Surface & Coatings Technology 259 (2014) 560–569

Contents lists available at ScienceDirect

Surface & Coatings Technology

j ourna l homepage: www.e lsev ie r .com/ locate /sur fcoat

Nickel-nitride-coated nickel foam as a counter electrode fordye-sensitized solar cells

Sun Ha Park a,b,1, Yong-Hun Cho c,1, Myounggeun Choi c, Hyelim Choi c, Jin Soo Kang a,b, Ji Hyun Um a,b,Jung-Woo Choi a,b, Heeman Choe c,⁎, Yung-Eun Sung a,b,⁎⁎a Center for Nanoparticle Research, Institute for Basic Science (IBS), Seoul 151-742, South Koreab School of Chemical & Biological Engineering, Seoul National University, Seoul 151-744, South Koreac School of Advanced Materials Engineering, Kookmin University, 861-1 Jeongneung-dong, Seongbuk-gu, Seoul 136-702, South Korea

⁎ Corresponding author. Tel.: +82 2 910 5020; fax: +8⁎⁎ Correspondence to: Y.-E. Sung, Center for NanopartiScience (IBS), Seoul 151-742, South Korea. Tel.: +82 2 88

E-mail addresses: heeman@kookmin.ac.kr (H. Choe), y1 These authors contributed equally to this work.

http://dx.doi.org/10.1016/j.surfcoat.2014.10.0300257-8972/© 2014 Elsevier B.V. All rights reserved.

a b s t r a c t

a r t i c l e i n f o

Article history:Received 27 May 2014Accepted in revised form 13 October 2014Available online 20 October 2014

Keywords:CoatingFoamNickelNitrideDSSCCounter electrode

This study employs for the first time surface-nitrided Ni foam as a counter-electrode free of Pt and transparentconducting oxides (TCOs) to realize a cost-effective counter electrode for dye-sensitized solar cells (DSSCs).This electrode simultaneously features high catalytic activity for triiodide reduction and high conductivity in asingle layer. The nitrided Ni foam is synthesized by nitridation treatment of open-cell Ni foam in an ammonia at-mosphere. This electrode presents high catalytic activity on the nitrided surface and easy electron transport abil-ity in the three-dimensional, interconnectedmetallic structure. This study provides a preliminary design conceptfor utilizing the nitrided Ni foam as a promising cost-effective counter electrode that does not require expensivePt and TCO.

© 2014 Elsevier B.V. All rights reserved.

1. Introduction

Solar cells have been drawing attention as advanced alternative en-ergy devices to generate energy without the emission of carbon dioxide[1]. Dye-sensitized solar cells (DSSCs) have received great interest fromboth academia and industry, because they are reported to have photo-voltaic efficiencies above 12% [2–4]. DSSCs usually consist of dye-adsorbed, mesoporous titanium dioxide (TiO2), an electrolyte contain-ing redox couples, and a counter electrode [5]. The typically usedcounter-electrode consists of platinum (Pt) catalyst deposited on glasscoated with transparent conducting oxide (TCO), which performs twocritical functions: it transfers electrons from an external circuit totriiodide and reduces triiodide ions in the electrolyte [6–8]. Many recentstudies have focused on finding inexpensive alternative materials withhigh electrocatalytic activity as a replacement for Pt, because Pt elec-trodes in DSSCs have several problems such as high material cost(more than ~8% of the total device cost [9]) and poor stability in corro-sive electrolytes despite Pt's exceptionally good catalytic activity [7,10].Various carbon-basedmaterials have receivedmuch attention as prom-ising alternatives to Pt and many research groups have reported high

2 2 910 4320.cle Research, Institute for Basic0 1889; fax: +82 2 888 1604.sung@snu.ac.kr (Y.-E. Sung).

photovoltaic performance of 6.6–9.7% efficiency by using various carbonmaterials such as graphite, active carbon, and carbon nanotubes [8,11,12]. Conducting polymers have also been used effectively as counterelectrodes in various redox electrolytes [13–15]. In addition, other ap-proaches involving applying new inorganic materials such asmetal car-bide, metal sulfide, and metal nitride in counter electrodes have showngood photovoltaic performance (2.7–8.3%) [16–18].

Glass coated with TCO (TCO-coated glass or TCO glass) that is used asa substrate for counter electrodes also has a drawback of cost, owing tothe expensive fabrication process. The TCO glasses commonly used inelectrodes are In-doped SnO2 (ITO) or F-doped SnO2 (FTO). Since theTCO substrate is responsible for more than 40% of the total device cost,DSSCs that use TCO-free counter electrodes are of great interest for reduc-tion in the production cost [19]. Pt-like active material coated onto metalfoil, such as Ti or stainless steel foil, is one of themostwidely investigatedTCO-free electrodes because of its low cost, superior conductivity, andfacile synthesis [20,21]. Several attempts have been made to constructcost-effectiveDSSCs using Pt- and TCO-free counter electrodes. For exam-ple, DSSCswith submicron-graphite counter electrodes have achieved 6%cell efficiency [19]. Metal carbide filmwith conductive carbon has shown6.6% efficiency [22]. The research groups of Park and Sung utilizedpoly(3,4-ethylenedioxythiophene) (PEDOT) and camphorsulfonic-acid-doped polyaniline (CSA-doped PANI) film as a TCO- and Pt-free counterelectrode for DSSCs, respectively, and demonstrated comparable photo-voltaic performance to DSSCs with a platinized TCO counter electrode[23,24].

