XPS study on the mechanism of starch-hematite surface chemical complexation

Minerals Engineering 110 (2017) 96–103

Contents lists available at ScienceDirect

Minerals Engineering journal homepage: www.elsevier.com/locate/mineng

XPS study on the mechanism of starch-hematite surface chemical complexation

MARK

Gabriela F. Moreiraa, Elaynne R. Peçanhab, Marisa B.M. Montea,b, Laurindo S. Leal Filhoa, ⁎ Fernando Stavalec, a

Instituto Tecnológico Vale - ITV, Avenida Juscelino Kubitschek, 31, Ouro Preto, MG 35.400-000, Brazil Laboratory for Surface Chemistry, Coordination of Mineral Processing – CETEM, Avenida Pedro Calmon, 900 Ilha da Cidade Universitária, Rio de Janeiro, RJ 21941972, Brazil c Centro Brasileiro de Pesquisas Físicas – CBPF/MCTI, Rua Xavier Sigaud 150, Rio de Janeiro, RJ 22290-180, Brazil b

A R T I C L E I N F O

A B S T R A C T

Keywords: Iron ore Starch Flotation XPS FTIR

Polysaccharides are some of the most widely employed flotation reagents in the mineral processing industry. Among several, starch is of particular importance for reverse flotation of iron ores. It is known to behave as an efficient depressant for hematite and, therefore, its interaction is of great relevance. In this paper, we investigate the surface chemistry of starch adsorbed onto hematite by means of X-ray photoelectron spectroscopy and Fourier transform infrared spectroscopy. Our results indicate that starch undergoes an important oxidation process under industrial gelatinization conditions, which favors the binding of starch molecules. Meanwhile, the oxide surface is subjected to full hydroxylation, leading to strong polysaccharide-metal hydroxide interaction. Previously proposed starch interaction mechanisms are discussed, and the importance of acid-base interactions is underscored.

1. Introduction Adsorption of specific molecules on metal oxide and hydroxide surfaces is of significant importance in a variety of fields, ranging from catalysis to corrosion science and semiconductor manufacture (Shrimali et al., 2016). Applications involving separation of minerals by flotation are particularly relevant for this paper (Araujo et al., 2005). In this field, a metal oxide surface immersed in aqueous solution undergoes several reactions at acid-base sites, including the formation of surface hydroxyl groups and eventual adsorption of organic molecules, resulting in hydrophobicity or hydrophilicity. The adsorbed species can interact upon the oxide surface in different ways, such as through electrostatic or hydrophobic interactions and chemical complexation, depending on the choice of appropriate aqueous solution conditions. Therefore, a significant step toward understanding their chemical interaction can be made by investigating surface composition and associated changes (Laskowski et al., 2007; Liu et al., 2000; Filippov et al., 2013). In this study, we focus on the interaction between iron oxide (hematite) and cornstarch, one of the mostly widely used organic depressants. Adsorption of starch on iron oxides has been widely investigated and debated in the literature as it represents a relatively

⁎

Corresponding author. E-mail address: [email protected] (F. Stavale).

http://dx.doi.org/10.1016/j.mineng.2017.04.014 Received 26 October 2016; Received in revised form 6 April 2017; Accepted 21 April 2017 0892-6875/ © 2017 Elsevier Ltd. All rights reserved.

inexpensive and environmentally-friendly separation agent, besides acting as a very effective depressant (Laskowski et al., 2007; Liu et al., 2000; Weissenborn et al., 1995; Kar et al., 2013; Tang and Liu, 2012; Pavlovic and Brandao, 2003; Martins et al., 2012). Moreover, the starch adsorption mechanism is of major importance not only for the separation of iron ores, but also because it is largely employed for a number of other minerals, including sulfide and phosphate ores (Laskowski et al., 2007; Liu et al., 2000; Filippov et al., 2013; Raju et al., 1997; Leal Filho et al., 2000). In short, the separation process is based on the adsorption of starch molecules onto iron oxide, which renders the surface hydrophilic, preventing, at the same time, the adsorption of hydrophobic amine molecules present in the aqueous solution. In earlier studies, nonselective hydrogen bonding and electrostatic forces were proposed as the primary adsorption mechanism for carbohydrate molecules, mainly because of the large number of hydroxyl groups on both starch and oxide surfaces (Balajee and Iwasaki, 1969). Later investigations suggested, however, that polysaccharide molecules might preferably adsorb through acid-base interactions, as discussed by Laskowski et al. in a recent review article (Laskowski et al., 2007). Current research on starch adsorption addresses the important role of metalhydroxylated oxide surface, supported by a number of studies using

Minerals Engineering 110 (2017) 96–103

G.F. Moreira et al.

starch at a 5:1 ratio and the starch solution was designed to yield an optimal concentration of 550 mg/L, just as employed industrially (Ma et al., 2011). For the starch-adsorbed samples, hematite was conditioned with a gelatinized starch suspension for 60 min at pH 10.5. The suspensions were filtered and dried in a vacuum desiccator for approximately 12 h at room temperature and later loaded into an ultra-high vacuum chamber for XPS or FTIR analysis.

