www.acsami.org Research Article Interplay between Electrolyte-Gated Organic Field-Effect Transistors and Surfactants: A Surface Aggregation Tool and Protecting Semiconducting Layer Qiaoming Zhang,∥ Adrián Tamayo,∥ Francesca Leonardi,* and Marta Mas-Torrent* Cite This: ACS Appl. Mater. Interfaces 2021, 13, 30902−30909 Read Online ACCESS Metrics & More Article Recommendations * sı Supporting Information ABSTRACT: Molecular surfactants, which are based on a water-insoluble tail and a water-soluble head, are widely employed in many areas, such as surface coatings or for drug delivery, thanks to their capability to form micelles in solution or supramolecular structures at the solid/liquid interface. Electrolyte-gated organic field-effect transistors (EGOFETs) are highly sensitive to changes occurring at their electrolyte/gate electrode and electrolyte/organic semiconductor interfaces, and hence, they have been much explored in biosensing due to their inherent amplification properties. Here, we demonstrate that the EGOFETs and surfactants can provide mutual benefits to each other. EGOFETs can be a simple and complementary tool to study the aggregation behavior of cationic and anionic surfactants at low concentrations on a polarized metal surface. In this way, we have monitored the monolayer formation of cationic and anionic surfactants at the water/electrode interface with p-type and n-type devices, respectively. On the other hand, the operational stability of EGOFETs has been dramatically enhanced, thanks to the formation of a protective layer on top of the organic semiconductor by exposing it to a high concentration of a surfactant solution (above the critical micelle concentration). Stable performances were achieved for more than 10 and 2 h of continuous operation for p-type and n-type devices, respectively. Accordingly, this work points not only that EGOFETs can be applied to a wider range of applications beyond biosensing but also that these devices can effectively improve their long-term stability by simply treating them with a suitable surfactant. KEYWORDS: electrolyte-gated organic field-effect transistors, surfactant, surface aggregation, protective top layer, long-term stability 1. INTRODUCTION conductors (OSCs) can be printed on flexible substrates Surfactants are amphiphilic molecules that contain a hydro- over large areas. In particular, electrolyte-gated organic field- philic polar head and a long hydrophobic tail.1−5 They are effect transistors (EGOFETs) are appealing for working in widely employed in many areas such as detergents and aqueous media.12−14 Their layout consists of exposing directly emulsifiers, in nano- and micro-particle preparation, and even the OSC toward an electrolyte, where a gate contact is also in protein research and DNA extraction.1,6−8 The most immersed. The application of a source−gate voltage yields the intriguing property of surfactants is their capability to form formation of two electrical double layers (EDLs): one placed at supramolecular structures in a solution like micelles and self- the gate/electrolyte interface and the other one at the OSC/ assembled structures at the solid/liquid or solid/air interfaces electrolyte interface. These EDLs, which exhibit high like admicelles, hemimicelles, and monolayers.1 The aggrega- capacitance values on the order of tens of μF·cm−2, determine tion of surfactants strongly depends on the concentration,2 the device’s electrical performance.15 For this reason, electrolyte ionic strength,3 temperature,9 and surface proper- EGOFETs have often been exploited as electrical transducers ties.10 A range of different techniques, including vibrational for biosensing by the proper modification of such interfaces sum frequency generation (SFG) spectroscopy combined with with receptor groups.16−18 Additionally, although much less total internal reflection Raman (TIR Raman) scattering1 or with surface plasmon resonance (SPR)2 and atomic force microscopy (AFM),3,11 are commonly employed to investigate Received: March 31, 2021 the aggregation of this class of molecules on surfaces. However, Accepted: June 8, 2021 these techniques are not easily available and, further, the data Published: June 22, 2021 interpretation can be complex. Organic electronic devices are raising considerable interest for applications requiring low cost since organic semi- © 2021 The Authors. Published by American Chemical Society https://doi.org/10.1021/acsami.1c05938 30902 ACS Appl. Mater. Interfaces 2021, 13, 30902−30909 ACS Applied Materials & Interfaces www.acsami.org Research Article Figure 1. (a) Scheme of the EGOFET structure. (b) Molecular structures of the OSCs, PS, and surfactants used in this work. Schematic detection procedure of (c) approach I and (d) approach II. explored, EGOFETs can also be employed as a research tool to this device. To obtain devices that are electrically stable in study chemical/physical processes or to monitor cell/micro- water, highly crystalline and homogeneous OSC thin films are organism activities,19−21 where such events can be followed by required to avoid electrochemical doping. Previously, we registering an easily readable electrical signal. However, despite showed that the deposition of blends of OSCs with an the high potential of these devices, some unsolved problems insulating polymer (i.e., polystyrene, PS) by the solution are hindering their technological transfer such as their long- shearing technique (i.e., bar-assisted meniscus shearing, term electrical stability in aqueous media. BAMS) is an effective strategy to realize high-performance In this work, we demonstrate the reciprocal benefits that EGOFET devices.12,15 This was ascribed to the high degree of surfactants and EGOFETs can offer to each other. First, we