561S.H. Park et al. / Surface & Coatings Technology 259 (2014) 560–569

Many transition metal nitrides such as MoN, Ni3N, TiN, and W2Nhave been reported previously as highly efficient counter electrodesfor DSSCs [17,25–28]. Despite an earlierwork by Jiang et al.,who appliedthe surface-nitrided Ni particle film (approximately 200 nmNi particlesonto the surface of themetallic Ni film 200 μm in thickness) as a Pt- andTCO-free counter electrode [25], this paper reports the first usage of ni-trided nickel foam as a three-dimensional porous Pt- and TCO-freecounter electrode for DSSCs. TheNi foamused in this study is amore ad-vanced three-dimensional porous structure than a structure of particles,because it is comprised of three-dimensionally interconnected metallicstruts with pores. As such, it exhibits comparable conductivity to that ofa metal foil, allowing metal foam to act as a good electron-transportingsubstrate from an external circuit to the redox electrolyte when com-pared to TCO. More importantly, the regularly-spaced open-pore struc-tures in themetal foam can provide a high specific surface area to redoxcouples in the electrolyte and reduce mass transport limitations of theelectrolyte in the electrode. To obtain a bi-functional structured elec-trode that features good electrocatalytic activity and electrical conduc-tivity, nitrided Ni foam was prepared through nitridation treatment inan ammonia atmosphere. The resulting nitrided Ni foam was observedto have high conductivity in the bulk Ni foamand great catalytic activityon the nitrided surface. The transitionmetal nitrides in the nitrided sur-face were able to show high electrocatalytic activity because of theirnoble-metal-like electronic structures to confirm that the nitrided Nifoam can be used as a new effective counter-electrode withoutemploying expensive Pt and TCO (Fig. 1).

The new DSSC assembled using the nitrided Ni foam in this studyexhibits a conversion efficiency approaching ~5%, which is compara-ble to that of the Pt-coated TCO electrode [29,30] and is consideredan excellent efficiency, given that this was measured with gel elec-trolyte and the Ni metal foam used in this preliminary study wasfar from being optimized in terms of its pore size range, becausethe Ni foam used in this study had relatively large pores on theorder of several hundred microns while smaller pores might providebetter results. This study thus illustrates the great potential for effi-cient and low-cost Pt- and TCO-free DSSC counter-electrodes onthe basis of an open-cell metal-foam-based configuration. In addi-tion, this study can be considered as a framework for designing ad-vanced DSSCs using an optimized metal-foam counter-electrode,because the experimental methods and insights gained here mayalso apply to other types of metallic-foam counter electrodes withfurther modified microstructural features, e.g., Ni or Ti foams withsmaller pore sizes and larger specific surface area [31–34].

Fig. 1. Schematic diagrams of a DSSC: (a) DSSC with the typical counter-electrode (Pt and FTO

2. Material and methods

2.1. Preparation of nitrided Ni metal foam

The Ni foam (99.9% pure, Metal Foam Korea, Republic of Korea) witha thickness of 500 μmandNi foil (99.9% pure, Sigma-Aldrich, USA)with athickness of 125 μm were prepared and compared for use as DSSC'scounter electrodes. The Ni foam with hollow strut walls was fabricated(Metal Foam Korea) through a Ni plating method with removable poly-mer foam as a precursor, which is schematically shown in Fig. 2. The pre-cursor polymer foamwas eventually removed thermally, leaving hollowstrut structures in the final product of Ni foam (110 ppi, Fig. 3(a)). The Nifoamwas immersed in a 0.1MaqueousNaOH solution anddried to elim-inate impurities from the foam surface. After the cleaning process, the Nifoam was chemically etched with sulfuric acid (H2SO4) or nitric acid(HNO3). 10 wt.% sulfuric acid or nitric acid was dissolved in distilledwaterwhere the temperaturewaswellmaintained; nitric acid treatmentwas carried out at room temperature whereas sulfuric acid treatmentwas carried out at 60 °C. The cleaned Ni foam was then immersed inthe prepared sulfuric or nitric solution for 20 min, which was stirredusing a glass rod to remove bubbles from the foam surface. The preparedNi foamwas nitrided in a box furnace at 450 °C for 2 h, in a flowing pureammonia atmosphere with a gas flow rate of 100 sccm. The nitrided Nifoam was then heated in a tubular furnace with an Ar/5% H2 mixturegas at 230 °C for 24 h to reduce the constituent metal oxides. The pre-pared nitrided Ni foam electrode was finally attached to a TCO-freeglass substrate using Surlyn (50 μm, Dupont, USA).

2.2. Electrode assembly for DSSCs

Themicroscale TiO2 filmwith an active area of 0.2 cm2was preparedby depositing a viscous TiO2 paste (Ti-Nanoxide D/SP, Solaronix Co.,Switzerland) onto the FTO-coated glass (FTO glass; 8Ω cm−2, PilkingtonTEC glass™, USA). TheTiO2 electrodewas immersed in 0.5mMN719dye(Ru[LL0-(NCS)2], L = 2,20-bipyridyl-4,40-dicarboxylic acid, L′ = 2,2′-bipyridyl-4,40-ditetrabutylammonium carboxylate, Dyesol, Australia)for 24 h. This dye-adsorbed TiO2 electrode was prepared for use as theworking electrode and the nitrided Ni foam for use as the counter elec-trode. The two electrodes were assembled with electrolyte into asandwich-type cell using thermal adhesive films (Surlyn: 50 μm,Dupont, USA). The gel-type electrolyte was obtained from Dyesol,Australia and was used as-received.

glass) and (b) DSSC with the Pt- and TCO-free counter-electrode with nitrided Ni foam.

Fig. 2. Schematic of the processing route of the open-cell Ni foam used in this study. Ni is plated onto the polymer precursor foam, which is removed thermally in the later stage of pro-cessing to result in the Ni foam with hollow strut walls.