infrared spectroscopy and adsorption curves, in particular the one by Weissenborn et al. (1995), Pavlovic and Brandao (2003), Tang and Liu (2012), Subramanian and Natarajan (1988), Lima and Brandao (1999). These studies suggest that shifts in the infrared vibrational bands are related to the glucopyranose ring, and CeH deformation occurs due to the attachment of two carbon atoms to an oxygen-terminated hematite surface. This model is consistent with the one previously proposed by Liu et al. (2000), Ravishankar and Pradip (1995), in which a polysaccharide metal ring complex formed between the organic molecule and the oxide surface. Actually, this acid-base interaction is likely to occur; however, evidence of whether this chemical interaction is weak or strong and of its bonding nature is lacking in the literature. To assess these mechanisms and to improve our understanding of chemical complexation, we investigated natural hematite (Fe2O3) surfaces after alkali treatment (pH 10.5) and starch adsorption, using diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS) and X-ray photoelectron spectroscopy (XPS). Through surface-sensitive chemical characterization, we found compelling evidence of iron oxyhydroxide formation under alkaline conditions and spectral signatures of starch oxidation followed by strong surface complexation of iron polysaccharide.

2.2. X-ray photoelectron spectroscopy measurements The experiments were performed using a SPECS UHV system (base pressure of 1 × 10−9 bar) equipped with a PHOIBOS 150 hemispherical electron analyzer using monochromatic Al-Kα radiation. The spectra were recorded with pass energy of 50 or 20 eV for surveys and high-resolution measurements, respectively. The spectrometer was previously calibrated using the Au 4f7/2 peak (84.0 eV), resulting in a full width half maximum (FWHM) of 0.7 eV for a sputtered metallic gold foil. The sample surface contained substantial amounts of adventitious carbon, but the samples were not sputtered, since preferential sputtering of oxygen over iron is known to reduce oxidation of original surface species (Biesinger et al., 2011). The binding energies were referenced by setting the adventitious carbon C 1s peak to 284.6 eV for hematite, Na 1s peak at 1071.5 eV for starch or the Fe 3p 56 eV for hematite after starch adsorption (Biesinger et al., 2011; Shchukarev and Korolkov, 2004). The high-resolution spectra of Fe 2p, O 1s, and C 1s regions were fitted using CasaXPS with a Shirley background and a Gaussian/Lorentzian line shape.

2. Experimental design 2.1. Materials Natural hematite (Fe2O3) powders (5–150 μm) were used. The samples were dry-ground and wet-sieved. Fraction sizes around 150 μm were used for XPS while those below 20 μm were used for DRIFTS and zeta potential measurements. The obtained fractions were washed with Milli-Q water. The chemical composition of the samples, determined by X-ray fluorescence (XRF) and inductively coupled plasma optical emission spectrometry (ICP-OES), is summarized in Table 1. The results indicate that the major components are Fe with minor to trace amounts of Ti, Cr, Al, Si, Ca, and Mn. Scanning electron microscopy images and energydispersive spectroscopy analyses (data not shown) also indicate microparticles in the samples are mostly composed of iron oxide with only trace amounts of other elements. The high-grade cornstarch employed in this study was supplied by VETEC Ltda. Three experimental tests were performed under the following conditions: (1) pure hematite, (2) hematite stirred in water at pH 10.5 and, (3) hematite stirred in gelatinized starch solution at pH 10.5. The conditions strongly dictates the adsorption of starch molecules onto the hematite surface and the chosen pH value is known to effectively boost flotation. Ultrapure deionized water was used for the preparation of all samples under the same ionic conditions (NaCl 10−3 M). Hydroxylated hematite was prepared by stirring the particulate samples in analytical grade NaOH solution at pH 10.5 for 60 min at room temperature. The gelatinized starch was prepared by adding NaOH and

2.3. Infrared Fourier transform spectroscopy measurements For IR spectral studies, DRIFTS and attenuated total reflectance (ATR) were employed using an IRPrestige-21 Shimadzu spectrophotometer at a resolution of 4 cm−1. DRIFTS and ATR measurements were employed to investigate the vibrational bands present in starch powder and in gelatinized starch film after alkali treatment. Further analyses were performed using DRIFTS to investigate hematite powder samples diluted to 10% by mass in a KBr matrix. Spectral changes were better visualized in the difference spectra by subtracting the bare hematite spectrum from the conditioned hematite one, as indicated in the Results section. 2.4. Zeta potential measurements Zeta potential measurements as a function of pH were obtained at constant ionic strength using the Malvern-Zetasizer system (Malvern Instruments, UK). The samples were thoroughly equilibrated for 60 min at each pH and the measurements were recorded in a dilute suspension of 1% solids by weight. NaCl (10−3 M) was used as an indifferent electrolyte and pH was adjusted using diluted HCl and NaOH solutions. Since electrokinetic studies are performed in liquid, the direct measurement of hematite in suspension with adsorbed starch was performed after adsorption tests. This technique requires dilute solutions; therefore, starch concentrations used in these tests were 12 mg/L, 25 mg/L (previously determined in studies to determine the best conditions), and 550 mg/L for comparison.

