Impact of Substrate and Process on the Electrical Performance of

Subscriber access provided by University of Winnipeg Library

Article

The Impact of Substrate and Process on the Electrical Performance of Screen-Printed Nickel Electrodes: Fundamental Mechanism of Ink Film Roughness Bilge Nazli Altay, Jerome S. Jourdan, Vikram Shreeshail Turkani, Hervé Dietsch, Dinesh Maddipatla, A. Pekarovicova, Paul D. Fleming, and Massood Z. Atashbar ACS Appl. Energy Mater., Just Accepted Manuscript • DOI: 10.1021/acsaem.8b01618 • Publication Date (Web): 27 Nov 2018 Downloaded from http://pubs.acs.org on November 30, 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 32 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 Energy Materials

The Impact of Substrate and Process on the Electrical Performance of Screen-Printed Nickel Electrodes: Fundamental Mechanism of Ink Film Roughness Bilge Nazli Altay1*, Jerome Jourdan2, Vikram S. Turkani3, Hervé Dietsch2, Dinesh Maddipatla3, Alexandra Pekarovicova1, Paul D. Fleming1, Massood Atashbar3 1 Department

of Chemical and Paper Engineering Department, Western Michigan University,

Kalamazoo MI 49008 2 BASF

Corporation, Wyandotte, MI 48192

3 Department

of Electrical and Computer Engineering, Western Michigan University,

Kalamazoo, MI 49008 *Corresponding

author: [email protected]

Keywords: nickel nanoparticles, thermoplastic polymer, printed electronics, 4-point probe, screen printing, surface roughness, electrode Graphical Abstract

ACS Paragon Plus Environment

1

ACS Applied Energy Materials 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

Page 2 of 32

ABSTRACT In recent years, traditional printing methods have been integrated to print flexible electronic devices and circuits. Since process requirements for electronics differ from graphic printing, the fundamentals require rediscovery mainly to optimize manufacturing techniques and to find cost reduction methods without compromising functional performance. In addition, alternative inks need to be formulated to increase the variety of functional inks and to pioneer new product developments. In this report, we investigate a thermoplastic-based nickel ink prototype to print electrodes using a screen printing process. Process fundamentals are explored, and cost reduction methods are addressed by studying the effect of substrate roughness, print direction and number of ink layers on the electrical performance of printed nickel. Multilayered electrodes are printed on paper and heat stabilized engineered film. A novel fundamental mechanism is found that explains the effect of substrate roughness on ink film roughness in screen printing, including the roughness measurement of the screen mesh wire that is reported for the first time. Results demonstrated that (i) surface roughness of substrates does not have significant effect on printed ink film roughness in screen printing, (ii) ink film thickness is higher on non-absorbent materials, while line gain is higher on absorbent materials, (iii) the effect of electrode orientation on electrical performance is insignificant (iv) the effect of substrate roughness on the electrical performance for the first print layer can be eliminated by printing multiple layers. The results significantly affect substrate choice and number of ink layers, which are considered as the major cost factors in the manufacturing of printed electronics. 1. INTRODUCTION In recent years, use of traditional printing processes has become a novel method to manufacture cutting-edge thin, flexible electronics in a time and cost-efficient manner. These additive process


2

Page 3 of 32 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


techniques involve continuous deposition of functional materials in a layer-on-layer manner on a thin platform for the fabrication of light weight and portable electronic devices [1-6]. In addition, printing techniques require fewer processing materials, produce significantly less wastage and offer large-area, high volume production in relatively shorter times for the fabrication of green electronics when compared to conventional photolithography and silicon-based techniques [7-12]. Polymer inks containing functional pigments enable fabricating electronics that have shapes predefined by the device layout on flexible substrates, such as papers, films, textiles and metals among others [13-15]. Despite many substrate properties, thermal and mechanical properties are highly critical for the precise registration of stacking functional materials upon each other during printing to create electronic devices [16]. In general, the key factor among the ink, substrate and printing processes that influences the profitability is the substrate [17]. A simplistic description of substrate cost is that the finer/more special they are, the more they cost. Table 1 lists typical cost items for a print manufacturer, indicating substrate is the first single-handed cost item. It is reported that 1% reduction in substrate cost translates to 34% increase in profit [17]. In terms of printed electronics, the pigment used in functional ink formulation, such as precious metals, can also add to the manufacturing cost. Table 1. Typical cost breakdown in printing manufacturing [17] Printing manufacturing cost Factory payroll Substrate Factory expenses Other materials and outside services Administrative expenses Selling expenses Operating profits