562 S.H. Park et al. / Surface & Coatings Technology 259 (2014) 560–569

2.3. Characterization and photoelectrochemical measurements

The thickness and morphology of the nitrided Ni foam were con-firmed by field emission-scanning electron microscopy (FE-SEM;SUPRA 55VP, Carl Zeiss, Germany) at an acceleration voltage of 2 kV.Atomic force microscopy (AFM; Innova SPM, Veeco, USA) in tappingmode was also used to examine the surface morphology of the nitridedNi foam. An X-ray photoelectron spectrometer (XPS; Thermo VG Scien-tific, USA)was used to determine the chemical composition of theNi ni-tride film after 300 s of etching using an Al Kα source. The results werecalibrated by referencing C1s at 285 eV and fitted using Avantage

10 µm

500 nm

(a)

(b)

10.94 µm

Fig. 3. SEM images of structure of the Ni foam electrode: (a) enlarged cross-sectionalimage of the nitrided Ni foam and (b) surface image of the nitrided Ni foam.

Software. The phase and structural changes in the films were examinedby X-ray diffraction (XRD; D-MAX2500-PC, Rigaku, Japan) with Cu-Kαradiation (λ = 1.54178 Å).

Electrochemical tests, including three-electrode cyclic voltammetry(CV) and a dummy cell test, were performed using a potentiostat/galvanostat (PGSTAT30, AUTOLAB PGSTAT30, The Netherlands) withan electrolyte of dissolved iodide redox couple in acetonitrile solutionof 10 mM LiI, 1 mM I2, and 0.1 M LiClO4 at a scan rate of 50 mV s−1.The in-situ optical transmittance change during the CV measurementwas observed using a He–Ne laser (λ = 633 nm) to identify and com-pare the electrochromic properties of the nitrided Ni foam sampleswith andwithout etching. The photocurrent density–voltage (j–V) char-acteristics of the assembled DSSCs and current transient measurementwere evaluated using a 500-W xenon isotope lamp (XIL; model05A50KS source units, SERIC Ltd., Japan) with a light intensity of 1 sun(100mWcm−2), whichwas adjusted using a Si reference solar-cell fab-ricated at the National Institute of Advanced Industrial Science andTechnology (AIST) in Japan. The j–V characteristic test was carried outunder front and rear illuminations. The incident photon-to-current effi-ciency (IPCE; the QEX10 solar cell quantum efficiency measurementsystem, PV Measurements Inc., USA) was measured from 300 to800nmunder short-circuit conditions. The electrical impedance spectra(EIS) were measured using an impedance analyzer (IM6, ZAHNER-Elecktrik GmbH & Co. KG, Germany) at a frequency ranging from 0.5to 105 Hz. The Nyquist plot of EIS was analyzed using an equivalent cir-cuit model presented in the literature [11,35].

3. Results and discussion

A comparison of field-emission scanning electron microscopy (FE-SEM) images of the nitrided Ni foam and foil is shown in Fig. 4. Thethicknesses of the nitrided Ni foil and foam are 117 and 458 μm, respec-tively. While the nitrided Ni foil has the typical dense structure withsmooth surface, the Ni foam has a three-dimensional porous structure.In particular, the Ni foam shows interconnected structures consistingof small and large open pores with a compact and uniform surface[35]. In addition, a triangle-shaped hollow space is observed inside theinterconnected structures, with ca. 10 μm wall thickness, as depictedin FE-SEM images of Fig. 3(a). The interconnected porous structure con-sists of small Ni struts, as shown in the magnified FE-SEM surface im-ages of Fig. 3(b). The density of the Ni foam was measured by physicalmeasurement, which was obtained as the ratio of the mass to volumeof a rectangular cuboid specimen. The Ni foam has a density of1.09 g cm−3, indicating a porous structure with 12.3 ± 0.6% relativedensity compared to the theoretical density of Ni foil (8.83 g cm−3),which was calculated by X-ray diffraction (XRD, PDF #01-089-7128).This suggests that the Ni foam has a highly porous structure with

Fig. 4. Comparison of FE-SEM images: the surface images of (a) the nitrided Ni foil and (b) nitrided Ni foam; cross-sectional images of (c) the nitrided Ni foil and (d) nitrided Ni foam.

Fig. 5. Comparison of XRD patterns of the Ni foil, nitrided Ni foil, Ni foam, and nitrided Nifoam.

563S.H. Park et al. / Surface & Coatings Technology 259 (2014) 560–569

87.7% porosity, which is expected to enhance direct contact with theelectrolyte and improve mass transport in the electrode [36–38].

The XRD patterns of the Ni foil, Ni foam, and their nitrided counter-parts are shown in Fig. 5. All of the diffraction peaks of the Ni foil and Nifoam after nitridation can be indexed to face-centered-cubic (f.c.c.) Ni(PDF #00-004-0850). A preferential orientation of Ni (200) at a 2θ of52° is obtained in the Ni foil and nitrided Ni foil, whereas a strongpeak at a 2θ of 44° is observed in the Ni foam and nitrided Ni foam. Al-though the XRD patterns of the Ni foams are different from those of theNi foil, they are similar to those of Ni particles [25]. The Ni foam com-prises interconnectedNi strutswell-oriented to the (111) plane, despitetheNi foil showingorientation to the (200) plane. This orientation of theelectrode in XRDpatterns ismaintained after the nitridation process. Onthe other hand, very small peaks around 2θ values of 42° and 52° corre-sponding to the peaks of NiO were detected in the nitrided Ni foam,which suggests that a small amount of oxidation of the Ni foil andfoam also occurred during nitridation at 450 °C in an ammonia atmo-sphere. Although a rather porous NiO oxide layer generally developson the surface of pure Ni, the amount of NiO formed at a relativelylow temperature of 450 °C for such a short time of 2 h in an ammoniaatmosphere is expected to be minimum, and hence, it is not seen inthe images in Fig. 4 [39].

