Acceptor Percolation Determines How Electron-Accepting Additives

May 31, 2018 - (6) In addition, electron injection in p-channel OFETs may cause ... are those that incorporate the donor–acceptor (D–A) alternatin...
0 downloads 0 Views 1MB Size
Subscriber access provided by Kaohsiung Medical University

Article

Acceptor Percolation Determines how Electron-Accepting Additives Modify Transport of Ambipolar Polymer Organic Field-Effect Transistors Michael J. Ford, Ming Wang, Karen C. Bustillo, Jianyu Yuan, Thuc-Quyen Nguyen, and Guillermo C. Bazan ACS Nano, Just Accepted Manuscript • DOI: 10.1021/acsnano.8b03006 • Publication Date (Web): 31 May 2018 Downloaded from http://pubs.acs.org on June 1, 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 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Nano

Acceptor Percolation Determines how ElectronAccepting Additives Modify Transport of Ambipolar Polymer Organic Field-Effect Transistors Michael J. Ford‡||¶*, Ming Wang||¶†, Karen C. Bustillo§, Jianyu Yuan¥, Thuc-Quyen Nguyen||¶†, Guillermo C. Bazan|‡||¶†* ‡

Materials Department, University of California, Santa Barbara, California, 93106, USA.

||

Center for Polymers and Organic Solids, University of California, Santa Barbara, California,

93106, USA. ¶

Mitsubishi Chemical Center for Advanced Materials, University of California, Santa Barbara,

California, 93106, USA. †

Department of Chemistry and Biochemistry, University of California, Santa Barbara, California,

93106, USA. §

National Center for Electron Microscopy, Molecular Foundry, Lawrence Berkeley National

Laboratory, Berkeley, California, 94720, USA. ¥

Institute of Functional Nano & Soft Materials (FUNSOM), Jiangsu Key Laboratory for Carbon-

Based Functional Materials & Devices, Soochow University, Suzhou, Jiangsu, 215123, P. R. China. *E-mail: [email protected], [email protected]

ACS Paragon Plus Environment

1

ACS Nano 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 25

KEYWORDS organic field-effect transistors, ambipolar, alternative acceptor, morphology, organic semiconductors, polymer semiconductors

Abstract

Organic field-effect transistors (OFETs) that utilize ambipolar polymer semiconductors can benefit from the ability of both electron and hole conduction, which is necessary for complementary circuits. However, simultaneous hole and electron transport in organic fieldeffect transistors results in poor ON/OFF ratios, which limits potential applications. Solution processing methods have been developed to control charge transport properties and transform ambipolar conduction to hole-only conduction. The electron acceptor phenyl-C61-butyric acid methyl ester (PC61BM), when mixed in solution with an ambipolar semiconducting polymer can truncate electron conduction. Unipolar p-type OFETs with high, well-defined ON/OFF ratios and without detrimental effects on hole conduction are achieved for a wide range of blend compositions, from 95:5 to 5:95 wt. % semiconductor polymer:PC61BM. When introducing the alternative acceptor N,N′-bis(1-ethylpropyl)-3,4:9,10-perylenediimide (PDI), high ON/OFF ratios are achieved for 95:5 wt. % semiconductor polymer:PDI; however, electron conduction increases for 50:50 wt. % semiconductor polymer:PDI. As described within, we show that electron conduction is practically eliminated when additive domains do not percolate across the OFET channel; i.e., electrons are “morphologically trapped”. Morphologies were characterized by optical, electron, and atomic force microscopy as well as X-ray scattering techniques. PC61BM

was substituted with an endohedral Lu3N fullerene, which enhanced contrast in electron

microscopy and allowed for more detailed insight into the blend morphologies. Blends with alternative, non-fullerene acceptors emphasize the importance of morphology and acceptor

ACS Paragon Plus Environment

2

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

ACS Nano

percolation, providing insights for such blends that control ambipolar transport and ON/OFF ratios. In the study of organic field-effect transistors (OFETs), the charge carrier mobility (µ) has been the preeminent figure of merit when comparing the efficacy of newly-designed organic semiconductors.1 The gradual channel approximation describes the current as a function of voltage as follows:2  =

×

 =

×





[  −    −

  

]

(  −  )

(1)

(2)

