In situ formation of LDH-based nanocontainers on the surface of AZ91 magnesium alloy and detailed investigation of their crystal structure

Ttsin Shulh, M.Serehnov, M.H.Iuzviuk, I.A.Zoklo, P.Krlov, N.Shrngl,D.C.F.Wieln, S.V.Lmk, A.A.Yremhenko, C.Blwert, M.L.Zhelukevih,e

aHelmholtz-Zentrum Hereon, Institute of Surface Science, Max-Planck-Straβe 1, Geesthacht 21502, Germany

b Condensed Matter Department, Petersburg Nuclear Physics Institute Named by B.P.Konstantinov of National Research Center “Kurchatov Institute”, MKR Orlova roshcha 1, Gatchina 188300, Russian Federation

c Helmholtz-Zentrum Hereon, Institute of Metallic Biomaterials, Max-Planck Straβe 1, Geesthacht 21502, Germany

d Department of Materials and Ceramic Engineering, CICECO – Aveiro Institute of Materials, University of Aveiro, Aveiro 3810-193, Portugal

e Faculty of Engineering, University of Kiel, Kaiserstraβe 2, Kiel 24143, Germany

Abstract

Keywords: Layered double hydroxides; Magnesium alloy; Chelating agent; Functional coating.

1.Introduction

Magnesium alloys constitute one of the most important groups of engineering materials [1,2,3].They are highly used for biomedical [4,5], transportation [6,7]and 3C [8]industries.However, their application is limited due to high reactivity, which requires adequate corrosion protection.Historically, the ability of LDH to form conversion coatings with an active response for corrosion protection was widely demonstrated for different kinds of aluminium alloys[9].LDH-based surface or post-treatments were suggested for bare [10,11],anodised [12–14]and plasma electrolytic oxidation (PEO)treated aluminium substrates [15–17].Following pioneering works of Buchheit on LDH formation for protection of aluminium alloys [18,19], a lot of attention was given to further LDH functionalization via intercalation of corrosion inhibitors in the structure [10,11,20–22].This approach was successful and it was further extended for anodised and PEO treated aluminium-based materials [12–17], where a combination of nanocontainers with ceramic layers can provide simultaneous corrosion and wear protection, combining “smart” and barrier protection.Thus, the application of LDHs was revealed as a promising approach for effective corrosion protection and can be extended for magnesium alloys.

LDHs are hydrotalcite-type structures with the general formula [MI/MII1–xMIIIx(OH)2]x+[An–]x/n?yH2O.In this formula,MI, MIIand MIIIare mono-, di- and trivalent metal ions, respectively, An–– interlayer anions andx– stoichiometric coefficient [23].The main advantage of LDH-based conversionlayers is that they can provide not only barrier protection but also active corrosion protectionon demand,due to a high ionexchange ability [24,25].However, this ion-exchange ability depends on the type of LDH.For example, for LDH with monovalent and divalent anions in the interlayer, the following order of ion-exchange ability was determined: LDH-I>LDH-NO3>LDH-Br>LDH-Cl>LDH-F>LDH-OH and LDH-SO4>LDH-CrO4>LDH-HAsO4>LDH-HPO4>LDH-CO3, respectively [26,27].

The high reactivity of magnesium and magnesium alloys requires different routes for LDH formation in comparison with aluminium-based materials [28–36].Even when successful synthesis is performed, it is difficult to produce the most preferential LDH modifications (with nitrate anions, which are easy to be exchanged by functional species).In 2007, Lin et al.reported a facile method for the synthesis of LDH-CO3by simple immersion of AZ91D alloy in a CO2-saturated water solution (pH 4.3) at 50 °C for different time periods.The resulting samples were found to be covered with Mg-Al LDH(Mg6Al2(OH)16CO3?4H2O) when the treatment period was at least 12 h [37].Later, the same authors suggested a twostep methodology with an overall reduced processing time,consisting of immersion in a CO32-/HCO3-solution at pH 6 as a first step, followed by a second step where pH was increased to 11.5 [38].

In the steam-coating method of LDH formation on magnesium substrate, ultrapure water [39–41]or aqueous solutions of aluminium nitrate [42,43]were used.It was shown that a mixture of Mg(OH)2and Mg-Al LDH carbonates was formed on the surface under these conditions.The main advantages of LDH formed via steam coating methods are the significant adhesion to the substrate and homogeneous structure with fine micro asperity.The resulting coatings provide significant corrosion protection for magnesium alloys.However, the use of autoclave is necessary to form LDH layers in steam, and the obtained Mg-Al LDH carbonates have a low ion-exchange ability [39,40,42,43].Besides, not only Mg-Al LDH forms on the surface of magnesium alloys using ultrapure water.Additional Mg(OH)2contaminations with small amounts of MgCO3and AlO(OH) were detected [41].

Previously it was demonstrated thatin situLDH formation on the surface of magnesium alloys in a treatment bath containing chelating agents allows to perform the synthesis without autoclave conditions and/or addition of carbonates in the electrolyte [44].The successful formation of LDH nanocontainers was shown in the presence of disodium salt of nitrilotriacetic acid (NTA) at a temperature of 95 °C and under ambient pressure conditions on the surface of AZ91 magnesium alloy.However, the drawback is that NTA disodium salt is suspected of causing cancer [45–47]; therefore, the search for alternative chelating agents becomes necessary.

An additional issue important to highlight is that it is sometimes difficult to identify which LDH is formed exactly on the surface.For instance, since the distance between cationic layers is almost identical for intercalated carbonate, nitrate and hydroxide between layers, the presented and verified literature data does not show a clear indication of which LDH is formed on magnesium alloy under the conditions of synthesis.This can be explained in a way, that the interlayer distance is defined by the “size” of intercalated anion and in our case is mainly determined by the diameter of oxygen in the anions (R(O) ～1.4[48]).

In the current work, we have selected diethylenetriaminepentaacetic acid (DTPA) pentasodium salt previously identified as chelating agent [49], forin situLDH preparation on AZ91 magnesium alloy since its chelating capacity of Mg2+and Al3+cations is close to that of NTA, according to the thermodynamic stability constants [50,51].Although DTPA is not known to cause cancer, it is indicated in the Safety Data Sheet (SDS) [52]as “suspected of damaging fertility and unborn child”.On the other hand, it was proposed forin vitroinhibition of human cytomegalovirus replication [53].Further studies of DTPA effects on the environment and humans are needed to prove or disprove these points.The aim of current work is to demonstrate that DTPA is a better alternative for LDH synthesis under atmospheric conditions, to identify the best parameters of synthesis and to understand what type of LDH is preferentially formed during the synthesis since it is important for further LDH functionalization.

2.Experimental

2.1.Materials

The LDH conversion layer was prepared on the surface of magnesium alloy AZ91 with the following composition(wt.%): Al – 8.60, Zn – 0.64, Mn – 0.22, Nd – 0.027, Si– 0.0073, Sn – 0.0053, La – 0.0046, Cu – 0.0023, Fe –0.0010, Ni – 0.0003, Be – 0.00076, Ti – 0.0007, Th<0.02,Ce<0.0009, Zr<0.0006, Pb<0.0004, Pr<0.0002, Ag<0.0001, Mg – balance.Samples with following dimensions were used: 15 ?15 ?4 mm.The samples were ground with SiC paper (800 and 1200 grit), rinsed with water and dried with compressed air.

