A Two-Year Evaluation of Corrosion-Induced Damage to Hot Galvanized Reinforcing Steel B500SP in Chloride Contaminated Concrete

Corrosion-induced damage to concrete reinforced with bars is a serious problem regarding technical and economic aspects and strongly depends on used materials, corrosion environment, and service life. Tests described in this paper refer to a two-year evaluation of the effectiveness of protection provided by zinc-coated low-carbon reinforcing steel of grade B500SP in concrete against chloride corrosion. Performed tests were comparative and included measurements conducted on four groups of concrete test elements with dimensions of 40 mm × 40 mm × 140 mm reinforced with a bar having a diameter of ϕ8 mm. Particular groups were a combination of different types of concrete with or without chloride additives, with galvanized or black steel. Chlorides as CaCl2 were added to the concrete mix in the amount of 3% of cement weight in concrete. Reinforced concrete specimens were periodically monitored within two years using the following techniques: linear polarization resistance (LPR) and electrochemical impedance spectroscopy (EIS). Polarization measurements were conducted in a three-electrode arrangement, in which a rebar in concrete served as a working electrode, stainless steel sheet was used as an auxiliary electrode, and Cl−/AgCl,Ag was a reference electrode. Comparative tests of changes in the density of corrosion current in concrete specimens without chloride additives basically demonstrated no development of corrosion, and possible passivation was expected in case of black steel. Higher densities of corrosion current were observed for galvanized steel during first days of testing. The reason was the dissolution of zinc after the contact with initially high pH of concrete pore solution. Six-month measurements demonstrated a higher density of corrosion current in concrete specimens with high concentration of chlorides, which unambiguously indicated corrosion in concrete reinforced with galvanized or black steel. Densities of corrosion current determined for selected specimens dramatically decreased after an 18-month interval in measurements. Corrosion was even inhibited on black steel as an insulating barrier of corrosion products was formed. The above observations were confirmed with structural studies using scanning electron microscopy (SEM) and energy dispersive spectroscopy (EDS) techniques. Results obtained from corrosion (LPR, EIS) and structural (SEM, EDS) tests on specimens of concrete reinforced with steel B500SP demonstrated a very favorable impact of zinc coating on steel by providing two-year protection against corrosion in the environment with very high chloride content.


Introduction
Issues addressed to the durability of reinforced concrete structures have been much tested and described in scientific papers of an interdisciplinary nature [1,2]. Bonds formed between concrete and

Materials
Twelve concrete test elements with dimensions of 40 mm × 40 mm × 140 mm, reinforced with single steel bars were prepared for tests. Two batches of standard concrete mixes were used to place the concrete [47]. In both cases the mix contained 489 kg/m 3 of CEM I 42.5-SR3/NA cement, 1669 kg/m 3 of sand and aggregate of grain size of 2-8 mm. Water-cement ratio (w/c) was 0.45. About 3% of CaCl 2 by cement weight, dissolved in batched water, was added to one of those mixes. In conclusion, six test elements were prepared from concrete without any additives, and another group of six specimens was prepared from concrete with chloride additives.
Each concrete test element was reinforced with a single ribbed rebar with a diameter of φ8 mm, made of steel of grade B500SP classified to steel class C of high ductility [48]. As claimed by the manufacturer, maximum values of additives to low-carbon steel for elements, in which content exceeds 0.5% should not be greater than 1.65% for Mn, 0.85% for Cu, and 0.6% for Si.
In accordance with the primary purpose of studies, half of rebars made of steel of grade B500SP, used as the reinforcement in concrete test elements, was coated in the HDG process. For that purpose, holders were welded to cleaned bars cut to 1.5 m long pieces to provide galvanizing in a vertical position. Rebars placed on plating racks were successively immersed in a degreasing, pickling and finally fluxing bath. Prior to the immersion in the galvanizing bath, rebars were dried in a dryer. Bathing in molten zinc, at a temperature of ca. 450 • C, lasted 120 s. The obtained zinc coatings on steel rebars had an average thickness of ca. 100 µm, which met standard requirements [49,50]. Before the tests, the thickness of the zinc coating was initially verified under the metallographic microscope Delta Optical MET-200-TRF (Delta Optical Ltd., Nowe Osiny, Poland).
Before concreting the specimens, galvanized or black steel rebars were cut into sections of 15 cm in length. Ends of each rebar were protected with shrink film against crevice corrosion in zones, where rebars expanded from concrete. Then, the rebars were stabilized in standard moulds for preparing specimens of masonry mortar, and then concreted. After two days from concreting, the specimens were stripped from the mould. Two weeks after concreting, polarization measurements of the reinforcement began.
In summary, six concrete test elements were reinforced with galvanized rebars, and other six with black steel rebars. As two concrete mixes were used during tests, the total number of twelve test elements was divided into four groups, each of them containing three identical specimens. To differentiate clearly those groups, different colors and symbols were assigned to each of them as shown schematically in Figure 1: • blue color and B symbol-black steel in concrete without any additives, • green color and G symbol-galvanized steel in concrete without any additives, • red color and B-Cl symbol-black steel in concrete with chlorides, • violet color and G-Cl symbol-galvanized steel in concrete with chlorides.
Materials 2020, 13, 3315 4 of 26 Each concrete test element was reinforced with a single ribbed rebar with a diameter of ϕ8 mm, made of steel of grade B500SP classified to steel class C of high ductility [48]. As claimed by the manufacturer, maximum values of additives to low-carbon steel for elements, in which content exceeds 0.5% should not be greater than 1.65% for Mn, 0.85% for Cu, and 0.6% for Si.
In accordance with the primary purpose of studies, half of rebars made of steel of grade B500SP, used as the reinforcement in concrete test elements, was coated in the HDG process. For that purpose, holders were welded to cleaned bars cut to 1.5 m long pieces to provide galvanizing in a vertical position. Rebars placed on plating racks were successively immersed in a degreasing, pickling and finally fluxing bath. Prior to the immersion in the galvanizing bath, rebars were dried in a dryer. Bathing in molten zinc, at a temperature of ca. 450 °C, lasted 120 s. The obtained zinc coatings on steel rebars had an average thickness of ca. 100 µm, which met standard requirements [49,50]. Before the tests, the thickness of the zinc coating was initially verified under the metallographic microscope Delta Optical MET-200-TRF (Delta Optical Ltd., Nowe Osiny, Poland).
Before concreting the specimens, galvanized or black steel rebars were cut into sections of 15 cm in length. Ends of each rebar were protected with shrink film against crevice corrosion in zones, where rebars expanded from concrete. Then, the rebars were stabilized in standard moulds for preparing specimens of masonry mortar, and then concreted. After two days from concreting, the specimens were stripped from the mould. Two weeks after concreting, polarization measurements of the reinforcement began.
In summary, six concrete test elements were reinforced with galvanized rebars, and other six with black steel rebars. As two concrete mixes were used during tests, the total number of twelve test elements was divided into four groups, each of them containing three identical specimens. To differentiate clearly those groups, different colors and symbols were assigned to each of them as shown schematically in Figure 1: • blue color and B symbol-black steel in concrete without any additives, • green color and G symbol-galvanized steel in concrete without any additives, • red color and B-Cl symbol-black steel in concrete with chlorides, • violet color and G-Cl symbol-galvanized steel in concrete with chlorides. Assigned symbols and colors are consequently used in this paper to describe and analyze performed tests.