In these equations Id is the drain current; W is the channel width; Ci is the dielectric areal capacitance; L is the channel length; Vg is the gate voltage; Vd is the source-drain voltage; and VT is the threshold voltage. Equation 1 is the general expression of the current-voltage characteristics, simplified to Equation 2 in the saturation regime valid when |Vd| ≥ |Vg – VT|. For complementary logic circuits, the ratio of the current in the ON-state relative to the current in the OFF-state, i.e., ON/OFF ratio (ION/IOFF), is also considered important to maximize since a smaller ION/IOFF reduces noise margins and increases static power consumption.3 Higher µ typically results in larger ION/IOFF, given that µ is proportional to the ON current. However, when ambipolar characteristics are observed, ION/IOFF may be difficult to interpret. For ambipolar semiconductors in particular, the OFF current cannot be explicitly evaluated since it is dependent on the source-drain voltage and gate voltage.4 Most often in the literature, ION/IOFF for ambipolar semiconductors has been defined by the ratio of the maximum and minimum currents. This particular way of determining ION/IOFF is less meaningful for complementary inverters, which require large ION/IOFF with IOFF minimized across a wide voltage range.5

ACS Paragon Plus Environment

3

ACS Nano 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

Despite

the

aforementioned

challenges

Page 4 of 25

with

managing

ION/IOFF,

ambipolar

semiconductors are interesting alternatives as the current-carrying material in complementary circuits; a single semiconductor utilized for both the p- and n-type components would eliminate the need for separate optimizations for each component and would reduce the total device fabrication cost. However, since ambipolar OFETs never reach a well-defined OFF-state, complementary circuits utilizing ambipolar semiconductors are typically inferior with respect to noise immunity and power consumption, when compared to two separately-optimized p- and norganic semiconductors.6 In addition, electron injection in p-channel OFETs may cause biasstressing instability observed under certain configurations.7 Various groups have proposed routes to truncate ambipolar conduction by selectively disrupting transport of one charge carrier while the counterpart flows unimpeded. Methods to modulate ambipolar conduction have involved channel polarity control,5 introduction of dopants,8 dielectric modifications,9–11 and contact modifications.12,13 More recently, we observed that solution-processable additives can also control ambipolar conduction and improve ION/IOFF.6,14 In our original work, where difficulties associated with ambipolarity were tamed,14 we noted that among the highest performing polymer semiconductors reported are those that incorporate the donor-acceptor (D-A) alternating structure (e.g., Ref. 15–19). In many cases, as in

poly[4-(4,4-dihexadecyl-4H-cyclopenta[1,2-b:5,4-b′]dithiophen-2yl)-alt-

[1,2,5]thiadiazolo[3,4-c]pyridine] (PCDTPT, see Figure 1a for molecular structure), a polymer semiconductor of particular interest due to its reported high µ,20–23 the D-A orbital interactions reduce the gap between the highest occupied molecular orbital (HOMO) and the lowest unoccupied molecular orbital (LUMO). For narrow-gap organic semiconductors, e.g., D-A polymers, ambipolar conduction becomes more favorable as the gap decreases and thus ION/IOFF

ACS Paragon Plus Environment

4

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

ACS Nano

is small under common OFET architectures.4,7 ION/IOFF of PCDTPT and other D-A polymer semiconductors could be improved by up to 5 orders of magnitude after introducing additives with appropriate LUMO (HOMO) energy levels to block electron (hole) transport, e.g., phenylC61-butyric acid methyl ester (PC61BM).6,14 For p-channel OFETs, fullerene additives effectively act as traps to block electron transport. Ambipolar transport was modulated to pchannel transport for a wide concentration range (i.e., 99:1-5:95 wt. % PCDTPT:PC61BM). For low fullerene loading, there are no percolation pathways to support electron transport in the fullerene phase. That higher fullerene content (e.g., 5:95 wt. % PCDTPT:PC61BM) does not exhibit increased electron current relative to more dilute fullerene conditions was thought to be consistent with the absence of percolating electron transport pathways due to phase separation (Figure 1b.) Percolation pathways of fullerene domains that become tortuous and disconnected, upon annealing for example, have been demonstrated for fullerenes in organic photovoltaics.24–26 Polymer:acceptor OFET active layers, especially at high weight content acceptor (e.g., 5:95 wt. % polymer:acceptor) have not yet been well-characterized. In analogy to the bulk heterojunction that is now well-studied, there is a need to understand the interplay between charge transport and morphology as well as the impact of processing and alternative (i.e., nonfullerene) electrontrapping additives in order to maximize ON/OFF ratios for ambipolar polymer blend OFETs. In this contribution, we examine how morphology impacts electron transport in PCDTPT:additive blends. The morphologies of PCDTPT incorporating PC61BM were investigated by electrical characterization methods, optical microscopy, atomic force microscopy (AFM), transmission electron microscopy (TEM), and grazing incidence wide-angle X-ray scattering (GIWAXS), which corroborated the hypothesis that percolating electron transport pathways influence trapping properties and ION/IOFF. For improved contrast in the TEM analysis,

