Microstructure, surface chemistry and electrochemical response of Ag|AgCl sensors in alkaline media

24 Pages·2017·2.52 MB·English

by  Farhad Pargar

J Mater Sci (2018) 53:7527–7550
Electronic materials
ELECTRONIC MATERIALS
Microstructure, surface chemistry and electrochemical
response of Ag|AgCl sensors in alkaline media
Farhad Pargar1,* , Hristo Kolev2 , Dessi A. Koleva1 , and Klaas van Breugel1
1Faculty ofCivilEngineering and Geosciences, Materials andEnvironment, Technische UniversiteitDelft, Stevinweg 1,
2628CN Delft,The Netherlands
2Institute ofCatalysis, Bulgarian Academy ofSciences, Acad. G.BonchevStr., bl.11,1113 Sofia,Bulgaria
Received:24October2017 ABSTRACT
Accepted: 27January 2018
Characterization of the Ag/AgCl electrode is a necessary step toward its
Published online: application as a chloride sensor in a highly alkaline medium, such as concrete.
6 February2018 The nucleation and growth of AgCl on Ag in 0.1 M HCl was verified through
cyclic voltammetry. Ag anodization was performed at current densities, deter-
(cid:2) The Author(s) 2018. This mined by potentiodynamic polarization in the same (0.1 M HCl) medium. The
article is an open access morphology and microstructure of the AgCl layers were evaluated via electron
publication microscopy, while surface chemistry was studied through energy-dispersive
spectroscopy and X-ray photoelectron spectroscopy. At current density above
2 mA/cm2, the thickness and heterogeneity of the AgCl layer increased. In this
condition, small AgCl particles formed in the immediate vicinity of the Ag
substrate,subsequentlyweakeningthebondstrengthoftheAg/AgClinterface.
Silveroxide-basedorcarbon-basedimpuritieswerepresentonthesurfaceofthe
sensor in amounts proportional to the thickness and heterogeneity of the AgCl
layer.Itisconcludedthatawell-definedlinkexistsbetweenthepropertiesofthe
AgCl layer, the applied current density and the recorded overpotential during
Aganodization.Theresultscanbeusedasarecommendationforpreparationof
chloride sensors with stable performance in cementitious materials.
Introduction discussion [1–3]. Discussions concern the OCP
responses of the sensors in alkaline medium and the
The determination of the chloride content in rein- observedscatteramongtheseresponses.Forinstance,
forced concrete (RC) structures is important for the similar Ag/AgCl sensors are reported to give a dif-
assessment of the probability of chloride-induced ferent response to identical alkaline medium, which
corrosion of steel reinforcement. Although the is a nonexpected outcome. While the sensor’s
application of Ag/AgCl sensors for the above pur- response is just a common open-circuit potential
pose is well known, the benefits and drawbacks of (OCP) record, i.e., the Ag/AgCl electrode potential
utilization of these sensors are still a subject of reading, the possible discrepancies will reflect the
Address correspondence toE-mail: [email protected]
https://doi.org/10.1007/s10853-018-2083-0
7528 J Mater Sci (2018) 53:7527–7550
time needed for establishing a dynamic equilibrium Different techniques are available for the AgCl for-
at the sensor/environment interface. Although fun- mation on a Ag substrate [13]. The AgCl layer can be
damental electrochemical thermodynamics and attached(physicallybond)ontheAgsubstratebydip-
kinetics can explain these discrepancies [4–12], prac- pingAginmoltensilverchloride,thelayercanbeelec-
tical application of the Ag/AgCl sensors for RC trochemicallyformedbyanodizationinHClsolution,or
structures is still to be justified. This justification chemically formed, by oxidation with aqueous FeCl
3
refers to specifying the limits for these sensors’ solution. The choice of a technique is determined by a
application and the reasons behind the existence of generalattempttoimprovetheperformanceofAg/AgCl
such limits. electrodesinviewoftheirsensitivitytovariousparam-
The motivation of this work lies in an attempt to eters, involved in the manufacturing process. For
clarify the origin of the unexpected results obtained instance,inthecaseofscreenprinting,aAg–AgClpaste
with Ag/AgCl sensors in alkaline medium. The lack isdepositedonametallicsurface(e.g.,Pt)toproducethe
ofagreementamongtheOCPresponseofthesensors Ag/AgClelectrode.Theadhesivepropertiesofthepaste
has been reported as partly attributable to the sen- to the supporting electrode might not be very good;
sors’ preparation methods [1]. Varying physical hence, the reliability of the sensor will depend on the
properties of the AgCl layer were considered as the homogeneityoftheAg–AgClpasteused[14].Ingeneral,
reason for OCP responses, different from those the relatively small contact area, i.e., weak adhesion
expected [2]. The potentially weak bond between the betweentheAgrodandAgCllayer,mayleadtooxida-
AgCllayerandtheAgsubstrateofthesensor,abond tion of AgCl at Ag/AgCl interface by hydroxide ions
which develops during sensor’s preparation, has fromthealkalinemediumatarelativelyfastrate[15].