Electrochemical Polarization Tests
Polarization measurements of the reinforcement in concrete test elements were conducted in a three-electrode arrangement illustrated in Figure 2. Galvanized or black steel rebar 1 embedded in concrete 2 with or without chloride additives served as the working electrode. The auxiliary electrode Assigned symbols and colors are consequently used in this paper to describe and analyze performed tests.

Electrochemical Polarization Tests
Polarization measurements of the reinforcement in concrete test elements were conducted in a three-electrode arrangement illustrated in Figure 2. Galvanized or black steel rebar 1 embedded in concrete 2 with or without chloride additives served as the working electrode. The auxiliary electrode 3 was made of a 2 mm thick stainless-steel sheet having the dimensions of 140 mm × 40 mm. A wet felt 4 with the dimensions of 140 cm × 40 cm provided the electric contact between the sheet 3 and the top surface of concrete. The reference electrode 5 (Cl − /AgCl,Ag) was placed in a vertical ballast guide 6 to provide its stabilisation during tests. The ballast 6 was used to provide the proper adhesion of the auxiliary electrode 3 through the felt 4 to the top surface of the concrete specimen 2. All three electrodes were connected to the potentiostat 7 Gamry Reference 600.
Materials 2020, 13, 3315 5 of 26 3 was made of a 2 mm thick stainless-steel sheet having the dimensions of 140 mm × 40 mm. A wet felt 4 with the dimensions of 140 cm × 40 cm provided the electric contact between the sheet 3 and the top surface of concrete. The reference electrode 5 (Cl − /AgCl,Ag) was placed in a vertical ballast guide 6 to provide its stabilisation during tests. The ballast 6 was used to provide the proper adhesion of the auxiliary electrode 3 through the felt 4 to the top surface of the concrete specimen 2. All three electrodes were connected to the potentiostat 7 Gamry Reference 600. Electrochemical tests were performed on the above measuring system using the following techniques: linear polarization resistance (LPR) and electrochemical impedance spectroscopy (EIS). The specimens were kept at ca. 20 ± 2 °C and the relative humidity of 50 ± 10%. Prior to beginning polarization tests on a given day, the top surface of the concrete specimen was immersed in tap water to the depth up to 5 ± 1 mm for ca. 30 min to ensure better conductivity of concrete cover in the tested reinforcement during measurements. After taking from water, the specimens were connected to the potentiostat 7 and changes of gradually stabilizing potential were observed with the reference electrode 5 for 60-120 min. When potential changes were at the level of 0.1 mV/s, EIS methods were performed on the steel reinforcement in concrete. Testing was performed at the fixed range of frequencies of 0.01 Hz-1 kHz in the potentiostatic mode with a disturbing sinusoidal signal at the potential amplitude of 10 mV over the corrosion potential. Then, after a few minutes or a longer interval when changes in the corrosion potential did not exceed the level of 0.1 mV/s, LPR tests were conducted in the potentiodynamic mode. The reinforcement was polarized at a rate of 1 mV/s within the range of potential changes from −150 mV to +50 mV regarding the corrosion potential.
The first measuring series began two weeks after placing specimens in concrete. Other five measuring series were conducted in one-and two-week intervals. At the next stage, six measuring series were taken at a two-week interval. Final measuring series took place in a month interval. After 18 months, prior to scheduled microscopic measurements, EIS and LPR tests were repeated but only on four specimens representing each of B, G, B-Cl, and G-Cl groups. At the final testing stage, results obtained from polarization measurements were compared with microscopic observations and analyses.

Structural Tests
The specimens for structural tests were cut out perpendicularly to the rebar axis. Metallographic specimens were prepared from those cut pieces of concrete with rebars made of steel B500SP. Tests on the macrostructure of concrete specimens were performed at the stand specially prepared for macro testing. And microstructure of concrete was tested using Olympus SZX9 stereo microscope Electrochemical tests were performed on the above measuring system using the following techniques: linear polarization resistance (LPR) and electrochemical impedance spectroscopy (EIS). The specimens were kept at ca. 20 ± 2 • C and the relative humidity of 50 ± 10%. Prior to beginning polarization tests on a given day, the top surface of the concrete specimen was immersed in tap water to the depth up to 5 ± 1 mm for ca. 30 min to ensure better conductivity of concrete cover in the tested reinforcement during measurements. After taking from water, the specimens were connected to the potentiostat 7 and changes of gradually stabilizing potential were observed with the reference electrode 5 for 60-120 min. When potential changes were at the level of 0.1 mV/s, EIS methods were performed on the steel reinforcement in concrete. Testing was performed at the fixed range of frequencies of 0.01 Hz-1 kHz in the potentiostatic mode with a disturbing sinusoidal signal at the potential amplitude of 10 mV over the corrosion potential. Then, after a few minutes or a longer interval when changes in the corrosion potential did not exceed the level of 0.1 mV/s, LPR tests were conducted in the potentiodynamic mode. The reinforcement was polarized at a rate of 1 mV/s within the range of potential changes from −150 mV to +50 mV regarding the corrosion potential.
The first measuring series began two weeks after placing specimens in concrete. Other five measuring series were conducted in one-and two-week intervals. At the next stage, six measuring series were taken at a two-week interval. Final measuring series took place in a month interval. After 18 months, prior to scheduled microscopic measurements, EIS and LPR tests were repeated but only on four specimens representing each of B, G, B-Cl, and G-Cl groups. At the final testing stage, results obtained from polarization measurements were compared with microscopic observations and analyses.