2.2.Chemicals

Aqueous solution of pentasodium salt of diethylenetriaminepentaacetic acid (DTPA, ～ 40% in H2O,C14H18N3Na5O10, Sigma-Aldrich, Germany), aluminium nitrate nonahydrate (Al(NO3)3?9H2O, 98.0–102.0%, Alfa Aesar, Kandel, Germany), sodium nitrate (NaNO3, ≥99%,Carl Roth, Karlsruhe, Germany), sodium hydroxide (NaOH,≥98%, Carl Roth, Karlsruhe, Germany), nitric acid (HNO3,65%, Merck KGaA, Darmstadt, Germany) have been used for LDH synthesis.All chemicals have been used without any further purification.2

.3.Synthesis

The LDH conversion layer was prepared in an aqueous solution containing DTPA chelating agent (0.001, 0.01 and 0.1 M concentrations were tested), 0.05 M Al(NO3)3and 0.25 MNaNO3.First,DTPA pentasodium salt was dissolved in deionized water and then Al(NO3)3and NaNO3salts were added.The pH of the obtained solution was adjusted to 8.0 ± 0.1 and 9.0 ± 0.1 using HNO3solution (10 times water diluted 65% HNO3) and to 10.0 ± 0.1 and 11.0 ± 0.1 using 1 M NaOH solution.The specimens of AZ91 magnesium alloy were immersed in the prepared treatment baths, preheated to different temperatures (room temperature, 60, 80 and 95 °C)under continuous stirring.Different synthesis times were investigated (15 min, 1, 3 or 6 h) in order to select the optimal conditions.After the synthesis, AZ91 samples were rinsed with deionized water and dried under air at room temperature.

2.4.Computational methods

Thermodynamic calculationsof the possibility of LDH formation on the surface of AZ91 magnesium alloy were performed using Hydra-Medusa software vers.of 16 Dec.2010 [54].Stability constants of DTPA as well as magnesium and aluminium complexes with DTPA have been added to the Hydra-Medusa software based on literature data (Log KAl= 18.51 and log KMg= 9.03 were used) [51,55,56].

Modeling of possible LDH crystal structureswas performed using FullProf Suite software using unit cell parameters of LDH-NO3, LDH-CO3and LDH-OH from literature(Table 1 and [57,58]), while unit cells of each expected structure were visualized using VESTA program, ver.3.4.5 [59].

Table 1Literature data of the lattice parameters of LDH-CO3, LDH-OH and LDHNO3.

2.5.Characterization

Phase analysisof the obtained LDH based nanocontainers was performed using Bruker D8 Advance diffractometer(Karlsruhe, Germany).The XRD measurements during the screening of the LDH synthesis parameters were performed with CuKαradiation (40 kV, 40 mA) in the 2θrange from 5 to 30° with step increment equal to 0.02° and exposure time equal to 1 s.For the detailed structural analysis of LDH, the same machine was used, however, the following parameters were selected: 2θrange from 5 to 70° with step increment of 0.02° and exposure time of 80 s.

Investigation of phase changesof the obtained LDH by temperature dependentin situXRD measurements was performed at the beamline P07b at PETRAIII storage ring(Deutsches Electronen Synchrotron, DESY, Hamburg, Germany).The energy of the X-ray beam was 81 keV and the wavelength was 0.1423.The beam size was set to 200 μm × 200 μm.A Perkin Elmer detector XRD1621 with 200 μm pixel size was used.LDH flakes scratched from the surface of AZ91 were used to perform these measurements in order to reduce the influence of the substrate during the measurements.The LDH flakes were placed in a borosilicate glass capillary with a diameter of 1.5 mm and a wall thickness of 19 μm.For the temperature control,a Linkam heating stage was installed into the beamline.The measurements were performed in the temperature range from 25 to 400 °C with a step of 10 °C after 5 min stabilization period.5 2D patterns were measured per temperature point with 2 s expose time each.This patterns were averaged at the end using PyFAI software [60].For easier visual understanding, the scattering angles of the averaged diffractograms were recalculated to CuKαradiation.

The characterization of surface morphologies, crosssections and elemental mapping with a low magnification were performed usingscanning electron microscopeTESCAN VEGA3(Brno,Czech Republic)equipped with energy dispersive X-ray spectrometer (EDS, Heidenrod, Germany).

Mapping of surface and cross-section analysis with higher magnification (indicated in the text below) were performed usingscanning electron microscopeTESCAN LYRA 3 with Dual Beam Field Emission (Brno, Czech Republic) equipped withEnergy dispersive X-ray Spectroscopydetector (AZtec with Ultim & Extreme Detectors).The images were obtained on the polished specimens (diamond paste, 1 μm) without any sputtering of the surface.

Raman spectrawere recorded by a confocal Raman microscope (Bruker, Sentera).All data acquisition was performed at 532 nm laser wavelength, 20 × objective lens, 25 mW of laser power, 50 μm aperture size, and 128 scans with an integration time of 5 s.Bruker OPUS 7.5.18 software was used to evaluate the Raman spectral data.

X-ray photoelectron spectroscopy(XPS) measurements were performed using a KRATOS AXIS Ultra DLD (Kratos Analytical, Manchester, United Kingdom) equipped with a monochromatic AlKαanode working at 15 kV (225W).For the survey spectra, a pass energy of 160 eV was used, while for the region spectra, the pass energy was 20 eV.The investigated area was 700 × 300 μm.For depth profiling, Ar etching was performed.The etching rate was 8 nm/min related to Ta2O5(acceleration voltage 3.8 kV with an extraction current of 160 μA).For all of the samples, charge neutralization was necessary.The evaluation and validation of the data were carried out with the software CASA-XPS version 2.3.18.Calibration of the spectra was done by adjusting the C1s signal to 284.5 eV.For deconvolution of the region files,background subtraction (linear or Shirley) was performed before calculation.

Fig.1.Thermodynamic calculation of the equilibrium composition in the solution containing DTPA, NaNO3, Al(NO3)3 and Mg2+ (as a product of AZ91 substrate dissolution).

Thermogravimetric analysis (TGA)was carried out using a Setaram SetSys 16/18 instrument (sensitivity 0.4 μg, alumina crucibles) in flowing oxygen (flow rate of 30 mL/min)with a constant heating rate of 2 K/min.LDH flakes scratched from the surface of AZ91 were used for measurements in order to reduce the influence of the substrate on the analysis.The initial sample mass was 29.5 mg.Thermogravimetric data were corrected for buoyancy effects by subtracting the corresponding baselines recorded under identical conditions using a dense inert alumina sample.

3.Results and discussion

3.1.Thermodynamic calculations

The possibility of LDH formation on the surface of AZ91 magnesium alloy was estimated via thermodynamic calculations using Hydra-Medusa software.The following concentrations of the components were selected: DTPA – 0.1 M,NaNO3– 0.25 M, Al(NO3)3– 0.05 M; the concentration of Mg2+was estimated as 0.1 M based on previously published work [61].