been suggested as the main reason for the poor per- One of the most frequently used methods for Ag/
formance of the sensor in alkaline medium [3], AgCl sensor preparation is anodization of Ag in a
although this has not been experimentally justified. HCl solution [16–19]. The low pH of a HCl solution
Hence, the experimental evidence to support this wouldlimitthepossibleAg Oformation[20,21]and
2
hypothesis is scarce [1, 4]. would result in AgCl formation on the Ag substrate.
The thermodynamic considerations and the kinet- A layer of AgCl with sufficient thickness was con-
ics of AgClformation on a Agsubstrate were subject sideredtobeofca.tensofthemicrometerrange[12].
to numerous works in the fields of electrochemistry With anodization, the rate of AgCl formation
and crystal growth since the early 1950s [5–11]. depends on the applied current density and the time
Despite the results in these thoroughly elaborated oftheanodization [22].Different currentdensityand
works, the properties and performance of Ag/AgCl anodizing time have been reported for the prepara-
electrodes in the alkaline medium are yet to be tion of sensors, e.g., 0.4 mA/cm2 for 30 min [16, 19],
affirmed, especially in view of their application as 2 mA/cm2 for 30 min [17], 0.2 mA/cm2 for 1 h [18],
chloride sensors for RC structures. 0.4 mA/cm2 for 2 h [3]. Logically, as with each elec-
One of the reasons for a limited number of reports trochemical process of this kind, the anodization
onthecorrelationbetweenthepropertiesoftheAgCl regime will affect the produced layer thickness, sur-
layer and the response of the sensor in alkaline face morphology and ionic/electron conductivity,
medium is related to unsuccessful attempts for respectively [5, 10]. These features of the AgCl layer
observing the inner morphology of the AgCl layer. may subsequently be reflected in the electrochemical
For instance, as reported in [4, 11, 12]: microscopic response of the sensor [20], especially in alkaline
observations of the Ag/AgCl interface was prob- medium.Forinstance,aporousmicrostructureofthe
lematic.TheattemptstostiffenthesamplebyCuand AgCl layer increases the probability of a mixed
Ni–P plating or liquid N (low temperature of potential by limiting the rate at which the sensor can
2
approx. – 190 (cid:3)C) was not successful, since the soft achieve equilibrium potential [23]. This should be
AgCl particles would spread over the entire surface avoided, if possible, in order to achieve a more
of the samples and mask the interface. Sample accurate and reliable sensor’s response.
preparation to address the above challenges is dis- Considering aforementioned, the focus of this
cussed in this work and is part of in-depth investi- paper is to correlate the properties of the AgCl layer
gation of the Ag/AgCl sensors’ performance in totheanodizationregimeandlinkthesetotheactual
alkaline medium. electrochemical response of the sensor in chloride-
J Mater Sci (2018) 53:7527–7550 7529
containing alkaline medium. The importance of this to 1 h, and (2) for equal duration of 1 h, but at four
correlation is in view of the research question: how different current densities, i.e., 0.5 mA/cm2 (regime
different morphology, microstructure and composi- A),1 mA/cm2(regimeB),2 mA/cm2(regimeC)and
tion of the AgCl layer are important for the accuracy 4 mA/cm2(regimeD).Thewiresanodizedinthefirst
ofasensoranddeterminationofthechloridecontent manner,i.e.,highestcurrentdensityandvaryingtime
in a relevant medium. The significance of the above of anodization, are designated as D-type sensors,
dependencies is also evaluated in view of the intrin- since they formed a parallel group of the D sensors,
sic conductivity of a AgCl layer (ionic and electron prepared via the second approach at 4 mA/cm2 for
conductivity), more importantly how these proper- 1 h anodization only. The D-type sensors were only
ties determine an electrochemical equilibrium at a used for microscopic investigations of the Ag/AgCl
sensor/solution interface, i.e., the sensor’s response, interfaceandassupportiveevidencetotheresultsfor
is also addressed. all other sensor types, including the discussion on
ohmic resistance of the AgCl layers. The experimen-
tal setup and instrumentation for sensors’ prepara-
tion were identical to the one for the CV and PDP
Experimental materials, methods
tests. The current regimes were chosen based on the
and technical background
PDP screening test as specified in ‘‘Electrochemical
Electrochemical tests on Ag in 0.1 M HCl tests on Ag in 0.1 M HCl’’ section. The thickness of
the AgCl layer was both experimentally determined
