Journal of Molecular Liquids 315 (2020) 113787

Contents lists available at ScienceDirect

Journal of Molecular Liquids

journal homepage: www.elsevier.com/locate/molliq

Combined experimental and computational studies on corrosion

inhibition of Houttuynia cordata leaf extract for steel in HCl medium
Le Thanh Thanh a, Nguyen Si Hoai Vu b,c, Phan Minh Quoc Binh a, Vinh Ai Dao b,c, Vu Thi Hanh Thu d,e,
Pham Van Hien b,c,⁎, Casen Panaitescu f,⁎⁎, Nguyen Dang Nam b,c,⁎
a
PetroVietnam University, 762 Cach Mang Thang Tam Street, Long Toan Ward, Ba Ria City, Ba Ria, Vung Tau Province 790000, Viet Nam
b
Future Materials & Devices Lab., Institute of Fundamental and Applied Sciences, Duy Tan University, Ho Chi Minh City 700000, Viet Nam
c
The Faculty of Environmental and Chemical Engineering, Duy Tan University, Da Nang 550000, Viet Nam
d
VNUHCM-University of Science, 227 Nguyen Van Cu Street, District 5, Ho Chi Minh City 700000, Viet Nam
e
Vietnam National University, Ho Chi Minh City 700000, Viet Nam
f
Department Engineering of Petroleum Processing and Environmental Protection, Petroleum-Gas University of Ploiesti, 100680 Ploiesti, Romania

a r t i c l e i n f o a b s t r a c t

Article history: The development of corrosion inhibitors that could spontaneously adsorb immediately on the metal surfaces
Received 20 April 2020 after they are damaged by electrochemical reactions would remarkably arise the improvement of cost-saving,
Received in revised form 3 July 2020 durability and environmental impact of metals. The use of leaf materials for metal corrosion remains an impor-
Accepted 8 July 2020
tant challenge due to the high corrosion protection, availability as natural materials, environmental friendly
Available online 9 July 2020
and low-cost. Herein, we report a Houttuynia cordata leaf extract as a green corrosion inhibitor that acts as a
Keywords:
mixed type of inhibitor for steel in an aqueous 0.1 M hydrochloric acid solution. It also performs the high inhibi-
Carbon steel tion performance (up to 98.3% by adding 1500 ppm inhibitor), spontaneous uniform and smooth film formation
Corrosion of the steel surface and less corrosion attack when the H. cordata leaf extract is added to the aqueous hydrochloric
Hydrochloric acid acid solution. In contrast to the inhibited systems, uninhibited specimen is severely corroded at high corrosion
Green inhibitor rate, which is associated with a significant degradation of the entire steel surface. Additionally, investigation on
Electrochemical methods good adhesion and thermal stability of the protective film, as well as enriched electron abundance of selected in-
Molecular dynamics simulation hibitor species are performed via molecular dynamics simulation method. Furthermore, insights of corrosion, in-
hibition and spontaneous film formation mechanisms are also proposed and discussed in detail.
© 2020 Elsevier B.V. All rights reserved.

1. Introduction affordable prices of steel compared to other metals bring on variety of

applications. Therefore, improving usage lifespan of steel from the so-
Although hydrochloric acid (muriatic acid) is a highly corrosive called “corrosion” phenomenon has been a major challenging issue of
chemical, it is one of versatile reagents for large-scale chemical indus- science and industry.
tries (e.g., steel pickling, etching operations) and small-scale applica- The use of corrosion inhibitors (CIs) has been extensively adaptable
tions (pH control or neutralization) [1]. By interacting with H+ and for some decades as a preferred effective means for corrosion control on
Cl− dissociated from HCl in water, ferrous materials as mild steel tend metals. Most of today's commercial inorganic and organic inhibitors
to reach the more meta-stable state through spontaneity of chemical performed high corrosion resistivity, such as chromate, silane, and
and electrochemical reactions, causing the considerable metal loss. It their derivatives; however, their known hazard threats on both
eventually leads to premature structural failure of material and conse- human and animal living environments are still problematic and led
quent expenses of replacement cost as well as environmental concerns to restrictive regulations [5–7], opening a key research area to exploring
at large [2–4]. In addition, outstanding mechanical properties and effective alternatives from natural sources with low-priced, abundant
and biodegradable capability. The inhibition principle of synthetic or-
⁎ Corresponding authors at: Future Materials & Devices Lab., Institute of Fundamental ganic compounds was found to basically lie on benefits of π-
and Applied Sciences, Duy Tan University, Ho Chi Minh City 700000, Viet Nam and The conjugated systems, highly electronegative functional groups contain-
Faculty of Environmental and Chemical Engineering, Duy Tan University, Da Nang ing nitrogen, oxygen, sulfur and heterocycles in improving targeting-
550000, Viet Nam. adsorption of inhibitive matters onto the surficial metal to effectively
⁎⁎ Corresponding author.
E-mail addresses: [email protected] (P. Van Hien),
block or actively react with the corrosive species [8,9]. Owing to incred-
[email protected] (C. Panaitescu), [email protected] (N.D. Nam) ibly rich sources of naturally synthesized organic compounds with ac-
. ceptable eco-friendliness, various plant extracts have been selected to

https://doi.org/10.1016/j.molliq.2020.113787
0167-7322/© 2020 Elsevier B.V. All rights reserved.
2 L.T. Thanh et al. / Journal of Molecular Liquids 315 (2020) 113787

be new desirable corrosion inhibitors [10–13]. In addition, organic com- then filtered by the filter paper for producing the H. cordata leaf's glue.
plexes from plant extracts can be naturally biodegradable along with The glue was concentrated by a rotary evaporator to obtain the con-
their simple and safe extraction process. A wide range of extracted densed glue named H. cordata leaf extract (HCLE). It is very hard to sep-
parts of plants have been widely studied and demonstrated to serve as arate the HCLE's components because their retention times are much
good corrosion inhibitors for aluminum and steel such as leaves closed. Therefore, the ethanol extract of H. cordata leaf will be used for
[14–19], seeds [20–24], roots [25,26] to fruits [27–29]. The sum of inhi- our investigation. Main components of HCLE were characterized by
bition activity of phytocompounds from plant extract typically is mixed- gas chromatography - mass spectroscopy 78 (GC–MS) with Agilent
type [30]. It could be explained by flexible electron consumption and/or 7890B for GC and Agilent 5977A for MS at 1 ml/min of flow rate. The sys-
donation of multi-functional organic groups to the metal surface to sup- tem was equipped with a column 19091S-433UI (constituted by 5%
port both chemical and physical adsorption on the metal surface to ob- phenyl-methyl polysiloxane) which is useful to isolate not only non-
struct total anodic and cathodic corrosion reactions [31]. polar volatile but also high-polar non-volatile organic compounds of
Herbs, widely distributed under tropical climate condition, are es- Houttuynia cordata leaf extract (HCLE). In addition, HCLE was also ex-
sential spices in traditional Vietnamese cousins for flavoring and natu- amined by attenuated total reflectance Fourier transform infrared spec-
rally medicating, and are abundant sources of non-toxic organic troscopy (ATR-FTIR, Alpha-FTIR spectrometer with ±1 cm−1 of the
compounds. Fish-mint is a well-known herb and traditionally indige- wavenumber accuracy and using Opus 3.1 software). HCLE was added
nous medicine of Southeast Asia, including Vietnam, belonging to family to 0.1 M HCl solution to achieve 0, 100, 500, 1000 and 1500 ppm for in-
Saururaceae and scientifically named as Houttuynia cordata. It is a peren- hibition and surface characterizations.
nial plant growing all year round with green, heart-shaped leaf. Re- Working electrodes for electrochemical and coupon tests were fabri-
searches ongoing for several years have reported that H. cordata are cated from steel sheet (steel compositions were given in Table S1 of
plentiful of flavonoids, alkaloids, and other polyphenols that are respon- Supplementary information) to get 10 mm of length and width and
sible for diverse pharmacological actives, such as relieving fever, reduc- 3 mm of height. All working electrodes and coupons were mounted
ing swelling, and enhancing digestive and urinative functions; by a low viscosity epoxy system with 17:1 ratio of resin and hardener
interestingly, H. cordata has been used for HIV-1 treatment and SARS from EpoFix without shrinkage at room temperature for further control-
prevention [32,33]. H. cordata thus is a promising “green” substitute ling 100 mm2 exposed area. The exposed area was rinsed with deion-
against acid-induced corrosion for undesirable inhibitors. ized water and ethanol after grounding by 1200-grit silicon carbide
Nowadays, thanks to breakthroughs in computer science, theoretical papers. All experiments were carried out in a naturally-aerated 0.1 M
calculation has been utilized as one of the most fundamental methods HCl solution at room temperature. Electrochemical tests were per-
for evaluation of chemical and electrochemical phenomenon at molec- formed using a VSP system of BioLogic Scientific Instruments. Steel, sil-
ular scale [34]. In corrosion science, it not only further provides a ver/silver chloride, and titanium mesh were respectively used as the
more certain affirmation of experimental results but also extend com- working, reference, and counter electrodes for three electrode system
prehension beyond experimental assessment into energetic and elec- in the electrochemical cell, which was kept in a Faraday cage for further
tronic properties of corrosion protective systems as well as their electrochemical experiments. After 1 h immersion in solution for stabi-
interaction and protection work on the metal substrate in any aggres- lizing the open-circuit potential (OCP), electrochemical impedance
sive medium via multiple calculations of density functional theory spectroscopy (EIS) measurements were carried out every 1 h over a pe-
(DFT) and molecular dynamic (MD) [35,36]. riod of 24 h at peak-to-peak amplitude, 10 mV of the sinusoidal pertur-
Even though H. cordata possesses necessary advantages to be a po- bation signals, and frequency range from 100 kHz to 10 mHz. After
tential inhibitor alternative, there is no current research reporting the which, linear polarization resistance (LPR) was also investigated in a po-
use of this natural product for steel against corrosion in acidic media tential range of ±20 mVOCP with 166 × 10−3 mV/s of scan rate. Potenti-
by a combined study using experimental and theoretical approaches. odynamic, anodic, and cathodic polarization (PD, AP, and CP) tests were
Otherwise, our preliminary experiments found that H. cordata leaf was undertaken after 24-h immersion at OCP. PDs were run at
obtained with a short extraction time in ethanol solution and competi- 166 × 10−3 mV/s of scan rate from a potential of −250 mV (vs. EOCP)
tive solubility in HCl solution, compared to those of Hibiscus sabdariffa to −250 mVAg/AgCl, whereas APs and CPs were also conducted at
leaf extract (HCLE) and Aganonerion polymorphum leaf-water extract 166 × 10−3 mV/s of scan rate from OCP to −300 and −1200 mVAg/
(APLWE), which could related to its rich sources of polar compounds AgCl of the anodic and cathodic potentials, respectively.
to enhance the adsorption on the steel surface and in turn improve cor- A cube with 15 mm of length and width and 5 mm of height was fab-
rosion inhibition efficiency. This work reports the corrosion inhibition ricated from steel sheet and these specimens were hung by Teflon tape
ability of H. cordata leaf extract (HCLE) on mild steel in 0.1 M hydrochlo- via a 1 mm-diameter hole drilled across the specimen for weight loss
ric acid, using potentiodynamic polarization (PD), electrochemical im- (WL) measurements followed ASTM G1 and G31 [37,38]. All weight
pedance spectroscopy (EIS), linear polarization resistance (LPR) and loss specimens were finished by grounding with 1200-grit SiC paper,
gravimetric measurement. The surface morphology and inhibitor com- cleaned by ultrasonic cleaner with ethanol for 5 min and dried before
position were characterized using scanning electronic microscopy measuring the exposed area and 24 h storing in automatic desiccator.
(SEM), attenuated total reflectance Fourier transform infrared (ATR- The specimens were then weighed to get an initial weigh (mi) and
FTIR) spectroscopy, gas chromatography–mass spectrometry (GC–MS) hung in 0.1 M HCl solution without and with HCLE addition for 24 h. Fi-
and X-ray photoelectron spectroscopy (XPS). Theoretical calculation nally, weight loss specimens were cleaned in 1000 ml solution made by
was performed to confirm experimental results and by far discuss in- distilled water, 500 ml hydrochloric acid, and 3.5 g hexamethylene
sights into corrosion inhibition mechanism of HCLE through its adsorp- tetramine for 5 min. These specimens were then cleaned and stored as
tion activities on mild steel in acidic solution. the same above processes before weighing the final weigh (mf). The
weight loss (w) was calculated by the results of subtraction between
2. Experimental section mi and mf. To ensure reproducibility and accurate weigh, three mea-
surements were run for each test and an electronic analytical balance
Ethanol (C2H5OH), hexamethylenetetramine, and hydrochloric acid with four decimal scale was used for measuring the weights.
(HCl) were purchased from Sigma Aldrich for supporting all experimen- Scanning electron microscopy measured by a Supra 55 VP SEM at
tal processes without any treatments. H. cordata leaves were picked up 20 kV and 10 mm of the accelerating voltage (EHT) and working dis-
from the East of South Vietnam, carefully cleaned by distilled water, tance (WD) was used for examining the surface morphology after
dried in furnace at 60 °C, and ground to be powder. This powder was 24 h immersion in 0.1 M HCl solution without and with HCLE addition.
then added to C2H5OH Soxhlet system to be extracted at 75 °C and In addition, surface roughness of steel surfaces after 24 h immersion in
 L.T. Thanh et al. / Journal of Molecular Liquids 315 (2020) 113787 3