% 26 19 16 15 11 11 2

3


Page 4 of 32

Prominent conductive pigments are silver, gold and copper-based [2, 3]. High price is a drawback for silver and gold, while cheaper copper is prone to oxidation and corrosion [18]. Nickel (Ni) on the other hand, an abundant element on earth, has high conductivity and high resistance to oxidation and corrosion [19, 20]. It is a cost-effective material relative to common precious metals and has higher temperature coefficient of resistance to a wider temperature range [21]. Aside from the conductivity, Ni is also ferromagnetic at room temperature [22, 23] which cannot be replaced by the forenamed elements. There is a deep interest in using Ni for the fabrication of temperature detectors, ultrasonic transducers [24], multilayer ceramic capacitors [25], optical and plasmonic antennas [26], interconnectors [19], magnetic sensors [27] and batteries [28]. Wetting properties and surface roughness of substrates, suitable ink rheology characteristics and proper curing are considered to play an important role in the electrical performance of functional ink films [29-32]. Wetting properties of substrates effect ink spreading and leveling behavior during printing, thus effect the homogeneity of the printed ink layer. The degree of wetting is usually determined by surface energy studies that involve measuring contact angles of liquids with known surface tension. Smaller contact angles (below 90°) correspond to high wettability, while larger contact angles (above 90°) to low wettability [33]. Besides wetting, substrate surface characteristics also influence ink leveling during printing. In general, printing inks spread and penetrate into the porous and rougher substrates [34]. Non-porous substrates do not absorb ink, and spreading is influenced by liquid surface tension, viscosity, surface energy and surface roughness [35]. The characteristics of penetration, spreading and leveling of inks effect ink film thickness consequently, thus electrical performance of printed electrodes [36]. General theory in Eq. 1, where ρ and t indicate volume resistivity and thickness, respectively, shows the dependence of sheet resistivity (R) of a material on its thickness:


4

Page 5 of 32 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


𝑅 = 𝜌/𝑡

(1).

Ink rheology plays a crucial role to obtain high quality and stable film thickness. Monitoring viscosity as a function of increased shear rate, increased temperature or simulating printing conditions before the real production reveals propriety information such as ink runnability for the chosen printing process, ink leveling and suitability for the production conditions, therefore eliminating problems regarding printability, material and time loss at early stages [37-40]. In this research, Ni ink rheology flow characteristics are measured in response to increasing shear rate and increasing temperature using a dynamic stress rheometer. The screen printing process was simulated to understand the Ni ink viscosity behavior when it is on a screen mesh, during printing and after printing. Wetting properties were explored by characterizing surface energy of substrates and liquid surface tension by contact angle and pendant drop methods, respectively, using a video system and the Owens-Wendt method [41]. Electrical performance of thermoplastic based screenprinted Ni electrodes was compared after printed on a traditional paperboard that has surface roughness in the micron range and a special heat stabilized film that has roughness in the nano range and corrected based on electrode geometry [42, 43]. To the best of our knowledge, this is the only research reported to date for substrate roughness effect on ink film roughness in screen printing, and for the measurement of surface roughness of the screen mesh wire. The effect of electrode orientation and printing multiple ink layers on the electrical performance was also studied. Surface topographies were characterized with a vertical scanning 3D optical interferometry (VSI). Pigment morphology, elemental analysis and pigment structure were investigated using a scanning electron microscopy (SEM), energy dispersive X-ray spectroscopy (EDS) and X-ray diffractometer, respectively.


5


Page 6 of 32

2. MATERIALS and METHODS 2.1. Materials A prototype thermoplastic based Ni ink (XCUS-78095, PPG Industrial Coatings, OH) with solid content of 75%, and density of 2.6 g/cc was used for screen printing the conductive traces. The substrates were polyethylene terephthalate (PET) (Melinex ST506, DuPont, US), and solid bleached sulphate board (SBS) (144 lb. 10 pt., C2S cover, International Paper, US). Ultra-filtered deionized (DI) water (Fisher Scientific, Fair Lawn, NJ) and diiodomethane (MI) (reagent plus 99%, density 3.325 g/cc; Sigma Aldrich, St. Louis, MO) were newly purchased and used as test liquids for the surface tension and contact angle measurements. A 10 mL syringe (BD Luer Lok, Becton Dickinson, Franklin Lakes, NJ) with a 0.55 mm needle tip (JG20-1.0., Jensen Global, Santa Barbara, CA) were used. 2.2. Ink rheology and screen printing simulation Rheological characterization (steady state flow test, viscosity vs. temperature) of the ink was performed using an AR 2000 dynamic stress rheometer equipped with a 20-mm 2° cone parallel plate geometry (TA Instrument, New Castle, DE). Initially, a steady state flow test was conducted for the Ni ink with shear rates varying from 0.1 to 3000 s-1 at 23°C. Then, viscosity behavior of the ink over varying temperatures was studied using a Peltier plate that was heated from 20 to 60°C at a constant shear rate. Later, TA Rheology Advantage Data Analysis software was used to assess the best rheology model fit based on steady rate versus viscosity. Finally, for the screen printing simulation, the 0.1 s-1 shear rate was considered when ink was on the mesh; 1000 s-1 shear rate to simulate ink behavior during printing and 0.1 s-1 shear rate for ink recovery after printing. The tests were repeated three times to estimate the standard deviations.


6

Page 7 of 32 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


2.3. Surface tension and surface energy estimations An FTA200 (First Ten Angstrom; Portsmouth, VA) video system of 60.9 frame rate (Figure 1) along with FTA32 software were employed to measure the surface tension of Ni ink as well as test liquids using the pendant drop method at ambient conditions. 3 ml of Ni ink was loaded in to the 10-ml syringe and the ink was dispensed at 1 µl/s speed through a 0.55 mm outer diameter needle. Five drops were dispensed and the average surface tension was measured. Similarly, the surface energies of the PET and SBS substrates were obtained by measuring the contact angles (sessile drop profile) of the two test liquids with the FTA 200 system using Owens-Wendt method.