thin-film crystallinity along with an extended homogeneity at show that EGOFETs can become an appealing alternative tool the long range (i.e., from several to hundreds of micrometers). to investigate the adsorption behavior of cationic and anionic Therefore, in this work, we have followed the same approach surfactants on a polarized metal surface (i.e., gate electrode) by p- or n-type EGOFETs, respectively. Second, we demonstrate and, as active materials, used PS-based blends of p-type OSC unprecedented long-term operation stability of EGOFETs 2,8-difluoro-5,11-bis(triethylsilylethynyl)anthradithiophene upon exposing the OSC layer to a high concentration (above (diF-TES-ADT) and n-type OSC N,N′-bis(n-octyl)-dicyano- the critical micelle concentration, CMC) of a surfactant perylene-3,4:9,10-bis-(dicarboximide) (PDI8CN2). The chem- solution due to the formation of a protective top layer on the ical structures of these materials are shown in Figure 1b (see OSC. Hence, our current work further evidences the high Section 4 for further details regarding the deposition of the potential of EGOFETs, proving that these devices can exhibit materials and device fabrication). improved electrical performance and that they can be applied Regarding surfactants, we selected cationic surfactant to a broader range of applications. cetyltrimethylammonium bromide (CTAB) and anionic surfactant sodium dodecyl sulfate (SDS) (Figure 1b, bottom). 2. RESULTS AND DISCUSSION Surfactants show different aggregation behaviors according to their concentration. The critical micelle concentration (CMC) The EGOFET architecture is shown in Figure 1a. It consists of an OSC layer bottom-contacted by two gold electrodes, refers to the minimum concentration required for micelle namely, source (S) and drain (D). The OSC is in electric formation in solution, which can then assemble in a different contact with a third Pt gate electrode (G) through an fashion once a surface is available. At high surfactant electrolyte. These devices operate at a very low operation concentrations above CMC, the morphology of surfactant voltage (<1 V) due to the high capacitance of electrical double aggregates on the solid surface vary from globular micelles to layers. The application of source−gate voltage (VGS) promotes cylindrical micelles and, finally, to layered films.11 At the other the formation of two EDLs at the electrolyte/OSC and extreme, the monolayer formation concentration (MFC), electrolyte/gate electrode interfaces,12,15 where the applied VGS which for a given surfactant is typically several orders of decreases steeply.22,23 The OSC represents the core element of magnitude below its CMC, does not allow the formation of 30903 https://doi.org/10.1021/acsami.1c05938 ACS Appl. Mater. Interfaces 2021, 13, 30902−30909 ACS Applied Materials & Interfaces www.acsami.org Research Article micelles in solution and, at the solid−liquid interface normally device response with 1 μM CTAB in Figure S1b, Supporting promotes the formation of a molecular monolayer.1,2,4,6 Information). Clearly, transfer characteristics are significantly 2.1. Approach I: EGOFETs as a Tool to Study the affected by the addition of CTAB in the media, as evidenced Monolayer Formation on a Polarized Metal Surface. In by the negative Vth shift accompanied by a decrease in the this part of the work, we aimed at exploiting EGOFETs as an source−drain current (IDS) in the saturation regime (Figure ultrasensitive tool to investigate the monolayer formation of 2a). It is also worth noting that the slopes of transfer curves ( surfactants at the water−metal interface, specifically on the IDS vs VGS curves) are not significantly altered in the whole EGOFET polarized Pt gate contact. For this purpose, to experiment (Figure S1c-d, Supporting Information), indicating monitor the cationic CTAB surfactant on a negatively charged that product [mobility × capacitance] (μ × cdl) is not much electrode, a p-type EGOFET was required. On the contrary, influenced (the average value is 0.23 μS·V−1 with 21% the anionic SDS was studied by employing an n-type variation). The same scenario is found when plotting the EGOFET. electrical characteristics in the linear regime (VDS = −0.1 V) Figure 2a displays the typical p-type transfer characteristics (Figure S2, Supporting Information). These observations can (VGS window ranging from 0.3 to −0.5 V) of an EGOFET be rationalized by considering the aggregation behavior of the CTAB cations (hereafter referred to as CTA+) on the negatively polarized Pt gate surface. This phenomenon leads to effective screening of the gate bias, which results in a negative shift of Vth, and thus, a higher VGS is required to create a conducting channel in the OSC. The same experiment was repeated by replacing CTAB with NaBr to keep the same anion while dramatically changing the cationic counterpart. In this case, a much less pronounced negative Vth shift was observed (see Figure S3, Supporting Information). This supports our previous assumption regarding the self-organ- ization of the large CTA+ cations on the gate electrode, which causes a stronger screening of the VGS. To quantify the dependence of Vth shift with respect to CTAB concentration, Vth was extracted in linear and saturation regimes (Section 4). The Vth shift (ΔVth) is defined as ΔVth = VCTAB th − VMilliQ th water , where VCTAB th is the Vth under different CTAB concentrations and VMilliQth water corresponds to the value extracted when Milli-Q water is employed as electrolyte media before the addition of the surfactant. The relationship between ΔVth and log[CTAB] is depicted in Figure 2b, where three different regions can be clearly observed: a first plateau region (I) at low surfactant concentrations (1 nM < c < 100 nM), a steep region (II) at medium concentrations (100 nM < c < 10 μM), and a second plateau (III) at high concentrations (10 μM < c < 1 mM). The CMC of CTAB in water is known to be 1 mM,24 whereas its MFC is around 1−10 μM in water.5 Taking this into consideration, region (I) can be attributed to the Figure 2. (a) I−V transfer characteristics of p-type EGOFETs (based adsorption of monomers on the gate caused by electrostatic on diF-TES-ADT:PS blend) in the saturation regime (VGS = −0.4 V) interactions between the positively charged head-group of using CTAB solutions in Milli-Q water as the electrolyte. (b) Average CTA+ and the negatively polarized Pt gate electrode, as threshold voltage shifts (ΔVth) plotted against different concen- sketched in Figure 2c. Such monomer aggregation results in trations of CTAB. These data were extracted from five devices. The negligible effects on the EGOFET response. In region (II), an EGOFET devices were exposed to CTAB solutions with concen- abrupt Vth shift due to effective screening of the gate bias is trations ranging from 1 nM to 1 mM in ascending order. (c) found, which might be indicative of the surfactant monolayer Schematic picture of CTA+ aggregation on the gate Pt surface. formation on the gate electrode. In region (III), above 10 μM, CTA+ ions fully cover the surface of the Pt gate electrode, and based on diF-TES-ADT:PS using Milli-Q water as electrolyte then, an increase in surfactant concentration is no further media (black curve). A stability check consisting of recording affecting the device’s electrical characteristics. several transfer characteristics was always performed to ensure The influence of the ionic strength was also explored by a reproducible electrical response before the subsequent phosphate-buffered solution (PBS, 1×, pH 7.4) as electrolyte surfactant detection test (Figure S1a, Supporting Information). media, giving a similar trend (Figure S4, Supporting Afterward, water solutions with different concentrations of Information). This result is also in agreement with the fact CTAB spanning from 1 nM to 1 mM were directly employed that, as expected, CTAB has a head-on orientation of the as electrolyte media in ascending order (Figure 1c, approach CTA+ ions on the negatively polarized Pt surface. This can be I). For each concentration, an additional stability check in affirmed because the Debye screening length (λD) sharply different CTAB concentrations was carried out to ensure the decreases with increasing ionic strength.6,25,26 At high ionic stability and reproducibility of the data (see, for instance, the strengths, the CTA+ charges would not be detected if they 30904 https://doi.org/10.1021/acsami.1c05938 ACS Appl. Mater. Interfaces 2021, 13, 30902−30909 ACS Applied Materials & Interfaces www.acsami.org Research Article Figure 3. (a) I−V characteristics in the saturation regime (VDS = 0.5 V) of an n-type EGOFET (based on PDI8CN2:PS blend) using SDS solutions of different concentrations in Milli-Q water as electrolyte media. (b) ΔVth vs log[SDS] extracted from three different devices. The n-type EGOFET device was exposed to SDS surfactant solutions from 1 nM to 10 mM in ascending order. were oriented far away from the Pt gate (i.e., λD is ∼100 nm in oscillating trend of the ΔVth is found, which seems to be in Milli-Q water, while it is <1 nm in PBS 1× solution27). A accordance with the work published by Song and co-workers, similar response was observed by replacing the Pt gate wherein SFG and SPR characterization techniques were electrode with a carbon composite-based electrode (Figure employed.2 They found that a DS− monolayer is formed at a S5, Supporting Information). Furthermore, cyclic voltammetry concentration of 0.2 mM by the initial adsorption of DS− (CV) and electrochemical impedance spectroscopy (EIS) molecules with their negatively charged head groups pointing experiments further support the formation of a CTA+ toward the positively polarized surface. Nevertheless, the monolayer on the Pt electrode (Figures S6 and S7, Supporting decrease of surface potential promotes that, at higher SDS Information). Finally, atomic force microscopy (AFM) images concentrations (3 mM < c < 8 mM), some SDS molecules on evaporated gold electrodes on Kapton after exposure to begin adsorbing in the opposite orientation (head groups CTAB at a concentration of 10 μM (region II) and applying a pointing toward the water phase). Such a process would voltage sweep from 0.3 to −0.5 V also confirmed the formation explain the reduction of VGS screening observed in our devices of a monolayer (Figure S8, Supporting Information). However, at SDS concentrations above 1 mM, as sketched in the inset of the presence of CTAB on the OSC surface was not observed Figure 3b. The CMC of SDS is reported to be around 7−8 under these conditions (Figure S9, Supporting Information). mM in deionized water;30 hence, above this concentration, Noticeably, the influence of CTAB aggregation on the OSC micelles are formed. surface was also explored by incubating the film with the All of these results have further been confirmed by EIS surfactant previously to perform the electrical test, but not measurements, in which the capacitance response also exhibits significant electrical effects were observed in the monolayer an oscillating trend in the same SDS concentration range formation concentration range (Figure S10, Supporting (Figure S11, Supporting Information). Noticeably, this Information). behavior disappeared when using PBS 1× as a medium It should be mentioned that previously an organic (Figure S12, Supporting Information), probably caused by the electrochemical transistor (OECT), whose working principle decrease of λD at high ionic strengths,26,31 which prevents the relies on doping/dedoping a