Non-Fullerene Based Printed Organic Photodiodes with High

Subscriber access provided by UNIV OF DURHAM

Organic Electronic Devices

Non-Fullerene Based Printed Organic Photodiodes with High Responsivity and MHz Detection Speed Noah Strobel, Mervin Seiberlich, Tobias Rödlmeier, Uli Lemmer, and Gerardo Hernandez-Sosa ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.8b16018 • Publication Date (Web): 15 Nov 2018 Downloaded from http://pubs.acs.org on November 19, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Non-Fullerene Based Printed Organic Photodiodes with High Responsivity and MHz Detection Speed Noah Strobel1,2, Mervin Seiberlich1,2, Tobias Rödlmeier1,2, Uli Lemmer1,3, Gerardo HernandezSosa1,2* 1

Light Technology Institute, Karlsruhe Institute of Technology, Engesserstrasse 13, 76131

Karlsruhe, Germany

2

InnovationLab, Speyerer Strasse 4, 69115 Heidelberg, Germany

3

Institute of Microstructure Technology, Karlsruhe Institute of Technology, Hermann-von-

Helmholtz-Platz 1, 76344 Eggenstein-Leopoldshafen, Germany

*

[email protected]

Keywords: organic photodiode, non-fullerene acceptor, spectral responsivity, digital printing, inkjet printing, aerosol-jet printing

1 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Abstract Digitally printed organic photodiodes (OPDs) are of great interest for the cost-efficient additive manufacturing of single and multi-device detection systems with full freedom of design. Recently reported high-performance non-fullerene acceptors (NFAs) can address the crucial demands of future applications in terms of high operational speed, tunable spectral response as well as device stability. Here, we present the first demonstration of inkjet and aerosol-jet printed OPDs based on the high-performance NFA, IDTBR, in combination with poly(3-hexylthiophene) (P3HT) exhibiting a spectral response up to the NIR. These digitally printed devices reach record responsivities up to 300 mA/W in the visible and NIR spectrum competing with current commercially available technologies based on Si. Furthermore, their fast dynamic response with cut-off frequencies surpassing 2 MHz outperforms most of the state-of the-art organic photodiodes. The successful process translation from spincoating to printing is highlighted by the marginal loss in performance compared to the reference devices, which reach responsivities of 400mA/W and detection speeds of more than 4 MHz. The achieved high device performance and the industrial relevance of the developed fabrication process provide NFAs with an enormous potential for the development of printed photodetection systems.


Page 2 of 23

Page 3 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Introduction Optical detection is of crucial importance in a variety of applications covering the fields of imaging, industrial sensing and medical diagnostics.1 These technological fields will greatly benefit from the further development of organic photodiodes to enhance existing optical systems2–6 due to their inherent characteristics, including mechanical flexibility, low-cost fabrication and additive integration.2,3,7–9 The vast majority of state-of-the-art organic photodiode (OPD) systems today are based on a photoactive material blend comprising a polymer governing the absorption range of the device, combined with a fullerene acceptor to enable efficient exciton dissociation and electron transport. In the past years, many research groups, including our own, have shown that such bulkheterojunction (BHJ) OPDs can reach performance values comparable and - in parts of the visible spectrum - even exceeding the figures of merit (FOMs) of various commercially available inorganic devices.10–12 However, these state-of-the-art OPDs are expected to exhibit disadvantages, which can be attributed to intrinsic properties of the fullerene acceptor.13,14 Aside from its limited stability and low optical absorption, its poor synthetic flexibility strongly limits the spectral selectivity of OPDs, which in most of the currently investigated systems, is defined by the choice of the polymer donor. Recently, a new class of non-fullerene acceptor (NFA) materials have shown excellent electrical properties combined with high absorption coefficients, yielding record solar cell performance.15–18 Furthermore, its flexible synthetic design enables a spectral response extending towards the near infrared (NIR) part of the spectrum, offering opportunities for the development of a new generation of OPDs. Particularly, optical systems with NIR detection and imaging capabilities are of essential importance to many applications in the fields of medical diagnosis, communication and gas sensing.19 Currently, one of the existing methods to extend the responsivity to the NIR rely on the exploitation of optical effects like cavity-enhanced sub-bandgap absorption20,21, which has proven to be very successful, but introduces a higher level of complexity 3 ACS Paragon Plus Environment