Figure 1. (a) FTA200 video system, (b)surface tension and contact angle measurements, (c) data collection, (d) equilibrium state contact angle analysis 2.4. Electrode fabrication and printed Ni ink film characterization


7


Page 8 of 32

A semiautomatic screen printer (MSP 485, HMI, Lebanon, NJ) with a stainless-steel screen of 325 mesh count, 28 µm wire diameter, 22.5° deflection angle and 12.7 μm MS-22 emulsion thickness was used for Ni ink deposition. Electrodes with 3 mm width and 26 mm length were deposited in three orientations: print direction (squeegee direction), 45°, and cross-print directions at a constant squeegee pressure (Figure 2a). The image in Figure 2b was printed to explore the thinnest line width (50µm, 100µm, 200µm, 500µm, 1mm and 3 mm) that the substrates can hold. The ink layers were printed wet-on-wet, then cured in a Yamato DX 300 convection oven for 5 minutes at 130ºC. Ethylene glycol monobutyl ether acetate (99%, Sigma Aldrich, St. Louis, MO) was used for cleanup. Substrate roughness, ink film roughness and ink film thickness were measured using a Contour GT-K vertical scanning 3D optical white light interferometer microscope (Bruker Corporation, Billerica, MA) with 5x lens at 100 µm back scan, 100 µm length and 5% threshold. Stainless steel mesh was measured with 50x lens at 5 µm back scan, 100 µm length and 5% threshold.

(a)

(b)

Figure 2. (a) Ni electrodes printed at three orientations: print direction (left), 45° (middle) and cross direction (right) (b) Electrodes with different line width 2.5.

SEM, EDS and XRD analysis

The printed Ni electrode surface was imaged using an SEM EVO MA-15 Variable-Pressure Scanning Electron Microscope (Carl Zeiss SMT Ltd., Pleasanton, CA) operating in high vacuum (HV) mode. HV imaging, after Au/Pd coating (1-minute) (Leica EM ACE200 Sputter Coater, Buffalo Grove, IL), provided higher-resolution images in both secondary electron imaging (SEI) and backscattered electron (BSE) imaging modes. In SEI 1° electrons bombarding the sample


8

Page 9 of 32 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


cause 2° electrons to leave the sample at a lower energy that can be deflected by a Faraday cage surrounding the electron detector. When the cage is positively-biased both high- and low-energy electrons are captured and a bright, highly-resolved image is generated. In BSE imaging the Faraday cage is negatively-biased, causing low-energy 2° electrons to be deflected and allowing only high-energy 1° electrons to be detected; this generates images based on atomic number (Z) differences with higher-Z elements appearing brighter. Energy dispersive X-ray spectroscopy (EDS) was performed using dual Bruker X-flash 6/30 X-ray Spectrometers. EDS makes use of Xrays generated by a sample as electrons from the SEM are bombarding it and provide qualitative and semi quantitative identification of the sample elemental composition. Structural characterization of printed Ni electrodes was carried out using a Philipps XRG 3100 diffractometer operating at 25 kV and 10 mA with Cu Ka radiation. 2.6.

Electrical measurements

3 mm printed Ni electrodes were used for the electrical performance study with a Keithley 2400 4-point probe sourcemeter. The Haldor Topsøe geometrical factor method was used for correction based on electrode geometry (ohms per square, Ω/□) [42, 43]. The results were quantified based on sheet resistivity and volume resistivity. Ink film thickness gathered from vertical scanning interferometry was used to estimate volume resistivity ( Ω m) of the printed Ni electrodes by multiplying by ink film thickness. 2.7.

Statistical analysis

The null hypotheses for the study were whether the statistical mean of electrical performance of printed Ni electrodes on SBS with respect to orientation, number of printed ink layers and substrate roughness were significantly different from PET, causing functional irregularities when printing and curing conditions were kept constant. The p-values gathered using the analysis of variance


9


Page 10 of 32

(ANOVA) statistical method with JMP 14 software. The significance level was set to α =0.05 (95%). The null hypotheses were rejected when p-value was less than 0.05. 3. RESULTS and DISCUSSION A very distinct rheology behavior is expected from screen printing inks due to the nature of the screen printing process. At low shear rates, a high viscosity is required for ink to stay stable on the printing plate before printing, however a very low viscosity is needed temporarily upon shear during printing, followed by recovery to a high viscosity after printing [37-39]. The experimental Ni ink yielded the desirable shear thinning behavior, high viscosity of 100 Pas at low shear rate, and 1 Pas at high shear rate as shown in Figure 3a. The low viscosity upon shear is very critical in screen printing to assure transferring patterns from plate to substrates. Critical material properties can be gathered from the viscosity versus shear rate flow by using rheology models [40]. The model helps to track the effect of ingredient changes in ink formulation or the ink behavior in different processing methods, such as printing or roller coating. Consistency value provided by the model is an intrinsic property of materials, which shows the time constant of a flow model. The reciprocal of the consistency provides a shear rate that shows the onset shear rate for the shear thinning. Using the Cross rheology model, Ni ink consistency was calculated as 0.0096 s. For the temperature ramp analysis, the shear rate of 104 s-1 (1/0.0096 s) was kept constant while increasing the temperature to explore viscosity behavior upon heating. From Figure 3b, it is observed that the viscosity of Ni ink decreased from 25 Pas to 12 Pas when the temperature was varied from 20°C to 60°C. The same viscosity decrease was observed in the viscosity versus shear rate data (Figure 1a) when shear rate changed from ~100 s-1 to 250 s-1. Since the change in viscosity corresponds to the same order of magnitude shear rate, Ni ink would be considered staying stable at varying temperatures in production cycle as well as storage conditions. In Figure 3c, the screen printing


