An In Vitro Corrosion Study of Open Cell Iron Structures with PEG Coating for Bone Replacement Applications

Iron-based substrates with polyethylene glycol coating were prepared as possible materials for biodegradable orthopedic implants. Biodegradable materials that provide mechanical support of the diseased tissue at the time of implanting and then disappear gradually during the healing process are sometimes favored instead of permanent implants. The implant degradation rate should match the time of the tissue regrowth. In this work, the degradation behavior of iron-based foams was studied electrochemically during immersion tests in Hanks’ solution. The corrosion rate of the polyethylene glycol-coated samples increased and the corrosion potential shifted to more negative values. This indicates an enhanced degradation rate as compared to the uncoated material, fulfilling the goal of being able to tune the degradation rate. It is the interfacial interaction between the hydrophilic polymer layer and the iron surface that is responsible for the enhanced oxidation rate of iron.


Introduction
Degradable materials represent a new generation of implantable systems.These materials are currently intensively studied, because of their potential use as implants in orthopedic and cardiovascular surgery [1].These novel kinds of implants can adapt to the needs of the human body in which they are implanted.Their role is to support initially the formation of the healing injured tissues and subsequently to degrade gradually, while the tissue function is repaired [2][3][4][5][6].The degradable Metals 2018, 8, 499 2 of 21 implant approach brings great benefits over the techniques used today.The biodegradable implants can enhance and facilitate the treatment and reduce the number of invasive procedures, as well as save costs [5][6][7][8][9][10][11][12][13].
Biomaterials suitable for degradable bone implant applications must exhibit appropriate mechanical [14] and biological properties [15][16][17].They must be flexible, mechanically tough, nontoxic, biocompatible [18][19][20], and their corrosion rate must match the intended purpose [21][22][23].Such materials are based on biocompatible metals such as magnesium, iron, and zinc [22,24,25] which gradually corrode.Scientists make great efforts to prevent the corrosion of industrial structures [26,27] and permanent medical devices [28].However, the corrosion process is desired in the degradation of temporary metal implants.There are a number of publications about degradable orthopedic implants based on magnesium [25,29]; however, fewer studies exist on biodegradable bone implants based on iron [30][31][32][33].Iron can provide certain advantages over magnesium-based devices, such as higher strength, hardness, wear resistance, and toughness.Magnesium has also a drawback because its corrosion rate is rather fast and during its oxidation a significant amount of hydrogen gas is released, which also causes appreciable local change in the pH [29,[34][35][36].Iron, with its slower corrosion rate, is considered an interesting candidate for use as a biodegradable implant material and preliminary in vivo studies have already shown that it has potential for medical applications [37].
Iron is eobiotic and present in all cells of the human body.It is essential in the formation of hemoglobin and helps to convert provitamin beta-carotene to active the vitamin [38].Its presence in the body arising from the corroding implant will therefore not interfere with normal physiological functions unless a threshold limit is exceeded [39].While the benefits that iron-based degradable biomaterials provide are promising, the iron degradation rate for this purpose is slow.This slow degradation rate can have similar disadvantages to those of permanent implants [40][41][42].To overcome this problem, addition of appropriate alloying elements to iron [43] or implant surface treatment, for example the creation of a porous polymer film, can hasten the degradation rate [44].
Coatings are used to alter the bioactivity and modify the surface properties of metallic implants.Up to 75% of coating materials for degradable iron-based biomaterials studied for use in bone repair and regeneration are ceramic coatings based on hydroxyapatite and other calcium phosphates.The second largest group of coating materials represents polymers.Bioceramic coatings improve biocompatibility and the bone bonding ability of metallic implants, in addition to and enhancing bone integration [45][46][47].The diverse effects of bioceramic coatings on the mechanical and corrosion behavior of coated materials were observed depending on the composition, amount, size, structure, morphology, and deposition method of the coating layer [45,[48][49][50].The combination of bioceramics with polymer in composite coating led to the desired mechanical properties (elastic modulus, tensile strength) and beneficial influence on cell adhesion and differentiation [33,51,52].The application of biodegradable polymer coatings represents a new approach to tailor the degradation behavior of iron materials for bone repair.The polymeric coating of the implant can help to improve osseointegration, bone growth, and regeneration [44].The poly(lactic-co-glycolic acid) (PLGA) coating increased the degradation rate and mechanical properties (compressive strength, toughness) of porous iron samples [53].Polyethylene glycol (PEG) is often used as the coating material for implants [54].PEG is a biocompatible polymer that is soluble in water.It is nontoxic, possesses non-antigenicity and immunogenicity, it does not damage biological molecules and cells, and it suppresses platelet adhesion.Because PEG is hydrophilic, it can dissolve in water, especially if the PEG film on the surface of materials is formed by physical absorption [55].
Because the material surface is in direct contact with body tissues, it is necessary to adjust its chemical, physical, biological, and morphological properties to a specific application [56].Moreover, the success of an implant depends also on the shape or form of the applied material [7].The shape, internal structure, and design of the implant must be adapted to the characteristics of the replaced tissue [56].The optimal design for bone applications seems to be an open cell structure that is similar to the bone tissue construction itself [57].Such a structure can be achieved by using powder metallurgy, which is a simple, inexpensive, and effective method [58][59][60].The mechanical properties of devices with an open cell structure are Metals 2018, 8, 499 3 of 21 determined by the material, porosity (morphology, size, and distribution of pores), and density of the created foam [61,62].The resulting product can have a near-net shape with uniform morphology [58,59,63], which can enable the use of the porous materials in very specific applications [64,65].
The aim of the present study was to prepare open cell structures based on carbonyl iron by the powder metallurgy method, coated with PEG.To the best of our knowledge, this is the first time that a polymeric coating layer was used to affect the degradation rate of iron foam biomaterials.The effect of the polymer coating on the corrosion properties of the prepared structures was evaluated.A combination of the nonhomogeneous corrosion attack (pitting) in the iron layer and the oxidative degradation of the PEG layer is suggested as the main form of degradation.The mechanical and tensile properties of coated iron foams are critical for their use for in vivo trauma instruments.The influence of PEG coating layer on biocompatibility and mechanical properties of these materials is the subject of future studies.

Cellular Structure Preparation
Cellular iron structures were prepared by the impregnation of polyurethane foam (PUR) (Filtren ® TM 25133 with the cell size 1060-1600 µm, Eurofoam TP spol.s r.o., Brno, Czech Republic) with a suspension formed by 7 g of carbonyl iron powder (CIP) provided by BASF, Ludwigshafen, Germany (type CC d50 3.8-5.3,3.8-5.3µm) in 6 mL of ultrapure water with 200 mg of gelatin.Gelatin (Sigma-Aldrich, St. Louis, MO 63103, USA) was dissolved in water at 60 • C. The composition of CIP was 99.5% Fe, 0.05% C, 0.01% N, and 0.18% O.The foam impregnation was carried out for 24 h.Then the prepared structures were sintered in a tube furnace Aneta 1 (ANETA, Trenčianská Teplá, Slovak Republic).At first, the formed material was held at 450 • C for 2 h in a nitrogen atmosphere.The final structure was obtained by sintering at 1120 • C for 1 h in a reduction atmosphere (90% N 2 , 10% H 2 ), resulting in the full removal of the polyurethane foam.