Linear sweep cyclic voltammetry (CV) and poten- (‘‘Morphology, microstructure and surface chemistry
tiodynamic polarization (PDP) were employed as of the AgCl layer’’ section) and theoretically
screening techniques to derive information for AgCl approximated by employing Eq. (1), which essen-
formation on the Ag substrate. Ag wires of 99.5% tially is a modification of Faraday’s law:
purity, 1 mm diameter, were supplied by Salomon’s i(cid:2)M(cid:2)t
X¼ ð1Þ
Metalen B.V. Netherlands. The CV and PDP tests F(cid:2)d
werecarriedoutin0.1 MHClsolution(pH * 1.4)in
where X is the thickness of the AgCl layer (cm); i is
a conventional three-electrode cell arrangement,
the applied current density (A/cm2); M is the
where the Ag wire (1 cm length) was the working
molecular weight of AgCl, 143.5 g/mol; t is the
electrode, a Pt mesh was the counter electrode and a
duration of anodization; F is the Faraday’s constant
saturated calomel electrode (SCE) was the reference
(F = 96500 C/mol/equiv); d is the density of the
electrode. The CV scans were performed from - 400
AgCl layer, 5.56 g/cm3. The theoretical approxima-
to? 150 mV(versusOCP)atascanrateof100 mV/s
tion of a AgCl layer after Ag anodization is also
withtheaimtoverifyAgClnucleationandgrowthin
reported in the literature [5].
thechosenmedium.PDPwasperformedintherange
of -200 to ? 1600 mV versus OCP at a scan rate of
Ag/AgCl sensors’ response in alkaline
0.5 mV/s.ThePDPtestresultsdeterminedtherange
medium
ofcurrentdensities,potentiallysuitableforanodizing
Ag in the chosen medium. The CV and PDP tests The sensors’ response ina modelalkalinemediumis
were performed using PGSTAT 302N potentiostat their open-circuit potential (OCP) over time. A brief
(Metrohm Autolab B.V., The Netherlands). background on OCP-related theory in view of the
Ag/AgCl electrode (or sensor) is as follows: record-
Ag anodization: preparation of Ag/AgCl ingtheOCPofthesensorreflectsthemainconceptof
sensors defining the chloride content at the sensor/medium
interface. Well known is that Ag/AgCl sensors are
The Ag wires were cleaned for 2 h in concentrated predominantlysensitivetochlorideions;hence,their
ammonia, then immersed in demineralized water response in chloride-containing medium is directly
overnight,priortoanodizationin0.1 MHClsolution. related to the activity (concentration, respectively) of
Anodization was carried out in two manners: (1) at thechlorideions.Thisrelationshipisgovernedbythe
4 mA/cm2withvaryinganodizationtimefrom 900 s Nernst equation (Eq. 2):
7530 J Mater Sci (2018) 53:7527–7550
RT Morphology, microstructure and surface
E¼E0Ag=AgCl(cid:3)2:303nFlg½aCl(cid:3)(cid:4) ð2Þ chemistry of the AgCl layer
where E is the measured OCP of the sensor versus a
The following procedure sequence (Fig. 1) was
reference electrode such as SCE, E0 is the stan-
Ag=AgCl employed prior to microscopic observations of the
dard electrode potential for the Ag/AgCl electrode Ag/AgCl interface: (1) a portion of the Ag wire was
(V),acl(cid:3) istheactivityofthechlorideions(mol dm-3) narrowed prior to anodization; (2) the narrowed
in the vicinity of the electrode, R is the gas constant portionwasstretchedfromthetwosidesofthenotch;
(J mol-1 K-1), F is the Faraday constant (C mol-1) (3)thisallowedinvestigationofa‘‘crosssection’’,i.e.,
and T is the absolute temperature (K). The chloride the parallel growth of the AgCl layer on the Ag
ions concentration (CCl(cid:3)) [molality] can be calculated substrate. The obtained cross section resembled a
further, using the relationship in Eq. (3), employing fracture surface. This enabled the inner morphology
the activity coefficient ccl(cid:3) [24]. oftheAgCllayertobewellobserved.Thisprocedure
acl(cid:3) ¼CCl(cid:3) (cid:2)cCl(cid:3) ð3Þ and sequence were specifically chosen for the pur-
pose of clear differentiation of the AgCl layers’ for-
Equation (2) is based on the electrochemical
mation on identically handled Ag substrates, but in
exchange equilibrium, presented by Eq. (4), which
conditions of different anodization regime (various
takes place on the surface of the sensor in chloride-
currentdensity,‘‘Aganodization:preparationofAg/
containing medium. In the alkaline medium of zero
AgCl sensors’’ section). This approach was chosen as
or low chloride ions content, the adopted OCP will
the only suitable one to address the objectives of this
deviate from the one following Eq. (2), since inter-
work and to overcome the challenges with Ag/AgCl
fering hydroxide ions will dominate the potential
sample preparation for microscopy purposes, as
response, i.e., E will not present the Ag/AgCl equi-
specifically outlined in the introduction section.