conducting polymer, was reported detection of the head-out orientation of DS− anions. AFM as a suitable platform for the detection of CTAB micelles.28 analysis of a planar gold gate electrode after being exposed to This was driven by the modification of the size and type of SDS at a concentration of 1 mM and applying a transfer ionic dopants occurring during CTAB micelle assembly. voltage sweep also confirmed the presence of a monolayer on However, with the here-in fabricated EGOFETs, we are able the electrode (Figure S13, Supporting Information). to study the CTAB monolayer formation, which takes place at Thus, EGOFETs represent a promising platform to study much lower concentrations. the surface assembly of ionic surfactants at low concentrations To verify the versatility of this approach, the surface on a polarized electrode using a simple electrical read-out. assembly of anionic surfactant SDS was investigated using an 2.2. Approach II: Surfactant Protecting Layer to n-type EGOFET based on a PDI8CN2:PS blend film.29 In this Enhance the Long-Term Stability of EGOFETs. Long- device, the Pt gate electrode is positively polarized during its term operational instability is a key factor for sensor operation. Figure 3a displays the transfer characteristics in the implementation in biological and biomedical applications.15 saturation regime of the device exposed to different SDS Stability is still a major drawback in EGOFETs, and it is mainly concentrations. Noticeably and as the previous example, the ascribed to the migration of ions into the OSC layer.15 electrical characteristics are significantly affected by the Therefore, finding a suitable route to protect the OSC film, presence of the SDS surfactant due to the screening of the without hampering the electrical performance, could be an gate voltage by the SDS anion (hereafter referred to as DS−) effective way to enhance the device’s stability. assembly on the gate electrode, resulting in a Vth shift toward It is known that when the concentrations of surfactant higher-voltage values together with a reduction of IDS (Figure solutions are at least 2 times higher than their CMC, micelles 3a). aggregate on surfaces, finally forming layered films.11 Taking Analyzing in detail the ΔVth shift as a function of the SDS this into account, we proceed in using this approach for concentration (Figure 3b), we observe that the results agree encapsulating the OSC films. Approach II consisted hence of with DS− monolayer formation in the concentration range placing a drop of the surfactant solution with a concentration from 1 μM to 1 mM. However, at higher concentrations, an above CMC on top of the OSC film for 20 min, followed by 30905 https://doi.org/10.1021/acsami.1c05938 ACS Appl. Mater. Interfaces 2021, 13, 30902−30909 ACS Applied Materials & Interfaces www.acsami.org Research Article Figure 4. (a) I−t plot of the normalized IDS of a p-type EGOFET (based on diF-TES-ADT:PS blend) recorded at VGS = −400 mV and VDS = −50 mV in the absence of SDS and after treating the organic semiconductor film with SDS. All data were recorded using Milli-Q water as the electrolyte. The inset is the schematic interpretation of the SDS aggregation on the diF-TES-ADT:PS blend surface. (b) I−V transfer characteristics in the linear regime using Milli-Q water as media recorded initially (black line), after treating the semiconductor with SDS (blue line), and after the current monitoring test of the SDS treated device (red line). Figure 5. (a) I−t plot of EGOFETs based on PDI8CN2:PS films. The normalized IDS is recorded at fixed operation voltage (VGS = 400 mV and VDS = 300 mV) using Milli-Q water as the electrolyte and after treating the OSC film with a 10 mM CTAB solution. The inset is the schematic view of CTAB aggregation on the OSC surface. (b) I−V transfer characteristics recorded in the saturation regime using Milli-Q water as electrolyte media of the device as prepared (black line), after CTAB treatment (blue line), and after the current monitoring test (red line). the removal of the solution and then drying with a N2 flow. monitoring. As shown in Figure 4b, in the transfer character- Then, the EGOFET was electrically tested in Milli-Q water istics registered after the SDS treatment (blue line), only a (Figure 1d). small positive Vth shift accompanied with IDS increase is found The first test was performed using the p-type EGOFET with respect to the initial device (black line). These results are based on diF-TES-ADT:PS and the anionic SDS surfactant. To attributed to the DS− anions that aggregate on the OSC optimize the SDS concentration, a series of different surface, inducing more holes in the conduction channel. After concentrations of SDS solutions in Milli-Q water were the current monitoring test (red line), the device shows a employed. Current monitoring tests under fixed VGS = −400 stable electrical performance with a slight IDS decrease and a mV and VDS = −50 mV were performed to test their protective significantly reduced hysteresis. It should be noticed that only efficacy. As displayed in Figure S14 (Supporting Information), when the device is operated at a high source−drain bias (VDS = the optimized SDS concentration was 50 mM. To highlight the −400 mV), the enhanced device stability is no more realized, long-term stability induced by the SDS layer, the normalized which could be attributed to the high longitudinal field present and original current monitoring data obtained from devices along the OSC channel (Figure S16, Supporting Information). with and without the SDS treatment are presented in Figures AFM was employed to study the morphology of the OSC 4a and S15 (Supporting Information), respectively. In the thin-film surface prior to and after exposure to the SDS absence