Coating of the Iron Samples
Iron foams in form of cylinders (Ø 10 mm, h 20 mm) with a density of about 24 kg/m 3 were coated with polyethylene glycol 4000 (PEG) (Sigma-Aldrich, St. Louis, MO 63103, USA) through a sol-gel process.Three solutions with 5, 10, and 15 wt.% PEG were prepared by dissolving the polymer in 96% ethanol (Mikrochem spol.s.r.o., Pezinok, Slovak Republic).The prepared solutions were equilibrated at room temperature for 24 h.Ultrasonically cleaned specimens (in acetone, ethanol, and distilled water for 10 min each) were immersed in the PEG solutions for 3 h at a laboratory temperature.Subsequently, the coated iron structures were dried for 3 h at 45 • C. The abbreviations for the final materials studied in this work are Fe-PEG1, Fe-PEG2, Fe-PEG3 for the samples immersed into 5, 10, and 15 wt.% PEG solution, respectively.

Material Characterization
The microstructure of the samples was visualized by a scanning electron microscope (SEM) coupled with an energy dispersive spectrometer (EDX) (JSM-7000F, JEOL, Tokyo, Japan with EDX INCA, Oxford Instruments, Abingdon, UK).
The specific surface area (S BET ) of the foams was determined by a low-temperature nitrogen adsorption method using a NOVA 1200 e Surface Area and Pore Size Analyzer (Quantachrome Instruments, London, UK).The values were calculated using the Brunauer-Emmett-Teller (BET) theory.
The FT-IR spectra were recorded using a Tensor 29 infrared spectrometer (Bruker Optik GmbH, Ettlingen, Germany) using the attenuated total reflection (ATR) method.
The Raman spectra were recorded using a Renishaw inVia spectrophotometer (Renishaw UK Sales Ltd., Wotton-under-Edge, UK).All spectra were recorded using a 532-nm laser from 100 cm −1 to 4000 cm −1 , at a 50% laser power.The samples were exposed to the laser for 10 s.All experiments were performed at room temperature.Each spectrum represents the average of five measurements.

Potentiodynamic Polarization
The electrochemical degradation tests were carried out using an Autolab PGSTAT 302 N potentiostat (Metrohm, Herisau, Switzerland).The measurements were conducted in a three-electrode system with an Ag/AgCl/KCl (3 mol/L) reference electrode, a platinum counter electrode, and an iron sample as the working electrode.Corrosion was studied in Hanks' solution (pH = 7.4) (composition in Table 1).A constant temperature of 37 ± 2 • C was maintained during the experiment.The potentiodynamic polarization studies were carried out from −900 mV to 200 mV (vs.Ag/AgCl/KCl (3 mol/L)) at a polarization scan rate of 0.1 mV/s.The corrosion rate (CR in mm/year) was calculated using Equation (1) based on ASTM G59 [66]: where j corr is the current density (µA/cm 2 ), K is a constant that defines the units for the corrosion rate (3.27 × 10 −3 for CR in mm/year), EW is the equivalent weight (27.92 g/eq for Fe), and ρ is the density of iron (7.86 g/cm 3 ).In each analysis three samples were studied.

Electrical Impedance Spectroscopy
Electrical impedance spectroscopy (EIS) measurements were performed on the samples immersed in Hanks' solution using an Autolab PGSTAT 302 N potentiostat (Metrohm, Herisau, Switzerland) and the same three-electrode setup as that used for the potentiodynamic polarization.The EIS measurements were carried out in the frequency range from 10 mHz to 100 kHz with an AC perturbation amplitude of 10 mV.The bias DC potential was equal to the open circuit potential (OCP) determined before the EIS measurements.Three groups of iron open cell structures were examined.The first group consisted of the samples not yet corroded in Hanks' solution.The next two ensembles were examined after immersion in Hanks' solution for 24 h and 48 h.

Static Immersion Test
At first, the samples were weighted to obtain initial weights and then the samples were ultrasonically cleaned in acetone and ethanol for 10 min each.Experimental specimens were immersed in 10 mL of Hanks' solution for 12 weeks and the temperature was kept at 37 • C using a heating mantle.Samples were taken out from the solution after every 7 days of treatment, ultrasonically cleaned in ethanol and distilled water for 10 min each, air dried, and weighed.The corrosion rate (CR in mm/year) was calculated from the mass changes using Equation (2) based on ASTM G31 standard [67]: where m i and m f are the initial mass (g) and final mass after corrosion (g), A is the exposed surface area of the sample (cm 2 ), ρ is the density (g/cm 3 ), and t is the immersion time (h).In each analysis three samples were studied.

Results
The amount of the polymeric coating deposited onto the surface of the sintered iron samples was determined from the mass differences after and before the coating layer deposition.The values were about 8 mg, 20 mg, and 35 mg for the Fe-PEG1, Fe-PEG2, and Fe-PEG3 samples, respectively.This corresponded to the content of the PEG in the coated samples of about 0.8 wt.%, 1.5 wt.%, and 3.0 wt.%, respectively.The interaction between PEG and iron metal is due to a dipole-cation binding between the ether group of the polymer and the positive charge of the partially oxidized surface of the carbonyl iron substrate.The specific surface area (S BET ) of all samples was measured.The values of S BET were 0.48 m 2 /g, 0.80 m 2 /g, and 0.61 m 2 /g for uncoated iron, Fe-PEG1, and the Fe-PEG2 samples, respectively.The value of S BET for Fe-PEG3 was below the detection limit of the instrument and thus led to an unsatisfactory measurement.The low value could be associated with lower porosity and smoothing of the edges of iron structures caused by the formation of a more coherent coating layer with a higher content of PEG.

Surface Morphology and Composition
The final state of the iron-based structures was achieved by sintering the iron-impregnated PUR foam.The appearance of the porous iron samples was similar to the void structure of the PUR foam.The sintered materials contained spherical macropores sized from 350 µm to 1700 µm.The strut thickness between individual macropores was about 100-250 µm (Figure 1).As can be seen from Figure 1, the addition of the polymer coating resulted in the smoothing of the edges of the iron structures.However, the insertion of the PEG film did not cause any significant change of the macropore size.The surface of the iron specimens without a polymer coating layer was covered with fine nodes with numerous pores (size 1-9 µm), as can be seen at a higher magnification (Figure 1b).These fine nodes were also present in the Fe-PEG1 sample (Figure 1c,d).However, the pores in these specimens were wholly or partially filled up with polymer and were smaller (size up to 3 µm) than the pores of the as-prepared iron samples without the PEG film.The roughness of the samples with higher amounts of PEG coating (Fe-PEG2 and Fe-PEG3) was completely smoothed out by the polymeric layer, and the surface of the prepared materials was more compact, uniform, and homogenous (Figure 1e-h).
where mi and mf are the initial mass (g) and final mass after corrosion (g), A is the exposed surface area of the sample (cm 2 ), ρ is the density (g/cm 3 ), and t is the immersion time (h).In each analysis three samples were studied.

