Patterned Liquid Metal Contacts for Printed Carbon Nanotube

Subscriber access provided by UCL Library Services

Article

Patterned Liquid Metal Contacts for Printed Carbon Nanotube Transistors Joseph B. Andrews, Kunal Mondal, Taylor Neumann, Jorge A Cardenas, Justin Wang, Dishit P. Parekh, Yiliang Lin, Peter Ballentine, Michael D. Dickey, and Aaron D Franklin ACS Nano, Just Accepted Manuscript • DOI: 10.1021/acsnano.8b00909 • Publication Date (Web): 09 May 2018 Downloaded from http://pubs.acs.org on May 9, 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 Nano

Patterned Liquid Metal Contacts for Printed Carbon Nanotube Transistors Joseph B. Andrews1, Kunal Mondal3, Taylor Neumann3, Jorge A. Cardenas1, Justin Wang1, Dishit P. Parekh3, Yiliang Lin3, Peter Ballentine1, Michael D. Dickey3*, Aaron D. Franklin1,2* 1

Department of Electrical and Computer Engineering, Duke University, Durham, NC 27708, USA 2

3

Department of Chemistry, Duke University, Durham, NC 27708, USA

Department of Chemical and Biomolecular Engineering, North Carolina State University, Raleigh, NC 27695, USA *Co-corresponding author Michael Dickey. Email: [email protected], Tel: +1-919-513-0273 * Co-corresponding author Aaron Franklin. Email: [email protected], Tel: +1-919-681-9471

Table of Contents Graphic

Abstract: Flexible and stretchable electronics are poised to enable many applications that cannot be realized with traditional, rigid devices. One of the most promising options for low-cost stretchable transistors are printed carbon nanotubes (CNTs). However, a major limiting factor in stretchable CNT devices is the lack of a stable and versatile contact material that forms both the interconnects 1 ACS Paragon Plus Environment

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 Nano

and contact electrodes. In this work, we introduce the use of eutectic gallium-indium (EGaIn) liquid metal for electrical contacts to printed CNT channels. We analyze thin-film transistors (TFTs) fabricated using two different liquid metal deposition techniques – vacuum filling PDMS microchannel structures and direct-writing liquid metals on the CNTs. The highest performing CNT-TFT was realized using vacuum-filled microchannel deposition with an in situ annealing temperature of 150 ºC. This device exhibited an on/off ratio of more than 104 and on-currents as high as 150 µA/mm – metrics that are on par with other printed CNT-TFTs. Additionally, we observed that at room temperature, the contact resistances of the vacuum-filled microchannel structures were 50% lower than those of the direct-write structures, likely due to the poor adhesion between the materials observed during the direct-writing process. The insights gained in this study show that stretchable electronics can be realized using low-cost and solely solution processing techniques. Furthermore, we demonstrate methods that can be used to electrically characterize semiconducting materials as transistors without requiring elevated temperatures or cleanroom processes.

Key words: liquid metal, direct-writing, eutectic gallium-indium (EGaIn), stretchable electronics, thin-film transistor (TFT); carbon nanotube (CNT); nanomaterials Accelerated interest in sensors and systems has driven the need for stretchable and flexible electronic devices.1,2 Applications including wearable sensors,3,4 biosensors,5–7 soft robotics,8 and many more require electronics with performance that is consistent regardless of the stretching/flexing that occurs in real-world, mobile environments.9–14 The most crucial component required for computation and signal processing is a transistor, which can also be used as a sensing device.

2 ACS Paragon Plus Environment

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

ACS Nano

Over the past 20 years, significant progress has been made in breaking from traditional, rigid electronics.15–20 The field originated with the invention of solution-processed, polymer semiconductors.15 These materials allow for different fabrication methods than traditional siliconbased transistors. While the polymer semiconductor devices presented something exciting, both their electrical performance and overall robustness proved to be limited. More recently, the advent of solution-processed nanomaterials has presented a more viable path towards developing truly flexible, printed electronic transistors with the needed performance and robustness.16,18–20 While there is an abundance of research work in the field of nanomaterial-based flexible electronics, there has been little exploration into developing truly stretchable electronics. One field of study that merits increased attention is the interface between stretchable conductors and semiconductors, a crucial component to the development of truly stretchable electronic systems. Single-walled carbon nanotubes (CNTs) are promising semiconducting nanomaterials because they are readily available, printable, and have excellent transport properties.21 Over the last decade, CNT thin-film transistors (CNT-TFTs) – with channels composed of dense, thin-film networks of CNTs – have exhibited performance well beyond any polymeric semiconductor-based devices with the ability to enable complex circuits and useful sensors.3,22,23 What’s more, in thin films of sufficient density, the connections between CNTs will not be broken under a certain amount of strain. In recent years, stretchable transistors that utilize semiconducting CNTs have been shown to remain viable during and after strain.24,25 However, the on-current and total resistance of the device is significantly altered by strain as low as 20 %. This is often due to failures in the source and drain contacts of the transistor. In our previous work, we have seen Ag nanoparticle contacts delaminate and crack under strains as low as 3.51%.20 Hence, there is a need for identifying stretchable conducting contacts for CNT thin-film transistors (CNT-TFTs).