that burdens the fabrication of the devices. A more processing-friendly approach is simply given by the incorporation of thick polymer:fullerene active layers as presented by Ardalan et al..22 This approach however negatively influences the responsivity in the visible regime and further increases recombination losses. Here, we utilize a NIR absorbing NFA, offering a much simpler route to combine spectral tunability with facile device fabrication. The acceptor of choice is IDTBR, which has a chemical structure consisting of an indacenodithiophene core with benzothiadiazol and rhodanine flanking groups and has recently shown impressive performance in organic photovoltaics (OPV).13,23,24 Nevertheless, the transition from a high-efficient OPV to a high-performance OPD requires the adoption of architecture, processing, and choice of interlayers in a way that maximizes the specific FOMs of optical sensors. These are represented by the specific detectivity (D*) as a measure for sensitivity, the spectral responsivity (SR) defining the wavelength selectivity in terms of light to current conversion, the 3 dB-cutoff frequency (f3dB) characterizing the detection speed and the linear dynamic range (LDR) demonstrating the window of linear response as a function of intensity. Notably, the processing of non-fullerene devices has so far mostly been limited to coating techniques with no lateral resolution.15,24,25 While this is relevant in OPV research due to their focus on large area processes, OPD fabrication should be oriented towards printing techniques as these offer high-throughput production combined with the ability to downscale device footprint for increased integration density. In particular, digital printing processes like inkjet and aerosol-jet offer great potential and industrial relevance for light detection systems, due to their high resolution, drop-on-demand deposition and roll-to-roll compatibility.7,26–28 In this work, we show digitally printed IDTBR:poly(3-hexylthiophene) (P3HT) OPDs reaching record responsivities of 300 mA/W at 750 nm as well as a fast temporal response with cut-off frequencies surpassing 2 MHz. Careful evaluation and characterization of the printing parameters 4 ACS Paragon Plus Environment

Page 4 of 23

Page 5 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


enabled us to fabricate opaque and semi-transparent devices using a combination of inkjet and aerosol-jet printing. The achieved FOMs including the spincoated reference devices, which reach responsivities of 400 mA/W and detection speed > 4 MHz are among the highest reported for OPDs in the visible and NIR wavelength region to date. Our results represent a step forward towards the implementation of NFAs in industrially relevant fabrication processes and strongly underline their great potential for optical detection systems.

Results and Discussion Benchmarking of Non-Fullerene OPDs Figure 1a presents the SR of non-fullerene P3HT:IDTBR OPDs with evaporated molybdenumtrioxide

(MoO3)

and

spin-cast

poly(3,4-ethylenedioxy-thiophene):poly(styrenesulfonate)

(PEDOT) as electron blocking layers and a schematic of its device architecture. Both devices had spin-cast zinc-oxide (ZnO) and active layers and were utilized to comprehensibly benchmark the material systems in an inverted architecture prior to the printing process. The blend shows strong a)

b)

Figure 1. (a) SR of OPDs with PEDOT and MoO3 as interlayers as well as a commercial Si-photodiode for comparison. The inset shows the device stack. (b) Relative change of SR when processed in ambient condition or in inert nitrogen atmosphere. The red line and green dotted line show the normalized absorbance spectra of P3HT and IDTBR respectively. 5 ACS Paragon Plus Environment


absorption up to 750 nm covering the entire visible spectrum and part of the NIR. This demonstrates the beneficial effect of the NFA, which considerably complements the SR of P3HT. The obtained peak SR of 400 mA/W is comparable to commercial silicon photodiodes in the entire absorption spectrum of the active blend and even superior for wavelengths below 600 nm (see Figure 1a). The data demonstrates a nearly identical performance of the devices with both interlayers, confirming the suitability of PEDOT to replace the evaporated MoO3 as a printable hole transport and electron blocking layer as well as a transparent electrode for the fully printed devices. The JV-characteristic during illumination with a solar simulator and the current response for different light intensities is depicted in Figure S1 of the supporting information and further confirms the suitability of PEDOT as a printable interlayer. Merely for positive bias, meaning the photovoltaic regime, a mismatch is observed, which stems from an increased series resistance. The response of the OPD is linear for a large range of light intensities reaching a linear dynamic range > 100 dB (see SI Figure S1). The effect of processing in ambient condition, as in the case of the printing process, is analyzed by fabricating a control device with a MoO3 interlayer in the glovebox. The normalized SR of the two respective devices is shown in Figure S1c. It can be observed that the SR is reduced in the region of the P3HT absorption range (350-650 nm) when processed in air. For wavelengths greater than 650 nm, corresponding to the IDTBR absorption, the SR-curves overlap. The averaged ratio of the SR in ambient and N2 atmosphere is displayed in Figure 1b. We observed a negative relative change in the entire region of P3HT absorption with only a marginal variation where IDTBR governs the SR (see Figure 1b). This effect can be attributed to a degradation of the P3HT under ambient conditions and the high air-stability of the non-fullerene acceptor.13,29


Page 6 of 23

Page 7 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


a)

b)

Figure 2. (a) Dark currents of OPDs with varying active layer thickness and processing condition. (b) Noise spectral density spectrum at 0 V and 2 V. The dotted and dashed-dotted lines mark the thermal and shot noise limits at 2 V respectively.

