nanomaterials Article Fully Inkjet-Printed Biosensors Fabricated with a Highly Stable Ink Based on Carbon Nanotubes and Enzyme-Functionalized Nanoparticles Mijal Mass 1 , Lionel S. Veiga 1 , Octavio Garate 1 , Gloria Longinotti 1 , Ana Moya 2,† , Eloi Ramón 2 , Rosa Villa 2,3 , Gabriel Ybarra 1, * and Gemma Gabriel 2,3, * 1 INTI-Micro y Nanotecnologías, Instituto Nacional de Tecnología Industrial (INTI), San Martín, Buenos Aires B1650WAB, Argentina; mmass@inti.gob.ar (M.M.); lveiga@inti.gob.ar (L.S.V.); ogarate@inti.gob.ar (O.G.); glonginotti@inti.gob.ar (G.L.) 2 Institut de Microelectrònica de Barcelona, IMB-CNM (CSIC), Campus Universitat Autònoma de Barcelona, Cerdanyola del Vallès, 08193 Barcelona, Spain; ana.moya@eurecat.org (A.M.); eloi.ramon@imb-cnm.csic.es (E.R.); rosa.villa@imb-cnm.csic.es (R.V.) 3 CIBER de Bioingeniería, Biomateriales y Nanomedicina (CIBER-BBN), 28029 Madrid, Spain * Correspondence: gybarra@inti.gob.ar (G.Y.); gemma.gabriel@imb-cnm.csic.es (G.G.) † Present Addresses: Eurecat, Centre Tecnològic de Catalunya, Functional Printing & Embedded Devices Unit, 08302 Mataró, Spain.   Abstract: Enzyme inks can be inkjet printed to fabricate enzymatic biosensors. However, inks Citation: Mass, M.; Veiga, L.S.; containing enzymes present a low shelf life because enzymes in suspension rapidly lose their Garate, O.; Longinotti, G.; Moya, A.; catalytic activity. Other major problems of printing these inks are the non-specific adsorption of Ramón, E.; Villa, R.; Ybarra, G.; enzymes onto the chamber walls and stability loss during printing as a result of thermal and/or Gabriel, G. Fully Inkjet-Printed mechanical stress. It is well known that the catalytic activity can be preserved for significantly longer Biosensors Fabricated with a Highly periods of time and to harsher operational conditions when enzymes are immobilized onto adequate Stable Ink Based on Carbon surfaces. Therefore, in this work, horseradish peroxidase was covalently immobilized onto silica Nanotubes and Enzyme-Functionalized nanoparticles. Then, the nanoparticles were mixed into an aqueous ink containing single walled Nanoparticles. Nanomaterials 2021, 11, carbon nanotubes. Electrodes printed with this specially formulated ink were characterized, and 1645. https://doi.org/10.3390/ enzyme electrodes were printed. To test the performance of the enzyme electrodes, a complete nano11071645 amperometric hydrogen peroxide biosensor was fabricated by inkjet printing. The electrochemical response of the printed electrodes was evaluated by cyclic voltammetry in solutions containing Academic Editor: redox species, such as hexacyanoferrate (III/II) ions or hydroquinone. The response of the enzyme Piersandro Pallavicini electrodes was studied for the amperometric determination of hydrogen peroxide. Three months after the ink preparation, the printed enzyme electrodes were found to still exhibit similar sensitivity, Received: 3 June 2021 demonstrating that catalytic activity is preserved in the proposed ink. Thus, enzyme electrodes can Accepted: 17 June 2021 be successfully printed employing highly stable formulation using nanoparticles as carriers. Published: 23 June 2021 Keywords: biosensors; electrochemical detection; inkjet printing; carbon nanotubes; carbon-ink Publisher’s Note: MDPI stays neutral electrodes; silica nanoparticles with regard to jurisdictional claims in published maps and institutional affil- iations. 1. Introduction Printed electronics are expected to have a great impact on the manufacturing of Copyright: © 2021 by the authors. functional electronic devices [1–7], especially in the biomedical field, where the fabrication Licensee MDPI, Basel, Switzerland. of low-cost single-use sensors and biosensors have led to interesting applications, such This article is an open access article as personalized medicine [8–12]. Inkjet printing (IJP) is one of the most used printing distributed under the terms and technologies for sensor production because it allows low production costs and high mass- conditions of the Creative Commons production of devices with the desired reproducibility [13–15]. A variety of functional Attribution (CC BY) license (https:// materials can be digitally deposited by IJP in microscale dimensions at low temperatures creativecommons.org/licenses/by/ on a wide variety of substrates. Furthermore, as a non-contact, mask-less deposition 4.0/). approach, it reduces fabrication time and costs, and permits customized geometries [16–19]. Nanomaterials 2021, 11, 1645. https://doi.org/10.3390/nano11071645 https://www.mdpi.com/journal/nanomaterials Nanomaterials 2021, 11, 1645 2 of 15 Recent research has also demonstrated that inkjet printing is a reliable solution for the fabrication of sensors and biosensors [20–23], making the technology advantageous for the development of green electronics [24] and reliable, easy-to-use diagnostic tests. Specifically in the biosensing field, IJP benefits from its versatility and high resolution for the development of prototypes. However, IJP of biomolecules presents some chal- lenges, especially regarding molecular stability. In this respect, non-specific adsorption of biomolecules onto the ink chamber walls and maintaining stability during printing, where the molecules suffer thermal and/or mechanical stress, have been reported as two major issues for IJP of biomolecules [21,25]. A range of additives have been employed in order to maintain the stability of biomolecules [26] with varying results. Although it is expected that some enzymes immobilized on a surface might exhibit a lower catalytic activity, it is well known that surface-immobilization often leads to improved stability, since their aggregation is no longer possible. Furthermore, as indicated by Rodrigues et al. [27], enzyme rigidification may lead to preservation of the enzyme properties under drastic conditions in which the enzyme tends to become distorted, thus decreasing its activity. Finally, as has been reported by Hoarau et al., methods of enzyme attachment can be fine-tuned so that activity and stability can be greatly enhanced [28]. On the other hand, carbon-based nanomaterials, especially carbon nanotubes (CNTs) and graphene, have been recognized as convenient materials for the construction of biosen- sors, since they can be employed as solid supports for immobilization of biomolecules and also provide a high electronic conductivity [29]. The high surface to volume ratio of CNTs enables an increase in the number of immobilized biomolecules, as well as reducing the time response of the biosensors [30]. In order to immobilize enzymes onto CNTs, they are usually oxidized to produce carboxylic groups that can act as anchoring points. For instance, a usual method to immobilize proteins onto CNTs is via the carbodiimide crosslinking reaction between carboxylic groups formed on the CNTs and free amino groups of enzymes [31]. However, it is well known that the oxidation of CNTs has a negative impact on their conductive and mechanical properties. Therefore, alternative methods of immobilization of biomolecules, which do not affect CNT electronic properties, are of interest. The use of particulated carriers such as polystyrene microspheres, onto which biomolecules can be immobilized, has been proposed as an alternative to direct linking to CNTs for the preparation of inks for printed enzyme electrodes [32]. This approach allows enzyme immobilization without significantly affecting the electrical and electrochemical behavior of CNTs. However, inks containing microspheres are not suitable for IJP, since particles with a size higher than 200 nm cause nozzle clogging. This problem might be solved by the use of an adequate nanocarrier and, in this sense, silica nanoparticles seem a convenient material of choice. Silica nanoparticles can be prepared by the Stöber method with a low dispersity and a controlled size between 9 and 800 nm [33], amenable to IJP. Furthermore, the surface of silica particles can be easily functionalized to provide amino or carboxylic groups onto which enzymes can be covalently linked [34]. In this paper, we propose a general approach to increase the stability of biomolecules in inks and during the printing by immobilizing enzymes onto silica nanoparticles. The incorporation of these nanoparticles in a single walled carbon nanotube (SWCNT) conduc- tive ink allows a simple and general procedure for the preparation of IJP enzyme electrodes. Our results show that this approach could be used to immobilize a considerable quantity of biomolecules without