Results
The amount of the polymeric coating deposited onto the surface of the sintered iron samples was determined from the mass differences after and before the coating layer deposition.The values were about 8 mg, 20 mg, and 35 mg for the Fe-PEG1, Fe-PEG2, and Fe-PEG3 samples, respectively.This corresponded to the content of the PEG in the coated samples of about 0.8 wt.%, 1.5 wt.%, and 3.0 wt.%, respectively.The interaction between PEG and iron metal is due to a dipole-cation binding between the ether group of the polymer and the positive charge of the partially oxidized surface of the carbonyl iron substrate.The specific surface area (SBET) of all samples was measured.The values of SBET were 0.48 m 2 /g, 0.80 m 2 /g, and 0.61 m 2 /g for uncoated iron, Fe-PEG1, and the Fe-PEG2 samples, respectively.The value of SBET for Fe-PEG3 was below the detection limit of the instrument and thus led to an unsatisfactory measurement.The low value could be associated with lower porosity and smoothing of the edges of iron structures caused by the formation of a more coherent coating layer with a higher content of PEG.

Surface Morphology and Composition
The final state of the iron-based structures was achieved by sintering the iron-impregnated PUR foam.The appearance of the porous iron samples was similar to the void structure of the PUR foam.The sintered materials contained spherical macropores sized from 350 µm to 1700 µm.The strut thickness between individual macropores was about 100-250 µm (Figure 1).As can be seen from Figure 1, the addition of the polymer coating resulted in the smoothing of the edges of the iron structures.However, the insertion of the PEG film did not cause any significant change of the macropore size.The surface of the iron specimens without a polymer coating layer was covered with fine nodes with numerous pores (size 1-9 µm), as can be seen at a higher magnification (Figure 1b).These fine nodes were also present in the Fe-PEG1 sample (Figure 1c,d).However, the pores in these specimens were wholly or partially filled up with polymer and were smaller (size up to 3 µm) than the pores of the as-prepared iron samples without the PEG film.The roughness of the samples with higher amounts of PEG coating (Fe-PEG2 and Fe-PEG3) was completely smoothed out by the polymeric layer, and the surface of the prepared materials was more compact, uniform, and homogenous (Figure 1e-h).The micrographs of the cross-sections of the iron samples with PEG coating pointing to the difference in porosity changing with PEG loading are shown in Figure 2. It is evident that the samples are highly porous and the polymer material is present mainly on the surface of the iron foam.The highest porosity among the tested samples was observed for the Fe-PEG1 sample.The Fe-PEG2 sample contained slightly smaller and less open pores.The Fe-PEG3 sample exhibited the lowest porosity due to the higher amount of PEG entering into the pores of the iron substrate, causing a reduction of the pore size.The micrographs of the cross-sections of the iron samples with PEG coating pointing to the difference in porosity changing with PEG loading are shown in Figure 2. It is evident that the samples are highly porous and the polymer material is present mainly on the surface of the iron foam.The highest porosity among the tested samples was observed for the Fe-PEG1 sample.The Fe-PEG2 sample contained slightly smaller and less open pores.The Fe-PEG3 sample exhibited the lowest porosity due to the higher amount of PEG entering into the pores of the iron substrate, causing a reduction of the pore size.Energy-dispersive X-ray (EDX) analysis was used to determine the content of the main components on the surface of the PEG-coated porous specimens.The average elemental composition of the surface of the coated iron samples is summarized in Table 2.The potential interaction between PEG and iron was investigated by means of infrared and Raman spectroscopy.Because of the fact that the intensities of the spectra of all Fe-PEG composites  The micrographs of the cross-sections of the iron samples with PEG coating pointing to the difference in porosity changing with PEG loading are shown in Figure 2. It is evident that the samples are highly porous and the polymer material is present mainly on the surface of the iron foam.The highest porosity among the tested samples was observed for the Fe-PEG1 sample.The Fe-PEG2 sample contained slightly smaller and less open pores.The Fe-PEG3 sample exhibited the lowest porosity due to the higher amount of PEG entering into the pores of the iron substrate, causing a reduction of the pore size.Energy-dispersive X-ray (EDX) analysis was used to determine the content of the main components on the surface of the PEG-coated porous specimens.The average elemental composition of the surface of the coated iron samples is summarized in Table 2.The potential interaction between PEG and iron was investigated by means of infrared and Raman spectroscopy.Because of the fact that the intensities of the spectra of all Fe-PEG composites Energy-dispersive X-ray (EDX) analysis was used to determine the content of the main components on the surface of the PEG-coated porous specimens.The average elemental composition of the surface of the coated iron samples is summarized in Table 2.The potential interaction between PEG and iron was investigated by means of infrared and Raman spectroscopy.Because of the fact that the intensities of the spectra of all Fe-PEG composites were too low, the spectra were normalized on an arbitrary scale.The results are presented in Figures 4 and 5.
Metals 2018, 8, x FOR PEER REVIEW 7 of 21 were too low, the spectra were normalized on an arbitrary scale.The results are presented in Figures 4 and 5.In the infrared spectrum of pure PEG, a very intense peak at 2881 cm −1 corresponding to the stretching vibration of C-H group was recorded.The bending vibrations corresponding to the same group were found at 1466 cm −1 and 1340 cm −1 and the stretching vibrations, corresponding to O-H and C=O groups, were registered at 1279 cm −1 and 1096 cm −1 , respectively.
The Raman spectrum (Figure 5) confirms the results obtained by infrared spectrometry (Figure 4).The internal vibrations of the methylene group give rise to peaks recorded at 2887 cm −1 and 1482 cm −1 .The intense peak recorded at 2887 cm −1 corresponds to stretching vibration, and the peak at 1482 cm −1 is the result of the bending vibration of a C-H group.The twisting vibrations corresponding to the same group were found at 1285 cm −1 and 1232 cm −1 , and the stretching vibrations corresponding to the C-C and C=O groups were registered at 1142 cm −1 and 1052 cm −1 , respectively.The bands at 876 cm −1 were recorded in the IR spectrum only for Fe-PEG1 and Fe-PEG2 samples, and those observed in the Raman spectra for all of the samples were assigned to the skeletal vibrations of PEG.The C-C-O bending vibration appears at 533 cm −1 .The band at 225 cm −1 corresponds to the skeletal deformation mode of PEG.In the infrared spectrum of pure PEG, a very intense peak at 2881 cm −1 corresponding to the stretching vibration of C-H group was recorded.The bending vibrations corresponding to the same group were found at 1466 cm −1 and 1340 cm −1 and the stretching vibrations, corresponding to O-H and C=O groups, were registered at 1279 cm −1 and 1096 cm −1 , respectively.
The Raman spectrum (Figure 5) confirms the results obtained by infrared spectrometry (Figure 4).The internal vibrations of the methylene group give rise to peaks recorded at 2887 cm −1 and 1482 cm −1 .The intense peak recorded at 2887 cm −1 corresponds to stretching vibration, and the peak at 1482 cm −1 is the result of the bending vibration of a C-H group.The twisting vibrations corresponding to the same group were found at 1285 cm −1 and 1232 cm −1 , and the stretching vibrations corresponding to the C-C and C=O groups were registered at 1142 cm −1 and 1052 cm −1 , respectively.The bands at 876 cm −1 were recorded in the IR spectrum only for Fe-PEG1 and Fe-PEG2 samples, and those observed in the Raman spectra for all of the samples were assigned to the skeletal vibrations of PEG.The C-C-O bending vibration appears at 533 cm −1 .The band at 225 cm −1 corresponds to the skeletal deformation mode of PEG.In the infrared spectrum of pure PEG, a very intense peak at 2881 cm −1 corresponding to the stretching vibration of C-H group was recorded.The bending vibrations corresponding to the same group were found at 1466 cm −1 and 1340 cm −1 and the stretching vibrations, corresponding to O-H and C=O groups, were registered at 1279 cm −1 and 1096 cm −1 , respectively.
The Raman spectrum (Figure 5) confirms the results obtained by infrared spectrometry (Figure 4).The internal vibrations of the methylene group give rise to peaks recorded at 2887 cm −1 and 1482 cm −1 .The intense peak recorded at 2887 cm −1 corresponds to stretching vibration, and the peak at 1482 cm −1 is the result of the bending vibration of a C-H group.The twisting vibrations corresponding to the same group were found at 1285 cm −1 and 1232 cm −1 , and the stretching vibrations corresponding to the C-C and C=O groups were registered at 1142 cm −1 and 1052 cm −1 , respectively.The bands at 876 cm −1 were recorded in the IR spectrum only for Fe-PEG1 and Fe-PEG2 samples, and those observed in the Raman spectra for all of the samples were assigned to the skeletal vibrations of PEG.The C-C-O bending vibration appears at 533 cm −1 .The band at 225 cm −1 corresponds to the skeletal deformation mode of PEG.