0.1 M HCl solution without and with HCLE addition were also character- chloride (Cl−) ion, and 556 H2O molecules (extended simple point
ized by atomic force microscopy (AFM, Bruker Multimode 8) at the av- charge SPC/E water model [43]), were created in dimensions
erage signal value. The surface film formed on steel after 24 h 24.8 × 24.8 × 37.9 Å3. The optimized inhibitor molecules were put on
immersion in 0.1 M HCl solution without and with HCLE addition was the constrained Fe(110) substrate in parallel configurations, then the
analyzed by X-ray photoelectron spectroscopy (XPS, Kratos Nova) and whole boxes were used for geometry optimization and dynamics simu-
attenuated total reflectance Fourier transform infrared spectroscopy lation for 0.5 ns using Forcite module [39] of the MS program. The
(ATR-FTIR) as shown above description. An Al Kα energy source at condensed–phase optimized molecular potentials for atomistic simula-
1486.6 eV was used in XPS measurements, while 160 eV of the pass en- tion studies (COMPASS) force field was applied to control the MD
ergy and 20 eV were set for survey and high-resolution scans. Further- simulation. The Ewald (accuracy of 10−5 kcal/mol) and atom–based
more, the hydrocarbon C 1s peak at 285.0 eV was used as a reference (cut–off distance 9.5 Å) summation methods were set for the electro-
for calibrating the binding energy. static and van der Waals interactions to integrate Newton's second
The corresponding chemical structures of HCLE substances were ob- law equations of motion by velocity Verlet method, respectively. The
tained from the open chemistry database PubChem for simulation pro- dynamics simulation was carried out at 298 and 333 K (25 and 60 °C)
cesses. Each substance was geometrically optimized using DMol3 by the use of canonical (NVT) ensemble, Andersen thermostat with a
module [39] in the Materials Studio (MS) 2017 with following configu- 1.0 ps timestep for a total 500,000 steps. The MD simulation at 25 and
rations: Becke-three-parameter-Lee-Yang- Parr (B3LYP) hybrid func- 60 °C is related to the experimental process. Specifically, 25 °C is the
tional was set to the maximum 10−5 Ha of energy, 2 × 10−3 Ha/Å of room temperature, while 60 °C is the extraction temperature. Energy
force, 10−6 Ha of scf tolerance, and 5 × 10−3 Ha of smearing. The DFT values of all boxes after dynamic equilibrium was exerted for calculation
semi-core pseudopotential were applied for core treatment with double of binding energy between inhibitor molecules and the steel substrate,
numeric polarization (DNP) basis set and 3.5-basis file. The Mulliken as followed Eqs. (6)–(8) [44]:
charge and Hirshfeld quadrupole were used for population analysis.
Eint ¼ Eall −ðEsurþsol þ Einhþsol Þ þ Esol ð6Þ
The frontier molecular orbitals and energies of an optimized structures
included: the Fukui function (the nucleophilic f(+) and electrophilic f
Ebind ¼ −Eint ð7Þ
(−)), the highest occupied/lowest unoccupied molecular orbitals
HOMO/LUMO, and the HOMO/LUMO energies (EHOMO and ELUMO). The
ΔEbind ¼ E333K 298K
bind −E bind ð8Þ
essential electronic characteristics (chemical bandgap (ΔE), chemical
hardness (η), electronegativity (χ), and the fraction of electron trans-
In Eqs. (6) and (7), Eint and Ebind are the interaction and binding en-
ferred (ΔN)) were computed using EHOMO/ELUMO and Eqs. (1)–(5) [40]:
ergies between inhibitor molecules and the steel substrate, respectively.
ΔE ¼ ELUMO −EHOMO ð1Þ Eall is the total energy of a box that mimics the substrate (sur) and inhib-
itor molecule (inh) in 0.1 M HCl solution (sol). Esur+sol, Einh+sol, and Esol
ELUMO −EHOMO stand for the total energy of the box contains surface and inhibitor, in-
η¼ ð2Þ
2 hibitor and 0.1 M HCl solution, and the 0.1 M HCl solution only, respec-
tively. In Eq. (8), ΔEbind represents the energy deviation between
−ðELUMO þ EHOMO Þ binding energy at different temperatures at which E298K 333K
bind and Ebind are
χ¼ ð3Þ
2 the binding energy at relevant temperatures.
χ −χinh
ΔN ¼
Fe  ð4Þ 3. Results and discussion
2 ηFe −ηinh