Further indication of this can be seen from the slight increase in dark current (Figure 2a) of the devices fabricated in ambient conditions, which could stem from oxygen doping of the semiconductors or trap induced dark current injection due to degradation of the polymer.30 Nonetheless, dark currents show a stronger influence of the thickness of the active layer, as can be seen when comparing the 200 nm thick devices with a device of 90 nm active layer thickness. While such thin active layers can be beneficial for organic semiconductors,31 they tend to be less robust regarding fabrication errors and shortcuts. Furthermore, they are prone to higher leakage stemming from shunt paths due to extended donor or acceptor domains through the entire device, higher sensitivity to roughness or spikes in the electrodes, consequently leading to increased dark current densities.11 This is especially critical for the sensitivity of OPDs as the dark current (idark) increases the noise floor and thus limits the minimal detectable signal. More specifically idark enhances the so called shot noise current (ishot) according to equation 1, where e is the elementary charge and B is the electrical bandwidth of the measurement system. 7 ACS Paragon Plus Environment


𝑖𝑠ℎ𝑜𝑡 = 2𝑒𝑖𝑑𝑎𝑟𝑘𝐵

( 1)

For an active layer thickness of 90 nm the dark current density was in the range of 100 µAcm-2 at a reverse bias voltage of -2 V. By increasing the thickness to 200 nm the dark current could be reduced by more than 1 order of magnitude considerably reducing the shot noise contribution. An even thicker layer of 470 nm enhances this effect by further lowering the dark current. However, the detection speed is also reduced from 4 to 1 MHz due to a simultaneous increase in the transit time of the charges as is shown in Figure S1d of the SI. Aside from shot noise there are additional sources, such as thermal and 1/f-noise, which increase the noise floor. The FOM, which relates the SR of the device to the noise limitations is the specific detectivity D*. It provides a point of comparison for devices of different geometry independent of the measurement systems bandwidth B and can be calculated by equation 2. Here, A is the device area, Snoise is the noise spectral density and inoise is the rms-noise current, which is often approximated as the product of a constant Snoise and 𝐵 (see detailed discussion in the SI).32 𝑆𝑅 𝐴

𝐷 ∗ = 𝑆𝑛𝑜𝑖𝑠𝑒 ≈

𝑆𝑅 𝐴𝐵 𝑖𝑛𝑜𝑖𝑠𝑒

( 2)

For our OPDs, we obtained Snoise by performing the Fourier-transform of the recorded dark current over time as described in 33. Figure 2b displays Snoise of devices with 200 nm active layer thickness for different reverse bias voltages. It can be observed that the noise decreases in the lower frequency range, approaching a frequency independent regime, i.e. white noise. For 0 V the noise is expected to be limited by thermal noise, while for increasing bias voltage shot noise should become dominant. To evaluate this, the theoretical shot and thermal noise spectral densities (Sshot and Sth) have been calculated from equations 3 and 4 and added to Figure 2b. 𝑆𝑠ℎ𝑜𝑡 = 2𝑒𝑖𝑑𝑎𝑟𝑘 𝑆𝑡ℎ𝑒𝑟𝑚𝑎𝑙 =

4𝑘𝐵𝑇 𝑅𝑠ℎ𝑢𝑛𝑡

( 3) ( 4)


Page 8 of 23

Page 9 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 3. The 3dB cut-off frequency of OPDs with MoO3 and PEDOT as electron blocking layers at a reverse bias voltage of -2 V. Where kB is the Boltzmann-constant, T is the temperature and Rshunt is the shunt resistance of the diode. For low frequencies < 10 Hz the measured noise is larger than the theoretical thermal and shot noise contribution, pointing to the frequency dependent sources (i.e. 1/f-noise) as the dominating limitation. For higher frequencies, Snoise approaches a frequency independent regime, i.e. white noise. At 0 V reverse bias, the noise in this regime is thermally limited as there is no current going through the OPD resulting in a shot noise of zero. At larger reverse bias voltages the noise spectral density increases and ishot dominates. This results in a frequency dependent D* which reaches 1.52 x 1011 Jones at 0 V and 1.74 x 1010 Jones at -2 V reverse bias and wavelength of 750 nm for frequencies above 10 Hz. Further details regarding the frequency dependence of D* are described in the supporting information. Dynamic characterization of the devices has been carried out, to determine the 3 dB-cutoff frequency (f3dB).34 This FOM is defined as the frequency for which the signal power from a modulated illumination source reaches 50 % of the steady-state amplitude. At this point the light signal is changing too fast for the OPD to reach equilibrium. Thus, this cutoff frequency describes 9 ACS Paragon Plus Environment


the detection speed of the OPD. While for certain applications like imager sensors a detection speed of hundreds of kHz is sufficient,25 other technological areas like communication depend on much higher speeds. The cut-off of an OPD is generally limited by either the transit time of the charge carriers through the device or the RC-constant resulting from the device geometry according to equation 5, where ftr represents the transit-limited and fRC the RC-limited cut-off frequency. 1 𝑓23𝑑𝐵

1

1

= 𝑓2 + 𝑓2 𝑅𝐶

𝑡𝑟

( 5)

The RC-limit can be estimated from R, the sum of the OPD’s series and the oscilloscopes load resistance, in addition to the geometric capacitance Cgeo using equation 6. 1

( 6)

𝑓𝑅𝐶 = 2𝜋𝑅𝐶𝑔𝑒𝑜

To optimize the dynamic response of an OPD, the device geometry therefore has to be considered carefully. Especially the thickness is crucial as it influences both limits. While a thin device will lead to a high capacitance, a thicker device would increase the transit time. For an active area of 1 mm², a resistance of 130 Ω, a device thickness of 200 nm and an assumed dielectric constant of 3, fRC results in 9 MHz. The measured bandwidth is displayed in Figure 3 for MoO3 and PEDOT as interlayers. Both devices reach a cutoff-frequency of more than 4 MHz with an applied reverse bias of -2 V. This result suggests that a layer thickness of 200 nm represents a well-balanced tradeoff between transit- and RC-limit. The achieved 3 dB-cutoff frequencies are among the highest values reported for solution processed OPDs32 and demonstrate the great potential of non-fullerene acceptor systems also from a dynamic perspective.