Immersion Tests
The preliminary results of the immersion corrosion tests are given in Reference [68].A more detailed study followed here.The micrographs of the corroded surface of all specimens after 4, 8, and 12 weeks of immersion in Hanks' solution at 37 • C are shown in Figures 6-8.The presence of degradation products was identified by the EDX analysis of the corroded sample surfaces [68].The amount of the corrosion products increased by extending the immersion in Hanks' solution from 4 to 12 weeks for all specimens.The non-homogeneous corrosion leading to the rupture of struts of the open cell porous solids and the formation of the thin layer of the corrosion products were observed on the surface of all specimens after 4 weeks of degradation, as documented in Figure 6.The degree of sintered samples' damage rose with the increasing time of immersion.As can be seen from Figure 7, the whole surface of the porous structure was completely covered with degradation products after 8 weeks of immersion in Hanks' solution.Extended crack formation, considerable dissolution of the struts, and the accumulation of degradation products on the surface of the cellular samples were observed, as shown in Figure 8, after 12 weeks of the immersion test.

Immersion Tests
The preliminary results of the immersion corrosion tests are given in Reference [68].A more detailed study followed here.The micrographs of the corroded surface of all specimens after 4, 8, and 12 weeks of immersion in Hanks' solution at 37 °C are shown in Figures 6-8.The presence of degradation products was identified by the EDX analysis of the corroded sample surfaces [68].The amount of the corrosion products increased by extending the immersion in Hanks' solution from 4 to 12 weeks for all specimens.The non-homogeneous corrosion leading to the rupture of struts of the open cell porous solids and the formation of the thin layer of the corrosion products were observed on the surface of all specimens after 4 weeks of degradation, as documented in Figure 6.The degree of sintered samples' damage rose with the increasing time of immersion.As can be seen from Figure 7, the whole surface of the porous structure was completely covered with degradation products after 8 weeks of immersion in Hanks' solution.Extended crack formation, considerable dissolution of the struts, and the accumulation of degradation products on the surface of the cellular samples were observed, as shown in Figure 8, after 12 weeks of the immersion test.

Immersion Tests
The preliminary results of the immersion corrosion tests are given in Reference [68].A more detailed study followed here.The micrographs of the corroded surface of all specimens after 4, 8, and 12 weeks of immersion in Hanks' solution at 37 °C are shown in Figures 6-8.The presence of degradation products was identified by the EDX analysis of the corroded sample surfaces [68].The amount of the corrosion products increased by extending the immersion in Hanks' solution from 4 to 12 weeks for all specimens.The non-homogeneous corrosion leading to the rupture of struts of the open cell porous solids and the formation of the thin layer of the corrosion products were observed on the surface of all specimens after 4 weeks of degradation, as documented in Figure 6.The degree of sintered samples' damage rose with the increasing time of immersion.As can be seen from Figure 7, the whole surface of the porous structure was completely covered with degradation products after 8 weeks of immersion in Hanks' solution.Extended crack formation, considerable dissolution of the struts, and the accumulation of degradation products on the surface of the cellular samples were observed, as shown in Figure 8, after 12 weeks of the immersion test.Mass loss was assessed by periodic weighing of the samples, which were first thoroughly rinsed and dried.The mass losses of iron samples immersed in Hanks' solution for a period of 12 weeks are listed in Table 3.The mass changes were caused by the generation of soluble or non-adhering corrosion products and by the dissolution of the sintered samples.An increase in mass loss was observed in all examined samples in the course of the whole period of the test.The values of the corrosion rate determined from the immersion test are shown in Table 4 and Figure 9.For the calculation of the corrosion rate of the Fe-PEG3 sample, we used the value of SBET for the bare Fe sample (0.48 m 2 /g) since the BET method did not provide satisfactory results for the surface area of this sample.The highest corrosion rates were observed after 4 or 6 weeks of immersion in Hanks' solution, with the exception of the sample with the highest content of the hydrophilic PEG coating, for which the highest corrosion rate was observed after 2 weeks of immersion (0.0919 mm/year).The values of the corrosion rate in the fourth week of immersion were 0.0236 mm/year, 0.0327 mm/year, 0.0493 mm/year, and 0.0555 mm/year, for the uncoated Fe as well as the Fe-PEG1, Fe-PEG2, and Fe-PEG3 samples, respectively.Afterwards, a gradual decline in the values of the degradation rate for all samples was observed, presumably due to the corrosion products' layer formation.Mass loss was assessed by periodic weighing of the samples, which were first thoroughly rinsed and dried.The mass losses of iron samples immersed in Hanks' solution for a period of 12 weeks are listed in Table 3.The mass changes were caused by the generation of soluble or non-adhering corrosion products and by the dissolution of the sintered samples.An increase in mass loss was observed in all examined samples in the course of the whole period of the test.The values of the corrosion rate determined from the immersion test are shown in Table 4 and Figure 9.For the calculation of the corrosion rate of the Fe-PEG3 sample, we used the value of S BET for the bare Fe sample (0.48 m 2 /g) since the BET method did not provide satisfactory results for the surface area of this sample.The highest corrosion rates were observed after 4 or 6 weeks of immersion in Hanks' solution, with the exception of the sample with the highest content of the hydrophilic PEG coating, for which the highest corrosion rate was observed after 2 weeks of immersion (0.0919 mm/year).The values of the corrosion rate in the fourth week of immersion were 0.0236 mm/year, 0.0327 mm/year, 0.0493 mm/year, and 0.0555 mm/year, for the uncoated Fe as well as the Fe-PEG1, Fe-PEG2, and Fe-PEG3 samples, respectively.Afterwards, a gradual decline in the values of the degradation rate for all samples was observed, presumably due to the corrosion products' layer formation.