ΦFe −χinh The steel corrosion mechanism in acidic media has been reported as
ΔE ¼ ð5Þ a sum of both anodic iron dissolution and cathodic hydrogen evolution.
2ηinh
As a result, the coupon weight is lightened with the exposing time. The
rate of steel corrosion (mm/yr) were calculated by the result of weight
It is noted that there are two offered equations to calculate ΔN, using
loss (w) using Eq. (9) following [37]:
chemical hardness (ηFe = 0 eV) and electronegativity (χFe = 7.0 eV) of
iron (Eq. (4)) [41] or workfunction (ΦFe = 4.82 eV) of the Fe(110) plane k1  w
(Eq. (5)) [42]. Corrosion rate ðmm=yr Þ ¼ ð9Þ
Atd
The Fe(110) surface was used for representing the steel substrate,
imported from the MS library. Fe crystal was geometrically optimized where k1 is a constant (8.76 × 104), A is the total exposed area (cm2), t is
using generalized gradient approximations (GGA), Perdew–Wang–91 the immersion time (h), and d is the density (g/cm3). Table 1 shows the
functional for the exchange–correlation energies. The Broyden– gravimetric results of mild steel after 24-h immersion in 0.1 M HCl solu-
Fletcher–Goldfarb–Shanno (BFGS) algorithm were employed for the tion with the absence and presence of different HCLE concentrations at
optimization process with 5 × 10−6 eV/atom of energy tolerance, room temperature. It is observed that the loss of weight decreased with
10−2 eV/Å of max force, and 2 × 10−6 GPa of max stress. The on–the– increasing HCLE concentration. Specifically, the weight loss of steel cou-
fly generation (OTFG) ultrasoft pseudopotential, Koelling–Hammon rel- pon is HCLE concentration-dependency wherein w values 118.5, 13.7,
ativistic treatment with 10−6 eV of cut–off energy, 5 × 10−6 eV/atom of 6.5, 3.1 and only 2.1 mg were obtained after 24-h immersion of steel
scf tolerance, and 10 × 10 × 10 in the reciprocal lattice (total 35 points) in solution with 0, 100, 500, 1000 and 1500 ppm HCLE, respectively.
of non–offset k–point Brillouin zone sampling, were applied for energy The lowest relevant corrosion rate of mild steel in the range of inhibitor
calculation. The optimization process and energy calculation were exe- concentrations, decreasing from 7.35 to 0.13 mm/yr, was achieved by
cuted using the CASTEP module [39] in the MS suite. After the optimiza- presence of 1500 ppm green inhibitor in HCl solution. In addition, gas
tion process, the Fe crystal was cleaved to (110) surface with three bubbles were observed to generate continuously on the steel surface
layers in–depth, and a 10 × 10 supercell was built to the final Fe(110) during 24-h immersion in solution containing only hydrochloric acid,
substrate. which is contrary to significantly low degree of bubble evolution when
The MD simulation was performed to mimic the interaction be- the steel coupon was exposed to solution with high content of HCLE. It is
tween inhibitor molecules and steel substrate in 0.1 M HCl referred to decline of the cathodic reactions on the steel surface by the
environment. To do this, several simulation boxes, containing the Fe intervention of this green inhibitor. Therefore, first investigation
(110) surface, inhibitor molecules, one hydronium (H3O+) ion, one through gravimetric measurement demonstrated the reduction of steel
4 L.T. Thanh et al. / Journal of Molecular Liquids 315 (2020) 113787

Table 1 Table 2
Weight loss results (mi, initial weight; mf, final weight; w, weigh loss; and CR, corrosion Corrosion parameters (Ecorr, corrosion potential; icorr, corrosion current density; βa, anodic
rate) of steel after 24 h immersion in 0.1 M HCl solutions containing different HCLE con- Tafel Slope; βc, cathodic Tafel slope and CR, corrosion rate) obtained from potentiody-
centrations at room temperature. namic polarization curves of steel after 24 h immersion in 0.1 M HCl solution containing
different HCLE concentrations at room temperature.
Specimens mi (g) mf (g) w (mg) CR (mm/yr)
Specimens Ecorr icorr βa -βc CR Inhibition
0 ppm 17.1748 17.0570 117.8 7.30
(mVAg/AgCl) (μA/cm2) (mV) (mV) (mm/yr) efficiency (%)
17.2115 17.0927 118.8 7.37
17.2075 17.0885 119.0 7.38 0 ppm −485 630.2 207 191 7.42
Average 118.5 7.35 −483 615.1 227 202 7.24
100 ppm 17.1713 17.1572 14.1 0.87 −486 627.5 213 211 7.38
17.4118 17.3986 13.2 0.82 Average −485 624.3 7.35
17.1328 17.1190 13.8 0.86 100 ppm −521 75.7 155 150 0.89
Average 13.7 0.85 −522 71.2 165 153 0.84
500 ppm 17.5593 17.5528 6.5 0.40 −520 70.5 155 160 0.83
17.4325 17.4259 6.6 0.41 Average −521 72.5 0.85 88.4
17.5600 17.5537 6.3 0.39 500 ppm −523 62.5 157 147 0.37
Average 6.5 0.40 −524 61.1 154 173 0.36
1000 ppm 17.6113 17.6085 2.8 0.17 −523 60.3 152 165 0.35
17.6992 17.6962 3.0 0.19 Average −523 61.3 0.36 90.2
17.3975 17.3939 3.6 0.22 1000 ppm −516 26.5 139 129 0.16
Average 3.1 0.19 −517 28.0 138 148 0.16
1500 ppm 17.6747 17.6725 2.2 0.14 −516 29.5 140 135 0.17
17.5205 17.5188 1.7 0.11 Average −516 28.0 0.17 95.5
17.7005 17.6981 2.4 0.15 1500 ppm −517 11.9 137 128 0.14
Average 2.1 0.13 −516 10.4 127 115 0.12
−515 9.9 122 120 0.12
Average −516 10.7 0.13 98.3

dissolution and hydrogen evolution are results of the effective inhibition

of the HCLE extract on steel corrosion in HCl medium.
Fig. 1(a) illustrates potentiodynamic polarization of mild steel after Cl− to steel corrosion. It is apparent that current densities of both anodic
24-h immersion in 0.1 M HCl solution without and with various concen- and cathodic branches were strongly reduced in solution containing the
trations of HCLE at room temperature. Corrosion parameters (Ecorr, cor- HCLE inhibitor. This reveals that electrochemical corrosion reactions at
rosion potential; icorr, corrosion current density; βa, anodic Tafel slope; the metal/electrolyte interface, that caused electron exchange, was ef-
βc, cathodic Tafel slope and CR, corrosion rate) identified by Tafel ex- fectively controlled by the added HCLE inhibitor. Especially, the signifi-
trapolation method are given in Table 2. PD result denotes the extreme cant decrease of corrosion current densities as the increasing
corrosion of steel in HCl solution with both active anodic and cathodic introduction of HCLE content into corrodent suggested greater inhibi-
reactions, which is attributed acceleration of dissociated ions H+ and tion of higher HCLE content for steel in the acidic solution. The PD curves

Fig. 1. (a) Potentiodynamic, (b) anodic and (c) cathodic polarization curves of steel after 24 h immersion in 0.1 M HCl solution containing different HCLE concentrations at room
temperature. All polarization measurements were scanned at 0.166 mV/s.
 L.T. Thanh et al. / Journal of Molecular Liquids 315 (2020) 113787 5

of inhibited systems identified the cathodic inhibition effect of HCLE (Z′) - imaginary impedance (Z″)). The depressed shape of arcs is
given their corrosion potential become mildly negative in respect to accounted to a complex phenomenon of corrosion and inhibition as
that of the uninhibited system. Whereas, the anodic inhibition control well as arbitrary distribution of inhomogeneities on the steel surface,
was further proved owing to the appearance of pitting potential be- leading to the capacitance deviation from ideal to non-ideal state [46].
tween −0.4 and − 0.3 VAg/AgCl in inhibited systems (also as an indica- The Nyquist plots revealed that the increase of inhibitor concentration
tion of film formation). Thus, the HCLE extract should be classified as a and immersion time resulted in the enhancement of the surface imped-
mixed inhibitor to affect both anodic and cathodic kinetics. Abiding by ance. Significantly small semicircles were attained for sample in HCl so-
Faraday's law, the corrosion rate (mm/yr) was calculated using current lution without inhibitor while the diameter was extended by the rise of
density extracted from PD method [45]: inhibitor concentration in the acidic solution, which is about 20, 30, 60
and 70 times larger when 100, 500, 1000 and 1500 ppm of inhibitor
k2  icorr  E:W: were added, respectively. This implies that higher HCLE concentration
Corrosion rate ðmm=yrÞ ¼ ð10Þ
d in the HCl solution, greater corrosion protection for the steel surface.
As the function of the immersion time, continuous enlargement of sur-
where k2 is a constant (3.27 × 10−3), icorr is the corrosion current den-
face impedance in solution with 1000 or 1500 ppm of HCLE is docu-
sity (μA/cm2), E. W. is the equivalent weight (g) and d is the metal den-
mented, differentiate to its uncertain development in solution without
sity (g/cm3). Also, the corrosion inhibition efficiency of HCLE was
and with low concentration of 100 or 500 ppm inhibitor. It is an indica-
computed using the following equation:
tion for the reinforcement of the protective film with high inhibitive
  content to fully, actively prevent H+ and Cl− from migrating to the
i0corr −icorr
inhibition efficiency ð%Þ ¼  100 ð11Þ steel surface. Furthermore, the presence of two depressed semicircles
i0corr of steel immersed in 1500-ppm containing solution with the clear ob-
servation of a large semicircle at medium-low frequency area implies
where i0corr and icorr is the corrosion current density (μA/cm2) of speci- the great resistance for charge transfer process occurring at the film/
mens immersed in HCl solution without and with HCLE inhibitor, steel substrate interface was achieved by inhibitor addition.
respectively. As can be seen in the Bode plot as shown in Fig. S1 of Supplementary
Results in Table 2 of Supplementary information evidently portrays information, only one peak obtained in the uninhibited system could be
that the corrosion current density decreased as the rise of HCLE concen- assigned for charge transfer process rather than film formation in highly
tration from 0 to 1500 ppm and its significant decline from 624.3 to 10.7 corrosive HCl solution. The presence of two peaks at high-medium and
(μA/cm2) is documented for the presence of 1500 ppm HCLE in 0.1 M medium-low frequency in the inhibited system is evident to the forma-
HCl solution. The monotonic decrease in corrosion rate and increase in tion of inhibitive film and the high resistivity for charge transfer process
corrosion inhibition efficiency of mild steel with increasing concentra- at film/steel interface. Additionally, the increment of phase angle with
tion of HCLE in HCl solution accounted for the significant support of increasing concentration of HCLE and immersion time is associated
HCLE in cutting down electron exchange and in turn reducing deteriora- with a more capacitive film formation. According to EIS data, the equiv-
tion at active surface sites. Compared to our previous studies, the corro- alent circuit offered for EIS result of steel in 0.1 M HCl solution (Fig. 2(f
sion inhibition efficiency at 1500 ppm is in ascending order: APLWE [14] and g)) contains: solution resistance (Rs); constant phase element of the
b HSLE [46] b HCLE. The unstable shift of corrosion potential to be more double layer (CPEdl) placed in parallel with charge transfer resistance
negative value with the rising addition of HCLE extract is probably given (Rct). Electrical components of the protective film as constant phase el-
by the random adsorption of inhibitor matters as well as their interac- ement (CPEfilm) and resistance (Rfilm) were further added for steel in
tion with corrosion products for construction of the protective film on HCLE-containing 0.1 M HCl solution. Constant phase element was
the steel surface. used rather than pure capacitor to provide the compensation for non-
In order to investigate the independent effect of HCLE inhibitor on ideal dielectric behavior and chaotic motion of matters at the electro-
partial anodic and cathodic reactions and disclose the prominent corro- lyte/working electrode interface ZCPE = CPE−1(jω)−n; n corresponds
sion inhibition behavior behind, anodic and cathodic polarization scans to the phenomenological coefficient. The n value demonstrates the na-
were additionally carried out for steel immersed in solution without ture of constant phase element (as a capacitance if n = 1; a Warburg im-
and with 1500 ppm of HCLE concentration, as plotted in Fig. 1(b and pedance if n = 0.5; a resistance if n = 0; and an inductance if n = (−1). j
c). The current density of partial electrochemical reactions was deter- = (−1)−1/2 and ω are the imaginary number and angular frequency
mined by intersection point of corrosion potential and extrapolated (rad s−1), respectively. The ZSimpWin software was used for EIS fitting
Tafel line. The Tafel regime for anodic oxidation reactions (Fig. 1(b)) and the variation of corresponding impedance parameters were plotted
witnessed a considerable decrease in anodic current density and the as the function of immersion time in Fig. 3.
presence of passivation region by using 1500 ppm HCLE inhibitor, The transient of protective film and charge transfer resistances with
which is justified by the formation of protective film when the HCLE time of specimens in solution without and with different HCLE concen-
was added. At inhibitor concentration of 1500 ppm, the cathodic current trations were plotted in Fig. 3(a and b). The significantly low charge
density (Fig. 1(c)) is observed to reduce by nearly 1.5 orders of magni- transfer resistance is observed for the steel electrode exposing to HCl so-
tude at Ecorr, which kept at a lower degree over a wide range of potential lution without HCLE addition during 24 h immersion. It seems to be due
from approximately −1.2 to −0.4 V vs. VAg/AgCl in comparison with that to electron exchange from corrosion reactions and the transfer of H+
of the uninhibited system. It is thought to be due to the surface film and Cl− from HCl solution to electrolyte/electrode interface, resulting
formed by HCLE matters to strongly impact the partial reduction reac- in corrosion acceleration of the steel for the long exposure time. As men-
tions occurring at cathodic sites. Therefore, it is noted according to po- tioned, the addition of HCLE inhibitor is demonstrated to increase the
larization data that the inhibitive species can simultaneously behave corrosion resistance with immersion time, however, the inhibited sys-
inhibition performance for corrosion cells under electrochemical force tem containing 100 ppm HCLE show that resistances (Rfilm and Rct)
of aggressive ions, including its preferable control for cathodic reactions are observed to decrease gradually after 14-h immersion. This appears
and effective constrain for anodic dissolution reactions through an ac- to prove the formation of a defective film that might be mainly consti-
tive film formation on the metal surface. The results explained why tuted by rust product and low inhibitive content to insufficiently cover
less weight loss and bubble evolved were obtained in inhibited system. and protect the whole metal surface, thereby causing less inhibition
Fig. 2 depicts the impedance responds occurring on the steel/acid in- performance and then benefiting penetration of aggressive ions to dam-
terface by introduction of different HCLE concentrations for 24 h in age the surface film in the long run. On the other hand, the significant
0.1 M HCl solution, plotted in the typical Nyquist form (real impedance increase in Rfilm and Rct with immersion time is seen with the rise of
6 L.T. Thanh et al. / Journal of Molecular Liquids 315 (2020) 113787

