co-workers. Liu et al. [5,9e14], found that the presence of iron(II) can clearly accelerate the transformation of ferrihydrite into different iron phases. High pH and fast heating rates favor the topotactic transformation of ferrihydrite in solution as well as the formation of hematite, while a low pH and low heating rates favor the dissolution/re-precipitation mechanism of ferrihydrite as well as the formation of goethite [11]. When the pH of a mixture of ferrihydrite and Fe(II) ions is raised from 5 to 9, the ferrihydrite is rapidly transformed into hematite particles at 100  C. The adsorption of Fe(II) and electron transfer between Fe(II) and Fe(III) accelerates the dissolution/reprecipitation mechanism and the topotactic transformation. It is thought that Fe(II), in the forms of FeOHþ and Fe(OH)2, plays the role of a catalyst [9]. Saric et al. [15] have concluded that iron oxides/hydroxides obtained by precipitation in FeCl3 solutions were strongly dependent on the experimental conditions. Powders obtained by precipitation with NaOH at pH 4.2 and 4.4 and aging at 90  C contained a-Fe2O3 as a single phase. When the pH of the system increased to 6.5, the precipitation of a-FeOOH particles was observed. In this work, a systematic study of the effect of pH in the phases of iron oxides/hydroxides obtained by chemical precipitation was performed. Iron(III) chloride nonahydrated was used as precursor, water as solvent, and sodium hydroxide as a base to promote the

precipitation reactions and adjust the pH. In addition, the effect of adding to the system different amounts of Fe(II) was also studied. Iron(II) chloride tetrahydrated was used either as a source of iron(II) ions for the iron oxide phase and to test its catalytic effect. The syntheses were carried out at three different pH levels (acid, neutral, alkaline) at two different molar ratios of iron(II)/iron(III) (0.02 and 0.5). A detailed description of the chemical and growth mechanisms involved in the synthesis of iron oxides/hydroxides, and particularly the effect of the presence of Fe(II) in the transformation of phases, is also presented based on a review of the open literature (Section 2). The materials obtained in this work were characterized by several chemical and structural analytical techniques in order to identify the formed iron phases and to discuss the probable mechanisms of phases' transformation/growth. In this context, the € ssbauer spectroscopy was of great value and helped to clarify Mo the samples composition in terms of iron phases and their amounts. For the first time in this context, the phase conversion was quantified giving a clear indication of the predominant transformation mechanisms. This work is a contribution for an improved understanding and control of the iron oxides/hydroxides formation by chemical precipitation, in order to allow the tailoring of the produced phases for subsequent applications.

2. Overview on the phenomena of the iron oxides/hydroxides growth and chemical transformation The hydrolysis and condensation chemistry of transition metals compounds in solution to produce metal oxides/hydroxides is well documented in the literature, such as in Livage et al. [16] and Jolivet [17]. Metal alkoxides can be used as precursors, but many of the alkoxides are difficult to obtain, expensive and sensitive to moisture, heat and light, making long term storage difficult. Metal salts are cheaper and easier to handle than metal alkoxides [18]. A metal salt, when in aqueous solution, converts into ionic species and the cations (MZþ) suffer solvation by water molecules according to:

MZþ þ H2 O/½M  OH2 Zþ

(1)

[M(OH2)n]Zþ species are formed, where the oxygen atoms of water are directed toward the metal cation (Livage, 1993). For example, iron cations form hexacoordinated aquo complexes e [Fe(OH2)6]Zþ [19]. Water behaves as a Lewis base, and the solvation process leads to the formation of a partially covalent bond, involving some electron transfer from the water molecule to the metal cation. This transference increases the positive partial charge on hydrogen atoms, weakening the OeH bond and, as a consequence, the water molecule becomes more acidic. According to the magnitude of the electron transfer, a deprotonation can take place as follows:

½M  OH2 Zþ 4½M  OHðZ1Þþ þ Hþ 4½M  OðZ2Þþ þ 2Hþ (2) In a noncomplexing aqueous media, a whole set of more or less deprotonated species can be present. Three kinds of ligands can be identified, an aquo ligand (OH2), a hydroxo ligand (OH) and an oxo ligand (O). The formula of any inorganic precursor after hydrolysis can be summarized as [MONH2Nh](Zh)þ, being N the coordination number and h the hydrolysis ratio. The hydrolysis ratio depends on the pH of the solution and the oxidation state of the metal cation. This dependence can be represented in a charge-pH diagram (Fig. 1).

Please cite this article in press as: J. Pedrosa, et al., Controlled phase formation of nanocrystalline iron oxides/hydroxides in solution e An insight on the phase transformation mechanisms, Materials Chemistry and Physics (2015), http://dx.doi.org/10.1016/j.matchemphys.2015.07.018

J. Pedrosa et al. / Materials Chemistry and Physics xxx (2015) 1e11

Fig. 1. Charge-pH diagram (adapted from Livage et al. [16]).

As shown in Fig. 1, low-valent cations (Z < 4), like iron(III) and iron(II), give rise to aquo complexes. An increase in pH leads to the formation of aquohydroxo and/or hydroxo complexes. The existence of MOH bonds promotes the condensation of the inorganic precursors [17]. After iron(III) hydrolysis, the hydroxylated complexes can condense by two simple distinct mechanisms: i) oxolation, with elimination of water and formation of hydroxo bridges between iron(III) atoms (reactions 3 and 4); ii) oxolation, with elimination of water and formation of oxo bridges between iron(III) atoms (reaction 5) [16].



