Lightweight, Flexible, High-Performance Carbon ... - ACS Publications

Jan 21, 2016 - Richard E. Smalley Institute for Nanoscale Science and Technology, Rice University, Houston, Texas 77005, United States. §. Communicat...
3 downloads 0 Views 1MB Size
Subscriber access provided by GAZI UNIV

Article

Lightweight, flexible, high-performance carbon nanotube cables made by scalable flow coating Francesca Mirri, Nathan Orloff, Aaron Michael Forster, Rana Ashkar, Robert James Headrick, E.Amram Bengio, Christian Long, April Choi, Yimin Luo, Angela R. Hight Walker, Paul Butler, Kalman B Migler, and Matteo Pasquali ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.5b11600 • Publication Date (Web): 21 Jan 2016 Downloaded from http://pubs.acs.org on January 26, 2016

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 free 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 accessible to all readers and 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.

ACS Applied Materials & Interfaces 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 28

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

Lightweight, flexible, high-performance carbon nanotube cables made by scalable flow coating** Francesca Mirri,1,2 Nathan D. Orloff,3,4 Aaron M. Forster,5 Rana Ashkar,6,7,8 Robert J. Headrick,2,9 E. Amram Bengio,1,2 Christian J. Long,10,11 April Choi,1 Yimin Luo,1 Angela R. Hight Walker,12 Paul Butler,6 Kalman B. Migler,4 Matteo Pasquali1,2,9* 1

Department of Chemical and Biomolecular Engineering, Rice University, Houston, Texas 77005, USA 2 Richard E. Smalley Institute for Nanoscale Science and Technology, Rice University, Houston, Texas 77005, USA 3 Communications Technology Laboratory, National Institute of Standard and Technology, Boulder, Colorado 80305, USA 4 Material Science and Engineering Division, National Institute of Standard and Technology, Gaithersburg, Maryland 20899, USA 5 Materials and Structural System Division, National Institute of Standard and Technology, Gaithersburg, Maryland 20899, USA 6 NIST Center for Neutron Research, National Institute of Standard and Technology, Gaithersburg, Maryland 20899, USA 7 Materials Science and Engineering Department, University of Maryland, College Park, Maryland 20742, USA 8 Biology and Soft Matter Division, Oak Ridge National Laboratory, Oak Ridge, TN, USA 37831 9 Department of Chemistry, Rice University, Houston, Texas 77005, USA 10 Center for Nanoscale Science and Technology, National Institute of Standard and Technology, Gaithersburg, Maryland 20899, USA 11 Maryland Nanocenter, University of Maryland, College Park, Maryland 20742, USA 12 Physical Measurement Laboratory, National Institute of Standard and Technology (NIST), Gaithersburg, Maryland 20899, USA

*Address correspondence to [email protected] **Partial contribution of the United States Government; not subject to copyright in the United States. Description of commercial products herein is for information only; it does not imply recommendation or endorsement by United States Government.

Keywords: carbon nanotubes, coaxial cables, dip-coating, attenuation, rheology.

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

Page 2 of 28

Abstract Coaxial cables for data transmission are ubiquitous in telecommunications, aerospace, automotive, and robotics industries. Yet, the metals used to make commercial cables are unsuitably heavy and stiff. These undesirable traits are particularly problematic in aerospace applications, where weight is at a premium and flexibility is necessary to conform with the distributed layout of electronic components in satellites and aircraft. The cable outer conductor (OC) is usually the heaviest component of modern data cables; therefore, exchanging the conventional metallic OC for lower weight materials with comparable transmission characteristics is highly desirable. Carbon nanotubes (CNTs) have been recently proposed to replace the metal components in coaxial cables; however, signal attenuation was too high in prototypes produced so far. Here, we fabricate the OC of coaxial data cables by directly coating a solution of CNTs in chlorosulfonic acid (CSA) onto the cable inner dielectric. This coating has an electrical conductivity that is approximately twoorders-of-magnitude greater than the best CNT OC reported in the literature to date. This high conductivity makes CNT coaxial cables an attractive alternative to commercial cables with a metal (tin-coated copper) OC, providing comparable cable attenuation and mechanical durability with a 97 % lower component mass.

2 ACS Paragon Plus Environment

Page 3 of 28

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

Introduction Coaxial cables are indispensable in modern technology and have a wide range of uses that span from navigation to telecommunication systems.1 Coaxial cables are used to propagate signals and require a center conductor (CC) and an outer conductor (OC) separated by an insulating layer (dielectric). The OC serves two functions: signal transmission as well as electromagnetic (EM) shielding to minimize EM interference.2 While EM shielding does not require high OC conductivity,3 signal loss (signal attenuation) through the transmission line is highly sensitive to the conductivity and architecture of the OC. Metals are generally used because of their high conductivity; however, their high density and limited fatigue resistance require braided shielding architectures and wire oversizing to meet mechanical specifications and reach appropriate OC conductance. Replacing the metals in conventional coaxial cables with lighter, fatigue-resistant materials has been a holy grail particularly in aerospace applications where weight reduction directly affects launch cost, payload, and fuel efficiency. In commercial and military aircrafts, as well as satellites and spacecrafts, this can lead to improved travel range, mission time, and reduced emissions.4 Size and weight reduction becomes even more compelling in the case of aircraft with limited space such as unmanned aerial vehicles (UAVs). Despite more than three decades of research, composite core-skin metal-polymer constructs (such as metallized PPTA (Aracon) and PBO (Amberstrand)) provide only minor advances and have limited operating ranges because of issues such as delamination.5 In principle, carbon nanotubes (CNTs) are an alluring alternative to conventional conductors used in coaxial data cables because they combine high strength, electrical and thermal conductivity with low density,6-9 which makes them ideal for applications where weight saving is a determining factor.4, 10 However, achieving electrical performance comparable to metals has been challenging.