Table 1 X-ray fluorescence and inductively coupled plasma optical emission spectroscopy (XRF and ICP-OES) of natural hematite. Element

Fe Ti Si Cr Al Mn Mg Ca Na

Chemical analysis (%) XRF

ICP

3. Results

67.60 0.66 0.15 0.21 0.14 < 0.10 < 0.10 < 0.10 –

64.80 0.64 0.10 0.15 0.15 0.14 0.036 0.076 0.062

3.1. Zeta potential measurements Fig. 1 depicts the zeta potential curves for gelatinized starch, pure hematite, and hematite after starch adsorption at different concentrations. The curve profile indicates that the gelatinized starch surface is negatively charged for the entire pH range, approaching a potential equal to zero between pH 3 and 1, but it was not possible to observe its isoelectric point (IEP). According to a previous study by Tang and Liu 97

Minerals Engineering 110 (2017) 96–103

G.F. Moreira et al.

Starch (powder) Gelatinized Starch 1650 1560 1410 1352

20

11

12

-10 -20

-40

Hematite Gelatinized Starch Hematite after starch adsorption (12 mg/L) Hematite after starch adsorption (25 mg/L) Hematite after starch adsorption (550 mg/L)

3470

-30

4000

pH

3600

3200

2800

2400

2000

845

10

928

9

754

8

990

7

1150

6

1080

5

1015

4

1450 1340 1150 1076 995 930 860 765

3

1640

2

2930 2890

1

3350

0

Transmittance

Zeta potential (mV)

10

1600

1200

800

400

-1

Wavenumber (cm )

Fig. 1. Zeta potential as a function of pH for hematite, gelatinized starch, and hematite after starch adsorption at different concentrations. NaCl (10−3 M) was used as electrolyte background with a starch concentration of 550 mg L−1.

Fig. 2. DRIFT spectrum of starch powder and ATR spectrum of gelatinized starch (after vacuum drying).

changes are also identified in the 1080–900 cm−1 region for the stretching mode of alcohol CeOH moieties coupled to CeC. The bands at 1015 cm−1 and 990 cm−1 appear sharper and intensified after gelatinization and have been previously assigned to amorphous starch stretching modes (Capron et al., 2007). Notably, the new peaks located at 1560 cm−1 and 1410 cm−1 indicate formation of carboxyl groups (COO−) during the gelatinization process. These bands have been related in previous studies to COO− asymmetric and symmetric stretching in carboxylic acid salts (Max and Chapados, 2004; Szymanski and Erickson, 1970). Therefore, the spectral signature reveals that the structural modification is connected to severe chemical changes as the insertion of carboxyl and alcohol groups are promoted and starch undergoes oxidation (Tang and Liu, 2012; Ragheb et al., 1995).

(2012), the negative surface charge originates from the dissociation of protons from hydroxyl (-OH) and carboxylic functional groups (-COOH). The hydroxyl groups are believed to be initially present in starch molecules as opposed to carboxylic groups introduced during gelatinization. The zeta potential characteristics for pure hematite, with an IEP of 4.5, are typical, and comparison with hematite conditioned with starch serves to address the chemical modifications induced on the oxide surface (Fuerstenaua, 2005; Kar et al., 2013). After adsorption, a substantial change was observed in the pure hematite IEP at both starch concentrations in solution (IEP of 3.1 for 12 mg/L and 3.5 for 25 mg/L). This behavior suggests a specific interaction between starch molecules and oxide surface since the IEP is drastically shifted to lower pH values (Fuerstenaua, 2005). Moreover, at a high concentration (550 mg/L), we note a very similar potential profile for starch and hematite with starch adsorption, indicating that we probably only observe the surface potential of the starch molecule in solution. Also, the curve close to pH 10.5 indicates that both starch and hematite are negatively charged, suggesting a relationship that cannot be based on purely electrostatic interaction. Additional information on the nature of this specific interaction is provided in the following sections.

3.3. XPS spectra of starch The surface chemical composition of starch films was examined by XPS. Fig. 3 shows the typical features observed for starch gelatinized at pH 10.5 using NaOH. The film surface contains carbon, oxygen, and sodium, with a C/O ratio of 0.9, which is relatively close to the expected calculated value of 0.8. The O 1s and C 1s peaks in the spectra were fitted using several components, as illustrated in Fig. 3b and c, respectively. The C 1s spectra have a main peak at 285 eV related to CeC/CeH groups (label C1) and several other components at 285.5, 286.8, 289.3 eV assigned to CeOH (label C2), CeO (label C3) and, the latter one, to COOH (label C4) groups, respectively. Note a large FWHM for the 286.8 eV component, which may account for different superimposed carbon bonds, such as CeO and O-glycosidic bonds (Yan et al., 2015; Yanga et al., 2009; Yamada et al., 2013; Desimoni et al., 1990). These peak assignments are perfectly consistent with those found for O 1s components, as shown in Fig. 3b. The O 1s region displays contributions from Na KLL Auger transition at 535.8 eV (label NaAuger), COOH at 531.2 eV (label O1), CeOH at 532.4 eV (label O2), and adsorbed water at 533.8 eV (label O3). Table 3 shows the overall peak positions and the FWHM obtained from the fitting of spectra components.