of SDS (black curve), the IDS exhibits a fast decrease in surfactant solution. As shown in Figure S17 (Supporting the first half-hour followed by a current increase in the Information), the AFM images reveal morphological features at following 30 min and then a slow but continuous decrease for the nanoscale level on top of the diF-TES-ADT:PS thin film, the remaining test time (more than 30% of the current is which can be attributed to the SDS assemblies. dropped after 10 h). In contrast, the EGOFETs treated with Similar results were obtained using films of only diF-TES- SDS show an initial 10% current decrease but, afterward, show ADT (i.e., without the binding PS polymer) or also using a stable current value for more than 10 h (red curve). This another anionic surfactant such as sodium octyl sulfate (SOS) enhanced and unprecedented operation stability is mainly (Figures S18 and S19, Supporting Information). Finally, a ascribed to the protective action provided by the SDS layer comparison experiment was carried out by substituting SDS for formed on top of the OSC film, as sketched in the inset of the cationic surfactant CTAB following the same experimental Figure 4a. Our observations are further confirmed by the protocol (Figure S17, Supporting Information). As displayed transfer characteristics recorded prior to and after I/t in Figure S20 (Supporting Information), no device stability 30906 https://doi.org/10.1021/acsami.1c05938 ACS Appl. Mater. Interfaces 2021, 13, 30902−30909 ACS Applied Materials & Interfaces www.acsami.org Research Article improvement was noticed in this case. This indicates that with 4. EXPERIMENTAL SECTION the cationic surfactant the protective layer on the OSC is not 4.1. Materials. The organic semiconductor materials used in this stably formed under the operational device conditions of a p- work, 2,8-difluoro-5,11-bias(triethylsilylethynyl)anthradithiophene type EGOFET, which can be ascribed to the polarization of the (diF-TES-ADT) and N,N′-bis(n-octyl)-dicyanoperylene-3,4:9,10-bis- device (the OSC is positively charged under operation). (dicarboximide) (PDI8CN2), were purchased from Lumtec and To further verify this and validate our methodology, Polyera Inc. (N1200), respectively. Polystyrene (PS, Mw = 10 000 and PDI8CN2:PS films were exposed to CTAB (10 mM solution) 280 000 g·mol−1), chlorobenzene (CB), and 2,3,4,5,6-pentafluor- and implemented as n-type EGOFET. As in the previous case, othiophenol (PFBT) were provided by Sigma-Aldrich. The AFM characterization shows the formation of a passivating cetyltrimethylammonium bromide (CTAB) and sodium dodecyl layer on top of the organic semiconductor film after exposing it sulfate (SDS) were also purchased from Sigma-Aldrich. All of the commercial materials were used as received without any further to CTAB (Figure S21, Supporting Information). It is purification. Platinum (Pt) wire (Φ =0.5 mm) and silver (Ag) wire important to highlight that n-type EGOFET devices suffer (Φ =0.5 mm) were obtained from Sigma-Aldrich. Multiwalled carbon much fast deterioration due to moisture and water.32 As nanotubes (>95% of carbon purity, 10−30 nm of outer diameter, and displayed in Figure 5a, the same stability enhancement about 5−15 μm of length) were purchased from SES Research observed previously with the p-type EGOFET was encoun- (Houston, TX). Pt wire was cleaned in hot sulfuric acid (0.1 M) for tered again here. Indeed, a sharp IDS increase in the first half- 30 min and rinsed with Milli-Q water prior to each electrical and hour followed by a constant IDS in the following 2 h was found. electrochemical measurement. Milli-Q water (18.2 MΩ·cm resistivity In stark contrast, the n-type EGOFET with no protection at 25 °C) was obtained from a Milli-Q Integral Water Purification shows fast and important deterioration with about 90% System. decrease of the initial current after 3 h (Figure S22, Supporting 4.2. EGOFET Device Fabrication. The diF-TES-ADT:PS10K (4:1) and PDI8CN2:PS280K (1:2) blend solutions used in this work Information). This behavior is further confirmed by comparing were prepared according to our previously published protocol.15,29 the transfer characteristics prior to and after the current The S/D electrodes (Cr/Au, 5/40 nm), with a geometrical ratio W/L monitoring test (Figures 5b and S22, Supporting Information). = 690 (channel width W = 20 700 μm and channel length L = 30 After protecting the organic semiconductor with CTAB, an μm), were deposited on Kapton polyimide film (75 μm thick, important shift of the transfer characteristics takes place. DuPont) by a thermal evaporation system (Evaporation System Auto Remarkably, after 3 h of continuous operation during the 306 from Boc Edwards) at a pressure of 2 × 10−4 Pa. Prior to the monitoring test, the transfer characteristics registered prove deposition of the OSC:PS blend solution, the S/D substrates were that the device is still working properly. cleaned in an ultrasonic bath with acetone and isopropanol for 15 min each, followed by an ozone treatment for 25 min. After that, the 3. CONCLUSIONS cleaned S/D electrodes were functionalized by PFBT through immersion of the substrate into PFBT/isopropanol (2 μL:1 mL) In conclusion, we have fabricated p-type and n-type EGOFETs solution for 15 min and then rinsed with isopropanol. Finally, all of based on printed OSC:PS blends and studied the effect of the the OSC:PS blend films were coated by the bar-assisted meniscus surface aggregation of surfactants on their interfaces, which is shearing (BAMS) technique at a fixed speed of 1 cm·s−1 and a plate highly influenced by the device polarization. First, the temperature of 105 °C in ambient conditions as previously reported.15 EGOFETs have been demonstrated to be a simple, 4.3. Surfactant