Fig. 2. Impedance (Nyquist) plots of steels during 24 h immersion in 0.1 M HCl solution containing: (a) 0, (b) 100, (c) 500, (d) 1000 and (e) 1500 ppm HCLE. EIS measurements were
carried out every 1 h over a period of 24 h and selected times were added for easily presenting and identifying the results. Proposed equivalent circuit for fitting the impedance
spectra of steel in 0.1 M HCl solution: (f) without and (g) with HCLE additions.

HCLE inhibitor concentration in which the addition of 1000 and Morphology of the corroded surfaces after 24 h of immersion in
1500 ppm work effectively. This can be interpreted by the reinforced 0.1 M HCl solution was characterized by SEM and results are presented
formation of the protective film and the reduction of charged species in Fig. 5. The distinctive morphologies are evident as a result of corro-
(corrosive ions, electron) at the film/steel substrate interface, respec- sion damage as shown through Fig. 5(a–e). Fig. 5(a) presents a harsh
tively. The resultant total impedance (Rtotal = Rct + Rfilm) and corrosion corrosion on the steel surface through a disappearance of original mor-
rate extracted by EIS and LPR (Figs. S2 and S3 of Supporting informa- phology such as grounding signals. This phenomenon is consistent with
tion) additionally pointed out the influence of HCLE on enhancing cor- the soap bubbles obtained during immersion process of steel in 0.1 M
rosion resistance of the steel surface by the HCLE matter-containing HCl solution for both electrochemical and coupon tests. It is fact that
film formation, which is more protective with the rise of inhibitor the steel is an active material in acid solution (pH ~1) where iron ions
amount added and exposure time of the steel surface to inhibited sys- and oxides/hydroxides are enriched in solution and steel surface, re-
tems. The decline in CPEfilm and CPEdl of steel during immersion time spectively. Meanwhile, hydrogen evolution is presented at the cathodic
in solution consisting of over 500 ppm inhibitor in Fig. 3(c and d) con- regime. Iron oxides/hydroxides formed on steel surface could be clear in
firmed the increasing construction of the capacitive film constituted the XPS measurement. The phenomenon of the soap bubbles was still
by multiple layers of absorbed species to effectively retard paths for mit- performed when steel immersed in 0.1 M HCl solution containing
igation of corrosion-induced ions from solution to the steel substrate. 100 ppm HCLE, indicating a significant corrosion in this steel surface.
High capacitances, rising with immersion time, are obtained for systems However, the steel surface immersed in 0.1 M HCl solution containing
with absence and presence of only 100 ppm HCLE, it is said to be owing 100 ppm HCLE in Fig. 5(b) shows a less corrosion attack compared to
to the concentration of conductive ions on the metal surface and/or the that of the surface immersed in 0.1 M HCl solution in Fig. 5(a), which
appearance of defective film formed on the metal surface that can be de- is corresponded with less soap bubbles during immersion process. Re-
graded after long time immersion. The correlative results of corrosion sults of insignificant corrosion attack were displayed when HCLE con-
rate after 24 h immersion in 0.1 M HCl solution among different mea- centration was increased in 0.1 M HCl solution given the appearance
surements PD, EIS, LPR and WL in Fig. 4 reconfirmed the profound effect of apparent grounding scratches on the SEM images as shown in Fig. 5
of HCLE inhibitor for steel in HCl solution, resulting in reduction of cor- (c–e). Aside from the evident scratches on the steel surface, there ap-
rosion rate with increasing introduction of HCLE wherein high concen- pears a thin film that significantly limited damage on the surface,
tration of over 500 ppm HCLE used is of a significantly low level of resulting in a more uniform morphology with increasing HCLE concen-
corrosion-induced damage. tration in solution. To appreciate the effect of HCLE on the steel surface
 L.T. Thanh et al. / Journal of Molecular Liquids 315 (2020) 113787 7

Fig. 3. Influence of HCLE on (a) protective film and (b) charge transfer resistances, (c) protective film and (d) double layer CPE magnitudes of steel in 0.1 M HCl solution.

roughness, an AFM study was undertaken after 24-h immersion in solu- surface roughness could be interpreted as being a result of HCLE's mol-
tion with different HCLE additions ranging from 500 to 1500 ppm and ecule adsorption on the steel surface to electrochemical corrosion pro-
the results are presented in Fig. 6. It is noticed that Ra (average rough- cess. Furthermore, an increase in HCLE concentration improved the
ness) was used instead of rms (mean roughness) due to a significant ef- corrosion mitigation and film formation, resulting in more uniformity
fect of peaks and valleys in an image. The results show that the obtained of the surface and hindering corrosion attack, as well as a raise in corro-
Ra values are 8.5, 5.2 and 2.5 nm for steels in 0.1 M HCl solution contain- sion resistance. These absorbed components were determined by ATR-
ing 500, 1000 and 1500 ppm, respectively. The low roughness magni- FTIR and XPS measurements and their absorption mechanism is able
tude obtained by immersion in high inhibitor-containing solution to be clear via MD technique.
could be related to self-assembly of organic molecules of HCLE on the Fig. S4(a) of Supplementary information displays the GC–MS spectra
steel surface [47]. Grounding scratches clearly observed in Fig. 5 of the dried HCLE extract that consists of many peaks, indicating the
(c) express the greatest corrosion protection of HCLE on the steel sur- presence of many compounds in the HCLE extract. However, there are
face in 1500 ppm-containing HCl solution. A significant difference of 14 major peaks including low polar compounds at the early of retention
times, low quantities and disintegrated compounds with small peaks in
the spectra, and others related to high polar compounds. The low polar
components in extracted products could be mainly counted for the si-
tosterol and lupenol, et al. [48]. The mainly natural compounds in
HCLE are observed as alkaloids, strigolactones and phenolic compounds.
Particularly, phenolic compounds containing nitrogen atom include nic-
otinamide, asimilobine, N-methyllaurotetanine, indole-3-carboxylic
acid and N-nornuciferine are identified according to the commercial
GC spectra library (NIST 2.0). These compounds are also consistent
with literatures [49–54] and chosen for theoretical methods of molecu-
lar dynamics simulations in this study in order to compare the detailed
electronic and molecular understandings concerning the adsorption
and active sites of inhibitors, as well as their effects on inhibition perfor-
mance. Furthermore, HCLE was also characterized by ATR-FTIR and the
result was presented in Fig. S4(b) of Supplementary information. The
carbonyl group and N\\H deformation mode could be seen around
1602–1649 cm−1 band while C_N and C_C in aromatic groups can
be attributed to the bands at 1508 cm−1. The N\\H bending vibrations
can be assigned around 1253–1400 cm−1 bands. The strong band at
1028 cm−1 can be accounted to the C\\C bending vibration [48]. Addi-
Fig. 4. Comparison of corrosion rate among PD, EIS, LPR and WL measurements of steel
after 24 h immersion in 0.1 M HCl solution with different inhibitor concentrations.
tionally, a series of peaks between 667 and 866 cm−1 is accorded with
Magnitudes of corrosion rate obtained from WL and other methods (PD, EIS and LPR) the stretching vibrations of aromatic C\\H and the ≡C-H in terminal al-
were calculated using Eqs. (9) and (10), respectively. kynes. Therefore, GC–MS and ATR-FTIR denoted that HCLE contains a
8 L.T. Thanh et al. / Journal of Molecular Liquids 315 (2020) 113787