2þ FeðOHÞðOH2 Þ5  3þ  5þ þ FeðOH2 Þ6 4 ðOH2 Þ5 FeðOHÞFeðOH2 Þ5 þ H2 O

occurs on the Fe(III) phase (Fig. 2 e step II) and, subsequently, the electron transfer between Fe(II) and interfacial Fe(III) takes place (Fig. 2 e step III). This electron transfer promotes a solid-state transformation (Fig. 2 e step IVa), in the case of the topotactic mechanism. On the other hand, it can also promote a reductive dissolution (Fig. 2 e step IVb), followed by condensation, dehydration and growth of the new phase (Fig. 2 e steps V,VI,VII). The topotactic transformation occurs within the solid phase, involving an internal rearrangement of the crystal. Due to the low mobility of the crystal network, a tridimensional correspondence between the initial and final structures is required [1]. The reconstructive transformation involves a dissolution/reprecipitation mechanism, therefore no structural relationship is required between the precursor species and the final product. Such mechanism depends on the solubility and dissolution rate of the precursor species. In the dissolution process, Fe(III) forms monomers in the solution and, with time, these monomers condense to form the new phase, as shown in the following reaction scheme [10]: dissolution

þ

Ferrihydrite ƒƒƒƒƒ! FeðOHÞ2

inorganic polymerization

 FeðOHÞ2þ ƒƒƒƒƒƒƒƒƒƒƒƒƒ! Product

(6)

3. Materials and methods 3.1. Synthesis of iron oxides/hydroxides

(3)

 2þ  4þ 2 FeðOHÞðOH2 Þ5 4 ðOH2 Þ4 FeðOHÞ2 FeðOH2 Þ4 þ 2H2 O (4)  2þ  4þ 2 FeðOHÞðOH2 Þ5 4 ðOH2 Þ5 FeOFeðOH2 Þ5 þ H2 O

3

(5)

For ferric complexes, condensation occurs from strongly acidic media (pH  1), by addition of a base (like NaOH) at room temperature, and leads almost instantly to ferrihydrite forms. The high reactivity of ferric species and the impossibility to stop the process at a molecular level makes the quantification of the proportion of hydroxo and oxo bridges in the material impossible [19]. Due to the poor structural organization of ferrihydrite, it ends up to being converted into more stable phases, transformation that is favored by its high surface area [7,20]. The conversion of ferrihydrite is slow at the catalyst absence, but its transformation can be accelerated by the addition of strong reductants [11]. Some additives, like Fe2þ, due to its strong reducing capacity, can induce the transformation of ferrihydrite into other phases [1]. Fe2þ can be added as an iron(II) salt and its reducing ability is stronger when sorbed on the ferric oxide in opposition to the dissolved state [21]. When dissolved, and depending on the pH value, Fe(II) hydrolyzes and forms new species, like Fe2þ, Fe(OH)þ, Fe(OH)2 and FeðOHÞ 3 [11]. The formed iron(II) complexes can then adsorb on the surface of the iron(III) oxide, weakening the FeeO bonds to neighboring atoms and leading to the detachment of iron(III) complexes to the solution. The transformation of ferrihydrite can occur by two different mechanisms, topotactic or reconstructive [9], and both involve an electron transfer between the Fe(II) and Fe(III) ions. The two processes are competitive and strongly dependent on the conditions of the reaction medium. Initially, the adsorption of the Fe(II) species

Ferric chloride hexahydrated (FeCl3.6H2O, 97% pure, SigmaeAldrich), ferrous chloride tetrahydrated (FeCl2.4H2O, 99% pure, SigmaeAldrich) and sodium hydroxide (NaOH, 98% pure, Panreac) were used as received, and double distilled water was produced in the laboratory. Before the synthesis, the following iron and sodium hydroxide solutions were prepared: 100 mL of 6 M sodium hydroxide solution, 25 mL of 1 M Fe(III) solution and 5 mL of 0.1 or 2.5 M Fe(II) solution, depending on the desired molar ratios of iron(II)/iron(III) required, i.e. 0.02 or 0.50, respectively. The ferric solution displayed a reddish brown color, and the ferrous solution exhibited a pale green color. The synthesis batches were carried out under nitrogen gas atmosphere (1 bar) and double distilled water was used in all the stages. The syntheses were performed in a glass reaction vessel (100 mL, Duran Group) with a head with four inlet necks, placed in a thermostatic bath at 25  C. 40 mL of water were added to this vessel and nitrogen gas was bubbled at the bottom of the vessel using a thin tube. The nitrogen flow rate was sufficient to allow for vigorous stirring of the reactor content. Temperature and pH sensors (pH 213 e Microprocessor pH meter, from HANNA Instruments) were placed in two of the inlet necks. The only remaining free inlet was used for the reagents addition. Firstly, the previously prepared Fe(III) solution was added to the water and then the sodium hydroxide solution was slowly added until the desired pH, according to Table 1. With the addition of the alkali solution, the formation of dispersed brownish flakes instantaneously occurred. Then, the Fe(II) solution (iron(II)/iron(III) ¼ 0.02 or 0.5) was added to the above mixture. The pH of the system was again adjusted to the chosen pH, and the total reaction volume adjusted to 100 mL with water. The formed slurry was left for 2.5 h in the reactor and then transferred to two falcon tubes (50 mL each) and washed three times with double distilled water using a centrifuge (5000 rpm for 10 min in each washing batch). The products were dried in an oven for three days, at 75  C, and then placed in a desiccator in nitrogen atmosphere. The abovementioned drying period was adjusted for

Please cite this article in press as: J. Pedrosa, et al., Controlled phase formation of nanocrystalline iron oxides/hydroxides in solution e An insight on the phase transformation mechanisms, Materials Chemistry and Physics (2015), http://dx.doi.org/10.1016/j.matchemphys.2015.07.018

4

J. Pedrosa et al. / Materials Chemistry and Physics xxx (2015) 1e11

Fig. 2. Topotactic and reconstructive transformation of ferrihydrite (Adapted from Liu et al. [9]).