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

ACS Nano

A promising candidate material to use as the conductor in flexible electronics is a liquid metal alloy, specifically eutectic gallium-indium, commonly known as EGaIn.8,26 One could argue that the best material to be encapsulated in an elastomer for stretchable applications would be a liquid, which remains continuous within the elastomeric structure even under large amounts of strain. When liquid metal is used as a wire, the only performance degradation occurs in a resistance change that is caused by an increase in length and a decreasing cross-sectional area.27 There are two common methods to deposit liquid metal at room temperature with excellent dimensional control: direct-write printing28,29 and microchannel injection/filling.26,27,30,31 Direct-write printing has the advantage of being an additive, maskless patterning technique, while microchannel injection/filling allows for much finer geometric control. While a large amount of research has been done on the fabrication and performance of liquid metals as stretchable conductors,32 even in using liquid metals as gate conductors for transistors33 and as a contact to graphene,34,35 there has been no work done regarding the use of a liquid metal as the source and drain contact material in a CNT-based transistor. Interestingly, the liquid metal mercury was heavily used in preliminary CNT conductance experiments shortly after their discovery. In these initial experiments, CNTs were dipped into mercury liquid metal crucibles to form conformal contacts to study conductance.36,37 These experiments provided the basis for initial electrical experiments with CNTs but have little influence on this study due to the lack of liquid metal patterning, and the simple two terminal devices that were created. In this work, we investigate the use of EGaIn as contact electrodes for printed CNT-TFTs. Our devices utilize aerosol jet printed CNTs for the semiconducting channel with eutectic galliumindium liquid metal for the source and drain contacts. The performance difference (in terms of contact resistance, on-current, and on-off current ratio) is studied for two distinct liquid metal


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

ACS Nano

deposition techniques: microchannel filling and contact printing. Methods to improve device performance are also explored, including in situ and post-process annealing. In addition, the simplicity with which this method forms electrical contacts may be appealing as a research tool for quickly characterizing traces of CNTs without the need for cleanroom processing (i.e. deposition of metal traces using vacuum processing). Results and Discussion To explore the contact interface between liquid metal and carbon nanotubes, substrategated carbon nanotube thin-film transistors were fabricated using a combination of aerosol jet printing and two distinct liquid metal deposition techniques. Substrate-gated transistors allow for a consistent and well understood gate/dielectric combination, which eliminates unnecessary variables in studying the source/drain contact interface. Conducting silicon substrates (p++, boron doped) act as the gate of the transistor, while a 300 nm thermally grown SiO2 insulating layer acts as the dielectric. The substrates were first cleaned using acetone, IPA, and DI water. Next, the substrates were exposed to an O2 plasma treatment to remove any organic contamination. The surface was then functionalized with a monomer (poly-L-lysine, 0.1 wt. % in H2O) through incubation to promote CNT adhesion. A CNT ink with a 99.9% semiconducting purity was printed using an aerosol jet printer. The aerosol jet printing operation works by atomizing liquid inks into an aerosol which are then carried to the substrate by an inert carrier gas. In addition, an inert gas flows annularly along the outer edge of the deposition nozzle to ensure that the aerosolized ink follows a focused path down to the substrate and prevents the nozzle from clogging. Due to the described operation, the printer is compatible with a wide variety of nanomaterial-based inks and allows for resolution down to 15 µm. The CNT ink was diluted to a concentration of 0.01 mg/ml and printed using a sheath gas 5 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

ACS Nano

flow of 40 sccm and a carrier gas flow of 27 sccm while the atomization current was held at 330 mA. A schematic of the printing process can be observed in Figure 1A. After printing, the channel was rinsed with toluene, dried with an N2 gas, and then baked at 150 °C to remove excess solvent and surfactant. The liquid metal source and drain contacts were then deposited using two common methods: vacuum filling patterned microchannels and direct-write printing. For the vacuum filling, the standard replica molding method was used to fabricate microchannels (~50 µm of channel height and varying width) composed of polydimethylsiloxane elastomers (PDMS) (Sylgard-184 from Dow Corning, USA). The process for fabricating the microchannels is exhibited in depth in Supplemental Figure S1. After replica molding of elastomers against a lithographic mold, a biopsy puncher was used to make an inlet hole with a diameter of 1 mm. There is no outlet. This molded PDMS was then bonded to the SiO2 wafer gate electrode surface with the help of oxygen plasma to complete the microfluidic device. The printed CNT layer on SiO2 wafer was covered by a thin paper to protect form plasma etching. A puddle of liquid metal (EGaIn, 75 wt. % gallium and 25 wt. % indium) was placed over the inlet and the device introduced inside a vacuum desiccator for half an hour to ensure complete removal of air from the microchannels. After releasing the vacuum, the desiccator chamber fills with air, which pushes the EGaIn into the microchannels using the pressure differential. This approach, described in detail in the literature, has the appeal that the microchannels fill in a hands-free manner.38,39 Source and drain electrodes were also deposited by direct-write printing. Direct-write printing has been shown to be an effective method for the deposition of high resolution stretchable conductors in both 2- and 3-dimensions.28,29,40,41 Contact printing removes the need for fabricating microfluidic channels and places the metal only where it is needed based on the digital pattern provided to a 3-axis motion-controlled printing stage. In this process, liquid metal is loaded in a 3