3 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

Page 4 of 28

In recent work, CNT sheets were wrapped around the cable dielectric to form the OC.11, 12Although the CNT sheet conductivity was high enough to provide EM shielding comparable to commercial metal OCs,12 the resulting cables did not meet the military standards for signal attenuation due to insufficient CNT sheet conductivity. In a follow-up paper, Jarosz et al.13 were able to meet signal attenuation military standards in short (8 cm including terminations) cables by wrapping the cable dielectric with vacuum filtered CNT buckypaper and overlaying palladium foil for ~ 2/3 of the cable length. This procedure left only ~ 2 cm of CNT OC exposed, too short to draw any conclusions on the effectiveness of the CNT OC versus the efects of the palladium foil and terminations. Moreover, despite the fact that tape-wrapped dielectric insulation has been in common use for nearly two centuries, it has largely been supplanted for specific commercial applications over the last few decades by extruded dielectric insulation. This trend towards extruded dielectric insulation has been driven by high scalability, low cost, and easy deployment for producing long cables.14 Here we show that solution-coated CNT OCs combine high electrical functionality, mechanical flexibility, and scalable manufacturing. Specifically, we show that a coaxial cable consisting of an inner copper conductor and seamless CNT OC meets data transmission requirements while providing a 97 % mass saving over its metal counterpart.

Results and discussion Cable fabrication The CNT coaxial cables were fabricated from RG-174/U coax (Figure 1a) and subsequently attached to female SubMiniature version A (SMA) connectors (Figure S1 in Supporting 4 ACS Paragon Plus Environment

Page 5 of 28

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

Information). In both CNT and commercial cables a copper wire serves as the CC and is enclosed by a polyethylene (PE) dielectric (Figure 1b). The dielectric is covered by the OC, which is a metal braid for the commercial cables and a CNT layer in CNT cables (Figure 1b). Both cables are insulated by nominally identical PVC jackets. CNTs (Unidym, Inc.*) were solution coated onto the cable PE dielectric by two methods: (a) discrete dip coating using a solution of CNTs in chlorosulfonic acid (CSA) (Figure 1c), followed by coagulation, washing in water, and overnight air drying; (b) continuous roll-to-roll wire coating (Figure 1d) with inline deposition of a CNTCSA solution, coagulation, water washing, and off-line overnight air drying. We used method (a) to produce the discrete lengths of cable used in this article; however, method (b) (Figure 1d) is better suited for industrial manufacturing and can be adapted to yield pre-metered coating thickness. For our cables, we controlled the CNT thickness by dip coating multiple layers (Figure S2 in Supporting Information), yielding CNT layer thicknesses of (13 ± 2) μm, (43 ± 4) μm, and (90 ± 14) μm. We used the modified Landau-Levich equation15 in conjunction with rheological measurements to design coating conditions. Specifically, in dip coating of cylindrical wires, it is well known that the coating thickness hdry is related to the withdrawal speed u by the power law relation ℎ𝑑𝑟𝑦 ~ 𝑢𝑝 ,15 where p =2n / (2n+1). Here, n is the power law exponent that controls fluid shear thinning in the expression 𝜂 = 𝐾𝛾̇ 𝑛−1 (lower values of n indicate stronger shear thinning, and n = 1 denotes Newtonian fluids, for which follow the classical Landau-Levich expression16 ℎ𝑑𝑟𝑦 ~ 𝑢2/3 ); here 𝜂 is the fluid viscosity, 𝛾̇ is the shear rate, and K is the consistency index.17 The coatings were fabricated by varying the withdrawal speed for two concentrations (1.3 % and 1 % by mass) and the coating thickness was determined by observing the samples by SEM. Figure 2a shows viscosity η versus shear rate 𝛾̇ . We observe that the CNT-CSA solutions display shear thinning over a broad 5 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

Page 6 of 28

range of shear rates, consistent with earlier reports on CNT solutions in acids,18, 19 as well as CNT suspensions in water.20-22 We find shear thinning exponents of n = 0.43 at 1 % and 0.28 for the 1.3 % solutions—for comparison, an isotropic semidilute solution of rods is expected to show an exponent of 0.5.23 These measured exponents are consistent with a biphasic system with an isotropic CNT solution in equilibrium with an interspersed liquid crystalline phase (more strongly shear thinning) whose fraction is growing with overall CNT concentration. Based on the measured rheological properties, we estimate that the coating thickness should scale with p = 0.46 for the 1 % solution and p = 0.36 for the 1.3 % solution. We investigated the validity of this scaling law by fitting the relationship of coating thickness to coating speed (Figure 2b). We find that the two estimates of the exponent p agree to within 3 % to 15 %, showing that classical scaling laws for polymer processing can be applied to designing solution processing of macroscopic CNT materials. Coating morphology To investigate the local morphology and structure of the CNT layer onto the cable dielectric, we used scanning electron microscopy (SEM), atomic force microscopy (AFM), polarized Raman spectroscopy, and small-angle neutron scattering (SANS). SEM and AFM indicated the presence of CNT bundles oriented parallel to the draw direction as shown in Figure 3a and 3b, respectively. The average order parameter of the CNT coating obtained by polarized Raman spectroscopy was 0.34 ± 0.14 (Figure S3 in Supporting Information). SANS is a useful technique to determine the presence of aligned structure in liquid and solid samples and estimate the degree of alignment. In this case, we used SANS to measure the alignment of the CNT bundles in the coating direction directly taking SANS measurements on the CNT coating peeled off from the dielectric. SANS