Page 10 of 23

Page 11 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Digital printing of Non-Fullerene based OPDs The transfer of the OPD fabrication to industrially relevant digital printing processes was carried out by bottom-up replacement of the process of all functional layers. We utilize inkjet printing for ZnO and the active layer and aerosol-jet printing for the top contact of the semitransparent device. In order to overcome unwanted drying effects, such as coffee ring formation or dewetting, and to achieve good printability, process development and ink-formulation steps (as explained in detail in the SI) are necessary. For the ZnO layer this was successfully achieved by addition of diethyeleneglycol and glycerine to a commercially available nanoparticle dispersion, resulting in homogenous layers. Similarly, the active NFA ink was modified by small fraction of tetralin, which reduced evaporation rate and minimized material flow towards the borders, due to its higher boiling point. The results of the surface evaluation of the printed layers are depicted in Figure S3. Figure 4 presents the characterization of FOMs for the printed devices and its comparison to the spin coated reference. We want to point out that the active layers of the reference devices are processed from the same solvent system as the printed devices (chlorobenzene and tetralin), and are therefore different from the device performance in the previous section (Figure 1) which were processed from chlorobenzene. The comparison of the SR between the devices processed from the pristine formulation and the tetralin containing inks is shown in a single graph in Figure S4a of the SI. We observed that the SR is slightly reduced in the range of 650-750 nm corresponding to the absorption range of IDTBR. This leads to the assumption that the micromorphology of the nonfullerene acceptor is altered by the presence of the tetralin. The increased drying time stemming from the higher boiling point of tetralin combined with the relatively low solubility of P3HT in it could result in a stronger aggregation of the non-fullerene acceptor similarly to how it has been reported for fullerene systems.35 A domain size surpassing the exciton diffusion length would then effectively reduce the charge separation efficiency. The higher roughness of the printed layer 11 ACS Paragon Plus Environment


a)

b)

c)

d)

Figure 4. (a) JV-curve under illumination and in the dark of spincoated (SC), inkjet and fully printed (inkjet + aerosol-jet) devices. (b) SR of the different opaque and semitransparent devices. (c) Noise spectral noise density of the semitransparent devices at 0 and 2V reverse bias as well as the calculated shot and thermal noise limits. (d) Photograph of a printed semi-transparent 3x3 OPD array with separated pixels. Inset: Micrograph of a single pixel. The ITO and PEDOT electrodes are marked for better visibility. evaluated by atomic force microscopy measurements (Figure S3b) shown in the supporting information further points in this direction. All dark currents are in the range of 10 µAcm-2. We attribute the main source of leakage to electron injection. This is supported by previous results25, where a 100 nm thick MoO3 electron blocking layer allowed much lower dark currents. The higher slope for the printed samples shows a less 12 ACS Paragon Plus Environment

Page 12 of 23

Page 13 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


efficient current rectification. Under illumination with 100 mWcm-2 the device comprising of an inkjet printed ZnO and active layer is almost identical to the fully spin coated device in regards to J-V characteristics. A photocurrent of 12.7 mA/cm² is observed for the reference while the inkjet printed device reached 11.4 mA/cm². Both were measured at a reverse bias of -2 V. The SR-curve in Figure 4b shows that the current loss is nearly homogeneous over the whole absorption range indicating similar micromorphology of the active layer and no negative influence from the printing process. At 750 nm and -2 V reverse bias, a peak responsivity of 0.34 and 0.28 AW-1 is reached for the spincoated and inkjet printed OPD respectively. This value is, to the best of our knowledge, the highest reported to date for printed OPDs in the NIR region and among the highest of NIR OPDs in general.19 Furthermore, semi-transparent devices are demonstrated by replacing the PEDOT/Silver bilayer electrode with an aerosol-jet printed high-conductive PEDOT electrode deposited on the inkjet printed ZnO/P3HT:IDTBR bilayer. The performance of this semitransparent device is depicted in Figure 4a and b. Slightly lower dark currents are observed for the printed electrodes, which might have resulted from a more defined active area. It has been reported how PEDOT can artificially increase the effective active area due to additional lateral conductive paths contributing to the dark current.36 Additionally, the lack of a reflective electrode induces a reduction of the magnitude as well as a change in the SR shape resulting from a shorter optical path, which especially affects the absorption of longer wavelengths (Figure 4b). This is true for electrodes printed on both the inkjet printed and spincoated active layers (see Figure S4b), confirming that the reduced SR can be fully attributed to the semi-transparent nature of the devices. Figure 4c depicts the noise spectral density of the printed semitransparent devices together with the thermal and shot noise limits, Sthermal and Sshot. Similarly to the spincoated reference Snoise is limited by thermal noise at 0 V. For increasing reverse bias, the noise rises. However, it approaches the shot noise limit much slower than the 13 ACS Paragon Plus Environment