The results of Hydra-Medusa thermodynamic calculations are presented in Fig.1.The magnesium complex with DTPA(MgC14H18N3O103–) is expected in the pH range from 5 to 12.The concentration of MgC14H18N3O103–increases up to a pH of 7 and reaches its maximum in the pH range from 7 to 10.At pH higher than 10.8, Mg2+is bonded preferentially to hydroxyl ions and Mg(OH)2is formed.However, in the pH range between 9 – 10, advantageous for LDH formation(indicated in blue in Fig.1), about 4% of free Mg2+cations exist and are available for LDH formation.Besides, upon formation of the first LDH phase, the free Mg2+gets depleted.This leads to the shift of the Mg-chelated/Mg-free dynamic equilibrium so that more Mg2+becomes available for LDH formation.In other words, one can expect that, in the case of suitable conditions (temperature, time) [22], LDH formation on the surface of AZ91 is possible in the presence of DTPA as a chelating agent.However, the surface contamination with Mg(OH)2seems to be unavoidable.

3.2.Optimization of synthesis parameters

In order to verify and optimize the LDH formation on the AZ91 surface under ambient pressure, different synthesis conditions (temperature, pH, time, concentration of chelating agent) were tested.Synthesis under each selected condition with respective XRD measurements of the obtained surfaces was tested at least three times in order to verify the reproducibility of the treatment.The initial results of LDH formation were checked using XRD by tracking (003) and (006)LDH reflections.The obtained patterns are presented in Fig.2.

Increasing the temperature (Fig.2a) of the treatment bath leads to an increase in the intensity of the Bragg reflections associated with LDH.It is important to highlight that temperatures below autoclave conditions are sufficient, and the formation of an LDH-based conversion layer takes place.In agreement with the thermodynamic calculations, the experimental results confirm that LDH formation is possible in the aqueous DTPA solution at pH between 9 and 10 (Fig.2b).However, at pH 9 still contaminations are dominating, thus in the frame of the current work, pH 10 was selected as optimal forin situLDH growth with high reproducibility.Studies of the influence of treatment duration on the LDH formation(Fig.2c) revealed that LDH flakes can be detected already after 1 h of hydrothermal treatment (based on XRD results),and homogeneous coverage is produced after 3 h treatment.A synthesis duration of more than 6 h was not investigated in the current work since the aim for further industrial application of the developed technology is to keep necessary resource consumption and cost as low as possible.Unfortunately, 0.001 M DTPA concentration was not sufficient and LDH formation was observed only in solutions with DTPA concentrations between 0.01 and 0.1 M (Fig.2d).Finally, the following conditions were selected as optimal and most reproducible for the based scientific driver study of LDH formation on the AZ91 in the presence of DTPA pentasodium salt:chelating agent concentration - 0.1 M, temperature - 95 °C,pH - 10 and duration of treatment - 6 h.However, milder and industrially relevant conditions also result in LDH formation(e.g.80 °C, 3 h and 0.01 M DTPA) which is even more attractive for industrial application.

Fig.2.XRD patterns of AZ91 sample after treatment in DTPA pentasodium salt solution under different conditions.(a) temperature (0.1 M solution of pentasodium salt of DTPA, pH 10, 6 h); (b) pH (0.1 M, 95°C, 6 h); (c) time (0.1 M, 95 °C, pH 10); (d) concentration of chelating agent (95 °C, pH 10, 6 h).

The surface morphology of AZ91 obtained as a result of the hydrothermal LDH synthesis in the presence of a DTPA chelating agent under previously optimized conditions was analysed using SEM (Fig.3a, b).The observed flakes are typical for LDH nanocontainers, however, the obtained surface is rough.This can be associated with continuous etching of the specimen during LDH growth and uneven dissolution of the substrate during synthesis.A simultaneous decrease in DTPA concentration in the treatment bath, temperature and treatment time also results in the formation of LDH-like flakes(Fig.3c,d).However,in this case,the obtained LDH coverage of the AZ91 surface is not fully uniform, and only “valleys”of LDH-like flakes are visible.

Overall, LDH on the surface of AZ91 magnesium alloy can be formed under ambient pressure in the presence of DTPA pentasodium salt.However, it is not possible to conclude which type of LDH is formed.This uncertainty exists since LDHs with different anions can have the same or similar gallery height values.Thus, (00l) peaks can have the same or quite close positions in the XRD pattern, and it is impossible to directly distinguish between them on the basis of XRD results presented in Fig.2.In order to clarify which anions are contained in the synthesised LDH (nitrate,carbonate or hydroxide), additional characterizations were performed.

3.3.Crystallographic modeling and validation

Using the experimental data shown in Fig.2 as well as extended XRD measurements of bare and LDH-treated AZ91(Fig.4), the crystallographic analysis of LDH nanocontainers obtained under the optimal conditions of synthesis was performed.The obtained diffraction pattern can be indexed in space groupR-3mtypical for LDH.This serves as a confirmation that the LDH-based nanostructure is formed on the surface of AZ91 rather than other layered compounds (e.g.oxonitrates [62,63]).

Based on the obtained XRD results and in order to identify the space available for the anions inside the host Mg-Al hydroxide layer, the interplanar distanced003was calculated using Bragg’s equation:

whereθ– Bragg angle (°),λ– wavelength ().

From the experimental data, the 2θposition of the (003)peak is 11.62(1).This gives a calculatedd003spacing equal to 7.6.Using this value, thec-parameter of the LDH unit cell was obtained using the interrelationc= 3?d003= 22.8.The known values of the lattice parameters of three possible compounds (LDH-CO3, LDH-OH and LDH-NO3) taken from the literature are given in Table 1 with the respective references.These values match either LDH-CO3or LDH-OH, but not LDH-NO3(Table 1 and Fig.6).

Based on the data presented in Table 1, XRD patterns of possible compounds were simulated using FullProf Suite software(Fig.5),and respective unit cells were constructed using Vesta software (Fig.6).As one can see, there is no significant difference between (00l) peak positions for LDH-OH and LDH-CO3.The model patterns of these two compounds fit the experimental curves best, contrary to LDH-NO3.Thus,it is clear that LDH-NO3is not formed under the selected conditions of synthesis.Since Mg-Al LDH-CO3can be syn-thesized in different forms, it is necessary to consider the possibilities of the formation of compounds that have crystal symmetry different from rhombohedralR-3m.Besides, the possibility of the coexistence of LDH-CO3and LDH-OH is not excluded.It is important to highlight that, for simulation of XRD patterns, only space group and lattice constants were employed.For more accurate simulation, it is necessary to use atom coordinates, polytype stacking, etc.

Table 2The elemental composition of LDH-coated surface of AZ91 according to EDS point analysis (Fig.7)

Fig.3.SEM photograph of the AZ91 surface after hydrothermal treatment in the solution containing DTPA pentasodium salt at different conditions: (a, b)0.1 M DTPA at pH 10 and 95 °C for 6 h, (c, d) 0.01 M DTPA at pH 10 and 80 °C for 3 h.

Fig.4.XRD patterns of AZ91 magnesium alloy before (a) and after the treatment (b) in the 0.1 M solution of DTPA at 95 °C and pH 10 for 6 h (LDH-peaks are indexed, non-LDH peaks are marked).