significant loss of their biological activity. As a proof a concept, we present the results for the preparation of an enzyme electrode prepared with horseradish peroxidase (HRP). Nanomaterials 2021, 11, 1645 3 of 15 2. Materials and Methods 2.1. Synthesis of SiO2 Nanoparticles Spherical SiO2 nanoparticles (SNPs) were prepared by a sol-gel process by controlled hydrolysis of tetraethyl orthosilicate (TEOS) in batch synthesis at room temperature. Firstly, 13.5 mL of deionized H2 O, 24.5 mL of anhydrous EtOH and 1.22 mL of concentrated NH3 were placed together in a glass flask. Then 830 µL of TEOS were quickly added under vigorous magnetic stirring (800 rpm) and the flask was closed to avoid reagents evaporation. After 15 min, the solution turned into a pale blue-white solution indicating the presence of colloidal SiO2 . The solution was left covered overnight on a rotary mixer. To remove by-products and solvents after the synthesis of the silica particles, they were centrifuged at 3000 rpm for 30 min (Mikro 1200, Hettich, Tuttlingen, Germany) and gently resuspended in 10 mL deionized water. After repeating the washing process three times, hydrodynamic size was measured by dynamic light scattering (DLS) with a DynaProNanostar from Wyatt Technology (Santa Barbara, CA, USA). Finally, the particles were dried until a solid was obtained, and then they were placed in an oven at 120 ◦ C and under vacuum for 48 h. After this stage, they were allowed to cool and then resuspended in H2 O. The solid was placed in a FalconTM type tube and H2 O was added until total resuspension was achieved. Finally, they were centrifuged again at 6000 rpm for 20 min with the addition of absolute EtOH twice. All chemicals were of reagent grade and used without further purification. The size and distribution were analyzed using field emission scanning electron microscopy (Quanta 250, FEI, Waltham, MA, USA). 2.2. Preparation of SNP-HRP 2.2.1. Functionalization of SNPs Firstly, the dried silica nanoparticles were treated at 120 ◦ C for 24 h to ensure the consolidation of the inorganic structure. Then, the particles were resuspended in 1.5 mL of EtOH and 200 µL of APTES was added. The reaction was allowed to stir overnight to obtain NH2 modified silica particles. Subsequently, they were centrifuged at 14,000 rpm for 5 min and washed with EtOH and three times with dimethylformamide (DMF). The SiO2 -NH2 NPs were modified into carboxylic groups. Although several methods of cova- lent enzyme immobilization may be used, our experience shows that enzymes conjugated using carbodiimide present a higher activity. Thus, amino groups in the NPs were changed to carboxylic groups in order to link them to amino residues in the enzyme forming an amide covalent link. The NH2 -modified particles dispersed in DMF were added to a previously prepared solution of 5 mL of 1% anhydrous succinic and 1.9 mL of pyridine. The mixture was stirred overnight. The nanoparticles were washed twice in DMF and resuspended in 5% HCl for 5 min. Finally, the particles were centrifuged at 8000 rpm for 10 min and resuspended in deionized water until the pH of deionized water was reached. The NPs were resuspended in 2 mL of 0.1 M phosphate buffer pH 7.0. 2.2.2. HRP Immobilization onto SiO2 Particles HRP was immobilized by coupling the amine groups of the enzyme with the car- boxylic groups of the SNPs. Firstly, the carboxylic groups were activated by adding 500 µL of a mixture of 0.1 M EDC and 25 mM NHS (in 0.1 M phosphate buffer pH 7.0) to 500 µL of COOH-modified particles. The mixture was incubated for 60 min with low vortexing. It was washed twice with 500 µL of 0.1 M phosphate buffer solution of pH 7.0 and sub- sequently 80 µL of HRP (0.002 g/80 µL in phosphate buffer) was added. They were left under stirring for 2 h at room temperature. Then, the particles were centrifuged for 20 min at 3000 rpm to avoid enzyme denaturation. The supernatant was removed and gently re- suspended in 500 µL of phosphate buffer. The above process was repeated 5 times. Finally, the SNPs were resuspended in 4 mL of phosphate buffer and stored in a refrigerator. Nanomaterials 2021, 11, 1645 4 of 15 2.3. SWCNT Aqueous Ink Single-walled carbon nanotubes (SWCNT) carboxylic acid functionalized with 90% carbon basis, with a diameter of 4–5 nm and a length of 0.5–1.5 µm (bundle dimensions, Sigma-Aldrich, St. Louis, MO, USA), sodium dodecyl sulfate (Sigma-Aldrich Chemie GmbH, Schnelldorf, Germany) and deionized water (conductivity less than 1 µS cm−1 ) were used for the preparation of the aqueous ink. Two inks were formulated with dif- ferent concentrations of SWCNT: 7.5 mg/mL and 5 mg/mL. The first (SWCNT-7.5) was formulated with 11.2 mg of SWCNT mixed with 8 mg of sodium dodecyl sulfate (SDS), which acts as a surfactant, in an Eppendorf® tube with 1.5 mL of deionized water at room temperature; while the second ink (SWCNT-5) was formulated with 7.1 mg of SWCNT mixed with 5.3 mg of SDS in 1.5 mL of deionized water. These mixtures were sonicated for 30 min and then centrifuged at 18,000 rpm for 10 min. The supernatant was used. The inks were allowed to stabilize for at least 12 h and then sonicated it again for 30 min before use. Both inks were prepared to be printed with IJP. 2.4. SWCNT-SNP-HPR Bio-Inks An aqueous bio-ink was formulated by incorporating silica nanoparticles (SNP, with a mean diameter of 66 nm ± 8 nm) with covalently immobilized HRP to the SWCNT ink prepared as previously described, fulfilling the necessary rheological conditions for IJP. In order to find an adequate formulation with good electrochemical response and easy printing, several inks were prepared combining the two carbon nanotubes inks (SWCNT-7.5 and SWCNT-5) with different concentrations of the enzyme-functionalized silica nanoparticles (SNP-HRP) (Table 1). Table 1. Electrodes printed with different combinations of SWCNT and SNP-HRP concentrations and number of layers, including the estimated masses of SWCNT and SNP-HRP deposited on the electrodes. SWCNT SNP-HRP Number of SWCNT Mass SNP-HRP Mass Concentration Concentration Layers (µg) (µg) (mg/mL) (mg/mL) 6 1.5 0 7.5 0 12 3.0 0 18 4.5 0 6 1.0 0 5 0 12 2.0 0 18 3.0 0 6 1.5 0.2 7.5 1 12 3.0 0.4 6 1.0 0.2 5 1 12 2.0 0.4 6 1.0 1.0 5 5 12 2.0 2.0 2.5. IJP Electrodes Process A three-electrode electrochemical cell was designed so that the dimensions were com- patible with the electrodes used for a previously reported portable multipotentiostat [34] (Supplementary Information Figure S1). Printing patterns were made using the Electronic Design Automation (EDA) layout software and imported with the Dimatix Bitmap edi- tor software. For the development of the conductive path, the working electrodes (WE) with a geometric surface area (GSA) of 0.78 mm2 and counter electrode (CE) with a GSA 5.25 mm2 , a commercially low curing Au nanoparticle ink were used (Drycure Au-JB 1010B, C-INK Co., Okayama, Japan). For the reference electrode (RE) with a GSA 2.55 mm2 , a silver nanoparticle ink (DGP-40LT-15C from the firm ANP, Sejong, Korea) was used. The passivation and protective layer of the electrodes was done using a dielectric PriElex® Nanomaterials 2021, 11, 1645 5 of 15 SU-8 ink (MicroChem, Westborough, MA, USA). Finally, a circular area with a diameter of 1 mm was printed onto the WE with the carbon nanotubes water-based ink. All inks show drop-on-demand (DoD) inkjet compatible specifications. The manufacturing process was performed without the need of temperature and humidity control in a standard laboratory environment. All electrodes were fabricated on a 125 µm thick polyethylene terephthalate (PET) substrate (Melinex ST504, DuPont Teijin Films, Chester, VA, USA), without any extra surface treatment, using a DoD Dimatix Materials Printer (DMP 2831 from Fujifilm Dimatix, Santa Clara, CA, USA) and three disposable and fillable cartridges containing 16 individually addressable nozzles each with a diameter of 21.5 µm and 10 pL nominal droplet volume (DMC-11610 from Dimatix Fujifilm, Lebanon, NH, USA). Each cartridge was filled with the above commented inks. Several samples of WEs were manufactured, combining different concentrations of carbon nanotubes and functionalized nanoparticles and also different numbers of layers. In the case of WEs printed with bio-ink (SWCNT-SNP-HRP), the reduction of printing times was prioritized so they were only manufactured with 12 and six layers. The details of these combinations can be seen in Table 1. 