Photo-Electrochemical Application of ZnOG Thin Film for in situ Monitoring of Steel Sour Corrosion

Further to traditional corrosion monitoring techniques for rated deteriorations, nowadays modern electrochemical monitoring methods are promising for the control of non-rated damage mechanisms. Considering carbon steel as the most commonly used alloy in the oil and gas industry, there are special grades under NACE MR0175 standard which are immune to sour corrosion. However, according to the industry reports, their immunity can be terminated by upset conditions or on site repairs. This issue will impose either a high operational risk or exorbitant maintenance and inspection costs. Hence, in this paper, a new monitoring technology framework is discussed to lessen a catastrophic failure risk for carbon steel under wet H2S corrosion. In this regard, the application of a developed hybridized ZnO-graphene micro-electrode (ZnOG) with a mix band gap of 1.17 eV is studied. Under hydrogen sulfide attack and when ZnOG hybrids are excited by UV illumination, their photo-electrochemical responses are analyzed. ZnOG hybrids emanate informative impedance signals in a response to the formation of ZnO(1 – x)Sx nano-crystals.


INTRODUCTION
One of the biggest challenges in corrosion management systems is when stress and corrosion incorporate in a damage mechanism [1,2]. Considering sour corrosion of steel in wet H 2 S environment, specific standards are issued, addressing material selection criteria in the construction phase [3]. However, during the commissioning phase, a material compliance to design standards may be terminated due to the maintenance, service upset, or bad operation. Calculation of a damage rate and its probability is known to be a pitfall to accurately quantify an associated risk [4,5]. Furthermore, inspections of such damage mechanisms are exorbitant or fairly effective. In view of that, appendix F of standard DNV-G101 states: "the nontime-dependent (non-rated) mechanisms are not considered suitable for direct control by inspection, but may, for example, require monitoring certain process parameters" [6].
In the modern electronic age, many of the process parameters such as temperature, pH, and flow are controlled by automated feedback controllers that have been possible only after the introduction of the associated reliable sensors. Likewise, for corrosion monitoring, many methods are developed [7]; but to the best of our knowledge regarding utilization of semiconductor science in sour corrosion monitoring, yet no method is reported.
Depending on a sour gas concentration, co-corrodants, pH, ambient temperature, and steel composition, different sulphide scales are formed on steel [8]. The most common scale is Pyrrhotite (F (1 -x) S); that uniform scale suppress both further corrosion and an induced hydrogen flux [9]. Otherwise, localized scales amplify sour corrosion and further hydrogen flux due to the FeS semiconducting assistance in electrochemical hydrogen formation [10]. In recent years, numerous papers have been published on ZnO-graphene (ZnOG) nano-structures [11][12][13] with different applications such as solar cells [14], sensors [15][16][17], supercapacitance [18], optoelectronic devices [19], and environmental issues [20], to name but a few. Besides, a new concept of a sensing mechanism in atomically thin transition metal dichalcogenides to detect molecules based on dark excitons was presented elsewhere [21]. However, in this paper, the ZnOG photo-electrochemical (PEC) response is discussed for monitoring the application.

MATERIALS AND METHODS
A novel ZnOG hybrid semiconductor micro-electrode was developed via supercritical isopropanol treatment, dip-coated on a carbon steel (St32) substrate. All preparation procedures and material properties investigations are reported in another article by the authors [22]. However, in this report ZnOG morphology is shown by means of a MIRA3 TESCAN Field Emission Scanning Electron Microscope (FE-SEM) and analyzed simultaneously using Energy Dispersive Spectroscopy (EDS).
In addition, PEC tests were conducted with the test media selected as 3.5 wt % NaCl and a wet H 2 S electrolyte composed of 2.1 g Na 2 S · 9H 2 O in 100 mL of distilled water, which resulted in H 2 S concentration of 4600 ppm. A three-electrode cell with a specimen electrode, a platinum counter electrode, and a saturated calomel electrode (SCE) was used in the tests according to ASTM G5/G59. PEC tests were conducted in both the normal mode (dark) and an excited state of ZnOG via UV illumination by an OSRAM ULTRAMED 400W/FDA metal halide, which emits 8W in the UVB range (280 ~ 315 nm) and 88 W in the UVA range (315 ~ 400 nm).