The nitridation and oxidation that occurred in the Ni foil and foamwere analyzed by X-ray photoelectron spectroscope (XPS) with Ar ionsputtering, as demonstrated in Fig. 6(c). The binding energy of the oxi-dized Ni state, NiO, on the surface is observed at 855.0 or 854.3 eV,which gradually disappears with increasing Ar ion sputtering time. Asshown in the XPS spectrum of the surface (Fig. 6), a larger amount ofNi nitride or metallic Ni is detected on the surface of the nitrided Nifoam, but a relatively smaller amount is observed on the surface of the ni-trided Ni foil. In contrast, the oxidized Ni, NiO, is mainly observed in thenitrided Ni foil electrode. This suggests that the Ni foam was able to benitrided on a very fine scale because of its porous structure and large sur-face area [25]. As shown in the XPS depth profiles in Fig. 6(a) and (b), therelative oxygen and nitrogen atomic concentrations are dramatically

(a) (b)

(c)

Inte

nsit

y (a

.u.)

Inte

nsit

y (a

.u.)

Binding energy (eV)860 859 858 857 856 855 854 853 852 851 850

Inte

nsit

y (a

.u.)

Binding energy (eV)

Nitrided Ni foil

Nitrided Ni foam

NiO

NiO

NiO

Ni3N Ni NiO

NiO

Ni3N

860 859 858 857 856 855 854 853 852 851 850

Ni(OH)2

Binding energy (eV)

860 859 858 857 856 855 854 853 852 851 850

Fig. 6. Ni2p XPS depth profile of the nitrided Ni foil and nitrided Ni foam with Ar ion sputtering: Ni2p XPS depth profiles of (a) the nitrided Ni foil and (b) nitrided Ni foam, and(c) comparison of Ni2p XPS spectra of the nitrided Ni foil and foam at the surface.

564 S.H. Park et al. / Surface & Coatings Technology 259 (2014) 560–569

decreased, and the metallic Ni state is clearly increased in both of the ni-trided Ni electrodes with increasing Ar ion sputtering time. The bindingenergy of 853.6 eV indicates the presence of Ni nitride in the bulk ofthe nitrided Ni foam, that of 852.3 eV indicates the presence of metallicNi, and that of 853.8 eV indicates the presence of Ni nitride in the bulkof the nitrided Ni foil [40].

The electrical resistance of the film plays a significant role in the per-formance of the electrode, which can also be affected by its bulk state.Indeed, the resistances of the nitrided electrodes differ. For example,the four-point probe measurement reveals a value of 3.3 mΩ cm2 forthe nitrided Ni foam electrode, which is only slightly larger than thatof the Ni foil without nitridation (2.9 mΩ cm2); therefore, the metallicNi in the bulk state is believed to provide good electron transfer abilityto the surface-nitrided Ni foam, which makes it a promising alternativeto the TCO substrate.

To increase the surface roughness of theNimetal foamelectrode, theNi foam was chemically etched with nitric acid or sulfuric acid becausethey are both effective pickling reagents and strong oxidizers [41]. Etch-ing the nitrided Ni foams with HNO3 and H2SO4 created a rougher sur-face compared to the nitrided Ni foil electrode [35–41]. A roughnesscomparison of the nitrided Ni foam and two etched nitrided Ni foams,Ni foam (H2SO4) and Ni foam (HNO3), is shown in the atomic force mi-croscope (AFM) images of Fig. 7(a)–(c) in comparison with roughnessof the Ni foil (Fig. 7(d)). The root mean square (RMS) roughnesses ofthe Ni foam, Ni foam (H2SO4), and Ni foam (HNO3) are respectively20.1, 45.7, 43.9 nm, all of which are significantly higher than that ofthe Ni foil (9.6 nm). The increased surface area is expected to influencethe electrocatalytic activity of the triiodide redox reaction and increase

the current density in solar-cell devices [41–43]. In addition, the wallthicknesses of the Ni foams were decreased with chemical etching be-cause the acids could dissolve the Ni metal easily. Although the thick-ness of the Ni foam strut decreased by acid treatment from ~19 μm to~14 μm as shown in Fig. 8, the triangle-shaped hollow space and thetotal thickness of the Ni foam electrode remained unchanged. Despiteno change in the entire sample thickness, therewasmass decreasemea-sured for both the Ni foams etchedwith nitric or sulfuric acid: from0.76to 0.62 g for the Ni foam etched with nitric acid and from 0.79 to 0.77 gfor the Ni foam etched with sulfuric acid. Moreover, these changes inmass imply that there were 2.5% and 0.4% increases in porosity for theNi foam samples after being etched with nitric and sulfuric acids,respectively.

As shown in Fig. 9(a), the XRD patterns of the Ni foams were notchanged after etched with acids. In particular, the preferential orienta-tion of Ni (111) at a 2θ of 44° was also observed in the nitrided Nifoam regardless of acid treatment. The binding energy of the nitridedNi foam electrode was slightly shifted to a higher value in the XPS anal-ysis of Fig. 9(b). As shown in the XPS spectrum of Ni2p, the binding en-ergy for metallic Ni was 852.7 eV and the Ni nitride peak also appearedfor the nitridedNi foamwithout acid treatment. In addition, the nitridedNi foams treated with sulfuric acid and nitric acid had binding energiesat 853.06 eV and 853.02 eV, respectively, which is attributed to the Nifoam being finely nitrided and oxidized owing to the increased surfacearea from chemical etching [25].