3.2. FTIR spectra of starch Fig. 2 illustrates the IR spectra of starch before and after gelatinization. The most pronounced bands are indicated and the major changes are related to the disruption of large starch molecules and to the loss of crystallinity during polymerization process. The spectra show significant differences between starch powder and gelatinized starch. Table 2 presents the assignments of infrared absorption bands for starch powder, starch after gelatinization, and starch after adsorption on hematite (Filippov et al., 2013; Weissenborn et al., 1995; Tang and Liu, 2012; Subramanian and Natarajan, 1988; Casu and Reggiani, 1964). Note that the most significant changes occur between OeH and CeOH stretching frequencies. The consequences of gelatinization can be observed throughout the infrared spectra. The most obvious changes are associated with the steep decline of intramolecular hydrogen bonds between starch molecules (3300–3600 cm−1) and the broader and shifted features in the 1420–1200 cm−1 region assigned to OeH coupled to CeH deformation. These modifications are characteristic of the amorphization process related to the decrease of intramolecular and intermolecular hydrogen bonds (Filippov et al., 2013; Weissenborn et al., 1995). Molecular

3.4. DRIFT spectra of hematite The DRIFT spectra of hematite, hematite stirred in water at pH 10.5, gelatinized starch, and starch adsorbed on hematite are shown in Fig. 4. The spectral changes for the stirred hematite and starch adsorbed on hematite are better visualized in the difference spectra displayed in the 98

Minerals Engineering 110 (2017) 96–103

G.F. Moreira et al.

Table 2 Assignment of infrared vibration bands for starch (powder), gelatinized starch, and starch adsorbed on hematite (Filippov et al., 2013; Weissenborn et al., 1995; Tang and Liu, 2012; Subramanian, 1988; Casu and Reggiani, 1964). Approximate band position (cm−1)

Vibrational assignment

Literature

Starch (powder)

OeH stretching, intramolecular hydrogen bond between hydroxyl groups attached to C-2 and C3́ from adjacent α-glucopyranose units

3470

3470

OeH stretching, intramolecular hydrogen bond CeH and CeH2 stretching Water HeOeH deformation COO− asymmetric stretching in carboxylic acid salts CeH2 deformation COO− symmetrical stretching in carboxylic acid salts OeH in-plane deformation coupled with CeH deformation CeOeC glycosidic linkage stretch coupled with CeOH stretch and OeH deformation CeO stretch coupled with CeC stretch and OeH deformation Ring vibration C1eH deformation. Equatorial (chair) hydrogen

3350 2930/2900 1650 1560 1455 1410 1420–1200 1150 1080–1000 935/765 890–845

3350 2930/2890 1640

figures on the right side of the vertical axis, whose spectra are subtracted from the pure hematite spectrum. Fig. 4a shows hematite spectra with several bands at 580 cm−1 and 490 cm−1 related to FeeO vibrational modes, as extensively reported in the literature (Rendon and Serna, 1981). Additional features related to overtone and combination bands of FeeO stretching vibrations near 1088 and 1038 cm−1 are quite clear as well (Subramanian and Natarajan, 1988; Rendon and Serna, 1981). The two prominent bands at 915 and 783 cm−1 are related in several studies to FeeOeH bending vibrations related to goethite phase (Rendon and Serna, 1981; Salama et al., 2015). As hematite is stirred at pH 10.5, both goethite bands show more intense and with higher definition at 908 and 793 cm−1. The bands at 580 to 490 cm−1 become broader with the new shoulder at 573 cm−1. A new shoulder unfolds at 700 cm−1 and indicates an out-of-plane deformation of -OH groups. These findings suggest formation of a goethite-like surface since broadening is known to intensify with the increase in the amount of substoichiometric iron oxide (Kustova et al., 1992). In

Gelatinized starch

Starch adsorbed on hematite 3300

1650 1560

2930/2897 1665 1530

1410 1352 1150 1080/1015/990 928/754 845

1395 1355 1150 1080/1045/1022 928/760 849

1450 1340 1150 1076/995 930/765 860

Table 3 XPS characterization of starch gelatinized with NaOH at pH 10.5. Peak components

BE (eV) (FWHM)

C 1s

CeC/CeH CeOH CeO COOH

284.9 285.5 286.8 289.3

O 1s

COOH CeOH CeO/water

531.2 (1.27) 532.4 (1.28) 533.8 (1.38)

Na KLL

NaAuger

535.8 (2.64)

(0.92) (1.66) (1.23) (0.90)

addition, bands in the range of 1600–1400 cm−1 related to HeOeH water and hydroxyl deformations significantly increase after stirring hematite at pH 10.5.

Fig. 3. (a) XPS survey spectra of gelatinized starch at pH 10.5 using NaOH. (b) and (c) High-resolution spectra of O 1s and C 1s regions, respectively.

99

Minerals Engineering 110 (2017) 96–103

G.F. Moreira et al.

Fig. 4. (a) The left axis shows the DRIFT spectrum of hematite and hematite stirred in water at pH 10.5 and the right axis shows the difference spectrum of hematite stirred in water at pH 10.5. (b) The left axis shows the ATR spectrum of gelatinized starch and the right axis shows the difference spectrum of starch adsorbed on hematite.

