Direct Wiring of Eutectic Gallium-Indium to a Metal Electrode for Soft

Subscriber access provided by UNIV OF LOUISIANA

Surfaces, Interfaces, and Applications

Direct Wiring of Eutectic Gallium-Indium to a Metal Electrode for Soft Sensor Systems Suin Kim, Jihye Oh, Dahee Jeong, and Joonbum Bae ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.9b05363 • Publication Date (Web): 08 May 2019 Downloaded from http://pubs.acs.org on May 8, 2019

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 26 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

Direct Wiring of Eutectic Gallium-Indium to a Metal Electrode for Soft Sensor Systems Suin Kim, Jihye Oh, Dahee Jeong, and Joonbum Bae

∗

Department of Mechanical Engineering, UNIST, Ulsan

E-mail: [email protected]

Phone: +82 (52)217 2335

Abstract For wider applications of the liquid metal-based stretchable electronics, an electrical interface has remained as a crucial issue, due to its fragile electromechanical stability and complex fabrication steps. In this study, a direct writing-based technique is introduced to form the writing paths of conductive liquid metal (eutectic Gallium-Indium, eGaIn) and electrical connections to o-the-shelf metal electrodes in a single process. Specically, by extending eGaIn wires written on a silicone substrate, the eGaIn wires were physically connected to the ve dierent metal electrodes, of which stability as an electrical connection was investigated. Among the ve dierent surface materials, the metal electrode nished by electroless nickel immersion gold (ENIG) had reproducible and low contact resistance without time-dependent variation. In our experiments, it was veried that the electrode part made by an ENIG-nished exible at cable (FFC) were mechanically (strain≤ 100%, pressure≤ 600kP a) and thermally (temperature≤ 180 ◦ C ) durable. By modifying trajectories of eGaIn wires, soft sensor systems were fabricated and tested to measure nger joint angles and ground reaction forces, composed with 10 sensing units, respectively. The proposed method enables the eGaIn-based soft sensors or circuits to be connected to the typical electronic components through a FFC or 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

weldable surfaces, using only o-the-shelf materials without additional mechanical or chemical treatments.

Keywords : Soft sensor

Eutectic Gallium-Indium, Liquid metal, Direct ink writing, Electrical interface,

.

Introduction Wearable devices have transformed into stretchable and deformable electronics, attached directly on the bodies as electronic skins. 1 Such a deformable device should be not only exible, but also stretchable to maintain its original functionality even under extreme mechanical strain from body movements. 2,3 For such applications, liquid-phase electronic devices have been actively developed based on the liquid metal, such as eutectic Gallium-Indium (eGaIn, 75wt% Ga and 25wt% In) due to its metallic conductivity and uidic recongurability. 4 A microudic channel lled with eGaIn has been built in an elastomeric body based on various fabrication techniques, which can be used as conductive paths for stretchable circuits or soft sensors. 5,6 Due to its superior stretchability, various types of wearable sensors have been developed, directly attached on the skins to measure nger, 7,8 wrist, 9 or leg motions. 10 Although eGaIn-based stretchable electronics have been studied vigorously, there have been two major technical barriers to wider applications to wearable sensors: a facile fabrication technique and a stable electrical connection. The various fabrication techniques for eGaIn-based stretchable electronics are well introduced in previous works. 5,6 Most fabrication methods require delicate hand skills and time-consuming fabrication steps, resulting in low productivity, low yield rate, and diculties in manufacturing system-level components including multiple microchannels (i.e. compactness) and modifying the design of the microchannel (i.e. programmability 11 ), because a customized mask or mold should be prepared. Furthermore, connection of electrodes within a single fabrication process have been achieved by none of the fabrication methods. Usually, wires are manually inserted into the 2