ACS Paragon Plus Environment

5

ACS Nano 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 25

an endohedral fullerene, Lu3N@C80-PCBEH (ethyl-hexyl, Lu-PCBEH) was also used. We also incorporated an alternative electron acceptor, N,N′-bis(1-ethylpropyl)-3,4:9,10-perylenediimide (PDI) with PCDTPT. Alternative nonfullerene additives 1) are technologically important due to lower cost and chemical tunability relative to fullerenes; 2) extend the generality of our chargetrapping method in showing how dilute trap conditions can also modify ambipolar transport to ptype transport; and 3) provide insight into how morphology dictates charge transport, since unlike fullerenes, higher concentrations of these alternative acceptors support electron conduction. These results offer morphological guidelines for successful tuning of charge transport properties in semiconductor blends. Results/Discussion

ACS Paragon Plus Environment

6

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

ACS Nano

Figure 1. a) Structures of molecules referred to in this manuscript. b) Cartoon of hypothesized morphology from a top view for high-fullerene-content OFETs. Color scheme: gold = source and drain contacts; green = polymer semiconductor; purple = fullerene; dark blue arrow = hole transport; pink arrow = electron transport; light blue = dielectric substrate. In the cartoon, fullerene phases are disrupted and thus cannot support electron conduction; hole

ACS Paragon Plus Environment

7

ACS Nano 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 25

conduction is uninterrupted adjacent to the dielectric and is thus unimpeded by the fullerene phase. Not drawn to scale. c)

Representative current-voltage characteristics for

PCDTPT:PC61BM blends with the wt. % listed relative to PCDTPT content with Vd = -80 V. These results are similar to those obtained in Ref. 14 and are reproduced for context. d) Representative current-voltage characteristics for PCDTPT:PDI blends with the wt. % listed relative to PCDTPT content with Vd = -80 V. Table 1. Device characteristics for PCDTPT:acceptor blends. The average of at least 8 devices and the standard deviations are reported. ION/IOFF is reported as a range of magnitudes and is the ratio of the maximum and minimum Id for those active layers that do not exhibit a true OFF state due to electron conduction. OFETs were measured by sweeping Vg forward and backwards at Vd = -80 V from Vg of 10 or 20 to -60 V, and µ and VT were calculated from Equation 2 when fitting between Vg of ~-30 to -50 V. Active Layer

µ (cm2 V-1 s-1)

VT (V)

ION/IOFF

Pristine PCDTPT

0.37±0.07

2.8±4.8

101-102

95:5 wt. % PCDTPT:PC61BM

0.35±0.05

-8.6±0.7

104-106

50:50 wt. % PCDTPT:PC61BM

0.30±0.10

-8.5±1.4

104-105

5:95 wt. % PCDTPT:PC61BM

0.27±0.03

-7.7±1.5

104-106

95:5 wt. % PCDTPT:PDI

0.35±0.06

-9.3±0.5

105

50:50 wt. % PCDTPT:PDI

0.22±0.07

-9.2±1.1

104-105

ACS Paragon Plus Environment

8

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

ACS Nano

5:95 wt. % PCDTPT:PDI

0.21±0.06

-4.9±1.5

102

To demonstrate the effect of electron-accepting additives when blended with PCDTPT, bottom-gate