spincoated reference sample. The mismatch to the calculated shot noise in the low frequency regime furthermore supports the necessity to calculate D* from the noise rather than the dark current only. The resulting average D* at – 2 V for frequencies greater than 1 kHz thus reaches 2.1x109 Jones. A photograph of a final semitransparent 3x3 OPD array is displayed in Figure 4d together with a micrograph of a single pixel. As all of the utilized digital techniques are drop on demand, the material is only deposited where needed, leading to spatially separated device, which prevents crosstalk and reduces materials consumption to a minimum. 2

norm. amplitude / dB

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

0 -2 -4

IJ IJ+AJP

-6

V bias = -2V

-8 -10 100

101

102

103

Frequency / kHz

Figure 5. The 3dB cut-off frequency of the opaque and semitransparent printed devices at a reverse bias of – 2 V. The dynamic characterization of the opaque and semitransparent printed NFA based OPDs is depicted in Figure 5. At -2 V, the semitransparent OPD reached cut-off frequencies of merely 250 kHz, limited by the PEDOT electrode series resistance of ~ 5 kΩ leading to an estimated RClimited cut-off frequency of 260 kHz. However, a cut-off frequency > 2 MHz is reached by the inkjet printed opaque device, which outperforms to the best of our knowledge any reported value for printed OPDs.32,37


Page 14 of 23

Page 15 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Conclusion In conclusion, we demonstrated for the first time digitally printed OPDs comprising of a high performance NFA. The fabricated OPDs based on an IDTBR:P3HT BHJ reach record SR values of up to 300mAW-1 in the NIR. The cut-off frequency was found to be in the range of 2 MHz at a reverse bias of -2 V, surpassing the detection speed of most comparable fullerene containing systems by more than an order of magnitude. The achieved high device performance and the industrial relevance of the developed fabrication process provide NFAs with an enormous potential for the development of printed photodetection systems. We expect that the digital additive manufacturing of OPDs based on NFAs will contribute to the advancement of flexible costefficient integrated optoelectronics with demanding operational speeds and high responsivity in a wide spectral range.



Experimental Methods Materials. The preparation of ZnO nanoparticle (Avantama N10) as well as the PEDOT dispersions (Hereaus VPAI-4083 and F-HC solar) was carried out in a clean room environment under ambient conditions. Active layer solutions (40g/L in chlorobenzene) and inks (see SI) consisting of 1:1 ratio of P3HT (Rieke Metals, MW=72800) and IDTBR (1-Materials) were prepared 24 h prior to deposition in a nitrogen filled glovebox. The chemical structures of the materials are shown in the SI. Device fabrication. All devices were fabricated via spincoating and printing in air unless mentioned otherwise in the text. Pestructured ITO substrates were cleaned with water and soap mechanically and subsequently in an ultrasonic bath in water, acetone and isopropanol for 10 min each. Parameters for spincoating or inkjet printing were selected to achieve layer thicknesses of 40 nm for the ZnO and 200 nm for the active layer. Reference devices had a spincast ZnO and active layer as well as spincast PEDOT or evaporated MoO3 as an EBL. The devices were finalized by a thermally evaporated silver electrode. Both the opaque and semi-transparent printed devices had a printed ZnO and active layer. They were printed with a Dimatix (DMP 2831) with 10 pL cartridges at 5 kHz jetting frequency and a printhead temperature of 50°C in a single pass process. The drop spacing was set to 15 µm and the cartridge was mounted at an angle of 23.2° for ZnO and 33.4° for the active layer. The opaque printed OPD had the same bilayer electrode as the reference device consisting of spincoated PEDOT and thermally evaporated silver. The semi-transparent device had a high conductive PEDOT electrode, which was aerosol-jet printed (Optomec AJ-300) using ultrasonic atomization and a nozzle with 200 µm diameter. The commercial F-HC solar dispersion was mixed with H2O in a ratio of 1:1. The printing parameters were set to a sheath and atomizer gas flow of 12 sccm and 16 sccm respectively at ultrasonication voltage of 45 V. A thermal annealing step was performed for all devices in the glovebox after deposition of the 16 ACS Paragon Plus Environment

