120023627_AL36_09_R1_060303

ANALYTICAL LETTERS Vol. 36, No. 9, pp. 20212039, 2003

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 DOI: 10.1081/AL-120023627 Copyright & 2003 by Marcel Dekker, Inc. 0003-2719 (Print); 1532-236X (Online) www.dekker.com

Electrochemical Characterization of Commercial and Home-Made Screen-Printed Carbon Electrodes

Aoife Morrin, Anthony J. Killard,* and Malcolm R. Smyth
National Centre for Sensor Research, School of Chemical Sciences, Dublin City University, Dublin, Ireland

ABSTRACT
Screen-printing technology is widely used for the mass-production of disposable electrochemical sensors. The practical utility of carbon screen-printed electrodes has been exploited, despite the fact that little is known about the nature of the electrode reactions. (Wang, J.; Pedrero, M.; Sakslumd, H.; Hammerich, O.; Pingarron, J. Electrochemical activation of screenprinted carbon strips. The Analyst 1996, 121 (3), 345350). Given the complexity of carbon electrodes in general, and dierences in the composition of commercial carbon inks, the question arises as to how such dierences and complexity aect their electrochemical reactivity. The aim of this

*Correspondence: Anthony J. Killard, National Centre for Sensor Research, School of Chemical Sciences, Dublin City University, Dublin 9, Ireland; Fax: 353 1 700 5703; E-mail: Tony.Killard@dcu.ie. 2021

AQ1

[3.6.20033:37pm]

[20212040]

[Page No. 2021]

f:/Mdi/Al/36(9)/120023627_AL_036_009_R1.3d

Analytical Letters (AL)

120023627_AL36_09_R1_060303

2022
43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84

Morrin, Killard, and Smyth work was to compare the electroactivity of both commercial electrodes and electrodes fabricated in-house from various commercial inks, in order to nd the electrode most suited to amperometric sensor work. Methods of analysis include cyclic voltammetry, amperometry and linear sweep voltammetry. It was found that the commercial working electrodes were not suited to the high current work of interest, due to their poor charge transfer properties. The in-house electrode had less resistive properties, and was more suited for high current amperometric sensing. Utilizing this electrode conguration, an optimal carbon paste was chosen for the working electrode. Key Words: Screen-printed electrode; Cyclic voltammetry; Linear sweep voltammetry; Amperometry; Charge transfer.

INTRODUCTION Carbon electrodes are particularly attractive for sensing applications. These materials have a high chemical inertness and provide a wide range of anodic working potentials with low electrical resistivity. They also have a very pure crystalline structure that provides low residual currents and a high signal to noise ratio.[2] Many of the devices reported rely on the use of carbon materials such as glassy carbon,[3] and carbon pastes.[4] Screen printing of the carbon ink for the fabrication of electrodes has realized commercial success in the glucose sensing eld.[5] Developed for the printing industry, this thick-lm technology has been adapted for the electronics industries and biosensor research. Screen-printed electrodes have low unit costs and are capable of undergoing mass production, while still maintaining adequate levels of reproducibility. They also have the advantages of miniaturization and versatility. Carbon ink used for working electrodes must contain a binder, solvent, and graphite particles. What is still of some concern with screenprinting, is the level of reproducibility in electrode production. This is mainly due to the nature of the carbon inksthe composition of which are proprietaryand the lack of control of the microscopic structure of individual electrodes. Grennan et al.[6] investigated the eects of the curing temperature on the physical and electrochemical characteristics of carbon paste C10903D14 (Gwent Electronic Materials). Improved sensor performance and decreased variability was demonstrated at elevated curing temperatures and this was associated with morphological changes to the carbon electrode surface. Wang et al.[7] compared the electrochemical

[3.6.20033:37pm]

[20212040]

[Page No. 2022]

f:/Mdi/Al/36(9)/120023627_AL_036_009_R1.3d

Analytical Letters (AL)

120023627_AL36_09_R1_060303

Screen-Printed Carbon Electrodes

85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126

2023

behavior and electroanalytical performance of thick lm carbon sensors on ceramic substrates fabricated from four dierent commercially available carbon inks. They found that C10903D14 (Gwent Electronic Materials) was optimal for amperometric sensing. This ink possessed an attractive electrochemical reactivity but was found to have high residual currents. This would render it most suited to amperometric work as this method is not dependent on background contributions. It would be less suited, however, to voltammetric or stripping voltammetry work. It is not just the interfacial region between solution and electrode that is important in determining the electrodes characteristics, but also the rest of the electrode, including the properties of the conducting path. Carbon inks may have higher resistivities than other types of conducting inks and so may not be suitable as a conductive layer, e.g., for high current work. Cui et al.[8] characterized a screen-printed strip comprising working, reference, and auxiliary electrodes. Silver acted as the conducting path. Erlenkotter et al.[9] used a similar format with on-board reference and auxiliary electrodes. However, the dierence was that carbon acted as the conducing path for the working and auxiliary electrodes. Both strips described potentially have dierent charge transfer properties due to their very dierent compositions, and although both strips were successful for their respective applications, they may not necessarily be suited to other applications. It is important when designing any type of screen-printed electrode that the charge transfer properties are suited to the end-use application.