scans indicate the presence of iron, oxygen, carbon, and traces of manganese on the sample surface. In the present study, the impact of Mn cations on surface chemistry is ruled out since it accounts for less than 1% of surface chemical composition. The most pronounced XPS spectral changes are observable at the high-resolution Fe 2p, O 1s and C 1s core levels shown in Fig. 6. In the high-resolution spectra of the Fe 2p region one can observe that the Fe+3 valence state is characterized by an asymmetric 2p3/2 peak and by the satellite feature at higher binding energy. These features are typical of hematite (Fe2O3) related to Fe+3 cations at octahedral sites. The binding energy for the spin-orbit splitting peaks and Fe 2p3/2 and 2p1/2, are 711.0 and 724.4 eV, respectively. When analyzing the 2p region, the intensity of each spin-orbit component can be fitted using two curves related to the different chemical environments of the surface cations. This is not accurate, though, since 2p states in transition metals are actually composed of multiplet splitting, as described by Gupta and Sen (Gupta and Sen, 1975) and extensively investigated in several recent studies (Biesinger et al., 2011; Gupta and Sen, 1975). Nevertheless, the fitted components serve as reference, indicating trends. The fitting shown in Fig. 6a indicates that Fe 2p3/2 and 2p1/2 envelopes can be fairly adjusted by two components each and by the corresponding satellite features. As hematite samples were stirred in water at pH 10.5, both peaks broadened and the satellite feature became fainter, as 2p peaks were kept following the same FWHM and the approximate BE position. The corresponding fitting is depicted in Fig. 6d. Note that one component increases, line shape 1, and the 2p3/2 peak envelope becomes more symmetric, suggesting changes on the surface iron coordination or subtle modifications in charge state (Welsh and Sherwood, 1989). Spectral changes induced by starch adsorption are difficult to identify since the surface is covered by a polysaccharide layer that is a few nanometers thick. Yet, by examining the changes depicted in Fig. 6g, one can observe that as the molecules interact with surface cations, line shape, they become more intense at the expense of line shape 2. Spectral features suggest that the hematite surface is either chemically or structurally modified when it interacts with alkaline water. In accordance with the detailed XPS study on iron oxides conducted by Grosvenor et al. (2004), 2p3/2 peaks are expected to exhibit an asymmetric tail to higher binding energy ascribed to surface structures/terminations due to a subtle changes of the crystal field

Fig. 4b shows the comparison between gelatinized starch and starch adsorbed on hematite, revealing several spectral modifications related to the adsorption of polysaccharide molecules on the surface. The most pronounced changes are summarized in Table 2 and indicate considerable shifts in the vibrational modes of carboxyl and alcohol groups. The starch bands assigned to alcohol CeOH moieties coupled to CeC firstly observed at 1015 cm−1 and 990 cm−1 (Fig. 2) shift to 1045 cm−1 and 1022 cm−1 (Fig. 4b). Additional changes are also visualized for carboxyl groups as vibrational bands shift from 1560 to 1530 cm−1 and from 1410 to 1395 cm−1. On the other hand, bands at 1150 and 1080 cm−1 hardly shift while their shape modifies slightly. Therefore, the most pronounced changes in starch adsorbed on hematite are linked to CeOH and COOe vibrational modes, suggesting that starch adsorption may occur through these functional groups. 3.5. XPS spectra of hematite Fig. 5 shows the XPS survey spectra for hematite, hematite stirred in water at pH 10.5, and hematite after starch adsorption. XPS survey

Fig. 5. XPS survey spectra of hematite, hematite stirred in water at pH 10.5, and after starch adsorption at pH 10.5.

100

Minerals Engineering 110 (2017) 96–103

G.F. Moreira et al.

Fig. 6. High-resolution XPS spectra of Fe 2p, O 1s, and C 1s regions for hematite (a), (b) and (c); hematite stirred in water at pH 10.5 (d), (e) and (f); and hematite after starch adsorption at pH 10.5 (g), (h) and (i). The labels indicate components of the main peaks.

assigned to residual water molecules on the surface. Adsorption of starch on hematite shows significant changes of C 1s and O 1s peak components compared to hematite stirred in water at pH 10.5. According to the spectra depicted in Fig. 6i, the C 1s main envelope contains contributions of several peaks, corresponding to CeC/CeH at 284.6 eV, CeO/COOe groups at 287.0 eV, carboxylic groups at 288.1 eV, in addition to the main component at 286.2 eV, assigned to CeOH. It should be noted that the starch adsorbed onto hematite displays pronounced and shifted CeOH components toward a higher binding energy in comparison to the gelatinized starch depicted in Fig. 3. This suggests that as starch adsorbs on the oxide surface, it strongly attaches to the surface, possibly through COO− and CeOH moieties. Further analyses of the O 1s peak are also very instructive for the assignments of the new carbon spectral features. The O 1s spectra can be fitted using four components related to FeeO in hematite at 529.9 eV, FeeOH in FeO(OH) at 531.4 eV, adsorbed water at 534.8 eV, and the most intense component, FeeOH∗ at 533.2 eV (Biesinger et al., 2011). The sharp increase of FeeOH∗ with the polysaccharide adsorption suggests that the molecules might be bound to the surface through this functional group. The fitted parameters are displayed in Table 4.