Fig. 5. SEM images of steel surface after 24 h immersion in the 0.1 M HCl solution containing: (a) 0, (b) 100, (c) 500, (d) 1000 and (e) 1500 ppm of HCLE.

mixture of more than alkaloids, strigolactones and phenolic com- also report that the peaks of C 1s and O 1s in the inhibited specimen
pounds. Fig. S4(c) shows ATR spectra of the surface film after 24 h im- are obviously stronger than that of the uninhibited specimen. The Fe
mersion in solution added different HSLE concentrations. The results 2p peaks are evidently obtained on the uninhibited specimen, whereas,
indicated that the same absorption bands are found on all steel surfaces very small Fe 2p peaks are shown in the inhibited system, manifesting
and whose intensity increased with an increase in HSLE concentrations. less metal dissolution when steel was immersed in 0.1 M HCl solution
It is observed that the bands around 1720 cm−1 are corresponded to containing 1500 ppm HSLE. This phenomenon obviously performed
carbonyl groups. C_N and C_C in aromatic groups could be accorded on the narrow scan of Fe 2p in Fig. 7(b), that presented six and four
with the peaks in bands from 1400 to 1500 cm−1. N\\H bending vibra- peaks on the spectra of steel surface immersion in 0.1 M HCl solution
tion and C\\C bending vibration could be observed at the bands around without and with 1500 ppm HSLE addition. These peaks can be assigned
1000–1300 cm−1 while the emergence of a sharp peak at 729 cm−1 to ferrous and the ferric compounds, satellite and metallic Fe around
might be related to aromatic C\\H and the ≡C-H in terminal alkynes. 710, 711, 712, 723, 724 and 726 eV [55]. It could be a result of iron dis-
There is a significant rise in intensity of these absorption peaks when solution to form the oxide and hydroxide products that were enriched
an increase in HSLE concentration was added to 0.1 M HCl solution, on the surface products for uninhibited specimen and lessened on the
evoking improved formation of HSLE species on the metal surface. Par- surface products for inhibited specimen by an inhibition for metal disso-
ticularly, a deviation of these peaks' position with that of HSLE powder lution. Fig. 7(c) pointed out an evident carbon contamination that could
ensured that the HSLE species formed on the surfaces as results of a be assigned for Fe(CO)5 peak at 288.8 eV, satellite peak at 287.1 eV, CO
prominent chemisorption process rather than physical accumulation and C peaks at 285.8 and 284.6 eV for steel surface immersed in unin-
of HSLE onto the steel surfaces. It is referred that the protective film hibited system. The strongly observed peaks of carbon contamination
on the steel surface is mainly contributed by corporation of components are detected on steel surface immersed in inhibited system, containing
from the HCLE extract to iron oxide/hydroxide after initial corrosion of a combination C-C=O, O-C=O, C_O, C_N of both single and double
steel. Furthermore, the increase of the absorption peak's intensity with bonds at 288.6 eV, C\\C single bonds at 287.1 eV, C\\O and C\\N single
increasing of HSLE concentration pinpointed a dependence of inhibition bonds at 286.0 eV and other carbon bonds including C, C\\C, C_C at
mechanism on HSLE concentrations. around 284.6 eV [56,57]. A slight shift of carbon peaks between uninhib-
To ensure the HSLE absorption on the steel surface, the adsorbent ited and inhibited system suggested an evident carbon contamination
surfaces after 24 h immersion in 0.1 M HCl solution without and with from inhibitor species formed on steel surface immersed in inhibited
1500 ppm HSLE addition were determined by XPS. Fig. 7(a) clearly system. In addition, nitrogen peaks are significantly observed on the
shows that the Fe 2p, C 1s and O 1s are observed on all specimen sur- steel surface immersed in inhibited system and not performed for unin-
faces, while new N element is observed on the steel surface immersed hibited system as described in Fig. 7(d). It indicated that N 1 s peaks can
in 0.1 M HCl solution containing 1500 ppm HSLE. Furthermore, results be assigned C\\N, N\\O and C_N bonds around 399.9 eV. These bonds

Fig. 6. AFM images of steel surface after 24 h immersion in the 0.1 M HCl solution containing: (a) 500, (b) 1000 and (c) 1500 ppm of HCLE.
 L.T. Thanh et al. / Journal of Molecular Liquids 315 (2020) 113787 9

Fig. 7. (a) Survey and narrow scan X-ray photoelectron spectroscopy spectra of (b) Fe, (c) C, (d) N, (e) P and (f) O existences on steel surfaces after 24 h immersion in solution containing 0
and 1500 ppm HCLE.
10 L.T. Thanh et al. / Journal of Molecular Liquids 315 (2020) 113787