EXPERIMENTAL Materials Horseradish peroxidase (HRP, 1100 U/mg and 1310 U/mg, P8672) was purchased from SigmaAldrich (Poole, Dorset, UK). Aniline was purchased from Aldrich (13,293-4), vacuum distilled and stored frozen under nitrogen. Thirty percent (v/v) hydrogen peroxide solution was purchased from Merck. Polyvinylsulphonate (PVS, 27,842-4), potassium hexacyanoferrate(II) (22,768-4) (potassium ferrocyanide trihydrate) and potassium hexacyanoferrate(III) (20,801-9) (potassium ferricyanide) were purchased from Aldrich. EuroashTM and UltraTM electrode strips were donated from Inverness Medical Ltd. EuroashTM, UltraTM, Ercon (661901), and LRH (C2010201R15) carbon paste inks were donated by Inverness Medical Ltd. (Inverness, Scotland). LifescanTM silver conductive ink was donated by Inverness Medical

[3.6.20033:37pm]

[20212040]

[Page No. 2023]

f:/Mdi/Al/36(9)/120023627_AL_036_009_R1.3d

Analytical Letters (AL)

120023627_AL36_09_R1_060303

2024
127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168

Morrin, Killard, and Smyth

Ltd. Seriwash universal screen wash (ZT639) was obtained from Sericol Ltd. (Kent, UK). Glassy carbon and silver/silver chloride (Ag/AgCl) electrodes were purchased from Bioanalytical Systems Ltd. (Cheshire, UK). The platinum mesh (29,809-3) was purchased from Aldrich.

Buers and Solutions Unless otherwise stated, all electrochemical measurements were carried out in phosphate buered saline (PBS), (0.1 M phosphate, 0.137 M NaCl, and 2.7 mM KCl), pH 6.8.

Instrumentation Screen-printing of in-house (noncommercial) electrodes was performed with a semi-automated DEK 248 printing machine (Weymouth, UK). Nylon screens with varying mesh thickness were used, and mounted at 45 to the print stroke. Blade rubber squeegees were employed, and a ood blade was utilized. All inks were cured in a conventional oven. All electrochemical protocols were performed either on a BAS100/W electrochemical analyzer with BAS100/W software, or a CHI1000 potentiostat with CHI1000 software, using either cyclic voltammetry or time-based amperometric modes. An Ag/AgCl reference electrode and a platinum mesh auxillary electrode were used for bulk electrochemical experiments. Scanning electron microscopy (SEM) was performed with a Hitachi S 3000N scanning electron microscope. An acceleration voltage of 20 kV was employed.

Screen-Printed Electrode Fabrication Five electrode types were fabricated for this study. Two were manufactured commercially (EuroashTM and UltraTM) and three by inhouse screen-printing (designated Ultra-inH, Ercon-inH, and LRH-inH, according to the working electrode carbon used). The structural characteristics of each of the electrodes are summarized in Table 1. The curing conditions for all inks are summarized in Table 2.

T1 T2

[3.6.20033:37pm]

[20212040]

[Page No. 2024]

f:/Mdi/Al/36(9)/120023627_AL_036_009_R1.3d

Analytical Letters (AL)

[3.6.20033:37pm]

Screen-Printed Carbon Electrodes

[20212040]

120023627_AL36_09_R1_060303

[Page No. 2025]

Table 1.

Summary of the composition of all electrodes used.

Electrode type EuroashTM carbon UltraTM carbon UltraTM carbon Ercon 661901 LRH C2010201R15 10 mm2 9 mm2 8 mm2

Conducting layer

Working electrode layer

Working electrode area

Insulation layer Ercon

Name EuroashTM UltraTM Ultra-inH Ercon-inH LRH-inH

EuroashTM Lifescan silver and EuroashTM carbon UltraTM carbon Lifescan silver Lifescan silver Lifescan silver

f:/Mdi/Al/36(9)/120023627_AL_036_009_R1.3d

UltraTM Ultra in-house Ercon in-house LRH in-house

2025

Analytical Letters (AL)

120023627_AL36_09_R1_060303

2026
169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210

Morrin, Killard, and Smyth Table 2. Curing conditions for all screen-printing inks used. Curing conditions Conventional oven @70 C for 6 min Conventional oven @70 C for 13 min

Ink type Lifescan silver ink Carbon inks donated by Inverness Medical Ltd: EuroashTM UltraTM Ercon 661901 LRH C2010201R15 Ercon insulation ink

Conventional oven @70 C for 15 min

(f)

(c) (e) (d) (a) (b)

Figure 1. Components of the in-house screen-printed electrode: (a) substrate, (b) Ag/AgCl conducting paths, (c) carbon working electrode, (d) Ag/AgCl auxillary electrode, (e) Ag/AgCl reference electrode, and (f ) insulation layer. (Working electrode area: 9 mm2).