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

ACS Nano

cc syringe barrel connected to a pressure actuator (Nordson EFD Ultimus V). Pulling a modest amount of vacuum (< 3 kPa) in the head space above the metal prevents the liquid metal from leaking from the syringe and allows control over the amount of metal being printed. When the liquid metal is brought in contact with a substrate, the metal adheres to the substrate, and as the stage moves, the metal is sheared out of the syringe, thereby depositing traces of liquid metal onto the substrate. The mechanical strength of the thin, passivating oxide skin that forms instantaneously on the surface of liquid metal helps to stabilize the printed geometries against the destabilizing effects of gravity and the high surface tension of the liquid metal. The computercontrolled motion of the nozzle allows for printing of arbitrarily complex liquid metal patterns. Figure 1D shows lines being printed using a 22-gauge conical polypropylene nozzle (nominal inner diameter of 0.413 mm). A photograph of the setup, coupled with a schematic, can be observed in Supplemental Figure S2.

Figure 1: Printed CNT-TFT with fabrication of liquid metal contacts. (A) Schematic illustrating the operation of the aerosol jet printer depositing the semiconducting CNTs. Schematics of the two methods for liquid metal contact deposition: (B) direct-write printing and (C) vacuum filling. Photographs of the (D) direct-write printing and (E) vacuum filling methods.


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

ACS Nano

Page 8 of 23

After the liquid metal electrodes were deposited, the completed transistors were electrically characterized by measuring the ID-VGS and ID-VDS relationships. The device characteristics for the microchannel and printed contacts, both done at room temperature with no post-anneal, can be seen in Figure 2. Both devices have a channel length of 200 µm and a channel width of 500 µm. The microchannel-deposited contact sample exhibits an on-current approximately 10x higher than that of the direct-write sample, as well as an on/off current ratio of 104. While the off-currents and threshold voltages are similar, the transconductance of the microchannel vastly outperforms that of the printed contacts, which is indicative of a better overall contact interface. The field-effect mobility for each device was estimated using the following equation: 𝜇"## = *

%&' ()

&' +,- ./0

. LCH and

WCH represent the channel length and width respectively, gm is the device transconductance, Cox is the oxide capacitance, and VDS is the drain to source voltage applied. The oxide capacitance was estimated using the parallel plate model, where 𝐶23 =

56 5,7

, which was calculated to be 1.15 x

10-8 F/cm2. The effective mobilities are estimated to be 3.42 and 2.35 cm2/(V•s) for the microchannel device and the direct-write printed device, respectively. Another interesting component, seen in the output curves (Figure 2B), is the evidence of a significant transport barrier at the liquid metal-CNT interface based on the nonlinear slope at low drain-source biases (VDS). This may be attributed to the oxide skin of the liquid metal, which contributes to the native, well understood Schottky barrier observed for typical CNT transistors.42 The hysteretic characteristics for transistors stemming from both deposition methods can be seen in Figure S3. The microchannel devices exhibit slightly reduced hysteresis in comparison with the contact printed due to the PDMS coverage of the CNT channel. Also, the microchannel contacts show promise as a stretchable contact interface. A two-terminal resistive device was fabricated using unsorted (conducting and


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

ACS Nano

semiconducting) CNTs and EGaIn. It was found that the resistance changed by a factor of 0.1 after an initial applied strain and then exhibited no change after additional applied strain (Figure S9).

Figure 2: Transistor characteristics for both the vacuum-filled (blue) and contact-printed (red) liquid metal contact-CNT devices. (A) Transfer and subthreshold characteristics at VDS = -5 V and (B) output characteristics with VGS varying from -40 V to -10 V in -10 V increments. Both devices had a channel length of 200 µm and a channel width of 500 µm. The poorer performance of the contact printed liquid metal device is attributed to the poor EGaIn surface adhesion to the CNTs, and thus overall CNT coverage, compared to the microchannel deposited liquid metal device. While the vacuum-filled microchannel forces the liquid metal into contact with the CNT thin-film via pressure, the contact printing relies on surface adhesion to help shear the metal from the nozzle. A control experiment was completed in which the direct-write printed sample was subjected to the same vacuum conditions as the microchannel samples. There was no improvement in the electrical performance, and therefore, the improvement of the microchannel devices cannot be attributed to the applied vacuum. The results from this control experiment can be seen in Figure S6. Optical images for each contact structure are shown in Figures 3A and 3B. The width of the contact printed electrodes can be seen to become thinner


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

ACS Nano

over the CNT channel, indicating poor surface adhesion, while no evidence of poor adhesion can be seen with the microchannel devices. The poor surface adhesion for the direct-write deposited liquid metal can be explained through the operating mechanism. As the needle moves across the printing surface with a standoff distance of 50 µm, the liquid metal adheres to the silicon wafer and shears out of the nozzle. Literature suggests that surface roughness – even on the order of a few nanometers – can significantly lower the adhesion force of the oxide-coated liquid metal to a substrate.43–47 Through atomic force microscopy, the CNT channel was found to have a measured RMS surface roughness of 3.96 nm (Supplemental Figure S4). This roughness is also evident in the SEM of the CNT channel displayed in Figure 3C. In addition, our experience printing on many substrates suggests that the oxide-coated metal is more likely to adhere to hydroxide-coated surfaces – such as the silicon wafer – than carbonaceous surfaces such as CNTs. This poor adhesion leads to a visible change exhibited in Figure 3B, between the printed features on SiO2 where the liquid metal wets uniformly and the CNT channel where the liquid metal rests, but in thinner traces. Furthermore, a crevice directly at the CNT channel (Figure 3D) corroborates our hypothesis of an interruption in surface adhesion at the SiO2-CNT junction.


