Corrosion Science 53 (2011) 23852392

Contents lists available at ScienceDirect

Corrosion Science
journal homepage: www.elsevier.com/locate/corsci

Inhibitory action of aqueous coffee ground extracts on the corrosion

of carbon steel in HCl solution
Vanessa Vasconcelos Torres a, Roberto Salgado Amado a, Camila Faia de S a, Tatiana Lopez Fernandez a,
Carlos Alberto da Silva Riehl b, Alexandre Guedes Torres c, Eliane DElia a,
a
Departamento de Qumica Inorgnica, Instituto de Qumica, UFRJ, Avenida Athos da Silveira Ramos 149, Centro de Tecnologia, Bloco A Laboratrio 634A, CEP 21941-909,
Cidade Universitria, Rio de Janeiro, RJ, Brazil
b
Departamento de Qumica Analtica, Instituto de Qumica, UFRJ, Avenida Athos da Silveira Ramos 149, Centro de Tecnologia, Bloco A Sala 507, CEP 21941-909,
Cidade Universitria, Rio de Janeiro, RJ, Brazil
c
Laboratrio de Bioqumica Nutricional e de Alimentos, Instituto de Qumica, UFRJ, Avenida Athos da Silveira Ramos 149, Centro de Tecnologia, Bloco A Sala 528A, CEP 21941-909,
Cidade Universitria, Rio de Janeiro, RJ, Brazil

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

Article history: The effects of aqueous extracts of spent coffee grounds on the corrosion of carbon steel in a 1 mol L1 HCl
Received 21 September 2010 were examined. Two methods of extraction were studied: decoction and infusion. The inhibition ef-
Accepted 19 March 2011 ciency of C-steel in 1 mol L1 HCl increased as the extract concentration and temperature increased.
Available online 25 March 2011
The coffee extracts acted as a mixed-type inhibitor with predominant cathodic effectiveness. In this
study, the adsorption process of components of spent coffee grounds extracts obeyed the Langmuir
Keywords: adsorption isotherm. The chlorogenic acids isolated do not seem to explain the corrosion inhibition
A. Steel
observed during the use of the coffee extracts.
B. EIS
B. Polarisation
2011 Elsevier Ltd. All rights reserved.
B. Weight loss

1. Introduction
Carbon steel is one of the most important alloys in existence
and has a wide variety of industrial applications [1]. Acidic solu-
tions are also commonly used throughout various industries, and
applications include acid pickling, industrial cleaning, acid rescal-
ing, oil-well acidication, and petrochemical processes. Because
of the widespread applications of both carbon steel and acidic solu-
tions, researchers have devoted attention to developing more-
effective, non-toxic corrosion inhibitors to reduce the destructive
effects of acids on metal surfaces [2]. Most inhibitors used in indus-
try are organic compounds primarily composed of nitrogen, oxy-
gen and sulphur atoms. Inhibitors containing multiple bonds play
an important role in facilitating the adsorption of these compounds
onto metal surfaces [3]. A bond can be formed between the elec-
tron pair and/or the p electron cloud of the donor atoms and the
metal surface, thereby reducing corrosive attack in an acidic med-
ium. Although there are many organic compounds with corrosion
inhibiting properties described in the literature, the range of
appropriate inhibitors for a particular system is very limited due
to the specicity of the inhibitors and the great variety of systems
in which corrosion can occur. Although many of these compounds
have high inhibition efciencies, many are undesirable due to their
toxicity in humans, dangerous effects on the environment and high
prices [1].
In recent years, research into the use of low-cost and eco-
friendly compounds as corrosion inhibitors for mild steel has
intensied [1,426]. Plant extracts are generally inexpensive and
can be obtained through simple extraction processes. In addition,
the development of technological uses for food residues may
reduce the volume of waste discarded and increase the economic
viability of appropriate waste management options.
The objective of this study was to investigate the inhibitory ef-
fects of aqueous coffee extracts as corrosion inhibitors for carbon
steel in 1 mol L1 hydrochloric acid. The extracts were obtained
from spent coffee grounds using two methods, decoction and infu-
sion, to test the usability of the coffee residue. Open-circuit
potential measurements, potentiodynamic polarisation curves,
electrochemical impedance measurements and weight loss mea-
surements were utilised.
2. Experimental
2.1. Inhibitor preparation

Corresponding author. Tel.: + 55 21 25627813; fax: + 55 21 25627559. Aqueous extracts of spent coffee grounds were obtained using
E-mail address: eliane@iq.ufrj.br (E. DElia). two methods: infusion and decoction. The type of coffee used in