Figure 1 depicts a schematic of the in-house screen-printed electrode with onboard reference and auxiliary electrodes (Ultra-inH, Ercon-inH, and LRH-inH). Electrodes were screen-printed onto a preshrunk PET substrate (a). Initially, a layer of three Ag/AgCl tracks were deposited as the conducting paths from electrodes to contacts for the reference, auxiliary, and working electrodes (b). A layer of carbon was deposited as the working electrode (c). The Ag/AgCl acted as both reference (d) and

F1

[3.6.20033:37pm]

[20212040]

[Page No. 2026]

f:/Mdi/Al/36(9)/120023627_AL_036_009_R1.3d

Analytical Letters (AL)

120023627_AL36_09_R1_060303

Screen-Printed Carbon Electrodes

211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252

2027

auxiliary (e) electrodes. Finally, an insulation layer was deposited to eliminate cross-talk and to dene the working electrode area (9 mm2) (f ).

Cyclic Voltammetry Glassy carbon or screen-printed electrodes were cycled in equimolar amounts of potassium ferrocyanide and potassium ferricyanide (1 103 M) using 1 M KCl as supporting electrolyte. Voltammograms were obtained using scan rates ranging from 10 to 100 mV s1 and at a sensitivity of 1 103 A V1 vs. Ag/AgCl under diusion limited conditions.

Determination of Heterogeneous Electron Transfer Rate Constants Heterogeneous electron transfer rate constants (k0) were calculated using the method of Nicholson[10] according to Eq. (1):   1=2  =2 nF DR k D0 v RT D0
0

where refers to a kinetic parameter, D0 is the diusion coecient for the ferricyanide (7.6 106 cm2 s1), DR is the diusion coecient for the ferrocyanide (6.3 106 cm2 s1), and  is the transfer coecient (0.5), R is the universal gas constant (8.314 JK mol1), T is the absolute temperature (K), n is the number of electrons transferred, and F is Faradays constant (96,485 C). values for the electrode systems were calculated with the aid of a solver program that generated the sixth polynomial plot of Ep vs. log ().

Electrode Pretreatment Procedure Glassy carbon electrodes were cleaned by successive polishing on aqueous slurries of 1, 0.3, and 0.05 mm alumina powder, followed by ultrasonic cleaning in Milli-Q water for 10 min. The electrodes were then placed in a solution of 0.2 M H2SO4. A single voltammetric cycle was carried out between 1200 mV and 1500 mV at 100 mVs1 vs. Ag/AgCl. The same voltammetric procedure was employed for cleaning the screen-printed electrodes.

[3.6.20033:37pm]

[20212040]

[Page No. 2027]

f:/Mdi/Al/36(9)/120023627_AL_036_009_R1.3d

Analytical Letters (AL)

120023627_AL36_09_R1_060303

2028
253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294

Morrin, Killard, and Smyth

Amperometric Electroanalytical Procedure The electroanalytical procedure was carried out according to Killard et al.[11]

RESULTS AND DISCUSSION Two commercially manufactured screen-printed working electrodes (WE) were examined; EuroashTM and One Touch UltraTM. These electrodes were manufactured by Inverness Medical Ltd., for glucose testing. The WE of the EuroashTM strip were composed of a silver and carbon conducting path, a carbon working electrode and an insulation layer to dene the electrode area. The UltraTM WE electrode, contained only carbon and insulation layers, relying on only carbon to act as the conductor and the electrode. The advantage of using less silver, or none at all, is to allow for reduced cost manufacturing. Electrochemical analyses were initially carried out on the commercial electrodes. However, subsequently the in-house artwork was designed as a result of nding that the commercial electrodes suered from severe charge transfer problems and were not suitable to this amperometric sensor work. This in-house electrode design (Fig. 1) did not encounter charge transfer diculties as the conducting tracks were composed solely of silver. It was used for the analysis of UltraTM, Ercon, and LRH inks and these electrodes are referred to in this section as Ultra-inH, Ercon-inH, and LRH-inH, respectively. Summaries of all inks used for fabrication and their respective curing conditions are given in Tables 1 and 2.