Structural Tests
The specimens for structural tests were cut out perpendicularly to the rebar axis. Metallographic specimens were prepared from those cut pieces of concrete with rebars made of steel B500SP. Tests on the macrostructure of concrete specimens were performed at the stand specially prepared for macro testing. And microstructure of concrete was tested using Olympus SZX9 stereo microscope and GX51 inverted microscope (LM) (Olympus Corporation, Tokyo, Japan).
Advanced tests on the specimen structure at selected points were performed with HITAHI S-3400N scanning electron microscope (SEM) (Hitachi, Tokyo, Japan) at an accelerating voltage within the range of 15-25 kV. Tests on chemical composition at selected micro-areas of the specimens were performed with the energy dispersive spectroscopy (EDS) using the spectrometer by THERMO NORAN company with SYSTEM SIX software (Thermo Fisher Scientific Inc., Waltham, MA, USA). As microscopic tests are invading into the structure of the specimen material, they were performed after the final stage of electrochemical tests. Therefore, those tests should be only related to results obtained after the last, 792 day of measurements. Figure 3 shows a comparison of results from 160 measurements of corrosion potential of the steel reinforcement in concrete from twelve test elements. Figure 3a presents results for concrete without any additives, whereas Figure 3b refers to concrete with calcium chloride. All measurements were performed against the silver chloride electrode. Advanced tests on the specimen structure at selected points were performed with HITAHI S-3400N scanning electron microscope (SEM) (Hitachi, Tokyo, Japan) at an accelerating voltage within the range of 15-25 kV. Tests on chemical composition at selected micro-areas of the specimens were performed with the energy dispersive spectroscopy (EDS) using the spectrometer by THERMO NORAN company with SYSTEM SIX software (Thermo Fisher Scientific Inc., Waltham Massachusetts, USA). As microscopic tests are invading into the structure of the specimen material, they were performed after the final stage of electrochemical tests. Therefore, those tests should be only related to results obtained after the last, 792 day of measurements. Figure 3 shows a comparison of results from 160 measurements of corrosion potential of the steel reinforcement in concrete from twelve test elements. Figure 3a presents results for concrete without any additives, whereas Figure 3b refers to concrete with calcium chloride. All measurements were performed against the silver chloride electrode. Core tests were conducted in less than six months and restarted after about an 18-month interval, but only on four selected specimens that were further subjected to microscopic tests.

Linear Polarization Resistance (LPR) Tests
Linear polarization resistance (LPR) tests were the last tests from the group of electrochemical Core tests were conducted in less than six months and restarted after about an 18-month interval, but only on four selected specimens that were further subjected to microscopic tests.

Linear Polarization Resistance (LPR) Tests
Linear polarization resistance (LPR) tests were the last tests from the group of electrochemical tests performed on the reinforced concrete specimen. Contrary to electrochemical impedance spectroscopy (EIS) tests regarded as complementary measurements, LPR tests were conducted on each test element from each measuring series. The whole period of testing produced a total of 160 polarization curves, in which exemplary shapes for three selected measuring series are illustrated in Figure 4. Characteristic time points for measurements were the following measuring days: 4, 74, and 792. The selected measuring days, marked with grey color in Figure 3, were to provide a more detailed presentation and analysis of test results from the initial phase of measurements (day 4), the halfway point of tests (day 74) and from the period directly before planned microscopic tests and analyses (day 792).
On day 4 of measurements, shapes of polarization curves ( Figure 4a) were more arranged showing the clear formation of four groups. The corrosion potential of the specimens with black steel was less negative than in case of galvanized steel. The mean potential value in the specimen groups B and B-Cl was −0.28 V and −0.45 V, respectively. For the groups G and G-Cl, it was equal to −0.73 V and −0.91 V, respectively. However, the most significant information obtained from the analysis of polarization curves was the polarization resistance R p was inversely proportional to the density of corrosion current i corr . The Stern-Geary Equation (1) was used to calculate values of i corr , taking into account coefficients of rectilinear slope for segments of polarization curves-anodic b a and cathodic b c .
Parameters of the Equation (1) taken for calculations of i corr are shown in Appendix A- Table A1, which contains results for only 28 out of 160 analyzed polarization curves. Mean densities of corrosion current for individual groups used further in the paper, demonstrated that the lowest value of i corr,mid = 0.31 µA/cm 2 was obtained for the specimens from the group B. Because of naturally high alkaline pH (pH > 12.5) of concrete pore solution without additives, the obtained current density could indicate the passivation of reinforcing steel in concrete. Adding calcium chloride to the concrete mix caused the same black steel started to corrode in the specimens from the group B-Cl with i corr.mid = 1.65 µA/cm 2 . And the zinc coating on the reinforcement also began to dissolve electrochemically at pH > 13.3 when in direct contact with highly alkaline pore solution. The mean density of corrosion current i corr.mid = 0.93 µA/cm 2 calculated for the group G could with high probability indicate that occurrence typical for the initial phase of concrete curing. The final, the highest mean density of corrosion current i corr.mid = 1.98 µA/cm 2 referred to galvanized reinforcing steel in concrete with chloride additives. For the specimen group G-Cl, it could indicate the accumulation of zinc corrosion caused by high pH of concrete pore solution and aggressive impact of chloride ions.
On day 74 of measurements, distributions of polarization curves ( Figure 4b) were slightly shifted when compared to the distribution of the same specimens presented in Figure 4a. The corrosion potential of black steel in concrete without chlorides, averaged for the same group of specimens, was considerably reduced to E corr.mid = −0.40 V, and in concrete with chlorides the potential slightly increased to E corr.mid = −0.37 V. For galvanized steel in concrete without chlorides, a clear drop in potential to the value E corr.mid = −0.85 V was observed with the simultaneous increase in potential to the following level  After about two years of measurements (day 792), prior to the commencement of microscopic tests, polarization curves for individual specimens representing each of four groups ( Figure 5) showed an interesting change. A noticeable change was found for black steel in concrete with chlorides (B7-Cl) because its corrosion potential largely increased to −0.04 V, but what is more important, the density of corrosion current dramatically decreased to just 0.01 µA/cm 2 . Taking into account the fact the analyzed reinforced concrete specimen was potentially the most vulnerable to The analysis of calculated mean densities of corrosion current on day 74 of measurements indicated in concrete with chlorides still low values of i corr.mid = 0.27 µA/cm 2 for black steel and a dramatic drop in i corr.mid to 0.14 µA/cm 2 for galvanized steel. The passivation of black steel in concrete without additives was a natural process, but no corrosion of zinc coating on steel reinforcement after several dozens of days in contact with the pore solution was desired despite the initially intensive dissolution of zinc. In concrete specimens with chlorides, the corrosion of black steel i corr.mid = 0.71 µA/cm 2 and galvanized steel i corr.mid = 1.16 µA/cm 2 was still observed on day 74 of measurements. But mean densities of corrosion current were slightly lower than on the first days of measurements.
After about two years of measurements (day 792), prior to the commencement of microscopic tests, polarization curves for individual specimens representing each of four groups ( Figure 5) showed an interesting change. A noticeable change was found for black steel in concrete with chlorides (B7-Cl) because its corrosion potential largely increased to −0.04 V, but what is more important, the density of corrosion current dramatically decreased to just 0.01 µA/cm 2 . Taking into account the fact the analyzed reinforced concrete specimen was potentially the most vulnerable to corrosion, the only possible explanation for the above situation was the high likelihood of the large accumulation of corrosion products (rust) around the steel rebar, and consequently, a serious limitation of direct access for oxygen and humidity to its surface. Hence, electrode processes on steel were stopped. For galvanized reinforcing steel in concrete with chlorides (G10-Cl), the corrosion current density dropped to 0.34 µA/cm 2 . The likely reason was the accumulation of corrosion products of zinc on the coating slightly slowing down its further dissolution. After two years of measurements, black steel (B1) and galvanized steel (G4) in the concrete specimens without chlorides had stable but low density of corrosion current (0.35 µA/cm 2 and 0.29 µA/cm 2 , respectively), which suggested the reinforcement corrosion was not likely to develop.
Materials 2020, 13, 3315 9 of 26 corrosion, the only possible explanation for the above situation was the high likelihood of the large accumulation of corrosion products (rust) around the steel rebar, and consequently, a serious limitation of direct access for oxygen and humidity to its surface. Hence, electrode processes on steel were stopped. For galvanized reinforcing steel in concrete with chlorides (G10-Cl), the corrosion current density dropped to 0.34 µA/cm 2 . The likely reason was the accumulation of corrosion products of zinc on the coating slightly slowing down its further dissolution. After two years of measurements, black steel (B1) and galvanized steel (G4) in the concrete specimens without chlorides had stable but low density of corrosion current (0.35 µA/cm 2 and 0.29 µA/cm 2 , respectively), which suggested the reinforcement corrosion was not likely to develop.