Page 2 of 26

Page 3 of 26 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


terminal of the channel, with mechanical sealing using silicone bonds. However, these connections are mechanically and electrically fragile, despite a lot of time and high dexterity of the hand are needed, as mentioned in many previous reports. 8,10 Mechanical durability of the interface is essential to prevent leakage of eGaIn from the microchannel. A stable electrical interface is also important to connect the microuidic device to other components (e.g. wire, connector, or electronics) for its proper functioning. More specically, interfacing the liquid metal with metal electrodes has been identied as a challenge due to alloying behavior of eGaIn with most metals 6,12,13 and minimal adhesion between eGaIn and most solids. 14 In detail, the silver electrode becomes electromechanically fragile due to alloying behavior of eGaIn. 13 To prevent the destructive behavior, indirect interfacing, such as inserting a solder ball, 15 graphene thin lm, 13 and z-axis conductive lm, 16 has been demonstrated. For the second issue, poor adhesion of eGaIn on most solids was reported in case of without rupturing the oxide skin, 14 resulting in inconsistent and unstable electrical connections between an electronic chip and eGaIn. 17 The poor wettability has been alleviated by applying HCl vapor 17 or patterning the conductor in honeycombed shape and using a nonionic surfactant, 18 resulting in increased mechanical contact area between eGaIn and the electrodes. However, such processes require specialized technique (e.g. sputtering) and additional materials or chemicals. Furthermore, the microuidic channel (sensors or electrical circuit) and the interface cannot be fabricated in a single process, which may increase complexity of manufacturing process. To deal with these technical issues, we propose a direct ink writing-based technique to form the conductive paths of eGaIn and the electrical connections in a single process without introducing additional chemicals and treatments. In the proposed technique, the eGaIn paths can be directly wired to o-the-shelf metal electrodes, by extending the eGaIn paths onto the metal electrodes. The aim of this study is to select a proper metal electrode satisfying electrical stability, among the commercially available nishing methods in printed circuit board (PCB) industry: electroless nickel immersion gold (ENIG), immersion silver 3



(Im-Ag), electroless nickel (E-less Ni), organic solderbility preservative (OSP), and immersion tin (Im-Sn). For each surface nish, stable connectivity with eGaIn is evaluated in terms of reproducibility and time-dependent sustainability of the contact resistance between eGaIn and the metal surface. Based on the comparison, a proper conductor material for the electrical interface is selected, of which mechanical and thermal stability are evaluated. As applications, two types of the soft sensor systems are fabricated and tested, including a glove and an insole-type sensors.

Result and Discussion

Direct Writing We tested direct writing on six dierent substrates as shown in Figure 1a. For test specimens, made-to-order boards (0.3 mm thickness including the metal plating) were prepared with metal traces (1 mm pitch, 20 pins, and 20 mm length) with four dierent surface nishes: ENIG, Im-Ag, E-less Ni, and OSP. Commercialized exible at cables (FFC) (1 mm pitch, 20 pins, and 0.15 mm thickness) nished by ENIG and Im-Sn were also prepared. Direct writing of eGaIn was attempted on each surface under the same writing conditions, including size of syringe tip, charged amount of eGaIn, distance between the syringe and the substrate, and speed of the syringe. To initiate direct writing, the droplet of eGaIn at the syringe tip should adhere to the surface, stretching and rupturing the oxide skin continuously. However, it was challenging for the droplet to adhere to the metal surfaces without disruption of the skin, resulting in extra-thin (or even disconnected) traces for the ENIG, E-less Ni, and OSP nished surfaces and a rolling lump of eGaIn for the surface nished with Im-Sn. Even under the same writing conditions, a line of eGaIn was continuously written on the surfaces nished with Im-Ag and covered with a silicone (Dragon Skin 30 made by Smooth On 19 ). The results can be understood by the dierent wettability of eGaIn on each surface, which can be elucidated by a height variation test, approximating a droplet of eGaIn at the syringe 4