Table 1 Samples identification according to the used synthesis conditions. pH variation after pH adjustment. Sample

pH

S1 S2 S3 S4 S5 S6

4 7 12 4 7 12

Iron(II)/iron(III) molar ratio 0.02

0.50

Initial pH

Final pH

DpH

4.1 7.4 12.3 4.3 7.2 12.2

3.8 7.0 12.0 4.1 6.4 11.7

0.3 0.4 0.3 0.3 0.8 0.5

constant sample weight. After the drying step, the material was rigid and fragmented. The final samples were composed by the mix of the material obtained in two replicas and designated as S1, S2, S3, S4, S5 and S6 e Table 1. 3.2. Materials characterization The six samples described in Table 1 were characterized by several techniques. A portion of each sample was milled in a mortar until a finely divided powder was obtained. This powder was used for samples analyzed by Fourier transform infrared spectroscopy, €ssbauer spectroscopy analyses. For the X-ray diffraction and Mo scanning electron microscopy, the material was analyzed as obtained after drying. 3.2.1. Fourier Transform Infrared Spectroscopy (FTIR) The FTIR analyses were performed to identify the functional groups existing in the chemical structure of the synthesized iron oxides/hydroxides. The infrared spectra of the materials were obtained using a FT/IR-4200 spectrometer from Jasco. The used wave number range was 400e4000 cm1. For these analyses, it was necessary to prepare circular and thin KBr pellets with the sample material. Thus 0.3 mg of sample was mixed with 80 mg of KBr and milled together in a small mortar. The KBr was previously milled and maintained in an oven at 60  C for 2 days. The mixture of KBr and iron oxides/hydroxides was also dried in the oven at 60  C for 1 day to avoid the interference of air humidity in the spectra. Just before each analysis, the prepared mixture was pressed in a Qwik

Handi-Press to obtain homogeneous and transparent pellets with 7 mm of diameter. The pellets were then placed in the equipment sample holder for analysis. 3.2.2. X-ray diffraction (XRD) X-ray diffraction patterns of the samples were obtained by means of a Philips X'Pert diffractometer, operating in BraggBrentano geometry (q-2q), using CoKa radiation (l ¼ 1.78897 Å) and an applied voltage and current of 40 kV and 35 mA, respectively. The scans were performed at room temperature and collected over the range of 5 e100 , with the step of 0.025 and the recording time of 0.5 s per step. The phases' identification was done using the software X'Pert Graphics & Identify and the ICDD (International Centre for Diffraction Data) database. The initial treatment of each diffractogram was performed using the Origin 8.6 software (OriginLab), by defining a baseline, smoothing the curves and adjusting a Voigt function to the data. After the peaks fitting, the program automatically provides the b value (FWHM e full width at half maximum) of each peak. It was then possible to estimate the crystallites size, t, of the obtained phases by using the Scherrer equation (Equation (7)):

t ¼ Kl=ðb cos qÞ

(7)

k is the shape factor (usually equal to 0.9), l is the radiation wavelength (Å), b corresponds to the full width at half maximum (FWHM) in radians and q is the Bragg angle (degrees). The unit cell dimensions of indexed iron phases were calculated by the equations presented by Smart and Moore [22]. €ssbauer Spectroscopy 3.2.3. Mo € ssbauer spectra were recorded in a WissEL spectrometer, at Mo room temperature, using a source of 57Co atoms in a matrix of Rh with an activity of 10 mCi. A foil of a-Fe was used as standard to obtain the reference value for the speed. The spectra were fitted by a set of Lorentzian lines determined by the least squares method. 3.2.4. Scanning electron microscopy (SEM) SEM micrographs were taken with a JSM-5310 electron microscope, from JEOL, operating at an accelerating voltage of 20 kV. A small portion of each sample was glued in a carbon tape to a

Please cite this article in press as: J. Pedrosa, et al., Controlled phase formation of nanocrystalline iron oxides/hydroxides in solution e An insight on the phase transformation mechanisms, Materials Chemistry and Physics (2015), http://dx.doi.org/10.1016/j.matchemphys.2015.07.018

J. Pedrosa et al. / Materials Chemistry and Physics xxx (2015) 1e11

5

metallic wafer (sample holder) and then coated with a thin layer of gold, using physical vapor deposition for 30 s. 4. Results and discussion 4.1. pH and temperature evolution during the syntheses The pH and temperature changes were recorded for all the syntheses. Fig. 3 shows the pH and temperature variation during the synthesis of sample S1. It is possible to verify that the initial ferric solution (time ¼ 0 s) has an acidic pH of approximately 1. From 0 s up to 1000 s, the addition of the ferrous solution as well as the pH adjustment were made, which correspond to a region of significant pH variation in the graph, as expected. During the time of 2.5 h that was given to complete the reactions, from 1000 to 10 000 s, the pH remains stable. For the other syntheses, the graphics are very similar, varying only in the final pH set-point. The pH variation, since the pH adjustment until 2.5 h, for all the samples, is shown in Table 1. It is shown that the pH slightly decreases during all the syntheses. This decrease in pH can be explained by the desorption of some Hþ initially adsorbed on the surface of the formed slurry, and by the hydrolysis reactions, which are accompanied by deprotonation [12]. The DpH observed for samples S5 and S6 is clearly higher than the values of the others samples. As can be seen later, those values can be justified by the occurrence of oxolation reactions, leading to an additional deprotonation step. The thermal bath used for the syntheses was set to a temperature of 25  C. In Fig. 3 it is possible to verify that the temperature of synthesis of sample S1 has not suffered a large deviation from the set point temperature. The same was observed for all the other syntheses. 4.2. Materials appearance Differences in the chemical moieties and bonds of the materials lead to variations in their electronic energetic levels and, therefore, changes on the wavelength of the light that they absorb and reflect, resulting in different colors of the material [23]. The color of iron oxides/hydroxides can help in the identification of the main constituent phase or as a measure of purity [24]. Fig. 4 shows the various colors exhibited by different phases of iron oxides/hydroxides. The final appearance of the milled iron oxides/ hydroxides samples is presented in the same figure. Samples S1, S2 and S4 show a brown color; S5 and S6 are also brown powders, but darker than the previous ones; S3 exhibits a yellow coloration. Based on the comparison of the two columns of the table, it is possible to make an initial identification of the probable phases present in the samples. Sample S3, due to its