RESULTS AND DISCUSSION
ZnOG composition and structure are reported in our previous article [22], proving a hybrid structure between zinc and graphene oxide. Moreover, a band gap and a luminescence trans-band energy study in another report of ours revealed a mix band gap of 1.17 eV with three trapping levels as an n-type semiconductor [23]. Nevertheless, ZnOG micrograph was depicted by FE-SEM and analyzed by EDS in Fig. 1. Figure 1a depicts the composition of a twin ZnOG structure by EDS; which proves the presence of carbon (graphene) in the ZnO structure. Figures 1b, 1c show thin film morphology and ZnOG button shape micrograph, respectively.
Subsequently, ZnOG electron-hole entanglement with the electrochemical redox reactions (Photo-electrochemical property) are studied. Figure 2 depicts the polarization behavior of ZnOG thin film on steel St32) substrate, in the normal mode (dark, Fig. 2a) and an excited state (UV illumination, Fig. 2b).   Besides, the polarization data are listed in Table 1. Both Fig. 2 and Table 1 demonstrate that steel St32 has almost the same corrosion potential (20 mV variance) in both electrolytes, but its corrosion rate in wet H 2 S is 44%, which is lower than its value in 3.5% NaCl, due to the formation of FeS protective layer [8]. Also the interface resistance (R p ) of the sample in wet H 2 S electrolyte is five times higher than that in 3.5% NaCl.
ZnOG in dark condition (Fig. 2a) has a nobler potential and a much lower redox current than steel St32, which is postulated as ZnO reactivity being much lower than that of iron and even zinc metal. Meanwhile, ZnOG has a higher corrosion rate in wet H 2 S compared to that in 3.5% NaCl electrolyte. This is caused by the formation of a non-uniform corrosion product of non-stoichiometry ZnS [24]. The nonuniformity of ZnS is attributed to the presence of graphene oxide flakes in the coating.
PEC potentiodynamic tests of excited ZnOG by UV illumination (Fig. 2b) indicate that excitation changes the ZnOG potential in the cathodic direction (more negative) in both electrolytes. The potential shift of ZnOG in wet H 2 S is seven times higher than that in 3.5% NaCl, while its PEC current is half. Furthermore, since FeS as a corrosion product have semiconductor properties, PEC study of steel St32 was also carried out in a wet H 2 S electrolyte under UV illumination, however, an imperceptible polarization alteration was observed, and the linear sweep voltammetry (LSV) diagram was almost the same as the one in the dark.
Electrochemical impedance spectra (EIS) in Nyquist diagrams (Fig. 3a) imply that, in a wet H 2 S electrolyte, there are higher impedances at the working electrode-liquid interface. For the steel St. 32 electrode, as mentioned before, this is due to the formation of a protective FeS film in the absence of oxygen. Likewise, a higher impedance of ZnOG in wet H 2 S, as already stated, is attributed to the formation of non-uniform non-stoichiometry ZnS.
PEC impedance of ZnOG in the Bode diagram (Fig. 3b) shows a lower impedance (|Z|) for excited

Carbon steel in wet H2S
Carbon steel in 3.5% NaCl ZnOG thin film in 3.5% NaCl

ZnOG thin film in wet H2S
Carbon steel in wet H2S ZnOG in frequencies lower than 1000 Hz. The Nyquist diagram shows a diffusing Warburg behavior for excited ZnOG, and at higher frequencies the imaginary part of impedance (Z'') is increased [25]. This is in agreement with a higher phase difference at low frequencies in the Bode diagram. At low frequencies, near DC, the phase difference can be a result of the capacitive or a reactive element. Pertinent equivalent circuits are depicted in Fig. 3c, by the Zview software version 3.5d.
ZnOG in a 3.5% NaCl electrolyte results in the same equivalent circuits both in normal and excited states, while excited ZnOG divulges one more constant phase element 3 (CPE3) in a wet H 2 S electrolyte in contrary to 3.5% NaCl. Besides, this element is attributed to the formation of non-stoichiometry ZnS in the form of activated ZnO (1 -x) S (x) semiconductor charge distribution [26]. The morphology of ZnO (1 -x) S (x) nano-crystals is revealed by FE-SEM after electrochemical tests, as illustrated in the background of  Fig. 1c it can be mentioned that the ZnOG disks are smooth prior to electrochemical tests but after them the corroded ZnOG micrographs, inspected by FE-SEM, show nano-sized crystals of ZnO (1 -x) S (x) on the ZnOG button.
Finally, the intercept of a tangential line of the linear section in the Mott-Schottky plot gives a flat band potential of -750 and -1420 mV in a wet H 2 S electrolyte for ZnOG in dark and under UV illumination, respectively (Fig. 4a). Accordingly, a positive slope in the linear section indicates ZnOG as a prevailing ntype semiconductor in dark; and a negative slope for excited ZnOG indicates a p-type behavior.
In conclusion, a band gap arrangement vs SCE in Fig. 4b shows that ZnOG Fermi level (electrolyte redox potential) shifts in the cathodic direction by UV illumination. An increment of ZnOG electronegativity is caused by the formation of active ZnO (1 -x) S (x) nano-crystals. ZnS has a more negative conduction  CONCLUSIONS Monitoring of a sour corrosion may be possible by monitoring the sulphur corrosion products formation. In this regard, a sensor mechanism was introduced based on a ZnOG hybrid semiconductor response to corroding environment, in the form of ZnO (1 -x) S (x) nano crystallization, where x depends on the H 2 S concentration, pH, oxygen content, and the temperature of the electrolyte. A sensing parameter can be defined as potential shift, in case of H 2 S presence in the electrolyte, or impedance alteration (CPE3) to monitor the sulphide formation rate. Although, the density functional theory calculations in [26] show that the band-gap energy of ZnO (1 -x) S (x) is not a linear function of the S content, it can be stated here that by choosing a proper frequency range, the S content can be monitored. However, further investigations are required to model the mechanism by using the quantum physics science. Also, experimental results should be normalized for different sour corrosion media, even regarding various H 2 S/CO 2 ratios.