This analysis allows to assume that obtained XRD pattern may belong to two coexisting phases, namely quintinite-3T (Mg4Al2(OH)12CO3?4H2O) [57]and manasseite(Mg6Al2(OH)16(CO3)?4H2O) [58].The first structure can bedescribed by theP3112[57]space group with lattice constantsa= 10.623(1)andc= 22.655(2).The second one belongs to theP63/mmc[58]space group with lattice constantsa= 6.141(3)andc= 15.475(2).These two compounds give a better convergence between experimental and calculated XRD curves during crystallographic fitting (profileRfactor is ca.6).Besides, the experimental XRD pattern may correspond to the combination of coexisting quintinite-3T and Mg-Al LDH-OH, with the unit cell parameters of the latter phase equal to a = 3.042(1)and c = 22.924(4)(profileR-factor is ca.7).Such uncertainty occurs due to the significant broadening of Bragg peaks.In order to obtain more accurate information about crystalline phases, additional studies are needed.

Fig.5.Simulated XRD patterns of (a) LDH-NO3, (b) LDH-OH and (c) LDH-CO3 structures, described by R-3m sp.gr; (d) quintinite and manasseite.Peak positions are labelled by small color lines.

Fig.6.Schematic representation of the unit cells of LDH-NO3, LDH-OH and LDH-CO3 with interplanar distances d003 for each compound.

Fig.7.Elemental mapping of the LDH on the surface of AZ91 obtained in 0.1 M solution of DTPA pentasodium salt at 95 °C and pH 10 during 6 h.

3.4.Elemental analysis

The maps of the main element distribution as well as elemental EDS point analysis of the coatings were recorded and analysed (Figs.7–8).Fig.7 demonstrates the surface view of the obtained LDH-based conversion layer as well as the mapping of the main layer-forming elements.An almost uniform distribution of elements is reached and can be interpreted as homogeneous coverage of the magnesium surface by LDH flakes.However, the aluminium distribution map shows some zones with a higher Al content.This can be associated with the presence of the Mg17Al12beta phase previously detected by the XRD analysis (Fig.4).According to the EDS point analysis results, the oxygen content predominantly prevails in the LDH layer in comparison with carbon and nitrogen(Table 2).This can be explained in a way that oxygen exists both in the host cationic layer as well as in the anionic-based interlayer, while carbonates are located only in the interlayer,and their relative concentration is lower.The presence of a small amount of nitrogen can be associated with the adsorbed nitrate from nitrate-containing treatment bath.

Table 3The elemental composition of the LDH layer obtained from cross-section by the point EDS analysis (Fig.8b)

Fig.8 presents the morphology and elemental distribution mappings obtained from the cross-section of LDH-treated AZ91.The thickness of the obtained LDH coatings varies from 8 to 16 μm.Several interesting effects can be observed in the presented cross-sections.First of all, a clear line of Zn enrichment can be detected at the interface.Less pronounced,but more extended enrichment of aluminium and oxygen saturation near the interface is also visible.In order to confirm and verify the noticed effects, a windowless EDS detector AZtec with Ultim & Extreme Detectors of Tescan Lyra microscope was used (Fig.8b).One can see a clear enrichment of three elements (O, Zn and Al) in the interface zone.The Zn and Al enrichment can indicate the selective dissolution of magnesium during LDH synthesis and the formation of aluminium- and zinc-rich layer under the LDH flakes.Themaps also show a uniform distribution of a low amount of nitrogen through the LDH layer.As was already mentioned previously, this effect can be explained by the adsorption of nitrates from a nitrate-containing solution during LDH formation.

Fig.8.Cross-section morphology and elemental maps of the AZ91 treated in 0.1 M solution of DTPA pentasodium salt at 95 °C for 6 h, pH 10 (a) in low and (b) high magnification.

In order to quantify the elemental composition of the LDH layer, an EDS analysis of the obtained coating was performed(Table 3).The ratio between Mg and Al is determined as approximately equal to 2:1 which is close to that expected for the quintinite-3T structure.From another hand, this ratio between Mg and Al can also be obtained by mixing quintinite-3T, Mg-Al LDH-OH and aluminium compounds (most probably amorphous Al2O3/Al(OH)3not detectable during XRD measurements) from the enriched layer, as demonstrated previously.To clarify this, a more detailed analysis is required.Overall, this result agrees well (except for carbon) with the EDS data collected from the surface(Table 2).The atomic ratio of O:C is close to 2:1 as well.The difference between O and C ratio obtained from the surface and by cross-sectional analysis can be explained by contamination of the surface with CO2from the environment and by remaining traces of epoxy used for embedment.

Table 4Elemental analysis of the LDH-treated surface of AZ91 during the XPS measurement before and after Ar etching

3.5.Analysis of involved species

The LDH flakes formed under optimal conditions were analysed using Raman spectroscopy in order to identify the species present in the structure.The spectra were recorded directly after the LDH synthesis as well as after 7 and 13 days of storage under air conditions (Fig.9).

All the recorded spectra have signatures that can be associated with those of the Mg-Al LDH family (150, 550,1055 cm-1).The 1055 cm-1band signal is known to correspond to the symmetric stretching mode of interlayer anions:NO3-and/or CO32-.Thus, 1055 cm–1can be attributed to CO32-groups loaded into LDH, but possibly also to NO3-.However, as described above, the presence of NO3–in the LDH interlayer was excluded by the XRD analysis, and the detected nitrate can be adsorbed from the treatment bath.The band at 550 cm-1can be associated with hydrogen bond of CO32-with interlayer H2O or with interlayer OH-.The 150 cm-1band corresponds to Me-O bonds.The signals in the range of 3000–3600 cm-1are associated with OH-and H2O [74–80].The obtained patterns demonstrate the presence of LDH-like structures which do not change over almost 2 weeks.

In order to quantify the species intercalated into the LDH,XPS analysis of the obtained LDH coating was performed(Figs.10, 11).XPS results confirm the presence of carbonates in the LDH.The O 1s signal can be attributed to hydroxide, water intercalated in LDH, and carbonate-ions.Additionally, still CO2from the environment might be adhered to the surface and contribute to O 1s signal.One can see that the amount of nitrogen is negligible and can be associated with nitrate possibly adsorbed from the treatment bath,as mentioned above.

The amount of carbon decreases with the increasing depth of the sample(approximately 80–90 nm)(Fig.11).This effect can be explained by the presence of CO2adsorbed from the environment at the very top surface.

The atomic ratio of magnesium to aluminium is around 1:1 according to the XPS analysis (Table 4).This difference with the EDS results for the top view can be explained in a way that, during the XPS measurements, only a thin layer on the surface was analysed (3–5 nm).In contrast, the penetration depth during the EDS analysis is a couple of micrometers with a huge accelerated volume, and a more averaged value is obtained.

Overall, based on the results of XRD, XPS and EDS as well as crystallographic modeling, we cannot exclude that synthesized structures contain interlayer hydroxides.The difference in the elemental concentrations (both from the LDH host and the interlayer) measured by different methods can be associated with different penetration depths of analysis during testing,as well as with different deviation of measurement results.The “surroundings” of the measured points are also playing their role during analysis: CO2from the atmosphere or carbon from epoxy resin increase the C signal near the surface, while the composition of the substrate influences the measured data near the interface.Nevertheless, the following LDH formula can be considered as the most general oneunder conditions of synthesis:

Fig.9.Raman spectra of LDH on the surface of AZ91 obtained directly after synthesis and after storage under ambient air conditions and room temperature for 7 and 13 days.

Fig.10.XRS survey spectra before Ar etching (a), after 600 s Ar etching (b) and after 12,000 s Ar etching (c) of the AZ91 sample treated in 0.1 M DTPA pentasodium salt solution at 95 °C and pH 10 for 6 h.