0010-938X/$ - see front matter 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.corsci.2011.03.021
2386 V.V. Torres et al. / Corrosion Science 53 (2011) 23852392
our investigation is called Pilo and is popular in Brazil. This com-
mercial type of coffee is a blend of Coffea arabic and Coffea
canephora, both of which are produced in So Paulo and in the Min-
as Gerais regions. In this study, commercial coffee (30 g) was used
for conventional extraction in an electric coffee maker. A commer-
cial paper lter and 300 mL of distilled water were used to prepare
a coffee beverage in the typical fashion. The residue retained in the
paper lter was dried in an oven at 50 C for 24 h. After drying, the
material was weighed (21.774 g) and divided into two 10 g
aliquots for the following extraction procedures: (a) a 10-min
decoction procedure in 100 mL of distilled boiling water and (b)
a 30-min infusion procedure in 100 mL of distilled water with an
initial temperature of 100 C. Both extracts were ltered, lyophi-
lised and stored at 4 C until analysis. The lyophilised extracts
of the spent coffee grounds were used as corrosion inhibitors for
carbon steel in 1 mol L1 hydrochloric acid.
2.2. Phenolic components of the coffee extracts
The phenolic antioxidants of the coffee extracts were analysed
using two methods: the colorimetric determination of total phen-
olics via reaction with the FolinCiocalteau reagent and an HPLC
analysis of the chlorogenic acids. The analysis of the total phenolics
in the aqueous coffee extracts was conducted essentially as
described elsewhere [27]. Freezedried samples were extracted
using acetone:water:acetic acid (70:29.5:0.5; v/v/v) and orbital
shaking. Samples were reacted with the FolinCiocalteau reagent,
the level of absorbance was read at 760 nm in a DU Beckman
Spectophotometer (Beckman; CA, USA) and contents of the total
phenolics were expressed as mg of gallic acid equivalents per gram
of freezedried extract (mg GAE/g). Chlorogenic acids constitute
the major phenolic components of roasted coffee and therefore
were analysed as representative of the phenolic components of
the samples. Chlorogenic acids were extracted from the freeze
dried samples as previously described [28]. The extracted samples
were analysed using HPLC in a Shimadzu liquid chromatograph
consisting of an LC-10AD vp quaternary pump and a SPD-M10A
vp diode array detector. Five microliters of each sample were in-
jected through a Rheodyne valve into a Kromasil C18 HPLC column
(150  4.6 mm, 5 lm, Eka Chemicals, NY, USA) and eluted using a
solvent gradient. The HPLC solvent gradient, peak assignment, and
quantitative analysis were used as previously described [29]. The
results obtained are expressed in mg of chlorogenic acids per g of
freezedried sample.
2.3. Electrochemical procedure
Working electrodes were prepared from steel specimens with
the following composition (wt.%): C: 0.18, P: 0.04, S: 0.05, Mn:
0.30, Si: trace, and Fe: balance. The electrodes were prepared by
embedding the steel rods in epoxy resin and exposing a surface
area of 1 cm2 to the electrolyte. Prior to each measurement, the
sample surfaces were abraded with 400, 600 and 1000 grade em-
ery paper, washed with double-distilled water, degreased with
acetone and dried in warm air.
All electrochemical measurements were conducted in a
thermostated conventional three-electrode Pyrex cell. A saturated
calomel electrode (SCE) and a large-area platinum wire were used
as the reference and auxiliary electrodes, respectively. The electro-
lyte was a 1 mol L1 HCl solution prepared from 37% HCl (pur-
chased from Merck Co. (Darmstadt Germany) and double
distilled water. All experiments were carried out in 100 mL of
non-stirred and naturally aerated electrolyte maintained at 25 C.
In all experiments, the carbon steel electrode was allowed to
reach its stable open-circuit potential (OCP), which occurred after
1 h. Electrochemical impedance measurements were performed
over a frequency range of 100 kHz to 4 mHz at the stable open-cir-
cuit potential with an AC wave of 10 mV. The polarisation curves
were also obtained after 1 h in the open-circuit potential. They
ran from the cathodic to the anodic direction within the range of
850 to 250 mV/SCE with a scan rate of 1.0 mV s1.
The electrochemical experiments were performed using a PC-
controlled VoltaLab 80 system with a PGZ402 potentiostat (Radi-
ometer Analytical) coupled with VoltaMaster 4 software.
The inhibition efciency (n%) was calculated from potentiody-
namic polarisation curves and electrochemical impedance dia-
grams as follows:
jcorr;0  jcorr
n%  100; 1
jcorr;0
where jcorr,0 is the density of the corrosion current in the absence of
the inhibitor and jcorr is the density of the corrosion current in the
presence of the inhibitor obtained from Tafel plots.
Rct  Rct;0
n%  100; 2
Rct
where is Rct,0 is the charge-transfer resistance in the absence of the
inhibitor and Rct is the charge-transfer resistance in the presence of
the inhibitor obtained from the electrochemical impedance
diagrams.
2.4. Weight loss experiment
C-steel specimens with the same composition used in the elec-
trochemical measurements were mechanically cut into 3.0 cm 
1.0 cm  1.0 cm sections, polished with a silicon carbide emery
cloth (grades 320, 400, 600 and 1000), washed with double dis-
tilled water, degreased with acetone and dried in warm air. Tripli-
cate specimens were immersed in the acid test solutions in the
absence and presence of 400 mg L1 aqueous extracts of used cof-
fee grounds obtained by infusion for 24, 48 and 72 h at 25 C. The
temperature was controlled using an aqueous thermostat. Speci-
mens were removed, rinsed in water and acetone, dried in warm
air and stored in a desiccator. The weight loss was determined
using an analytical balance with a precision of 0.1 mg. The inhibi-
tion efciency (n%) was obtained using the equation
W0  W
n%  100; 3
W0
where W0 and W are the weight loss in the absence and presence of
the extract, respectively.
The effect of temperature on the corrosion rate for steel cou-
pons in 1 mol L1 HCl solution at 25, 35, 45 and 55 C was also
examined. This experiment was performed in the absence and
presence of 400 mg L1 aqueous coffee ground extracts obtained
using the infusion process with an immersion period of 2 h.
In the present study, each experiment was repeated three times
under the same conditions, and the relative differences between
the replicate experiments were found to be smaller than 3%, indi-
cating good reproducibility. The average of the three replicate val-
ues was used to further process the data.
2.5. Surface analysis
The specimens used to examine the surface morphology were
immersed in 1 mol L1 HCl in the absence and presence of
400 mg L1 of aqueous coffee ground extracts obtained via decoc-
tion and infusion at 25 C for 2 h. The analysis was performed on
a CARL ZEISS EVO MA-10 scanning electronic microscope. The
accelerating voltage was 20 kV.
 V.V. Torres et al. / Corrosion Science 53 (2011) 23852392 2387