Page 16 of 23

Page 17 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


PEDOT at 140°C for 10 min. Finished devices were encapsulated with a glass slide and a UVadhesive. Layer characterization. UV-Vis absorption was measured by illumination with an AvaLight-DHS-BAL light source and detection with an AvaSpec ULS3648 spectrometer. AFM images were taken with a DME Dual-Scope atomic force microscope, while optical microscope images were recorded with a Nikon Eclipse 80i. Layer thickness and profiles were measured with a Veeco Dektak 150 profilometer. Steady state measurements. Current voltage measurements were carried out using a source meter unit (SMU) (Keithley 2636A) in the dark and under illumination with a solar simulator (LOC) or a 500 mW Laser (PGL FS-VH) for linear dynamic range evaluation. The JV-curves in this manuscript are the closest to the average of all devices with a rectification ratio greater than 103 at ± 2 V in the dark. Additional measured JV-curves and information on the reproducibility of the printing process are described in the SI and displayed in Figure S5. Spectral responsivity was measured by illumination of the OPDs with monochromatic light created by a combination of a Xenon-discharge lamp (450 W Osram XBO), a monochromator (Acton, SP-2150i) and a chopper wheel. The intensity was calibrated with a reference photodiode (Thorlabs, FDS100). The generated current was amplified (Femto DHPCA-100) and recorded with a lock-in amplifier (SR830, Stanford Research Systems). Bias voltages were applied with the aforementioned SMU. Detection speed. Cut-off frequencies are measured by recording the current response of the devices to a pulse train of square light pulses, which is generated by driving a 520 nm diode laser (Oxxius LBX 520) with a wave generator (Agilent 33522A). The duty cycle was 50 % and the frequency was swept from 1 kHz to 20 MHz. After amplification of the OPD signal it was recorded with an oscilloscope (Agilent DSO 6102A).



Noise characterization. Spectral noise density measurements are carried out in an electromagnetically shielded metallic box. The signal is amplified by a variable gain low noise current amplifier (Femto DLPCA-200) before being read by the aforementioned SMU. Importantly, the amplifier is directly connected to the box without a cable to minimize in-coupling of noise from the environment. Bias voltage is applied by an isolated voltage source (Stanford Research Systems SIM 928). The measured time signal is multiplied with the Hann window function and transformed to a frequency spectrum by a fast-Fourier transformation.


Page 18 of 23

Page 19 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Acknowledgements The authors acknowledge the financial support of the German Ministry for Education and Research (BMBF) through Grant No. FKZ:13N13691. The authors further thank M. Held and N. Lam for helpful discussions.

Supporting Information The supporting Information file contains chemical structures of the materials, additional current voltage curves, a discussion about electrical noise with measured spectra, a description of the inkformulation process, printed layer characterization, visualization of processing effects on spectral responsivity and evaluation of printing reproducibility.



References (1) (2) (3) (4) (5) (6) (7) (8) (9)

(10) (11) (12)

(13)

(14)

Nabet, B. Photodetectors : Materials, Devices and Applications; Elsevier, 2016; pp 1–525. Lochner, C. M.; Khan, Y.; Pierre, A.; Arias, A. C. All-Organic Optoelectronic Sensor for Pulse Oximetry. Nat. Commun. 2014, 5, 1–7. https://doi.org/10.1038/ncomms6745. Jansen-van Vuuren, R. D.; Armin, A.; Pandey, A. K.; Burn, P. L.; Meredith, P. Organic Photodiodes: The Future of Full Color Detection and Image Sensing. Adv. Mater. 2016, 28, 4766–4802. https://doi.org/10.1002/adma.201505405. Baeg, K. J.; Binda, M.; Natali, D.; Caironi, M.; Noh, Y. Y. Organic Light Detectors: Photodiodes and Phototransistors. Adv. Mater. 2013, 25 (31), 4267–4295. https://doi.org/10.1002/adma.201204979. Pierre, A.; Gaikwad, A.; Arias, A. C. Charge-Integrating Organic Heterojunction Phototransistors for Wide-Dynamic-Range Image Sensors. Nat. Photonics 2017, 11 (3), 193–199. https://doi.org/10.1038/nphoton.2017.15. Pierre, A.; Arias, A. C. Solution-Processed Image Sensors on Flexible Substrates. Flex. Print. Electron. 2016, 1 (4), 1–26. https://doi.org/10.1088/2058-8585/1/4/043001. Eckstein, R.; Strobel, N.; Rödlmeier, T.; Glaser, K.; Lemmer, U.; Hernandez-Sosa, G. Fully Digitally Printed Image Sensor Based on Organic Photodiodes. Adv. Opt. Mater. 2018, 1701108 (step II), 1701108. https://doi.org/10.1002/adom.201701108. Binda, M.; Natali, D.; Iacchetti, A.; Sampietro, M. Integration of an Organic Photodetector onto a Plastic Optical Fiber by Means of Spray Coating Technique. Adv. Mater. 2013, 25 (31), 4335–4339. https://doi.org/10.1002/adma.201301285. Akkerman, H.; Peeters, B.; van Breemen, A.; Shanmugam, S.; Tordera, D.; van der Steen, J.-L.; Kronemeijer, A. J.; Malinowski, P.; Roose, F. De; Cheyns, D.; Genoe, J.; Dehaene, W.; Heremans, P.; Gelinck, G. 39-3: Printed Organic Photodetector Arrays and Their Use in Palmprint Scanners. SID Symp. Dig. Tech. Pap. 2018, 49 (1), 494–497. https://doi.org/10.1002/sdtp.12609. Kielar, M.; Dhez, O.; Pecastaings, G.; Curutchet, A.; Hirsch, L. Long-Term Stable Organic Photodetectors with Ultra Low Dark Currents for High Detectivity Applications. Sci. Rep. 2016, 6 (July), 1–11. https://doi.org/10.1038/srep39201. Pierre, A.; Deckman, I.; Lechêne, P. B.; Arias, A. C. High Detectivity All-Printed Organic Photodiodes. Adv. Mater. 2015, 27 (41), 6411–6417. https://doi.org/10.1002/adma.201502238. Gong, X.; Tong, M.; Xia, Y.; Cai, W.; Moon, J. S.; Cao, Y.; Yu, G.; Shieh, C.-L.; Nilsson, B.; Heeger, A. J. High-Detectivity Polymer Photodetectors with Spectral Response from 300 Nm to 1450 Nm. Science 2009, 325 (5948), 1665–1667. https://doi.org/10.1126/science.1176706. Baran, D.; Ashraf, R. S.; Hanifi, D. A.; Abdelsamie, M.; Gasparini, N.; Röhr, J. A.; Holliday, S.; Wadsworth, A.; Lockett, S.; Neophytou, M.; Emmott, C. J. M.; Nelson, J.; Brabec, C. J.; Amassian, A.; Salleo, A.; Kirchartz, T.; Durrant, J. R.; McCulloch, I. Reducing the Efficiency-Stability-Cost Gap of Organic Photovoltaics with Highly Efficient and Stable Small Molecule Acceptor Ternary Solar Cells. Nat. Mater. 2017, 16 (3), 363–369. https://doi.org/10.1038/nmat4797. Baran, D.; Gasparini, N.; Wadsworth, A.; Tan, C. H.; Wehbe, N.; Song, X.; Hamid, Z.; Zhang, W.; Neophytou, M.; Kirchartz, T.; Brabec, C. J.; Durrant, J. R.; McCulloch, I. Robust Nonfullerene Solar Cells Approaching Unity External Quantum Efficiency Enabled by Suppression of Geminate Recombination. Nat. Commun. 2018, 9 (1), 1–9. 20 ACS Paragon Plus Environment