Voltammetric Performance of Screen-Printed Electrodes The ferri/ferrocyanide redox couple was the redox system used for comparing the voltammetric behavior of screen-printed electrodes. Figure 2 displays the cyclic voltammograms of the redox couple at a glassy carbon electrode for comparison purposes, the commercial carbon strip electrodes (EuroashTM and UltraTM) and the commercial inks printed in-house (Ercon-inH, LRH-inH, and Ultra-inH). The mean peak separations and anodic ( jp,a) and cathodic ( jp,c) peak current densities are also illustrated in Fig. 3 (n 3). The commercial electrodes, EuroashTM and UltraTM, yielded very poor reversibility with Ep values of 471 (56) mV and 416 (37) mV, respectively and jp,a values of 12.313 (2.025) mA cm2 and 15.107 (2.638) mA cm2, respectively.

F2

F3

[3.6.20033:37pm]

[20212040]

[Page No. 2028]

f:/Mdi/Al/36(9)/120023627_AL_036_009_R1.3d

Analytical Letters (AL)

120023627_AL36_09_R1_060303

Screen-Printed Carbon Electrodes

295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314
Current (A) Current (A)
4e-5 2e-5 2e-5 4e-5 -4.0e-5 0.6 0.4 0.2 0.0 2.0e-5 2e-5 4.0e-5 4e-5

2029

Current (A)

Current (A)

0.0

-2.0e-5

-2e-5

(a)

-4e-5 0.6 0.4 0.2 0.0

(b)

4e-5

Current (A)

Current (A)

-2e-5

-2e-5

-4e-5 0.6 0.4 0.2 0.0

(c)

-4e-5 0.6 4e-5 0.4 0.2 0.0

(d)

315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336

2e-5

2e-5

-2e-5

-2e-5

-4e-5 0.6 0.4 0.2 0.0

(e)

-4e-5 0.6 0.4 0.2 0.0

(f)

Potential (V)

Potential (V)

Figure 2. Cyclic volatmmograms for dierent electodes in 1 103 M ferri/ ferrocyanide and 1 M KCl supporting electrolyte. (a) Glassy carbon, (b) EuroashTM, (c) UltraTM, (d) Ercon-inH, (e) LRH-inH, and (f ) Ultra-inH. The commercial electrodes ((b), (c)) showed very poor reversibility. Using the in-house design, ((d), (e), (f )) reversibility improved, with the Ultra-inH exhibiting the best behavior.

This behavior was attributed to the poor charge transfer of the electrodes. The conducting paths (composed of segments of silver and carbon for EuroashTM and fully carbon for UltraTM electrodes) from WEs to contacts had resistive properties that may have become a signicant limiting factor in charge transfer. It resulted in the poor reversibility of the redox couple and low jp,a values. This initial work motivated the inhouse electrode artwork to be designed where the conducting tracks were

[3.6.20033:37pm]

[20212040]

[Page No. 2029]

f:/Mdi/Al/36(9)/120023627_AL_036_009_R1.3d

Analytical Letters (AL)

120023627_AL36_09_R1_060303

2030
337 338 339
E (mV)
700
4e-6

Morrin, Killard, and Smyth

600

Current Density (A mm2)

500

2e-6

340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378

400

jp,a
0

300

jp,c

200

-2e-6

100

-4e-6

EuroflashTM UltraTM

Ultra-inH

LRH-inH

Ercon-inH

EuroflashTM UltraTM

Ultra-inH

LRH-inH

Ercon-inH

Electrode

Electrode

(a)

(b)

Figure 3. (a) Cyclic voltammetric peak separations (Ep) and (b) anodic and cathodic peak current densities for 1 103 M ferri/ferrocyanide and 1 M KCl for each of the screen-printed electrodes (n 3).

composed solely of silver. Each in-house thick-lm carbon electrode exhibited dierent electron-transfer reactivities towards ferri/ferro, with the Ultra-inH electrode yielding the most reversible behavior. For example, the redox couple gave a Ep value of 264 (7) mV for UltrainH, as compared to 314 (25) mV and 562 (52) mV for Ercon-inH and LRH-inH, respectively. The Ultra-inH also oered the highest jp values and lowest overvoltage of all the in-house electrodes (i.e., anodic peak potentials for ferrocyanide of 392 mV, compared to 452, 460, 410, and 439 mV for UltraTM, EuroashTM, Ercon-inH, and LRH-inH, respectively). Of all the electrodes examined, the Ultra-inH electrode exhibited the best behavior towards the redox couple. It was observed immediately that the commercial electrodes, manufactured by Inverness Medical Ltd., were not suited to present purposes because of poor charge transfer properties. The in-house design had more optimal charge transfer properties, and in conjunction with the UltraTM commercial ink as the WE, behaved as the best screen-printed electrode. This work demonstrates the importance of optimizing both the conducting path and the carbon of the WE when designing a new screen-printed electrode. Both parameters have profound eects on the behavior of the electrode. All k0 values are given in Table 3. Recalling that for all the screenprinted electrodes the Ep values were considerably greater than the 59 mV value expected for Nernstian one-electron reactions, k0 values were then also inevitably low compared to glassy carbon. Commercial electrodes (EuroashTM and UltraTM) exhibited k0 values 2000-fold and 1250-fold lower than that obtained for glassy carbon, respectively. The LRH-inH electrode proved the poorest with regard to k0, being