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

ACS Nano

Figure 3: Optical images of the (A) vacuum-filled microchannel device and (B) contact-printed device taken at 2.5x magnification. (C) Representative SEM image of the dense network of CNTs within the printed, semiconducting thin-film channel. (D) SEM image of a tilted sample at the interface between direct printed liquid metal and the CNT film.

After identifying that the vacuum-filled microchannel was the best technique for depositing liquid metal electrodes, efforts were made to improve the contact performance through thermal annealing during and after the deposition. Devices were fabricated with in situ annealing temperatures of 80 and 150 ºC, while another device was exposed to a post-deposition anneal of 150 ºC. It is important to note is that all devices are initially annealed at 150 ºC directly after the CNT channels are deposited. The following experiments, however, also include an additional annealing step either during or after the liquid metal deposition. The electrical characteristics, including a bar graph displaying the differences in on-current, can be seen in Figure 4. The in situ anneal led to the most significant increase in on-current. We are unsure why it improves the performance, but possible reasons could be a lower viscosity of the metal, changes in surface interactions (better wetting), or the in situ anneal could drive some contaminant molecules off the CNTs prior to the liquid metal flooding the microchannel. Overall, the in situ anneal at 150 ºC yielded an excellent on-off current ratio of more than 104 and an on-current that is greater than 2x 11 ACS Paragon Plus Environment

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

ACS Nano

the on-current of a post-annealed device at the same temperature. Furthermore, the on-current of the 150 ºC in situ anneal is 50% greater than that of an in situ anneal at 80 ºC, indicating that higher temperature anneals have more positive effects. Note, the on-current extraction was adjusted for the variation in threshold voltage. The transistor characteristics for many different lengths from a single representative device can be seen in supplemental Figure S5.

Figure 4: Impact of annealing on vacuum-filled microchannel liquid metal-CNT contacts. (A) Subthreshold, (B) transfer, and (C) output characteristics for vacuum-filled liquid metal contacted CNT-TFTs with various deposition temperatures and post-process anneals. The purple and gray represent in situ anneals at 150 ºC and 80 ºC, respectively, while the maroon displays the characteristics from a device with a 150 ºC post-deposition anneal. (D) Comparison of the oncurrent (extracted at consistent overdrive voltage to account for threshold voltage differences) for each process technique. All devices had a channel length of 200 µm and a channel width of 500 µm.

Another crucial parameter to understanding the metal-semiconductor interface is the contact resistance. To experimentally estimate this parameter, TLM (transmission line model) structures were fabricated, consisting of an array of adjacent devices of varying channel length on the same, continuous CNT thin film trace. The contact resistance is extracted by plotting the


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

ACS Nano

resistance for each device vs. the channel length. After performing a linear fit, the y-intercept will represent the resistance of a device at a channel length of 0 µm. One half of this value can then be used to experimentally estimate the resistance that stems from a single contact.48,49 Through this method, we can ignore variations due to fluctuating sheet resistances and simply observe the metrics associated with the liquid metal-CNT interface. A plot displaying the extraction of the contact resistance, coupled with a bar graph of the contact resistances vs. fabrication method can be found in Figures 5A and 5B, respectively. All resistances were obtained at a VDS of -1 V and an overdrive gate voltage of VGS-VT = -30 V. The overdrive gate voltage will adjust for variation in threshold voltage among the devices. Through the contact resistance extraction, we obtain a more detailed glimpse into the benefits of the microchannel. The resistance associated with the contact for the microchannel device with no anneal, 117 MΩ•µm, is 33% of the contact resistance associated with the direct printing, 350 MΩ•µm. This is consistent with the observed differences in on-current of the devices. Furthermore, we identified that the 150 ºC in situ annealed device had the least contact resistance: 45.61 MΩ•µm. This value is of the same magnitude as the ≈ 13 MΩ•µm that is observed from fully developed, printed silver nanoparticle top contacts for CNT-TFTs.50 The decrease in contact resistance, relative to the direct-write samples, most likely arises from the liquid metal adhesion and conformal coverage of the network of CNTs. The microchannel forces the liquid metal into a more conformal interface with the CNT trace, helping to overcome the surface forces that clearly affect the printed EGaIn contacts. Heating during filling of the liquid metal in the microchannel further improves this contact. It should be noted that there are other ways to potentially improve the contact resistance. One possible option is the addition of a sparse, metallic adhesion layer. Additional experiments


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

ACS Nano

were attempted to validate this approach, in which a sparse layer of silver nanoparticles was printed onto the CNTs (covering the entire CNT film, without discrimination of channel or contact regions) prior to liquid metal deposition. Unfortunately, this attempt led to transistors with extremely low on/off ratios, without significant contact resistance reduction (see Fig. S7); however, a more thorough study of the right combination of ink, print parameters, and thermal treatments could yield considerable improvement in contact resistance.