3. Results and discussion
3.1. Electrochemical experiments
3.1.1. Potentiodynamic polarisation curves
Fig. 1A and B present the potentiodynamic polarisation curves
of C-steel in a 1 mol L1 HCl solution in the absence and presence
of aqueous coffee ground extracts obtained via decoction and infu-
sion at 25 C. The respective electrochemical parameters (i.e., cor-
rosion potential (Ecorr), corrosion current density (jcorr), and the
anodic (ba) and cathodic (bc) Tafel constants, as shown in Table
1), were derived from Tafel plots.
From the potentiodynamic polarisation curves, it can be seen
that the extracts caused a decrease in both anodic and cathodic
current densities, most likely due to the adsorption of the organic
compounds present in the extracts at the active sites of the elec-
trode surface. This also slowed both metallic dissolution and
hydrogen evolution and consequently slowed down the corrosion
process. The two extracts exhibited the same behaviour.
2 process. The level of inhibition efciency calculated based on the
10
10
1 83.5 and between 78.2 and 88.1 in the concentration range of
(A)
-2
Current Density / mA cm
Table 1 indicates that, in the presence of the coffee extracts,
both stable OCP and Ecorr (derived from the Tafel plots) shifted
cathodically with respect to the blank (734 mV and 1028 mV,
respectively). An inhibitor can be classied as cathodic or anodic
if the difference in corrosion potential is more than 85 mV with re-
spect to the corrosion potential of the blank [24]. Such results will
indicate that the coffee extracts act as a mixed-type inhibitor with
predominant cathodic effectiveness. These results show that the
coffee extracts can retard both anodic and cathodic reactions under
open-circuit and polarised conditions.
The corrosion current density (jcorr) decreased as the amount of
the inhibitor increased, and this decrease was slightly more inu-
ential in the case of the extract obtained via infusion. The cathodic
Tafel slopes (bc) did not change signicantly with the addition of
the extracts (Table 1), indicating that the adsorbed inhibitor mole-
cules do not affect hydrogen evolution: i.e., hydrogen evolution is
diminished exclusively by the surface-blocking effect. The anodic
region of the potentiodynamic polarisation curves can be seen as
indicating an expressive increase in the anodic Tafel slopes (ba)
with the addition of the inhibitor. These results show that the ad-
sorbed inhibitor to the carbon-steel modies the metal dissolution
jcorr values obtained in the absence and presence of coffee extracts
from decoction and infusion, respectively, falls between 78.7 and
100400 ppm.