librium only, but also reflect a mixed potential of a
The AgCl layers were analyzed, using Environ-
Ag/AgCl/Ag O interface. In this case, the exchange
2 mental Scanning Electron Microscopy (ESEM), Phi-
equilibrium presented by Eq. (5) will be relevant.
lips-XL30 equipped with an energy-dispersive
AgþCl(cid:3) $AgClþe(cid:3) ð4Þ spectrometer (EDS). The samples were examined
under accelerating voltage of 20 kV in high vacuum
2AgClþ2OH(cid:3) $Ag Oþ2Cl(cid:3)þH O ð5Þ
2 2 mode.
MoredetailsonthetheoreticalaspectsofAg/AgCl The surface chemistry was evaluated through
response in alkaline medium, and relevant con- X-ray photoelectron spectroscopy (XPS). The mea-
straints,arereportedinarecentreviewonthesubject surements were taken using an ESCALAB MkII (VG
and references therein [25]. Scientific) electronspectrometer at abasepressure in
The OCP of the chloride sensors, as produced in the analysis chamber of 5 9 10-10 mbar using twin
different anodization regimes, were recorded in anode MgK /AlK X-ray source with excitation
a a
chloride-containing cement extract. OCP was moni- energies of 1253.6 and 1486.6 eV, respectively. The
tored by using the specified in ‘‘Electrochemical tests XPS spectra were recorded at the total instrumental
on Ag in 0.1 M HCl’’ section electrochemical setup. resolution (as it was measured with the FWHM of
The cement extract (CE) of pH ca. 12.8 was obtained Ag3d5/2 photoelectron line) of 1.06 and 1.18 eV for
by mixing cement powder (CEM I 42.5N) and dem- MgK andAlK excitationsources.Theprocessingof
a a
ineralized water at the ratio of 1:1. The bottled mix- themeasuredspectraincludedasubtractionofX-ray
ture was rotated for 24 h, followed by filtration for satellitesandShirley-typebackground[28].Thepeak
obtaining the extract. The chemical composition of positions and peak areas were evaluated by a sym-
CE is as follows: Ca—201 mg/l; K—3.85 mg/l; Na— metrical Gaussian–Lorentzian curve fitting. The rel-
1.33 mg/l; Al—4 mg/l and Fe\1 mg/l. Sodium ativeconcentrationsofvariouschemicalspecieswere
chloride was added as a solid to the desired con- determined by normalization of the peak areas to
centrations of 20 and 260 mM in the CE (CE is a their photoionization cross sections, calculated by
generally used model medium for electrochemical Scofield [29].
studies, e.g., for steel [26, 27], since it resembles the
pore water of cement-based materials).
J Mater Sci (2018) 53:7527–7550 7531
Fig.1 ProceduresequenceforthepreparationofacrosssectionoftheAg/AgClsensor:anarrowingtheAgwirepriortoanodization;
bstretching fromboth sides after anodization; cas aprepared crosssection for ESEMobservations.
Results and discussion forbothinterfacialanddiffusioncontrolofthecrystal
growth kinetics [39]. In other words, as seen in
AgCl nucleation and growth Fig. 2a), the hysteresis loop, forming in the reverse
scanoftherecordedCVforAgin0.1 MHClsolution,
CV tests
the cross-over point at ca. 40 mV and the peak C2,
define the nucleation of AgCl on the Ag substrate.