Solution Preparation. Surfactant solutions with complimentary, and highly sensitive tool to investigate the different concentrations were prepared in Milli-Q water and adsorption behavior of cationic (i.e., CTAB) and anionic (i.e., phosphate-buffered saline (PBS 1×, pH 7.4 with 137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 2 mM KH2PO4). All of the original SDS) surfactants on the gate surface of EGOFETs. Employing surfactant solutions were prepared at a concentration of 100 mM by a p-type EGOFET, the formation of a monolayer of the CTA+ dissolving the corresponding surfactant salts and then stirring at 40 °C cations on the negatively polarized Pt gate electrode could be for at least 1 h to completely dissolve the powder. Afterward, the monitored. This was evidenced by a sharp Vth variation in the solutions were diluted as required with Milli-Q water or PBS solution. concentration range (1 μM < c < 10 μM). Additionally, with an All of the surfactant solutions were freshly prepared before each n-type EGOFET, the surface assembly of SDS was explored. experiment. The results are in agreement with a head-on monolayer 4.4. Electrical Measurements. Electrical measurements, includ- formation up to a concentration of 1 mM. Afterward, the ing transfer, output characteristics, and long-term stability tests, were oscillating trend of Vth found indicates that some SDS recorded in ambient conditions using an Agilent B1500A semi- molecules upturn adsorbing on the surface in the opposite conductor device analyzer connected to the samples with a Karl SÜ SS probe station. For the surfactant aggregation behavior study orientation. (approach I), a poly(dimethylsiloxane) (PDMS) pool was placed In the second part of the work, surfactants were exploited to on top of the interdigitated area of the EGOFETs to confine 30 μL of spontaneously form a protective layer for the OSC thin films to target aqueous solution as electrolyte media and a gate electrode (Pt prolong EGOFETs’ long-term stability. In particular, forming a wire, carbon paste electrode), as depicted in Figure 1c. The layer of the anionic SDS surfactant on top of a p-type OSC surfactants, dissolved in Milli-Q water or PBS 1× solution, were unprecedentedly enhanced the device’s stability for more than employed directly as electrolyte media during the EGOFET tests in a 10 h in continuous operation. Similarly, the cationic CTAB concentration spanning from nanomolar to millimolar. Carbon paste surfactant improved the water stability of an n-type OSC for electrode was fabricated according to a previously reported more than 2 h. protocol.33,34 The scan rate of transfer characteristics was set to Our results point not only that EGOFETs can be applied to ∼70 mV·s−1 unless otherwise specified. To investigate the in situ long-term EGOFET monitoring after the a wider range of applications beyond biosensing but also that formation of a surfactant protecting layer on top of the OSC these devices can exhibit long-term stability when operating in (approach II), a drop of surfactant solution (∼50 μL) with target a water environment by simply treating them with a suitable concentrations (1, 5, 10, 50, and 100 mM) was placed on top of the surfactant. Hence, low-cost EGOFET devices might become a active area and kept for 20 min under ambient conditions to allow the simple and versatile tool for analysis in laboratories covering self-assembly of a compact surfactant layer on top of the different fields. semiconductor (Figure 1d, left). Then, the solution was removed 30907 https://doi.org/10.1021/acsami.1c05938 ACS Appl. Mater. Interfaces 2021, 13, 30902−30909 ACS Applied Materials & Interfaces www.acsami.org Research Article by gentle rinsing and dried with N2. A large PDSM pool (∼500 μL) Additional electrical transfer characteristics of EGO- was fixed on top of the interdigitated area followed by the injection of FETs, CV and EIS characterizations, AFM images, Milli-Q water into the pool as electrolyte media (Figure 1d, right). determination of the optimized surfactant concentration During the long-term stability test, a parafilm was sealed on top of the for the formation of a protective layer on top of OSC in pool to minimize solvent evaporation during the test. IDS was continuously recorded with an interval time of 1 s at VGS = −400 mV p-type EGOFETs, and additional I−t stability plots and VDS = −50 or −100 mV for p-type devices, whereas for n-type (PDF) devices, we applied VGS = 500 mV and VDS = 100 mV. To quantitatively describe the properties of EGOFETs, the classical MOSFET model has been employed to extract the figure of merits, including the threshold voltage (Vth), total electrical double layers ■ AUTHOR INFORMATION Corresponding Authors capacitance (Cdl), and charge carrier mobility (μ). According to the Francesca Leonardi − Institut de Ciència de Materials de MOSFET model, the IDS values in linear and saturation regimes are Barcelona, ICMAB-CSIC, 08193 Barcelona, Spain; Present approximately determined by the following equations Address: OnePlanet Research Center, Wageningen Campus, Plus Ultra II, Bronland 10, 6708 WH W Wageningen, The Netherlands.; IDS = cdlμ(VGS − Vth)VDS(linear) L Email: Francesca.Leonardi@imec.nl W Marta Mas-Torrent − Institut de Ciència de Materials de IDS = cdlμ(VGS − Vth)2 (saturation) Barcelona, ICMAB-CSIC, 08193 Barcelona, Spain; 2L orcid.org/0000-0002-1586-005X; Email: mmas@ where W is the channel width and L is the channel length. icmab.es 4.5. Morphological Characterization. Atomic force microscopy (AFM) images were obtained working with a 5100 SPM system from Authors Agilent Technologies in tapping mode. For approach I, gold gate Qiaoming Zhang − School of Physical Science and electrodes were evaporated on Kapton and analyzed by AFM after