Page 4 of 26

Page 5 of 26 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


D

(1,*

WV HW : ,' OD LQJ L WV U2

,P$J

(OHVV1L

263

,P6Q

'UDJRQ6NLQ

E

HW 9+ G HK

ศ

WFH UW 6

Figure 1: (a) Direct writing of eGaIn on the metal electrodes and the silicone substrate. (b) Height variation wettability tests of eGaIn on each surface. and withdrawing the syringe after the syringe reached the surface completely. 14 Figure 1b shows the original droplet and the maximally stretched (right before being detached from the substrate) shape of the droplet. As shown in the result, the droplet of eGaIn was easily separated from the ENIG, E-less Ni, OSP, and Im-Sn plated surfaces, while even pillars of eGaIn were built by high wettability of eGaIn on the Im-Ag-plated and silicone-coated surfaces. The moderate wettability of eGaIn on the Im-Ag surface is attributed to the vigorous alloying of eGaIn with silver. 20 The observed wetting behaviors of eGaIn originates from the nanoscale topology of the oxide skin, resulting in poor wettability on most solids and moderate wettability on the substrate with air permeability and viscoelastic property, such as silicone. 14 As a result, writing eGaIn directly on the metal electrodes is challenging except for the Im-Ag surface. As an alternative, we indirectly connected eGaIn wires as follows; 1) the eGaIn wires were extended from the silicone substrate, 2) raised in the vertical direction (building a pillar of eGaIn) around the electrode, 3) placed on the electrode surface, 4) stretched over the electrode, and 5) cut to nish writing. Such a method is physically feasible, because even a free standing structure can be built by eGaIn. 21 As shown in Figure 2a, the suggested method allowed reproducible writing of eGaIn on the dierent metal electrodes in spite of 5



(a)

(iii)

ENIG

Im-Ag

OSP

E-less Ni

Im-Sn (FFC)

Page 6 of 26

Electrode part Wire part

(b)

Sensing part (c)

(d)

ENIG

(FFC)

Figure 2: (a) Written eGaIn traces on the dierent metal electrodes. (b) Conguration of a soft sensor system including sensing, wire, and electrode parts. (c), (d) Extended view of the electrode part in the soft sensor system. poor wettability of eGaIn on the surfaces. Based on the proposed technique, all components in a soft sensor system can be written in a single process, e.g. a sensing glove with 10 sensing units (serpentine patterns) shown in Figure 2b. 22 The both terminals of each serpentine pattern are extended to the adjacent area of the electrode through the wire parts (Figure 2c), and connected to the metal electrode (Figure 2d). The electrode terminals in the opposite side can be accessed by electronic components (e.g. connectors, wires, and chips).

Electrical Stability We measured the resistance between the both terminals of the metal electrode, of which the other sides were connected through a 50 mm long eGaIn wire. The eGaIn wires can be written in the consistent and reproducible manner as proven in our previous work, 23 variation in each measurement might come from dierence in interfacial resistance between the eGaIn wire and the metal electrodes. To prevent oxidization and contamination of the 6


Page 7 of 26 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


Table 1: Resistance between the metal electrodes connected through an eGaIn wire (µ and σ are mean and standard deviations of the samples, respectively, and N is the number of the samples).

µ[Ω] σ[Ω] σ/µ N

ENIG 1.449 0.1129 0.0779 30

ENIG (FFC) 1.304 0.0449 0.0334 30

Im-Ag 1.445 0.1849 0.1279 30

E-less Ni 20.99 12.460 0.5936 27

OSP 8.595 7.3649 0.8569 28

Im-Sn 3.592 3.1735 0.8835 27

metal surfaces, the specimens remained packed in vacuum sealed bags, during delivery and storage before conducting the experiments. In addition, because only small amount of eGaIn (< 0.1 ml) was used, oxidization level of eGaIn hardly diers during the experiments. A total of 30 wires were measured for each surface material. The measured values are described in Table 1 and visualized as box plots shown in Figure 3. As shown in the result, the resistance values in the OSP, E-less Ni, and Im-Sn nished surfaces were signicantly larger and uncertain (large σ/µ), and even failures in connection (> kΩ) were observed (N < 30). In case of the ENIG and Im-Ag nished surfaces, resistance values were small and consistent (small σ/µ) without failure (N = 30). In a previous report, uncertain contact resistances were also observed, attributed to insucient mechanical contact area between eGaIn and a metal electrode. 17 Our experimental results, in case of the OSP, E-less Ni, and Im-Sn plated surfaces, can be understood similarly, because the proposed direct writing-based connection might not ensure a large mechanical contact area between eGaIn and a metallic surface. Nonetheless, stable electrical connection was immediately produced on the ENIG and Im-Ag plated surfaces. This result stems from good electrical conductivity of Au and Ag (about ten times higher than Ni and Sn) and alloying of eGaIn with Au and Ag. Cu has high conductivity and can be alloyed with eGaIn, but a layer of the organic compound may act as a diusion barrier in case of the OSP nish. To investigate time-dependent stability, resistance values of 10 wires were observed over 8 hours and after 6 months. As shown in Figure 4 (a), Each data set was represented as a box 7