0
10
3.1.2. Electrochemical impedance spectroscopy (EIS)
Fig. 2AC illustrate the electrochemical impedance diagrams for
10
-1 C-steel in a 1 mol L1 HCl solution in the absence and presence of
aqueous extracts from coffee grounds. Table 2 summarises the
Free impedance data from EIS experiments carried out in both the ab-
-2
10 100 mg L
-1 sence and the presence of increasing extract concentrations. In ex-
-1
tract-free solutions, only one depressed capacitive loop was
200 mg L observed and that loop can be attributed to the time constant of
-3 -1
10 300 mg L the charge transfer and the double layer capacitance. Such a
400 mg L
-1 depression is characteristic of solid electrodes and is often ascribed
to dispersion effects, which have been attributed to roughness and
-4
10 inhomogenities on the surface during corrosion [18,21]. This
-800 -700 -600 -500 -400 -300 -200
behaviour is unaffected by the presence of the inhibitor, indicating
Potential / mV/SCE
the activation-controlled nature of the reaction during a one-
2 charge transfer process. The intersection of this semicircle with
10
the real axis at high frequencies produced a value of approximately
1.25 X cm2 for the ohmic resistance (Rs) of the solution. The
1 charge-transfer resistance (Rct) values were calculated based on
10
the difference in impedance values at lower and higher frequen-
-2

cies. The double layer capacitance (Cdl) was calculated using the
Current Density / mA cm

0 (B) equation below:

10

1
C dl ; 4
10
-1 2pfmax Rct

where fmax is the frequency at which the imaginary component of

free
10
-2
the impedance is maximal. A Cdl value of 475 lF cm2 was deter-
-1
100 mg L mined for the C-steel electrode in 1 mol L1 HCl. The electrochem-
-1
200 mg L ical impedance diagrams obtained in the presence of the extract
-3
10 300 mg L
-1 also show only one depressed capacitive loop. Based on Table 2, it
-1
is clear that the Rct values increased and the Cdl values decreased
400 mg L with increasing inhibitor concentration. These results may be
-4
10 attributable to the adsorption of the components in the aqueous
-800 -700 -600 -500 -400 -300 used coffee ground extracts onto the metal/solution interface. In-
Potential / mV/SCE deed, this hypothesis is corroborated by the anodic and cathodic
Fig. 1. Polarisation curves for C-steel in 1 mol L1 HCl in the absence and presence
polarisation curves and the corrosion potential results. A compari-
of aqueous coffee ground extract at various concentrations: 100, 200, 300 and son of the inhibition efciencies obtained using the decoction and
400 mg L1. Extract obtained via decoction (A) and infusion (B). infusion methods shows that the values for the two methods are
2388 V.V. Torres et al. / Corrosion Science 53 (2011) 23852392

Table 1
Electrochemical parameters for C-steel in 1 mol L1 HCl in the absence and presence of aqueous coffee ground extract at various concentrations: 100, 200, 300 and 400 mg L1.
Extract obtained via decoction and infusion.

Medium [Inhibitor] (mg L1) OCP (mV/SCE) Ecorr (mV/SCE) jcorr (mA cm2) ba (mV/dec) bc (mV/dec) n (%)
Blank 459 461 0.309 48 101
BCDEC extract
100 470 489 0.0659 103 101 78.7
200 482 489 0.0650 102 105 79.0
300 467 486 0.0605 91 108 80.4
400 493 489 0.0509 100 105 83.5
BCINF extract
100 488 487 0.0674 105 107 78.2
200 466 472 0.0402 77 106 87.0
300 471 471 0.0389 83 118 87.4
400 477 483 0.0369 91 126 88.1

7 Table 2
(A) Free
Impedance data for C-steel in 1 mol L1 HCl in the absence and presence of aqueous
6
2

coffee ground extract at various concentrations: 100, 200, 300 and 400 mg L1.
-Imaginary Part / cm

5 Extracts were obtained via decoction and infusion.

4 Medium [Inhibitor] Rct fmax CdI n

(mg L1) (X cm2) (Hz) (lF cm2) (%)
3
Blank 10.6 31.6 475
2
BCDEC extract
1 100 69 40.0 58 85
200 90 31.6 56 88
0
0 2 4 6 8 10 12 14 300 114 31.6 44 91
2 400 160 40.0 25 93
Real Part / cm
BCINF extract
100 87 40.0 46 88
100
(B) 100 mg L
-1 200
300
107
156
31.6
25.0
47
41
90
93
2

80
-Imaginary Part / cm

-1
200 mg L 400 198 20.0 40 95
-1
60 300 mg L
-1
400 mg L
40
The variation in h according to the extract concentration indi-
20 cates the adsorption isotherm that describes the system. The t
to the Langmuir isotherm was determined by plotting C/h versus
0 C according to the following equation:
0 40 80 120 160 200
2
Real Part / cm
C 1
140 C; 5
(C) h K
2
-Imaginary Part / cm