10

Page 11 of 32

process is depicted to explore Ni ink printability and the ink recovery. The viscosity is ~100 Pas at 0.1 s-1, roughly corresponding to the ink at rest on the screen plate mesh. When high shear of 1000 s-1 is applied (with a squeegee during screen printing), viscosity decreases significantly to 1 Pas, thus ink can flow through the screen mesh resulting in ink transferring to the substrate. Once the printing is done, the viscosity recovers to ~90 Pas. The ability of ink to have lower viscosity temporarily upon shear, then recovering to its original state when shear removed, called thixotropy, is a preferred property of screen inks for the ease of processability [39]. (a)

(b)

120 100

30 25

Viscosity (Pa.s)

Viscosity (Pa.s)

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


80 60 40

20 15 10 5

20

0

0 0.1

1

10 100 Shear rate (1/s)

1000

10000

20

30

40 50 Temperature (ºC)

60

(c)


11


Page 12 of 32

Figure 3. (a) Ni ink viscosity as a function of shear rate at room temperature, (b) Ni ink viscosity as a function of temperature change at shear rate of 104 s-1, (c) Ni ink viscosity during screen printing simulation: screen on mesh at shear rate of 0.1 s-1, ink transfer shear rate of 1000 s-1, ink recovery at shear rate of 0.1 s-1.

(a) (b) Figure 4. (a) 3D profilometry image of PET film surface: 17 nm, (b) 3D profilometry image of SBS paperboard surface: 1009 nm. The roughness characterization of the PET film and SBS paperboard is shown in Figure 4a, b. The 3D optical imaging verifies that the PET film surface is significantly smoother than the surface of SBS paperboard. The measured root mean squared surface roughness is 17±2 nm for the PET and 1009±10 nm for the SBS, representing the PET is smoother than SBS paperboard by two orders of magnitude. The SBS surface is rougher due to the cellulosic fibers and filler content, and is presumed to be more hydrophilic relative to the more hydrophobic surface of PET. To determine this, pendant drop and contact angle methods were used to estimate surface energy values of the substrates. First, the surface tensions of DI water and MI were measured. Table 2 shows that the


12

Page 13 of 32 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


tensions of test liquids are slightly different from their literature values [44]. This could be due to the following factors: (a) variation in the purity levels of the chemical, (b) prolonged time periods of chemicals remaining stagnant in the laboratory, (c) contamination of chemicals and (d) degradation of principle materials in the chemicals. Based on the contact angle data (Table 3) and the Owens-Wendt method [41], surface energies of PET and SBS are found to be 45.97±2.66 and 40.28±3.40 mN/m, respectively. The reputable criterion for an ink to spread well on a substrate is that the substrate surface energy must be higher than the surface tension of the ink [45]. Ni ink surface tension was measured as 15.51 mN/m, therefore a good ink spreading on substrates was expected. Table 2. Surface tension of liquids Test liquids DI MI Ni ink

Surface tension (mN/m) 71.36 (±0.87) 48.03 (±0.60) 15.51 (±0.28)

Dispersive 21.80 45.73

Polar 49.56 2.33

Table 3. Contact angle of liquids on substrates Test fluids

DI

MI

PET film

69.02° (±0.07)

23.48° (±0.34)

SBS board

74.73° (±1.30)

37.06° (±0.41)

Besides ink spreading, connectivity in printed lines is also essential for electrodes to function properly. Figure 5 and 6 show the line connectivity in the microscopy images of thermally cured printed electrodes on PET and SBS, respectively. After 1, 2 and 4 print passes, PET film fails to hold the 50 µm thinnest line, while it is visible on the SBS paperboard after only 1 print pass. The print quality difference suggests that the surface energy differences between the substrates could be the reason, leading to different packing densities of Ni particles during printing. Ni ink more freely wets the SBS surface, thus particle packing would be less compared to ink on the PET.


13


Page 14 of 32

500µm

Figure 5. Printed Ni electrodes on PET film. 1 pass (left), 2 passes (middle) and 4 passes (right)

Figure 6. Ni electrodes on SBS paperboard. 1 pass (left), 2 passes (middle) and 4 passes (right) The morphology of printed and cured Ni pigment at various magnifications was captured using a scanning electron microscope. Unit particles were observed (Figure 7) as agglomerates with a of 1 – 5 µm in size after the curing process. The morphology is varied from spherical to flat plates and cubes that form cubic-to-spherical agglomerates. Similar morphologies observed by other researches are suggesting cubic morphology for the Ni ink [46, 47]. EDS analysis showed four main peaks belonging to Ni, carbon (C), gold (Au) and palladium (Pd) (Figure 8). Ni and C are associated with the electrodes and substrate respectively, while Au and Pd signals are from the sputter coating involved in the procedure of SEM sample preparation. The printed electrodes showed no significant contamination from inorganic or organic sources.


14

Page 15 of 32 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 7. SEM images of printed Ni ink: Magnified (a) 10K X, (b) 5K X

Figure 8. Energy dispersive X-ray spectroscopy (EDS) Figure 9 shows the X-ray diffractogram of the cured printed Ni electrodes cured (annealed). The measurements were performed for Bragg angle (2Ɵ) ranging from 40° to 80°. The step size and the dwell time of the measurements were 0.1° and 2 s, respectively. Bragg reflections were observed corresponding to (111), (200), and (220) planes of a typical face-centered cubic (fcc) Ni [47, 48]. No other phase was observed in the diffractogram indicating that printed Ni particles had single phased cubic structure, as observed in the SEM. The crystallite size (D) of the printed Ni was calculated to be 22 nm using Scherrer’s formula [50] shown in Eq. (2).