The formation of AgCl on a Ag substrate, from
Next, the cross-over point remains constant with
nucleation and growth to reduction and recovery of
subsequent scans (Fig. 2a, inlet), which proves an
the AgCl layer in various solutions, has been largely
interfacial control of the AgCl growth kinetics.
studied over the past decades. Both analytical and
ThefollowingfeaturesoftheCVresponse(Fig. 2a)
experimental works are reported and readily avail-
can also be noted: practically zero current on the
able [7, 9, 30–34]. Studying these processes generally
forwardscans(alsoconfirmingtheinertnessofAgin
means recording the metal’s response to a variety of
these conditions) was followed by a rapid rise of
imposed signals (waves). Linear sweep CV, for
anodic current, observed at 80 mV in the first scan
example, provides invariable information on the
and 54 mV in the subsequent scans. This single
nucleation process, while a PDP response would
anodic peak (A1) is a Ag dissolution peak in the 1st
define the range of current densities for Ag
scan and a stripping peak in the subsequent scans,
anodization.Itisnotasubjecttothispapertodiscuss
corresponding to the removal of already formed at
indetailtheelectrochemistrybehindAgClnucleation
negative potentials deposits on the Ag surface. A
andgrowth,buttoratherclarifythechosenapproach
cathodic shift for the base of the anodic peak was
toward Ag/AgCl sensors preparation. CV and PDP
observed with subsequent scans, at the point of the
were employed to that end, as screening techniques.
abrupt rise of anodic current (80 mV) (Fig. 2a, inlet).
Figure 2 depicts CV scans for Ag in 0.1 M HCl,
This shift was more significant between the 1st and
Fig. 2a, and the PDP response in 0.1 M HCl, Fig. 2b.
2nd CV scans (from 80 to 54 mV), marginal between
The aspects of employing CV and the theoretical
the 2nd and 5th scans (54–49 mV), and not relevant
interpretation of CV scans are reported in detail
afterward—toward the 20th scan. Prior to the catho-
[35–38] and will be only commented here to the
diccurrentmaximumonthereversescan(peakC2),a
extent of addressing the objectives of this work.
nearly linear current–potential region on either side
When a phase nucleation on a metal surface is
of the cross-over point (at 40 mV) can be observed
involved(asAgClonAg),theCVcurve(Fig. 2a)will
(Fig. 2a and inlet), where the cross-over potential
reflect a number of characteristic features: (1) an
would equal to the metal–metal ion reversible
enhanced anodic and cathodic peak currents sepa-
potential (in this case a slight deviation from the
ration, which is different from the otherwise
standard potential of 22 mV for the reaction equilib-
observed more narrow response for a general mass rium at a Ag/Ag? interface was observed). The
transport controlled CV; (2) a characteristic cross-
cathodicpeakC1(- 55 mV)wasrecordedonlyinthe
over in the reverse branch of the CV scan will be
1st scan and can be associated with: (1) the very first
observed(Fig. 2a,pointatE = 40 mV);alongwith(3)
stageofdeposition(initiation)ofAgClnuclei[10];(2)
thepresenceofacurrentmaximainthereversedscan canbeduetoincreasedAg?concentration,following
(peak C2 in Fig. 2a), caused by the phase nucleation
the first anodic current increase [40]; (3) and/or can
in the forward scan. The theory and observable
be due to the contribution of an initially smaller size
characteristic features in a CV scan were formulated
7532 J Mater Sci (2018) 53:7527–7550
Fig.2 a Anoverlay of the 1st,2nd, 5th,10th and20thCV scanfor Agin 0.1M HCl;bPDPcurvefor Agin 0.1M HCl.
AgCl nuclei on the Ag substrate. Similar variation in is that AgCl is known to have a limited electron
cathodic peaks potential and currents within AgCl conductivity, together with a limited ionic conduc-
formation on a Ag substrate was reported to be sus- tivity (e.g., for Ag? and Cl- ions, [43]. This will
tained in all scans and linked to the contribution of impede a subsequent growth of AgCl on the Ag
large (e.g., peak C2) or smaller (e.g., peak C1) AgCl surface, once a layer of a certain thickness had
nuclei [10]. However, the mechanism here is obvi- already developed. As shown in Fig. 2a, the current
ously different: a reoccurrence of the peak C1 with at anodic and cathodic peaks increases only initially,
subsequent scans was not recorded (Fig. 2a). With together with a slight shift of peak potentials (this is
scanning both anodic and cathodic peaks currents between the 1st and the 5th scans). With subsequent
initially increased, Fig. 2a, as expected, while the scanning no further increase was observed, main-
cross-over point remained at the identical current– taining the same peaks potential and, as aforemen-
potential location, irrespective of the scan number tioned, a constant position of the current cross-over
(Fig. 2a and inlet). A constancy of the cross-over point (inlet in Fig. 2a). These features would be con-
pointwouldmeantheformationofcrystalsofalarge sistent with a 2D (rather than 3D) growth process
size, while in contrast, a cathodic shift of this point [44], and stabilization of the AgCl layer on the Ag
would be denoted to diffusion-controlled crystal substrate.Theactualmorphologyandmicrostructure
growth and will also imply a decreasing crystal size ofAgCl,grownontheAgsubstrate,willbediscussed
[39]. A cathodic shift of the cross-over point (Fig. 2a) indetailin‘‘Surfacemorphologyandmicrostructure’’
wasnotobserved.Therefore,thepeakC1mostlikely section. The ionic and electron conductivity of the
reflects an increased Ag? concentration during the AgCl layer will be discussed in ‘‘Correlation of sen-
initial Ag dissolution, followed by initiation of AgCl sors’ response, surface properties and ohmic resis-
nuclei formation. The interfacial controlled crystal tance of AgCl’’ section in relation to the
growth of AgCl nuclei of sufficiently large size is morphological observations.