2.6. Fabrication of Inkjet Printed Electrode Optimization of the IJP process was performed as previously reported [35]. The manufacture of the electrodes was carried out following the sequence of steps as shown in Figure 1, in which SWCNT illustrates the strategy used for the printed electrodes with SWCNT ink and SWCNT-SNP-HRP the WE printed with bio-ink. This second strategy assumed that the enzymes functionalized in the bio-ink could not be exposed to tempera- tures higher than 37 ◦ C, as this could affect the enzyme activity. As a first step (Figure 1-1), we printed the WE, CE and conductive paths with the Au nanoparticles ink, setting a drop spacing (DS) of 15 µm, which is equivalent to a printing resolution of 1693 dpi (dots per inch). Subsequently, the cartridge containing the Au ink was replaced with one containing Ag nanoparticle ink to print the pseudo reference electrodes (pRE) (Figure 1-2). In this case, the printing was done with a DS of 40 µm (resolution of 635 dpi). They were then subjected to thermal drying of 90 ◦ C for 10 min and then to a sintering process in an oven 30 min at 120 ◦ C (Figure 1-3). This step is where the final electrical properties of the inks are achieved. The pRE was chlorinated by cyclic voltammetry (CV) in 0.1 M HCl, scanning potential from 0 to 0.2 V against Ag/AgCl commercial RE (Metrohm, Herisau, Switzerland Germany) and Pt CE (Metrohm) at 20 mV/s to obtain a stable pRE (Figure 1-4) [36]. For the WE printed with the SWCNT ink, the next step was the printing of the WE area of 1 mm in diameter with the CNT water-based ink with a DS of 15 µm (1693 dpi) (Figure 1-5.A). Subsequently, they were subjected to a drying process of 20 min at 120 ◦ C and then sintering at 140 ◦ C for another 20 min (Figure 1-6.A). Afterwards, the PriElex® SU8 dielectric ink was printed with a DS of 15 µm (1693 dpi) as a protective layer for the conductive paths and delimiting the active area for the WE and the contact pads of the electrodes (Figure 1-7.A). In this case, the curing was carried out first on a hot plate at 100 ◦ C to evaporate the solvents and then with UV lamp exposure for 30 s, to generate the polymerization of the ink by cross-linking (Figure 1-8.A). For the electrodes printed with the bio-ink (SWCNT-SNP-HRP), after the stage of shaping the conductive paths of the electrodes, the dielectric layer was also printed with PriElex® SU8, in the same way as previously described (Figure 1-5.B,6.B). In the last step, the bio-ink was printed on the active area of WE also with a DS of 15 µm (Figure 1-7.B), but in this case the curing was carried out at room temperature for 24 h (Figure 1-8.B). This curing strategy assumed that the bio-ink enzymes could not be subjected to higher than room temperature in order to avoid damage and losing enzymatic activity in the electrodes. Nanomaterials 2021, 11, 1645 6 of 15 Nanomaterials 2021, 11, x FOR PEER REVIEW 6 of 16 Figure1.1.(Top) Figure (Top) Fabrication Fabrication steps steps of three-electrode of three-electrode electrochemical electrochemical cellon cell based based on IJP. SWCNT: IJP. SWCNT: WE WE printed printed only with SWCNT and SWCNT-SNP-HRP: WE printed with bio-inks. (Bottom) Photograph only with SWCNT and SWCNT-SNP-HRP: WE printed with bio-inks. (Bottom) Photograph of the printed of the printed electrodes steps and several all inkjet printed electrochemical sensors onto a flexible electrodes steps and several all inkjet printed electrochemical sensors onto a flexible substrate. substrate. 2.7. Microscopic and Electrical Characterization of Electrodes 2.7. Microscopic and Electrical Characterization of Electrodes The printed electrodes were morphologically characterized with 3D optical perfilome- The printed ter, confocal electrodes mode (MS10 PLµNEOX were Sensorfare, morphologically Terrassa,characterized Barcelona, Spain)with and 3Doptical optical perfilometer, confocal mode (MS10 PLµ NEOX Sensorfare, Terrassa, microscopy (DM 4000M, Leica, Tokyo, Japan). WE diameters were measured with the Barcelona, Spain) and optical microscopy (DM 4000M, Leica, Tokyo, Japan). WE diameters ImageJ image processing program (LOCI, University of Wisconsin, Madison, WI, USA). were measured with the ImageJ Scanning imageMicroscopy Electron processing(SEM,program (LOCI,Carl Auriga-40, University of Wisconsin, Zeiss, Jena, Madison, Germany) images wereWI, USA). also Scanning taken Electron of surface and Microscopy (SEM,made the cross section Auriga-40, with Carl Zeiss, Jena, a Focused Germany) Ion Beam (FIB, images Zeiss were also 1560XB) taken of surface of electrodes printed and thethe with cross section hybrid ink.made SEMwith imagesa Focused Ion Beam were obtained (FIB, from Zeiss a cast 1560XB) drop of electrodes deposition of SNP printed with a FEIwith the hybrid Quanta ink.field 250 cold SEM SEM images were obtained microscope operatedfrom at 30a kV cast and the mean diameter was determined as the average size of 100 nanoparticles. Resistance30 drop deposition of SNP with a FEI Quanta 250 cold field SEM microscope operated at ofkV and the mean a conductive layerdiameter was determined was determined as the average with a semiconductor size of analyzer parameter 100 nanoparticles. (B1500A, Resistance Agilent, Santa of Clara, a conductive CA, USA) layer was determined connected with a semiconductor to a semi-automatic parameter probe station (Cascade analyzer SUMMIT Microtech (B1500A, 12161B-6, Agilent, Santa Clara,OR, Beaverton, CA,USA). USA)The connected resistanceto of a semi-automatic probe the conductive paths station was (Cascade measured withMicrotech two tips ofSUMMIT the probe 12161B-6, Beaverton,toOR, station connected twoUSA). SMUsThe resistance of the B1500Aof the conductive between both ends. paths was measured A fixed voltage waswith two tips injected of theatprobe (starting with − 0.01 Vstation connected to two 0.0001 V steps) and SMUsthe of current throughbetween the B1500A the pathboth wasends. measured. A fixed voltage was injected (starting at 0.01 V with −0.0001 V steps) and the current through the path was measured. 2.8. Electrochemical Measurements 2.8. The Electrochemical Measurements electrochemical characterization of the sensors was performed with an 8-channel potentiostat 1030A Electrochemical Analyzer The electrochemical characterization of (CH Instruments, the sensors Bee Cave,with was performed TX, anUSA). Con- 8-channel trol experiments were performed using a commercial Ag/AgCl (3 M KCl) RE potentiostat 1030A Electrochemical Analyzer (CH Instruments, Bee Cave, TX, USA). (DRIREF-2, World ControlPrecision Instrument, experiments were Sarasota, performed FL,using USA)aand 99.9% platinum commercial Ag/AgClwire(3 (Alfa Aesar, M KCl) RE GmbH (DRIREF-2, World Precision Instrument, Sarasota, FL, USA) and 99.9% platinumand & Co KG, Karlsruhe, Germany) as CE. All reagents were of analytical grade wire used (AlfaasAesar, received. GmbH & Co KG, Germany) as CE. All reagents were of analytical grade and used as received. Nanomaterials2021, Nanomaterials 2021,11, 11,1645 x FOR PEER REVIEW 77of 16 of15 3. Results and Discussion 3. Results and Discussion The main goal of this work was to develop an ink formulation with enzymes The main goal of this work was to develop an ink formulation with enzymes im- immobilized onto nanoparticles in order to increase the stability of the enzymes in inkjet mobilized onto nanoparticles in order to increase the stability of the enzymes in inkjet printed electrodes. However, several issues had to be taken into account to achieve this printed electrodes. However, several issues had to be taken into account to achieve this end. The most challenging ones were to achieve good printability avoiding nozzle end. The most challenging ones were to achieve good printability avoiding nozzle obstruc- obstruction, a good electrochemical response towards redox probes, and an adequate tion, a good electrochemical response towards redox probes, and an adequate analytical analytical performance of the enzyme-based electrodes. Therefore, the choice of performance of the enzyme-based electrodes. Therefore, the choice of nanomaterials nanomaterials with with optimumtheir optimum characteristics, characteristics, concentration, their andconcentration, the number ofand the layers printed number wasof printed layers carefully was carefully considered considered and assessed. and assessed. Figure 2 shows optical microscopy Figure 2 shows optical microscopy images imagesof ofdifferent differentWE: WE:(a) (a)aaAuAuelectrode, electrode,(b)(b)aa 12-layer SWCNT (7.5 mg/mL) electrode, and (c) a 12-layer SWCNT-SNP-HRP 12-layer SWCNT (7.5 mg/mL) electrode, and (c) a 12-layer