Electrochemical Impedance Spectroscopy (EIS) Tests
Electrochemical impedance spectroscopy (EIS) tests on the specimens of concrete with reinforcement were additional to core polarization tests performed by linear polarization resistance (LPR) method. Impedance tests focused only on three selected measuring series on day 4, 74, and 792. Figure 6a illustrates in the complex plane results from impedance tests on the steel reinforcement in concrete from 12 specimens on day 4 of measurements, whereas Figure 6b presents the same results as the function of phase shift angle ϕ against the logarithm of measurement frequency f. For the specimens from groups B and G made of concrete without chlorides, spectra at the Nyquist plots ( Figure 6a) exhibited capacitive behavior similar in shape to parts of semi-circles, in which

Electrochemical Impedance Spectroscopy (EIS) Tests
Electrochemical impedance spectroscopy (EIS) tests on the specimens of concrete with reinforcement were additional to core polarization tests performed by linear polarization resistance (LPR) method. Impedance tests focused only on three selected measuring series on day 4, 74, and 792. Figure 6a illustrates in the complex plane results from impedance tests on the steel reinforcement in concrete from 12 specimens on day 4 of measurements, whereas Figure 6b presents the same results as the function of phase shift angle φ against the logarithm of measurement frequency f. For the specimens from groups B and G made of concrete without chlorides, spectra at the Nyquist plots ( Figure 6a) exhibited capacitive behavior similar in shape to parts of semi-circles, in which extrapolation to the horizontal axis of real impedance provided the possibility of estimating charge transfer resistance R t across the interface. It is interesting to note that semi-circles referring to galvanized steel had slightly smaller conventional diameter. The probable reason was the dissolution of the zinc coating in highly alkaline (pH > 13.3) of concrete pore solution.
For the specimens from groups B-Cl and G-Cl made of concrete with chlorides, the size of spectra at the Nyquist plot was much smaller and the second time-constant was clearly noticeable. Shapes of those spectra were rather characteristic for the development of metal (Fe or Zn) corrosion in concrete. Observations made for the Nyquist plot ( Figure 6a) were additionally confirmed by the Bode plot (Figure 6b). Low frequencies typical for metal in concrete without additives (groups B and G) displayed quite wide angles of phase shift (−20 • -−40 • ), whereas the specimens of concrete with chlorides (groups B-Cl and G-Cl) reached the maximum level of −15 • . The observed shapes of those spectra, which were very distinctive, quite explicitly indicated the corrosion of metal in the specimens from groups B-Cl and G-Cl. For galvanized steel (group G) the situation was slightly less favorable, which could suggest not so intensive processes of zinc dissolution.
The above qualitative analysis of obtained spectra was confirmed by the quantitative analysis. For that purpose, the equivalent electrical circuit illustrated in Figure 6(a1), was proposed for spectra of concrete specimens without chlorides. It was the classic Randles circuit, in which the double layer capacity was replaced with a constant phase element CPE 0 . Parameter R c is resistance of concrete, R t is charge transfer resistance across the interface. The electric circuit shown in Figure 6(a2), which simulated the porous coating, was used for concrete specimens with chlorides. Besides already mentioned parameters of the Randles circuit, that circuit had two characteristics of the electrochemical zinc coating on the steel reinforcement: R f -resistance of the coating layer and CPE f -constant phase element characterizing its quasi capacity. The Simplex method was used to find numerical values of elements in equivalent circuits, and those values are presented in Appendix A- Table A2. The summary table presenting calculated results, also contain numerical parameters of constant phase elements: CPE 0 -Y 0 and α 0 and CPE f -Y f and α 0 .
On day 74 of measurements, the impedance spectra at the Nyquist plot ( Figure 7a) obtained for all four groups of the specimens were similar to spectra from day 4 ( Figure 6a). The main difference referred to the specimens with galvanized steel, for which the spectra at the Nyquist plot had a slightly different shape resembling a typical image of metal passivated in concrete at low frequencies.
That observation was confirmed by the Bode plot (Figure 7b), at which phase shift angles of galvanized steel at low frequencies were greater by ca. 5 • (with reference to absolute values) than for black steel. Relationships between shapes of impedance spectra were rather typical for zinc behavior in hardened concrete without any additives.
For the specimens of concrete with chlorides, day 74 of measurements revealed a small drop in the intensity of corrosion processes of galvanized and black steel. The presented deduction resulted from the quantitative analysis on the basis of the equivalent electrical circuits illustrated in Figure 7(a1,a2), similar to diagrams shown in Figure 6(a1,a2). That phenomenon could be explained by a gradual accumulation of corrosion products of iron and zinc on the surface of steel reinforcement in the specimen, which additionally were deposited in concrete pores. Dissolution of metals (Fe and Zn) was slightly inhibited mainly due to the limited access of oxygen to the surface of both metals.  reinforcing steel in concrete was inhibited despite the presence of aggressive chloride ions. A continuous increase in the volume of corrosion products was supposed to inhibit seriously the access of oxygen to the steel surface, and consequently to eliminate the agent required for the development of electrochemical corrosion of steel in concrete.  