Fig.11.Depth profile of the LDH-treated surface over 12,000 s of Ar etching.

wherexis responsible for the ratio between magnesium and aluminium,ais associated with the ratio between carbonates and hydroxides in the LDH interlayer,γandχrepresent intercalated and physically adsorbed water, respectively.For this equation, it is important to highlight thatxis always in a range between 0 and 1, whileais positive and alwaysa

In order to verify and confirm (or exclude) the inclusion of two different anions into the LDH galleries (hydroxide and carbonate), thermogravimetric analysis (TGA) of the LDH flakes was performed in a flow of oxygen (Fig.12).Such conditions were selected to avoid interaction between LDH structure and possible traces of magnesium alloy originating from the substrate and resulting in the formation of spineltype compounds with undefined chemical composition.

Taking into account the literature data of the thermal behaviour of LDH [81–87], the observed weight losses on heating can be assigned to the following processes:

1.<100°C: desorption of physically adsorbed water (parameterχin Eq.(3))

The structure of LDH remains unchanged in this temperature range.All the anions from the interlayer responsible for the charge compensation of the cationic host layers stay inside.

2.100°C

Water molecules are released from the interlayer gallery.The structure of LDH layers starts to collapse.However,water molecules are neutral species; therefore, no charge redistribution between the cationic host layer and interlayer should be expected.

3.210°C

where H2O(hostlayer)is associated with water released from the double hydroxide host cationic layer, and H2O(interlayer)indicates water originating from the charge balanced anions in the interlayer.

During stage 3, a full collapse of the LDH structure takes place.Two principal processes occur during this time (indicated as 3a and b in Fig.12) [84]:

- The release of initially charged species (hydroxides and carbonates) from the interlayer and release hydroxides from the host layer.

- The charge compensation associated with the released species happens due to the recrystallization of the host cationic layers and the formation of oxides and mixed oxides (spinel-type) structures.

Fig.12.TGA (a) and differential thermogravimetric (DTG) (b) curves of LDHs obtained from the surface of AZ91 after hydrothermal synthesis in the presence of DTPA pentasodium salt as a chelating agent.

4.460-500 °C: oxidation of Mg [88]

This process is not directly associated with the LDH decomposition and takes place only as a parallel secondary reaction: the LDH flakes were contaminated with traces of soft magnesium substrate at the stage of LDH powder preparation by scratching.Oxidation of magnesium in an oxygen atmosphere results in a minor weight gain at 460–500 °C.

For the quantification of anions released during the decomposition phase (3), fitting of the respective part of the DTG curve was performed with a Lorenz functions (Fig.13).

Careful analysis showed that the adequate fitting of the experimental data requires three contributions, which can be assigned to the release of three kinds of species in the temperature range between 210 and 460 °C.Available literature reports suggest that the decomposition of nitrate occurs at slightly higher temperatures (400–600 °C) compared to carbonate (300–400 °C) [82,83].Therefore, considering the temperature range corresponding to the collapse of the structure,the obtained TGA data support the assumption that the LDH structure is composed preferentially of hydroxide and carbonate anions rather than nitrate groups in the interlayer.This result is in good agreement with XRD and XPS data reported above.

In order to identify and assign these three decomposition sub-steps to particular species released from the structure,first of all, the identification of respective weight fractions was performed via integration of areas of identified DTG peaks.The calculated data showed that the following mass fraction can be associated with each peak: A peak – 8.81 wt.%, B peak – 3.73 wt.% and C peak – 13.22 wt.% (total initial mass of the used sample was 29.5 mg).

Fig.13.Fitting of the DTG curve using three Lorenz distribution functions (A–C – resolved curves).

In the next step, the obtained weight losses were associated with the actual species that could be released from the structure (Table 5).

Based on the ratio between anions, one can assume three possible combinations of hydroxides and carbonates existing in the LDH interlayer (cases: 2, 3, 6 in Table 5).The other three combinations (cases: 1, 4 and 5) can be excluded from the consideration since the amount of water associated with the interlayer hydroxides is larger than the amount of water associated with the host layer hydroxides, according to Eqs.(3)–(5).The calculation of coefficients in formula (2) would result in a parametera > 2, which is not possible.

Table 5Calculation of the possible structure of LDH

Additionally, the analysis of literature data [26,87,89]also allows us to exclude more cases from consideration:

- Following the thermal stability of the LDH structure,one can expect that water associated with the host layer will be released only after the release of both anions from the interlayer, thus case 3 can be excluded.

The higher charge of the carbonate anion in comparison with the hydroxide anion allows for more effective charge compensation of host cationic layers and thus a higher affinity to the host layers [26],thus case 6 is not possible.

Overall, one can expect the following order of species release on heating LDH:

However, the temperature ranges where the release of interlayer carbonates and hydroxides from the host layer takes place are very close [90].

Using the TGA results, in the frame of this work, it was confirmed that the LDH-OH/CO3phase suggested by the crystallographic modeling is the most probable phase forming during the LDH synthesis on the AZ91 magnesium alloy in the presence of the DTPA chelating agent.

3.6.Temperature dependent in situ XRD measurements

A similar LDH structure collapse is observed during the XRD measurements under elevated temperature(Fig.14).Following the (003) and (006) LDH reflections, one can see that there is a phase transformation occurring at ca.260–270 °C.This is seen by the shift and disappearance of typical LDH reflections.The observed results are in good agreement with the TGA data suggesting that the release of interlayer water and initial collapse of the LDH structure starts from 210 °C.

Summarizing, the crystallographic modeling demonstrated the possibility of coexistence of two phases: either quintinite-3T and manasseite or quintinite-3T and MgAl LDH-OH.Additional analytical studies did not exclude the presence of hydroxides anions in the LDH interlayer, while thermogravimetric analysis supports the possible formation of quintinite-3T and MgAl LDH-OH mixture.Thus, the overall general formula of LDH synthesized in the presence of DTPA as a

chelating agent can be represented as Mg-Al LDH-OH/CO3.The synthesis can be performed under ambient pressure at 95 °C without autoclave conditions.During the LDH formation in a hydrothermal bath, selective dissolution of magnesium takes place, and enrichment of the interface with zincand aluminium-based compounds may occurs.This finding needs to be verified in the future.Following the literature data published previously,one can expect such behaviour during conversion treatment [91,92].Moreover, redeposition of alloying elements from an active substrate (e.g.magnesium[93]and aluminium [94]) was already demonstrated in different processes.However, from the point of future corrosion protections of magnesium, the obtained results are very important.Interface enrichment with zinc and aluminium compounds can cause the acceleration of magnesium corrosion due to the formation of local galvanic couples where magnesium as a more active metal dissolves preferentially.As a result,parental LDH itself may not provide effective improvement of corrosion resistance and needs to be combined with suitable corrosion inhibitors loaded into the LDH.An important point here is that probably corrosion inhibitors suitable for zinc and aluminium,rather than just species inhibiting corrosion of magnesium,should be investigated.This assumption will be verified in our future works.

Fig.14.XRD patterns illustrating the thermal decomposition of LDH due to heating.The data were collected from LDH flakes scratched from the surface of AZ91 after hydrothermal synthesis in the presence of DTPA pentasodium salt as a chelating agent.The image on the right shows the magnified sections of the spectra.The patterns are shifted vertically relative to each other for better visualization.