A comparison of electrocatalytic activity for triiodide reduction isshown in the j–V curves for the I−/I3− redox of the three different nitrid-ed Ni foams in Fig. 10(a). According to the j–V curves in Fig. 10(a), the

500 nm

500 nm 500 nm

Nitrided Ni (foam)

Nitrided Ni (HNO3)

500 nm

Nitrided Ni (H2SO4)

Nitrided Ni foil

(a) (b)

(c) (d)

Fig. 7. Comparison of AFM images of the three different nitrided Ni foams and nitrided Ni foil: (a) the Ni foam, (b) Ni foam (H2SO4, 20 min), (c) Ni foam (HNO3, 20 min) and (d) Ni foil.

565S.H. Park et al. / Surface & Coatings Technology 259 (2014) 560–569

nitrided Ni foam electrodes exhibit only cathodic peaks, whereas the Ptelectrode generally exhibits both a cathodic peak and an anodic peak[44]. Furthermore, the acid treatment in the Ni foam greatly influencedthe electrochemical activity of the iodide reduction. The reduction-peakcurrent densities of the Ni foam (HNO3) (−2.52mA cm−2) and Ni foam(H2SO4) (−2.45 mA cm−2) are significantly higher than that of the Nifoam (−1.6 mA cm−2). Fig. 10(b) shows the linear relationship

2 μm

(a)

2 μm

(b)

Fig. 8. Change in thickness with acid treatment: SEM cross-sectional images of the (a) Nifoam and (b) Ni foam (HNO3, 20 min).

between the scan rate and cathodic peak current. The adsorption ofthe iodide species was slightly affected by the redox reaction on theelectrode surface and the major rate-determining step of the reactionwas the diffusion process [10,45]. In addition, the lower slope of the ni-tridedNi foamowing to the faster redox reactionwas altered by the acidtreatment; in particular, the Ni foam (HNO3) electrode showed thehighest slope among them [45]. The nitrided Ni foam (HNO3) thus hassuperior electrocatalytic activity for triiodide reduction.

The stability of the nitrided Ni foam electrode was measured by cy-clic voltammetry (CV) cycling tests. Iodine typically reacts with metalduring corrosion, which causes iodine depletion in the electrolyte [10].Although the loss of iodine by metal corrosion typically changes thecolor of the electrolyte from yellow to transparent, the electrolyteremained yellow throughout the CV cycling measurements; further-more, the absorbance also did not show any change in color after 100CV cycles [23,46–48]. Only a peak at 360 nm, contributed by the iodideanion, was observed in the UV–vis absorbance spectra of Fig. 11(a),which suggests that the nitrided Ni foams were not degraded by anycorrosion of themetal substrate in the iodide electrolyte because the ni-trided Ni layer on the surface functioned as a protective layer for themetallic Ni foam [23]. As shown in Fig. 11(b), the cathodic current den-sity only slightly changed after 100 CV cycles from−0.8 V to+0.5 V (vs.Ag/AgCl (sat. KCl)), which implies that the redox reaction of iodine waswell maintained in the nitrided Ni foam electrode after long electro-chemical tests. Therefore, the nitrided Ni foam demonstrated good re-versibility and stability as an electrode in the iodide redox couple.

Fig. 12 shows a comparison of DSSC performance of the counter elec-trodes with Pt-coated ITO glass, nitrided Ni foil, and nitrided Ni foam.Table 1 also lists the performance of DSSCs with these counter elec-trodes. The conventional DSSC system consisting of a photoanode,N719 dye-adsorbed 10 μm anatase TiO2 film on the FTO substrate, and

(a) (b)

Fig. 9. Comparisons of (a) XRD patterns, and (b) Ni2p XPS spectra of the nitrided Ni foam, nitrided Ni foam etchedwith H2SO4 (20min) and nitrided Ni foam etchedwith HNO3 (20 min).

566 S.H. Park et al. / Surface & Coatings Technology 259 (2014) 560–569

gel-type iodide redox couple,was tested under 1-sun front-illuminationconditions. The nitrided Ni foil counter-electrode shows a poor conver-sion efficiency of 1.20% owing to its low electrocatalytic activity for thereduction of triiodide. These low conversion efficiency and fill factor

(a)

(b)

Fig. 10. Comparison of redox performances of the three different nitrided Ni foams underiodide redox electrolyte: (a) j–V curves at 50 mV s−1 and (b) relationship between (scanrate)−1/2 (v−1/2) and cathodic peak current density (jpc).

(FF) are attributed to the nonporous texture and hence, low density ofactive surface area. On the other hand, the photovoltaic performanceof the nitrided Ni foam counter-electrode is clearly enhanced, becauseof its much higher active surface area from its three-dimensionally

-j pc

(mA

cm-2

)

Cycle number

Nitrided Ni foam (HNO3)

Nitrided Ni foam

10

8

6

4

2

0

(b)

(a)

Abs

orba

nce

(a.u

.)

Wavelength (nm)

0 20 40 60 80 100

300 400 500 600 700 800 900

Before CV

After CV

Nitrided Ni foam (H2SO4)

Fig. 11. Comparison of redox performances of the three different nitrided Ni foams underiodide redox electrolyte: (a) comparison of UV–vis spectra of electrolyte before and afterCV measurement and (b) changes in the cathodic peak current density (jpc) during 100CV cycles.

(a)

(c)

(b)

0 10 20 30 40 50 60

0.0 -0.1 -0.2 -0.3 -0.4 -0.5 -0.6 -0.7 -0.8

12

10

8

6

4

2

0

1.0

0.8

0.6

0.4

0.2

0.0

9 10 11 12 13 14 15

Fig. 12. Comparison of DSSC performance of the nitrided Ni foam and nitrided Ni foil counter electrodes under iodide redox electrolyte: (a) current density–voltage curves of the DSSCs,(b) plots of the current transient measurement, and (c) Nyquist plots of the DSSCs from EIS curves.