Page 20 of 23

Page 21 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(15)

(16) (17)

(18) (19) (20)

(21)

(22) (23)

(24)

(25)

(26)

(27)

https://doi.org/10.1038/s41467-018-04502-3. Ye, L.; Xiong, Y.; Zhang, Q.; Li, S.; Wang, C.; Jiang, Z.; Hou, J.; You, W.; Ade, H. Surpassing 10% Efficiency Benchmark for Nonfullerene Organic Solar Cells by Scalable Coating in Air from Single Nonhalogenated Solvent. Adv. Mater. 2018, 30 (8), 1705485. https://doi.org/10.1002/adma.201705485. Hou, J.; Inganäs, O.; Friend, R. H.; Gao, F. Organic Solar Cells Based on Non-Fullerene Acceptors. Nat. Mater. 2018, 17, 119–128. https://doi.org/10.1038/NMAT5063. Wadsworth, A.; Moser, M.; Marks, A.; Little, M. S.; Gasparini, N.; Brabec, C. J.; Baran, D.; McCulloch, I. Critical Review of the Molecular Design Progress in Non-Fullerene Electron Acceptors towards Commercially Viable Organic Solar Cells. Chem. Soc. Rev. 2018, 1–30. https://doi.org/10.1039/C7CS00892A. Lin, Y.; Wang, J.; Zhang, Z.-G.; Bai, H.; Li, Y.; Zhu, D.; Zhan, X. An Electron Acceptor Challenging Fullerenes for Efficient Polymer Solar Cells. Adv. Mater. 2015, 27 (7), 1170– 1174. https://doi.org/10.1002/adma.201404317. Liu, X.; Lin, Y.; Liao, Y.; Wu, J.; Zheng, Y. Recent Advances in Organic Near-Infrared Photodiodes. J. Mater. Chem. C 2018, 6 (14), 3499–3513. https://doi.org/10.1039/c7tc05042a. Tang, Z.; Ma, Z.; Sánchez-Díaz, A.; Ullbrich, S.; Liu, Y.; Siegmund, B.; Mischok, A.; Leo, K.; Campoy-Quiles, M.; Li, W.; Vandewal, K. Polymer:Fullerene Bimolecular Crystals for Near-Infrared Spectroscopic Photodetectors. Adv. Mater. 2017, 29 (33), 1702184. https://doi.org/10.1002/adma.201702184. Siegmund, B.; Mischok, A.; Benduhn, J.; Zeika, O.; Ullbrich, S.; Nehm, F.; Böhm, M.; Spoltore, D.; Fröb, H.; Körner, C.; Leo, K.; Vandewal, K. Organic Narrowband NearInfrared Photodetectors Based on Intermolecular Charge-Transfer Absorption. Nat. Commun. 2017, 8, 1–6. https://doi.org/10.1038/ncomms15421. Armin, A.; Vuuren, R. D. J.; Kopidakis, N.; Burn, P. L.; Meredith, P. Efficiency Manipulation of Organic Photodiodes. Nat. Commun. 2015, 6, 1–8. https://doi.org/10.1038/ncomms7343. Holliday, S.; Ashraf, R. S.; Wadsworth, A.; Baran, D.; Yousaf, S. A.; Nielsen, C. B.; Tan, C.-H.; Dimitrov, S. D.; Shang, Z.; Gasparini, N.; Alamoudi, M.; Laquai, F.; Brabec, C. J.; Salleo, A.; Durrant, J. R.; McCulloch, I. High-Efficiency and Air-Stable P3HT-Based Polymer Solar Cells with a New Non-Fullerene Acceptor. Nat. Commun. 2016, 7, 11585. https://doi.org/10.1038/ncomms11585. Gasparini, N.; Salvador, M.; Strohm, S.; Heumueller, T.; Levchuk, I.; Wadsworth, A.; Bannock, J. H.; de Mello, J. C.; Egelhaaf, H.-J.; Baran, D.; McCulloch, I.; Brabec, C. J. Burn-in Free Nonfullerene-Based Organic Solar Cells. Adv. Energy Mater. 2017, 7 (19), 1700770. https://doi.org/10.1002/aenm.201700770. Gasparini, N.; Gregori, A.; Salvador, M.; Biele, M.; Wadsworth, A.; Tedde, S.; Baran, D.; McCulloch, I.; Brabec, C. J. Visible and Near-Infrared Imaging with Nonfullerene-Based Photodetectors. Adv. Mater. Technol. 2018, 3 (7), 1800104. https://doi.org/10.1002/admt.201800104. Eckstein, R.; Rödlmeier, T.; Glaser, T.; Valouch, S.; Mauer, R.; Lemmer, U.; HernandezSosa, G. Aerosol-Jet Printed Flexible Organic Photodiodes: Semi-Transparent, Color Neutral, and Highly Efficient. Adv. Electron. Mater. 2015, 1 (8), 1–7. https://doi.org/10.1002/aelm.201500101. Azzellino, G.; Grimoldi, A.; Binda, M.; Caironi, M.; Natali, D.; Sampietro, M. Fully Inkjet-Printed Organic Photodetectors with High Quantum Yield. Adv. Mater. 2013, 25 21 ACS Paragon Plus Environment