6 ACS Paragon Plus Environment

Page 7 of 28

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

measurements on the CNT coating show a strong anisotropy of the scattering signal, confirming the preferential alignment along the cable axis (Figure 4a-b). The degree of CNT bundle alignment was quantified from annular averages (more details in Supporting Information) of the 2D scattering profile (Figure 4a-b) resulting in the intensity spectrum shown in Figure 4c. The alignment angle and the angular distribution were obtained from fits to a Gaussian distribution of the hump-like structures (Figure 4c) and were found to be 98° ± 20.4°. It is worth noting that in this case the alignment angle is determined by the orientation of the cable axis relative to the horizontal axis on the detector. The degree of alignment of the CNT bundles was obtained from fitting the 1D annularly-averaged data (Figure 4c) to the Maier-Saupe distribution of the form 

F (Q,  )   an P2 n (cos  ) , 24, 25 where φ is the alignment angle obtained from the Gaussian fits, n 0

P2n are even Legendre polynomials and an’s are the fit parameters (more details in Supporting Information, Figure S4). We truncated the series to the first five terms of the expansion, which sufficiently reproduced the measured signal. The alignment factor Af is obtained from the fit parameter a1 as Af = a1/5,25 and is ≈ 0.323 for the current sample. In general, Af takes values between zero for randomly oriented fibers and 1 in the case of perfect alignment. The value of 0.323 is in agreement with SEM images that show alignment in the coating direction as well as isotropically-oriented CNTs and confirms the order parameter found by polarized Raman spectroscopy and with alignment factor values typical of aligned soft matter systems, including fibrin, worm-like micelles and polymers.25 At 1.3 % by mass, the high aspect ratio CNTs (~ 4,000) form a liquid crystalline26 CNT solution (Figure S5 in Supporting Information), and the significant shear in the drawing direction during coating results in substantial CNT alignment along the cable axis. Generally, CNT alignment is responsible for improved eletrical conductivity in the alignment direction,27-29 However, higher alignment (order parameter

>

0.3) is required to see an 7

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

Page 8 of 28

improvement in conductivity. Recent studies on dip coated films from CNT-CSA solutions showed isotropic conductivity despite the weak alignment in the coating direction because of the presence of well-packed CNT ropes orthogonal to the coating direction, likely due to log-rolling dynamics of liquid crystalline domains in shear flow.18 In addition, since doping is known to reduce the junction resistance of the CNTs,30, 31 a less pronounced effect of alignment on the anisotropic condictivity is expected for CNT doped aligned coatings. Cable electrical properties under mechanical fatigue We tested the fatigue resistance of 14.2 cm long commercial cables and CNT cables in a threepoint geometry test (Figure S6, Supporting Information). During the test, the direct current (DC) resistance of the CNT layer was measured continuously as a function of bending cycles (Figure 5a). Before and after the bending test, we characterized the signal transmission (specifically, insertion loss) and we plotted the change in transmission (transmission of the cable after the mechanical test normalized by the transmission of the cable before the mechanical test) as a function of frequency (Figure 5b). The cable insertion loss is indicative of how well the signal propagates through the transmission line and accounts for the signal loss through the cable and the one due to the connectors. Cyclical bending increased the DC resistance of the cables with thin CNT layers ((13 ± 2) μm and (43 ± 4) μm); however, the effect of fatigue was negligible for (90 ± 14) μm thick CNT coating after 10,000 cycles. Consistent with the DC measurements, we found that commercial cables and cables with thick CNT coatings showed no change in transmission due to fatigue (Figure 5b)—this is expected for commercial cables because their metal mesh is designed to endure the bend radius (27 mm). Examination of the CNT cables (Figure S7) showed that cables with thinner coatings were damaged because of the geometrical mismatch between

8 ACS Paragon Plus Environment

Page 9 of 28

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

their size and the inner diameter of the insulating jacket (which was sized for the much thicker commercial cable metal braid (~ 500 μm in thickness)). Cable electrical properties In order to show the inherent trade-off between conductivity and weight (Figure 6a) and to select the optimal CNT source material, we made OCs with CNTs from different sources, measured the DC conductivity of the OC normalized by its density (specific conductivity), and plotted it versus the OC linear density (mass per unit length). We used CNTs from CCNI (as in Ref.32 , aspect ratio ~ 2410), from Unidym (similar growth technology as CCNI, higher aspect ratio ~ 4,000), and OCSiAl (recent high-volume manufacturer, lower aspect ratio ~ 1310). We measured the coating conductivity using two techniques: (1) four point probe method directly on the OC and (2) measuring the two point OC resistance of terminated cables of different length (more details in Supporting Information, Figure S8). Due to the elimination of the probe contact resistance, method (1) gives higher values of specific conductivity than method (2). However, due to the braided structure of the commercial cable OC, only method (2) could be used to determine the commercial cable OC conductivity. We note that the coating thickness did not affect the specific conductivity of the OC (Figure S9b in Supporting Information and Figure 6a), indicating that the process produces coatings with comparable structure. Figure 6a summarizes the specific conductivity values versus linear density found for the commerical metal braid and the CNT coatings made with different CNT sources. Unidym CNT coatings gave the highest specific conductivity (1.5 kSm2/kg measured with method (2) and 2 kSm2/kg using method (1)), comparable to that of tinned copper. The conductivity of Unidym CNT coatings is 650 kS/m, 33 % higher than CCNI CNT coatings and twice the conductivity of OCSiAl CNT coatings of comparable thickness. Also, Unidym CNTs showed a conductivity two orders of magnitude higher 9 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