Electrochemical Corrosion Behavior
The corrosion behavior of the prepared porous samples was studied by the anodic polarization method in Hanks' solution at 37 °C.Representative potentiodynamic polarization curves (Tafel plots) for PEG-coated samples, together with the uncoated iron sample, are shown in Figure 10.The reproducibility of the Tafel plots was good.It should be noted that the BET method did not provide a satisfactory result for the surface area of the Fe-PEG3 sample due to the low signal, and thus we used the value of SBET for the bare Fe sample (0.48 m 2 /g) to calculate the current density of this sample.The corrosion potential (Ecorr vs. Ag/AgCl/KCl (3 mol/L)) and corrosion current density (jcorr) were both calculated from the intersection of the extrapolated anodic and cathodic Tafel lines.Since the obtained curves did not exhibit the typical Tafel form, they were evaluated for both Tafel and non-Tafel behavior.A non-Tafel behavior of the examined material can be caused by the inhomogeneity of the sample surface and irregular structure of the porous samples [69].The values of Ecorr (vs.Ag/AgCl/KCl (3 mol/L)), jcorr, and CR determined for Tafel and non-Tafel behavior of curves are listed in Table 5.

Electrochemical Corrosion Behavior
The corrosion behavior of the prepared porous samples was studied by the anodic polarization method in Hanks' solution at 37 • C. Representative potentiodynamic polarization curves (Tafel plots) for PEG-coated samples, together with the uncoated iron sample, are shown in Figure 10.The reproducibility of the Tafel plots was good.It should be noted that the BET method did not provide a satisfactory result for the surface area of the Fe-PEG3 sample due to the low signal, and thus we used the value of S BET for the bare Fe sample (0.48 m 2 /g) to calculate the current density of this sample.The corrosion potential (E corr vs. Ag/AgCl/KCl (3 mol/L)) and corrosion current density (j corr ) were both calculated from the intersection of the extrapolated anodic and cathodic Tafel lines.Since the obtained curves did not exhibit the typical Tafel form, they were evaluated for both Tafel and non-Tafel behavior.A non-Tafel behavior of the examined material can be caused by the inhomogeneity of the sample surface and irregular structure of the porous samples [69].The values of E corr (vs.Ag/AgCl/KCl (3 mol/L)), j corr , and CR determined for Tafel and non-Tafel behavior of curves are listed in Table 5.The presence of a PEG coating on the surface of iron samples resulted in a higher susceptibility to corrosion, expressed by a shift of the corrosion potential to more negative values and an increase in the current density.Corrosion rates determined from Tafel plots by Tafel evaluation were 0.438 mm/year, 0.536 mm/year, 0.617 mm/year, and 0.703 mm/year for Fe, Fe-PEG1, Fe-PEG2, and Fe-PEG3 samples, respectively.Corrosion rates determined by non-Tafel evaluation were 0.360 mm/year, 0.494 mm/year, 0.596 mm/year, and 0.670 mm/year for Fe, Fe-PEG1, Fe-PEG2, and Fe-PEG3 samples, respectively.The highest value of jcorr was observed for the Fe-PEG3 sample and the lowest jcorr was obtained for the bare Fe sample.The same behavior was observed for Tafel and non-Tafel evaluation, but jcorr values obtained for the non-Tafel evaluation were slightly lower.The shift of Ecorr (vs.Ag/AgCl/KCl (3 mol/L)) to more negative values indicates that the polarization occurred mostly at the cathode and hence the corrosion rates of the PEG-coated iron samples were cathodically controlled [53].The value of the corrosion potential decreased with increasing amounts of PEG on the surface of the porous iron material.
The EIS measurements were performed before immersion (0 h) and after 24 h and 48 h immersion of the samples in Hanks' solution.Nyquist plots of the three cellular samples obtained before and after immersion in Hanks' solution are shown in Figure 11.The corrosion process of iron materials in a solution with neutral pH is influenced by the formation of corrosion products at the surface.This corrosion product layer acts as an inhibitor and the properties of the layer are dependent on the medium in which the corrosion takes place, on the hydrodynamic conditions, and on the duration of exposure [70].The Nyquist plots indicate the same trends as the results obtained from the potentiodynamic polarization curves.The impedance value of the tested samples decreased in the sequence: Fe, Fe-PEG1, Fe-PEG2, Fe-PEG3 (Figure 12), indicating the increase in corrosion susceptibility.The same trend was observed for all samples before immersion and after immersion in Hanks' solution for 24 h and 48 h.An increase in the impedance value was observed for each sample with prolonged time of immersion in Hanks' solution.The lowest impedance values were detected for the specimens before immersion, and the highest values were observed for the samples after 48 h of immersion.Considering that the Nyquist plots of the Fe sample after 24 h and 48 h of immersion included mainly a linear diffusion part, it can be concluded that a charge transfer occurred rapidly.The presence of a PEG coating on the surface of iron samples resulted in a higher susceptibility to corrosion, expressed by a shift of the corrosion potential to more negative values and an increase in the current density.Corrosion rates determined from Tafel plots by Tafel evaluation were 0.438 mm/year, 0.536 mm/year, 0.617 mm/year, and 0.703 mm/year for Fe, Fe-PEG1, Fe-PEG2, and Fe-PEG3 samples, respectively.Corrosion rates determined by non-Tafel evaluation were 0.360 mm/year, 0.494 mm/year, 0.596 mm/year, and 0.670 mm/year for Fe, Fe-PEG1, Fe-PEG2, and Fe-PEG3 samples, respectively.The highest value of j corr was observed for the Fe-PEG3 sample and the lowest j corr was obtained for the bare Fe sample.The same behavior was observed for Tafel and non-Tafel evaluation, but j corr values obtained for the non-Tafel evaluation were slightly lower.The shift of E corr (vs.Ag/AgCl/KCl (3 mol/L)) to more negative values indicates that the polarization occurred mostly at the cathode and hence the corrosion rates of the PEG-coated iron samples were cathodically controlled [53].The value of the corrosion potential decreased with increasing amounts of PEG on the surface of the porous iron material.