Fig. 3. pH and temperature variations during the synthesis of sample S1.

Fig. 4. Characteristic color of different iron oxides/hydroxides and appearance of the obtained materials. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

yellow color, is expected to be mainly constituted by goethite or akaganeite, but existence of additional phases cannot be excluded. Samples S1, S2, S4, S5 and S6, with different brown tones, can be associated to ferrihydrite, maghemite or feroxyhyte. The darker tone of samples S5 and S6 can be related to the presence of magnetite. 4.3. Chemical/structural characterization of the iron oxides/ hydroxides 4.3.1. FTIR The obtained FTIR spectra for all the samples are presented in Fig. 5, where various regions are distinguished to show the existent chemical groups. As can be seen, the spectra of samples S1, S2 and S4 are identical, the spectra of samples S5 and S6 are also similar and the sample S3 spectrum is different from the others. The wavenumbers of the band locations are summarized in Table A1, in the appendix. The band in the range of 3000e3800 cm1 can be ascribed to the stretching vibrations of OH groups in the iron oxyhydroxides and adsorbed H2O molecules [6,25]. The peaks between 1000 and 1700 cm1 can be attributed to bending vibrations of OH groups [26]. For samples S3eS6, there are two sharp peaks in the interval 2300e2400 cm1. These peaks are also detectable in samples S1 and S2. They are attributed to the absorption of radiation by CO2 existing in the analysis atmosphere [27]. Sample S3 presents two sharp peaks around 800 and 900 cm1 attributed to the bending vibration of the group eOH … O in iron oxyhydroxides [28]. This spectral feature is typical of goethite [1]. The small disturbances around 800 and 900 cm1 in other samples can also indicate the presence of goethite at minor amount. Transmittance bands between 400 and 700 cm1 are characteristic of FeeO bonds [28]. In particular, sample S3 presents a broad peak at 636 cm1, and samples S5 and S6 a broad double peak with minima at 633 and 581 cm1, for sample S5, and at 633 and 577 cm1, for sample S6. The other samples present less defined bands in this region. Samples S3eS6 also exhibit a flat band

Please cite this article in press as: J. Pedrosa, et al., Controlled phase formation of nanocrystalline iron oxides/hydroxides in solution e An insight on the phase transformation mechanisms, Materials Chemistry and Physics (2015), http://dx.doi.org/10.1016/j.matchemphys.2015.07.018

6

J. Pedrosa et al. / Materials Chemistry and Physics xxx (2015) 1e11

Fig. 5. FTIR spectra of the obtained products.

centered at ~450 cm1. From the literature, it can be concluded that the bands observed near 590 cm1 correspond to the stretching vibrations of FeeO, with iron occupying tetrahedral sites, and the bands at ~450 cm1 can be assigned to the stretching vibrations of FeeO, with iron occupying octahedral sites; these two bands are characteristic of magnetite [29] and samples S5 and S6 are the only ones that present both bands. It should be noted that the peak wavenumber of each type of vibration can suffer a deviation due to the crystalline degree, aggregation and particle shape difference [12]. In summary, sample S3 is probably mainly composed by goethite, whereas samples S5 and S6 exhibit a large number of FeeO bonds and an almost flat curve between 800 and 1000 cm1, thus they can be associated with the phases magnetite and/or maghemite. Samples S1, S2 and S4 seem to have the same iron oxide/hydroxide phase in its constitution, presenting some amount of OH groups. However, it is not possible to clearly identify this phase based only on these results. 4.3.2. XRD The XRD patterns obtained for samples S1 to S6 are shown in Fig. 6, together with diffractograms of different iron oxides/hydroxides phases from the ICDD database for comparison. The selected iron oxides/hydroxides phases from the database were ferrihydrite, goethite, magnetite and maghemite (the registration numbers of these phases in the ICDD database are also presented in Fig. 6). The peaks of intensity less than 10% were omitted in the ICDD reference patterns. Samples S1, S2 and S4 show only two very broad peaks centered at ~42 and ~74 (2q), and the patterns reveal that the samples are composed of an amorphous material. These two peaks can be explained by the presence of ferrihydrite, magnetite and/or maghemite, since they coincide with the two most intense peaks from the referred phases. However, in samples S1 and S2, the existence of magnetite and maghemite is unlikely due to the low iron(II)/iron(III) ratio (0.02) used for their syntheses. Moreover, the pattern fits well with the two outer lines of ferrihydrite, and the obtained phase is probably the two-line ferrihydrite (see Section 1), that is typically amorphous. The more intense peaks of sample S3 match perfectly with all