Page 10 of 28

than previously reported CNT OCs.11 The specific conductivity of the CCNI CNT coating is about 67 % lower than the value of aligned CNT fibers32 made with similar CNTs from the same producer (~ 1.5 kSm2/kg by method (1), likely because of lower packing density). The high conductivity of these CNT coatings (in particular of the Unidym CNTs) is due to the high CNT aspect ratio, quality (Raman spectroscopy of the Unidym CNT material showed a G/D ratio of ~ 40, Supporting Information, Figure S10), and to doping intrinsic to the dissolution in CSA. The CNT coating electrical properties are stable over time (Figure S9 in Supporting Information), likely because the residual acid is entrapped inside the CNT coating33 and is characterized by high boliling point. Because of their higher DC conductivity, Unidym CNT coatings were chosen for the characterization of cable attenuation properties. We measured the alternating current (AC) electrical properties of the CNT and commercial cables over a frequency range of 50 MHz to 3 GHz with a broadband, multiline-thru-reflect (TRL) technique using an open-short-load-through (OSLT) corrected vector network analyzer.34 To perform the multiline TRL technique, we fabricated six cables for each CNT coating thickness and for the commercial cables. The nominal lengths of the cables — (5.7, 7.4, 14.2, 22.2, 30.8, 36.9) cm — were chosen to optimally extract the propagation constant 𝛾 as a function of frequency.35, 36 𝛾 is a complex, frequency-dependent parameter (𝛾 = 𝛼 + 𝑖𝛽) that describes how an AC signal (or data) changes as a function of position along a coaxial cable. The real part of the propagation constant is the attenuation (or loss) per unit length 𝛼 as a function of frequency (Figure 6b), and 𝛽 is the phase constant. Lower values of 𝛼 indicate smaller losses through the transmission line, i.e., higher quality cables. We first measured the OSLT corrected complex scattering (S-) parameters of each cable, and then used the multiline TRL algorithm to extract the propagation constant. Because of its firm basis in circuit theory,34, 35, 37 the multiline TRL technique is considered the 10 ACS Paragon Plus Environment

Page 11 of 28

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

most accurate method to obtain γ and allows for the determination of γ and 𝛼 solely coming from the transmission line without the contribution of the connectors. This method allows the estimation of cable attenuation even with short data cable lengths and is thus particularly appropriate when testing cables made with experimental materials. In order to evaluate the correctness of the results obtained by the multiline algorithm, the OSLT corrected S-parameters were fitted to a distributed network model by a least-squares algorithm,38 which uses the length of the cable as the only input parameter (thinner line in Figure 6b). This confirmed that exchanging the commercial metal mesh for the CNT layer only influenced the distributed resistance per unit length (cable distributed resistance versus frequency, Figure S11 in Supporting Information). As expected, increasing the thickness of the CNT layer decreased the attenuation constant (Figure 6b) in agreement with literature,11 and improved the CNT cable quality to a value that is comparable to that of the commercial cables. Specifically, the (90 ± 14) μm CNT cable met the military standard MIL-C-17 at 1 GHz (a reference frequency for military specifications) and was only 30 % higher in attenuation at 400 MHz and about two-fold higher below 100 MHz. A thicker CNT layer or more conductive CNTs would decrease cable distributed resistance leading to lower attenuation values also at low frequencies. While we used RG-174/U, recent work on CNT cables has focused on replacing the metal OC with a KAuBr4 doped CNT OC in RG-58/U coaxial cable, that is a similar cable type with larger diameter size.11 Since different cable types have different attenuation requirements and different OC mass, we present our findings in terms of normalized attenuation at 1 GHz and a parameterized mass. We normalize the attenuation α by the military standard attenuation reported by MIL-C-17 (α0) at 1 GHz for each cable type (dashed line in Figure 6c) and the mass of the CNT OC (m) by the corresponding mass of the commercial cable (m0) for the specific cable type. When 11 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

Page 12 of 28

plotting α/α0 versus m/m0 (Figure 6c), values closer to the origin have lower attenuation and lower weight. Our cables showed comparable attenuation to commercial cables with an OC 30 times lighter. When compared to published results on CNT cables (squares in Figure 6c),11 we improved the attenuation two-fold without compromising mass and while avoiding the need for doping with gold salts, thus producing the best attenuation values to date for CNT cables by an attractive, scalable process. Conclusions In this work we showed that coaxial cable OCs can be solution coated directly onto the dielectric to manufacture data cables with seamless CNT OCs. The coating thickness can be tuned by varying the coating conditions. Due to the shear applied in the coating direction, the CNT coating shows a preferential CNT bundle alignment in the withdrawal direction. The cables made with a CNT OC meet military attenuation specifications at 1 GHz and have durability comparable to commercial cables. To our knowledge, these cables show the best values of attenuation for CNT cables made out of CNT-only OCs and have an OC mass 97 % lighter than conventional metal braids with an overall cable weight reduction of ~ 50 %. When compared to recent published results, the improved properties of our CNT cables can be attributed to the highly crystalline, long CNTs used as well as improved CNT packing, key properties needed to achieve high electrical conductivity.32 From this research, we can conclude that a 1 m length of CNT cable would have 1.5 dB signal attenuation and weigh approximately 7 g, whereas a commercial metal cable would have 1 dB of signal attenuation and weigh 13 g. Despite the lower CNT OC conductivity compared to the metal OC, the CNT cable achieves comparable signal attenuation values. This is because the microwave current propagates in both CC and OC (in opposite directions) equally. To make the total current equal, the charges are distributed about the cross sectional geometry at a given 12 ACS Paragon Plus Environment

Page 13 of 28

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

frequency. Because the current density in the OC is smaller (due to the OC larger cross-sectional area), the losses in the OC contribute less than the losses in the CC for the signal attenuation. Therefore the lower conductivity of a CNT OC affects attenuation only slightly, while dramatically reducing weight. These results show that when adopting a new material such as CNT in a metal-replacement application, it is crucial to reconsider device design and devise and adopt different device architectures that are enabled by those properties of CNTs that are uncommon in metals (in this case, flexibility and fatigue resistance) and can compensate for the relatively lower conductivity of CNT materials. Further refinements to the architecture, together with relatively minor improvement on CNT wire coating process and CNT quality, are likely to yield in the near future viable commercial CNT data cables for the aerospace market.