4.Conclusions

LDH was synthesisedin situon the surface of AZ91 magnesium alloy in the presence of pentasodium salt of DTPA in the electrolyte.As the optimal conditions of LDH formation,the following parameters were identified:DTPA concentration equal to 0.1 M, temperature of 95 °C, and pH of the solution equal to 10.The optimal duration of LDH growth was defined as 6 h.For further industrial application, the synthesis can be additionally optimized to obtain LDH under milder conditions (e.g.0.01 M DTPA concentration, 80 °C and 3 h).It was shown that LDH-based mixture with the general formula Mg-Al LDH-OH/CO3is formed on the surface under these conditions of synthesis as verified by the combination of SEM-EDS, XRD, XPS, Raman spectroscopy, TGA and crystal modeling simulation.

During the LDH formation in hydrothermal bath, selective dissolution of magnesium might occur and, as a consequence,enrichment of the interface with zinc- and aluminium-based compounds takes place.Thus, additional investigation of corrosion inhibitors suitable for intercalation into LDH intergallery for interface stabilization in the presence of corrosive media should be performed.

Contribution

Synthesis of LDH, SEM and cross-section characterization of obtained structures, writing first draft – T.S., selection of chelating agents – M.S.and S.V.L., calculation of Medusa diagram – T.S., M.S.and S.V.L., XRD characterization of obtained surfaces – M.S., D.C.F.W., M.H.I.and I.A.Z., Raman spectroscopy and XPS characterization – N.S., thermogravimetric analysis – A.A.Y.and P.K., supervision – C.B.and M.L.Z.All authors discussed the results and edited this article.

Declaration of Competing Interest

No conflict of interests can be declared.

Acknowledgments

This research was carried out with the financial support of the I2B fund (Helmholtz Association) in frame of MUFfin project as well as ACTICOAT project in frame of Era.Net RUS Plus Call 2017.A.A.Y.gratefully acknowledges financial support within the project CICECO - Aveiro Institute of Materials (UIDB/50011/2020 & UIDP/50011/2020) financed by national funds through the FCT/MCTES and when appropriate co-financed by FEDER under the PT2020 Partnership Agreement.All authors thank PETRA III (DESY, Hamburg)for the beamtime at P07b granted.The technical support of Mr.Gert Wiese during work with TESCAN LYRA 3 microscope and Mr.Ulrich Burmester during specimens preparation is gratefully acknowledged.

References-

[1]J.Song, J.She, D.Chen, F.Pan, J.Magnes.Alloy.8 (2020) 1–41,doi:10.1016/j.jma.2020.02.003.

[2]T.Xu, Y.Yang, X.Peng, J.Song, F.Pan, J.Magnes.Alloy.7 (2019)536–544, doi:10.1016/j.jma.2019.08.001.

[3]H.Yang, L.Wu, B.Jiang, W.Liu, J.Song, G.Huang, D.Zhang, F.Pan,J.Mater.Sci.Technol.62 (2021) 128–138, doi:10.1016/j.jmst.2020.05.067.

[4]V.K.Bommala, M.G.Krishna, C.T.Rao, J.Magnes.Alloy.7 (2019)72–79, doi:10.1016/j.jma.2018.11.001.

[5]Z.Z.Yin, W.C.Qi, R.C.Zeng, X.B.Chen, C.D.Gu, S.K.Guan,Y.F.Zheng, J.Magnes.Alloy.8 (2020) 42–65, doi:10.1016/j.jma.2019.09.008.

[6]M.K.Kulekci, Int.J.Adv.Manuf.Technol.39 (2008) 851–865, doi:10.1007/s00170-007-1279-2.

[7]S.Ehrenberger, H.Dieringa, H.E.Friedrich, https://cdn.ymaws.com/www.intlmag.org/resource/resmgr/docs/membership_central/mg_industry_studies/IMA_Life_Cycle_Analysis_Stud.pdf, 2013 (accessed 09 April 2020).

[8]S.N.Mathaudhu, A.A.Luo, N.R.Neelameggham, E.A.Nyberg,W.H.Sillekens, Essential Reading in Magnesium Technology, Springer International Publishers, 2016, doi:10.1007/978-3-319-48099-2.

[9]A.C.Bouali, M.Serdechnova, C.Blawert, J.Tedim, M.G.S.Ferreira,M.L.Zheludkevich, Appl.Mater.Today 21 (2020) 100857–100898,doi:10.1016/j.apmt.2020.100857.

[10]J.Tedim,M.L.Zheludkevich,A.N.Salak,A.Lisenkov,M.G.S.Ferreira,J.Mater.Chem.21 (2011) 15464–15470, doi:10.1039/c1jm12463c.

[11]J.Tedim, M.L.Zheludkevich, A.C.Bastos, A.N.Salak, A.D.Lisenkov,M.G.S.Ferreira, Electrochim.Acta 117 (2014) 164–171, doi:10.1016/j.electacta.2013.11.111.

[12]B.Kuznetsov, M.Serdechnova, J.Tedim, M.Starykevich, S.Kallip,M.P.Oliveira, T.Hack, S.Nixon, M.G.S.Ferreira, M.L.Zheludkevich,RSC Adv.6 (2016) 13942–13952, doi:10.1039/c5ra27286f.

[13]Y.Li, S.Li, Y.Zhang, M.Yu, J.Liu, J.Alloy.Compd.630 (2015)29–36, doi:10.1016/j.jallcom.2014.12.176.

[14]D.Mata, M.Serdechnova, M.Mochedano, C.L.Mendis, S.V.Lamaka,J.Tedim, T.Hack, S.Nixon, M.L.Zheludkevich, RSC Adv.7 (2017)35357–35367, doi:10.1039/c7ra05593e.

[15]M.Mohedano, M.Serdechnova, M.Starykevich, S.Karpushenkov,A.C.Bouali, M.G.S.Ferreira, M.L.Zheludkevich, Mater.Des.120(2017) 36–46, doi:10.1016/j.matdes.2017.01.097.

[16]F.Chen,P.Yu, Y.Zhang,J.Alloy.Compd.711(2017)342–348,doi:10.1016/j.jallcom.2017.04.016.

[17]A.C.Bouali, E.A.Straumal, M.Serdechnova, D.C.F.Wieland,M.Starykevich,C.Blawert,J.U.Hammel,S.A.Lermontov,M.G.S.Ferreira, M.L.Zheludkevich, Appl.Surf.Sci.494 (2019) 829–840, doi:10.1016/j.apsusc.2019.07.117.

[18]R.G.Buchheit, S.B.Mamidipally, P.Schmutz, H.Guan, Corrosion 58(1) (2002) 3–14, doi:10.5006/1.3277303.

[19]W.Zhang, R.G.Buchheit, Corrosion 58 (7) (2002) 591–600, doi:10.5006/1.3277650.

[20]M.L.Zheludkevich, S.K.Poznyak, L.M.Rodrigues, D.Raps, T.Hack,L.F.Dick, T.Nunes, M.G.S.Ferreira, Corros.Sci.52 (2010) 602–611,doi:10.1016/j.corsci.2009.10.020.

[21]T.Stimpfling, F.Leroux, H.Hintze-Bruening, Colloids Surf.A 458(2014) 147–154, doi:10.1016/j.colsurfa.2014.01.042.