the peaks of goethite, although this phase does not explain the presence of some small peaks, indicating the presence of other minor phases. The patterns observed in samples S5 and S6 match perfectly with the diffraction patterns of magnetite and maghemite. Due to the similar patterns between these phases, it is not easy to distinguish them by this technique. The two small elevations observed at ~36 and ~53 (2q) can be explained by the two most intense peaks of NaCl, indicating a less effective washing of these samples due to the stronger agglomeration of the magnetic particles. The Scherrer equation allowed the estimation of the average size of the crystallites, based on the major peak of the diffractogram of each sample. The calculated lattice parameters and crystallite sizes are presented in Table 2. From Table 2, it is possible to see that the calculated lattice parameters agree well with the database values, confirming the phase selection. The values obtained for the size of crystallites of the samples S1, S2 and S4 (~1 nm) confirm the low crystallinity of these samples. Even for the other samples, the crystallites sizes (~10e14 nm) remain in the nanometric range (<100 nm). €ssbauer spectroscopy 4.3.3. Mo €ssbauer spectra obtained for the six samples at room The Mo temperature are shown in Fig. 7. The solid lines represent the fit of a set of Lorentzians to the experimental data. The hyperfine parameters obtained from these curves for the prepared samples are summarized in Table 3. The fit lines of samples S1, S2 and S4 show only broad quadrupole doublets, with the parameters indicated in Table 3. The assignment of doublet signals to a particular iron phase is not direct, but ferrihydrite is most likely the phase that explains better € ssbauer and XRD results obtained for these samples [30]. the Mo Cornell and Schwertmann [1] and Murad and Schwertmann [31] mentioned that the poor order and the different states of crystal€ ssbauer linity of ferrihydrite cause significant variations in their Mo spectra, leading to difficulties in the interpretation of the parameters and to find consistent values in the literature for comparison. The parameters obtained for these samples are consistent with the presence of Fe(III) in octahedral sites and are in the range of the values obtained by Murad and Schwertmann [31], for different

Please cite this article in press as: J. Pedrosa, et al., Controlled phase formation of nanocrystalline iron oxides/hydroxides in solution e An insight on the phase transformation mechanisms, Materials Chemistry and Physics (2015), http://dx.doi.org/10.1016/j.matchemphys.2015.07.018

J. Pedrosa et al. / Materials Chemistry and Physics xxx (2015) 1e11

7

types of ferrihydrite, being also indicative of the variability in the parameters for this phase. The spectrum of sample S3 is a superposition of two sextets and one doublet. The strongest sextet, according to its parameters (Table 3), may be attributed to the presence of goethite, which constitutes 52.32% of the sample. The other sextet can be explained by the existence of a-Fe, showing a null isomer shift, a hyperfine magnetic field of 33.0 T [32] and corresponding to 37.45% of the sample. The presence of a-Fe in the sample can be due to the occurrence of oxidationereduction reactions between the iron ions and the oxygen present in the reaction system. The parameters of the doublet indicate the presence of Fe3þ atoms in tetrahedral sites or probably Fe3þ atoms in distorted octahedral sites, constituting 10.24% of the sample. In relation to the Fe3þ atoms in tetrahedral sites, with the respective obtained parameters, it was not possible to establish a clear connection to a specific iron phase. However supposing the existence of Fe3þ atoms in distorted octahedral sites, they can be attributed to the existence of ferrihydrite which has not yet been converted into a more stable phase. The spectra of samples S5 and S6 are composed by three sextets. The sextet with an isomer shift of 0.35 mm s1 is attributed to the presence of goethite, constituting 16.12% of sample S5 and 32.2% of sample S6. The other two sextets indicate the existence of magnetite, corresponding the sextet with H ~45 T to Fe2.5þ on octahedral sites, that represents 44.08% and 55.4% of samples S5 and S6, respectively; and the sextet with H ~49 T to Fe3þ on the tetrahedral sites, constituting 39.07% and 12.3% of samples S5 and S6, respectively [1]. From the previous results, it was observed that for acidic pH (S1 and S4) ferrihydrite is not converted into other iron phases. This fact was verified for the both Fe(II)/Fe(III) ratios used. The low pH create a positive charge shell around the ferrihydrite particles, disfavoring the adsorption of the Fe(II) species to their surface and so preventing their catalytic effect. In sample S2, the increase of pH to 7 did not lead to the formation of new phases, possibly due to the low ratio of Fe(II)/Fe(III) used. Sample S5, also with pH ¼ 7 but with a Fe(II)/Fe(III) ratio of 0.5, has led to the transformation of ferrihydrite into magnetite and goethite. In this case the high concentration of Fe(II) has led to the complete conversion in other phases, being thus verified its catalytic effect. The dominant amount of magnetite reveals that these are suitable conditions to obtain a good yield of this phase. Sample S3, even with a low Fe(II)/Fe(III) ratio (0.02), was almost completely converted into other phases (goethite and a-Fe), having left over about 10% of unconverted ferrihydrite. In this case the conversion may be due to the high pH (12) of the medium, which favors the occurrence of condensation reactions. Sample S6, with a high pH and Fe(II)/Fe(III) ratio, was completely converted into magnetite and goethite, being this conversion due to the effective catalytic effect of Fe(II) and the favoring of condensation reactions by the high pH.

Fig. 6. XRD patterns of the obtained products and ICDD reference patterns of selected phases for comparison.