could be ascribed to the nitrogen from the organic nitrogen groups. The oxygen) out of a total of 15 atoms. Additionally, NI has three double
above observations are consistent with oxygen peaks shown in Fig. 7(f). bonds and one amide functional group that enhanced the negatively
As listed in Table S4 of Supplementary information, the steel surface im- charge effect in the bonds and atoms. As a result, the number of nucle-
mersed in the uninhibited system displays oxygen from iron oxides and ophilic sites overwhelmed electrophilic sites.
hydroxides at four peaks around 533.5 (cyan) + 532.2 (blue), 530.8 The f(−) and f(+) sites have a close relationship with the HOMO
(green) and 529.3 eV (red), attributed to absorbed water in the hy- and LUMO regions. While the f(−)/f(+) sites of an inhibitor molecule
drated iron oxide/hydroxide [58,59], iron hydroxide (Fe(OH)2/Fe(OH) point out its binding position to the steel surface, the HOMO/LUMO re-
3/FeO(OH)), and iron oxide (Fe2O3/Fe3O4), respectively. Whereas, only gions show how easy can electrons transfer between them. Since the
three peaks at 530.7, 532.0 and 533.5 eV are observed on steel surface anodic areas on the steel surface after initial corrosion has positively
immersed in the inhibited system. The strong peak at 532.0 eV can be charged, inhibitor molecules can afterward adsorb on through f(+)
assigned for oxygen of iron hydroxide and C_O double bond. This phe- sites on their structures. Then HOMO orbitals donate electrons to attach
nomenon can be referred to the hydrated iron hydroxide that was ini- to the substrate and block anodic areas from attacks of aggressive spe-
tially formed and anchored by inhibitor species on the top of this layer cies. Fig. 9 portrays the HOMO and LUMO regions of main organic com-
to form the absorbed inhibitive layer reducing the oxide formation pounds derived from the HCLE extract. The HOMO/LUMO regions of all
from corrosion reactions, resulting in the reduction of the peak at substances are seen to allocate over hydrocarbon chains rather than
529.3 eV in the inhibited system. Other oxygen peaks are also ascribed functional groups. Theoretically speaking, compared to π bonds from
to N\\O and C\\O bindings of organic compounds. It is noted that hydrocarbon chains, functional groups can establish deeper HOMO/
these peaks were shifted and stronger than that observed on uninhib- higher LUMO regions that do not contribute to the electron donation/
ited system, ensuring the absorbed products from inhibitor species. Fur- acceptance to the steel surface. Besides the investigation on distribution
thermore, small peaks of P are also observed on the inhibited system, of the HOMO/LUMO regions on inhibitor structures, their energy calcu-
recommending that a small amount of the organic phosphoric groups lation is also further theoretical approach to gain better understanding
was also contributed to the protective film formed on steel surface. for corrosion inhibition performance of the multi-component inhibitor
XPS results indicated that a strong decrease in Fe peaks occurred, (HCLE extract). Quantum chemical parameters of main component de-
exhibiting a less dissolution of steel immersed in inhibited system. Fur- rived from the HCLE extract, consisting of ΔE, η, χ and ΔN, were calcu-
thermore, the significant increase and shift of C and O peaks, new N and lated using Eqs. (1) to (5) and presented in Table S3 of Supplementary
P peaks could be the result of inhibitor molecules' adsorption. By a com- information.
bined assessment of XPS, SEM and AFM results, it is apparent that steel The more effective inhibitors, based on quantum chemical aspect,
surface was completely covered by the adsorbed HCLE species at high must donate and would be more comfortable to give their electrons to
inhibitor concentration to form uniform and smooth surface film. The the steel substrate, which are identified by more positive ΔN and higher
surface analysis results are thus fully consistent with above electro- (less negative) EHOMO. Accordingly, Table S3 clearly indicated that inhib-
chemical results. itors are arranged in descending order of corrosion inhibition perfor-
GC–MS data revealed five main substances of HCLE as nitrogen- mance as ML N NN N AS N ID N NI. Besides, ML had the lowest
containing phenolic compounds, including asimilobine (AS), nicotin- chemical bandgap and hardness (ΔN and η), which demonstrated this
amide (NI), indole-3-carboxylic acid (ID), N-methyllaurotetanine (ML) inhibitor as the most stable and soft compound (following the hard
and N-nornuciferine (NN). Fukui functions f(+) and f(−) as well as and soft acids and bases (HSAB) theory). Therefore, ML inhibitor is
the energy regions HOMO and LUMO with related chemical structures found to be the good inhibition compound of the HCLE extract by its
are shown in Figs. 8 and 9, relatively. According to Fig. 8, electron easiness to donate electrons to the steel substrate. In contrast, NI with
cloud in the sigma bond was attracted to nitrogen and oxygen atoms, the simplest chemical structure displays highest value of ΔE and η, lead-
thereby making these atomics negatively charged (-I inductive effect) ing to its difficulty to donate electrons. It can be concluded that NI is not
and thus creating a nucleophilic f(+) site. Besides, functional groups a good inhibition part of the HCLE extract. By comparison with our last
such as =C=O, –C(=O)OH, –OH and –C(=O)NR2 also had a strong at- study [11], the fraction of electron transferred (ΔN) of HCLE's com-
traction to the electron clouds of the hydrocarbon branch, which inten- pounds is higher than that of A. polymorphum leaf-water extract
sively polarize bonding sites to be nucleophilic sites. These sites tended (APLWE)'s compounds. By using Eq. (4), except for NI, other main inhib-
to share the same plane to the molecular plane, except the case of NI itors of the HCLE extract show ΔN from 0.69 to 0.89 eV, compared to a
molecule. Such nucleophilic sites can attract positive ions H3O+ in the lower range 0.618 to 0.693 eV of main inhibitors of the APLWE extract.
corrosive environment and Fen+ from the solution/steel substrate inter- Thus, it can be judged that HCLE extract is likely to be better than inhi-
face through electrostatic force. It explained why the decline of cathodic bition effect than APLWE in the same aggressive HCl solution. The calcu-
hydrogen evolution and the formation of an organic-Fen+ film on the lated ΔN using Eq. (5) had smaller values but has a good agreement
steel surface happened, thereby leading to the reduction of corrosion with the previous one, which is also added for the later comparison.
consequence. In contrast, nitrogen and oxygen atoms were also electron On the ground of quantum chemical theory, main components of the
attractive (electrophilic f(−)) sites due to their high electronegativity. HCLE extract are potential inhibitors at which point the positive ΔN
Moreover, the conjugate effect made the sigma bonds (C\\C single tells the convenient transfer of electrons from main inhibitor molecules
bonds) on the benzene ring lose their electrons to become positive of the HCLE extract to the steel substrate, leading to the reduction of cor-
charged areas, turning them to be electrophilic sites. Interestingly, the rosion phenomenon. The quantum chemical results offer parallel
effect is supported by the electron attraction from functional groups, adsorbed configurations of the inhibitor molecules on steel substrate
resulting in the formation of prominent electrophilic sites at the sigma via distributions of f(+), f(−), HOMO/LUMO orbitals. It was found
bond nearby functional groups. It is observed that the electrophilic that inhibitors of HCLE extract can attach to the metal substrate at posi-
sites, in this case, distributed along the perpendicular direction to the tions containing highly electronegative atoms (N and O) as well as bind-
molecular plane, excluding that of NI molecule. As a result, electrophilic ing sites of hydrocarbon branches with functional groups. These
sites can pull Cl− ions out of the steel surface by the electrostatic attrac- configurations were also applied for the following MD calculation,
tion force, thus reducing the possibility of pitting corrosion caused by which aim to give visual look on the interaction between main compo-
such type of ions. Besides, electron-concentrated areas are able to nents of the HCLE extract and HCl solution consisting of water, dissoci-
have p-d interaction with vacant 3d orbitals of Fe. The reason why NI ated ions H3O+ and Cl− through the influence of temperature and time.
had anomalous properties might originate from its chemical structure. The Fe(110) plane was chosen to represent the steel substrate due to
NI is a smallest molecule in comparison to other substances in HCLE, its primary orientation in Fe crystal. Dimensions of simulation boxes
which has three highly electronegativity atoms (two nitrogen and one were optimized to ensure that the density of 0.1 M HCl solution is
 L.T. Thanh et al. / Journal of Molecular Liquids 315 (2020) 113787 11

Fig. 8. The quantum chemical simulation for nucleophilic and electrophilic Fukui sites of HCLE's substances. The red and blue colors in chemical structure part represent oxygen and
nitrogen atoms, respectively.

1.0 g/cm3. The COMPASS force field was applied to provide the more ac- adsorption state on the steel substrate with time but tended to detach
curate fit for the steel substrate and inhibitor molecules in a corrosive from the substrate after heating, which can be explained by its easy con-
environment. The applied temperatures 298 and 333 K (25 and 60 °C) nection and thermal instability to the steel substrate, respectively. NI
is corresponded to the room temperature and drying temperature of molecule is seen to slowly absorb on the steel surface; however, it
H. cordata leaf before extraction, respectively. Side-view images of moved towards the acidic solution after high system temperature was
each simulation box were captured after simulation, including two applied. The NN molecule is far away from the steel substrate in all
snapshots at 100 ps and 500 ps at 298 K and one snapshot at 500 ps three simulation conditions, also indicating the low corrosion perfor-
and 333 K, as displayed in Fig. 10. Increases in setting time and setting mance of this HCLE extract's component.
temperature can express changes in inhibitor/the steel substrate inter- Quantitative binding energy (Ebind) of main substances of the HCLE
actions with time and their thermal withstand, respectively. Fig. 10 extract to the steel substrate at 298 and 333 K are listed in Table 3 of
show that positions of AS and ML on the steel substrate remain un- Supplementary information. Based on Ebind, the order of inhibition per-
changeable with time and temperature. The results indicated that formance at 298 K is ML N AS N ID ≈ NI ≈ NN. Setting temperatures on a
these molecules quickly attached to the steel substrate with high bind- big system to test the thermal endurance of subsystems is one of the
ing energy, which is associated with their superior corrosion efficiency most significant advantages of MD simulation. At 333 K, the order of in-
for steel in HCl solution. The similar events are not observed on other hibition performance changed following the clear order ML N AS N NI N ID
main substances of the HCLE extract. The ID molecule kept their similar N NN. The negligible Ebind decrease of ML as well as the interesting Ebind
12 L.T. Thanh et al. / Journal of Molecular Liquids 315 (2020) 113787

Fig. 9. The quantum mechanics simulation for HOMO and LUMO regions of HCLE's substances. The red and blue colors in chemical structure part represent Oxygen and Nitrogen atoms,
respectively.

increase of AS molecule for steel in HCl solution with temperature rise AS molecule showed endothermic adsorption and others presented
confirmed their high thermal withstand, thereby suggesting them as exothermic adsorption when they were contacted to the steel surface
potential inhibitors for steel. The low Ebind of ID and NN which signifi- in HCl solution. It can be concluded that adsorption of the HCLE extract
cant decreased after higher temperature (333 K) was added, especially is the prominent exothermic process. The MD simulation results pro-
the negative value −0.7 eV achieved from the simulation result of the vided the clear support for the great corrosion protection of the HCLE
NN molecule, highlights less corrosion performance of these HCLE sub- extract for steel in HCl solution. ML and AS were found as high active,
stances. According to previous literatures [60], inhibitors with Ebind protective corrosion inhibition components of the HCLE extract, which
greater than 40 kJ/mol (0.415 eV) are classified to follow chemisorption can be ascribed to the antioxidant property of methoxy–phenolic
on the steel surface whereas the lower Ebind shows a kind of groups. Such an antioxidant phenolic compound can salvage free
physisorption. It is observed that all the HCLE's substances are chemi- dislocated ions such as H3O+ and Cl− in HCl medium by donating/con-
sorption at 298 K while only ID and NN turned to physisorption at suming their electrons, or chelating with Fen+ cations to form a more
333 K. The binding energy deviation (ΔEbind) decides whether the ad- stable film containing inhibitors [61]. Although other main components
sorption process of an inhibitor on the steel surface is endothermic of the HCLE extract such as ID, NI and NN might not be effective inhibi-
(positive ΔEbind) or exothermic (negative ΔEbind). Besides, the higher tors, they are expected to cooperate with ML and AS as well as minor or-
ΔEbind (more positive and less negative), the stronger the adhesion of ganic substances of the HCLE extract to co-form a high resistive film,
the molecule to the substrate (Table 3). The data manifested that only which can protect the steel surface from acid-induced corrosion).
 L.T. Thanh et al. / Journal of Molecular Liquids 315 (2020) 113787 13

Fig. 10. The side view of molecular dynamics simulation for HCLE's substances after 100 ps at 298 K and 500 ps at 298 and 333 K. The red and blue colors in chemical structure represent
Oxygen and Nitrogen atoms, respectively.