T3

[3.6.20033:37pm]

[20212040]

[Page No. 2030]

f:/Mdi/Al/36(9)/120023627_AL_036_009_R1.3d

Analytical Letters (AL)

120023627_AL36_09_R1_060303

Screen-Printed Carbon Electrodes

379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420

2031

Table 3. Table of heterogeneous electron transfer rate constant (k0) for glassy carbon and each of the screenprinted electrodes. Electrode Glassy carbon EuroashTM UltraTM Ercon-inH LRH-inH Ultra-inH k0 (cm s1) 5.9 102 2.83 105 4.7 105 1.74 104 1.67 105 3.09 104

3500-fold lower than glassy carbon. Ercon-inH and Ultra-inH both had the best k0 values of the screen-printed electrodes, yielding k0 values only 300-fold and 200-fold lower than glassy carbon, respectively. Thus, Ultra-inH exhibited the best k0 value, even if this was still two orders of magnitude lower than glassy carbon. Such decreases in the electrontransfer reactivity may be consistent with the composition of the ink, being composed only partly of conductive carbon particles. In view of the proprietary composition of all the inks, it is dicult to explain why the Ultra-inH electrode displayed the most favorable redox behavior. Observed changes in redox behavior may be dictated by varying graphite content (good redox behavior suggests a high graphite loading), the nature of the graphite particles, and the presence or absence of an adherent (inhibitory) organic layer. Further studies employing energy dispersive x-ray analysis (EDX), and scanning electrochemical microscopy (SECM) may help to establish the relationship between carbon content and electrode performance. Although Ultra-inH was shown to have the best behavior of all the screen-printed electrodes to ferri/ferrocyanide, its behavior was still far from ideal. Attempts to improve its behavior (by electrochemical pretreatment and optimization of curing conditions) were carried out and discussed in a later section.

Amperometric Performance of Screen-Printed Electrodes The electrochemical performance of the screen-printed electrodes was investigated by incorporating them into a batch cell set-up.[9] Previous work by this group had used these types of screen-printed carbon electrodes as the basis of a biosensor using electrodeposited

[3.6.20033:37pm]

[20212040]

[Page No. 2031]

f:/Mdi/Al/36(9)/120023627_AL_036_009_R1.3d

Analytical Letters (AL)

120023627_AL36_09_R1_060303

2032
421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462
0 30 40

Morrin, Killard, and Smyth

Current (A)

20

10

-10 0 2 4 6 8 10

[H2O2] mM

Figure 4. Amperometric sensor responses for successive additions of 0.5 mM hydrogen peroxide to a solution containing 2 mg/mL horseradish peroxidase. Electrodes used were (f) Ultra-inH, () Ercon-inH, and (m) LRH-inH. Electrodes held at 100 mV vs. Ag/AgCl wire electrode.

conducting PANI/PVS lms onto which was deposited HRP or antibodies. The nature of these biosensors has been described elsewhere.[11,12] Briey, PANI/PVS was deposited on the surface of the electrode and the potential was cycled the required number of times. No protein was immobilized onto the surface of the polymer. Ultra-inH, Ercon-inH, and LRH-inH electrodes were subjected to successive additions of 0.5 mM hydrogen peroxide added freshly to a solution of 2 mg mL1 horseradish peroxidase and the amperometric response monitored. All three sensors responded to the changes in peroxide concentration (Fig. 4). Similar response times and noise levels were observed (data not shown). Ultra-inH oered the highest sensitivity (4 mA mM1 peroxide), with Ercon-inH exhibiting a slightly lower sensitivity (3.2 mA mM1 peroxide). LRH-inH showed the poorest sensitivity (1.8 mA mM1 peroxide). This correlates with the voltammetric behavior. Ultra-inH exhibited the highest sensitivity in terms of jp values while LRH-inH exhibited the lowest values. Amperometric experiments could not be carried out on either of the commercial electrodes as the charge transfer properties of the electrodes hindered deposition of adequate polymer. For example, the in-house

F4

[3.6.20033:37pm]

[20212040]

[Page No. 2032]

f:/Mdi/Al/36(9)/120023627_AL_036_009_R1.3d

Analytical Letters (AL)

120023627_AL36_09_R1_060303

Screen-Printed Carbon Electrodes

463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 491 492 493 494 495 496 497 498 499 500 501 502 503 504
1500 1000 500 Potential (mV) 0 -500 -4e-5 Current (A) 2e-5 4e-5

2033

strips required seven cycles to immobilize polymer to the required thickness, whereas the commercial electrodes needed 20 cycles in order to reach only one fth the required thickness. The experiments on these strips were abandoned at this point.