bottom-contact

PCDTPT:PC61BM

OFETs

were

fabricated

(Active

layer/(Au/Ni)/silane-treated SiO2/doped Si) as in previous reports.6,14 Devices were measured in the saturation regime, utilizing Equation 2 to calculate relevant parameters (Table 1). Ambipolar transport characteristics are evident in transfer curves of pristine PCDTPT (Figure 1c, dashed grey trace). Hole conduction becomes dominant between Vg = 0 to -10 V. Electron conduction is also occurring, turning on and increasing as Vg becomes more positive. Note that in these cases the devices do not have a well-defined OFF state; therefore, the minimum current is used for ION/IOFF values reported in this manuscript. Upon addition of 5 wt. % PC61BM, electron conduction is suppressed and hole transport is unaffected. The ION/IOFF increases from ~101-102 in PCDTPT to ~104-106 in PCDTPT:PC61BM blends. This increase was expected for dilute concentrations of PC61BM where pathways to support electron transport are unlikely to form. As PC61BM loading increases, electron conduction remains low. Again, hole transport is unaffected, which is consistent with a morphology of disrupted pathways for electron transport, even at 5:95 PCDTPT:PC61BM. Electrons are trapped due to the morphology, which results in an absence of fullerene domains percolating across the channel. The thin film morphology was thus examined to investigate this hypothesis, but first we compare PCDTPT:PC61BM OFETs to an alternative acceptor blend. PCDTPT:PDI OFETs were also fabricated in the same architecture as the PCDTPT:PC61BM devices. Relevant device characteristics are summarized in Table 1. Like

ACS Paragon Plus Environment

9

ACS Nano 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 25

PC61BM, 95:5 wt. % PCDTPT:PDI is also effective at converting ambipolar to p-only transport, with ION/IOFF = 2.5±0.6 × 105 (Figure 1d). However, upon further PDI loading, the measured OFF current increased by ~3 orders of magnitude. Conducting pathways of PDI likely allow for electron conduction. In this case, hole conduction was also suppressed, indicating that PDI may be more miscible than PC61BM or may solidify concomitantly with PCDTPT, disrupting the internal film morphology and hole transport pathways. Although just one PDI analog was studied in detail here, unipolar p-channel transport was observed for three other PDI derivatives (Figure S1). To clarify the differences observed between PCDTPT:PC61BM and PCDTPT:PDI OFET blends, the morphologies of the thin films, especially of PCDTPT:PC61BM, were studied in detail.

ACS Paragon Plus Environment

10

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

ACS Nano

Figure 2. Optical microscopy, AFM, and GIWAXS of PCDTPT:PC61BM blends. (a-c) Optical microscopy (left) and AFM (right) images of corresponding PCDTPT:PC61BM blends with wt. % given relative to the amount of PCDTPT. Optical microscopy images are 280 µm wide. Scale bars in atomic force microscopy are 1 µm (a,b) and 4 µm (c). (d) In-plane segment plots from GIWAXS of PCDTPT:PC61BM blends. The associated PC61BM peak at 0.7 Å-1 has been highlighted.

ACS Paragon Plus Environment

11

ACS Nano 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 25

PCDTPT:PC61BM blends were first examined by optical and atomic force microscopy. Pristine PCDTPT films are smooth and featureless by optical microscopy (Figure 2a). By AFM, fiber bundle formation less than ~100 nm in width,21 and a root-mean-square roughness (RMS) of 0.89 nm are observed. Addition of PC61BM, e.g., 50:50 wt. % PCDTPT:PC61BM (Figure 2b), results in microscale features forming on the film surface. AFM reveals domains of ~300 nm, which are larger and less fiber-like than pristine PCDTPT (RMS = 2.71 nm). The change in morphology is attributed to the formation of PC61BM domains. For 5:95 wt. % PCDTPT:PC61BM (Figure 2c), microscale features similar to those seen in the 50:50 wt. % case are observed but in greater quantity, providing support that these domains correspond to PC61BM crystallites. Such large-scale domains made AFM analysis difficult due to roughness. Careful selection of scanning location, where domains were not observed by optical microscopy (Figure S2), permitted imaging that revealed ~2 µm structures with an RMS of 73.6 nm. Domains observed by optical microscopy along with the mesoscale structure observed by AFM are evidence of poor PC61BM percolation, which is consistent with high ION/IOFF ratios in 5:95 wt. % PCDTPT:PC61BM (Figure 1a). Thus, PC61BM likely acts as an electron trap due to the tortuous and disrupted morphology in high PC61BM weight content blends. GIWAXS can probe the crystallinity of the thin film to complement the observations by microscopy.27 For 95 wt. % PCDTPT:PC61BM, no characteristic peak of PC61BM (i.e., at qxy = ~0.7 Å-1) is evident, indicating mixed polymer:fullerene domains or a subthreshold detection level (Figure 2d). Pure domains of PC61BM crystallites appear in 50:50 wt. % PCDTPT:PC61BM, as indicated by the peak at ~0.7 Å-1 and consistent with the microscopy data. The peak at ~0.7 Å-1 is also observed for 20:80, 10:90, and 5:95 wt. % PCDTPT:PC61BM. The peak at ~1.8 Å-1, attributed to PCDTPT π-π stacking, is apparent in all blend compositions,