Table 2 Lattice parameters and crystallites size obtained for the samples S1 e S6, based on the most probable phase constituent of the sample. Sample

Indexed phase

Crystallite size, t/Å

Lattice parameters (Å) Calculated

S1 S2 S3 S4 S5 S6

Ferrihydrite Ferrihydrite Goethite Ferrihydrite Magnetite Magnetite

ICDD database

a

b

c

a

b

c

5.21 5.11 4.58 4.97 8.37 8.40

e e 10.01 e e e

9.54 9.49 3.03 9.46 e e

5.08 5.08 4.61 5.08 8.37 8.37

e e 9.96 e e e

9.40 9.40 3.02 9.40 e e

10.0 11.5 146.7 10.0 120.8 109.2

Please cite this article in press as: J. Pedrosa, et al., Controlled phase formation of nanocrystalline iron oxides/hydroxides in solution e An insight on the phase transformation mechanisms, Materials Chemistry and Physics (2015), http://dx.doi.org/10.1016/j.matchemphys.2015.07.018

8

J. Pedrosa et al. / Materials Chemistry and Physics xxx (2015) 1e11

€ssbauer spectra, obtained at room temperature, for S1eS6 samples. The solid lines represent the fits of a set of Lorentzians to the experimental data. Fig. 7. Mo

Table 3 €ssbauer parameters resulting from the fits to data shown in Fig. 7.a Mo Sample S1 S2 S3

S4 S5

S6

Doublet Doublet Sextet Sextet Doublet Doublet Sextet Sextet Sextet Sextet Sextet Sextet

d (mm s1)

DEQ (mm s1)

H (T)

G (mm s1)

Area (%)b

Phasec

0.34(1) 0.34(1) 0.35 0.0 0.30(1) 0.35(1) 0.32(1) 0.32(1) 0.35 0.30(1) 0.31(1) 0.35

0.70(1) 0.68(1) 0.30 0.0 0.68(1) 0.68(1) 0.005(1) 0.006(1) 0.30 0.004(1) 0.022(1) 0.30

e e 38.0 33.0 e e 44.5 48.5 38.0 45.0 49.0 38.0

0.46(1) 0.43(1) 0.47(1) 1.29(1) 0.53(1) 0.41(1) 0.96(1) 0.70 1.24(1) 1.11(1) 0.48 1.56(1)

100 100 52.32 37.45 10.24 100 44.08 39.7 16.12 55.4 12.3 32.2

VI

Fe3þ Fe3þ Goethite a-Fe IV Fe3þ/VIFe3þ VI Fe3þ Magnetite Magnetite Goethite Magnetite Magnetite Goethite VI

a €ssbauer parameters: isomer shift (d), quadrupole splitting (DEQ), hyperfine magnetic field (H) and line width (G). Isomer shifts are given relatively to a-Fe. The values Mo not showing errors have been fixed in the fitting procedure. b Area under the peak (%) is related to the relative abundance of each iron site. c IV Fe3þ e iron(III) in tetrahedral sites; VIFe3þ e iron(III) in octahedral sites.

Liu et al. [9] and Andreeva et al. [33] have concluded that the conversion of ferrihydrite into goethite proceed through a dissolution/reprecipitation mechanism (reconstructive transformation). Usman et al. [34] reported the formation of magnetite by interactions of ferrihydrite and Fe(II); the presence of Fe(II) induced structural modifications, promoting a topotactic transformation of ferrihydrite. If the transformation into magnetite and goethite occurs by the mechanisms mentioned above, it becomes evident with the current work that a competition between the two mechanisms exists, since some amount of goethite also appeared in conjunction with magnetite in samples S5 and S6. Moreover, it can be concluded that

the pH strongly influences the extension of conversion by both mechanisms. On the other hand, the amount of Fe(II) clearly defines, at high pH, the predominant mechanism. 4.4. Materials microstructure observed by SEM All samples were inspected by SEM. The images of the surfaces at two magnifications are shown in Fig. 8. Sample S1 is composed of agglomerates of spherical particles of approximately 1 mm in diameter. As referred in literature, these spheres can be associated to the presence of ferrihydrite [1], which validates the previous conclusions.

Please cite this article in press as: J. Pedrosa, et al., Controlled phase formation of nanocrystalline iron oxides/hydroxides in solution e An insight on the phase transformation mechanisms, Materials Chemistry and Physics (2015), http://dx.doi.org/10.1016/j.matchemphys.2015.07.018

J. Pedrosa et al. / Materials Chemistry and Physics xxx (2015) 1e11

Fig. 8. SEM images of samples S1eS6 with lower magnifications (on the left) and a magnification of 20000 (on the right).

Please cite this article in press as: J. Pedrosa, et al., Controlled phase formation of nanocrystalline iron oxides/hydroxides in solution e An insight on the phase transformation mechanisms, Materials Chemistry and Physics (2015), http://dx.doi.org/10.1016/j.matchemphys.2015.07.018