15


(2)

0.9λ

𝐷 = βcosθ

where, λ is the X ray wavelength, θ is the Bragg diffraction angle and β is the full width at half maximum (FWHM) of the largest diffraction intensity peak. Finally, it is important to understand that the size obtained from Scherrer’s formula is the apparent crystalline size and is not necessarily the same as the particle size in SEM since several crystals can agglomerate are considered to form particles [51]. The shapes of the agglomerates do not necessarily correspond to the crystal symmetry. The cubic morphology from the X-ray diffraction is unambiguous regarding the underlying cubic morphology. [111] Intensity (arbitrary units)

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

Page 16 of 32

[200] [220]

40

45

50

55

60 2Ɵ (degree)

65

70

75

80

Figure 9. X-ray Diffractogram of printed Ni electrode Resistivity measurement is one of the main characterization methods for electrodes [52]. Here we used both sheet resistivity and volume resistivity for comparison by incorporating ink film thicknesses gathered from the vertical scanning interferometry (Figure 10a). Fundamentally, thinner ink layers show higher resistance relative to thicker ink layers and the lowest resistance is preferable for applications where conductivity is needed. Figure 11 shows that 1, 2 and 4 print passes on PET resulted in 28.0±0.2 µm, 30.3±0.4 µm and 33.7±0.1 µm ink film thickness and 19.8±1.1 µm, 22.0±2.1 µm and 24.8±1.1 µm on SBS, respectively. This is expected since Ni ink


16

Page 17 of 32 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


more freely wets the SBS surface, and absorbs into the SBS structure due to permeability. Previous studies determined that ink film thickness can be decreased by increasing the squeegee pressure depending on the binder content or by diluting the functional ink [40, 53]. In this study, the thickness difference is attributed to the absorbency of the substrate since the squeegee pressure was kept constant. Ni ink particles pack more densely on PET relative to SBS, causing higher ink film thickness due to the non-absorbent nature of PET film. Ink thickness difference between the substrates was found significant, (1-layer: p=.0002, 2-layers: p=.0026, 4-layers: p=.0001). However, line gain was higher on SBS due to its rougher and more porous surface (Figure 12). Although line gain and ink film thicknesses varied significantly between the PET and SBS ink layers, the difference in the ink film roughness is found to be insignificant (Figure 11) (1-layer: p=.1539, 2-layers: p=.1340, 4-layers: p=.9640). Ink roughness was measured as 2.58±0.03 µm, 2.54±0.06 µm and 2.46±0.06 µm on PET, and 2.87±0.28 µm, 2.72±0.16 µm and 2.47±0.15 µm on SBS for 1, 2 and 4 print passes, respectively. Considering the significant difference between the roughness of the PET and SBS, it can be said that substrate roughness has no significant effect on ink film roughness in screen printing. The major contributor to the ink film roughness is found to be the screen printing plate, stainless steel mesh surface, since it is in one-to-one contact with the ink surface during printing. Figure 10b verifies that the average of measured mesh roughness is 2.96±0.92 µm. The slight difference between the screen mesh roughness and Ni ink film roughness is attributed to the ink leveling upon the absence of shearing force after printing and drying consistent with the rheology results (Figure 3) [40].


17


Page 18 of 32

a)

b) Figure 10. (a) Sample of 3D profilometry image of ink film thickness, (b) Surface roughness of stainless steel mesh: 2.96 µm


18

PET, µm PET, Ra

SBS, µm SBS, Ra

10

Ni on PET

40

8

30

6

20

4

10

2

0

0 1 pass

2 passes

4

4 passes

Figure 11. Ink film thickness and ink film roughness at different print passes

Printed line gain (%)

50

Ink film thickness (µm)

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


Ink film roughness - Ra (µm)

Page 19 of 32

Ni on SBS

3

2

1

0 1 pass

2 passes

4 passes

Figure 12. Line gain at different print passes

The effect of electrode orientation on volume resistivity on both substrates was found insignificant (SBS: p=.7826, PET: p=.8729) (Figure 13). Therefore, results were only compared as a function of number of print passes. Sheet resistivity is the term commonly used in industry and can be tailored by changing ink film thickness. The sheet resistivity of printed Ni electrodes on both substrates reduced as the ink film thickness increased, although it showed a slight increase at 4layers (Figure 14). The reason would be the multiple print passes that was performed using weton-wet method, which does not include curing after each print pass. Since curing time was kept constant at the experimental design for all the print passes, it caused thicker ink layers to be cured improperly, and most likely resulted in increase in the apparent sheet resistivity for the 4-layer ink film. With 1-layer Ni electrode, 6.6 Ω/□ sheet resistivity was obtained on PET, while 14.5 Ω/□ on SBS (p=.0004). When multiple ink layers were printed, the difference became insignificant. 5.5 and 6.6 Ω/□ were obtained for 2-layers and 11.3 and 10.6 Ω/□ were obtained for 4-layers on PET and SBS, respectively (2-layers: p=.1305, 4-layers: p=.7793). The resistivity difference for 1layers can be attributed to the subjacent substrate surface roughness. However, the first ink film roughness takes over the influence for the following ink layers, ensuring insignificance difference in sheet resistivity (Figure 15).