Linear Sweep Voltammetric Performance of Screen-Printed Electrodes The background current of thick-lm carbon electrodes is strongly aected by the carbon ink employed.[7] Figure 5 compares the background voltammograms for the dierent carbon electrodes in degassed phosphate buer (pH 6.8). Several electrodes of each type were analyzed and Fig. 5 shows data representative of all analyses. LRH-inH exhibited the widest potential window particularly with respect to the cathodic potential limit (i.e., high hydrogen overvoltage). Its potential window had a range of 1150 to 300 mV, where the nonfaradaic current remained constant (5.2 mA) in this electrolyte solution. The background current of Ercon-inH was narrow and poor, exhibiting a lot of interference. Ultra-inH also had a narrow potential window (1097 to 60 mV) but

F5

490

-2e-5

Figure 5. Linear sweep voltammograms in degassed PBS buer ( pH 6.8). Electrodes used were (solid line) LRH-inH, (long dash) Ultra-inH, and (short dash) Ercon-inH.

[3.6.20033:37pm]

[20212040]

[Page No. 2033]

f:/Mdi/Al/36(9)/120023627_AL_036_009_R1.3d

Analytical Letters (AL)

120023627_AL36_09_R1_060303

2034
505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546

Morrin, Killard, and Smyth

was not aected by major interferences. The non-faradaic current was of the same magnitude as for LRH-inH. The anodic potential limits (i.e., oxygen overvoltage) were approximately the same for each of the inks. A carbon ink possessing a narrow potential is not necessarily a negative property for amperometric sensing. It should be noted that an electrode of choice for xed potential amperometric biosensors need not necessarily have the widest potential window as amperometric measurements are less aected by dierences in the background contributions, as they are usually performed after the decay of transient currents to steady state values.[7] However, the nonfaradaic background current measured in linear sweep voltammetry, could potentially have an eect on the sensitivity of the electrode. The background current can limit the lowest current that can be measured, and so could aect the detection limits of an assay.

Optimization of Ultra-inH Although Ultra-inH did exhibit the best properties of all electrodes, for the purpose of designing an electrode suited towards amperometric sensing, there were major concerns that the UltraTM ink for the WE was still not ideal. This was highlighted in the cyclic voltammetric study of ferrocyanide. Attempts to decrease the Ep values were done by varying the curing temperature and length of curing time of the carbon ink, and also the eect of electrochemical pretreatment was studied.

Curing Parameters Due to the composition of carbon inks, the parameters of curing can have a profound eect on their performance.[6] Ep values and ip values for the ferri/ferrocyanide redox couple were monitored over a range of curing temperatures and it was found that above a temperature of 70 C, Ep values increased greatly and the ip,a decreased (Fig. 6). Carbon inks may be composed of three basic constituents: graphite, vinyl, or epoxy-based polymeric binders and solvent to enhance the inks anity for the substrate in terms of adhesion, and to improve viscosity for the screen-printing process. It has been suggested that increases in curing temperature may result in evaporation of the solvent and decomposition of the polymeric binder to give a greater denition of the graphite or carbon particles. This would mean that the increases in

F6

[3.6.20033:37pm]

[20212040]

[Page No. 2034]

f:/Mdi/Al/36(9)/120023627_AL_036_009_R1.3d

Analytical Letters (AL)

120023627_AL36_09_R1_060303

Screen-Printed Carbon Electrodes

547 548 549 550
Ip,a (mA)
-2.5 300 -3.0 400

2035

553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588

-2.0

E (mV)

551 552

200

-1.5

100

-1.0

0 25 40 70 120 150

Curing temperature (C)

Figure 6. Mean cyclic voltammetric peak separations (n 4) (bar) and mean anodic peak currents (n 4) (line) for 1 103 M ferrocyanide and 1 M KCl for varying curing temperatures of the UltraTM ink on the Ultra-inH electrode. A curing temperature of 70 C showed optimal voltammetric performance for Ep and ip,a values.

temperature should result in an increase in the microparticulate nature of the carbon and greater denition of the graphite particle surface area.[6] According to this theory, electron transfer rates should increase as the graphite particle surface area becomes increasingly dened (i.e., with increasing temperature). This behavior was not observed for the Ultra ink. Increases in temperature above 70 C resulted in reduced reaction kinetics. This demonstrates that that the individual nature of the ink and its unknown constituents can have a profound eect on its characteristics. The Ultra ink seemed to maintain a very dened microparticulate character (Fig. 7), which was quite dierent from the Gwent electrode.[6] This dierence in behavior could be attributed to dierent solvents with dierent evaporation rates, dierent graphite particles or binders or other additives, and their relative concentrations/ solubilities etc. If the viscosity of inks dier, the lm thickness after a single print will most likely also dier and may cause a dierence in the electrochemical and physical characteristics. Below 70 C, Ep values were relatively constant. There were no resistive eects observed that one might expect if there was excess polymeric binder present at the electrode surface due to curing at lower temperatures. ip,a values peaked at 70 C, and this temperature was chosen as optimum, as greatest sensitivity was exhibited at this curing temperature.