(28)

(29) (30) (31) (32) (33) (34) (35) (36) (37)

(47), 6829–6833. https://doi.org/10.1002/adma.201303473. Pace, G.; Grimoldi, A.; Natali, D.; Sampietro, M.; Coughlin, J. E.; Bazan, G. C.; Caironi, M. All-Organic and Fully-Printed Semitransparent Photodetectors Based on Narrow Bandgap Conjugated Molecules. Adv. Mater. 2014, 26 (39), 6773–6777. https://doi.org/10.1002/adma.201402918. Kielar, M.; Daanoune, M.; François-Martin, O.; Flament, B.; Dhez, O.; Pandey, A. K.; Chambon, S.; Clerc, R.; Hirsch, L. Insights into the Failure Mechanisms of Organic Photodetectors. 2018, 1700526. https://doi.org/10.1002/aelm.201700526. Fallahpour, A. H.; Kienitz, S.; Lugli, P. Origin of Dark Current and Detailed Description of Organic Photodiode Operation Under Different Illumination Intensities. IEEE Trans. Electron Devices 2017, 64 (6), 2649–2654. https://doi.org/10.1109/TED.2017.2696478. Tait, J. G.; Paetzold, U. W.; Cheyns, D.; Turbiez, M.; Heremans, P.; Rand, B. P. Interfacial Depletion Regions: Beyond the Space Charge Limit in Thick Bulk Heterojunctions. ACS Appl. Mater. Interfaces 2016, 8 (3), 2211–2219. https://doi.org/10.1021/acsami.5b10891. García de Arquer, F. P.; Armin, A.; Meredith, P.; Sargent, E. H. Solution-Processed Semiconductors for next-Generation Photodetectors. Nat. Rev. Mater. 2017, 2 (3), 16100. https://doi.org/10.1038/natrevmats.2016.100. Heinzel, G.; Rüdiger, A.; Schilling, R. Spectrum and Spectral Density Estimation by the Discrete Fourier Transform (DFT), Including a Comprehensive List of Window Functions and Some New Flat-Top Windows; 2002. Strobel, N.; Eckstein, R.; Lehr, J.; Lemmer, U.; Hernandez-Sosa, G. Semiconductor: Insulator Blends for Speed Enhancement in Organic Photodiodes. Adv. Electron. Mater. 2018, 1700345. https://doi.org/10.1002/aelm.201700345. Hoth, C. N.; Choulis, S. A.; Schilinsky, P.; Brabec, C. J. High Photovoltaic Performance of Inkjet Printed Polymer:Fullerene Blends. Adv. Mater. 2007, 19 (22), 3973–3978. https://doi.org/10.1002/adma.200700911. Herrbach, J.; Revaux, A.; Vuillaume, D.; Kahn, A. P-Doped Organic Semiconductor: Potential Replacement for PEDOT:PSS in Organic Photodetectors. Cit. Appl. Phys. Lett 2016, 109 (073301), 1–5. https://doi.org/10.1063/1.4961444. Pace, G.; Grimoldi, A.; Sampietro, M.; Natali, D.; Caironi, M. Printed Photodetectors. Semicond. Sci. Technol. 2015, 30 (10), 104006. https://doi.org/10.1088/02681242/30/10/104006.


Page 22 of 23

Page 23 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


TOC:


Non-Fullerene Based Printed Organic Photodiodes with High

Recommend Documents