ACS Paragon Plus Environment

12

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

ACS Nano

including the 5:95 wt. % PCDTPT:PC61BM. PC61BM thus does not alter the internal ordering of PCDTPT crystalline domains, which was expected since µ did not change as PC61BM content increased. GIWAXS corroborates the microscopy data, showing that fullerene domains grow as fullerene content increased, and GIWAXS also provides an explanation as to why fullerene additives do not influence hole transport.

ACS Paragon Plus Environment

13

ACS Nano 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 25

Figure 3. TEM of 5:95 wt. % PCDTPT:Lu-PCBEH. a) HAADF-STEM image. Green square defines the region mapped by EDS. Bright contrast corresponds to the Lu-rich phase and dark contrast to the Lu-poor phase; b) EDS map of S-K edge (purple) and Lu-L edge (green); c) EDS spectra from the two phases. Green spectrum is from Lu-rich regions and purple spectrum is from sulfur-rich/Lu-poor regions. TEM was used to probe the thin film regions. Contrast in TEM typically relies on differences in electron density.28,29 Since carbon is the majority constituent of organic semiconductors, electron density contrast between two organic semiconductors is often too low to be useful or relies on significant differences in crystallinity. For the films that were studied for this report, phase separation is apparent between PCDTPT and PC61BM for 5:95 wt. % PCDTPT:PC61BM (Figure S3). Large scale phase separation is consistent with optical and atomic force micrographs, resembling a tortuous network of disrupted domains. To compare, optical microscopy, AFM and TEM of 5:95 wt. % PCDTPT:PDI showed continuous crystalline domains (Figures S4-S5), which is consistent with higher electron conduction in these blends relative to 5:95 wt. % PCDTPT:PC61BM. However, we were interested in also probing areas corresponding to the PCDTPT-rich portion of the film adjacent to the dielectric to better understand the function of the electron acceptor in these blends. From the initial morphology examination, we could not rule out the presence of mixed phase or amorphous PC61BM domains and were thus interested in probing amorphous regions that would be difficult to observe under typical bright field TEM. Thus we substituted PC61BM with Lu-PCBEH (see Figure S6 for molecular structure); the endohedral Lu atoms were anticipated to enhance Z-contrast and allowed for probing the thin film by energy dispersive X-ray spectroscopy (EDS).29 We note that the miscibility of Lu-

ACS Paragon Plus Environment

14

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

ACS Nano

PCBEH relative to PC61BM was found to be dependent on the host polymer.28-30 Various control experiments were therefore carried out to confidently assess that Lu-PCBEH impacts electronic properties similarly to PC61BM (see Supporting Information). We first note that the electron affinity of Lu-PCBEH is lower relative to PC61BM by ~0.2-0.3 eV.31 Thus, Lu-PCBEH may be less effective as an electron acceptor for PCDTPT relative to PC61BM (Figure S7). Lu-PCBEH, which has a similar electron affinity relative to PCDTPT,32 was then introduced as an additive to OFETs that used poly[2,6-(4,4-bis-alkyl-4H-cyclopenta[2,1-b;3,4-b']-dithiophene)-alt-4,7-(2,1,3benzothiadiazole)] (PCDTBT), which has a lower electron affinity relative to PCDTPT. LuPCBEH was found to be equally effective as PC61BM in improving ION/IOFF of PCDTBT OFETs (Figures S7, S8). The morphologies of PCDTBT:Lu-PCBEH blends are similar to that of PCDTPT:Lu-PCBEH blends (Figure S9), which increased our confidence that the morphologies observed by electron microscopy could be correlated with PC61BM blends. The internal organizations of PCDTPT:Lu-PCBEH blends were characterized by high angle annular dark field (HAADF) scanning transmission electron microscopy (STEM), which exploits the Z-contrast between the two components. Images were acquired using a FEI Titan 80300 operating at 60kV. A 5:95 wt. % PCDTPT:Lu-PCBEH blend annealed at 200oC will be the focus of this discussion, but other conditions were measured as well (Figures S10-S16). For 5:95 wt. % PCDTPT:Lu-PCBEH, large phase-separated features that