19


8 6

50

layer - PET layer - PET layer - PET layer - SBS layer - SBS layer - SBS

4 2

PET, µm PET, Ω/□

SBS, µm SBS, Ω/□

40

15

30 10 20 5

10 0

0 Cross direction

Figure 13. Volume resistivity as a function of electrode orientation

2 passes

4 passes

Figure 14. Ink film thickness in relation to sheet resistivity

100

Surface Roughness (µm)

0 1 pass

Print direction

PET, SBS, PET, SBS,

10

µm µm Ω/□ Ω/□

21 16

1 11 0.1

6

0.01

Sheet resistance (Ω/□)

45 degree

20

Sheet resistance (Ω/□)

1 2 4 1 2 4

Ink film thickness (µm)

10 Volume resistivity (x10-4 Ω·m)

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

Page 20 of 32

1 Substrate

1 pass

2 passes 4 passes

Figure 15. Surface roughness in relation to sheet resistivity 4. CONCLUSIONS A thermoplastic based nickel conductive ink has been screen-printed to fabricate electrodes. The electrical performance, ink film roughness and ink film thickness have been studied by increasing the number of printed layers and by changing the electrode orientation on two distinctly different substrates, PET film and SBS paperboard. We have shown the definite benefit of choosing a rougher paper


20

Page 21 of 32 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


substrate over a high smoothness engineered film substrate to provide similar electrical performance. Considering the cost of SBS paperboard being 35x less expensive relative to PET film, significant reduction in the manufacturing cost can be achieved for many applications just by altering substrate choice. The

homogenous

electrical

response

is

systematically

achieved

that

is

independent of substrate and print direction (electrode orientation). The roughness of the PET film is found to be 60x smoother than the SBS paperboard, although only the first printed layer showed slight difference in sheet resistivity that can be attributed to the substrate roughness, yet no difference was observed on the ink film roughness. The effect on ink film roughness in screen printing was found dictated by the roughness of screen plate mesh wire, which was quantified for the first time. The lowest sheet resistivity has been found 5.5 Ω/□ on PET and 6.6 Ω/□ on SBS paperboard when two layer of Ni ink were printed. The electrical properties of the printed layers were in the same order of magnitude, confirming compatible electrical performance between the substrates. The highest ink film thickness of 32 µm has been found on the PET film, while the lowest thickness of 20 µm was on SBS paperboard. Higher


21


Page 22 of 32

ink film thickness means more ink usage during production that adds to the cost of electronic fabrication. AUTHOR INFORMATION Corresponding Author *Bilge Nazli Altay Chemical and Paper Engineering Department, A-217, Western Michigan University, Kalamazoo, MI 49008-5462 Author Contributions The authors declare that they have no competing interests. All authors have given approval to the final version of the manuscript. Funding Sources This research was funded by FTA, Rossini North America Research Scholarship. (the word “partially” was deleted, as there is no other funder. With their help, we were able to purchase some of the chemicals during experimental trials) ACKNOWLEDGMENT


22

Page 23 of 32 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


The authors would like to thank Dr. Paul Cheetham and PPG Electronic Materials Group for providing the experimental nickel ink; Dr. Md Arifur Rahman, Nicholas Branham and Joseph Lengyel at BASF Corporation for their insightful discussion and help during experiments; and Dr. James Atkinson from LTI Printing, Karen Cumbie and Randall Linville from Westrock and Adam Young from Tekra Corp. for their help determining the cost of substrates.


23


Page 24 of 32

REFERENCES [1]

Leenen, M. A. M.; Arning, V.; Thiem, H.; Steiger, J.; Anselmann, R. Printable electronics: Flexibility for the future. Physica Status Solidi (A) Applications and Materials Science 2009, 206, 588-597.

[2]

Gamato, D. R.; Brazis, P.; Kalyanasundaram, K.; Zhang, J. (Eds.). Printed organic and molecular electronics 2004, Norwell, MA: Kluwer Academic Publishers.

[3]

Altay, B. N. Smart ink for Flexo Rossini scholarship winner examines printed electronics. Flexo 2016, 41, 26-32.

[4]

Kim, K. S.; Zhao, Y.; Jang, H.; Lee, S. Y.; Kim, J. M.; Kim, K. S.; Ahn, J.-H.; Kim, P.; Choi, J.-Y.; Hong, B. H. Large-scale pattern growth of graphene films for stretchable transparent electrodes. Nature 2009, 457, 706-710.

[5]

Bae, S.; Kim, H.; Lee, Y.; Xu, X.; Park, J.-S.; Zheng, Y.; Balakrishnan, J.; Lei, T.; Kim, H. R.; Song, Y.; Kim, Y.-J.; Kim, K. S.; Özyilmaz, B.; Ahn, J.-H.; Hong, B. H.; Iijima, S. Roll-to-roll production of 30-inch graphene films for transparent electrodes. Nature Nanotechnology 2010, 5, 574-578.

[6]

Arias, A. C.; MacKenzie, J. D.; McCulloch, I.; Rivnay, J.; Salleo, A. Materials and applications for large area electronics: solution-based approaches. Chem. Rev. 2010, 110, 3-24.

[7]

Eshkeiti, A.; Reddy, A. S. G.; Emamian, S.; Narakathu, B. B.; Joyce, M. J.; Joyce, M. K.; Fleming, P. D.; Bazuin, B. J.; Atashbar, M. Z. Screen printing of multilayered hybrid


24

Page 25 of 32 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


printed circuit boards on different substrates. IEEE Transactions on Components Packaging and Manufacturing Technology 2015, 5, 415-421. [8]

Zhu, H.; Narakathu, B. B.; Fang, Z.; Aijazi, A. T.; Joyce, M. K.; Atashbar, M.; Hu, L. A gravure printed antenna on shape-stable transparent nanopaper. Nanoscale 2014, 6, 91109115.