SWCNT-SNP-HRP electrode electrode (with55mg/mL (with mg/mL andand55 mg/mL mg/mL concentration concentration ofof SWCNT SWCNT and and SNP-HRP, SNP-HRP,respectively). respectively).These These are typical images representative of the obtained electrodes. Both Au and are typical images representative of the obtained electrodes. Both Au and SWCNT printed SWCNT printed electrodespresented electrodes presentedaahomogeneous homogeneoussurfacesurfaceininthe theoptical opticalmicroscopy microscopyimages. images.However, However, in Figure 2c, printing lines can be observed on the SWCNT-SNP-HRP in Figure 2c, printing lines can be observed on the SWCNT-SNP-HRP electrode. The electrode. The characteristic gold color is seen in the Au electrode, while both SWCNT electrodes characteristic gold color is seen in the Au electrode, while both SWCNT electrodes present present aadark darkcolor colorcompletely completelycovering coveringthetheunderlying underlyingAu Ausurface. surface.The TheSWCNT SWCNTelectrode electrodeinin Figure 2b is darker than the SWCNT-SNP-HRP one, probably due to Figure 2b is darker than the SWCNT-SNP-HRP one, probably due to the different mass the different massofof SWCNTdeposited SWCNT deposited(3 (3µg µ gvs. vs.22µg). µ g).As Asthe the number numberof of printed printed layers layers of of SWCNTs SWCNTsincreases, increases, theblack the blackcolor colorbecomes becomesmoremorehomogeneous homogeneousand andintense intenseon onthe theoverall overallelectrode electrodeareaareadue due tothe to thepresence presenceof ofnanotubes nanotubes(Supplementary (SupplementaryInformation InformationFigure FigureS2). S2). Figure2.2. Optical Figure Optical microscopy microscopy images imagesof ofprinted printedelectrodes: electrodes:(a)(a) 1-layer Au, 1-layer (b)(b) Au, 12-layer SWCNT 12-layer (7.5 SWCNT mg/mL of SWCNT), and (c) 12-layer SWCNT-SNP-HRP (5 mg/mL of SWCNT and 5 mg/mL of SNP- (7.5 mg/mL of SWCNT), and (c) 12-layer SWCNT-SNP-HRP (5 mg/mL of SWCNT and 5 mg/mL of HRP). SNP-HRP). Theelectrodes The electrodeswereweredesigned designedtotohavehavean anarea areaofof0.78 0.78mm mm22and andaadiameter diameterof of1000 1000µm.µ m. The actual measured diameters were 1064 µ m, 1126 µ m and The actual measured diameters were 1064 µm, 1126 µm and 1129 µm for Au, SWCNT and 1129 µ m for Au, SWCNT and SWCNT-SNP-HRP electrodes, respectively. Therefore, the dimensions, SWCNT-SNP-HRP electrodes, respectively. Therefore, the dimensions, as well as the area as well as the area ofthe of the electrodes, electrodes, werewere close close to thoseto ofthose of thedesign, the original original design, which confirm which confirm the the compatibility of the materials and the printing process. Finally, it can be seen that the insulatingthat compatibility of the materials and the printing process. Finally, it can be seen the layers insulatingcovers effectively layers the effectively conductive covers pathsthe and conductive delimits an paths and delimits approximate an area. circular approximate circular The area. resistance along the Au printed paths with a length of 13.5 mm was measured and an The resistance average alongvalue resistance the Au of 63 ± 12 Ωpaths printed with adeviation) (standard length ofwas13.5obtained. mm wasTherefore, measured and an average negligible ohmic resistance falls were to value of 63 ± 12 be expected dueΩ (standard deviation) to the conductive was paths. obtained. Typical Therefore, current values negligible ohmic falls were to be expected due to the conductive were in the order or below 10 µA, so IR drops in the conductive paths were below 1 mV. paths. Typical current valuesThewere in the concentration SWCNTs order or belowused 10 µ A, so IR in the drops inks hadinathe conductive major effect onpaths were below the printability. 1 mV.concentrations higher than 7.5 mg/mL were used, the nozzles clogged during the When The SWCNTsElectrodes first impressions. concentrationcouldused in the inks be printed using had a majorinks SWCNT effect with onathe printability. concentration of 7.5 mg/mL. When However, concentrations afterthan higher a few 7.5(from mg/mL 5 to 8) days were used,ofthe use,nozzles the nozzles clogged clogged duringand the the firstprinthead had Electrodes impressions. to be discarded. could be In printed contrast, printability using SWCNTgreatly inks withimproved when theof a concentration SWCNT 7.5 mg/mL. concentration However, after was alower than 5 to few (from mg/mL, 8) daysbeing of use,possible to print the nozzles for at clogged andleast the 3printhead months keeping had to the be nozzles discarded. working. In addition, In contrast, inks prepared printability with this when greatly improved SWCNT the concentration SWCNT concentrationwere less was sensitive lowertothanprinting 5 mg/mL,parameters such as to being possible waveform, print for voltages at least 3 applied monthstokeeping the nozzle, theetc., whichworking. nozzles had to be In carefully chosen addition, inksforprepared more concentrated with this SWCNT SWCNT inks (Supplementary concentration were lessInformation sensitiveFigure S3). The to printing final mass such parameters of deposited SWCNT as waveform, could voltages be estimated using the concentration of SWCNT, drop spacing and the volume of ink applied to the nozzle, etc., which had to be carefully chosen for more concentrated SWCNT inks (Supplementary Information Figure S3). The final mass of deposited Nanomaterials 2021, 11, 1645 SWCNT could be estimated using the concentration of SWCNT, drop spacing and8 of the 15 volume of ink deposited per layer, as reported in Table 1. Since the final response of the SWCNT electrodes depended on the amount of deposited SWCNT, electrodes printed with 12 layers and a SWCNT concentration of 7.5 mg/mL were equivalent to those printed deposited with per layer, 18 layers with aasconcentration reported in Table of 51.mg/mL. Since the final response Therefore, of the a better SWCNT electrodes printability could be depended on the amount of deposited SWCNT, electrodes printed obtained with a SWCNT concentration of 5 mg/mL at the expense of the need with 12 layers and a of printing SWCNT concentration more layers. of 7.5 mg/mL were equivalent to those printed with 18 layers with a concentration All printed electrodes were electrochemically tested in different solutions in order toa of 5 mg/mL. Therefore, a better printability could be obtained with determineconcentration SWCNT of 5 mg/mL their capacitance, at the expense ofactive the electrochemically the need area,ofand printing more layers. the electrochemical All printed reversibility electrodes towards were electrochemically redox probes, tested in (III/II) such as hexacyanoferrate different andsolutions in order hydroquinone. to determine their capacitance, the electrochemically active area, and the electrochemical Firstly, the capacitive behavior of the electrodes was studied by CV. Figure 3a shows reversibility towards redox probes, such as hexacyanoferrate (III/II) and hydroquinone. the voltammograms obtained in phosphate-buffered saline (PBS) solution for a Au Firstly, the capacitive behavior of the electrodes was studied by CV. Figure 3a shows electrode and SWCNT printed electrodes on a Au. A capacitive behavior can be observed the voltammograms obtained in phosphate-buffered saline (PBS) solution for a Au electrode for all electrodes. A significant increase in the capacitive current with the number of and SWCNT printed electrodes on a Au. A capacitive behavior can be observed for all printed layers is observed and can be ascribed both to the higher roughness of the electrodes. A significant increase in the capacitive current with the number of printed layers electrodes as observed by confocal microscopy (discussed below) as well as the high is observed and can be ascribed both to the higher roughness of the electrodes as observed surface to volume ratio characteristic of nanomaterials. The values of capacitance obtained by confocal microscopy (discussed below) as well as the high surface to volume ratio as the slope capacitive current vs. scan rate were 320, 770 and 1420 µ F cm−2 for, 6-, 12- characteristic of nanomaterials. The values of capacitance obtained as the slope capacitive (Supplementary Information Figure S4), and −18-layer SWCNT printed electrodes, current vs. scan rate were 320, 770 and 1420 µF cm 2 for, 6-, 12-(Supplementary Information respectively. The obtained average value of specific capacitance of these SWCNT Figure S4), and 18-layer SWCNT printed electrodes, respectively. The obtained