Figure 5: Contact resistance extraction for various liquid metal deposition techniques. (A) Transmission line model (TLM) plots used for extracting sheet and contact resistance and (B) bar graphs of the extracted contact resistances for all deposition and treatment techniques used for the liquid metal contacts.

Conclusion Overall, eutectic gallium-indium liquid metal alloy (EGaIn) has been shown to be a viable contact material for CNT-based transistors. We have investigated two methods of liquid metal deposition and conclusively demonstrated the benefits and limitations to each. The highest performing transistor was realized using liquid metal contacts deposited using vacuum-filled PDMS micro-channels with an in situ anneal of 150 ºC. This device exhibited an on-off current


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

ACS Nano

ratio of more than 104, and an on-current of 150 µA/mm at VDS = -5 V with a channel length of 200 µm. The contact resistance from the vacuum-filled microchannel metal electrodes was found to be approximately 67% lower than the contact resistance associated with the 3D printed liquid metal contacts. Furthermore, the contact resistance from devices fabricated with the in situ annealed deposition techniques exhibited values on par with non-stretchable nanomaterial-based contacts. Looking forward, liquid metal offers many advantages over more traditional, solutionprocessed conductive materials, while providing the added benefit of being extremely stretchable without degradation. Hence, this work provides a significant advancement towards robust stretchable electronics from CNT-based devices with liquid metal contacts. It also provides a useful laboratory tool for making source and drain contacts to devices in a simple manner. Materials and Methods Substrate cleaning: Prior to depositing the CNTs, the silicon wafers were first rinsed using DI water. Next, the wafers were ultrasonicated in acetone and subsequently IPA for 5 minutes each. The wafers were again rinsed using DI water and dried with an N2 gas. Finally, the wafers underwent an O2 plasma cleaning treatment for 4 minutes at 100 W under vacuum. Wafer Functionalization: Prior to printing the semiconducting CNTs, the substrate was functionalized by incubation in poly-l-lysine (0.1 % w/v in H2O) for 5 minutes to promote CNT adhesion. After incubation, the remaining monomer was rinsed off using DI water and the sample was dried with N2 gas. Printing CNT Channels: The semiconducting CNT ink (IsoSol-S100) with a purity of 99.9% was procured from Nanointegris Inc., Canada. The received concentration was 0.05 mg/ml s-CNTs dispersed in an aromatic solution. The ink was further diluted with toluene to a final concentration of 0.01 mg/ml. The ink was printed using an aerosol jet printer (AJ-300 from Optomec Inc., USA) following ultrasonication for 1 hour to reduce CNT aggregation. The ink was printed using a 150 15 ACS Paragon Plus Environment

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

ACS Nano

µm diameter nozzle, with sheath and carrier gas flow rates set to 40 and 27 sccm, respectively. The nozzle speed was fixed at 1 mm/s. Both the water bath and the printing enclosure were at room temperature during printing, while the platen was held at 50 ºC to facilitate solvent evaporation. After printing, the substrate was rinsed with toluene and then annealed at 150 ºC for 30 minutes to further remove excess solvent/surfactant. Vacuum Filling Microchannels with Liquid Metal: Microchannels were first fabricated using the standard replica molding method of casting silicone over a topographical mold created using photolithography. The microchannels had various lengths and widths and a 50 µm height. The microchannels were bonded to the silicon substrate containing the CNT channel with the microchannels perpendicular to the printed CNT channel. Next, the microchannel was punctured to allow an inlet for the liquid metal. A drop of liquid metal (eutectic gallium-indium, EGaIn) was placed over the inlet and the sample was placed in a vacuum desiccator (Thermo-Scientific Vacuum Oven). The sample was then subjugated to a light vacuum for 30 minutes. The sample is then reintroduced to atmospheric pressure, forcing the liquid metal drop to fill the channels in a hands-free manner. This process is similar to prior work with EGaIn.38,39 The process was completed with the oven temperature held at 80 ºC and 150 ºC for the in situ annealed samples. Printing Eutectic Gallium Indium Liquid Metal (EGaIn): The direct-write printing was carried out using a custom-made setup as shown in the Supplemental Figure S2. EGaIn is placed in a 3 cc syringe barrel that is connected to a pressure actuator (Nordson EFD Ultimus V). A 22-gauge conical polypropylene needle is then connected to the syringe barrel, and a pressure of < 3 kPa is used to hold the liquid metal in the needle. The syringe is then loaded onto a numerically controlled X-Y-Z stage. The liquid metal is deposited by initially contacting the surface with the needle. As the needle moves across the surface the liquid metal is deposited onto the substrate due to the adhesion of the liquid metal to the surface and the resulting shear forces. The thickness of the deposited liquid metal lines was approximately 100 µm. The lines were printed on the SiO2 substrate perpendicular to the printed CNTs. Characterizations of Printed Devices: The optical images of the liquid metal-CNT devices were obtained using a Zeiss Axio microscope and a digital camera. The SEM images were acquired 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