120 100 mg L
-1

100 200 mg L
-1
where K is the adsorption equilibrium constant. Fig. 3 displays lin-
80 300 mg L
-1
ear plots with high correlation coefcients of 0.9997 and 0.9998 and
60 400 mg L
-1 a slope of 1.034 and 1.024 for the decoction and infusion extracts,
40 respectively. This behaviour suggests that the compounds present
in the coffee ground extracts adsorbed onto the C-steel surface were
20
consistent with a Langmuir adsorption isotherm, indicating the ab-
0 sence of interaction forces between the adsorbed molecules. The
0 40 80 120 160 200 240 280
2
decoction and infusion extracts behaved similarly; the values of
Real Part / cm the adsorptive equilibrium constants at 25 C are 0.0571 and
0.0645 L mg1. It is important to emphasise that some authors have
Fig. 2. Impedance diagrams obtained for C-steel in 1 mol L1 HCl solution in the
absence (A) and presence of aqueous coffee ground extract at different concentra- compared the adsorptive equilibrium constant values obtained
tions: 100, 200, 300 and 400 mg L1. Extract obtained via decoction (B) and infusion from natural product extracts with those of good acid inhibitors
(C). such as Tween-85 (0.9608 L mg1 for K) but that such a comparative
analysis is inappropriate if the molecular mass of the components
responsible for the adsorption process is unknown [19]. Therefore,
quite similar for each concentration, although the infusion method it is also impossible to discuss the behaviour of adsorption iso-
yielded slightly higher n values at all concentrations tested. therms supporting the use of natural product extracts as inhibitors
Inhibition efciency is directly proportional to the fraction of in terms of thermodynamic parameters, such as the standard free
the surface covered by the adsorbed molecules (h), which was cal- energy of adsorption value (DGads) [18], if the chemical nature of
culated using the data obtained from electrochemical impedance the adsorbed substance is unknown. Unfortunately, the literature
diagrams and the equation h = n/100 (see Eq. (2)). has employed this incorrect approach [21].
 V.V. Torres et al. / Corrosion Science 53 (2011) 23852392 2389

500 Table 3
(A) C-steel weight loss data when 1 mol L1 HCl was used in the absence
450 and presence of 400 mg L1 aqueous coffee ground extract obtained
via infusion for varying periods of time: 24, 48 and 72 h at 25 C.
400
Medium Wcorr (mg cm2 h1) n (%)
350
-1

Blank, 24 h 2.86
C/ / mg L

300 With inhibitor, 24 h 0.079 97

Blank, 48 h 1.58
250 With inhibitor, 48 h 0.146 91
Blank, 72 h 1.19
200 With inhibitor, 72 h 0.236 80

150

100
Table 4
50 C-steel weight loss data when 1 mol L1 HCl was used in the
absence and presence of 400 mg L1 aqueous coffee ground extract
0 obtained via infusion with 2 h of immersion time at 25, 35, 45 and
0 50 100 150 200 250 300 350 400 450 500
55 C.
-1
C / mg L
Temperature (C) Wcorr Wcorr n (%)
(g cm2 h1) (g cm2 h1)
500 Blank With inhibitor
(B)
450 25 0.0027 0.00064 76
35 0.0045 0.00072 84
400 45 0.0073 0.00096 87
55 0.0130 0.0012 91
-1

350
C/ / mg L

300

250 free and inhibited acid media increased as the temperature

increased. Additionally, the inhibition efciency of the coffee
200
grounds extract increased with temperature.
150 The apparent activation energy associated with C-steel corro-
sion in free and inhibited acid solution was determined using an
100
Arrhenius-type plot according to the following equation:
50
Ea
0 log W corr log A; 6
0 50 100 150 200 250 300 350 400 450 500 2:303RT
-1
C / mg L
where Wcorr is the corrosion rate, Ea is the apparent activation
Fig. 3. Langmuir adsorption isotherm for aqueous coffee ground extract on a C-steel energy, A is the frequency factor, T is the absolute temperature
surface in a 1 mol L1 HCl solution: (A) decoction and (B) infusion. and R is the molar gas constant. Arrhenius plots of (log Wcorr)
against (1/T) for C-steel in a 1 mol L1 HCl solution in the absence
and presence of the coffee ground extracts are shown graphically
3.2. Weight loss measurements in Fig. 4. The apparent activation energy obtained for the corrosion
process in the free acid solution was found to be 41.9 kJ/mol and
The results of the weight loss measurements associated with 16.8 kJ/mol in the presence of the inhibitor. Notably, the energy
the corrosion of C-steel in a 1 mol L1 HCl solution in the absence
and presence of 400 mg L1 aqueous coffee ground extracts ob-
tained via infusion for varying immersion times (24, 48 and 72 h) -1.6 Free
at 25 C are provided in Table 3. These assays were critical in deter- With inhibitor
-1.8
mining the stability of the extract with time. It was observed that
the rate of C-steel corrosion signicantly decreased upon the addi- -2.0
-2 -1