567S.H. Park et al. / Surface & Coatings Technology 259 (2014) 560–569

interconnected structure of Ni struts [25]. The short-circuit current den-sity (jsc) of the DSSCs indeed changed from 3.3 mA cm−2 to9.6 mA cm−2 and the FF slightly increased from 0.59 to 0.60. The nitrid-ed Ni foam counter-electrode also exhibited higher catalytic activity inthe electrochemical impedance spectroscopy (EIS) curves (Fig. 12(c))compared to that of the nitrided Ni foil counter-electrode. The semicir-cle in the higher frequency region represents the charge-transfer pro-cess of the electrolyte/electrode interface, whereas that in the lowerfrequency region represents the Nernst diffusion impedance of I−/I3−

redox species within a thin layer in the electrolyte [25]. The nitridedNi foam counter-electrode showed lower charge transfer resistance(Rct) in the range of high frequency in the EIS. On the other hand, thesurface charge-transfer resistance of the nitride Ni foil electrode washigher, indicating the poorer electrocatalytic activity. The diffusion re-sistance (Ws) values appear to decrease in the nitrided Ni foamcounter-electrode due to the three-dimensionally interconnected po-rous structures. Higher catalytic activity and easy electrolyte diffusionwere demonstrated when the Ni foam electrode was used as a DSSCcounter-electrode. Furthermore, mass transport limitations of the

Table 1Summary of the j–V characteristics of the dye-sensitized solar cells (DSSCs) with the ni-trided Ni foam, nitrided Ni foil, and Pt counter electrodes.

Material Voc/V jsc/mA cm−2 FF η/%

Nitrided Ni foam −0.67 9.62 0.60 3.88Nitrided Ni foil −0.62 3.25 0.59 1.20Pt −0.74 11.11 0.67 5.57

The dimension of the TiO2 electrode was 0.20 cm2.

electrolyte were influenced considering that both the nitrided Ni foamand foil were used in the same form of thick film. As shown in the cur-rent transient test (Fig. 12(b)), the thickdimension of thenitridedNi foilcaused a dramatic decrease in initial photocurrent density generated inthe cell, whereas the nitridedNi foam could sustain the initial photocur-rent density because the diffusion limitation of the photocurrent wasreduced owing to its three-dimensionally interconnected porousmorphology [49].

In an attempt to further improve the DSSC performance, the nitridedNi foam counter-electrode was chemically etched and also used as acounter electrode, as illustrated in Fig. 13 and Table 2. The short-circuit current density of the DSSCs was particularly increased withchemical etching of the Ni foam electrode, from 9.6 (unetched nitridedNi foam) to 10.7 (nitrided Ni foam etched with HNO3) and 11.3 (nitrid-ed Ni foam etched with H2SO4) mA cm−2. This is attributed to therougher surface created by chemical etching, which provides a largerspecific surface area to the iodide redox couple in the electrolyte. Thedecline in Rct in EIS of Fig. 13(b) also indicates the improvement in io-dide reduction owing to the surface roughened by chemical etching. Inaddition, the sum of all resistances of the DSSC with the nitrided Nifoamwas lower than that of the DSSC with Pt-coated FTO glass becauseof the inherently lower resistance of the metallic Ni foam electrode;therefore, this low resistance benefited the FF and cell conversion effi-ciency of the DSSC device with the nitrided Ni foam.

To obtain efficient light harvesting in DSSC devices, the use of scat-tering layers has been attempted previously [50,51]. Since thecounter-electrode substrate can act as a back reflector, theNimetal sub-strate might also increase the light harvesting in a DSSC due to its highreflectance property [23,50]. As shown in Fig. 13(d), the reflectance of

(a) (b)

(c) (d)

Fig. 13. Comparison of DSSC performance of the three different nitridedNi foams and Pt counter electrodes: (a) current density–voltage curves of the DSSCs, (b) Nyquist plots of the DSSCsfrom EIS curves, (c) IPCE spectra of the DSSCs, and (d) %R spectra of nitrided Ni foams.

568 S.H. Park et al. / Surface & Coatings Technology 259 (2014) 560–569

the Ni foam is indeed larger than that of the Pt-coated TCO electrode, al-though the reflectancewas decreased by nitridation treatment owing tothe formation of Ni nitride on the surface. After nitridation under an am-monia atmosphere, both the color and %R spectra of the Ni foam werechanged by acid treatment; additionally, the two different types ofchemically-etched nitrided Ni foam, Ni foam (H2SO4) and Ni foam(HNO3), showed different %R spectra. While the Ni foam electrodeetchedwith H2SO4 exhibited amaximum reflectance at the wavelengthof ~550 nm, theNi foamelectrode etchedwithHNO3 showed the lowestreflectance at ~500 nm.Given that thewavelength of 510–550 nm is thetypical range for themaximum absorption value of the Ru dye used as asensitizer of DSSCs, the Ni foam counter electrode etched with H2SO4

appears to have achieved the enhanced light harvesting in the DSSCdue to the high reflectance at ~550 nm, which is also in agreementwith the tendency observed in the IPCE spectra of Fig. 13(c) [15].