[9]

Kunnari, E.; Valkama, J.; Keskinen, M.; Mansikkamäki, P. Environmental evaluation of new technology: printed electronics case study. J. of Cleaner Production 2009, 17, 791-799.

[10] Faddoul, R.; Reverdy-Bruas, N.; Blayo, A.; Haas, T.; Zeilmann, C. Optimisation of silver paste for flexography printing on LTCC substrate. Microelectronics Reliability 2012, 52, 1483-1491. [11] Turkani, V. S.; Maddipatla, D.; Narakathu, B. B.; Bazuin, B. J.; Atashbar, M. Z. A Carbon Nanotube Based NTC Thermistor using Additive Print Manufacturing Processes. Sensors and Actuators A 2018, 279, 1-9. [12] Liu, Y.; Pharr, M.; Salvatore, G. A. Lab-on-skin: A review of flexible and stretchable electronics for wearable health monitoring. ACS Nano 2017, 11, 9614-9635. [13] Correia, V.; Mitra, K. Y.; Castro, H.; Rocha, J. G.; Sowade, E.; Baumann, R. R.; LancerosMende, S. Design and fabrication of multilayer inkjet-printed passive components for printed electronics circuit development. Journal of Manufacturing Processes 2018, 31, 364371.


25


Page 26 of 32

[14] MacDonald, W. Engineered films for display technologies. Journal of Materials Chemistry 2014, 14, 4-10. [15] Avuthu, S. G. R.; Wabeke, J. T.; Narakathu, B. B.; Maddipatla, D.; Arachchilage, J. S.; Obare, S. O.; Atashbar, M. Z. A Screen Printed Phenanthroline-Based Flexible Electrochemical Sensor for Selective Detection of Toxic Heavy Metal Ions. IEEE Sensors Journal 2016, 16, 8678–8684. [16] Bolduc, M.; Trudeau, C.; Beaupré, P.; Cloutier, S. G.; Galarneau, P. Thermal Dynamics Effects using Pulse-Shaping Laser Sintering of Printed Silver Inks. Scientific Reports 2018, 8. [17] Davis, R. H. The impact of cost on profitability in the printing industry. Printing Industry of America 2017. [18] Määttänen, A.; Ihalainen, P.; Pulkkinen, P.; Wang, S.; Tenhu, H.; Peltonen, J. Inkjetprinted gold electrodes on paper: Characterization and functionalization. ACS Applied Materials and Interfaces 2012, 4, 955-964. [19] Wells, R. C. Relative Abundance of Nickel in The Earth's Crust. United States Department of the Interior 1943. [20] Lockett, V.; Hartman, A.; Gustafson, J.; Lowenthal, M.; Ray, W. Nickel inks and oxidation resistant and conductive coatings. Patent No. US20140183421, 2014. [21] Kasanami, T.; Tani, H.; Fujiwara, S. Temperature sensor. Patent No. 5,242,225, 1993.


26

Page 27 of 32 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


[22] Campos, C. D. M.; Flacker, A.; Vaz, A. R.; Moshkalev, S. A.; Nobrega, E. G. O. Electroplated nickel permanent magnetic films over polymeric membranes. J. Electrochem. Soc. 2011, 158, D330-D334. [23] Dlubak, B.; Martin, M-B.; Weatherup, R.S.; Yang, H.; Deranlot, C.; Blume, R.; Schloegl, R.; Fert, A.; Anane, A.; Hofmann, S.; Seneor, P.; Robertson. J. Graphene-passivated nickel as an oxidation-resistant electrode for spintronics. ACS Nano 2012, 6, 10930-10934. [24] Crescini, D.; Marioli, D.; Taroni, A. Thick-film ultrasonic transducer with integrated temperature detector. Sensors and Materials 2000, 12, 15-24. [25] Park, S.; Kim, H. Flash light sintering of nickel nanoparticles for printed electronics. Journal of Thin Solid Films 2014, 550, 575-581. [26] Chen, J.; Albella, P.; Pirzadeh, Z.; Alonso-González, P.; Huth, F.; Bonetti, S.; Bonanni, V.; Åkerman, J.; Nogués, J.; Vavassori, P.; Dmitriev, A.; Aizpurua, J.; Hillenbrand, R. Plasmonic nickel nano-antennas. Small 2011, 7, 2341-2347. [27] Morten, B.; Prudenziati, M.; De Cicco, G.; Bianco, A.; Montesperelli, G.; Gusmano, G. Thick-film magnetoresistors and related sensors. Measurement Science and Technology 1997, 8, 21-28. [28] Young, K.; Wang, L.; Mays, W.; Reichman, B.; Chao-Ian, H.; Wong, D.; Nei, J. Nickel hydroxide positive electrode for alkaline rechargeable battery. Patent No. US9,935,315 B2, 2018.