The EIS measurements were performed before immersion (0 h) and after 24 h and 48 h immersion of the samples in Hanks' solution.Nyquist plots of the three cellular samples obtained before and after immersion in Hanks' solution are shown in Figure 11.The corrosion process of iron materials in a solution with neutral pH is influenced by the formation of corrosion products at the surface.This corrosion product layer acts as an inhibitor and the properties of the layer are dependent on the medium in which the corrosion takes place, on the hydrodynamic conditions, and on the duration of exposure [70].The Nyquist plots indicate the same trends as the results obtained from the potentiodynamic polarization curves.The impedance value of the tested samples decreased in the sequence: Fe, Fe-PEG1, Fe-PEG2, Fe-PEG3 (Figure 12), indicating the increase in corrosion susceptibility.The same trend was observed for all samples before immersion and after immersion in Hanks' solution for 24 h and 48 h.An increase in the impedance value was observed for each sample with prolonged time of immersion in Hanks' solution.The lowest impedance values were detected for the specimens before immersion, and the highest values were observed for the samples after 48 h of immersion.Considering that the Nyquist plots of the Fe sample after 24 h and 48 h of immersion included mainly a linear diffusion part, it can be concluded that a charge transfer occurred rapidly.The EIS spectra of the uncoated and the PEG-coated samples were modeled using the chosen equivalent circuits shown in Figure 13, where Rs is the solution resistance, RL1 is the pore resistance, RL2 is the resistance of polymer layer or corrosion products, and Rct is the charge transfer resistance.QL1, QL2, and Qdl are the constant phase elements (CPE) which correspond to the PEG film, the passive film, and the electrical double layer, respectively.It should be observed that other equivalent circuit arrangements with seven components could be drawn, with an identical electrical response (albeit different component values) and an overall satisfactory fit.This particular circuit was chosen because   The EIS spectra of the uncoated and the PEG-coated samples were modeled using the chosen equivalent circuits shown in Figure 13, where Rs is the solution resistance, RL1 is the pore resistance, RL2 is the resistance of polymer layer or corrosion products, and Rct is the charge transfer resistance.QL1, QL2, and Qdl are the constant phase elements (CPE) which correspond to the PEG film, the passive film, and the electrical double layer, respectively.It should be observed that other equivalent circuit arrangements with seven components could be drawn, with an identical electrical response (albeit different component values) and an overall satisfactory fit.This particular circuit was chosen because The EIS spectra of the uncoated and the PEG-coated samples were modeled using the chosen equivalent circuits shown in Figure 13, where R s is the solution resistance, R L1 is the pore resistance, R L2 is the resistance of polymer layer or corrosion products, and R ct is the charge transfer resistance.Q L1 , Q L2 , and Q dl are the constant phase elements (CPE) which correspond to the PEG film, the passive film, and the electrical double layer, respectively.It should be observed that other equivalent circuit arrangements with seven components could be drawn, with an identical electrical response (albeit different component values) and an overall satisfactory fit.This particular circuit was chosen because of its convergence and also because it fit the results for all three samples, thus making the comparison of the different impedance behaviors relevant.The solid/solution interface was characterized by the charge transfer resistance (R ct ) and by the double layer capacitance (Q dl ), which depends on the increase of the wetted metal area.Q dl was calculated (taking the CPE value as the capacitance) according to the following equation: where f max is the frequency at which the imaginary impedance component is at a maximum (which in this case was 0.01 Hz).The R ct values (Table 6) for all samples before and after immersion in Hanks' solution increased in the following order: Fe-PEG3, Fe-PEG2, Fe-PEG1, Fe.The highest R ct values were found for the pure iron structure and the lowest values were obtained for Fe-PEG3 sample in all cases.The value of R ct decreased with prolonged immersion for each individual sample.The highest Q dl values were obtained for the Fe-PEG3 material and the lowest for the Fe specimen before and after corrosion in Hanks' solution.An increase of Q dl for the Fe sample, while a decrease of Q dl values of PEG-coated samples, were observed with prolonged immersion from 0 to 48 h (Table 6).This decrease indicates a loss of wetted area on the coated surface due to the enhanced formation of degradation products as a result of the faster degradation.Yusop et al. [53] obtained similar results during the examination of porous iron structures with PLGA coating.They noticed lower impedance and R ct values for samples with polymer coating compared to uncoated iron samples.This trend was also observed in our experiment.of its convergence and also because it fit the results for all three samples, thus making the comparison of the different impedance behaviors relevant.The solid/solution interface was characterized by the charge transfer resistance (Rct) and by the double layer capacitance (Qdl), which depends on the increase of the wetted metal area.Qdl was calculated (taking the CPE value as the capacitance) according to the following equation: where fmax is the frequency at which the imaginary impedance component is at a maximum (which in this case was 0.01 Hz).The Rct values (Table 6) for all samples before and after immersion in Hanks' solution increased in the following order: Fe-PEG3, Fe-PEG2, Fe-PEG1, Fe.The highest Rct values were found for the pure iron structure and the lowest values were obtained for Fe-PEG3 sample in all cases.The value of Rct decreased with prolonged immersion for each individual sample.The highest Qdl values were obtained for the Fe-PEG3 material and the lowest for the Fe specimen before and after corrosion in Hanks' solution.An increase of Qdl for the Fe sample, while a decrease of Qdl values of PEG-coated samples, were observed with prolonged immersion from 0 to 48 h (Table 6).This decrease indicates a loss of wetted area on the coated surface due to the enhanced formation of degradation products as a result of the faster degradation.Yusop et al. [53] obtained similar results during the examination of porous iron structures with PLGA coating.They noticed lower impedance and Rct values for samples with polymer coating compared to uncoated iron samples.This trend was also observed in our experiment.

Degradation Mechanism
Open cell iron structures with PEG coatings are novel degradable biomaterials.There are known processes of degradation of the individual components of the prepared samples.However, the degradation behavior of the Fe-PEG composite material has not yet been investigated.The iron degradation process in aqueous solution begins with the oxidation of iron.It consists of an electrochemical process that starts with the following anodic (4) and cathodic (5) reactions:

Degradation Mechanism
Open cell iron structures with PEG coatings are novel degradable biomaterials.There are known processes of degradation of the individual components of the prepared samples.However, the degradation behavior of the Fe-PEG composite material has not yet been investigated.The iron degradation process in aqueous solution begins with the oxidation of iron.It consists of an electrochemical process that starts with the following anodic (4) and cathodic (5) reactions: The reaction of the Fe 2+ cations with OH − anions results in the formation of ferrous hydroxide (6).Some of the Fe 2+ cations can be oxidized to Fe 3+ ions (7), if oxygen and alkaline pH are present.Then, these ions can react to form ferric hydroxide (8).
In the presence of oxygen and chloride ions, FeO(OH) precipitates and the formation of magnetite (9) occurs: During in vitro tests a layer of precipitated degradation products forms, which usually exhibits a brownish color and consists of hydroxides originating from the reactions above [71,72].