Page 8 of 26

(a) (b) 2

1.5

1

ENIG

ENIG

ENIGFFC

ENIGFFC

Im-Ag

Im-Ag

OSP

E-less Ni

Im-SnFFC

Figure 3: Resistance between terminals of the metal electrodes connected through an eGaIn wire (50 mm). (a) Each box includes N data points, where the central red line indicates the median, the blue bock indicates range from 25th to 75th percentiles, and the whiskers indicate the extreme data range without outliers (plotted as red '+' symbols). (b) Extended view of the results from the ENIG and Im-Ag nished surfaces. plot (circular points, bold lines, and solid lines indicate medians, 25 to 75 percentile ranges, and minimum to maximum ranges, respectively). The resistance values were stable in the ENIG and Im-Ag nished surfaces, while vigorous time-dependent variations were observed in the OSP and Im-Sn nished surfaces. This time-dependent variation of resistance is attributed to variation in mechanical contact area due to gradual alloy formation between eGaIn and metals. 17,27 In case of the ENIG and Im-Ag nished surfaces, the resistance values were decreased after 6 months, which might attributed to gradually progressed alloy formation. None of the wires were disconnected, which can be an evidence for long-term stability of the connections. For the OSP and Im-Sn plated surfaces, the similar amount of the initial contact area was not enough to acquire suciently low contact resistance, due to lower conductivity of the surface materials (organic compounds or tin) than Au or 8


Page 9 of 26

(1,*

'XULQJ KRXUV

(1,*))&

$IWHU PRQWKV

'XULQJ KRXUV

263

$IWHUGD\V ฬ

'XULQJ KRXUV

$IWHU PRQWKV

$IWHUPRQWKV ฬ

& ))

G

'XULQJ KRXUV

$IWHU PRQWKV

%DUHVXUIDFH $IWHUGD\V ห

H*D,Q

,P$J

& ))

(OHVV1L 263 ,P6Q

263 ,P6Q

$IWHU PRQWKV

,P6Q))&

(1,*

(1,*

F

(1,*

(1,* ,P$J

& ))

$IWHU PRQWKV

263

(OHVV1L

ฬ

'XULQJ KRXUV

(OHVV1L

,P$J

:LWKLQPLQXWHV (1,*

E

$IWHU PRQWKV

(OHVV1L

'XULQJ KRXUV

,P$J

(1,*

D

,P6Q

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 4: (a) Time-dependent variation in resistance, Images captured by (b) a microscope (B011 5MP, Supereyes 24 ) (c) a stereo zoom microscope (AXIO ZOOM, ZEISS 25 ), and (d) a laser confocal microscope (OLS3100, Olympus 26 ). 9