13 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

Page 14 of 28

Methods Carbon nanotube coating and characterization CNTs were purchased from Unidym, Inc.*, OCSiAl, Inc.*, and Continental Carbon Nanotechnologies, Inc.* (CCNI) and mixed as received at the concentration of 1-1.3 % by mass in CSA (Sigma Aldrich*) using a speed mixer (DAC 150.1 FV-K, Flack Tek Inc.*).18 After coating the dielectric with the CNT solution at 100 mm/s (Figure 1c), the coated dielectric was coagulated in ether for 1 hour, followed by an isopropanol wash for 30 min, then a water bath for 1 hour. The coated dielectric was then air dried at room conditions overnight. Because a higher solution concentration yields thicker coatings (Figure 2), we used 1.3 % by mass solution to produce the cables reported in Figures 3-6. Once the coating was dried, we estimated the mass of the CNT coating by first cutting a segment of coated dielectric. We then melted the PE dielectric in a bath of dichlorobenzene heated at 150 ºC for 20 minutes, followed by a 20 minute bath in dichloromethane at ambient conditions to remove the dichlorobenzene. The CNT coating was dried in the oven at 100 ºC for 10 minutes and measured by a Citizen* microbalance. We determined the thickness of the coating by SEM imaging and a microcaliper (Figure S1 in Supporting Information). We verified CNT material quality by Raman spectroscopy (Figure S10 in Supporting Information). Rheology The viscosity versus shear rate of CNT-CSA solutions with mass fraction of 1 and 1.3 % was measured by an RDA III* strain-controlled rheometer with shallow cup geometry. The solution was loaded between the two plates in an inert condition to avoid the acid from reacting with the moisture in the atmosphere. A layer of ultra low viscosity Fluorinert FC-72 (ACROS Organics*) was placed on the top of the loaded shallow cup to isolate the acid from the environment. 14 ACS Paragon Plus Environment

Page 15 of 28

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

Additionally, a layer of low viscosity silicon oil was added on the top to limit the evaporation of FC-72 during the experiment. Morphology characterization SEM images of the coating on the dielectric were taken by a microscope (FEI Quanta 400ESEM FEG*). AFM measurements (Cypher AFM, Asylum/Oxford Instruments*) were performed in tapping mode at ambient conditions. We used a silicon cantilever (AC160TS-R3, Olympus*) with a spring constant of approximately 30 N/m, a resonant frequency of 295 kHz, a free oscillation amplitude of 100 nm and an imaging set point ratio of approximately 85 %. The linear tip speed during scanning was about 25 µm/s. Small-angle neutron scattering (SANS) measurements were performed at the NG7 30m SANS beam line at the NIST Center for Neutron Research (NCNR). The measurements were done using the standard SANS configurations, covering a Q-range of 0.003 Å-1 to 0.55 Å-1. Data reduction was performed using NCNR Igor macros and data fitting using the SasView software (www.sasview.org). The SANS measurements were performed on a piece of the CNT coating that was carefully peeled off the dielectric and sandwiched between two glass slides. Mechanical testing Mechanical testing was conducted using a 3-point bend test fixture attached to an MTS* servohydraulic load frame (Model 312, 100 kN) equipped with a 15 kN actuator. The upper grip fixture held the 220 N compression/tension load cell with the anvil attached to push down on the cable. The support and loading anvils were equipped with 10 mm diameter bearings. The span between support anvils was 60 mm. A schematic of the experimental loading configuration is provided in Figure S6 in Supporting Information. The cable was kept in tension across the support anvils by steel springs with spring constants of (488 ± 2) mN/mm that were attached to the rigid 15 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

Page 16 of 28

coax connections to isolate the fragile coax fitting from stress. The coiled springs were anchored to aluminum supports rigidly fixed to the 3-point bend fixture. The pre-tension axial force on the cable was (0.7 ± 0.02) N which gave an approximate axial stress of 35 kPa on the cable. The cable was fatigued by positioning the anvil in direct contact with the cable at zero normal load on the load cell. The loading anvil was stationary and the support anvils, connected to the actuator, were oscillated using a triangular ramp with amplitude of 16 mm deflection at a rate of 5 Hz for 10,000 cycles. This motion resulted in bending the length of cable through a (27 ± 2) mm radius of curvature. The normal load at maximum displacement was (13 ± 1) N and the axial force, based on spring displacement, was approximately 4.6 N. During the fatigue test, the DC resistance measurements were taken using a Keithley 1700 multimeter* after letting the cable settle for a period of two minutes to allow for thermal dissipation. Electrical DC and AC characterization The CNT coaxial cables were fabricated from RG-174/U coax (Figure 1a) and subsequently attached to female SMA connectors (Supporting Information, Figure S1). DC resistances were measured with Keithley 2000* multimeter. The microwave electrical measurements were performed on a vector network analyzer Hewlett Packard 8720D* that was corrected with openshort-load (OSLT) lumped-element calibration artifacts. The StatistiCAL software package (http://www.nist.gov/pml/electromagnetics/related-software.cfm) was used to perform the multiline thru-reflect-line analysis.