Discussion
The presence of a PEG layer on the surface of iron samples was proven by SEM micrographs, EDX analysis, and FT-IR and Raman spectroscopy.From the SEM micrographs (Figure 1) and EDX analyses (Table 2) of the surface of sintered samples, it can be seen that the polymeric layer on the surface of the Fe-PEG2 sample is discontinuous compared to the Fe-PEG3 sample, which contains only polymer on the surface.Hence, zero weight percent of iron was detected on the surface the Fe-PEG3 material by EDX analysis (Table 2).The content of carbon in the examined samples increased with increasing amounts of PEG on the surface of the iron cellular structures.The small amount of carbon and oxygen originating from the carbonyl iron powder was also noted on the surface of the Fe sample without the PEG coating layer.
In order to investigate the surface properties of the prepared PEG-coated porous materials, the specific surface area value of all samples was measured.In general, it can be seen that the values of S BET are very low.This is contrary to the study of Park et al. [73]; however, the authors of that work used a different synthetic approach which favored the formation of a porous composite.In a very similar report to ours [74], the S BET values were not reported.The introduction of the low amount of PEG into the structure of iron results in only a slight specific surface area increase of the whole material, while the value of the specific surface area decreased with increasing content of the PEG coating.The S BET value of the sample with the lowest content of PEG (Fe-PEG1) increased (0.80 m 2 /g) as compared to the bare iron sample (0.48 m 2 /g).This is probably due to the formation of a discontinuous PEG layer consisting of small isolated islands of PEG on the surface of the Fe substrate.The next increase in the PEG content led to a decrease in the S BET value to 0.61 m 2 /g for Fe-PEG2 and to a value below the detection limit for Fe-PEG3.
The results of the infrared spectroscopy presented in Figure 4 are in good agreement with recent literature [75,76].The intensity of the spectra of the samples with the low PEG content (Fe-PEG1 and Fe-PEG2) was quite low, and therefore some peaks are not interpreted, as they are within the noise level.Although no significant changes of the peak positions in the Fe-PEG samples in comparison to pure PEG were seen, some slight changes took place.This is in agreement with studies dealing with Au/PEG [77] or Se/PEG composites [78].
The Raman spectrum (Figure 5) confirms the results obtained by infrared spectrometry (Figure 4).The obtained Raman spectra (Figure 5) correlate well with other publications [79,80].The peak positions of Fe-PEG samples are almost identical to those of pure PEG in the infrared spectra as well.The intensity of the Raman spectra, similar to the intensity of the infrared spectra, was also very low.
The corrosion behavior of the sintered iron samples was examined by potentiodynamic polarization, electrochemical impedance, and an immersion test.The extension of corrosion during the static immersion test was observed by SEM micrographs (Figures 6-8).
An increase in mass loss in all examined samples during the whole period of the test was observed from the mass loss values of iron samples immersed in Hanks' solution, as listed in Table 3.More significant mass loss was observed for the coated samples compared to the bare Fe sample, indicating faster degradation [68].This might be associated with the local change of pH due to PEG hydrolysis [81].The Fe-PEG3 structure showed the most distinctive mass loss up until the fourth week of immersion.Thereafter, the Fe-PEG2 specimen exhibited somewhat higher values of mass loss [68].The total mass loss after 12 weeks of immersion was about 6% for the uncoated Fe sample, while it was 10-12% for the PEG-coated samples.Qi et al. [82] demonstrated that the coating of biodegradable aliphatic polyester could accelerate iron corrosion due to the acidic degradation products.
The PEG coating caused an increase in the corrosion rate of the coated iron samples, as compared to the uncoated sample (Table 4, Figure 9).Moreover, the degradation accelerated with the increased PEG content.The Fe-PEG3 specimen degraded the quickest out of all of the tested materials.The values of the corrosion rate in the twelfth week of immersion were 0.0217 mm/year, 0.0230 mm/year, 0.0357 mm/year, and 0.0406 mm/year, for the uncoated Fe, and for the Fe-PEG1, Fe-PEG2, and Fe-PEG3 samples, respectively.Also, Yusop et al. [53] achieved higher levels of degradation rate in their materials with PLGA compared to bare iron samples, due to the effect of PLGA hydrolysis.
The corrosion rates calculated from the electrochemical test using Equation ( 1) are summarized in Table 5.The PEG film caused an increase in the degradation rate of the samples with a polymeric layer compared to the iron sample without PEG coating.The highest values of the degradation rate calculated from the Tafel and non-Tafel evaluation were obtained for the Fe-PEG3 sample, while the Fe specimen exhibited the smallest corrosion rate.These results correlate with the findings of the immersion corrosion tests and are in good agreement with the study of Yusop et al. [53].They discovered that PLGA coating on an iron biomaterial leads to the increase of E corr and j corr , as well as the increase of the corrosion rate.The corrosion rates determined from Tafel plots were 0.11 mm/year, 0.42 mm/year, and 0.72 mm/year for porous pure iron, PLGA-coated porous iron, and PLGA-infiltrated porous iron, respectively.
The corrosion mechanism of iron in a physiological solution was extensively studied.In the case of PEG, a hydrophilic compound [83], the most common degradation pathway is hydrolysis or oxidation [73,84].At pH 7.4, the PEG chain is cleaved into oligomers with different numbers of repeating units [85].The degradation products of PEG can be eliminated in vivo by the kidneys [86].The interfacial interaction between the hydrophilic polymer layer and the iron surface generates the increasing degradation of iron.The degradation begins by the oxidative degradation of the PEG coating layer followed by the formation of defects in the polymer layer because the hydrophilic polymers undergo bulk degradation [87].The defects allow the penetration of a solution into the iron substrate.The process of PEG degradation also induces the release of hydrogen atoms [88].Protons can react with the electrons at the iron surface, leading to hydrogen evolution, passing iron ions into the degradation medium and its reduction, and thereafter to the inhibition of oxygen diffusion as a consequence of corrosion products' layer formation.All of these processes enable and enhance the iron degradation [89].A schematic representation of the Fe-PEG sample degradation process is shown in Figure 14.
substrate.The process of PEG degradation also induces the release of hydrogen atoms [88].Protons can react with the electrons at the iron surface, leading to hydrogen evolution, passing iron ions into the degradation medium and its reduction, and thereafter to the inhibition of oxygen diffusion as a consequence of corrosion products' layer formation.All of these processes enable and enhance the iron degradation [89].A schematic representation of the Fe-PEG sample degradation process is shown in Figure 14.