The last series of impedance measurement was made on day 792, that is, after about two years of measurements and included only four specimens (B1, G4, B7-Cl, G10-Cl) being the one-element representation of all four groups (B, G, B-Cl, G-Cl) of test elements. The Nyquist plot (Figure 8a) shows one dominating spectrum (B7-Cl) and three considerably smaller spectra near the beginning of the coordinate system. For better clearness of the results, those three spectra are again presented below, in the second complex plane with 20-time smaller values of complex and real impedance. The clear dominance of one spectrum over others indicated the impedance of the whole steel-concrete system in the specimen B7-Cl was about 20 times greater than the impedance of the other three specimens. As all impedance measurements on the final measuring series (day 792) were conducted within a broader range of frequencies 0.01 Hz-100 kHz, particularly the spectrum B7-Cl had a rather interesting shape. It had (from the left side) a high-frequency, slightly oblate semi-circle expressing electrochemical properties of concrete, and from the point of contraflexure a rectilinear part changing into the arc shape. The last described part could indicate diffusion control over the corrosion process. The spectrum shape for B7-Cl in the complex plane and at the Bode plot demonstrated a very strong slowdown on day 792 of measurements, comparable to the situation, in which dissolution of reinforcing steel in concrete was inhibited despite the presence of aggressive chloride ions. A continuous increase in the volume of corrosion products was supposed to inhibit seriously the access of oxygen to the steel surface, and consequently to eliminate the agent required for the development of electrochemical corrosion of steel in concrete. The qualitative evaluation and adjustment with the Simplex method of the theoretical spectrum, according to the diagram shown in Figure 8a1 confirmed the above observations as the determined charge transfer resistance Rt reached 1680 kΩcm 2 , which corresponded to the very low density of corrosion current icorr = 0.01 µA/cm 2 Impedance spectra of the other three specimens (B1, G4, G10-Cl) at the Nyquist plot were rather similar in shape, particularly with reference to the low-frequency range describing the properties of metal in concrete. The qualitative analysis conducted with the electrical equivalent circuit illustrated in Figure 8a2 also confirmed similar values of charge transfer resistance Rt = 34 k-50 kΩcm 2 , which corresponded to very low densities of corrosion current icorr = 0.29-0.35 µA/cm 2 . According to criteria published in [51], determined for rebars in concrete, such low densities of corrosion current usually indicate a very low level of steel corrosion. Moreover, each spectrum had a noticeably different inflexion point (that is, the closest point to the horizontal axis of real impedance), from which electrochemical properties of concrete were revealed within the low-frequency range. Diversified inflection points of those spectra indicated different values of concrete resistivity during measurements under the impact of alternating current flowing between reinforcing steel and the auxiliary electrode, placed on the top surface of the concrete specimen. The analyzed cases showed concrete in the specimen G4 had the lowest resistivity, and concrete in the specimen G10-Cl had the highest resistivity.  The qualitative evaluation and adjustment with the Simplex method of the theoretical spectrum, according to the diagram shown in Figure 8(a1) confirmed the above observations as the determined charge transfer resistance R t reached 1680 kΩcm 2 , which corresponded to the very low density of corrosion current i corr = 0.01 µA/cm 2 Impedance spectra of the other three specimens (B1, G4, G10-Cl) at the Nyquist plot were rather similar in shape, particularly with reference to the low-frequency range describing the properties of metal in concrete. The qualitative analysis conducted with the electrical equivalent circuit illustrated in Figure 8(a2) also confirmed similar values of charge transfer resistance R t = 34 k-50 kΩcm 2 , which corresponded to very low densities of corrosion current i corr = 0.29-0.35 µA/cm 2 . According to criteria published in [51], determined for rebars in concrete, such low densities of corrosion current usually indicate a very low level of steel corrosion. Moreover, each spectrum had a noticeably different inflexion point (that is, the closest point to the horizontal axis of real impedance), from which electrochemical properties of concrete were revealed within the low-frequency range. Diversified inflection points of those spectra indicated different values of concrete resistivity during measurements under the impact of alternating current flowing between reinforcing steel and the auxiliary electrode, placed on the top surface of the concrete specimen. The analyzed cases showed concrete in the specimen G4 had the lowest resistivity, and concrete in the specimen G10-Cl had the highest resistivity.

Results from Structural Tests and Their Analysis
Corrosion tests indicated the change in parameters of corrosion current over time which could suggest structural changes at the concrete-rebar interface. Therefore, microstructure and chemical compositions at the interface between concrete and rebar made of steel B500SP were analyzed for all four tested cases after completing corrosion tests: • black reinforcing steel B500SP in concrete without any additives, • galvanized reinforcing steel B500SP in concrete without chlorides, • black reinforcing steel B500SP in concrete with chlorides • galvanized reinforcing steel B500SP in concrete with chlorides.

General Description of the Microstructure of Concrete Reinforced with Rebars Made of Steel B500SP (Stereo, LM)
Tests on concrete-reinforcement joints began from analyzing its microstructure after completing corrosion tests, using the stereo microscope ( Figure 9). Figure 9a presents the cross-section (with dimensions of 40 mm × 40 mm) of the tested concrete specimen with a galvanized rebar concreted in the central point, made of ribbed steel of grade B500SP and with a diameter of φ8 mm. Figure 9b illustrates the close-up of the cross-section of steel rebar with a clearly visible, thin, silver and circumferential zinc coating. There are noticeable particles of coarse and fine (sand) aggregates near the rebar. Aggregates are tied together with cement gel. Figure 9c,d show another close-up of a strongly diversified thickness of the zinc coating, whose thickest layer is observed near oblique ribs of the reinforcing steel.
The tests conducted on the microstructure of reinforced concrete specimens using the light microscope revealed good conditions of joints between concrete and the reinforcement for rebars made of black steel (Figure 10a). And simultaneous observations confirmed previous conclusions made on the basis of tests using the stereo microscope (Figure 9) concerning the non-uniform thickness of zinc coating on rebars (Figure 10b).