According to the combined results of electrochemical methods, sur- is exposed to acidic environment of HCl, resulting in a damaged surface
face analysis techniques and the state-of-the-art approach via simula- and continuous gas evolution, as following typical reactions:
tion tools, the film formation mechanism of H. cordata leaf extract for
−
corrosion protection of steel in hydrochloric acid solution is proposed Anodic reaction : Fe→Fenþ
ads þ ne ð12Þ
as shown in Fig. 11. Iron electrochemically oxidizes when carbon steel
Cathodic reaction : 2Hþ þ 2e− →2Hads →H2 ↑ ð13Þ

The decrease of hydrogen ions leads to the increase of local pH, lead-
Table 3 ing to the formation of corrosion rust, abiding by the simple transforma-
Binding energy from molecular dynamics calculation of substances from HCLE extract at
298 and 333 K.
tion process as follows [62]:

Compound AS ID NI ML NN
Fe2þ
ads →FeOOH→Fe3 O4 →Fe2 O3 ð14Þ
At 298 K 2.0 1.0 0.9 3.8 0.8 In addition, because the exposure of the steel surface to HCl solution
At 333 K 2.4 0.3 0.6 3.6 −0.7 cause positive charge (Fen+), small radius Cl− ions with negative charge
ΔEbind (eV) 0.4 −0.7 −0.3 −0.2 −1.5
can easily diffuses inwardly and absorb on the metal substrate where
14 L.T. Thanh et al. / Journal of Molecular Liquids 315 (2020) 113787

Fig. 11. Schematic figure of the film formation mechanism of Houttuynia cordata leaf extract's species for protecting steel in 0.1 M HCl solution: (a) without inhibitor, (b) with low inhibitor
concentration and (c) with high inhibitor concentration.

they can work as catalysts for the formation of intermediate corrosion with functional groups tend to link to positive ions H3O+ to reduce their
products, leading to acceleration of the anodic corrosion reaction of attacks to the steel surface and also act as a bridge to Fen+ ions from the
steel as described [19]: solution/steel substrate interface for the formation of the protective
film. Organic compound-Fe complexes might be further formed by in-
− −
Fe þ H2 O þ Cl ↔½FeClOHads þ Hþ þ e− ð15Þ teraction between p electrons of organic compound and free-electron
orbitals 3d of Fen+. Organic functional group-metal linkages might
−
½FeClOHads ↔½FeClOH þ e− ð16Þ also be explained by chelate effect of organic ligands to metal ions
[63,64]. In addition, it is possible that other organic compounds from
− the HCLE extract might somehow concur with the main inhibition com-
½FeClOH þ Hþ → Fe2þ þ Cl þ H2 O ð17Þ
ponents (AS and ML) in forming the Fe-organic complexes via interac-
tion between these compounds. Secondly, functional groups
H. cordata leaf extract was found to have diverse organic structures
containing highly electronegative atoms N and O and sigma bonds
containing multiple bonds and functional groups which was expected
nearby them became electron-consuming sites to catch negative ions
to bear similar inhibition mechanism as reported synthetic organic
Cl−, which could help the steel surface prevent localized corrosion ac-
compounds. After 24 h immersion of steel in the 1500 ppm inhibited
celeration. The interaction reduction of active ions H3O+ and Cl− to
system, polarization data unraveled the mixed inhibition behavior
the steel substrate and presence of the protective film formation also re-
with preferable cathodic control for this inhibitive species, this means
strict the charge transfer process taking place at the solution/steel sub-
both the transformation of hydron to hydrogen and the steel dissolution
strate interface. The reactions for adsorption of HCLE inhibitor on the
were effectively prevented. Analyses of chemical compositions con-
steel surface are thus suggested:
firmed the presence of organic functional groups on the metal surface
after immersed in inhibited systems in which peak shifts of organic con- h i2−α−β
−
stituents in ATR-FTIR and XPS spectra disclosed the prominent chemi- Fe2þ þ αH2 O þ βCl þ γðHCLEÞ↔ FeðHCLEÞγ ðOHÞα ðClÞβ ↓
sorption activity of HCLE. The quantum chemical calculation indicated þ αHþ ð18Þ
that key positions established by effects of highly electronegative C, O,
N, P and π-conjugated bonds as well as heterocycles on the organic It is inferred that the film formation of precipitated [Fe
frameworks play different roles in donating or consuming electrons (HCLE)γ(OH)α(Cl)β]2-α-β species on the surface of steel is responsible
when the inhibitor come to interact with the steel substrate and acidic for the achieved effective corrosion inhibition. A great agreement of im-
solution. By MD simulation, AS and ML were identified as two main pedance response characterized by EIS and LPR proved the highly resis-
components of the HCLE extract that possess great corrosion inhibition tive film formation in inhibited solution containing from 100 to
performance wherein their methoxy–phenolic groups were considered 1500 ppm HCLE. However, the low HCLE concentration of 100 ppm re-
to be primarily responsible for steel anti-corrosion in HCl solution. The sulted in reduction of film and charge transfer resistance after long time
corrosion protection of HCLE on the steel surface can be presented in immersion, implying degradation of the protective film or increasing
two ways. Firstly, nitrogen and oxygen from heterocyclic compounds existence of charged species (electrons, Cl− and H+) at the film/steel
and functional groups as well as binding sites of hydrocarbon branches substrate interface, respectively. It is worth commenting that the
 L.T. Thanh et al. / Journal of Molecular Liquids 315 (2020) 113787 15

formed flimsy and porous layer of low inhibitive content is probably a Declaration of competing interest
reservoir of “weak” points that become channels for accession and con-
tinuous attacks of H+ and Clˉ, accordingly leading to collapse of film The authors declare that they have no known competing financial
structure over exposing time in solution (Fig. 11). The rough surface interests or personal relationships that could have appeared to influ-
with random distributions of soap bubbles and surrounding cracks ence the work reported in this paper.
was seen in SEM result (Fig. 5(b)) of steel immersed in 0.1 M HCl solu-
tion containing 100 ppm HCLE. Acknowledgment
High concentration of HCLE (500 to 1500 ppm) is related to effective
government of corrosion inhibition where low corrosion rate, less This work is funded by PetroVietnam University under grantcode
weight loss and higher corrosion resistance were achieved. Enhance- GV1909.
ment of corrosion resistance as time and concentration dependencies
and the rise in IR peak intensity of organic functional groups is an indi- Appendix A. Supplementary data
cation of increasing accumulation of highly inhibitive content to form
the film with high densification and compactness. The inhibition Supplementary data to this article can be found online at https://doi.
model for highly inhibited system in Fig. 11 describes the action of the org/10.1016/j.molliq.2020.113787.
adsorbed film to act as a barrier layer to either anchor and productively
prevent corrosion-induced ions H+ and Cl−. Obtained corrosion protec- References
tion of HCLE at 1500 ppm was found to be better than those of APLWE
[1] N.N. Greenwood, A. Earnshaw, Chemistry of the Elements, Elsevier, Amsterdam,
and HCLE extract [14,46]. Netherlands, 2012.
A concern about the findings of HCLE inhibitor is that the used [2] N.D. Nam, P.T.N. Ha, H.T. Anh, N.T. Hoai, P.V. Hien, J. Saudi Chem. Soc. 23 (2019)
amount of HCLE for steel corrosion in 0.1 M HCl should not be lower 30–42.
[3] V.S. Sastri, Challenges in Corrosion: Costs, Causes, Consequences, and Control, John
than 500 ppm while higher concentration than 1500 ppm can reduce
Wiley & Sons, Hoboken, New Jersey, USA, 2015.
the solubility of this inhibitor. Therefore, there is the need for purifying [4] A. Groysman, Corrosion for Everybody, Springer Science & Business Media, Berlin/
effective inhibitor components from HCLE extract on purposes of reduc- Heidelberg, Germany, 2009.
ing inhibitor concentration and enhancing corrosion inhibition [5] O. Gharbi, S. Thomas, C. Smith, N. Birbilis, npj Mater. Degrad 2 (2018) 1–8.
[6] S.E. Manahan, Toxicological Chemistry and Biochemistry, CRC Press, Boca Raton,
performance. Florida, USA, 2002.
[7] M. Forsyth, B. Hinton, Rare Earth-based Corrosion Inhibitors, Woodhead Publishing,
4. Conclusions Cambridge, UK, 2014.
[8] S. Bashir, H. Lgaz, I.-M. Chung, A. Kumar, Chem. Pap. 73 (2019) 2255–2264.
[9] F. Branzoi, A. Băran, A. Ludmila, E. Alexandrescu, Chem. Pap. (2020)https://doi.org/
The work demonstrated the H. cordata leaf extract as a potentially 10.1007/s11696-020-01242-x.
green corrosion inhibitor for steel in an aqueous hydrochloric acid solu- [10] N.S.H. Vu, P.V. Hien, T.V. Man, V.T.H. Thu, M.D. Tri, N.D. Nam, Materials 11 (2018) 59.
tion 0.1 M with high inhibition performance that can combine impor- [11] S.H. Zaferani, M. Sharifi, D. Zaarei, M.R. Shishesaz, J. Environ. Chem. Eng. 1 (2013)
652–657.
tant properties of spontaneous film formation such as higher film and [12] P.B. Raja, M.G. Sethuraman, Mater. Lett. 62 (2008) 113–116.
charge transfer resistances, lower film and double layer CPE magnitudes [13] N. Soltani, N. Tavakkoli, A. Attaran, B. Karimi, M. Khayatkashani, Chem. Pap. 74
and potential components of the protective film formed on the steel sur- (2020) 1799–1815.
[14] P.V. Hien, N.S.H. Vu, L.X. Bach, N.Q. Tran, V.A. Dao, Q.T. Trinh, N.D. Nam, New J. Chem.
face when HCLE concentration was increased in the aqueous hydrochlo-
43 (2019) 15646–15658.
ric acid solution 0.1 M. Furthermore, a number of surface analysis [15] Q. Wang, B. Tan, H. Bao, Y. Xie, Y. Mou, P. Li, D. Chen, Y. Shi, X. Li, W. Yang,
parameters including surface morphology and roughness and protec- Bioelectrochemistry 128 (2019) 49–55.
tive film components systematically ensured the formation of protec- [16] M. Mobin, M. Basik, Y. El Aoufir, J. Mol. Liq. 286 (2019), 110890.
[17] N. Asadi, M. Ramezanzadeh, G. Bahlakeh, B. Ramezanzadeh, J. Chin. Inst. Chem. Eng.
tive film on the steel surface and reaching a basic understanding of 95 (2019) 252–272.
this corrosion inhibitor system. Additionally, we also demonstrated [18] X. Zheng, M. Gong, Q. Li, L. Guo, Sci. Rep. 8 (2018) 9140.
HCLE acted as a mixed corrosion inhibitor which should be generally [19] N.S.H. Vu, P.V. Hien, M. Mathesh, V.T.H. Thu, N.D. Nam, ACS Omega 4 (2019)
146–158.
suitable for preventing a wide range of the aggressive HCl corrosion sys-
[20] H. Hassannejad, A. Nouri, J. Mol. Liq. 254 (2018) 377–382.
tems. DFT and MD simulations suggested the important role of highly [21] G. Bahlakeh, A. Dehghani, B. Ramezanzadeh, M. Ramezanzadeh, J. Mol. Liq. 293
electronegative atoms, multiple bonds, heterocycles and functional (2019), 111559.
groups in forming electron-donating or-consuming sites, thereby [22] L.L. Liao, S. Mo, H.Q. Luo, N.B. Li, J. Colloid Interface Sci. 499 (2017) 110–119.
[23] P.E. Alvarez, M.V. Fiori-Bimbi, A. Neske, S.A. Brandán, C.A. Gervasi, J. Ind. Eng. Chem.
explaining the effect of HCLE to reduce steel corrosion by catching disso- 58 (2018) 92–99.
ciated H+, Clˉ ions and cooperating with Fe+ to form a protective com- [24] A.H. Al-Moubaraki, A.A. Al-Howiti, M.M. Al-Dailami, E.A. Al-Ghamdi, J. Environ.
plex film. ML and AS were found as two main HCLE's substances that Chem. Eng. 5 (2017) 4194–4205.
[25] R. Haldhar, D. Prasad, A. Saxena, P. Singh, Mater. Chem. Front. 2 (2018) 1225–1237.
bear great corrosion inhibition, high adhesion and thermal withstand
[26] H.S. Gadow, M.M. Motawea, RSC Adv. 7 (2017) 24576–24588.
to the steel surface. Finally, our well-defined design and inhibitor selec- [27] L.L. Liao, S. Mo, H.Q. Luo, N.B. Li, J. Colloid Interface Sci. 520 (2018) 41–49.
tion offer the simple procedures for characterizing the green corrosion [28] A. Dehghani, G. Bahlakeh, B. Ramezanzadeh, M. Ramezanzadeh, J. Mol. Liq. 279
inhibitors' properties in the laboratorial practices. The underway study (2019) 603–624.
[29] Z. Sanaei, M. Ramezanzadeh, G. Bahlakeh, B. Ramezanzadeh, Ind. Eng. Chem. 69
will be focused on the refinement of this extractive glue for reducing in- (2019) 18–31.
hibitor concentration without degradation of this observed high inhibi- [30] S.A. Umoren, M.M. Solomon, I.B. Obot, R.K. Suleiman, J. Ind. Eng. Chem. 76 (2019)
tion performance. 91–115.
[31] M. Goyal, S. Kumar, I. Bahadur, C. Verma, E.E. Ebenso, J. Mol. Liq. 256 (2018)
565–573.
CRediT authorship contribution statement [32] D. Cheng, L. Sun, S. Zou, J. Chen, H. Mao, Y. Zhang, N. Liao, R. Zhang, Molecules 24
(2019) 1835.
Le Thanh Thanh: Methodology, Investigation, Funding acquisi- [33] K. Shingnaisui, T. Dey, P. Manna, J. Kalita, J. Ethnopharmacol. 220 (2018) 35–43.
[34] C.D. Taylor, P. Marcus, Molecular Modeling of Corrosion Processes: Scientific Devel-
tion. Nguyen Si Hoai Vu: Formal analysis. Phan Minh Quoc Binh: opment and Engineering Applications, John Wiley & Sons, Hoboken, New Jersey,
Formal analysis, Supervision. Dao Vinh Ai: Formal analysis. Vu Thi USA, 2015.
Hanh Thu: Formal analysis. Pham Van Hien: Methodology, Investi- [35] A. Thomas, P.R. Ammal, A. Joseph, Chem. Pap. 74 (2020) 3025–3037.
[36] E.A. Yaqo, R.A. Anaee, M.H. Abdulmajeed, I.H.R. Tomi, M.M. Kadhim, Chem. Pap. 74
gation, Funding acquisition. Casen Panaitescu: Supervision. Nguyen
(2020) 1739–1757.
Dang Nam: Writing - review & editing, Conceptualization, [37] ASTM G1-03, Standard Practice for Preparing, Cleaning, and Evaluating Corrosion
Supervision. Test Specimens, ASTM International, West Conshohocken, PA, USA, 2011.
16 L.T. Thanh et al. / Journal of Molecular Liquids 315 (2020) 113787