F7

[3.6.20033:37pm]

[20212040]

[Page No. 2035]

f:/Mdi/Al/36(9)/120023627_AL_036_009_R1.3d

Analytical Letters (AL)

120023627_AL36_09_R1_060303

2036
589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630

Morrin, Killard, and Smyth

Figure 7. Typical scanning electron micrograph of the Ultra ink cured at 70 C for 13 min (2500 X magnication). The surface topography shows good denition of graphite particles.

Scanning electron microscopy (SEM) was carried out on all of these electrodes, and no discernible dierences were noted at the dierent curing temperatures. Therefore, any changes in charge transfer eects by varying the curing temperature cannot be attributed to physical changes in the morphology of the surfaces. Figure 7 shows a typical UltraTM ink carbon surface cured at 70 C. The eect of curing time on electrode performances was monitored at a curing temperature of 70 C. Over a range of 020 min, there did not appear to be any signicant eect on electrode performance. Ep values for ferrocyanide did not vary signicantly (data not shown). This data appears to suggest that curing had very little impact on the performance of the ink at all. Even at 0 min (no curing), Ep was only marginally higher. This correlates with the curing temperature data in that up to about 70 C, there is very little gain in electrochemical performance. Heat may serve only to physically dry the constituents of the ink. This may also suggest that any solvents present are extremely volatile or extremely non-volatile and ink-drying may be brought about by polymerization processes rather than evaporative processes. It certainly suggests a very dierent composition for the Ultra ink, as opposed to the Gwent ink previously characterized.[6] Since curing temperature and time were not found to be such important factors, it was decided not to deviate from the standard

[3.6.20033:37pm]

[20212040]

[Page No. 2036]

f:/Mdi/Al/36(9)/120023627_AL_036_009_R1.3d

Analytical Letters (AL)

120023627_AL36_09_R1_060303

Screen-Printed Carbon Electrodes

631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672

2037

using conditions of 70 C for 13 min. For all future work, these curing parameters would be used for the UltraTM ink.

Electrochemical Pretreatment Pre-treatment of working electrodes is a method employed by many researchers in order to enhance the electrochemical activity of their screen-printed electrodes.[1,8,13] It is generally agreed that pretreatment eectively removes organic binders and contamination that occur at electrode surfaces such as carbon and gold and may bring about an increase in the numbers of chemically reactive sites on the electrode surface. Wang et al.[1] employed an electrochemical pretreatment method involving short preanodization (30 s to 3 min in the 1.5 to 2.0 V range) of screen-printed electrodes in phosphate buer solution (0.05 M). This pretreatment method appeared to increase the surface functionalities and roughness or to remove surface contaminants and resulted in enhanced electrochemical activity. Electrochemical pretreatment of electrodes can also be carried out by cycling the potential in acidic media. Gue et al.[13] simply used a chemical cleaning step with sulphuric acid and hydrogen peroxide solution for gold microelectrodes. This step was critical for sensor sensitivity. The electrochemical pretreatment method of Killard et al.[11] has been employed in this work. Cycling the screen-printed electrode in sulphuric acid (0.2 M) is believed to have the eect of stripping the surface of the carbon electrode. Any insulative materials present at the surface may be removed. The procedure may even have the eect of renewing the surface by removing the whole outer layer of the ink. To assess the eect of electrode pretreatment on the UltraTM ink, the electrodes were subjected to varying numbers of cycles in 0.2 M H2SO4, and the eect of this on electrode behavior was examined by looking at the ferri/ferro couple. By electrochemically pretreating the Ultra electrode, its behavior towards the ferri/ferrocyanide redox couple improved dramatically. Ep values decreased by 50%. Before pretreatment, electrodes were exhibiting an average Ep value of 222 mV (RSD 2.0%, n 9). After pretreatment, this was reduced to 112 mV (RSD 3.5%, n 9). ip current values also increased as a result. One pretreatment cycle was sucient to observe this behavior. Increasing the number of pretreatment cycles did not have a signicant eect. k0 values increased from 3.09 104 cm s1 (no pretreatment) to 3.97 103 cm s1 (pretreated); a 10-fold improvement. These gures suggest that the electrochemical pretreatment of the screen-printed

[3.6.20033:37pm]

[20212040]

[Page No. 2037]

f:/Mdi/Al/36(9)/120023627_AL_036_009_R1.3d

Analytical Letters (AL)

120023627_AL36_09_R1_060303

2038
673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714

Morrin, Killard, and Smyth

electrode greatly improved their electrochemical performance. After pretreatment, the kinetics and charge transfer rates at the UltraTM electrode were enhanced greatly.