Ag, resulting in decreased contact resistance according to time-dependent wetting of eGaIn onto the surface even within 8 hours. In case of the Im-Sn nished surface, the increasing tendency in resistance was observed also, which might be caused by decreased mechanical contact area between eGaIn and the surface or destructive behavior of eGaIn to the metal trace as observed in Ref. 13 In case of E-less Ni, no signicant time-dependent variations were observed due to poor wettability of eGaIn on Ni. 17 Except for the ENIG and Im-Ag nished surfaces, at least one abnormally high contact resistance (> 100 Ω) was observed and remained, even after 6 months. To observe the alloy formation, the metal surfaces interfacing with eGaIn wires were observed by microscopes (Figure 4b-d). The images in Figure 4b were captured by a simple microscope (B011 5MP, Supereyes 24 ) within a short time period (< 30 min). In the results, only the Im-Sn plated surfaces shows apparent alloy formation. The images in Figure 4c were captured by a stereo zoom microscope (AXIO ZOOM, ZEISS 25 ) after 7 days and 6 months, respectively. In case of the ENIG and Im-Ag nished surfaces, certain area of the metal electrode was alloyed with eGaIn, while the E-less Ni and OSP nished surfaces had no observable alloy formation. In case of Im-Sn nished surfaces, vigorous alloy formation was observed, even the plated surface was inated due to inux of eGaIn. Interestingly, as shown in Figures 4b,c, the speed of alloy formation was dierent in the ENIG-nished board and FFC, which might be due to dierent surface roughness. To investigate surface roughness and alloy formation, the bare and the eGaIn-interfaced surfaces were observed by a laser confocal microscope (OLS3100, Olympus 26 ) with 100 times magnication (Figure 4d). The tip of the eGaIn trace was observed to conrm dierence in surface roughness due to alloy formation. In case of the ENIG and E-less Ni nished surfaces, clear alloyed regions were observed. The alloy formation on the E-less Ni nished surface was not observed by optical microscopes shown in Figures 4b,c, which might indicate the thickness of the alloy was too thin to be observed. In case of the Im-Ag and Im-Sn nished surfaces, vigorous alloy formation was observed, resulting in obvious dierence in surface 10


Page 10 of 26

Page 11 of 26 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


FFC

217.5 mm 37.5 mm 57.5 mm

5

Data points Linear fit

4 3 2 1 0

0

50

100 150 Length [mm]

200

Figure 5: Contact resistance between eGaIn and the ENIG nished electrode. roughness. In case of the OSP-nished surface, notable alloy formation was not observed. Considering smaller and stable contact resistance is preferred not to impede normal function of a circuit, the metal electrodes plated by OSP, E-less Ni, and Im-Sn are not appropriate as a direct interface to eGaIn. Also, vigorous and rapid alloying of Ga with Ag is reported in previous studies, which might lead to uncertain and mechanically sensitive connection. 13 In contrast, contact resistance between eGaIn and the Im-Ag plated surface was stable and sufciently low (similar with the ENIG plated surface) in our results. However, strong adhesion of eGaIn on the Im-Ag plated surface occasionally causes diculty in nishing the writing (cutting the eGaIn trace) on the surface, resulting in an extended eGaIn pillar. As a result, the ENIG nished electrode was selected as the best candidate among the o-the-shelf metal electrodes for direct wiring of eGaIn. Transmission line method was applied to evaluate contact resistance between the eGaIn

11



wire and the ENIG-nished surface using an ENIG-nished FFC. 28 As shown in Figure 5, a clear linear tendency was observed, as the length of the eGaIn wires increased, due to the stable contact resistance between eGaIn and the ENIG-nished surface. The measured contact resistance (half of the y-intercept value) was about 0.2702 Ω (width of the eGaIn wire≈ 300 µm), comparable to typical cm-scale wiring and contacts. 16

Mechanical and Thermal Durability To test durability against mechanical strain, the specimen and the experimental setup were prepared as shown in Figure 6a. As a specimen, the written eGaIn wires connected to an ENIG-nished FFC on the base silicone layer was sealed by pouring a second silicone layer. The Dragon Skin 30 was used for the silicone layers, the boundary surfaces between eGaIn, silicone, and the FFC were lled with an adhesive silicone material for higher durability against leakage. The specimen was stretched as shown in Figure 6b, with strain up to 100% and 150% at the vicinity of the electrode. The average resistance (R) of the 10 eGaIn wires in the specimen were measured under recursive stretching cycles. The normalized resistance change ∆R/R0 is plotted in Figure 6c. In case of the 100 % strain, none of the wires was broken even after 1000 cycles, but at least one wire was broken after tens of cycles with 150 % strain. The leftmost wire of the specimen rst lost its connection due to delamination between the two silicone layers initiated from the interfacial area and resultant burst of the microchannel as shown in Figure 6c. As an evidence for no failure in all the wires, the low peak values of ∆R/R0 are plotted as shown in Figure 6d, which maintained its value very stably (

Direct Wiring of Eutectic Gallium-Indium to a Metal Electrode for Soft

Recommend Documents