Among the three different Ni foam counter-electrodes used in thisstudy, the nitridedNi foametchedwith HNO3 shows the best DSSC elec-trode performance by achieving a jsc, Voc, FF, and cell efficiency of10.7mA cm−2, 0.69 V, 0.64, and 4.73%, respectively. Taking into accountthe reference DSSC cell with a Pt-coated TCO electrode displaying a cellefficiency of 5.57%, the 4.73% cell efficiency of the nitrided Ni foamcounter-electrode etchedwith HNO3 is considered excellent, particular-ly because this was accomplished by using a foam pore structure farfrom being optimized and by not using the costly Pt and TCO materialsin the counter electrode. This first investigation on the DSSC cell

Table 2Summary of the j–V characteristics of the dye-sensitized solar cells (DSSCs) with the threedifferent nitrided Ni foam and Pt counter electrodes.

Material Voc/V jsc/mA cm−2 FF η/%

Ni foam 0.67 9.62 0.60 3.88Ni foam (H2SO4) 0.66 11.3 0.59 4.45Ni foam (HNO3) 0.69 10.7 0.64 4.73Pt 0.74 11.1 0.67 5.57

The dimension of the TiO2 electrode was 0.20 cm2.

assembled with the nitrided Ni foam counter-electrode indeed has aplenty of room for further improvement such as using Ni foams with amuch finer pore size distribution. This study can be regarded as a pre-liminary framework for designing advanced DSSCs utilizing three-dimensionally structured metallic-foam-based counter-electrode withfurther modifications. Based on this observation, further improved ni-trided Ni foams can be considered promising for cost-effective Pt- andTCO-free DSSC counter-electrodes.

4. Conclusions

The nitrided Ni foam electrode was applied as a Pt- and TCO-freecounter electrode to realize a new cost-effective counter electrode forDSSCs. The nitrided layer on the Ni foam demonstrated good electrocat-alytic activity in the iodide redox electrolyte, due to its noble-metal-likeelectronic structures. Since this nitrided Ni foam consists of three-dimensionally interconnected metallic struts, it can provide good elec-tron transfer paths. In addition, the open-cell porous structures in theNi foams can provide larger specific surface areas to redox couples andreduce mass transport limitations of the electrolyte in contact withthe electrode as compared to the commonNi foil.With further improve-ment, nitrided Ni foams with their inherent three-dimensionally-connected structure can be considered a promising solution to a cost-effective Pt- and TCO-free DSSC counter-electrode.

Acknowledgments

This work was supported by Institute of Basic Science (IBS) andthe Global Frontier R&D Program on Center for Multiscale EnergySystem funded by the National Research Foundation (NRF) underthe Ministry of Science, ICT & Future, Korea (2011-0031571). Thisstudy was also supported by another program funded through theNRF (NRF-2012K1A3A1A12054894). YC and HC also acknowledge fi-nancial supports from Priority (2012-0006680) and Pioneer (2011-0001684) Research Centers Programs through NRF.

569S.H. Park et al. / Surface & Coatings Technology 259 (2014) 560–569

References

[1] M. Grätzel, Nature 414 (2001) 338–344.[2] B. O'Regan, M. Grätzel, Nature 353 (1991) 737–740.[3] A. Yella, H.W. Lee, H.N. Tsao, C. Yi, A.K. Chandiran, Md.K. Nazeeruddin, E.W.G. Diau,

C.Y. Yeh, S.M. Zakeeruddin, M. Grätzel, Science 334 (2011) 629–634.[4] J.H. Yum, E. Baranoff, F. Kessler, T.Moehl, S. Ahmad, T. Bessho, A.Marchioro, E. Ghadiri,

J.E. Moser, C. Yi, Md.K. Nazeeruddin, M. Grätzel, Nat. Commun. 17 (3) (2012) 631.[5] Y. Luo, D. Li, Q. Meng, Adv. Mater. 21 (2009) 4647–4651.[6] M. Wu, X. Lin, Y. Wang, L. Wang, W. Guo, D. Qi, X. Peng, A. Hagfeldt, M. Grätzel, T.

Ma, J. Am. Chem. Soc. 134 (2008) 3419–3428.[7] T.N. Murakami, M. Grätzel, Inorg. Chim. Acta 361 (2008) 572–580.[8] S. Ahmad, J.H. Yum, H.J. Butt, M.K. Nazeeruddin, M. Grätzel, Chem. Phys. Chem. 11

(2010) 2814–2819.[9] G. Hashmi, K. Miettunen, T. Peltola, J. Halme, I. Asghar, K. Aitola, M. Toivola, P. Lund,

Renew. Sustain. Energy Rev. 15 (2011) 3717–3732.[10] A. Hauch, A. Georg, Electrochim. Acta 46 (2001) 3457–3466.[11] L. Kavan, J.H. Yum, M.K. Nazeeruddin, M. Grätzel, ACS Nano 5 (2011) 9171–9178.[12] M. Wu, X. Lin, T. Wang, J. Qiu, T. Ma, Energy Environ. Sci. 4 (2011) 2308–2315.[13] S. Ahmad, J.H. Yum, Z. Xianxi, M. Grätzel, H.J. Butta, M.K. Nazeeruddin, J. Mater.

Chem. 20 (2010) 1654–1658.[14] J. Burschka, V. Brault, S. Ahmad, L. Breau, M.K. Nazeeruddin, B. Marsan, S.M.

Zakeeruddin, M. Grätzel, Energy Environ. Sci. 5 (2012) 6089–6097.[15] Q. Tai, B. Chen, F. Guo, S. Xu, H. Hu, B. Sebo, X.Z. Zhao, ACS Nano 5 (2011)

3795–3799.[16] M. Wang, A.M. Anghel, B. Marsan, N.L. Cevey Ha, N. Pootrakulchote, S.M.