resemble spinodal

decomposition are observed (Figure 3a). Fiber formation, akin to morphologies observed by AFM for pristine PCDTPT, are observed in the darker areas of Figure 3a. Increasing the contrast of the image shows fiber formation more clearly (Figure S17). An area was selected for EDS mapping (Figure 3b, Figure S18). EDS mapping indicates that the white regions of Figure 3a are Lu-rich whereas the dark regions are Lu-poor. EDS profiles (Figure 3c) were extracted from

ACS Paragon Plus Environment

15

ACS Nano 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 25

the Lu-rich areas and the Lu-poor areas (Figure S18). Signals from low-Z elements like C, N, O, and S dominate the low energy region of the spectra (from ca. 0-3.5 keV). Contribution from the Au grid is observed at ca. 9.7 keV. For evaluation of Lu-content of each region, we focus on the Lα-edge at ca. 7.7 keV. There was negligible Lu signal in the Lu-poor region, which suggests a very low to nearly-zero concentration of mixed phase regions. Thus, the low electron current for high weight content fullerene blends is justified: fullerene percolation does not occur and cannot support electron transport. Electron acceptor additives must disrupt morphologies to effectively truncate electron transport at a wide range of additive concentrations, without impacting the organization of the hole transport layer. Conclusions The control of ON/OFF ratios, especially for ubiquitous ambipolar polymer semiconductors, has the potential to significantly impact future technologies in the area of organic electronics. As described, we combined different semiconductors and additives and examined the morphological features that enable electron trapping and hole conduction across a range of compositions (from 95:5 to 5:95 wt. % polymer:additive). Different additives are tolerant of changes in composition to different extents. PC61BM is effective from 95:5 to 5:95 wt. % PCDTPT:PC61BM since PC61BM apparently assembles in domains that do not percolate across the entire channel, that do not disrupt hole transport, and that are adjacent to the transport layer and/or electrodes to block electron transport. Morphological investigation by optical microscopy, AFM, GIWAXS, and TEM indicate 1) the presence of unperturbed PCDTPT domains that supported hole conduction and 2) disrupted PC61BM domains that can act as morphological traps. When an endohedral Lu3N fullerene is used in place of PC61BM, HAADFSTEM and EDS analyses can be done to clarify how fullerenes control electron transport even at

ACS Paragon Plus Environment

16

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

ACS Nano

high weight content. Alternative electron acceptors were also examined. PDI is effective at 95:5 wt. % PCDTPT:PDI. At higher concentrations of PDI, electron conduction is increased, and hole conduction decreases. The morphology of PCDTPT:PDI blends showed percolating domains that were attributed to PDI at 5:95 wt. % PCDTPT:PDI, which is likely why electron conduction increased. It is important that OFET transport properties, particularly ON/OFF ratios, can be tuned by solution-processable additives. The choice of acceptor and blend composition dictate the final morphology, which is critical for modulating ambipolar transport. These results provide the following morphological guidelines for future research in similar semiconductor blends: additive domains should be disrupted so the acceptor cannot support electron (or hole) conduction; an uninterrupted pathway for hole (or electron) conduction should exist for the current-driving semiconductor layer; and the additive should assemble close enough to the charge transport layer to influence characteristics. These types of semiconductor blends provide a useful platform for achieving high ON/OFF ratios for efficient complementary devices. Methods/Experimental Thin Film Processing: Stock solutions of PCDTPT (1 Materials), PDI, and PC61BM (Solenne) were dissolved in chlorobenzene and mixed by volume to obtain the corresponding blend weight ratio. The blend concentration was 5 mg/mL with respect to the total PCDTPT + PC61BM content. Thin films were processed by spin-coating at 2500 RPM or by blade-coating at a speed of 1.2 mm/s with a substrate temperature of 100oC. The blade-coater and stage was built using a LTA-HS actuator and integrated CONEX-CC controller. The motor controls the motion of the blade in the lateral direction while a micrometer is used to control the blade height relative to a hot plate, which is used as the surface for the substrate. The blade angle was set to 60o relative to the plane of the substrate. The blades used were glass microscope slides, which were cleaned by