27


Page 28 of 32

[29] Šandrej, M.; Mikula, M.; Dvonka, V. Properties of Conductive Features Printed on Papers and Foils. Acta Chimica Slovaca 2011, 4, 59 – 67. [30] Timoshevskii, V. The influence of surface roughness on electrical conductance of thin Cu films: An ab initio study. Journal of Applied Physics 2008, 103, 113705. [31] Rossnagel, S. M.; Kuan, T. S. Alteration of Cu conductivity in the size effect regime. Journal of Vacuum Science & Technology B: Microelectronics and Nanometer Structures Processing, Measurement, and Phenomena 2004, 22, 240. [32] Žveglič, M.; Hauptman, N.; Gunde, M. K. Screen-printed electrically conductive functionalities in paper substrates. Materials and Technology 2011, 45, 627-632. [33] Yuan, Y.; Lee, T. R. Contact angle and wetting properties. G. Bracco, B. Holst (eds.), Surface Science Techniques, Springer Series in Surface Sciences 2013, 51. [34] Clarke, A.; Blake, T. D.; Carruthers, K.; Woodward, A. Spreading and Imbibition of Liquid Droplets on Porous Surfaces. Langmuir 2002, 18, 2980–2984. [35] Jung, S.; Hutchings, I. M. The impact and spreading of a small liquid drop on a non-porous substrate over an extended time scale. Soft Matter 2012, 8, 2686. [36] Li, M. –Y.; Yang, M.; Vargas, E.; Neff, K.; Vanli, A.; Liang, R. Analysis of variance on thick-ness and electrical conductivity of carbon nanotube thin films. Measurement Science and Technology 2016, 27, 9. [37] Woo, K.; Jang, D.; Kim, Y.; Moon, J. Relationship between printability and rheological behavior of ink-jet conductive inks. Ceramics International 2013, 39, 7015-7021.


28

Page 29 of 32 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


[38] Hemmati, S.; Barkey, D. P.; Gupta, N. Rheological behavior of silver nanowire conductive inks during screen printing. J. Nanopart. Res. 2016, 18, 249. [39] Somalu, M. R.; Brandon, N. P. Rheological studies of nickel/scandia-stabilized-zirconia screen printing inks for solid oxide fuel cell anode fabrication. J. of the American Ceramic Society 2012, 95, 1220-1228. [40] Phair, J. W.; Lundberg, M.; Kaiser, A. Leveling and thixotropic characteristics of concentrated zirconia inks for screen printing. Rheologica Acta 2009, 48, 121-133. [41] Owens, D. K.; Wendt, R. C. Estimation of the Surface Free Energy of Polymers. J. of Applied Polymer Science 1969, 13, 1741-1747. [42] Siegfried E. A new parameter based on graphene for characterizing transparent, conductive materials. Carbon 2009, 47, 2936-2939. [43] Vuorinen, T.; Zakrzewski, M.; Rajala, S.; Lupo, D.; Vanhala, J.; Palovuori, K.; Tuukkanen, S. Printable, transparent, and flexible touch panels working in sunlight and moist environments. Advanced Functional Materials 2014, 24, 6340-6347. [44] Fowkes, F. Attractive forces at interfaces. Industrial & Engineering Chemistry 1964, 56, 40-52. [45] Hrehorova, E.; Rebros, M.; Pekarovicova, A.; Fleming, P. D.; Bliznyuk, V.N. Characterization of conductive polymer inks based on PEDOT:PSS. TAGA Proceedings 2008, 219-231.


29


Page 30 of 32

[46] LaGrow, A. P.; Ingham, B.; Cheong, S.; Williams, G. V. M.; Dotzler, C.; Toney, M. F.; Jefferson, D. A.; Corbos, E. C.; Bishop, P. T.; Cookson, J.; Tilley, R. D. Synthesis, Alignment, and Magnetic Properties of Monodisperse Nickel Nanocubes. Journal of American Chemical Society 2012, 134, 855−858. [47] Chen, H.; Xu, C.; Chen, C.; Zhao, G.; Liu, Y. Flower-like hierarchical nickel microstructures: Facile synthesis, growth mechanism, and their magnetic properties. Materials Research Bulletin 2012, 47, 1839-1844. [48] Hemalatha, M.; Suriyanarayanan, N.; Prabahar, S. Nanoscale synthesis and optical features of nickel nanoparticles. Optik-International Journal for Light and Electron Optics 2014, 125, 1962-1966. [49] Jian, X.; Zhou, Z.; Wu, S.; Chen, L.; Zeng, Q.; Wang, C. Controllable preparation of Ni nanoparticles for catalysis of coiled carbon fibers growth. Nanoscale research letters 2014, 9, 370. [50] Rahman, M. T.; McCloy, J.; Ramana, C. V.; Panat, R. Structure, electrical characteristics, and high-temperature stability of aerosol jet printed silver nanoparticle films. Journal of Applied Physics 2016, 120, 075305. [51] Jensen, H.; Pedersen, J. H.; Jørgensen, J. E.; Pedersen, J. S.; Joensen, K. D.; Iversen, S. B.; Gaard, E. G. Determination of size distributions in nanosized powders by TEM, XRD, and SAXS. Journal of Experimental Nanoscience 2006, 1, 355-373. [52] Koon, D. W.; Knickerbocker, C. J. What do you measure when you measure resistivity?. Review of Scientific Instruments 1992, 63, 207-210.


30

Page 31 of 32 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


[53] Cao, X.; Chen, H.; Gu, X.; Liu, B.; Wang, W.; Cao, Y.; Wu, F.; Zhou, C. Screen printing as a scalable and low-cost approach for rigid and flexible thin-film transistors using separated carbon nanotubes. ACS Nano 2014, 8, 12769-12776.


31



Page 32 of 32

Impact of Substrate and Process on the Electrical Performance of

Recommend Documents