tion of coffee ground extracts for all immersion times. These re- -2.2
log Wcorr / g cm h

sults indicate the inhibitory effect of this extract on C-steel

-2.4
corrosion in acidic solution. It is also clear that inhibition efciency
decreases with time, as it dropped from 97% after 24 h of immer- -2.6
sion to 80% after 72 h. There was some deterioration of the protec- -2.8
tive layer formed in the presence of the aqueous coffee ground
-3.0
extracts on the C-steel surface. These results are important because
the immersion times normally used in the literature are much low- -3.2
er than those tested in the experiments presented here. The data -3.4
also indicate that inhibition efciency can remain relatively high
with long periods of immersion. -3.6
0.0030 0.0031 0.0032 0.0033 0.0034
The effect of temperature on the corrosion of C-steel in a -1
1 mol L1 HCl solution ranging from 25 to 55 C with 2 h of immer- 1/T / K
sion time is presented in Table 4. The experiments were performed Fig. 4. Arrhenius plots for the corrosion rate for C-steel in a 1 mol L1 HCl solution
in the absence and presence of 400 mg L1 aqueous coffee ground in the absence and presence of 400 mg L1 aqueous coffee ground extract obtained
extracts obtained by infusion. The corrosion rates for steel in both via infusion.
2390 V.V. Torres et al. / Corrosion Science 53 (2011) 23852392

barrier of the corrosion reaction decreased in the presence of the the main corrosion inhibitors present in extracts of natural prod-
inhibitor. The inhibition efciency level also increased as the tem- ucts [23,26]. It emerged that the amount of phenolics overall and
perature increased (Table 4). Bouyanzer et al. have traced such the amounts of individual chlorogenic acids were slightly higher
observations to specic interactions between the steel surface and (approximately 1520% higher) in the extract obtained via the
the inhibitor [17,30]. The lower value of Ea in the corrosion process infusion of spent coffee grounds than in the extract obtained using
in the presence of inhibitor, as compared to that in a free acid solu- decoction (Table 5). Although it could be expected that decoction
tion, can be attributed to the chemisorption of the inhibitor on the would be a more effective method of extracting water-soluble
steel surface. compounds, exposure to a temperature of 100 C during decoction
may have resulted in the degradation of the compounds investi-
3.3. Phenolic components of the coffee extracts gated and therefore in higher levels of antioxidant components
in the infusion extract. These higher levels of phenolic compounds
Phenolic compounds and alkaloids are substances normally in the infusion extract might in turn explain the slightly higher de-
found in coffee samples [31] and are generally considered to be gree of corrosion inhibition observed for the infusion extract
(Tables 1 and 2). In contrast, the aqueous extracts of spent coffee
Table 5 grounds were found to be similar in terms of their relative distribu-
Phenolic components and total phenolics of freezedried aqueous extract from spent
tion of chlorogenic acids. These results are consistent with all of
coffee grounds.
the electrochemical analyses, which revealed relatively similar
Phenolic components Coffee extracts behaviour. Free caffeic acid was not detected in either type of spent
BCDEC extract BCINF extract coffee extract.
3-CQA 10.5 14.4 The major phenolic compounds in coffee are chlorogenic acids,
5-CQA (mg/g) 14.4 20.9 which are monoesters or diesters of quinic acid with trans-
3-FQA (mg/g) 3.3 4.8 hydroxycinnamic acids, such as cafeic and ferulic acids. The con-
4-CQA (mg/g) 11.4 15.8
tents of the chlorogenic acids in the freezedried aqueous extracts
5-FQA (mg/g) 1.7 3.3
4-FQA (mg/g) 1.3 1.9
in the present study (Table 5) are comparable to those reported for
3,4-diCQA (mg/g) 3.4 5.2 instant coffee samples [32], indicating that the residual amounts of
3,5-diCQA (mg/g) 2.3 3.3 phenolic compounds in spent coffee grounds were easily extract-
4,5-diCQA (mg/g) 1.1 2.0 able with water in this study. The relatively high amounts of phe-
CFQA (mg/g) 2.0 3.2
nolic components in spent coffee grounds indicate that these
CGA (mg/g) 51.3 74.9
Total phenolics (mg GAE/g) 64.2 77.2 compounds are candidates to inhibit the corrosion of carbon steel.