reflected by the shape of the CVs with subsequent
scans, where stabilization in both anodic and catho- PDP test
dic currents was observed between the 5th and 20th
cycles(Fig. 2a),attributabletothecoverageoftheAg The PDP test was performed to determine the range
substrate by a AgCl layer. of suitable current densities for AgCl formation. At
Similar features in a CV response were attributed potentials more noble than the corrosion potential
to a 3D growth of AgCl for Ag/AgCl interfaces in (Ecorr = 20 mV vs. SCE), the PDP curve shows a
variouselectrolytes[30,41,42].However,tobenoted sharp increase in anodic current, Fig. 2b. This
J Mater Sci (2018) 53:7527–7550 7533
increase in current with anodic polarization was Sensors’ response: OCP records
almost potential independent until ca. 150 mV, cor-
respondingtoananodiccurrentmaximumof4 mA/ Background and related general considerations
cm2. Stabilization followed after this point and
Evaluating the sensors’ response to alkaline medium
independence of current from potential with further
isimportantinviewoftheirpracticalapplicationina
polarization was observed. At potential values more
(reinforced) concrete system. In chloride-containing
anodic than 300 mV, a limitation of anodic current
medium the sensors’ OCP will return the chloride
was relevant with a current stabilization at approx.
2 mA/cm2. This limitation in the current density can ions concentration in that medium (in accordance
withthepreviouslyintroducedEqs. 2,3and4).Inthe
be interpreted as the coverage of the Ag surface by a
chloride-free environment, as would be in a concrete
AgCllayer.BasedonthePDPtestresult(Fig. 2b),the
system where chlorides did not yet penetrate the
current density range for AgCl layer formation is
betweenca.0.05and4 mA/cm2,with4 mA/cm2asa matrix, the sensors’ OCP will reflect a mixed poten-
maximum and 0.5–2 mA/cm2 as an average range, tial and a dynamic exchange equilibrium, similar to
Eq. (5), will be relevant. For the former situation, the
where the Ag substrate is supposedly uniformly
OCPisexpectedtoshiftincathodicdirectionandwill
covered by AgCl. Therefore, for the Ag anodization
be determined by the activity of the chloride ions
process in 0.1 M HCl and Ag/AgCl sensor’s prepa-
(e.g., for chloride content in the range of
ration, respectively, four current densities in the
rangeof:0.5–4 mA/cm2wereemployed(regimesA– 250–500 mM, the OCP value of a chloride sensor
DinFig. 3).Thecurrentdensitiesof2and4 mA/cm2 should read between 6 and 23 mV vs. SCE [45]). For
the latter case, chloride-free alkaline medium, the
canbereferredtoashigh current densities, while0.5
and 1 mA/cm2 are considered as average to low OCP values are expected to be more anodic (ca.
100–150 mV vs. SCE, [13, 45, 46]), due to a mixed
current densities.
potential,arisingfromOH-ionsinterference[1].The
The anodization regimes A–D in Fig. 3 are from
levelofOH-ionsinterferencewill,therefore,depend
this point forward used as sensors’ designation, i.e.,
onthechloride ionsconcentration(i.e.,theOH-/Cl-
sensors A, B, C and D are discussed in the following
ratio)andwilldominateintheabsenceofCl-ions.A
sections. Figure 3 also gives the theoretically
detection limit, largely determining a chloride sen-
approximated thickness of the AgCl layer in each
sor’s response, is reported to be 10 mM chloride
regime, together with the experimentally derived
content [45]. Although OH- ions interference cannot
one. The former was calculated using the previously
be excluded in the hereby employed alkaline solu-
introducedEq. (1);thelatterissubjecttopresentation
tionsofpHca.12.8,thechloridecontentinthemodel
and discussion in ‘‘Surface morphology and
media was adjusted to be above the reported detec-
microstructure’’ section.
tionlimitof10 mM.Thiswasalsodoneinviewofthe
objective of this work to elucidate the effect of AgCl
morphology and microstructure on the response of
sensor within chloride ions detection, rather than a
silveroxidecontributiontothesensors’performance.