Zakeeruddin, M. Grätzel, J. Am. Chem. Soc. 131 (2009) 15976–15977.[17] G.R. Li, J. Song, G.L. Pan, X.P. Gao, Energy Environ. Sci. 4 (2011) 1680–1683.[18] M. Wu, X. Lin, A. Hagfeldt, T. Ma, Angew. Chem. Int. Ed. 50 (2011) 3520–3524.[19] G. Veerappan, K. Bojan, S.W. Rhee, ACS Appl. Mater. Interfaces 3 (2011) 857–862.[20] K. Miettunen, I. Asghar, X. Ruan, J. Halme, T. Saukkonen, P. Lund, J. Electroanal.

Chem. 653 (2011) 93–99.[21] J.H. Park, Y. Jun, H.G. Yun, S.Y. Lee, M.G. Kang, J. Electrochem. Soc. 155 (7) (2008)

F145–F149.[22] Y. Wang, M. Wu, X. Lin, Z. Shi, A. Hagfeldt, T. Ma, J. Mater. Chem. 22 (2012)

4009–4014.[23] K.S. Lee, H.K. Lee, D.H. Wang, N.G. Park, J.Y. Lee, O.O. Park, J.H. Park, Chem. Commun.

46 (2010) 4505–4507.[24] S.H. Park, K.H. Shina, J.Y. Kim, S.J. Yoo, K.J. Lee, J. Shin, J.W. Choi, J. Jang, Y.E. Sung, J.

Photochem. Photobiol. A 245 (2012) 1–8.

[25] Q.W. Jiang, G.R. Li, S. Liu, X.P. Gao, J. Phys. Chem. C 114 (2010) 13397–13401.[26] Q.W. Jiang, G.R. Li, X.P. Gao, Chem. Commun. (2009) 6720–6722.[27] M. Wu, Q. Zhang, J. Xiao, C. Ma, X. Lin, C. Miao, Y. He, Y. Gao, A. Hagfeldt, T. Ma, J.

Mater. Chem. 21 (2011) 10761–10766.[28] H. Xu, X. Zhang, C. Zhang, Z. Liu, X. Zhou, S. Pang, X. Chen, S. Dong, Z. Zhang, L. Zhang,

P. Han, X. Wang, G. Cui, ACS Appl. Mater. Interfaces 4 (2012) 1087–1092.[29] J. Lee, B. Yoo, T. Suresh, M. Kang, R. Vital, K. Kim, Electrochim. Acta 54 (2009)

4365–4370.[30] E. Stathatos, P. Lianos, A.S. Vuk, B. Orel, Adv. Funct. Mater. 14 (2004) 45–48.[31] H.Y. Hsueh, Y.C. Huang, R.M. Ho, C.H. Lai, T.M. Makida, H. Hasegawa, Adv. Mater. 23

(2011) 3041–3046.[32] T.A. Schaedler, A.J. Jacobsen, A. Torrents, A.E. Sorensen, J. Lian, J.R. Greer, L. Valdevit,

W.B. Carter, Science 334 (2011) 962–965.[33] Y. Chino, D.C. Dunand, Acta Mater. 56 (2008) 105–113.[34] D.C. Dunand, Adv. Eng. Mater. 6 (2004) 369–376.[35] K.M. Lee, C.Y. Hsu, P.Y. Chen, M. Ikegami, T. Miyasaka, K.C. Ho, Phys. Chem. Chem.

Phys. 11 (2009) 3375–3379.[36] M. Grdeń, M. Alsabet, G. Jerkiewicz, ACS Appl. Mater. Interfaces 4 (2012)

3012–3021.[37] J. Broiek, F. Grande, J.T. Anderson, A. Keys, Ann. N. Y. Acad. Sci. 110 (1963)

113–140.[38] K.A. Hing, S.M. Best, W. Bonfield, J. Mater. Sci. Mater. Med. 10 (1999) 135–145.[39] H. Choe, D.C. Dunand, Mater. Sci. Eng. A 384 (2004) 184–193.[40] E.C.M. Pennings, W. Grellner, J. Am. Ceram. Soc. 72 (1989) 1268–1270.[41] Y. Wang, Z.W. Fu, X.L. Yue, Q.Z. Qin, J. Electrochem. Soc. 151 (2004) E162–E167.[42] H.G. Yun, Y. Jun, J. Kim, B.S. Bae, M.G. Kang, Appl. Phys. Lett. 93 (2008) 133311.[43] S. Cho, S.H. Hwang, C. Kim, J. Jang, J. Mater. Chem. 22 (2012) 12164–12171.[44] Z. Li, B. Ye, X. Hu, X. Ma, X. Zhang, Y. Deng, Electrochem. Commun. 11 (2009)

1768–1771.[45] Q. Li, J. Wu, Q. Tang, Z. Lan, P. Li, J. Lin, L. Fan, Electrochem. Commun. 10 (2008)

1299–1302.[46] T. Ma, X. Fang, M. Akiyama, K. Inoue, H. Noma, E. Abe, J. Electroanal. Chem. 574

(2004) 77–83.[47] X. Fang, T. Ma, M. Akiyama, G. Guan, S. Tsunematsu, E. Abe, Thin Solid Films 472

(2005) 242–245.[48] M. Toivola, F. Ahlskog, P. Lund, Sol. Energy Mater. Sol. Cells 90 (2006) 2881–2893.[49] B.M. Klahr, T.W. Hamann, J. Phys. Chem. C 113 (2009) 14040–14045.[50] K.S. Lee, J.H. Yun, Y.H. Han, J.H. Yim, N.G. Park, K.Y. Cho, J.H. Park, J. Mater. Chem. 21

(2011) 15193–15196.[51] J. Ferber, J. Luther, Sol. Energy Mater. Sol. Cells 54 (1998) 265–275.