[38] ASTM G31-72, Standard Practice for Laboratory Immersion Corrosion Testing of [52] K. Nemoto, S. Tanaka, M. Konno, S. Onozawa, M. Chiba, Y. Tanaka, Y. Sasaki, R.
Metals, ASTM International, West Conshohocken, PA, USA, 2004. Okubo, T. Hattori, Tetrahedron 72 (2016) 734–745.
[39] Material Studio 17.1.0.48, Dassault Systèmes, San Diego, USA2017. [53] A.F. Rossini, A.C.A. Muraca, G.A. Casagrande, C. Raminelli, J. Org. Chem. 80 (2015)
[40] M.S. Masoud, M.K. Awad, M.A. Shaker, M.M.T. El-Tahawy, Corros. Sci. 52 (2010) 10033–10040.
2387–2396. [54] H. Guinaudeau, M. Leboeuf, A. Cavé, J. Nat. Prod. 42 (1979) 325–360.
[41] V.S. Sastri, J.R. Perumareddi, Corrosion 53 (1997) 617–622. [55] R. Paul, C. Sarkar, Y. Yan, Q.T. Trinh, B.S. Rao, C.W. Pao, J.F. Lee, W. Liu, J. Mondal,
[42] A. Kiejna, T. Ossowski, Encyclopedia of Interfacial Chemistry: Surface Science and ChemCatChem (2020) https://doi.org/10.1002/cctc.202000072.
Electrochemistry, Elsevier, Oxford, 2018. [56] Q.T. Trinh, K.F. Tan, A. Borgna, M. Saeys, J. Phys. Chem. C 117 (2013) 1684–1691.
[43] H.J.C. Berendsen, J.R. Grigera, T.P. Straatsma, J. Phys. Chem. 91 (1987) 6269–6271. [57] C. Sarkar, S. Pendem, A. Shrotri, D.Q. Dao, P. Pham Thi Mai, T. Nguyen Ngoc, D.R.
[44] A.O. de la Roza, G.A. DiLabio, Non-covalent Interactions in Quantum Chemistry and Chandaka, T.V. Rao, Q.T. Trinh, M.P. Sherburne, ACS Appl. Mater. Interfaces 11
Physics: Theory and Applications, Elsevier, 2017. (2019) 11722–11735.
[45] ASTM G102-89, Standard Practice for Calculation of Corrosion Rates and Related In-
[58] Q.T. Trinh, K. Bhola, P.N. Amaniampong, F.o. Jérôme, S.H. Mushrif, J. Phys. Chem. C
formation From Electrochemical Measurements, ASTM International, West
122 (2018) 22397–22406.
Conshohocken, PA, USA, 2010.
[59] P.N. Amaniampong, Q.T. Trinh, K. De Oliveira Vigier, D.Q. Dao, N.H. Tran, Y. Wang,
[46] N.T. Hoai, P.V. Hien, N.S.H. Vu, D.L. Son, T.V. Man, M.D. Tri, N.D. Nam, Chem. Pap. 73
M.P. Sherburne, F.o. Jérôme, J. Am. Chem. Soc. 141 (2019) 14772–14779.
(2019) 909–925.
[47] H. Asakawa, S. Matsui, Q.T. Trinh, H. Hirao, Y. Inokuma, T. Ogoshi, S. Tanaka, K. [60] J.M. Winterbottom, M. King, Reactor Design for Chemical Engineers, CRC Press, New
Komatsu, A. Ohta, T. Asakawa, J. Phys. Chem. C 124 (2020) 7760–7767. York, USA, 1999.
[48] A.K. Satapathy, G. Gunasekaran, S.C. Sahoo, K. Amit, P.V. Rodrigues, Corros. Sci. 51 [61] M. Herrero, M. Castro-Puyana, E. Ibáñez, A. Cifuentes, Liquid Chromatography,
(2009) 2848–2856. Elsevier, 2017 359–380.
[49] A. Matsuoka, T. Isogawa, Y. Morioka, B.R. Knappett, A.E. Wheatley, S. Saito, H. Naka, [62] J. Alcántara, B. Chico, J. Simancas, I. Díaz, M. Morcillo, Materials 10 (2017) 406.
RSC Adv. 5 (2015) 12152–12160. [63] R. Singuru, Q.T. Trinh, B. Banerjee, B. Govinda Rao, L. Bai, A. Bhaumik, B.M. Reddy, H.
[50] J. Liu, C.-J. Li, L. Ni, J.-Z. Yang, L. Li, C.-x. Zang, X.-Q. Bao, D. Zhang, D.-M. Zhang, RSC Hirao, J. Mondal, ACS Omega 1 (2016) 1121–1138.
Adv. 5 (2015) 80553–80560. [64] M. Nihei, H. Ida, T. Nibe, A.M.P. Moeljadi, Q.T. Trinh, H. Hirao, M. Ishizaki, M.
[51] C.P. Ndi, M.J. Sykes, D.J. Claudie, R.A. McKinnon, S.J. Semple, B.S. Simpson, Aust. J. Kurihara, T. Shiga, H. Oshio, J. Am. Chem. Soc. 140 (2018) 17753–17759.
Chem. 69 (2016) 145–151.