9

10

J. Pedrosa et al. / Materials Chemistry and Physics xxx (2015) 1e11

For sample S2, also composed by ferrihydrite, the spherical particles are not so easily seen. This fact can indicate that more dense agglomerates of ferrihydrite were formed. € ssbauer spectrum of sample S3 has indicated the presThe Mo ence of goethite (52%). The SEM images of sample S3 (Fig. 8) show needle shaped crystals with a length apparently less than 0.5 mm. The obtained images are in agreement with other SEM images €ssbauer found in literature for goethite [35]. Although the Mo spectrum for this sample have presented the existence of other two phases, by the SEM analysis it was not possible to observe the presence of other types of crystals. The images of sample S4 show the clustering of two distinct sizes of particles: some less than 0.5 mm in size and others over ~1 mm in size. In spite of this heterogeneity of particles sizes, the shape of the particles is consistent with the presence of ferrihydrite. € ssbauer spectra for samples S5 and S6 have indicated the The Mo presence of two different phases, magnetite and goethite. However, in the SEM images, none of the samples show a texture similar to the needle shaped crystals found in sample S3, corresponding to the goethite phase. This can be related to the fact that goethite is in a much smaller amount than magnetite. The images of samples S5 and S6 show a denser material when compared to that of the other samples, and they are composed by particles much less than 1 mm in size. 5. Conclusions In the present study the influence of pH and Fe(II)/Fe(III) ratio in the controlled formation of iron oxides/hydroxides by chemical precipitation is reported. According to the results obtained from XRD and SEM, the obtained iron oxides/hydroxides consisted of agglomerates of nanometric crystallites ranging from ~1 to ~15 nm. Different shapes of the particles were observed by SEM, according to the predominant phase in the material. The existing phases in the materials were €ssbauer Spectroscopy. identified by FTIR, XRD and Mo The syntheses carried out at pH 4, led exclusively to the formation of ferrihydrite, independently of the Fe(II)/Fe(III) ratio. At pH 7 with a Fe(II)/Fe(III) ratio of 0.02, the formation of ferrihydrite was also observed. For low Fe(II)/Fe(III) ratio and a pH 12, the conversion of ferrihydrite into goethite was observed. At a high Fe(II)/Fe(III) ratio at pH 7 and 12, a mixture of magnetite and goethite was formed. Both mechanisms, topotactic and reconstructive, are strongly influenced by the pH, being the extent of the transformations enhanced by higher pH values. At high pH values the predominant mechanism is defined by the ratio of Fe(II)/Fe(III); a higher ratio favors the topotatic transformation of ferrihydrite in magnetite, while a low ratio leads to a dominant reconstructive mechanism. Acknowledgments B.F.O. Costa acknowledge the support funds from FEDER (Programa Operacional Factores de Competitividade COMPETE)/FCT e ~o para a Cie ^ncia e Tecnologia under the project PEst-C/FIC/ Fundaça UI0036/2011. Appendix Characteristic vibration frequencies of FTIR spectra of the synthesized materials are shown in Table A1.

Table A1 Characteristic vibration frequencies (cm1) of FTIR spectra of the synthesized materials. Vibration assignmenta

Sample S1

S2

S3

3151

3126

3124 2360 2341 1653 1557 1339

S4

S5

3361

1401

570 a

2360 2341 1616

890 796 636

832 616

453

463

552

S6 3421

2360 2341 1540

2360 2341 1635 1540

969

991

839 633 581 456

812 633 577 441

O-Hs O-Hs CO2 CO2 O-Hb O-Hb O-Hb O-Hd (Fe hydroxides) O-Hd (Fe hydroxides) O-Hd (Fe hydroxides) FeeOs FeeOs FeeOs

s e stretching, b e bending; d e deformation.

References [1] R.M. Cornell, U. Schwertmann, The Iron Oxides: Structure, Properties, Reactions, Occurences and Uses, second ed., Wiley-VCH, New York, 2003. [2] L. Machala, R. Zboril, A. Gedanken, Amorphous Iron(III) oxide e a review, J. Phys. Chem. 111 (2007) 4003e4018. [3] M. Faraji, Y. Yamini, M. Rezaee, Magnetic nanoparticles: synthesis, stabilization, functionalization, characterization, and applications, J. Iranian Chem. Soc. 7 (2010) 1e37. [4] M. Mohapatra, S. Anand, Synthesis and applications of nano-structured iron oxides/hydroxides e a review, Int. J. Eng. Sci. Technol. 2 (2010) 127e146. [5] H. Liu, M. Ma, M. Qin, L. Yang, Y. Wei, Studies on the controllable transformation of ferrihydrite, J. Solid State Chem. 183 (2010) 2045e2050. [6] M. Ristic, E. De Grave, S. Music, S. Popovic, Z. Orehovec, Transformation of low crystalline ferrihydrite to a-Fe2O3 in the solid state, J. Mol. Struct. 834e836 (2007) 454e460. [7] T. Hiemstra, W. Van Riemsdijk, A surface structural model of ferrihydrite I: sites related to primary charge, molar mass, and mass density, Geochim. Cosmochim. Acta 73 (2009) 4423e4436. [8] M. Niederberger, N. Pinna, Metal Oxide Nanoparticles in Organic Solvents: Synthesis, Formation, Assembly and Application, Springer, London, 2009. ISBN: 9781848826700. [9] H. Liu, Y. Wei, Y. Sun, The formation of hematite from ferrihydrite using Fe(II) as a catalyst, J. Mol. Catal. A Chem. 226 (2005) 135e140. [10] H. Liu, P. Li, M. Zhu, Y. Wei, Y. Sun, Fe(II)-induced transformation from ferrihydrite to lepidocrocite and goethite, J. Solid State Chem. 180 (2007) 2121e2128. [11] H. Liu, H. Guo, P. Li, Y. Wei, The transformation of ferrihydrite in the presence of trace Fe(II): the effect of the anionic media, J. Solid State Chem. 181 (2008) 2666e2671. [12] H. Liu, P. Li, B. Lu, Y. Wei, Y. Sun, Transformation of ferrihydrite in the presence or absence of trace Fe(II): the effect of preparation procedures of ferrihydrite, J. Solid State Chem. 182 (2009) 1767e1771. [13] H. Liu, Y. Wang, Y. Ma, Y. Wei, P. Guoqiang, The microstructure of ferrihydrite and its catalytic reactivity, Chemosphere 79 (2010) 802e806. [14] B. Liu, H. Guo, L. Ping, H. Liu, Y. Wei, D. Hou, Comparison study on transformation of iron oxyhydroxides: based on theoretical and experimental data, J. Solid State Chem. 184 (2011) 2139e2144. [15] A. Saric, S. Music, K. Nomura, S. Popovic, Microstructural properties of Feoxide powders obtained by precipitation from FeCl3 solutions, Mater. Sci. Eng. B Adv. 56 (1998) 43e52. [16] J. Livage, M. Henry, C. Sanchez, Sol-gel chemistry of transition metal oxides, Prog. Solid State Chem. 18 (1988) 259e341. [17] J.P. Jolivet, Metal Oxide Chemistry and Synthesis: From Solution to Solid State, Wiley, 2000. ISBN: 0471970565. [18] S. Sakka, in: H. Kozuka (Ed.), Handbook of Sol-gel Science and Technology: Processing, Characterization and Applications, Sol-Gel Processing, vol. I, Kluwer Academic Publishers, 2005.
ac, E. Tronc, Iron oxide Chemistry. From molecular clusters [19] J.P. Jolivet, C. Chane to extended solid networks, Chem. Commun. 5 (2004) 481e487. [20] J. Zhao, F. Huggins, Z. Feng, G. Huffman, Ferrihydrite: surface structure and its effects on phase transformation, Clay Clay Miner. 42 (1994) 737e746. [21] B.H. Jeon, B. Dempsey, W. Burgos, Kinetics and mechanisms for reactions of Fe(II) with Iron(III) oxides, Environ. Sci. Technol. 37 (2003) 3309e3315. [22] L. Smart, E. Moore, Solid State Chemistry: An Introduction, third ed., CRC Press, New York, 2005. ISBN:0203610636. [23] CCMR, Electron Excitation Responsible for Colors of Objects, CCMR e Cornell