Technology, Southwest University, 400715 Chongqing, P. R. being exposed to ML and CMC concentration of CTAB (10 μM and China; Institut de Ciència de Materials de Barcelona, 10 mM, respectively) and SDS (1 and 50 mM, respectively) while a ICMAB-CSIC, 08193 Barcelona, Spain voltage sweep was applied (from 0.3 to −0.5 V for CTAB and from −0.3 to 0.5 V for SDS) to them using a Novocontrol Alpha-AN Adrián Tamayo − Institut de Ciència de Materials de impedance analyzer equipped with POT/GAL 30V/2A. To check the Barcelona, ICMAB-CSIC, 08193 Barcelona, Spain formation of the surfactant protective layer on the OSC thin film, the Complete contact information is available at: freshly prepared OSC thin films were first investigated. Then, the https://pubs.acs.org/10.1021/acsami.1c05938 samples were dipped into a surfactant solution at a concentration above its CMC for 20 min and then slightly rinsed and dried with N2 Author Contributions gas. All images were recorded in ambient conditions and analyzed ∥ using Gwyddion 2.47 software. Q.Z. and A.T. equally contributed to this work. 4.6. Electrochemical Impedance Spectroscopy (EIS). Cyclic Notes voltammetry (CV) and electrochemical impedance spectroscopy The authors declare no competing financial interest. (EIS), including capacitance and phase angle spectra, were carried out using a Novocontrol Alpha-AN impedance analyzer equipped with a POT/GAL 30V/2A electrochemical interface in a frequency range between 105 and 10−1 Hz. CV measurements were carried out ■ ACKNOWLEDGMENTS This work was funded by DGI (Spain) through project in a conventional three-electrode electrochemical cell using a Pt wire GENESIS PID2019-111682RB-I00, the Generalitat de Cata- as the working electrode (Φ = 0.5 mm), a second Pt wire as the lunya (2017-SGR-918), and the Spanish Ministry of Economy counter electrode (Φ = 0.5 mm), and an Ag/AgCl quasi-reference and Competitiveness through the “Severo Ochoa” Programme electrode (Φ = 0.5 mm). Electrochemical responses were recorded in for Centers of Excellence in R&D (FUNFUTURE CEX2019- 0.1 M KCl aqueous solution containing 0.1 mM [Fe(CN)6]3−/4− as 000917-S). A.T. acknowledges the FPU fellowship and is the redox marker. During the CV test, a 100 mM CTAB solution in enrolled in the UAB Materials Science PhD program. F.L. Milli-Q water was added to the electrochemical cell to achieve the gratefully acknowledges the “Juan de la Cierva” programme. ■ desired concentration of CTAB in the working solution. The scan rate of the CV measurements was set to 10 mV·s−1. For the impedance REFERENCES measurements, a square Au electrode (S = 1 mm2) was employed as the counter electrode, a Pt wire (Φ = 0.5 mm) served as the working (1) Tyrode, E.; Rutland, M. W.; Bain, C. D. Adsorption of CTAB on electrode, and Ag/AgCl wire (Φ = 0.5 mm) was used as the quasi- Hydrophilic Silica Studied by Linear and Nonlinear Optical reference electrode. Surfactant solutions of different concentrations Spectroscopy. J. Am. Chem. Soc. 2008, 130, 17434−17445. (in ascending order) were used as media confined with a PDMS pool (2) Song, S. H.; Koelsch, P.; Weidner, T.; Wagner, M. S.; Castner, D. on the counter electrode. Prior to each EIS test, the CV measurement G. Sodium Dodecyl Sulfate Adsorption onto Positively Charged was implemented to guarantee the migration and adsorption of CTA+ Surfaces: Monolayer Formation with Opposing Headgroup Orienta- or DS− on the Pt surface, which was obtained by defining the DC tions. Langmuir 2013, 29, 12710−12719. voltage in the same window (0.3 to −0.5 V, a step of −0.1 V in the (3) Sakai, H.; Nakamura, H.; Kozawa, K.; Abe, M. Atomic Force CTAB case, while 0 to 0.5 V, a step of 0.1 V in the SDS case) used for Microscopy Observation of the Nanostructure of Tetradecyltrime- the EGOFET transfer characteristics. Then, the surfactant solutions thylammonium Bromide Films Adsorbed at the Mica/Solution were replaced by Milli-Q water. During EIS tests, DC and AC signals Interface. Langmuir 2001, 17, 1817−1820. were fixed to 0 and 20 mV, respectively. (4) Jiménez-Á ngeles, F.; Khoshnood, A.; Firoozabadi, A. Molecular ■ Dynamics Simulation of the Adsorption and Aggregation of Ionic ASSOCIATED CONTENT Surfactants at Liquid-Solid Interfaces. J. Phys. Chem. C 2017, 121, 25908−25920. * sı Supporting Information (5) Hu, C.; Dang, X.; Hu, S. Studies on Adsorption of The Supporting Information is available free of charge at Cetyltrimethylammonium Bromide at Carbon Paste Electrode and https://pubs.acs.org/doi/10.1021/acsami.1c05938. the Enhancement Effect in Thyroxine Reduction by Voltammetry and 30908 https://doi.org/10.1021/acsami.1c05938 ACS Appl. Mater. Interfaces 2021, 13, 30902−30909 ACS Applied Materials & Interfaces www.acsami.org Research Article Electrochemical Impedance Spectroscopy. J. Electroanal. Chem. 2004, Transistors-Opportunities for Biochemical Sensing and Extracellular 572, 161−171. Signal Transduction. J. Mater. Chem. B 2013, 1, 3728−3741. (6) Atkin, R.; Craig, V. S. J.; Wanless, E. J.; Biggs, S. Mechanism of (23) Tibaldi, A.; Fillaud, L.; Anquetin, G.; Woytasik, M.; Zrig, S.; Cationic Surfactant Adsorption at the Solid-Aqueous Interface. Adv. Piro, B.; Mattana, G.; Noël, V. Electrolyte-Gated Organic Field-Effect Colloid Interface Sci. 2003, 103, 219−304. Transistors (EGOFETs) as Complementary Tools to Electro- (7) Tsagkaropoulou, G.; Allen, F. J.; Clarke, S. M.; Camp, P. J. Self- chemistry for the Study of Surface Processes. Electrochem. Commun. Assembly and Adsorption of Cetyltrimethylammonium Bromide and 2019, 98, 43−46. Didodecyldimethylammonium Bromide Surfactants at the Mica- (24) Häckl, K.; Li, H.; Aldous, I. M.; Tsui, T.; Kunz, W.; Abbott, A. Water Interface. Soft Matter 2019, 15, 8402−8411. P.; Warr, G. G.; Atkin, R. Potential Dependence of