[22]M.Serdechnova, A.N.Salak, F.S.Barbosa, D.E.L.Vieira, J.Tedim,M.L.Zheludkevich, M.G.S.Ferreira, J.Solid State Chem.233 (2016)158–165, doi:10.1016/j.jssc.2015.10.023.

[23]D.G.Evans, R.C.T.Slade, in: Structure and Bonding, 119, Springer-Verlag, 2006, pp.1–87, doi:10.1007/430_005.

[24]F.Wu, J.Liang, Z.Peng, B.Liu, Appl.Surf.Sci.313 (2014) 834–840,doi:10.1016/j.apsusc.2014.06.083.

[25]F.Zhang, Z.G.Liu, R.C.Zeng, S.Q.Li, H.Z.Cui, L.Song, E.H.Han,Surf.Coat.Technol.258 (2014) 1152–1158, doi:10.1016/j.surfcoat.2014.07.017.

[26]S.Miyata, Clays Clay Miner.31 (1983) 305–311, doi:10.1346/CCMN.1983.0310409.

[27]S.M.Auerbach, K.A.Carrado, P.K.Dutta, M.Dekker, Handbook of Layered Materials, Basel Inc., New York, 2004 edited by.

[28]L.Guo, W.Wu, Y.Zhou, F.Zhang, R.Zeng, J.Zeng, J.Mater.Sci.Technol.34 (2018) 1455–1466, doi:10.1016/j.jmst.2018.03.003.

[29]L.Wu, X.Ding, Z.Zheng, A.Tang, G.Zhang, A.Atrens, F.Pan, J.Mater.Sci.Technol.64 (2021) 66–72, doi:10.1016/j.jmst.2019.09.031.

[30]J.Chen, L.Wu, X.Ding, Q.Liu, X.Dai, J.Song, B.Jiang, A.Atrens,F.Pan, J.Mater.Sci.Technol.64 (2021) 10–20, doi:10.1016/j.jmst.2019.10.007.

[31]T.Hu, Y.Ouyang, Z.H.Xie, L.Wu, J.Mater.Sci.Technol.92 (2021)225–235, doi:10.1016/j.jmst.2021.03.021.

[32]J.Chen, J.Feng, L.Yan, H.Li, C.Xiong, S.Ma, J.Magnes.Alloy.9(2021) 1019–1027, doi:10.1016/j.jma.2020.05.019.

[33]X.Wang,C.Jing,Y.Chen,X.Wang,G.Zhao,X.Zhang,L.Wu,X.Liu,B.Dong, Y.Zhang, J.Magnes.Alloy.8 (2020) 291–300, doi:10.1016/j.jma.2019.11.011.

[34]W.Wu,X.Sun,C.L.Zhu,F.Zhang,R.C.Zeng,Y.H.Zou,S.Q.Li,Prog.Org.Coat.147 (2020) 105746, doi:10.1016/j.porgcoat.2020.105746.

[35]X.Sun, Q.S.Yao, Y.C.Li, F.Zhang, R.C.Zeng, Y.H.Zou, S.Q.Li,Front Mater Sci 14 (2020) 426–441, doi:10.1007/s11706-020-0522-8.

[36]Z.M.Qiu, R.C.Zeng, F.Zhang, L.Song, S.Q.Li, Trans.Nonferrous Met.Soc.China, 30 (2020), 2967–2979.10.1016/S1003-6326(20)65435-8.

[37]J.K.Lin, C.L.Hsia, J.Y.Uan, Scr.Mater.56 (2007) 927–930, doi:10.1016/j.scriptamat.2007.02.020.

[38]J.K.Lin, J.Y.Uan, Corros.Sci.51 (2009) 1181–1188, doi:10.1016/j.corsci.2009.02.007.

[39]T.Ishizaki, S.Chiba, K.Watanabe, H.Suzuki, J.Mater.Chem.A 1(2013) 8968–8977, doi:10.1039/c3ta11015j.

[40]T.Ishizaki, S.Chiba, H.Suzuki, ECS Electrochem.Lett.2 (5) (2013)C15–C17, doi:10.1149/2.006305eel.

[41]K.Nakamura,M.Tsunakaw,Y.Shimada,A.Serizawa,T.Ishizaki,Surf.Coat.Technol.328 (2017) 436–443, doi:10.1016/j.surfcoat.2017.08.060.

[42]N.Kamiyama, G.Panomsuwan, E.Yamamoto, T.Sudare, N.Saito,T.Ishizaki, Surf.Coat.Technol.286 (2016) 172–177, doi:10.1016/j.surfcoat.2015.11.051.

[43]T.Ishizaki, N.Kamiyama, K.Watanabe, A.Serizawa, Corros.Sci.92(2015) 76–84, doi:10.1016/j.corsci.2014.11.031.

[44]T.N.Shulha, M.Serdechnova, S.V.Lamaka, D.C.F.Wieland,K.N.Lapko,M.L.Zheludkevich,Sci.Rep.8(2018)16409,doi:10.1038/s41598-018-34751-7.

[45]National Toxicology Program, Public Health Service.https://ntp.niehs.nih.gov/ntp/roc/content/profiles/nitrilotriaceticacid.pdf, 2016 (accessed 29 April 2021).

[46]Safety Data Sheet N9877, https://www.sigmaaldrich.com/DE/en/sds/sial/n9877, 2019 (accessed 23 Juni 2021).

[47]Safety Data Sheet N0128, https://www.sigmaaldrich.com/DE/en/sds/sigma/n0128, 2019 (accessed 23 Juni 2021).

[48]R.D.Shannon, Acta Cryst.A32 (1976) 751–767, doi:10.1107/S0567739476001551.

[49]S.V.Lamaka, B.Vaghefinazari, D.Mei, R.P.Petrauskas, D.H?che,M.L.Zheludkevich, Corros.Sci.128 (2017) 224–240, doi:10.1016/j.corsci.2017.07.011.

[50]R.M.Smith, A.E.Martel, Critical Stability Constants, 6, Plenum Press,New York, 1989 Second Supplement.

[51]J.A.Dean, Lange’s handbook of chemistry, 15th ed., McGraw-Hill Inc.,New York, 1999.

[52]Safety Data Sheet D6518, https://www.sigmaaldrich.com/DE/en/sds/sial/d6518, 2021 (accessed 23 Juni 2021).

[53]J.Cinatl, F.Hoffmann, J.Cinatl, B.Weber, M.Scholz, H.Rabenau,F.Stineker, H.Kabickova, M.Blasko, H.W.Doerr, Antiviral Res.31(1996) 23–34, doi:10.1016/0166-3542(95)00833-0.

[54]I.Puigdomenech, 16 December 2010.https://sites.google.com/site/chemdiagr/home (accessed 2017).

[55]V.N.Tihonov, Ananliticheskaya Himiya Aluminiya (in Russian), Nauka,Moskow, 1971.

[56]R.Pˇribil, Analytical Applications of EDTA and Related Compounds,1st ed., Pergamon Press, Oxford, New York, 1972.

[57]Quintinite-3T Mineral Data.http://www.webmineral.com/data/Quintinite-3T.shtml#.XoNjP4gzY2w (accessed April 2020).