3-CQA: 3-caffeoylquinic acid; 5-CQA: 5-caffeoylquinic acid; 3-FQA: 3-feruloylquinic

acid; 4-CQA: 4-caffeoylquinic acid; 5-FQA: 5-feruloylquinic acid; 4-FQA: 4-feru- 3.4. Studies with isolated 5-caffeoylquinic acid
loylquinic acid; 3,4-diCQA: 3,4-dicaffeoylquinic acid; 3,5-diCQA: 3,5-dicaffeoylqui-
nic acid; 4,5-diCQA: 4,5-dicaffeoylquinic acid; CFQA: Total caffeoylferuloylquinic Isolated 5-caffeoylquinic acid, as indicated in Fig. 5, was studied
acids; CGA: Total chlorogenic acids.
using electrochemical impedance in 1 mol L1 HCl because it is the
major chlorogenic acid that was found in both of the extracts. The
corresponding results are presented in Fig. 6 and Table 6. Again, the
HO CO2H electrochemical impedance diagrams obtained using the chloro-
genic acid show only one depressed capacitive loop with a slight
O increase in charge transfer resistance. Table 6 lists a selection of
electrochemical parameters for C-steel in 1 mol L1 HCl in the
HO absence and presence of the standard. The stable open-circuit po-
O
tential was slightly more negative when chlorogenic acid was used.
OH The inhibition efciency levels of the chlorogenic acid were 28%
OH
and 33% in the presence of 100 and 200 mg L1, respectively. These
OH values of inhibition efciency are several times lower than those
observed for the coffee extracts (Tables 1 and 2). It should be noted
Fig. 5. Chemical structure of 5-caffeoylquinic acid. that, although the concentration levels of the total chlorogenic

10
free
-1
5-CQA 100 mg L
8
-1
2

5-CQA 200 mg L
-Imaginary Part / cm

0
0 4 8 12 16 20
2
Real Part / cm

Fig. 6. Impedance diagrams for C-steel in a 1 mol L1 HCl solution in the absence and presence of 5-caffeoylquinic acid (5-CQA).
 V.V. Torres et al. / Corrosion Science 53 (2011) 23852392 2391

Table 6 extracts, the corrosion inhibition efciency was approximately

Electrochemical parameters of C-steel in 1 mol L1 HCl in the absence and presence of 90% for all coffee extracts. Thus, it would seem that isolated chlor-
5-caffeoylquinic acid (5-CQA).
ogenic acids do not seem to be responsible for the corrosion inhi-
Medium OCP (mV/SCE) Rct fmax CdI (lF cm2) n (%) bition results observed for the coffee extracts. In the present
(X cm2) (Hz) study, it is not possible to determine what components present
Blank 459 10.6 31.6 475 in the coffee extracts created their relatively high ability to inhibit
100 mg L1 5-CQA 464 14.8 31.6 340 28 corrosion. However, it can be speculated that other components
200 mg L1 5-CQA 472 15.8 79.4 127 33
present in the coffee extracts might have precipitated the observed
effects and that a complex synergistic relationship between chlor-
ogenic acids and other components of coffee may have been
responsible for the results of this study.

3.5. Surface analysis

Fig. 7 shows a SEM photograph recorded for C-steel samples

immersed for 2 h in a 1 mol L1 HCl solution in the absence (A)
and presence of 400 mg L1 of aqueous coffee grounds extracts
obtained using decoction (B) and infusion (C) at 25 C. The mor-
phology in Fig. 7A shows the type of rough surface characteristic
of the uniform corrosion of C-steel in acid as previously reported
[18], indicating that in extract-free solution, the surface is highly
corroded. C-steel immersed in 1 mol L1 solution with either type
of coffee extract was noted to have a smooth surface (Fig. 7B and
C), indicating that the surface was protected by the extracts. These
results corroborate all of the electrochemical and weight loss
measurements.

4. Conclusions

Aqueous extracts of coffee grounds can act as an effective, nat-

urally derived, green corrosion inhibitor for C-steel in a 1 mol L1
HCl solution.

(1) All electrochemical results, including the slightly more

negative values for OCP and corrosion potential, the indi-
cations of inhibitory action in the anodic and cathodic
polarisation curves and the electrochemical impedance
measurements, showed that the examined extracts acted
as mixed-type inhibitors with predominant cathodic
effectiveness.
(2) The adsorption of the studied extracts was consistent with
the Langmuir adsorption isotherm.
(3) The inhibition efciency of C-steel in 1 mol L1 HCl increased
with the temperature increases.
(4) The apparent activation energy (Ea) of C-steel dissolution
decreased when the extract was used.
(5) The action of the extract as a corrosion inhibitor for C-steel
in acid solution can be attributed to a strong chemisorptive
bond.
(6) The slight differences between the coffee extracts in terms of
chlorogenic acid composition and total phenolic contents
are consistent with the slight differences observed between
the two extracts in terms of inhibitory efciency.
(7) The presence of isolated chlorogenic acids does not seem to
explain the corrosion inhibition observed for the coffee
extracts. In the present study, it is impossible to determine
Fig. 7. SEM photograph (2000) of C-steel: immersed in 1 mol L1 HCl in the what components present in the coffee extracts caused their
absence (A) and presence of 400 mg L1 of aqueous coffee ground extract obtained relatively high ability to inhibit corrosion.
via decoction (B) and infusion (C).