Please cite this article in press as: J. Pedrosa, et al., Controlled phase formation of nanocrystalline iron oxides/hydroxides in solution e An insight on the phase transformation mechanisms, Materials Chemistry and Physics (2015), http://dx.doi.org/10.1016/j.matchemphys.2015.07.018

J. Pedrosa et al. / Materials Chemistry and Physics xxx (2015) 1e11

[24]

[25] [26] [27]

[28]

[29]

[30]

Center for Materials Research: An NSF MRSEC, July 2012. http://www.ccmr. cornell.edu/education/ask/?quid¼1297. I. Bruce, J. Taylor, M. Todd, M. Davies, E. Borioni, C. Sangregorio, Synthesis, characterization and application of silica-magnetite nanocomposites, J. Magn. Magn. Mater. 284 (2004) 145e160. J. Pattanayak, V. Rao, H. Maiti, The hydrolysis product of ferric nitrate in sodium hydroxide, J. Mater. Sci. 25 (1990) 2245e2248. M. Gotic, S. Music, Mossbauer, FT-IR and FE SEM investigation of iron oxides precipitated from FeSO4 solutions, J. Mol. Struct. 834 (2007) 445e453. L.E. Amand, C. Tullin, Theory Behind FTIR AnalysisdApplication Examples from Measurement at the 12 MW Circulating Fluidized Bed Boiler at Chalmers, CECOST, Lund University, Sweden, 1999. ~es, A. Moutinho, I. Seabra, B. Costa, H. Sousa, A. Portugal, CharacterL. Dura ization of iron(III) oxide/hydroxide nanostructured materials produced by solegel technology based on the Fe(NO3)3
9H2OeC2H5OHeCH3CHCH2O system, Mater. Chem. Phys. 130 (2011) 548e560. T. Ozkaya, M. Toprak, A. Baykal, H. Kavas, Y. Koseoglu, B. Aktas, Synthesis of Fe3O4 nanoparticles at 100 C and its magnetic characterization, J. Alloys Compd. 472 (2009) 18e23. C. Cromphaut, V. Resende, E. De Grave, A. Van Alboom, R. Vandenberghe,

[31] [32]

[33]

[34]

[35]

11

G. Klingelhofer, Characterisation of the magnetic iron phases in Clovis class rocks in Gusev crater from the MER Spirit Mossbauer spectrometer, Geochim. Cosmochim. Acta 71 (2007) 4814e4822. E. Murad, U. Schwertmann, The Mossbauer spectrum of ferrihydrite and its relations to those of other iron oxides, Am. Mineral. 65 (1980) 1044e1049. €ssbauer spectroscopy of minerals, in: T.J. Ahrens (Ed.), C. McCammon, Mo Mineral Physics and Crystallography: a Handbook of Physical Constants, American Geophysical Union, Washington D.C, 1995, pp. 332e347. D. Andreeva, I. Mitov, T. Tabakova, V. Mitrov, A. Andreev, Influence of iron(II) on the transformation of ferrihydrite into goethite in acid medium, Mater. Chem. Phys. 41 (1995) 146e149.
goire, P. Faure, C. Ruby, FeII M. Usman, M. Abdelmoula, K. Hanna, B. Gre induced mineralogical transformations of ferric oxyhydroxides into magnetite of variable stoichiometry and morphology, J. Solid State Chem. 194 (2012) 328e335. G. Montes-Hernandez, P. Beck, F. Renard, E. Quirico, B. Lanson, R. Chiriac, N. Findling, Fast precipitation of acicular goethite from ferric hydroxide gel under moderate temperature (30 and 70 C degrees), Cryst. Growth Des. 11 (2011) 2264e2272.

Please cite this article in press as: J. Pedrosa, et al., Controlled phase formation of nanocrystalline iron oxides/hydroxides in solution e An insight on the phase transformation mechanisms, Materials Chemistry and Physics (2015), http://dx.doi.org/10.1016/j.matchemphys.2015.07.018