average electrodes was 2.6 F g−1 , in accordance with previous reports stating typical values range value of specific capacitance of these SWCNT electrodes was 2.6 F g−1 , in accordance with from 2 to 45 previous F g−1 stating reports [37]. typical values range from 2 to 45 F g−1 [37]. Figure 3. Figure Electrochemical response 3. Electrochemical response of of printed printed electrodes. electrodes. (a) (a) CVs CVs in in PBS PBS buffer buffer of of pH pH 7.4 7.4 obtained obtained for for Au Au and and 6-, 6-, 12-, 12-, and and 18-layer SWCNT printed electrodes electrodes prepared prepared with with aa concentration concentrationof of7.5 7.5mg/mL. mg/mL. (b) CVs obtained for Au and 6-, 12-, 12-, and and 18-layer SWCNT 18-layer SWCNTprintedprintedelectrodes electrodesprepared preparedwith a concentration with of 7.5 a concentration ofmg/m in a in 7.5 mg/m solution of 10 of a solution mM 10hexacyanoferrate (III/II) mM hexacyanoferrate (III/II) and 0.1and M KNO0.1 M KNO 3 . (c) CVs3. obtained (c) CVs for obtained for Au Au printed printeda 12-layer electrode, electrode, a 12-layer of 7.5 mg/mL of 7.5 mg/mLofconcentration concentration SWCNT printed of electrode SWCNT printed electrode and a 12-layer and a 12-layerprinted SWCNT-SNP-HRP SWCNT-SNP-HRP electrode with printed 5 mg/mL electrode with 5concentration and 5 mg/mL mg/mL andof5SWCNT mg/mLand concentration SNP-HRP inofa SWCNT and solution of SNP-HRP 4 mM in a solution hydroquinone in PBSofof 4 mM hydroquinone pH 7.4. in PBS ofwere All voltammograms pH 7.4. All voltammograms acquired at a scan rate of were 0.05 Vacquired s−1 . at a scan rate of 0.05 V s−1. Secondly, the electrochemical reversibility of the electrodes was assessed using hexa- Secondly, cyanoferrate the electrochemical (III/II) and hydroquinone reversibility as redox probes.of theFigure electrodes wasthe 3b shows assessed using CVs obtained hexacyanoferrate for a Au and three(III/II) SWCNT andprinted hydroquinone electrodes,as redox with 6,probes. 12 andFigure 3b shows 18 layers, the CVs in a solution of obtained for a Au and three 10 mM hexacyanoferrate SWCNT (III/II). printed It can be seen thatelectrodes, with 6, 12 and the oxido-reduction 18 layers, in a of hexacyanoferrate solution of 10 (III/II) was mM hexacyanoferrate mostly inhibited in as-printed (III/II).AuIt electrodes, can be seensincethatnothesignificant oxido-reduction anodic of or hexacyanoferrate (III/II) was mostly inhibited in as-printed cathodic peaks can be observed. The initial electrochemical response could be Au electrodes, since no greatly significant improved by anodic or cathodic cleaning peaks the surface ofcan the be Auobserved. electrodes, The initial electrochemical a process known as surface response activa- could be greatly tion [35]. improvedthe After activating by gold cleaning the surface surface, anodic of the and Au electrodes, cathodic a process peaks could known be obtained; as surface the however, activation observed [35]. peakAfter activating potential the gold difference surface, of 214 mV wasanodic andhigh. rather cathodic peaks In contrast, could be obtained; well-defined however,peaks voltammetric the observed peak potential were obtained difference for SWCNTs of 214 mV electrodes. was rather Potential peak high. In contrast, differences of 91, well-defined 101 and 179 mV voltammetric were obtained peaks were6-,obtained when 12- and for SWCNTs 18-layer electrodes. electrodes were Potential peak differences used, indicating that electronof 91, 101 and transfer 179 mVpresented reactions were obtained when a greater 6-, 12- degree of and 18-layer electrochem- ical reversibility electrodes for aindicating were used, 6-layer SWCNT electrode. that electron transferAs reactions can be seen, peak current presented a greaterincreases degree with the amount of SWCNT deposited on the electrode, since the peak current peaks for a 18-layer electrode was more than twice as much as the 6-layer SWCNT one. This considerable increase in peak current observed with the increasing amount of SWCNT can be attributed to an increase in the electroactive area of the electrode. The electroactive area A could be determined by recording CVs at different scan rates (Supplementary Information, Figure S5a,b). As expected, the current peak followed a linear Nanomaterials 2021, 11, 1645 9 of 15 dependence (Supplementary Information Figure S5c) on the square root of the scan rate as expressed by the classic Randles-Sevcik Equation (1) [38]:  1/2 3/2 3/2 D i p = 0.4463 n F AC v1/2 (1) RT where ip is the current peak, n the number of electrons transferred in the redox event, F the Faraday constant, D the diffusion coefficient, C the concentration of the analyte, ν the scan rate, R the gas constant and T the temperature. Electroactive areas of 1.8, 2.6 and 4.1 mm2 were determined for 6-, 12- and 18-layer SWCNT-7.5 printed electrodes. Therefore, SWCNT presented an electroactive area higher than the geometrical area. This result might be connected to the high roughness of the electrode surface [39]. In order to study this hypothesis, SEM, FIB cross-sections and confocal images of Au and SWCNT printed electrodes were acquired to evaluate the thickness and roughness of these electrodes. Confocal images in Figure 4 show average thickness and roughness (standard devia- tion) values for 1-layer Au; 6, 12 and 18-layer SWCNT-5 and SWCNT-7.5 printed electrodes, whose values are detailed in Table 2. Table 2. Average thickness and roughness of electrodes printed with different combinations of SWCNT concentrations and number of layers and a layer of Au printed. Concentration of Number of Thickness Roughness Printed Electrode SWCNT [mg/mL] Layers (ave) [nm] (sd) [nm] 6 1100 315 Au + SWCNT 5 12 1148 340 18 1452 342 6 1118 261 Au + SWCNT 7.5 12 1429 290 18 2107 384 Au 0 1 230 65 Figure 5a shows the SEM image of the surface of a SWCNT printed electrode, in which the fibrous nature of the CNT can be seen. The cross section made by FIB (Figure 5b) shows the compact layer formed by the printing of Au nanoparticle ink (orange shaded), forming a uniform film of 430 nm thickness with a low roughness. On the contrary, SWCNT printed films with a thickness of 1.4 µm (blue shaded) show a rough surface with the presence of voids in the bulk of the stack layers. Considering that confocal images were taken on dried electrodes, the increased rough- ness and electroactive area observed for SWCNT electrodes suggest that SWCNT can penetrate in the solution beyond the diffusion layer and were able to collect redox species from a higher volume than that allowed for planar electrodes under semi-infinite diffusion conditions. Thus, the high values obtained for the electroactive area suggests that SWCNT were not confined to the electrode surface and were able to react with redox species beyond the diffusion layer as usual for planar electrodes. Figure 3c shows the CVs obtained for an Au printed electrode and a 12-layer SWCNT printed electrode in a solution containing 4 mM hydroquinone in a PBS buffer of pH 7.4. For Au electrodes used without further treatment, the oxidation of hydroquinone was mostly inhibited. Since direct reaction on the underlying Au substrate was mostly absent, the electrochemical response can be attributed to SWCNT, whose electrocatalytic properties to the oxidation of hydroquinone have been previously reported [31]. It is worth noting that the addition of silica nanoparticles did not inhibit the oxidation of hydroquinone by SWCNT. As can be seen in Figure 3c, the voltammogram for SWCNT-SNP-HRP elec- trodes presented a lower peak current for the oxidation of hydroquinone, possibly as a consequence of a lower electroactive area as the silica nanoparticles are dielectric. Nanomaterials 2021, Nanomaterials 2021,11, 11,1645 x FOR PEER REVIEW 10 of 16 10 of 15 Figure 4. Figure 4. Confocal Confocal relief relief images images and and profilometry profilometry of of WE WE printed printed with with (a) (a) 6-, 6-, (b) (b) 12-, 12-, (c) (c) 18-layer 18-layer of of SWCNT-5 SWCNT-5 ink ink (left) (left) and and (d) 6-, (e) 12-, (f) 18-layer of SWCNT-7.5 ink (right). (d) 6-, (e) 12-, (f) 18-layer of SWCNT-7.5 ink (right). Figure In 5a shows conclusion, thevoltammetric this SEM image of the surface analysis showsofthat a SWCNT SWCNTprinted printedelectrode, in electrodes which the fibrous nature of the CNT can be seen. The cross section made by present a high capacitance, with values typical for this material. With regard to the FIB (Figure 5b) shows the compact electrochemical layer reversibility, theformed