ACS Nano

using a FEI XL30 (SEM-FEG, USA). The devices were electrically characterized using a B1500 Device Analyzer (Agilent, USA) and Signatone probe station. Supporting Information The Supporting Information for this article can be found online and contains a detailed process flow for the microchannel fabrication, a photograph of the direct-write setup, a characteristic hysteresis curve, an AFM of the CNT channel, transistor curves at various channel lengths, a before and after transistor curve of the direct-write sample being subjected to the same conditions as the vacuum deposited liquid metal, a transistor curve of a device fabricated with an Ag nanoparticle adhesion layer, upside down optical images of the devices fabricated on glass, and a two-terminal device’s characteristics under strain. Acknowledgements Parekh, Neumann, and Dickey are grateful for support from the National Science Foundation (ERC EEC-1160483 and CMMI-1362284). References (1)

Rogers, J. A.; Someya, T.; Huang, Y. Materials and Mechanics for Stretchable Electronics. Science. 2010, 327, 1603–1607.

(2)

Nathan, A.; Ahnood, A.; Cole, M. T.; Lee, S.; Suzuki, Y.; Hiralal, P.; Bonaccorso, F.; Hasan, T.; Garcia-Gancedo, L.; Dyadyusha, A.; Haque, S.; Andrew, P.; Hofmann, S.; Moultrie, J.; Chu, D.; Flewitt, A.; Ferrari, A.; Kelly, M.; Robertson, J.; Amartunga G.; Milne, W. Flexible Electronics: The Next Ubiquitous Platform. Proc. IEEE. 2012, 100, 1486–1517.

(3)

Jung, Y.; Park, H.; Park, J.-A.; Noh, J.; Choi, Y.; Jung, M.; Jung, K.; Pyo, M.; Chen, K.; Javey, A.; Cho, G. Fully Printed Flexible and Disposable Wireless Cyclic Voltammetry Tag. Sci. Rep. 2015, 5, 8105.

(4)

Khan, S.; Tinku, S.; Lorenzelli, L.; Dahiya, R. S. Flexible Tactile Sensors Using Screen17 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

ACS Nano

Printed P(VDF-TrFE) and MWCNT/PDMS Composites. IEEE Sens. J. 2015, 15, 3146– 3155. (5)

Kim, D.; Viventi, J.; Amsden, J. J.; Xiao, J.; Kim, Y.; Blanco, J. A; Panilaitis, B.; Frechette, E. S.; Contreras, D.; Kaplan, D. L.; Omenetto, F. G.; Huang, Y.; Hwang, K. C.; Zakin, M. R.; Litt, B.; Rogers, J. A. Dissolvable Films of Silk Fibroin for Ultrathin, Conformal BioIntegrated Electronics. Nat. Mater. 2011, 9, 511–517.

(6)

Sarkar, D.; Liu, W.; Xie, X.; Anselmo, A. C.; Mitragotri, S.; Banerjee, K. MoS2 FieldEffect Transistor for Next-Generation Label-Free Biosensors. ACS Nano 2014, 8, 3992– 4003.

(7)

Xiang, L.; Wang, Z.; Liu, Z.; Weigum, S. E.; Yu, Q.; Chen, M. Y. Inkjet Printed Flexible Biosensor Based on Graphene Field Effect Transistor. IEEE Sens. J. 2016, 16, 1–1.

(8)

Kubo, M.; Li, X.; Kim, C.; Hashimoto, M.; Wiley, B. J.; Ham, D.; Whitesides, G. M. Stretchable Microfluidic Radiofrequency Antennas. Adv. Mater. 2010, 22, 2749–2752.

(9)

Manunza, I.; Bonfiglio, A. Pressure Sensing Using a Completely Flexible Organic Transistor. Biosens. Bioelectron. 2007, 22, 2775–2779.

(10) Kim, S.; Bito, I.; Jeong, S.; Tentzeris, M. M. A Flexible Hybrid Printed RF Energy Harvester Utilizing Catalyst - Based Copper Printing Technologies for Far-Field RF Energy Harvesting Applications. 2015, 339, 0–3. (11) Someya, T.; Sekitani, T.; Iba, S.; Kato, Y.; Kawaguchi, H.; Sakurai, T. A Large-Area, Flexible Pressure Sensor Matrix with Organic Field-Effect Transistors for Artificial Skin Applications. 2004, 101, 9966–9970. (12) Larson, C.; Peele, B.; Li, S.; Robinson, S.; Totaro, M.; Beccai, L.; Mazzolai, B.; Shepherd, R. Highly Stretchable Electroluminescent Skin for Optical Signaling and Tactile Sensing.