16 ACS Paragon Plus Environment

Page 17 of 28

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

Conflict of interest The authors declare the following competing financial interest(s): M.P. and F.M. are authors of the PCT international application PCT/US15/12938 describing the method for coating coaxial coatings of carbon nanotubes. M.P. and F.M. have a financial interest in DexMat, Inc., which is commercializing the technology reported in this publication.

Acknowledgments We acknowledge discussions pertaining to high frequency measurement and modeling with J. C. Booth and R. McMichael at National Institute of Standards and Technology (NIST). We thank J. Obrzut for providing the network analyzer, G. Cheng for helpful discussions on Raman spectroscopy, and A. L. Forster for helping in performing the cable mechanical tests. Research was supported by Air Force Office of Scientific Research (AFOSR) grants FA9550-12-1-0035, FA9550-09-01-0370, Air Force Research Laboratories (AFRL) agreement FA8650-07-2-5061, the Robert A. Welch Foundation (C-1668) (for F.M., M.P., A.C., Y.L., E.A.B.); a Cooperative Research Agreement (CRA) between the Rice University and NIST grant 70NANB12H188 (for N.D.O., F.M., K.M., A.R.H.W.), a Cooperative Research Agreement (CRA) between the University of Maryland and NIST grant 70NANB10H193 (for C.J.L.). The mechanical testing of the cable was funded through the Service Life of Nano-enabled Structural Polymer Composites (#7314003.000) (for A.M.F). R.J.H. was supported by a NASA Space Technology Research Fellowship (NSTRF14), grant number NNX14AL71H. Some of the measurements were conducted at the Center for Nanoscale Science and Technology, a user facility at NIST. This work utilized NIST Center for Neutron Research (NCNR) facilities supported in part by the National Science Foundation under Grant No. DMR-0944772 (for R.A. and P.B.). 17 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

Page 18 of 28

Supporting information available: Assembly of the CNT cables, CNT coating thickness measurement, polarized Raman spectroscopy on the CNT coating, SANS analysis and fitting, micrographs of the CNT-CSA liquid crystals, mechanical test set up, DC conductivity of the inner and outer conductors in the cables, Raman spectroscopy characterization of the CNT powder before dissolution in acid, specific conductivity of the CNT coating, distributed resistance and inductance of the commercial and CNT cables. This material is available free of charge via the Internet at http://pubs.acs.org.

18 ACS Paragon Plus Environment

Page 19 of 28

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

References 1. Jarosz, P.; Schauerman, C.; Alvarenga, J.; Moses, B.; Mastrangelo, T.; Raffaelle, R.; Ridgley, R.; Landi, B., Carbon Nanotube Wires and Cables: Near-Term Applications and Future Perspectives. Nanoscale 2011, 3, 4542-4553. 2. Pozar, D. M., Microwave Engineering. Hoboken, NJ : John Wiley: 2005. 3. Chung, D., Electromagnetic Interference Shielding Effectiveness of Carbon Materials. Carbon 2001, 39, 279-285. 4. De Volder, M. F.; Tawfick, S. H.; Baughman, R. H.; Hart, A. J., Carbon Nanotubes: Present and Future Commercial Applications. Science 2013, 339, 535-539. 5. Laforgue, A.; Ajji, A.; Robitaille, L., Development of Conductive Polymeric Fibers for Flexible Electronics Applications. In 2009 CF/DRDC International Defence Applications of Materials Meeting, Victoria, BC, Canada, April 8, 2009, Victoria, BC, Canada, 2009; pp 1-8. 6. Treacy, M. M. J.; Ebbesen, T. W.; Gibson, J. M., Exceptionally High Young's Modulus Observed for Individual Carbon Nanotubes. Nature 1996, 381, 678-680. 7. Ruoff, R. S.; Lorents, D. C., Mechanical and Thermal Properties of Carbon Nanotubes. Carbon 1995, 33, 925-930. 8. Baughman, R. H.; Zakhidov, A. A.; de Heer, W. A., Carbon Nanotubes - the Route toward Applications. Science 2002, 297, 787-792. 9. Sreekumar, T. V.; Liu, T.; Kumar, S.; Ericson, L. M.; Hauge, R. H.; Smalley, R. E., SingleWall Carbon Nanotube Films. Chem. Mat. 2003, 15, 175-178. 10. Baur, J.; Silverman, E., Challenges and Opportunities in Multifunctional Nanocomposite Structures for Aerospace Applications. MRS Bull. 2007, 32, 328-334. 11. Jarosz, P. R.; Shaukat, A.; Schauerman, C. M.; Cress, C. D.; Kladitis, P. E.; Ridgley, R. D.; Landi, B. J., High-Performance, Lightweight Coaxial Cable from Carbon Nanotube Conductors. ACS Appl. Mater. Interfaces 2012, 4, 1103-1109. 12. Harvey, S. E. In Carbon as Conductor: A Pragmatic View, 61st International Wire & Cable Symposium, 2012. 13. Jarosz, P.; Shaukat, A.; Mastrangelo, T.; Schauerman, C.; Cress, C. D.; Ridgley, R.; Landi, B., Coaxial Cables with Single-Wall Carbon Nanotube Outer Conductors Exhibiting Attenuation/Length within Specification. Micro & Nano Letters 2012, 7, 959-961. 14. Thue, W. A., Electrical Power Cable Engineering. Marcel Dekker: New York: 1999. 15. de Ryck, A.; Quéré, D., Fluid Coating from a Polymer Solution. Langmuir 1998, 14, 19111914. 16. Levich, B.; Landau, L., Dragging of a Liquid by a Moving Plate. Acta Physicochim. URSS 1942, 17, 42-54. 17. Makosko, C. W., Rheology: Principles, Measurements, and Applications. Wiley-vch: 1994. 18. Mirri, F.; Ma, A. W. K.; Hsu, T. T.; Behabtu, N.; Eichmann, S. L.; Young, C. C.; Tsentalovich, D. E.; Pasquali, M., High-Performance Carbon Nanotube Transparent Conductive Films by Scalable Dip Coating. ACS Nano 2012, 6, 9737-9744. 19. Davis, V. A.; Ericson, L. M.; Parra-Vasquez, A. N. G.; Fan, H.; Wang, Y. H.; Prieto, V.; Longoria, J. A.; Ramesh, S.; Saini, R. K.; Kittrell, C.; Billups, W. E.; Adams, W. W.; Hauge, R. H.; Smalley, R. E.; Pasquali, M., Phase Behavior and Rheology of Swnts in Superacids. Macromolecules 2004, 37, 154-160.