Conclusions
A metallurgical replication method was used to prepare iron-based open cell structures.This material is intended to constitute the structural element for a possible temporary in vivo implant.The iron surface was coated throughout the open cell structure by polyethylene glycol to various degrees.The PEG coating was accomplished by a sol-gel process.The surface of the porous structures was rendered more compact and homogenous with increasing amounts of the polymer.
The presence of the PEG coating on the sample surface is instrumental to enhance the rate of iron metal degradation.The PEG layer caused a shift of the corrosion potential to more negative values together with an increase in the corrosion rate as compared to uncoated iron materials.Impedance measurements show in a series of Nyquist plots that the charge transfer resistance of the studied samples was the highest for the uncoated iron specimens and decreased continuously with increasing amounts of PEG.Thus, the degradation rate increased with increasing amounts of PEG.A combination of the pitting corrosion in the iron layer and the oxidation of the PEG layer are suggested as the two main reasons for the enhanced degradation.
Coating the sintered open cell foam iron samples with a PEG layer led to an increase in the degradation rate in a tunable fashion based on the initial amount of PEG deposited on the substrate.This identifies the material and the method of its treatment as a promising approach to fabricating biodegradable implant materials for bone applications.Future experiments will aim to verify the biocompatibility and mechanical properties of such structures.

Figure 3
Figure 3 shows distribution of the polymer coating layer (black layer emphasized by the arrows) on the surface of the Fe-PEG3 sample.The polymer layer is considerably discontinuous.The approximate thickness of the PEG coating obtained from the micrographs of the cross-sections of the coated samples was 150-1000 µm.

Figure 3 .
Figure 3. Representative cross-sectional micrograph of the polymer coating layer on the surface of Fe-PEG3 sample containing, 3.0 wt.% PEG (white arrows point to the interface).

Figure 3
Figure 3 shows distribution of the polymer coating layer (black layer emphasized by the arrows) on the surface of the Fe-PEG3 sample.The polymer layer is considerably discontinuous.The approximate thickness of the PEG coating obtained from the micrographs of the cross-sections of the coated samples was 150-1000 µm.

Figure 3
Figure 3 shows distribution of the polymer coating layer (black layer emphasized by the arrows) on the surface of the Fe-PEG3 sample.The polymer layer is considerably discontinuous.The approximate thickness of the PEG coating obtained from the micrographs of the cross-sections of the coated samples was 150-1000 µm.

Figure 3 .
Figure 3. Representative cross-sectional micrograph of the polymer coating layer on the surface of Fe-PEG3 sample containing, 3.0 wt.% PEG (white arrows point to the interface).

Figure 3 .
Figure 3. Representative cross-sectional micrograph of the polymer coating layer on the surface of Fe-PEG3 sample containing, 3.0 wt.% PEG (white arrows point to the interface).

Figure 4 .
Figure 4. FT-infrared spectra of pure PEG (black line, bottom) and the Fe-PEG composites (in order from bottom to top: Fe-PEG1-red line, Fe-PEG2-orange line, Fe-PEG3-blue line).

Figure 4 .
Figure 4. FT-infrared spectra of pure PEG (black line, bottom) and the Fe-PEG composites (in order from bottom to top: Fe-PEG1-red line, Fe-PEG2-orange line, Fe-PEG3-blue line).

Figure 4 .
Figure 4. FT-infrared spectra of pure PEG (black line, bottom) and the Fe-PEG composites (in order from bottom to top: Fe-PEG1-red line, Fe-PEG2-orange line, Fe-PEG3-blue line).

Figure 9 .
Figure 9. Corrosion rate of open cell iron samples with and without the PEG film obtained from immersion tests in Hanks' solution.

Figure 9 .
Figure 9. Corrosion rate of open cell iron samples with and without the PEG film obtained from immersion tests in Hanks' solution.

Figure 10 .
Figure 10.Potentiodynamic polarization curves of open cell iron samples with and without the PEG film obtained in Hanks' solution at pH 7.4 and 37 °C at a scan rate 0.1 mV/s.

Figure 10 .
Figure 10.Potentiodynamic polarization curves of open cell iron samples with and without the PEG film obtained in Hanks' solution at pH 7.4 and 37 • C at a scan rate 0.1 mV/s.

Figure 12 .
Figure 12.Nyquist plots of open cell iron samples with and without PEG coating before immersion (a) and after immersion in Hanks' solution for 48 h (b).

Figure 12 .
Figure 12.Nyquist plots of open cell iron samples with and without PEG coating before immersion (a) and after immersion in Hanks' solution for 48 h (b).

Figure 12 .
Figure 12.Nyquist plots of open cell iron samples with and without PEG coating before immersion (a) and after immersion in Hanks' solution for 48 h (b).

Figure 13 .
Figure13.Equivalent circuit used for the modeling of the electrical impedance spectroscopy (EIS) spectra of sintered iron samples with and without PEG coating before degradation and after immersion in Hanks' solution for 24 and 48 h.Rs is the solution resistance, RL1 is the pore resistance, RL2 is the resistance of polymer layer or the corrosion products, and Rct is the charge transfer resistance.QL1, QL2, and Qdl are the constant phase elements.

Figure 13 .
Figure13.Equivalent circuit used for the modeling of the electrical impedance spectroscopy (EIS) spectra of sintered iron samples with and without PEG coating before degradation and after immersion in Hanks' solution for 24 and 48 h.R s is the solution resistance, R L1 is the pore resistance, R L2 is the resistance of polymer layer or the corrosion products, and R ct is the charge transfer resistance.Q L1 , Q L2 , and Q dl are the constant phase elements.

Figure 14 .
Figure 14.Schematic representation of the Fe-PEG sample degradation process.Figure 14.Schematic representation of the Fe-PEG sample degradation process.

Figure 14 .
Figure 14.Schematic representation of the Fe-PEG sample degradation process.Figure 14.Schematic representation of the Fe-PEG sample degradation process.

Table 2 .
Chemical composition of the surface of sintered iron samples, uncoated and coated with different amounts of the PEG layer.

Table 2 .
Chemical composition of the surface of sintered iron samples, uncoated and coated with different amounts of the PEG layer.

Table 2 .
Chemical composition of the surface of sintered iron samples, uncoated and coated with different amounts of the PEG layer.

Table 3 .
The average values of mass losses of open cells iron samples after 12 weeks of static corrosion in Hanks' solution.

Table 3 .
The average values of mass losses of open cells iron samples after 12 weeks of static corrosion in Hanks' solution.

Table 4 .
The values of static corrosion rate for each material immersed in Hanks' solution for 2, 4, 6, 8, 10, and 12 weeks.

Table 4 .
The values of static corrosion rate for each material immersed in Hanks' solution for 2, 4, 6, 8, 10, and 12 weeks.

Table 5 .
Values of Ecorr, jcorr, and CR for the uncoated Fe sample and samples with PEG film evaluated for the Tafel and non-Tafel behavior of potentiodynamic polarization curves.

Table 5 .
Values of E corr , j corr , and CR for the uncoated Fe sample and samples with PEG film evaluated for the Tafel and non-Tafel behavior of potentiodynamic polarization curves.

Table 6 .
Impedance parameters of uncoated and PEG-coated open cell Fe samples at fmax = 0.01 Hz.

Table 6 .
Impedance parameters of uncoated and PEG-coated open cell Fe samples at f max = 0.01 Hz.