General Description of the Microstructure of Concrete Reinforced with Rebars Made of Steel B500SP (Stereo, LM)
Tests on concrete-reinforcement joints began from analyzing its microstructure after completing corrosion tests, using the stereo microscope ( Figure 9). Figure 9a presents the cross-section (with dimensions of 40 mm × 40 mm) of the tested concrete specimen with a galvanized rebar concreted in the central point, made of ribbed steel of grade B500SP and with a diameter of ϕ8 mm.   Figure 9c,d show another close-up of a strongly diversified thickness of the zinc coating, whose thickest layer is observed near oblique ribs of the reinforcing steel.
The tests conducted on the microstructure of reinforced concrete specimens using the light microscope revealed good conditions of joints between concrete and the reinforcement for rebars made of black steel (Figure 10a). And simultaneous observations confirmed previous conclusions

Black Reinforcing Steel B500SP in Concrete with and without Chlorides
The analysis of the microstructure of joints between concrete without chlorides and rebar made of B500SP black steel ( Figure 11) showed a difference in concrete structure regarding its phase composition (Figure 11a-image from BSE detector) and rebar correctly embedded in concrete. The analysis of chemical composition at the interface between concrete and rebar ( Figure 12) indicated the presence of elements being components of concrete or steel rebar. Oxygen (Figure 12b), probably in the form of oxide, was found on the rebar surface. Oxidation of zinc coating is a characteristic effect of its destruction [52].

Black Reinforcing Steel B500SP in Concrete with and without Chlorides
The analysis of the microstructure of joints between concrete without chlorides and rebar made of B500SP black steel ( Figure 11) showed a difference in concrete structure regarding its phase composition (Figure 11a-image from BSE detector) and rebar correctly embedded in concrete. The analysis of chemical composition at the interface between concrete and rebar ( Figure 12) indicated the presence of elements being components of concrete or steel rebar. Oxygen (Figure 12b), probably in the form of oxide, was found on the rebar surface. Oxidation of zinc coating is a characteristic effect of its destruction [52].

Black Reinforcing Steel B500SP in Concrete with and without Chlorides
The analysis of the microstructure of joints between concrete without chlorides and rebar made of B500SP black steel ( Figure 11) showed a difference in concrete structure regarding its phase composition (Figure 11a-image from BSE detector) and rebar correctly embedded in concrete. The analysis of chemical composition at the interface between concrete and rebar ( Figure 12) indicated the presence of elements being components of concrete or steel rebar. Oxygen (Figure 12b), probably in the form of oxide, was found on the rebar surface. Oxidation of zinc coating is a characteristic effect of its destruction [52].  As expected, no traces of corrosion damage at the interface between steel and concrete were found, what was normal in the highly alkaline environment of concrete. Results from polarization tests on corrosion current density also indirectly suggested the passivation of steel, which was typical for such a system.
The analysis of joints between concrete with chlorides and black reinforcing steel B500SP indicated a clear change in the nature of the joint when compared to concrete without chlorides ( Figure 11). The presence of chlorides in concrete resulted in a large number of corrosion products at the rebar/concrete interface ( Figure 13). Corrosion products of the steel reinforcement (rust) did not only accumulate on the surface of the rebar, but also penetrated the porous structure of concrete. A considerable concentration of Cl and oxygen (Figure 14b) was found in these products which probably indicated the presence of chlorides and oxides in that area.  As expected, no traces of corrosion damage at the interface between steel and concrete were found, what was normal in the highly alkaline environment of concrete. Results from polarization tests on corrosion current density also indirectly suggested the passivation of steel, which was typical for such a system.
The analysis of joints between concrete with chlorides and black reinforcing steel B500SP indicated a clear change in the nature of the joint when compared to concrete without chlorides ( Figure 11). The presence of chlorides in concrete resulted in a large number of corrosion products at the rebar/concrete interface ( Figure 13). Corrosion products of the steel reinforcement (rust) did not only accumulate on the surface of the rebar, but also penetrated the porous structure of concrete. A considerable concentration of Cl and oxygen (Figure 14b) was found in these products which probably indicated the presence of chlorides and oxides in that area.
No concrete cracking in the interfacial zone was seen which indicated critical stresses blowing out the concrete cover from the inside did not occur. It should be noticed that the layer of corrosion products was thick enough to make the diffusion of oxygen to the steel surface very difficult. That effect was evident in results from impedance tests (Figure 8a) due to the appearance of the additional time constant.
indicated a clear change in the nature of the joint when compared to concrete without chlorides ( Figure 11). The presence of chlorides in concrete resulted in a large number of corrosion products at the rebar/concrete interface ( Figure 13). Corrosion products of the steel reinforcement (rust) did not only accumulate on the surface of the rebar, but also penetrated the porous structure of concrete. A considerable concentration of Cl and oxygen (Figure 14b) was found in these products which probably indicated the presence of chlorides and oxides in that area.  the rebar/concrete interface ( Figure 13). Corrosion products of the steel reinforcement (rust) did not only accumulate on the surface of the rebar, but also penetrated the porous structure of concrete. A considerable concentration of Cl and oxygen (Figure 14b) was found in these products which probably indicated the presence of chlorides and oxides in that area.