byofthe oxidation printing of Au(II) hexacyanoferrate nanoparticle ink (orange and hydroquinone was mostly inhibited in as-printed Au electrodes. In contrast, SWCNT-7.5 presented aOn shaded), forming a uniform film of 430 nm thickness with a low roughness. the good contrary, SWCNT printed films with a electrochemical response for both redox probes.thickness of 1.4 µ m (blue shaded) show a rough surface with Since the6-presence both of voids and 12-layer in the showed SWCNT bulk of the stackelectrochemical a good layers. response, these numbers of layers were maintained during the fabrication of enzyme electrodes with enzymes immobilized onto nanoparticles. SiO2 nanoparticles were chosen because they can be easily prepared with a high degree of quality by the Störber sol-gel process and with Nanomaterials 2021, 11, 1645 11 of 15 Nanomaterials 2021, 11, x FOR PEER REVIEW 11 o a size amenable to IJP (Supplementary Information Figure S6a). Besides, the surface of silica nanoparticles can be functionalized to create anchoring points for biomolecules. Figure 5. Figure 2-layer 5. Au2-layer WE printed Au WE with 12-layer printed of 12-layer with 7.5 mg/mLof of 7.5SWCNT mg/mL(a) ofSEM image SWCNT (a)ofSEM the top viewofofthe image itstop surface and (b) the corresponding cross section. view of its surface and (b)Authelayers are shadedcross corresponding in orange andAu section. thelayers SWCNTs ones arein are shaded shaded orangeinand blue.the SWCNTs ones are shaded in blue. Considering that confocal images were taken on dried electrodes, the increa roughness and electroactive The enzyme-immobilized area observed silica nanoparticles SWCNT for SWCNT electrodes ink electrodes were suggest testedthat SWC for the detectioncanofpenetrate hydrogeninperoxide, the solution usingbeyond the diffusion hydroquinone as layer a redox andmediator. were ableThe to collect re species from reduction of hydrogen a higher peroxide, volumeby catalyzed than HRP,thatisallowed for planar accompanied by electrodes the oxidationunder of semi-infi hydroquinone, diffusion which acts conditions. as an electronThus,donorthe high valuesformed [31].The obtained for the species oxidized electroactive can bearea sugg electrochemicallythat SWCNT detected wereby either not CV confined to the electrode surface or by chronoamperometry if theand were able electrode to react with re potential species beyond the diffusion layer as usual for planar is set at a convenient value. Cathodic linear polarization curves present increasing current electrodes. values with increasing Figure 3c showsperoxide hydrogen the CVs obtained for an Au concentration, as printed can be electrode and a 12-layer seen in Figure 6a. SWC printed electrode in a solution containing 4 mM hydroquinone Figure 6b shows the current transients obtained at a constant applied potential of −0.23 V in a PBS buffer of pH For Au electrodes used without further treatment, before and after the addition of hydrogen peroxide. Similar results were obtained printing the oxidation of hydroquinone 6 and 12 layers, mostly inhibited. albeit slightly Sincecurrent higher direct reaction values were on theobtained underlying Au substrate for 12-layer was mostly abs electrodes. the electrochemical response can be attributed Figure 6c shows the current-concentration curve obtained for the 6-layer of SWCNT- to SWCNT, whose electrocatal SNP-HRP (with 5 mg/mL and 5 mg/mL concentration of SWCNT and SNP-HRP, respec- [31]. properties to the oxidation of hydroquinone have been previously reported worth noting tively) at a measurement timethat thes addition of 60 of silica for electrodes nanoparticles printed 1 day and did90not days inhibit after the the oxidation ink preparation. As can be seen, the performance of both enzyme electrodes was similar.for SWC hydroquinone by SWCNT. As can be seen in Figure 3c, the voltammogram SNP-HRP electrodes presented a lower−2peak current for the oxidation of hydroquin The obtained sensitivity was around 57 +/− 3 µA cm mM−1 , a value comparable to the possibly as a consequence of a lower electroactive area as the silica nanoparticles ones obtained using enzyme electrodes prepared by other means, such as oxygen plasma dielectric. oxidation of carbon-based electrodes followed by enzyme immobilization [31,40]. Addi- In conclusion, this voltammetric analysis shows that SWCNT printed electro tionally, the stability of enzyme electrodes was also evaluated in similar fashion. Enzyme present a high capacitance, with values typical for this material. With regard to electrodes were prepared and calibration curves were obtained after 1, 30 and 60 days electrochemical reversibility, the oxidation of hexacyanoferrate (II) and hydroquin since preparation. As can be seen in Figure 6d, they exhibited a similar sensitivity. The was mostly inhibited in as-printed Au electrodes. In contrast, SWCNT-7.5 presente results presented in Figure 6c,d show that both the bio-ink and the printed enzyme elec- good electrochemical response for both redox probes. trodes were highly stable. Enzyme electrodes were printed presenting a similar sensitivity Since both 6- and 12-layer SWCNT showed a good electrochemical response, th (55 +/− 4 µA cm−2 mM−1 ) and a higher stability as previous approaches. numbers of layers were maintained during the fabrication of enzyme electrodes w It is worth noting that the concentration of SNP-HRP had a significant impact on enzymes immobilized onto nanoparticles. SiO2 nanoparticles were chosen because t the response ofcan thebebiosensor. It was observed that ink formulations prepared using easily prepared with a high degree of quality by the Störber sol-gel process concentrations ofwith a size amenablethan SNP-HRP lower to IJP1 (Supplementary mg/mL were not enough toFigure Information coverS6a). the surface Besides, the sur electrode (Supplementary Informationcan of silica nanoparticles Figure S6b). Therefore, be functionalized these concentrations to create anchoring points offor SNP-biomolecu HRP were not further considered. On the other hand, if the concentration The enzyme-immobilized silica nanoparticles SWCNT ink electrodes were tested was too high, the electrochemical the performance of the SWCNT-SNP-HPR detection of hydrogen peroxide, usingelectrode hydroquinonewas impaired. as a redox When mediator. electrodes withreduction concentrations of 5 mg/mL or higher of SNP-HRP were of hydrogen peroxide, catalyzed by HRP, is accompanied by the oxidatio used, repeated measurements showed an increasing hydroquinone, which actssignal as until a stable, an electron reproducible donor current oxidized [31].The formed value was species can obtained. This result can be interpreted in terms of a certain blockage electrochemically detected either by CV or by chronoamperometry if the electrof the electrochemical activity of SWCNT-SNP-HPR potential is set films, at awhich could bevalue. convenient reversed after repeated Cathodic measurements, linear polarization curves pre possibly due toincreasing a reconfiguration current valuesin thewithfilm structure. increasing On theperoxide hydrogen other hand, electrodesas can be s concentration, printed with SNP-HPR in Figureconcentrations 6a. Figure 6b lower showsthan the 5current mg/mL immediately transients responded obtained to at a constant app hydrogen peroxide generating potential a reproducible of −0.23 V before andelectrochemical after the addition signal as shown in of hydrogen Figure 6c. peroxide. Similar res oxidation of carbon-based electrodes followed by enzyme immobilization [31,40]. Additionally, the stability of enzyme electrodes was also evaluated in similar fashion. Enzyme electrodes were prepared and calibration curves were obtained after 1, 30 and 60 days since preparation. As can be seen in Figure 6d, they exhibited a similar sensitivity. Nanomaterials 2021, 11, 1645 The results presented in Figure 6c,d show that both the bio-ink and the printed enzyme 12 of 15 electrodes were highly stable. Enzyme electrodes were printed presenting a similar sensitivity (55 +/− 4 µ A cm−2 mM−1) and a higher stability as previous approaches. Figure6.6.