(Grosvenor et al., 2004). These differences in surface chemical composition may account for the observed peak asymmetry exhibited even by the pristine hematite sample, whose preparation consists exclusively of sample washing. Surface composition is mostly modified by stirring hematite at pH 10.5 as the surface and subsurface species are eventually fully converted to highly hydroxylated iron species. This process leads to the relatively symmetric and broad 2p peaks observed in Fig. 6d. In what follows, starch adsorption on hematite modifies the Fe+3 ligand field again and results in additional asymmetry of the 2p3/2 peak, as indicated when line shape 1 regains intensity at the expense of line shape 2 (Fig. 6g). Additional evidence for this mechanism is provided by the analysis of the O 1s region. The O 1s core level spectra for hematite, hematite stirred in water at pH 10.5, and starch adsorbed on hematite are shown in Fig. 6b, e, and h. Note in Fig. 6b that the peak envelope can be fitted using three components ascribed to FeeO within Fe2O3 at 529.9 eV (label O1), OH groups bonded to FeO within FeO(OH) at 531.1 eV (label O2) and to residual CeO or water molecules on the surface at 532.8 (label O3). Regarding hematite stirred at pH 10.5, there was a slight increase in the OH component, a particular rise in the component labeled as O3 at 533.0 eV, and appearance of a fourth component at 535.0 eV (label O4). Notably, a relative amount of residual carbon (Fig. 6f) appears unchanged, ensuring that O 1s components are unlikely to be linked to spurious surface contaminations. The reported O ls spectra for pure Fe2O3 consist of a single peak at approximately 529.9 eV which, depending on the sample preparation conditions, may display a small shoulder related to adsorbed OH groups (Brundle et al., 1977). There were, however, peaks at higher binding energy resembling FeeOeFe and FeOeOH bonds in oxyhydroxide such as α-FeOOH and γ-FeOOH (Welsh and Sherwood, 1989; Descostes et al., 2000; Asami and Hashimoto, 1977). Therefore, the rise in O3 suggests that FeOeOH with similar chemistry/structure to that of oxyhydroxide is formed by converting hematite to a fully hydroxylated surface. On the other hand, persistence of the O1 component indicates that hematite retains its stoichiometry and that hydroxylation is limited, to some extent, to the sample surface. Thus, stirring the hematite sample in water at pH 10.5 is likely to produce a small amorphous-like FeOOH phase on the hematite surface. Finally, the component labeled as O4 is

4. Discussion Results show that hematite undergoes severe hydroxylation under alkaline conditions, forming an oxyhydroxide surface, called FeO(OH), and other phases alike. The sorption affinity of goethite is directly related to its surface structure and composition and has therefore been a subject of several scientific investigations (Jones et al., 2000; Trainor et al., 2004; Cudennec and Lecerf, 2006; Leeuw and Cooper, 2007). In the theoretical studies of Jones et al. (2000) and Trainor et al. (2004), among others, the authors predicted that the hematite surface under very good hydroxylation conditions is converted to a fully hydroxylated Fe-terminated or O-terminated surface (Ghose et al., 2010; Yin et al., 2007). These studies suggested that the type of hydroxyl coordination is related to the surface activity of metal cations. Consequently, the onset of oxy-hydroxide formation is assumed to rearrange the hematite surface structure and results in a highly organic reactive surface that dictates chemical complexation with starch molecules. Moreover, these 101

Minerals Engineering 110 (2017) 96–103

G.F. Moreira et al.

Table 4 XPS characterization of Fe 2p, O 1s, and C 1s peak components and the corresponding FWHM in eV for hematite, hematite stirred in water at pH 10.5, and hematite after starch adsorption at pH 10.5. Peak components

Hematite

Hematite at pH 10.5

After starch adsorption

Fe 2p

2p3/2 component 1 2p3/2 component 2 2p3/2 satellite

710.3 (2.01) 712.1 (3.52) 718.8 (7.03)

710.2 (2.60) 712.4 (3.09) 718.2 (6.27)

709.9 (2.24) 712.0 (3.35) 718.4 (7.88)

C 1s

CeC/CeH CeOH CeO/COO− COOH

284.6 (1.23) 286.0 (2.13) – 288.6 (1.74)

284.7 (1.43) 285.7 (2.30) – 288.4 (1.80)

284.6 286.2 287.0 288.1

(1.53) (1.6) (1.63) (1.80)

O 1s

FeeO FeeOH CeO/water FeeOH* Water

529.9 (1.24) 531.1 (1.98) 532.8 (1.91) – –

529.9 531.4 – 533.0 535.0

529.9 531.4 – 533.2 534.8

(1.4) (2.11)

findings are in close agreement with the models proposed by Liu et al. (2000), Weissenborn et al. (1995), and Ravishankar et al. regarding the importance of metal hydroxylation for starch adsorption efficiency (Liu et al., 2000; Weissenborn et al., 1995; Ravishankar and Pradip, 1995). Furthermore, the analyses of IR fingerprints indicate that the glycosidic group of starch molecules persists and that alcohols and carboxylic groups shift after adsorption onto hematite. Thus, a probable chemical complexation must involve an oxidized and relatively intact molecule since glycosidic linkages are preserved. XPS results add to this proposal as CeOH, COO−, and COOH moieties are persistent organic groups on the surface after starch adsorption and, consequently, they must be related to those which strongly interact with the oxy-hydroxide surface. Though the direct nature of chemical interaction is difficult to be assigned, the significant shift of CeOH binding energy suggests a specific interaction, such as a coordinate bond between the functional groups of the molecule and the FeO(OH)-like surface via oxygen atoms.