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

ACS Nano

Science. 2016, 351, 1071–1074. (13) Andrews, J. B.; Cardenas, J. A.; Mullett, J.; Franklin, A. D. Fully Printed and Flexible Carbon Nanotube Transistors Designed for Environmental Pressure Sensing and Aimed at Smart Tire Applications. IEEE SENSORS 2017, 1–3. (14) Andrews, J.; Cao, C.; Brooke, M.; Franklin, A. Noninvasive Material Thickness Detection by Aerosol Jet Printed Sensors Enhanced through Metallic Carbon Nanotube Ink. IEEE Sens. J. 2017, 17, 4612–4618. (15) Sheraw, C. D.; Zhou, L.; Huang, J. R.; Gundlach, D. J.; Jackson, T. N.; Kane, M. G.; Hill, I. G.; Hammond, M. S.; Campi, J.; Greening, B. K.; Francl, J.; West, J. Organic Thin-Film Transistor-Driven Polymer-Dispersed Liquid Crystal Displays on Flexible Polymeric Substrates. Appl. Phys. Lett. 2002, 80, 1088–1090. (16) Das, S.; Gulotty, R.; Sumant, A. V.; Roelofs, A. All Two-Dimensional, Flexible, Transparent, and Thinnest Thin Film Transistor. Nano Lett. 2014, 14, 2861–2866. (17) Zhu, H.; Hu, L.; Cumings, J.; Huang, J.; Chen, Y. Highly Transparent and Flexible Nanopaper Transistor. ACS Nano 2013, 2106–2113. (18) Cao, Q.; Kim, H.; Pimparkar, N.; Kulkarni, J. P.; Wang, C.; Shim, M.; Roy, K.; Alam, M. a; Rogers, J. A. Medium-Scale Carbon Nanotube Thin-Film Integrated Circuits on Flexible Plastic Substrates. Nature 2008, 454, 495–500. (19) Cao, Q.; Hur, S. H.; Zhu, Z. T.; Sun, Y.; Wang, C.; Meitl, M. a.; Shim, M.; Rogers, J. a. Highly Bendable, Transparent Thin-Film Transistors That Use Carbon-Nanotube-Based Conductors and Semiconductors with Elastomeric Dielectrics. Adv. Mater. 2006, 18, 304– 309. (20) Cao, C.; Andrews, J. B.; Franklin, A. D. Completely Printed, Flexible, Stable, and


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

ACS Nano

Hysteresis-Free Carbon Nanotube Thin-Film Transistors via Aerosol Jet Printing. Adv. Electron. Mater. 2017, 3, 5, 1-10. (21) Park, S.; Vosguerichian, M.; Bao, Z. A Review of Fabrication and Applications of Carbon Nanotube Film-Based Flexible Electronics. Nanoscale 2013, 5, 1727. (22) Ha, M.; Xia, Y.; Green, A. A.; Zhang, W.; Renn, M. J.; Kim, C. H.; Hersam, M. C.; Frisbie, C. D. Printed, Sub-3V Digital Circuits on Plastic from Aqueous Carbon Nanotube Inks. ACS Nano 2010, 4, 4388–4395. (23) Ha, M.; Seo, J.-W. T.; Prabhumirashi, P. L.; Zhang, W.; Geier, M. L.; Renn, M. J.; Kim, C. H.; Hersam, M. C.; Frisbie, C. D. Aerosol Jet Printed, Low Voltage, Electrolyte Gated Carbon Nanotube Ring Oscillators with Sub-5 ms Stage Delays. Nano Lett. 2013, 13, 954– 960. (24) Chortos, A.; Koleilat, G. I.; Pfattner, R.; Kong, D.; Lin, P.; Nur, R.; Lei, T.; Wang, H.; Liu, N.; Lai, Y. C.; Kim, M. G.; Chung, J. W.; Lee, S.; Bao, Z. Mechanically Durable and Highly Stretchable Transistors Employing Carbon Nanotube Semiconductor and Electrodes. Adv. Mater. 2016, 28, 4441–4448. (25) Xu, F.; Wu, M.; Safron, N. S.; Roy, S. S.; Jacobberger, R. M.; Bindl, D. J.; Seo, J.; Chang, T.; Ma, Z.; Arnold, M. S. Highly Stretchable Carbon Nanotube Transistors with Ion Gel Gate Dielectrics. Nano Lett. 2014, 14, 682–686. (26) Dickey, M. D.; Chiechi, R. C.; Larsen, R. J.; Weiss, E. A.; Weitz, D. A.; Whitesides, G. M. Eutectic Gallium-Indium (EGaIn): A Liquid Metal Alloy for the Formation of Stable Structures in Microchannels at Room Temperature. Adv. Funct. Mater. 2008, 18, 1097– 1104. (27) Park, Y. L.; Chen, B. R.; Wood, R. J. Design and Fabrication of Soft Artificial Skin Using