Reinforcing Steel B500SP with Zinc Coating, in Concrete with and without Chlorides
The analysis of the joint between concrete without chlorides and the reinforcing steel B500SP with zinc coating demonstrated a clear difference the coating thickness ( Figure 15). It varied from more than ten (78.4 µm) to over 600 µm, but usually was within the range from 150 µm to 250 µm. The thickest zinc coating (620 µm) was near one of oblique ribs of the rebar. The variable thickness of the zinc coating is an effect directly resulting from the technology of the galvanizing process, but it can also be the result of the preparation of metallographic samples-the zinc "rubs" on the surface of the specimen. In this case, it is the effect of galvanizing technology. The analysis of chemical composition for the section of the zinc coating indicated its great uniformity ( Figure 16). That result was confirmed by testing the relative concentration elements on the surface (mapping) of the concrete-rebar joint ( Figure 17).
Polarization tests on the galvanized reinforcement in concrete without chlorides indicated some increase in corrosion current density on first days of measurements when compared to black steel. This effect is usually noticed during the first weeks of concrete curing. In this case, the microscopic analysis referred to a two-year old concrete, for which processes of dissolution of zinc coating on steel are not usually observed or have been stabilized. Therefore, it is difficult to determine the original thickness of the zinc coating. It may be wondered whether it's very non-uniform thickness observed under the microscope was only caused by the specificity of the technological process of the formation of hot-dip galvanized zinc coating or the effect of partial dissolution of zinc in the highly alkaline environment of concrete pore solution.
of the zinc coating is an effect directly resulting from the technology of the galvanizing process, but it can also be the result of the preparation of metallographic samples-the zinc "rubs" on the surface of the specimen. In this case, it is the effect of galvanizing technology. The analysis of chemical composition for the section of the zinc coating indicated its great uniformity (Figure 16). That result was confirmed by testing the relative concentration elements on the surface (mapping) of the concrete-rebar joint (Figure 17).  it can also be the result of the preparation of metallographic samples-the zinc "rubs" on the surface of the specimen. In this case, it is the effect of galvanizing technology. The analysis of chemical composition for the section of the zinc coating indicated its great uniformity (Figure 16). That result was confirmed by testing the relative concentration elements on the surface (mapping) of the concrete-rebar joint (Figure 17).  Polarization tests on the galvanized reinforcement in concrete without chlorides indicated some increase in corrosion current density on first days of measurements when compared to black steel. This effect is usually noticed during the first weeks of concrete curing. In this case, the microscopic analysis referred to a two-year old concrete, for which processes of dissolution of zinc coating on steel are not usually observed or have been stabilized. Therefore, it is difficult to determine the original thickness of the zinc coating. It may be wondered whether it's very non-uniform thickness observed The analysis of joint between concrete with chlorides and galvanized reinforcing steel B500SP indicated significant non-uniformity of the zinc coating on the rebar with reference to its thickness and morphology (Figure 18), and a similar situation was observed for the bond when no chlorides were present. The thickness of the zinc coating varied from as low as 100 µm (94.4 µm) to over 400 µm (434 µm) (Figure 18a). Morphology of the coating across its cross-section ( Figure 18d) suggested a diverse growth of crystallites in individual micro-areas. The analysis of chemical composition across the coating section (Figures 19 and 20) demonstrated considerable changes, particularly for concentration of Cl, which could locally reach high values (Figure 20b). Chlorine was also the element, which occurred along with the zinc coating (Figure 20c,g). The comparison of the thickness of zinc coating on steel in concrete without chlorides ( Figure  15) and with chlorides ( Figure 18) (though they are different specimens) indicated the presence of chlorides significantly reduced its thickness. That phenomenon was expected as a high concentration of chlorides always causes the corrosion of zinc coating [33]. The analysis of microscopic images The thickness of the zinc coating varied from as low as 100 µm (94.4 µm) to over 400 µm (434 µm) (Figure 18a). Morphology of the coating across its cross-section ( Figure 18d) suggested a diverse growth of crystallites in individual micro-areas. The analysis of chemical composition across the coating section (Figures 19 and 20) demonstrated considerable changes, particularly for concentration of Cl, which could locally reach high values (Figure 20b). Chlorine was also the element, which occurred along with the zinc coating (Figure 20c,g). The comparison of the thickness of zinc coating on steel in concrete without chlorides ( Figure  15) and with chlorides ( Figure 18) (though they are different specimens) indicated the presence of chlorides significantly reduced its thickness. That phenomenon was expected as a high concentration of chlorides always causes the corrosion of zinc coating [33]. The analysis of microscopic images showed the zinc coating at any place was completely damaged. Hence, high densities of corrosion protected by zinc. Corrosion products of zinc, with a smaller volume than those of iron, also did not cause cracks in the concrete cover of the reinforcement. Apparently, corrosion products of zinc filling the porous structure of concrete in the interfacial zone had enough space, and no undesirable critical values of tensile stress in concrete were triggered.

Discussion
The analysis of changes in a function of time of corrosion potential Ecorr of steel reinforcement in concrete very clearly demonstrated ( Figure 4) that corrosion potential of black reinforcing steel, both in case of concrete without additives (group B), and with chlorides (group B-Cl) was relatively close within the range of −0.1-−0.5 V. While evaluating obtained results with reference to standard criteria [53] for testing potential of the reinforcement in reinforced concrete structures, the information about the corrosion probability of 50%-95% was obtained after calculating results per the potential of copper-sulphate electrode. It should be pointed out that the lowest values of potential up to the level of −0.45 V were observed for concrete with chlorides during the first weeks of testing. And distributions of corrosion potential of galvanized reinforcing steel in line with expectations were characterized by lower values which directly resulted from zinc position in the electrochemical series of metals. In concrete without chlorides, zinc as less noble metal than iron, being the main component of tested low-carbon reinforcing steel, had the potential of ca. 0.3-0.5 V lower than the potential of black steel when in direct contact with concrete pore solution. For concrete with chlorides, the measured corrosion potential of galvanized reinforcement demonstrated the similar difference in potentials. However, the spread of potential values for the specimens from the group G-Cl was considerably greater than for the group B-Cl.
From a scientific perspective, the analysis of results from electrochemical tests (particularly density of corrosion current) seems to be very interesting during a 2-year period of performed tests. Figure 21 presents the comparison of changes in corrosion current densities icorr within two years of measurements, obtained on the basis of analyses of polarization curves of reinforcing steel in the concrete specimens. Results for galvanized steel (G) and black steel (B) in concrete without additives are shown in Figure 21a, whereas results for similar rebars in concrete with chlorides are illustrated in Figure 21b. The comparison of the thickness of zinc coating on steel in concrete without chlorides ( Figure 15) and with chlorides ( Figure 18) (though they are different specimens) indicated the presence of chlorides significantly reduced its thickness. That phenomenon was expected as a high concentration of chlorides always causes the corrosion of zinc coating [33]. The analysis of microscopic images showed the zinc coating at any place was completely damaged. Hence, high densities of corrosion current recorded during tests were always related to the dissolution of zinc, and not reinforcing steel protected by zinc. Corrosion products of zinc, with a smaller volume than those of iron, also did not cause cracks in the concrete cover of the reinforcement. Apparently, corrosion products of zinc filling the porous structure of concrete in the interfacial zone had enough space, and no undesirable critical values of tensile stress in concrete were triggered.