OCP records
As already introduced, the sensors’ response is the
OCPrecorditself,reflectingthedynamicequilibrium
at the sensor/environment interface. The main
objective here was to determine the accuracy of the
chloride sensors to reflect the chloride content in the
alkaline medium. What should also be noted, is that
the time to reach a stable OCP for each sensor will
determine its sensitivity and stability. Since the
Fig.3 Anodizationregimesforsensorpreparation(1-hduration)
model medium was identical, fluctuations or
andthicknessofthedepositedAgCllayeronthesilversubstrate.
7534 J Mater Sci (2018) 53:7527–7550
unexpectedperformancewouldberelatedtotheAg/ increase in chloride concentration from 20 to
AgCl interface and/or the AgCl properties, 260 mM. This is irrespective of the sensor type and
respectively. anodization regime, respectively.
TheOCPrecordsinchloride-containingCE(20and TheresultsdepictedinFig. 4alsoreflectadifferent
260 mM chloride content) are presented in Fig. 4. As patternofperformanceforthedifferentsensors:OCP
canbeobserved,increasingthechloridecontentfrom stabilizationdependsnotonlyonthechloridecontent
20 (Fig. 4a) to 260 mM (Fig. 4b) results in a cathodic in the medium, but also on the obviously different
shift of the OCP values for all sensors. For sensors A properties of the AgCl layer (which is in addition to
and B, the initial values, e.g., within 60–120 s of the thickness variation, Fig. 3). As shown in Fig. 4a,
immersion,readbetween90 mV(for20 mMchloride approximately 60 s were needed for a stable OCP of
content) and 25 mV (for 260 mM chloride content), sensors A and B, prepared at low current densities
while before stabilization the initial OCP values for (Fig. 3), while more than 1800 s were required for
sensors C and D are between 40 (Fig. 4a) and 15 mV sensorsCandD(withathickerAgCllayer,Fig. 3)to
(Fig. 4b).At the end of the test, the final OCPs for all reach a stable value. This observation was relevant
sensors correspond to the chloride content in the for both 20 and 260 mM chloride concentrations
medium—for 20 mM chloride content, the adopted (Fig. 4a, b). The increase in the AgCl layer thickness
values were between 80 and 89 mV, for 260 mM from* 8 lm(sensorA)to* 15 lm(sensorB)hasa
solution—the final values were between 24 and negligible effect on the OCP stabilization time.
25 mV. This is as expected and reflects the relevant However, the increase in AgCl layer thickness from
chloride content, since theoretically (following Eqs. 2 15 lm for sensor B to 20 lm for sensor C and 40 lm
and 3), at 20 and 260 mM chloride concentration, a for sensor D, has an obvious influence on the OCP
chloride sensor should read 89 and 23 mV versus stabilization time. Consequently, the observed dif-
SCE, respectively [45]. ference (1 min for sensors A and B vs. 30 min for
To be noted is that the OCP records in CE with sensors C and D) should be attributable to a specific
20 mM chloride concentration showed a variation of AgCl layer morphology and microstructure, rather
50 mV at the beginning of the test, narrowing down thantothicknessalone.Tobenotedisthattheabove
to 7 mV after 7200 s (Fig. 4a). In 260 mM chloride- potentialdifferencesanddeviationfromtheexpected
containing medium a smaller deviation was performancewasalsorecordedinthesamerangefor
observed,i.e.,lessthan10 mVatthebeginningofthe replicate specimens (three replicate specimens were
test, decreasing to 1 mV after 7200 s (Fig. 4b). The tested, but results for theseare not included in Fig. 4
results show that both OCP variation and the time for simplicity). For instance, in less than a 100 s,
needed for OCP stabilization are reduced upon an sensor A reached to the stable potentials of 85 ± 1.5
Fig.4 Evolution of OCP values of the Ag/AgCl sensors in chloride-containing cement extract: a CE?20mM chloride content;
bCE?260mM chloride content.