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

ACS Nano

Embedded Microchannels and Liquid Conductors. IEEE Sens. J. 2012, 12, 2711–2718. (28) Boley, J. W.; White, E. L.; Chiu, G. T. C.; Kramer, R. K. Direct Writing of Gallium-Indium Alloy for Stretchable Electronics. Adv. Funct. Mater. 2014, 24, 3501–3507. (29) Ladd, C.; So, J. H.; Muth, J.; Dickey, M. D. 3D Printing of Free Standing Liquid Metal Microstructures. Adv. Mater. 2013, 25, 5081–5085. (30) Park, C. W.; Moon, Y. G.; Seong, H.; Jung, S. W.; Oh, J. Y.; Na, B. S.; Park, N. M.; Lee, S. S.; Im, S. G.; Koo, J. B. Photolithography-Based Patterning of Liquid Metal Interconnects for Monolithically Integrated Stretchable Circuits. ACS Appl. Mater. Interfaces 2016, 8, 15459–15465. (31) Joshipura, I. D.; Ayers, H. R.; Majidi, C.; Dickey, M. D. Methods to Pattern Liquid Metals. J. Mater. Chem. C 2015, 3, 3834–3841. (32) Trlica, C.; Parekh, D. P.; Panich, L.; Ladd, C.; Dickey, M. D. 3-D Printing of Liquid Metals for Stretchable and Flexible Conductors. 2014, 90831D. (33) Chortos, A.; Lim, J.; To, J. W. F.; Vosgueritchian, M.; Dusseault, T. J.; Kim, T. H.; Hwang, S.; Bao, Z. Highly Stretchable Transistors Using a Microcracked Organic Semiconductor. Adv. Mater. 2014, 26, 4253–4259. (34) Ordonez, R. C.; Ha Yashi, C. K.; Torres, C. M.; Hafner, N.; Adleman, J. R.; Acosta, N. M.; Melcher, J.; Kamin, N. M.; Garmire, D. Conformal Liquid-Metal Electrodes for Flexible Graphene Device Interconnects. IEEE Trans. Electron Devices 2016, 63, 4018–4023. (35) Secor, E. B.; Cook, A. B.; Tabor, C. E.; Hersam, M. C. Wiring Up Liquid Metal: Stable and Robust Electrical Contacts Enabled by Printable Graphene Inks. Adv. Electron. Mater. 2018, 4, 1–6. (36) Wilson, N. R.; Cobden, D. H.; Macpherson, J. V. Single-Wall Carbon Nanotube Conducting


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

ACS Nano

Probe Tips. J. Phys. Chem. B 2002, 106, 13102–13105. (37) Frank, S.; Frank, S.; Poncharal, P.; Wang, Z. L.; Heer, W. A. Carbon Nanotube Quantum Resistors. Science. 2012, 1744, 10–13. (38) Lin, Y.; Gordon, O.; Khan, M. R.; Vasquez, N.; Genzer, J.; Dickey, M. D. Vacuum Filling of Complex Microchannels with Liquid Metal. Lab Chip 2017, 17, 3043–3050. (39) Bharambe, V.; Parekh, D. P.; Ladd, C.; Moussa, K.; Dickey, M. D.; Adams, J. J. VacuumFilling of Liquid Metals for 3D Printed RF Antennas. Addit. Manuf. 2017, 18, 221–227. (40) Parekh, D. P.; Ladd, C.; Panich, L.; Moussa, K.; Dickey, M. D. 3D Printing of Liquid Metals as Fugitive Inks for Fabrication of 3D Microfluidic Channels. Lab Chip 2016, 16, 1812– 1820. (41) Ahn, B. Y.; Duoss, E. B.; Motala, M. J.; Guo, X.; Park, S. Il; Xiong, Y.; Yoon, J.; Nuzzo, R. G.; Rogers, J. A.; Lewis, J. A. Omnidirectional Printing of Flexible, Stretchable, and Spanning Silver Microelectrodes. Science 2009, 323, 1590–1593. (42) Heinze, S.; Tersoff, J.; Martel, R.; Derycke, V.; Appenzeller, J.; Avouris, P. Carbon Nanotubes as Schottky Barrier Transistors. Phys. Rev. Lett. 2002, 89, 2–5. (43) Doudrick, K.; Liu, S.; Mutunga, E. M.; Klein, K. L.; Damle, V.; Varanasi, K. K.; Rykaczewski, K. Different Shades of Oxide: From Nanoscale Wetting Mechanisms to Contact Printing of Gallium-Based Liquid Metals. Langmuir 2014, 30, 6867–6877. (44) Kramer, R. K.; Boley, J. W.; Stone, H. A.; Weaver, J. C.; Wood, R. J. Effect of Microtextured Surface Topography on the Wetting Behavior of Eutectic Gallium-Indium Alloys. Langmuir 2014, 30, 533–539. (45) Liu, T.; Sen, P.; Kim, C. J. Characterization of Nontoxic Liquid-Metal Alloy Galinstan for Applications in Microdevices. J. Microelectromech. Syst. 2012, 21, 443–450.


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

ACS Nano

(46) Naidich, J. V.; Chuvashov, J. N. Wettability and Contact Interaction of Gallium-Containing Melts with Non-Metallic Solids. J. Mater. Sci. 1983, 18, 2071–2080. (47) Israelachvili, J. Intermolecular And Surface Forces; 3rd ed.; Academic Press: Burlington, MA, 2011, 415-467. (48) Reeves, G. K.; Harrison, H. B. Obtaining the Specific Contact Resistance from Transmission Line Model Measurements. IEEE Electron Device Lett. 1982, 3, 111–113. (49) Smith, J. T.; Franklin, A. D.; Farmer, D. B.; Dimitrakopoulos, C. D. Reducing Contact Resistance in Graphene Devices through Contact Area Patterning. ACS Nano 2013, 7, 3661–3667. (50) Cao, C.; Andrews, J. B.; Kumar, A.; Franklin, A. D. Improving Contact Interfaces in Fully Printed Carbon Nanotube Thin-Film Transistors. ACS Nano 2016, 10, 5221–5229.


Page 22 of 23

Patterned Liquid Metal Contacts for Printed Carbon Nanotube

Recommend Documents