[58]Manasseite Mineral Data.http://webmineral.com/data/Manasseite.shtml(accessed April 2020).

[59]K.Momma,F.Izumi,J.Appl.Crystallogr.44(2011)1272–1276,doi:10.1107/S0021889811038970.

[60]J.Kieffer, D.Karkoulis, J.Phys.Conf.Ser.425 (2013) 202012, doi:10.1088/1742-6596/425/20/202012.

[61]O.V.K.aravai, A.C.Bastos, M.L.Zheludkevich, M.G.Taryba,S.V.Lamaka, M.G.S.Ferreira, Electrochim.Acta 55 (2010) 5401–5406,doi:10.1016/j.electacta.2010.04.064.

[62]C.Ehrhardt, M.Gjikaj, W.Brockner, Thermochim.Acta 432 (2005)36–40, doi:10.1016/j.tca.2005.04.010.

[63]I.V.Morozov, K.O.Znamenkov, Y.M.Korenev, O.A.Shlyakhtin,Thermochim.Acta 403 (2003) 173–179, doi:10.1016/S0040-6031(03)00057-1.

[64]S.Cherepanova, N.Leont’eva, V.Drozdov, V.Doronin, Acta Crystallogr., Sect.A Found.Adv.72 (2016) 651–659, doi:10.1107/S2053273316014868.

[65]S.Babay, M.Toumi, J.Nanotechnol.Mater.Sci.4 (2) (2017) 112–120,doi:10.15436/2377-1372.17.1688.

[66]M.Bellotto, B.Rebours, O.Clause, J.Lynch, D.Basin, E.Elkaim, J.Phys.Chem.100 (1996) 8527–8534, doi:10.1021/jp960039j.

[67]A.V.Lukashin, S.V.Kalinin, M.P.Nikiforov, A.I.Zhirov, L.I.Kheifets,V.I.Privalov, A.A.Vertegel, MRS Online Proc.Libr.547 (1998) 239–242, doi:10.1557/PROC-547-239.

[68]S.J.Mills, A.G.Christy, J.M.R.GéNin, T.Kameda, F.Colombo, Mineral.Mag.76 (5) (2012) 1289–1336, doi:10.1180/minmag.2012.076.5.10.

[69]W.Tongamp, Q.Zhang, F.Saito, J.Mater.Sci.42 (2007) 9210–9215,doi:10.1007/s10853-007-1866-5.

[70]S.L.Wang, P.C.Wang, Colloids Surf.A 292 (2007) 131–138, doi:10.1016/j.colsurfa.2006.06.014.

[71]S.Mondal, S.Dasgupta, K.Maji, Mater.Sci.Eng.C 68 (2016) 557–564, doi:10.1016/j.msec.2016.06.029.

[72]A.N.Ay, B.Zümreoglu-Karan, L.Mafra, Z.Anorg.allg.Chem.635(2009) 1470–1475, doi:10.1002/zaac.200801287.

[73]Q.Z.Yang, J.Yang, C.K.Zhang, Int.J.Pharm.326 (2006) 148–152,doi:10.1016/j.ijpharm.2006.06.037.

[74]IRDG Infrared & Raman Discussion Group, Internet J.Vib.Spectrosc.(2021) http://www.irdg.org/the-infrared-and-raman-discussion-group/ijvs/reference-spectra/.(accessed April 2021).

[75]Spectral Database for Organic Compounds SDBS.https://sdbs.db.aist.go.jp/sdbs/cgi-bin/direct_frame_top.cgi (accessed April 2021).

[76]RRUFF, Quintinite R070709.https://rruff.info/chem=Mg,%20C,%20O/display=default/R070709 (accessed April 2021).

[77]RRUFF, Hydrotalcite R060527.https://rruff.info/chem=Mg,%20C,%20O/display=default/R060527 (accessed April 2021).

[78]RRUFF, Hydrotalcite R050457.https://rruff.info/chem=Mg,%20C,%20O/display=default/R050457 (accessed April 2021).

[79]F.Theiss, A.López, R.L.Frost, R.Scholz, Spectrochim.Acta A 150(2015) 758–764, doi:10.1016/j.saa.2015.06.002.

[80]J.T.Kloprogge, D.Wharton, L.Hickey, R.L.Frost, Am.Mineral.87(2002) 623–629, doi:10.2138/am-2002-5-604.

[81]S.Miyata, Clays Clay Miner.28 (1) (1980) 50–56.

[82]A.W?egrzyn, A.Rafalska-?asocha, D.Majda, R.Dziembaj,H.Papp, J.Therm.Anal.Calorim.99 (2010) 443–457, doi:10.1007/s10973-009-0190-5.

[83]Z.P.Xu, H.C.Zeng, Chem.Mater.13 (2001) 4564–4572, doi:10.1021/cm010347g.

[84]T.Stanimirova,N.Piperov,N.Petrova,G.Kirov,Clay Miner.39(2004)177–191, doi:10.1180/0009855043920129.

[85]L.Pesic, S.Salipurovic, V.Markovic, D.Vucelic, W.Kagunya,W.Jones, J.Mater.Chem.2 (10) (1992) 1069–1073, doi:10.1039/JM9920201069.

[86]R.L.Frost, W.N.Martens, K.L.Erickson, J.Therm.Anal.Calorim.82(2005) 603–608.

[87]R.Sasai, H.Sato, M.Sugata, T.Fujimura, S.Ishiha, K.Deguchi,S.Ohki, M.Tansho, T.Shimizu, N.Oito, M.Numoto, Y.Fujii,S.Kawaguchi, Y.Matsuoka, K.Hagura, T.Abe, C.Moriyoshi, Inorg.Chem.58 (2019) 10928–10935, doi:10.1021/acs.inorgchem.9b01365.

[88]G.Moser, C.Sch?nnenbeck, V.Tschamber, A.Brillard, J.F.Brilhas,Combust.Flame 230 (2021) 111419, doi:10.1016/j.combustflame.2021.111419.

[89]W.T.Reichle, S.Y.Kang, D.S.Everhardt, J.Catal.101 (1986) 352–359,doi:10.1016/0021-9517(86)90262-9.

[90]R.Yahyaoui, P.E.S.Jimenes, L.A.P.Maqueda, K.Nahdi, J.M.C.Luque,Thermochim.Acta 667 (2018) 177–184, doi:10.1016/j.tca.2018.07.025.

[91]J.Han, K.Ogle, J.Electrochem.Soc.164 (14) (2017) C952–C961,doi:10.1149/2.0341714jes.

[92]P.Volovitch, M.Serdechnova, K.Ogle, Corrosion 68 (6) (2012) 557–570, doi:10.5006/i0010-9312-68-6-557.

[93]D.H?che, C.Blawert, S.V.Lamaka, N.Scharnagl, C.Mendis,M.L.Zheludkevich, Phys.Chem.Chem.Phys.18 (2016) 1279–1291,doi:10.1039/c5cp05577f.

[94]A.C.Bouali, M.H.Iuzviuk, M.Serdechnova, K.A.Yasakau, D.Drozdenko, A.Lutz, K.Fekete, G.Dovzhenko, D.C.F.Wieland, H.Terryn,M.G.S.Ferreira, I.A.Zobkalo, M.L.Zheludkevich, J.Phys.Chem.C 125 (21) (2021) 11687–11701, doi:10.1021/acs.jpcc.1c02281.