19 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

Page 20 of 28

20. Dan, B.; Irvin, G. C.; Pasquali, M., Continuous and Scalable Fabrication of Transparent Conducting Carbon Nanotube Films. ACS Nano 2009, 3, 835-843. 21. Hobbie, E. K., Shear Rheology of Carbon Nanotube Suspensions. Rheol. Acta 2010, 49, 323-334. 22. Allen, R.; Bao, Z.; Fuller, G. G., Oriented, Polymer-Stabilized Carbon Nanotube Films: Influence of Dispersion Rheology. Nanotechnology 2013, 24, 015709. 23. Larson, R. G., The Structure and Rheology of Complex Fluids. Oxford University Press, New York: 1999. 24. Walker, L. M.; Wagner, N. J., Sans Analysis of the Molecular Order in Poly (Γ-Benzyl LGlutamate)/Deuterated Dimethylformamide (Pblg/D-Dmf) under Shear and During Relaxation. Macromolecules 1996, 29, 2298-2301. 25. Weigandt, K. M.; Porcar, L.; Pozzo, D. C., In Situ Neutron Scattering Study of Structural Transitions in Fibrin Networks under Shear Deformation. Soft Matter 2011, 7, 9992-10000. 26. Davis, V. A.; Parra-Vasquez, A. N. G.; Green, M. J.; Rai, P. K.; Behabtu, N.; Prieto, V.; Booker, R. D.; Schmidt, J.; Kesselman, E.; Zhou, W.; Fan, H.; Adams, W. W.; Hauge, R. H.; Fischer, J. E.; Cohen, Y.; Talmon, Y.; Smalley, R. E.; Pasquali, M., True Solutions of SingleWalled Carbon Nanotubes for Assembly into Macroscopic Materials. Nat. Nanotechnol. 2009, 4, 830-834. 27. Fischer, J. E.; Zhou, W.; Vavro, J.; Llaguno, M. C.; Guthy, C.; Haggenmueller, R.; Casavant, M. J.; Walters, D. E.; Smalley, R. E., Magnetically Aligned Single Wall Carbon Nanotube Films: Preferred Orientation and Anisotropic Transport Properties. J. Appl. Phys. 2003, 93, 2157-2163. 28. Pint, C. L.; Xu, Y. Q.; Morosan, E.; Hauge, R. H., Alignment Dependence of OneDimensional Electronic Hopping Transport Observed in Films of Highly Aligned, Ultralong Single-Walled Carbon Nanotubes. Appl. Phys. Lett. 2009, 94, 182107. 29. Zamora-Ledezma, C.; Blanc, C.; Puech, N.; Maugey, M.; Zakri, C.; Anglaret, E.; Poulin, P., Conductivity Anisotropy of Assembled and Oriented Carbon Nanotubes. Phys. Rev. E 2011, 84, 062701. 30. Nirmalraj, P. N.; Lyons, P. E.; De, S.; Coleman, J. N.; Boland, J. J., Electrical Connectivity in Single-Walled Carbon Nanotube Networks. Nano Lett. 2009, 9, 3890-3895. 31. Znidarsic, A.; Kaskela, A.; Laiho, P.; Gaberscek, M.; Ohno, Y.; Nasibulin, A. G.; Kauppinen, E. I.; Hassanien, A., Spatially Resolved Transport Properties of Pristine and Doped Single-Walled Carbon Nanotube Networks. J. Phys. Chem. C 2013, 117, 13324-13330. 32. Behabtu, N.; Young, C. C.; Tsentalovich, D. E.; Kleinerman, O.; Wang, X.; Ma, A. W. K.; Bengio, E. A.; ter Waarbeek, R. F.; de Jong, J. J.; Hoogerwerf, R. E.; Fairchild, S. B.; Ferguson, J. B.; Maruyama, B.; Kono, J.; Talmon, Y.; Cohen, Y.; Otto, M. J.; Pasquali, M., Strong, Light, Multifunctional Fibers of Carbon Nanotubes with Ultrahigh Conductivity. Science 2013, 339, 182186. 33. Green, M. J.; Young, C. C.; Parra-Vasquez, A. N. G.; Majumder, M.; Juloori, V.; Behabtu, N.; Pint, C. L.; Schmidt, J.; Kesselman, E.; Hauge, R. H.; Cohen, Y.; Talmon, Y.; Pasquali, M., Direct Imaging of Carbon Nanotubes Spontaneously Filled with Solvent. Chem. Commun. 2011, 47, 1228-1230. 34. Williams, D. F.; Wang, C.; Arz, U. In An Optimal Multiline Trl Calibration Algorithm, IEEE MTT-S Int. Microwave Symp. Dig., IEEE; 1999: 2003; pp 1819-1822. 35. Marks, R. B., A Multiline Method of Network Analyzer Calibration. IEEE Trans. Microwave Theory Tech. 1991, 39, 1205-1215. 20 ACS Paragon Plus Environment