Acknowledgements

acids (CGA, Table 5) during the electrochemical analyses of the cof- The authors thank Petrobras and the National Council of Tech-
fee extracts were relatively low, varying from 7.5 to 30 and 5.1 to nological and Scientic Development (Brazil) for research fellow-
21 ppm, respectively, in the assays of infused and decocted coffee ship support.
2392 V.V. Torres et al. / Corrosion Science 53 (2011) 23852392

References [19] X.H. Li, S.D. Deng, H. Fu, J. Appl. Electrochem. 40 (2010) 16411649.
[20] I. Radojeic, K. Berkovic, S. Kovac, J. Vorkapic-Furac, Corros. Sci. 50 (2008) 1498
1504.
[1] A.Y. El-Etre, Mater. Chem. Phys. 108 (2008) 278282.
[21] S.A. Umoren, I.B. Obot, N.O. Obi-Egbedi, J. Mater. Sci. (2009) 274279.
[2] S. Cheng, S. Chen, T. Liu, X. Chang, Y. Yin, Electrochim. Acta 52 (2007)
[22] M.I. Awad, J. Appl. Electrochem. 36 (2006) 11631168.
59325938.
[23] P.C. Okafor, M.E. Ikpi, I.E. Uwah, E.E. Ebenso, U.J. Ekpe, S.A. Umoren, Corros. Sci.
[3] M.S. Morad, A.M.K. El-Dean, Corros. Sci. 48 (2006) 33983412.
50 (2008) 23102317.
[4] I.H. Farooqi, A. Hussain, M.A. Quraishi, P.A. Saini, Anti-Corros. Meth. Mater. 46
[24] A.K. Satapathy, G. Gunasekaran, S.C. Sahoo, K. Amit, P.V. Rodrigues, Corros. Sci.
(1999) 328331.
51 (2009) 28482856.
[5] A.Y. El-Etre, Corros. Sci. 40 (1998) 18451850.
[25] A.M. Abdel-Gaber, B.A. Abd-El-Nabey, M. Saadawy, Corros. Sci. 51 (2009)
[6] A.Y. El-Etre, M. Abdallah, Corros. Sci. 42 (2000) 731738.
10381042.
[7] A.Y. El-Etre, Corros. Sci. 43 (2001) 10311039.
[26] K.W. Tan, M.J. Kassim, Corros. Sci. 53 (2011) 569574.
[8] A.Y. El-Etre, M. Abdallah, Z.E. El-Tantawy, Corros. Sci. 47 (2004) 385395.
[27] A.L. Waterhouse, Determination of total phenolics, in: R.E. Wrolstad (Ed.),
[9] A.Y. El-Etre, Appl. Surf. Sci. 252 (2006) 85218525.
Current Protocols in Food Analytical Chemistry, John Wiley & Sons, Inc., New
[10] Y. Li, P. Zhao, Q. Liang, B. Hou, Appl. Surf. Sci. 252 (2005) 12451253.
York, 2002.
[11] A.Y. El-Etre, Corros. Sci. 45 (2003) 24852495.
[28] A. Farah, T. De Paulis, L.C. Trugo, P.R. Martin, J. Agric. Food Chem. 53 (2005)
[12] A. Bouyanzer, B. Hammouti, L. Majidi, Mater. Lett. 60 (2006) 28402843.
15051513.
[13] P.B. Raja, M.G. Sethuraman, Pigment Resin Technol. 38 (2009) 3337.
[29] A. Farah, T. De Paulis, D.P. Moreira, L.C. Trugo, P.R. Martin, J. Agric. Food Chem.
[14] P.B. Raja, M.G. Sethuraman, M. Gopalakrishnan, Mater. Lett. 62 (2008) 113116.
54 (2006) 374381.
[15] K.S. Parikh, K.J. Joshi, Trans. SAEST 39 (2004) 2935.
[30] F.S. de Souza, A. Spinelli, Corros. Sci. 51 (2009) 642649.
[16] A.Y. El-Etre, Bull. Electrochem. 22 (2006) 7580.
[31] M.C. Monteiro, L.C. Trugo, Quim Nova. 28 (2005) 637641.
[17] A. Bouyanzer, B. Hammouti, Pigment Resin Technol. 33 (2004) 287292.
[32] L.C. Trugo, R. Macrae, Analyst 109 (1984) 263266.
[18] J.C. da Rocha, J.A.C.P. Gomes, E. DElia, Corros. Sci. 52 (2010) 23412348.