(a)(a)CVs Figure CVsandand(b)(b)chronoamperometric chronoamperometriccurves curvesforforananapplied appliedpotential −0.23 potentialofof−0.23 VV obtained for obtained for a 6-layer a 6-layer (dotted (dotted lines)lines) and 12-layer and 12-layer (solid (solid lines)printed lines) with with printed SWCNT-SNP-HRP SWCNT-SNP-HRP with a with a concentration concentration of 5 mg/mLof of5SWCNT mg/mL of andSWCNT 5 mg/mL and 5ofmg/mL SNP-HRP,of SNP-HRP, respectively, respectively, withoutwithout and and after after the the of addition addition of H 2O2 to a final concentration of 3 mM in a solution of 4 mM hydroquinone in PBS of pH H2 O2 to a final concentration of 3 mM in a solution of 4 mM hydroquinone in PBS of pH 7.4 at a scan 7.4 at a scan rate−of 0.05 V s−1. (c) Dependence of current density on H2O2 concentration for two rate of 0.05 V s 1 . (c) Dependence of current density on H2 O2 concentration for two SWCNT-SNP- SWCNT-SNP-HRP (6-layer with 5 mg/mL and 5 mg/mL concentration of SWCNT and SNP-HRP, HRP (6-layerelectrodes respectively) with 5 mg/mL printedand with5 mg/mL concentration 1 day (open of SWCNT circles) and 90 daysand(fullSNP-HRP, respectively) circles) after ink electrodes printed preparation. Linear with 1 day (open regression circles)as is shown anda 90dotted days (full line.circles) afterwas Current ink preparation. measured by Linear regression is shown chronoamperometry atas60 as with dottedan line. Current applied wasofmeasured potential −0.23 V in by0.1 chronoamperometry M phosphate buffer ofat pH607.4 s with and 4 mM hydroquinone. an applied potential of −(d) 0.23Dependence V in 0.1 Mof current density phosphate buffer on H2O7.4 of pH 2 concentration for three and 4 mM hydroquinone. SWCNT-SNP-HRP (d) Dependence ofelectrodes measured current density on H1,2 O30 and 60 days after being prepared under same 2 concentration for three SWCNT-SNP-HRP electrodes conditions as described for Figure 6c. measured 1, 30 and 60 days after being prepared under same conditions as described for Figure 6c. ItThis is worth fact noting can bethat the concentration further understood byof SNP-HRP hadmorphological studying the a significant impact on the and deposition response thickness of the material over the electrode. Figure 7a shows the SEM image of theusing of the biosensor. It was observed that ink formulations prepared electrode surface modified with SWCNT-SNP-HPR ink with 7.5 mg/mL and 1 mg/mL concentration of SWCNT and SNP-HRP, respectively. It is possible to distinguish the typical structure that the SWCNTs forms and among them the nanoparticles. When the concentration of SNP-HRP was 5 mg/mL or higher, the top view of the electrode (Supplementary Information, Figure S6c) was completely covered with nanoparticles. In Figure 7b,c, cross sections of 6- and 12-layer of SWCNT-SNP-HPR electrodes are shown. The 2-layer Au, highlighted in orange, presented a thickness of approx. 900 nm, while the thickness of SWCNT-SNP-HPR films was approximately 220 nm and 900 nm for 6-layer (7.5 mg/mL of SWCNTs and 1 mg/mL of SNP-HRP) and 12-layer (5 mg/mL of SWCNTs and 5 mg/mL of SNP-HRP) respectively. To summarize, enzyme electrodes were successfully printed using a bio-ink containing SWCNT and silica nanoparticles bearing immobilized HRP. The number of printed layers greatly influenced the performance of the enzyme electrodes, and the quantity of SWCNT- SNP-HPR deposited had to be optimized in order to obtain a high sensitivity, covering the electrode surface without forming thick films. concentration of SNP-HRP was 5 mg/mL or higher, the top view of the electrode (Suppl. Info., Figure S6c) was completely covered with nanoparticles. In Figure 7b,c, cross sections of 6- and 12-layer of SWCNT-SNP-HPR electrodes are shown. The 2-layer Au, highlighted in orange, presented a thickness of approx. 900 nm, while the thickness of SWCNT-SNP- Nanomaterials 2021, 11, 1645 HPR films was approximately 220 nm and 900 nm for 6-layer (7.5 mg/mL of SWCNTs13and of 15 1 mg/mL of SNP-HRP) and 12-layer (5 mg/mL of SWCNTs and 5 mg/mL of SNP-HRP) respectively. Figure 7. Figure 7. (a)Top (a)Topview viewSEMSEM images images of surface of 12-layer of surface SWCNT-SNP-HRP of 12-layer SWCNT-SNP-HRP printed electrode. printed (b,c) electrode. Focused ion beam cross-section image of electrodes printed showing the different layers and SiO2 (b,c) Focused ion beam cross-section image of electrodes printed showing the different layers and nanoparticles for (b) a 6-layer SWCNT-SNP-HRP (7.5 mg/mL of SWCNTs and 1 mg/mL of SNP- SiO 2 nanoparticles HRP) for (b) printed electrode anda 6-layer SWCNT-SNP-HRP (c) of 12-layer SWCNT-SNP-HRP (7.5 mg/mL of SWCNTs (5 mg/mL of SWCNTs andand 1 mg/mL 5 mg/mLof SNP-HRP) printed electrode and (c) of 12-layer SWCNT-SNP-HRP (5 mg/mL of SNP-HRP) printed electrode. Orange and blue shades correspond to printed Au and SWCNT- of SWCNTs and 5SNP-HRP mg/mL of SNP-HRP) printed films, respectively. electrode. Orange and blue shades correspond to printed Au and SWCNT-SNP-HRP films, respectively. To summarize, enzyme electrodes were successfully printed using a bio-ink 4.containing ConclusionsSWCNT and silica nanoparticles bearing immobilized HRP. The number of printed layerselectrodes Enzyme greatly influenced the performance were successfully of theaenzyme printed using bio-ink electrodes, containing and SWCNT the quantity and silicaof SWCNT-SNP-HPR nanoparticles bearingdeposited immobilizedhad HRP. to be optimized The enzyme in electrodes order to obtain a high retained the sensitivity, catalytic covering activity for attheleast electrode surface 3 months. Wewithout forming have shown thatthick HRP films. enzymes immobilized onto silica nanoparticles can be used in the formulation of ink to print enzyme electrodes. 4. Conclusions The catalytic activity remains stable for a longer time allowing for a longer shelf life. Amperometric biosensorswere Enzyme electrodes weresuccessfully fully printed demonstrating printed the great using a bio-ink potential containing of this and SWCNT type of technology silica for thebearing nanoparticles design, immobilized developmentHRP. and manufacture The enzyme of disposable electrodes biosensors, retained the thanks to activity catalytic flexibility forinatdesign, least 3the small amount months. We haveofshown material used that HRP and the automation enzymes of the immobilized multilayer printing process. onto silica nanoparticles In used can be combination with simple in the formulation andtoaccessible of ink electronics, print enzyme this electrodes. type of digital production approach is a very attractive option for the volume The catalytic activity remains stable for a longer time allowing for a longer shelf life. production of highly stable,biosensors Amperometric inkjet printed werebiosensors. fully printed demonstrating the great potential of this type of technology for the design, development and manufacture of disposable Supplementary Materials: The following are available online at https://www.mdpi.com/article/10 .3390/nano11071645/s1, Figure S1: Design of printed sensors, Figure S2: Images of a WE printed with SWCNT-5 ink, Figure S3: Waveform and parameters used for printing, Figure S4: Plot of capacitive current vs. scan rate for Au and SWCNTs printed electrodes, Figure S5: CVs obtained for a printed electrode in solution of 10 mM hexacyanoferrate (III/II) and Randles-Sevcik plots, Figure S6: SEM images of SiO2 nanoparticles and WE printed with SWCNT-SNP-HRP inks. Author Contributions: Conceptualization, M.M., G.Y. and G.G.; methodology, M.M.; validation, A.M., G.L. and G.Y.; formal analysis, M.M. and G.Y.; investigation, M.M., L.S.V. and O.G.; resources, E.R. and G.G.; writing—original draft preparation, M.M. and L.S.V.; writing—review and editing, G.Y. and G.G.; visualization, G.Y.; supervision, G.Y., G.G. and E.R.; project administration, G.G. and E.R.; funding acquisition, R.V., G.G. and E.R. All authors have read and agreed to the published version of the manuscript. Funding: This research was funded by EMHE program “Enhancing Mobility in Health and En- vironment” and I-COOP2019 financed by the State Agency “Consejo Superior de Investigaciones Científicas” CSIC (EMHE-CSIC), references– MHE-200037 and COOPA20377. Data Availability Statement: Data is contained within this article and the Supplementary Information. Acknowledgments: This work was supported by the Spanish government funded project by MICIU RTI2018–102070-B-C21 and RTI2018-096786-B-I00 (MINECO/FEDER, EU). Authors also want to thank the support of the SU-8 Unit of the CIBER in Bioengineering, Biomaterials and Nanomedicine (CIBER-BBN) at the IMB-CNM (CSIC) of ICTS “NANBIOSIS”. This work has also made use of the Spanish ICTS Network MICRONANOFABS partially supported by MEINCOM. The authors also want to thank to Luciano Patrone and Liber Solé and Abert Guerrero, for the FIB/SEM images ob- tained at INTI-CMNB and IMB-CNM, respectively, and Marta Duch Llobera for the images obtained in the 3D optical profilometer of the clean room of the IMB-CNM. M.M. also wants to thank the support of the INTI-Micro and Nanotechnologies, especially to theDepartment of Microelectronic Prototyping and Printed Electronics and Liliana Fraigi. The graphical abstract made use of HRP repre- Nanomaterials 2021, 11, 1645 14 of 15 sentation by Manuel Almagro Rivas taken from https://commons.wikimedia.org/wiki/ (accessed on 22 June 2021) File:HRP-xray.png and by Evannovostro taken from https://www.shutterstock.com/es (accessed on 22 June 2021) ID: 780740077. Conflicts of Interest: The authors declare no conflict of interest. References 1. Singh, M.; Haverinen, H.M.; Dhagat, P.; Jabbour, G.E. Inkjet Printing—Process and Its Applications. Adv. Mater. 