Discussion
The analysis of changes in a function of time of corrosion potential E corr of steel reinforcement in concrete very clearly demonstrated (Figure 4) that corrosion potential of black reinforcing steel, both in case of concrete without additives (group B), and with chlorides (group B-Cl) was relatively close within the range of −0.1-−0.5 V. While evaluating obtained results with reference to standard criteria [53] for testing potential of the reinforcement in reinforced concrete structures, the information about the corrosion probability of 50%-95% was obtained after calculating results per the potential of copper-sulphate electrode. It should be pointed out that the lowest values of potential up to the level of −0.45 V were observed for concrete with chlorides during the first weeks of testing. And distributions of corrosion potential of galvanized reinforcing steel in line with expectations were characterized by lower values which directly resulted from zinc position in the electrochemical series of metals. In concrete without chlorides, zinc as less noble metal than iron, being the main component of tested low-carbon reinforcing steel, had the potential of ca. 0.3-0.5 V lower than the potential of black steel when in direct contact with concrete pore solution. For concrete with chlorides, the measured corrosion potential of galvanized reinforcement demonstrated the similar difference in potentials. However, the spread of potential values for the specimens from the group G-Cl was considerably greater than for the group B-Cl.
From a scientific perspective, the analysis of results from electrochemical tests (particularly density of corrosion current) seems to be very interesting during a 2-year period of performed tests. Figure 21 presents the comparison of changes in corrosion current densities i corr within two years of measurements, obtained on the basis of analyses of polarization curves of reinforcing steel in the concrete specimens. Results for galvanized steel (G) and black steel (B) in concrete without additives are shown in Figure 21a, whereas results for similar rebars in concrete with chlorides are illustrated in Figure 21b. As shown in Figure 21a, black reinforcing steel in concrete without additives (B) generally had stable but the low density of corrosion current through the whole period of measurements, which in accordance with criteria for evaluating corrosion risk to steel in concrete, could indicate the passivation. Microscopic tests (Figure 12b) demonstrated the presence of oxygen on the surface of steel rebar which probably indicated the presence of a passive layer of oxides that is characteristic for highly alkaline concrete pore solution (pH > 12.5).
In case of galvanized steel reinforcement in concrete without chlorides (G), elevated values of corrosion current density (oscillating around 1 µA/cm 2 ) were observed in the initial phase of measurements (until day 18 of measurements). Then, corrosion current density stabilized at the level of 0.3-0.5 µA/cm 2 , which was a slightly better result than for black steel. The already mentioned quite a rapid increase in corrosion rate for galvanized steel in concrete without chloride content was related to zinc corrosion in highly alkaline concrete pore solution. At pH > 13.3 corrosion of galvanized coat on reinforcement was initiated.
Considering galvanized and black steel in concrete with chlorides, dynamics of changes in corrosion current densities and spread of those values during two-year measurements was very diversified (Figure 21b). During first 74 days of measurements, density of corrosion current for black steel did not exceed the level of 1.3 µA/cm 2 , whereas for galvanized steel values of icorr were higher, even up to 2 µA/cm 2 . After day 74, a further increase in corrosion current densities was observed for both galvanized and black steel. However, a wider spread of values was found for reinforcing steel As shown in Figure 21a, black reinforcing steel in concrete without additives (B) generally had stable but the low density of corrosion current through the whole period of measurements, which in accordance with criteria for evaluating corrosion risk to steel in concrete, could indicate the passivation. Microscopic tests (Figure 12b) demonstrated the presence of oxygen on the surface of steel rebar which probably indicated the presence of a passive layer of oxides that is characteristic for highly alkaline concrete pore solution (pH > 12.5).
In case of galvanized steel reinforcement in concrete without chlorides (G), elevated values of corrosion current density (oscillating around 1 µA/cm 2 ) were observed in the initial phase of measurements (until day 18 of measurements). Then, corrosion current density stabilized at the level of 0.3-0.5 µA/cm 2 , which was a slightly better result than for black steel. The already mentioned quite a rapid increase in corrosion rate for galvanized steel in concrete without chloride content was related to zinc corrosion in highly alkaline concrete pore solution. At pH > 13.3 corrosion of galvanized coat on reinforcement was initiated.
Considering galvanized and black steel in concrete with chlorides, dynamics of changes in corrosion current densities and spread of those values during two-year measurements was very diversified (Figure 21b). During first 74 days of measurements, density of corrosion current for black steel did not exceed the level of 1.3 µA/cm 2 , whereas for galvanized steel values of i corr were higher, even up to 2 µA/cm 2 . After day 74, a further increase in corrosion current densities was observed for both galvanized and black steel. However, a wider spread of values was found for reinforcing steel protected with the coating. Within that period, values of corrosion current density of the specimens with galvanized steel reached very high values of 3.8 µA/cm 2 . According to microscopic tests, zinc was found to be the only corrodible metal (Figure 20) on the undamaged structure of steel matrix which, considering the operational safety of reinforced concrete structures, is very important and positive information.
To sum it up, there was very strong corrosion of the reinforcement within the whole discussed period of two years of electrochemical testing according to criteria [51] for evaluating corrosion risk to reinforcing steel. Interestingly, after an 18-month interval in measurements determined densities of corrosion current dramatically decreased on day 792 of measurements. Corrosion of black steel was just inhibited as an insulating barrier from corrosion products of iron was formed. The microscopic tests ( Figure 13) showed that a layer of corrosion products was thick enough to hinder diffusion of oxygen required for corrosion development. Diffusion-controlled electrochemical processes were also confirmed by the Nyquist plot ( Figure 8a). It should be pointed out that the specimens were kept under rather stable thermal and humid conditions (temperature 20 ± 2 • C and relative humidity 50 ± 10%) during a very long, 18-month interval in measurements.
It is worth mentioning that despite an intensive development of corrosion on black steel in concrete with chlorides, no cracks were found on concrete surface of the test elements. It could be explained by a small diameter of rebars with reference to the thickness of the concrete cover. Thus, the volume of accumulating corrosion products did not exceed the available volume of concrete pores adjacent to rebars. Consequently, no critical values of tensile stress in concrete were observed.

Conclusions
The following conclusions can be drawn from two-year comparative tests on efficiency of the hot-dip galvanized zinc coating applied on reinforcing steel B500SP in concrete.

•
Comparative tests of changes in the density of corrosion current in the concrete specimens without chlorides basically demonstrated no development of corrosion, and possible passivation was expected in case of black steel. Higher densities of corrosion current were observed for galvanized steel during the first days of testing. The reason was the dissolution of zinc after the contact with initially high pH of concrete pore solution.

•
Six-month measurements demonstrated a higher density of corrosion current in the concrete specimens with high 3% concentration of chlorides, which unambiguously indicated corrosion in concrete reinforced with galvanized or black steel. Microscopic observations indicated zinc was the only metal corroding with the reinforcement made of galvanized steel, while the rebar surface showed no signs of corrosion.

•
Densities of corrosion current determined for selected specimens dramatically decreased after an 18-month interval in measurements. Corrosion was inhibited on black steel as an insulating barrier of corrosion products was formed. That result was also confirmed with microscopic observations. • The analysis of structural bonds in steel without or with zinc coating, containing concrete without or with chlorides demonstrated that the coating formed during hot dip galvanizing had varied thickness measured in the circumference at the cross-section of the rebar where the thickest layer of zinc was always observed at oblique ribs of the rebar.
• Additionally, the structural tests showed that the chemical composition at the section of the zinc coating was very homogeneous, and the layer of corrosion products indicated an increased content of oxygen, particularly for zinc coating in concrete with chlorides.

Conflicts of Interest:
The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.
Appendix A  Table A2. Comparison of results from analyzing impedance spectra of reinforcing steel in concrete, obtained for three selected measuring series on day 4, 74, and 792.