(1.49) (1.96) (1.97) (1.99)

(1.75) (2.15)

References Araujo, A.C., Viana, P.R.M., Peres, A.E.C., 2005. Reagents in iron ores flotation. Miner. Eng. 18, 219–224. Asami, K., Hashimoto, K., 1977. The X-ray photo-electron spectra of several oxides of iron and chromium. Corros. Sci. 17, 559–570. Balajee, S.R., Iwasaki, I., 1969. Trans. Soc. Min. Eng. AIME 244, 401. Biesinger, M.C., Paynec, B.P., Grosvenor, A.P., Laua, L.W.M., Gerson, A.R., 2011. R. St.C. Smart, resolving surface chemical states in XPS analysis of first row transition metals, oxides and hydroxides: Cr, Mn, Fe, Co and Ni. Appl. Surf. Sci. 257, 2717–2730. Brundle, C.R., Chuang, T.J., Wandelt, K., 1977. Core and valence level photoemission studies of iron oxide surfaces and the oxidation of iron. Surf. Sci. 68, 459–468. Capron, I., Robert, P., Colonna, P., Brogly, M., Planchot, V., 2007. Starch in rubbery and glassy states by FTIR spectroscopy. Carbohyd. Polym. 68, 249–259. Casu, B., Reggiani, M., 1964. Infrared spectra of amylose and its oligomers. J. Polym. Sci.: Part C 7, 171–185. Cudennec, Y., Lecerf, A., 2006. The transformation of ferrihydrite into goethite or hematite, revisited. J. Solid State Chem. 179, 716–722. Descostes, M., Mercier, F., Thromat, N., Beaucaire, C., Gautier-Soyer, M., 2000. Use of XPS in the determination of chemical environment and oxidation state of iron and sulfur samples: constitution of a data basis in binding energies for Fe and S reference compounds and applications to the evidence of surface species of an oxidized pyrite in a carbonate medium. Appl. Surf. Sci. 165, 288–302. Desimoni, E., Casella, G.L., Morone, A., Salvi, A.M., 1990. XPS determination of oxygencontaining functional groups on carbon-fibre surfaces and the cleaning of these surfaces. Surf. Interface Anal. 15, 627–634. Filippov, L.O., Severov, V.V., Filippova, I.V., 2013. Mechanism of starch adsorption on Fe–Mg–Al-bearing amphiboles. Int. J. Miner. Process. 123, 120–128. Fuerstenaua, D.W., 2005. Pradip, Zeta potentials in the flotation of oxide and silicate minerals. Adv. Coll. Interface. Sci. 114–115, 9–26. Ghose, S.K., Waychunas, G.A., Trainor, T.P., Eng, P.J., 2010. Hydrated goethite (αFeOOH) (1 0 0) interface structure: Ordered water and surface functional groups. Geochim. Cosmochim. Acta 74, 1943–1953. Grosvenor, A.P., Kobe, B.A., Biesinger, M.C., McIntyre, N.S., 2004. Investigation of multiplet splitting of Fe 2p XPS spectra and bonding in iron compounds. Surf. Interface Anal. 36, 1564–1574. Gupta, R.P., Sen, S.K., 1975. Calculation of multiplet structure of core p -vacancy levels. II. Phys. Rev. B 12, 15–19. Jones, F., Rohl, A.L., Farrow, J.B., van Bronswijk, W., 2000. Molecular modeling of water adsorption on hematite. Phys. Chem. Chem. Phys. 2, 3209–3216. Kar, B., Sahoo, H., Rath, S.S., Das, B., 2013. Investigations on different starches as depressants for iron ore flotation. Miner. Eng. 49, 1–6. Kustova, G.N., Burgina, E.B., Sadykov, V.A., Poryvaev, S.G., 1992. Vibrational spectroscopic investigation of the goethite thermal decomposition products. Phys. Chem. Miner. 18, 379–382. Laskowski, J.S., Liu, Q., O’Connor, C.T., 2007. Current understanding of the mechanism of polysaccharide adsorption at the mineral/aqueous solution interface. Int. J. Miner. Process. 84, 59–68. Leal Filho, L.S., Seidl, P.R., Correia, J.C.G., Cerqueira, L.C.K., 2000. Molecular modelling of reagents for flotation processes. Miner. Eng. 13, 1495–1503. Leeuw, N.H., Cooper, T.G., 2007. Surface simulation studies of the hydration of white rust Fe(OH)2, goethite α-FeO(OH) and hematite α-Fe2O3. Geochim. Cosmochim. Acta 71, 1655–1673. Lima, R.M.F., Brandao, P.R.G., 1999. Investigation on the selectivity in the inverse flotation of iron ores by infrared spectrometry. In: Polymers in Minerals Processing, J. S. (Ed.), Montreal, pp. 139–152. Liu, Q., Zhang, Y., Laskowski, J.S., 2000. The adsorption of polysaccharides onto mineral surfaces: an acid/base interaction. Int. J. Miner. Process. 60, 229–245. Ma, X., Marques, M., Gontijo, C., 2011. Int. J. Miner. Process. 100, 179–183. Martins, M., Lima, N.P., Leal, L.S., 2012. Filho, Depressão de minerais de ferro: influência da mineralogia, morfologia e pH de condicionamento. Revista Escola de Minas 65, 393–399.