J Mater Sci (2018) 53:7527–7550 7535
and25 ± 0.2 mVinCEwith20and260 mMchloride observed, the morphological features and packing of
concentrations, respectively. In contrast, a relatively the AgCl layer are different in each case. For sensors
stable OCP for the sensor D was observed after A and B (Fig. 5a, b), ‘‘packed-piled’’ AgCl particles
1800 s, reading ca. 78 ± 3 mV for CE with 20 mM were well observed on the Ag substrate. Addition-
chloride concentration and 23 ± 2 mV for CE with ally, the AgCl layer appears to be relatively uniform,
260 mM chloride concentration. presenting rounded, 1–2 lm in (top) size AgCl par-
Additionally, the following observation is impor- ticles, and a distinct boundary between these. Along
tanttobespecificallyaddressed:intheinitialperiod, with clearly observable boundaries, specifically for
thehigherinstabilityofsensorsCandD,comparedto sensor A (Fig. 5a), well defined are the surface
sensors A and B, was obvious (Fig. 4a, b, records openings of micro-channels. For sensor B (Fig. 5b),
prior to 1200 s). The result means that sensors C and these features become slightly distorted, the surface
Dinitiallyreportan‘‘overestimated’’chloridecontent morphology, however, maintaining similar to the
(a more cathodic OCP), while sensors A and B give sensor A appearance.
an almost immediate accurate response. In other In contrast, the AgCl layer for sensors C and D
words, the potentially different microstructural becameamosaicofcomplexpatterns(Fig. 5c,d).The
properties of the Ag/AgCl interface for A and B high current densities, as employed for 1 h in the
versusCandDsensorsshouldberesponsibleforthe production of sensors C and D (2 and 4 mA/cm2),
observeddiscrepancies.Exceptthealreadydiscussed apparently induced the formation of elongated and
considerations, this hypothesis follows the common discontinuous, ‘‘twisted’’ AgCl particles. The ‘‘inner’’
knowledge that limitation of ion transport of any microstructure of the AgCl layer cannot be judged
kindattheinterfacesensor/medium(orlimitationof from surface observations only, but can be clearly
electron transport along the sensors’ conductive sur- visualized on a cross section of the Ag/AgCl inter-
face) will be reflected in variations in the time to face. Figure 6 presents the cross sections for sensors
establishanequilibriumcondition.Theseaspectswill A and D, together with relevant EDS analysis in
be discussed in ‘‘Correlation of sensors’ response, depth of the sections, toward the Ag substrate (the
surface properties and ohmic resistance of AgCl’’ resultsforsensorsAandBweresimilar,aswellasfor
section in correlation to the experimental results on sensorsCandD,micrographsofthecrosssectionfor
microstructure of the Ag/AgCl interface. Surface B and C sensors are, therefore, not presented). The
chemistry, although supposed to be similar, might EDS analysis in Fig. 6 for the interfaces in sensors A
also play a role in the sensor’s response. Therefore, and D confirms both Ag and Cl in the AgCl layer,
thenextstepinthisworkisanin-depthinvestigation while only Ag at the Ag substrate, hence confirming
of the surface morphology, microstructure and sur- the expected qualitative outcome for the Ag/AgCl
face chemistry of all sensors, which are discussed in interface.
the following sections. For sensor A (Fig. 6a), a visuallywelladhered and
homogeneous AgCllayer can beobserved. Thegrain
Surface morphology and microstructure boundaries in depth of the layer can be well seen,
showing a well packed and almost perpendicular
In this section, electron microscopy (Figs. 5, 6 and 7) orientation toward the Ag substrate. This upright
andEDSanalysis(Fig. 6)arediscussedforallsensors particles’orientation would account foracontinuous
(A, B, C and D), produced for identical anodization micro-channel network in depth of the layer. The
time of 1 h, but at different current densities (Fig. 3). contribution of such a morphology and microstruc-
Additionally, electron microscopy of the Ag/AgCl ture to the performance of the sensor would be
interface for D-type sensors, anodized at the highest reflectedinarapidresponsetotheenvironment.This
current density of 4 mA/cm2 and varying anodiza- is because a (vertically) open, continuous network of
tion time, is also presented (Fig. 7) as supportive micro-channels would account for reduced limita-
evidence for the discussion in this and the next tions to ion transport. This was as actually observed,
sections. OCP records’’ section and Fig. 4, where an almost
Figure 5 depicts ESEM micrographs from the top instant accurate response of sensors A and B to the
surface of the sensors, visualizing the AgCl layers, pre-defined chloride content was recorded, together
obtained in each anodization regime. As can be with a fast OCP stabilization. These microstructural
7536 J Mater Sci (2018) 53:7527–7550
Fig.5 ESEMimagesoftheAgCllayeronthesurfaceofsensorsAat0.5mA/cm2(a),Bat1mA/cm2(b),Cat2mA/cm2(c)andDat
4mA/cm2(d).
Fig.6 CrosssectionandEDSspectraoftheAg/AgClinterfacein is chosen to allow visibility of the bi-layer structure in sensor D,
sensorAat0.5mA/cm2(a)andsensorDat4 mA/cm2(b).Note together with the fullAgCl layer thickness.
Agsubstratein(b)isnotseenatthis92500magnification,which