CONCLUSION Initially, commercial screen-printed electrodes were examined with a view to using them for amperometric immunosensing. It was found however, that although the working electrodes of the strips may have been suitable, the charge transfer properties of the strips were not high enough for the high current work of interest. This was due to the fact that the electrode surfaces and conducting paths were too resistive and hindered the required current ow from the working electrode to the potentiostat. A new in-house electrode was designed with a silver conducting path. The charge transfer properties of the electrode were not limiting, and this design was used for the electrochemical analysis of various working electrode carbon inks. The inks were analyzed using voltammetry, linear sweep voltammetry, and amperometry and it was found that the Ultra-inH electrode had the most preferable electrochemical properties (i.e., a k0 value of 3.09 104 cm s1, and a high sensitivity in the amperometric experiments). These properties were further enhanced by electrochemical pretreatment rendering it the most suitable for amperometric sensing.

REFERENCES 1. Wang, J.; Pedrero, M.; Sakslumd, H.; Hammerich, O.; Pingarron, J. Electrochemical activation of screenprinted carbon strips. The Analyst 1996, 121 (3), 345350. 2. Zhang, S.; Wright, G.; Yang, Y. Materials and techniques for electrochemical biosensor design and construction. Biosens. & Bioelectron. 2000, 15 (56), 273282. 3. Bin, L.; Smyth, M.R.; OKennedy, R. Immunological activities of IgG antibody on precoated Fc receptor surfaces. Anal. Chim. Acta 1996, 331 (29R32R), 97102. 4. Ciana, L.D.; Bernacca, G.; Bordin, F.; Fenu, S.; Garetto, F. Highly sensitive amperometric measurement of alkaline phosphatase activity with glucose oxidase amplication. J. Electroanal. Chem. 1995, 382 (12), 129135.

[3.6.20033:37pm]

[20212040]

[Page No. 2038]

f:/Mdi/Al/36(9)/120023627_AL_036_009_R1.3d

Analytical Letters (AL)

120023627_AL36_09_R1_060303

Screen-Printed Carbon Electrodes

715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756

2039

5. Hart, J.; Wring, S. Recent developments in the design and application of screen-printed electrochemical sensors for biomedical, environmental and industrial analyses. Trends in Anal. Chem. 1997, 16 (2), 89103. 6. Grennan, K.; Killard, A.J.; Smyth, M.R. Physical characterization of a screen-printed electrode for use in an amperometric biosensor system. Electroanal. 2001, 13 (89), 745750. 7. Wang, J.; Tian, B.; Nascimento, V.B.; Angnes, L. Performance of screen-printed carbon electrodes fabricated from dierent carbon inks. Electrochim. Acta 1998, 43 (23), 34593465. 8. Cui, G.; Yoo, J.; Lee, J.; Yoo, J.; Uhm, J.; Cha, G.; Nam, H. Eect of pretreatment on the surface and electrochemical properties of screen-printed carbon paste electrodes. The Analyst 2001, 126 (8), 13991403. 9. Erlenkotter, A.; Kottbus, M.; Chemnitius, G. Flexible amperometric transducers for biosensors based on a screen-printed three electrode system. J. Electroanal. Chem. 2000, 481 (1), 8294. 10. Nicholson, R.S. Theory and application of the cyclic voltammetry of electrode reaction kinetics. Anal. Chem. 1965, 37 (11), 13511355. 11. Killard, A.J.; Zhang, S.; Zhao, H.; John, R.; Iwuoha, E.I.; Smyth, M.R. Development of an electrochemical ow injection immunoassay (FIIA) for the real-time monitoring of biospecic interactions. Anal. Chim. Acta 1999, 400 (13), 109119. 12. Killard, A.J.; Micheli, L.; Grennan, K.; Franek, M.; Kolar, V.; Moscone, D.; Palchetti, I.; Smyth, M.R. Amperometric separation-free immunosensor for real-time environmental monitoring. Anal. Chim. Acta 2001, 427, 173180. 13. Gue, A.; Tap, H.; Gros, P.; Maury, F. A miniaturized silicon based enzymantic biosensor: towards a generic structure and technology for multi-analytes assays. Sens. Actuat. B 2002, 82 (23), 227232. Received February 10, 2003 Accepted March 9, 2003

AQ2

[3.6.20033:37pm]

[20212040]

[Page No. 2039]

f:/Mdi/Al/36(9)/120023627_AL_036_009_R1.3d

Analytical Letters (AL)

[3.6.20033:37pm]

[20212040]

[Page No. 2040]

f:/Mdi/Al/36(9)/120023627_AL_036_009_R1.3d

Analytical Letters (AL)