Surfactant (8) Hideshima, S.; Saito, M.; Fujita, K.; Harada, Y.; Tsuna, M.; Adsorption at the Graphite Electrode/Deep Eutectic Solvent Sekiguchi, S.; Kuroiwa, S.; Nakanishi, T.; Osaka, T. Label-Free Interface. J. Phys. Chem. Lett. 2019, 10, 5331−5337. Detection of Allergens in Food via Surfactant-Induced Signal (25) Gebbie, M. A.; Dobbs, H. A.; Valtiner, M.; Israelachvili, J. N. Amplification Using a Field Effect Transistor-Based Biosensor. Sens. Long-Range Electrostatic Screening in Ionic Liquids. Proc. Natl. Acad. Actuators, B 2018, 254, 1011−1016. Sci. U.S.A. 2015, 112, 7432−7437. (9) Kawai, T.; Kamio, H.; Kondo, T.; Kon-No, K. Effects of (26) Adibnia, V.; Shrestha, B. R.; Mirbagheri, M.; Murschel, F.; De Concentration and Temperature on SDS Monolayers at the Air - Crescenzo, G.; Banquy, X. Electrostatic Screening Length in “Soft” Solution Interface Studied by Infrared External Reflection Spectros- Electrolyte Solutions. ACS Macro Lett. 2019, 8, 1017−1021. copy. J. Phys. Chem. B 2005, 109, 4497−4500. (27) Kergoat, L.; Piro, B.; Berggren, M.; Pham, M. C.; Yassar, A.; (10) Islam, M. N.; Sharker, K. K.; Sarker, K. C. Salt-Induced Horowitz, G. DNA Detection with a Water-Gated Organic Field- Modulation of the Krafft Temperature and Critical Micelle Effect Transistor. Org. Electron. 2012, 13, 1−6. (28) Tarabella, G.; Nanda, G.; Villani, M.; Coppedè, N.; Mosca, R.; Concentration of Benzyldimethylhexadecylammonium Chloride. J. Malliaras, G. G.; Santato, C.; Iannotta, S.; Cicoira, F. Organic Surfactants Deterg. 2015, 18, 651−659. Electrochemical Transistors Monitoring Micelle Formation. Chem. (11) Zhao, F.; Du, Y.; Yang, P.; Li, X.; Tang, J. Adsorption Behavior Sci. 2012, 3, 3432−3435. of Hexadecyltrimethylammonium Bromide (CTAB) to Mica (29) Campos, A.; Riera-Galindo, S.; Puigdollers, J.; Mas-Torrent, M. Substrates as Observed by Atomic Force Microscopy. Sci. China, Reduction of Charge Traps and Stability Enhancement in Solution- Ser. B: Chem. 2005, 48, 101−106. Processed Organic Field-Effect Transistors Based on a Blended n- (12) Leonardi, F.; Casalini, S.; Zhang, Q.; Galindo, S.; Gutiérrez, D.; Type Semiconductor. ACS Appl. Mater. Interfaces 2018, 10, 15952− Mas-Torrent, M. Electrolyte-Gated Organic Field-Effect Transistor 15961. Based on a Solution Sheared Organic Semiconductor Blend. Adv. (30) Hu, K.; Bard, A. J. Characterization of Adsorption of Sodium Mater. 2016, 28, 10311−10316. Dodecyl Sulfate on Charge-Regulated Substrates by Atomic Force (13) Liao, C.; Yan, F. Organic Semiconductors in Organic Thin-Film Microscopy Force Measurements. Langmuir 1997, 13, 5418−5425. Transistor-Based Chemical and Biological Sensors. Polym. Rev. 2013, (31) Brown, M. A.; Goel, A.; Abbas, Z. Effect of Electrolyte 53, 352−406. Concentration on the Stern Layer Thickness at a Charged Interface. (14) Torsi, L.; Magliulo, M.; Manoli, K.; Palazzo, G. Organic Field- Angew. Chem., Int. Ed. 2016, 55, 3790−3794. Effect Transistor Sensors: A Tutorial Review. Chem. Soc. Rev. 2013, (32) Porrazzo, R.; Luzio, A.; Bellani, S.; Bonacchini, G. E.; Noh, Y. 42, 8612−8628. Y.; Kim, Y. H.; Lanzani, G.; Antognazza, M. R.; Caironi, M. Water- (15) Zhang, Q.; Leonardi, F.; Casalini, S.; Temiño, I.; Mas-Torrent, Gated n-Type Organic Field-Effect Transistors for Complementary M. High Performing Solution-Coated Electrolyte-Gated Organic Integrated Circuits Operating in an Aqueous Environment. ACS Field-Effect Transistors for Aqueous Media Operation. Sci. Rep. Omega 2017, 2, 1−10. 2016, 6, No. 39623. (33) Muñoz, J.; Leonardi, F.; Ö zmen, T.; Riba-Moliner, M.; (16) Manoli, K.; Magliulo, M.; Mulla, M. Y.; Singh, M.; Sabbatini, González-Campo, A.; Baeza, M.; Mas-Torrent, M. Carbon-Paste L.; Palazzo, G.; Torsi, L. Printable Bioelectronics to Investigate Nanocomposites as Unconventional Gate Electrodes for Electrolyte- Functional Biological Interfaces. Angew. Chem., Int. Ed. 2015, 54, Gated Organic Field-Effect Transistors: Electrical Modulation and 12562−12576. Bio-Sensing. J. Mater. Chem. C 2019, 7, 14993−14998. (17) Casalini, S.; Dumitru, A. C.; Leonardi, F.; Bortolotti, C. A.; (34) Tamayo, A.; Muñoz, J.; Mas-Torrent, M. Electronic Perform- Herruzo, E. T.; Campana, A.; De Oliveira, R. F.; Cramer, T.; Garcia, ance of Polymer Carbon-Paste Nanoallotropes from 0D to 3D as R.; Biscarini, F. Multiscale Sensing of Antibody-Antigen Interactions Novel Gate Electrodes in Water-Gated Organic Field-Effect by Organic Transistors and Single-Molecule Force Spectroscopy. ACS Transistors. Adv. Electron. Mater. 2020, 6, No. 2000431. Nano 2015, 9, 5051−5062. (18) Ricci, S.; Casalini, S.; Parkula, V.; Selvaraj, M.; Deniz, G.; Greco, P.; Biscarini, F.; Mas-Torrent, M. Label-Free Immunodetec- tion of α -Synuclein by Using a Microfluidics Coplanar Electrolyte- Gated Organic Field-Effect Transistor. Biosens. Bioelectron. 2020, 167, No. 112433. (19) Le Gall, J.; Mouillard, F.; Le, T. N.; Vu, T. T.; Mattana, G.; Brayner, R.; Zrig, S.; Noël, V.; Piro, B. Monitoring Photosynthetic Microorganism Activity with an Electrolyte-Gated Organic Field Effect Transistor. Biosens. Bioelectron. 2020, 157, No. 112166. (20) Fillaud, L.; Petenzi, T.; Pallu, J.; Piro, B.; Mattana, G.; Noel, V. Switchable Hydrogel-Gated Organic Field-Effect Transistors. Lang- muir 2018, 34, 3686−3693. (21) Kyndiah, A.; Leonardi, F.; Tarantino, C.; Cramer, T.; Millan- Solsona, R.; Garreta, E.; Montserrat, N.; Mas-Torrent, M.; Gomila, G. Bioelectronic Recordings of Cardiomyocytes with Accumulation Mode Electrolyte Gated Organic Field Effect Transistors. Biosens. Bioelectron. 2020, 150, No. 111844. (22) Cramer, T.; Campana, A.; Leonardi, F.; Casalini, S.; Kyndiah, A.; Murgia, M.; Biscarini, F. Water-Gated Organic Field Effect 30909 https://doi.org/10.1021/acsami.1c05938 ACS Appl. Mater. Interfaces 2021, 13, 30902−30909