2010, 22, 673–685. [CrossRef] 2. Søndergaard, R.R.; Hösel, M.; Krebs, F.C. Roll-to-Roll fabrication of large area functional organic materials. J. Polym. Sci. Part B 2013, 51, 16–34. [CrossRef] 3. Khan, S.; Lorenzelli, L.; Dahiya, R. Technologies for Printing Sensors and Electronics over Large Flexible Substrate: A Review. IEEE Sens. J. 2015, 15, 3164–3185. [CrossRef] 4. Cui, Z. Printed Electronics: Materials, Technologies and Applications, 1st ed.; John Wiley & Sons: Singapore, 2016. [CrossRef] 5. Correia, V.; Mitra, K.Y.; Castro, H.; Rocha, J.G.; Sowade, E.; Baumann, R.R.; Lanceros-Mendez, S. Design and fabrication of multilayer inkjet-printed passive components for printed electronics circuit development. J. Manuf. Proc. 2018, 31, 364–371. [CrossRef] 6. Sousa, R.E.; Costa, C.M.; Laceros-Méndez, S. Advances and Future Challenges in Printed Batteries. ChemSusChem 2015, 8, 3539–3555. [CrossRef] [PubMed] 7. Cruz, F.S.M.; Rocha, L.A.; Viana, J.C. Printing Technologies on Flexible Substrates for Printed Electronics. Flex. Electron. 2018, 47–70. [CrossRef] 8. Harsanyi, G. Sensors in Biomedical Applications. Fundamentals, Technology and Applications, 1st ed.; CRC Press: Boca Raton, FL, USA, 2000; pp. 223–298. [CrossRef] 9. Komuro, N.; Takaki, S.; Suzuki, K.; Citterio, D. Inkjet printed (bio)chemical sensing devices. Anal. Bioanal. Chem. 2013, 405, 5785–5805. [CrossRef] [PubMed] 10. Lesch, A.; Jović, M.; Baudoz, M.; Zhu, Y.; Tacchini, P.; Gumy, F.; Girault, H.H. Point-of-Care Diagnostics with Inkjet-Printed Microchips. ECS Trans. 2017, 77, 73. [CrossRef] 11. Yang, W.; Li, N.; Zhao, S.; Yuan, Z.; Wang, J.; Du, X.; Wang, B.; Cao, R.; Li, X.; Xu, W.; et al. A Breathable and Screen-Printed Pressure Sensor Based on Nanofiber Membranes for Electronic Skins. Adv. Mater. Technol. 2017, 1700241. [CrossRef] 12. Abellán-Llobregat, A.; Jeerapan, I.; Bandodkar, A.; Vidal, L.; Canals, A.; Wang, J.; Morallón, E. A stretchable and screen-printed electrochemical sensor for glucose determination in human perspiration. Biosens. Bioelectron. 2017, 91, 885–891. [CrossRef] 13. Gao, M.; Li, L.; Song, Y. Inkjet printing wearable electronic devices. J. Mater. Chem. C 2017, 5, 2971–2993. [CrossRef] 14. Andò, B.; Baglio, S.; Bulsara, A.R.; Emery, T.; Marletta, V.; Pistorio, A. Low-Cost Inkjet Printing Technology for the Rapid Prototyping of Transducers. Sensors 2017, 17, 748. [CrossRef] [PubMed] 15. Cinti, S.; Colozza, N.; Cacciotti, I.; Moscone, D.; Polomoshnov, M.; Sowade, E.; Baumann, R.R.; Arduini, F. Electroanalysis moves towards paper-based printed electronics: Carbon black nanomodified inkjet-printed sensor for ascorbic acid detection as a case study. Sens. Actuators B Chem. 2018, 265, 155–160. [CrossRef] 16. Calvert, P. Inkjet Printing for Materials and Devices. Chem. Mater. 2001, 13, 3299–3305. [CrossRef] 17. Huang, Q.; Zhu, Y. Printing Conductive Nanomaterials for Flexible and Stretchable Electronics: A Review of Materials, Processes, and Applications. Adv. Mat. Technol. 2019, 1800546. [CrossRef] 18. Sundriyal, P.; Bhattacharya, S. Inkjet-Printed Sensors on Flexible Substrates. In Environmental, Chemical and Medical Sensors. Energy, Environment, and Sustainability; Bhattacharya, S., Agarwal, A., Chanda, N., Pandey, A., Sen, A., Eds.; Springer: Singapore, 2018; pp. 89–113. [CrossRef] 19. Kenry, J.C.Y.; Lim, C.T. Emerging flexible and wearable physical sensing platforms for healthcare and biomedical applications. Microsyst. Nanoengineering. 2016, 2, 16043. [CrossRef] 20. Li, J.; Rossignol, F.; Macdonald, J. Inkjet printing for biosensor fabrication: Combining chemistry and technology for advanced manufacturing. Lab. Chip 2015, 15, 2538–2558. [CrossRef] 21. Sui, Y.; Zorman, C.A. Review—Inkjet Printing of Metal Structures for Electrochemical Sensor Applications. J. Electrochem. Soc. 2020, 167, 037571. [CrossRef] 22. Distler, T.; Boccaccini, A.R. 3D printing of electrically conductive hydrogels for tissue engineering and biosensors—A review. Acta Biomater. 2020, 101, 1–13. [CrossRef] 23. Moya, A.; Gabriel, G.; Villa, R.; Del Campo, F.J. Inkjet-printed electrochemical sensors. Curr. Opin. Electrochem. 2017, 3, 29–39. [CrossRef] 24. Sfragano, P.S.; Laschi, S.; Palchetti, L. Sustainable Printed Electrochemical Platforms for Greener Analytics. Front. Chem. 2020, 8, 644. [CrossRef] [PubMed] 25. Setti, L.; Fraleoni-Morgera, A.; Mencarelli, I.; Filippini, A.; Ballarin, B.; Di Biase, M. An HRP-based amperometric biosensor fabricated by thermal inkjet printing. Sens. Actuators B Chem. 2007, 126, 252–257. [CrossRef] 26. Wang, H.; Pan, Q.; Wang, G. A Biosensor Based on Immobilization of Horseradish Peroxidase in Chitosan Matrix Cross-linked with Glyoxal for Amperometric Determination of Hydrogen Peroxide. Sensors 2005, 5, 266–276. [CrossRef] Nanomaterials 2021, 11, 1645 15 of 15 27. Rodrigues, R.C.; Ortiz, C.; Berenguer-Murcia, A.; Torres, R.; Fernandez-Lafuente, R. Modifying enzyme activity and selectivity by immobilization. Chem. Soc. Rev. 2013, 42, 6290. [CrossRef] 28. Hoarau, M.; Badieyan, S.; Marsh, E.N. Immobilized enzymes: Understanding enzyme—Surface interactions at the molecular level. Org. Biomol. Chem. 2017, 15, 9539–9551. [CrossRef] 29. Cho, I.H.; Kim, D.H.; Park, S. Electrochemical biosensors: Perspective on functional nanomaterials for on-site analysis. Biomater. Res. 2020, 24, 6. [CrossRef] 30. Zhang, X.; Li, C.R.; Wang, W.C.; Xue, J.; Huang, Y.L.; Yang, X.X.; Tan, B.; Zhou, X.P.; Shao, C.; Ding, S.J.; et al. A novel electro- chemical immunosensor for highly sensitive detection of aflatoxin B1 in corn using single-walled carbon nanotubes/chitosan. Food Chem. 2016, 192, 197–202. [CrossRef] 31. Garate, O.; Veiga, L.; Medrano, A.V.; Longinotti, G.; Ybarra, G.; Monsalve, L.N. Waterborne carbon nanotube ink for the preparation of electrodes with applications in electrocatalysis and enzymatic biosensing. Mater. Res. Bull. 2018, 106, 137–143. [CrossRef] 32. Molinari, J.; Florez, L.; Medrano, A.; Monsalve, L.; Ybarra, G. Electrochemical Determination of β-Lactoglobulin Employing a Polystyrene Bead-Modified Carbon Nanotube Ink. Biosensors 2018, 8, 109. [CrossRef] 33. Fernandes, R.S.; Raimundo, I.M.; Pimentel, M.F. Revising the synthesis of Stöber silica nanoparticles: A multivariate assessment study on the effects of reaction parameters on the particle size. Colloids Surf. A Physicochem. Eng. Aspects. 2019, 577, 1–7. [CrossRef] 34. Cortina, M.E.; Melli, L.J.; Roberti, M.; Mass, M.; Longinotti, G.; Tropea, S.; Lloret, P.; Rey Serrantes, D.A.; Salomon, F.; Lloret, M.; et al. Electrochemical Magnetic Microbeads-basedBiosensor forPoint-of-Care Serodiagnosis of Infectious Diseases. Biosens. Bioelectron. 2016, 80, 24–33. [CrossRef] [PubMed] 35. Moya, A.; Sowade, E.; Del Campo, F.J.; Mitra, K.Y.; Ramon, E.; Villa, R.; Baumann, R.R.; Gabriel, G. All-inkjet-printed dissolved oxygen sensors on flexible plastic substrates. Org. Electron. 2016, 39, 168–176. [CrossRef] 36. Jin, X.; Lu, J.; Liu, P.; Tong, H. The electrochemical formation and reduction of a thick AgCl deposition layer on a silver substrate. J. Electroanal. Chem. 2003, 542, 85–96. [CrossRef] 37. Gradzka, ˛ E.; Winkler, K. Recent Progress on Electrochemical Capacitors Basedon Carbon Nanotubes. In Carbon Nanotubes—Recent Progress, 1st ed.; Rahman, M., Asiri, A., Eds.; IntechOpen: London, UK, 2018. [CrossRef] 38. Bard, A.; Faulkner, L. Electrochemical Methods: Fundamentals and Applications, 2nd ed.; John Wiley & Sons, Inc.: New York, NY, USA, 2000; pp. 226–260, ISBN 0-471-04372-9. 39. Moore, K.J.; Flavel, B.S.; Yu, J.; Abell, A.D.; Shapter, J.G. Increased redox-active peptide loading on carbon nanotube electrodes. Electrochem. Acta 2013, 89, 206–211. [CrossRef] 40. Molinari, J.; Moina, C.; Ybarra, G. Electrochemical Immunosensor for the determination of β-Casein. J. Electrochem. Sci. Eng. 2015, 5, 9–16. [CrossRef]