5. Conclusions This study demonstrated that starch gelatinization at pH 10.5 amorphizes, highly oxidizes, and exposes reactive organic groups for adsorption onto hematite. These changes could be readily observed as the interaction between starch and hematite significantly modifies the zeta potential profile curve. Furthermore, FTIR and XPS spectral changes served as evidence of surface chemical modification as starch adsorbs onto the oxide and forms a surface chemical complex. The nature of these interactions cannot be related to simply electrostatic interactions and acid-base interaction is thus assumed. Our results support previously proposed models in which a strong starch-hematite interaction occurs via formation of a polysaccharide-metal hydroxide ring. Moreover, we provide additional evidence on the importance of oxidization in gelatinized starch prepared with NaOH as a way to enhance chemical complexation. In view of the current results, the use of chemically modified starches apparently improves the efficiency of reverse flotation of iron ore since the effects of starch oxidation can increase the amount of carbonyl and carboxyl groups in its structure.

Acknowledgments This work was partially supported by CNPq, FAPERJ, Alexander von Humboldt Foundation (AvH), and Max-Planck Gesellschaft Partner Group Programm (MPG). F. Stavale thanks the Surface and Nanostructures Multiuser Lab and the Lab of Biomaterials at CBPF. Finally, we thank Prof. Rafael Cassaro/IQ-UFRJ, Dr. Virginia Nykänen/ ITV-MI, and Dr. Emília Annese/LNLS for several fruitful discussions.

102

Minerals Engineering 110 (2017) 96–103

G.F. Moreira et al.

Szymanski, H.A., Erickson, R.E., 1970. Infrared Absorption Bands. Springer Science +Business Media New York, New York. Tang, M., Liu, Q., 2012. The acidity of caustic digested starch and its role in starch adsorption on mineral surfaces. Int. J. Miner. Process. 112–113, 94–100. Trainor, T.P., Chaka, A.M., Eng, P.J., Newville, M., Waychunas, G.A., Catalano, J.G., Brown Jr., G.E., 2004. Structure and reactivity of the hydrated hematite (0 0 0 1) surface. Surf. Sci. 573, 204–224. Weissenborn, P.K., Warren, L.J., Dunn, J.G., 1995. Selective flocculation of ultrafine iron ore. 1. Mechanism of adsorption of starch onto hematite. Colloids Surf. A: Physicochem. Eng. Aspects 99, 11–27. Welsh, I.D., Sherwood, P.M.A., 1989. Photoemission and electronic structure of FeOOH: distinguishing between oxide and oxyhydroxide. Phys. Rev. B 40, 6386–6392. Yamada, Y., Yasuda, H., Murota, K., Nakamura, M., Sodesawa, T., Sato, S., 2013. Analysis of heat-treated Graphite oxide by X-ray photoelectron Spectroscopy. J. Mater. Sci. 48, 8171–8198. Yan, Y., Miao, J., Yang, Z., Xiao, F.-X., Yang, H.B., Liu, B., Yang, Y., 2015. Carbon nanotube catalysts: recent advances in synthesis, characterization and applications. Chem. Soc. Rev. 44, 3295–3346. Yanga, D., Velamakanni, A., Bozoklu, G., Park, S., Stoller, M., Piner, R.D., Stankovich, S., Jung, I., Field, D.A., Ventrice Jr., C.A., Ruoff, R.S., 2009. Chemical analysis of graphene oxide films after heat and chemical treatments by X-ray photoelectron and Micro-Raman spectroscopy. Carbon 47, 145–152. Yin, S., Ma, X., Ellis, D.E., 2007. Initial stages of H2O adsorption and hydroxylation of Feterminated a-Fe2O3(0 0 0 1) surface. Surf. Sci. 601, 2426–2437.

Max, J.-J., Chapados, C., 2004. Infrared spectroscopy of aqueous carboxylic acids: comparison between different acids and their salts. J. Phys. Chem. A 108, 3324–3337. Pavlovic, S., Brandao, P.R.G., 2003. Adsorption of starch, amylose, amylopectin and glucose monomer and their effect on the flotation of hematite and quartz. Miner. Eng. 16, 1117–1122. Ragheb, A.A., Abdel-Thalouth, I., Tawfik, S., 1995. Gelatinization of starch in aqueous alkaline solutions. Starch/Stärke 47, 338–345. Raju, G.B., Holmgren, A., Forsling, W., 1997. Adsorption of Dextrin at Mineral/Water Interface. J. Colloid Interface Sci. 193, 215–222. Ravishankar, S.A., Pradip, N.K., 1995. Khosla, Selective flocculation of iron oxide from its synthetic mixtures with clays: a comparison of polyacrylic acid and starch polymers. Int. J. Miner. Process. 43, 235–247. Rendon, J., Serna, C.J., 1981. IR spectra of powder hematite: effects of particle size and shape. Clay Miner. 16, 375–381. Salama, W., Aref, M.E., Gaupp, R., 2015. Spectroscopic characterization of iron ores formed in different geological environments using FTIR, XPS, Mössbauer spectroscopy and thermoanalyses. Spectrochim. Acta Part A: Mol. Biomol. Spectrosc. 136, 1816–1826. Shchukarev, A.V., Korolkov, D.V., 2004. XPS study of group IA carbonates. Eur. J. chem. 347, 2. Shrimali, K., Jun, J., Hassas, B.V., Wang, X., Miller, J.D., 2016. The surface state of hematite and its wetting characteristics. J. Colloid Interface Sci. 477, 16–24. Subramanian, K.A., Natarajan, 1988. Some studies on the adsorption behaviour of an oxidised starch onto hematite. Miner. Eng. 1, 241–254.

103