Page 21 of 28

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

36. Lee, C.-H.; Orloff, N. D.; Birol, T.; Zhu, Y.; Goian, V.; Rocas, E.; Haislmaier, R.; Vlahos, E.; Mundy, J. A.; Kourkoutis, L. F.; Nie, Y.; Biegalski, M. D.; Zhang, J.; Bernhagen, M.; Benedek, N. A.; Kim, Y.; Brock, J. D.; Uecker, R.; Xi, X. X.; Gopalan, V.; Nuzhnyy, D.; Kamba, S.; Muller, D. A.; Takeuchi, I.; Booth, J. C.; Fennie, C. J.; Schlom, D. G., Exploiting Dimensionality and Defect Mitigation to Create Tunable Microwave Dielectrics. Nature 2013, 502, 532-536. 37. DeGroot, D. C.; Jargon, J. A.; Marks, R. B. In Multiline Trl Revealed, ARFTG Conference Digest, Fall 2002. 60th, 5-6 Dec. 2002; 2002; pp 131-155. 38. Booth, J. C.; Orloff, N. D.; Mateu, J.; Janezic, M.; Rinehart, M.; Beall, J. A., Quantitative Permittivity Measurements of Nanoliter Liquid Volumes in Microfluidic Channels to 40 Ghz. IEEE Trans. Instrum. Meas. 2010, 59, 3279-3288. 39. Gutfinger C.; Tallmadge J. A., Films of Non-Newtonian Fluids Adhering to Flat Plates. AIChE J. 1965, 11, 403-413.

21 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

Page 22 of 28

FIGURES

Figure 1| (a) A photograph of a CNT coaxial cable with SubMiniature version A (SMA) connectors (Inset: SMA connector at an auxiliary view). (b) Top, schematics of a CNT coaxial cable compared to a conventional commercial cable. Bottom, photographs of the CNT coaxial cables and the conventional commercial cable with the different coatings revealed. (c) Laboratorybased dip coating process used to coat the coaxial cables for the data presented in Figure 3 and Figure 4. (d) Scalable roll-to-roll wire coating for the fabrication of CNT coatings.

22 ACS Paragon Plus Environment

Page 23 of 28

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

Figure 2| Coating thickness can be tuned by varying the solution concentration and coating speed. (a) Viscosity versus shear rate for CNT-CSA solutions with mass fractions of (1 and 1.3) % Unidym CNTs where n represents the power law exponent; each data set represents an average of 3 samples independently prepared. (b) Coating thickness versus withdrawal speed for (1 and 1.3) % by mass solutions; thickness measurements were obtained by SEM. The exponent p calculated using n from the rheology data and the one predicted by lubrication analysis39 agree within 15 % for 1 % solution and 3 % for 1.3 % solution.

23 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

Page 24 of 28

Figure 3| (a) Scanning electron microscope micrographs of the CNT layer for each thickness value. The draw direction (arrow) shows that the CNT bundles oriented along the draw direction. (b) Atomic force microscope (AFM) images of a (90 ± 14) μm CNT coating on a coated coaxial cable showing bundle alignment and uniform coverage. Far left shows an optical image of the investigated surface.

24 ACS Paragon Plus Environment

Page 25 of 28

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

Figure 4| (a) 2D pattern of the normalized scattering intensity obtained by small-angle neutron scattering (SANS) measurements on the CNT coating clearly indicating aligned CNT bundles, as evident from the strong anisotropy in the signal. (b) Fit of the 2D SANS signal in (a) to a model of aligned CNTs. (c) Annular intensity average of the 2D scattering profile in (b). The red line represents the best fit.

25 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

Page 26 of 28

Figure 5| (a) DC resistance of the CNT (Unidym) cables. During the 10,000 cycle bending test, the relative DC resistance of the 90 μm CNT cable increased by about 1 %, demonstrating excellent mechanical durability at 27 mm bending radius. (b) Change in transmission (insertion loss) relative to the initial value, which shows that the thickest CNT coating retained their AC performance even after 10,000 bending cycles.

26 ACS Paragon Plus Environment

Page 27 of 28

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

Figure 6| (a) Specific conductivity of the OC (conductivity normalized by density) as a function of the coaxial cable OC mass per unit length (linear density). Error bars represent ± 1 standard deviation. The squares represents the values of specific conductivity measured by method (1), while the circles the ones measured by method (2). U stands for Unidym CNT coatings, C stands for CCNI CNT coatings, and O stands for OCSiAl CNT coatings. (b) The attenuation constant versus frequency for the different Unidym CNT coaxial cables and the commercial cables. The multiline algorithm (solid lines) and least-squares fit (thinner lines) were used to extract the attenuation constant. The uncertainty (shaded regions) was computed by error propagation. The purple dots represent the military standard for attenuation at 0.05, 0.1, 0.4, and 1 GHz for RG174/U (1.5 dB/m or 45 dB/100ft). (c) Normalized attenuation (α/αo) versus normalized mass (m/mo) for the Unidym CNT coaxial cables and commercial cable. Attenuation (α) was normalized by military standard attenuation (αo) at 1 GHz for the RG-174/U cable type (1.5 dB/m, dashed line). Squares are published work on RG-58/U cables11, compared to their military standard attenuation (red dashed line). The yellow square represents the KAuBr4 doped coating from reference11. Values closer to the origin have improved attenuation and lower mass. The error bars for our work are within the marker size.

27 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

Page 28 of 28

TOC

28 ACS Paragon Plus Environment