ACS Paragon Plus Environment

17

ACS Nano 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 25

piranha solution. The substrate was coated by injecting 8 µL of solution in between a ~100-200 µm channel formed between the blade and the substrate. Films were annealed at 200oC unless specified otherwise. The substrates were cleaned by sonication in acetone and isopropanol for 3 minutes each and then dried in an oven under ambient atmosphere at 120 oC for 10 minutes. The substrates were treated with UV-O3 for 15 minutes and then passivated by using decyltrichlorosilane (Gelest Chemicals) from a 0.2 vol% solution in toluene at 80 oC for 25 minutes. The substrates were then rinsed and sonicated in toluene and dried under nitrogen flow prior to film deposition. For OFETs, source and drain contacts (5 nm Ni/50 nm Au) were deposited using a standard two-step photolithography. The Ni adhesion layer was etched by submerging in dilute Ni etchant after deposition. Morphological Characterization: GIWAXS measurements were performed at beamline 7.3.3 at the Advanced Light Source with an X-ray wavelength of 1.2398 Å at a 300 mm sample detector distance. The measurements were calibrated using an AgB Standard. Samples were scanned in a He environment at an incident angle of 0.12o. Micrographs were obtained using an Olympus MX51 microscope at 31.5x magnification with differential interference contrast. AFM images were obtained in air using an Innova AFM. TEM samples were prepared by first spincoating a 10 mg/mL solution of sodium polystyrene sulfonate in water on top of a silicon dioxide substrate. A thin (ca. 30 nm) layer of Cyclotene (Dow Chemical Company, also referred to as BCB) was used as a film support layer. The semiconductor blend was spun on top and floated off in onto a ultrathin carbon film on lacey carbon Au TEM grid (Ted Pella). HAADF-STEM images were obtained using an FEI TitanX 60-300 TEM operating at 60 kV with a HAADF detector. A Bruker windowless EDS detector was used for EDS.

ACS Paragon Plus Environment

18

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

ACS Nano

Device Characterization: The final OFET architecture (from bottom to top) for bottom gate, bottom contact field-effect transistors was Si (500 µm)/DTS-treated SiO2 (300 nm)/Ni (5 nm)/Au (50 nm)/blend layer. Cylcotene (30 nm) was also used as a dielectric layer for some OFETs referenced in the SI. The mobility for blend devices was obtained by fitting the Equation 2 to the saturation regime transfer characteristics, where W is the channel width (1 mm), L is the channel length (5-160 µm), and Ci is the gate dielectric layer capacitance per unit area (10 nF/cm2). Devices were measured under nitrogen in a glovebox using a Signatone 1160 probe station and Keithley 4200 semiconductor parametric analyzer. Mobility values calculated from a gate voltage range of about -30 V to -50 V at a source-drain voltage of -80 V.

ASSOCIATED CONTENT Supporting Information. Additional microscopy and current-voltage data is available free of charge via the Internet at http://pubs.acs.org. AUTHOR INFORMATION Corresponding Authors *[email protected], [email protected] Author Contributions The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. Funding Sources This work is supported by Mitsubishi Chemical Center for Advanced Materials.

ACS Paragon Plus Environment

19

ACS Nano 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 25

The authors declare no competing financial interest ACKNOWLEDGMENT We thank Dr. Zachary Henson for the synthesis of the swallow-tail PDI derivatives. This research used resources of the Advanced Light Source, which is a DOE Office of Science User Facility under contract no. DE-AC02-05CH11231. Work at the Molecular Foundry was supported by the Office of Science, Office of Basic Energy Sciences, of the U.S. Department of Energy under Contract No. DE-AC02-05CH11231. The MRL Shared Experimental Facilities are supported by the MRSEC Program of the NSF under Award No. DMR 1720256; a member of the NSF-funded Materials Research Facilities Network. J. Yuan would like to acknowledge the support from the Natural Science Foundation of Jiangsu Province of China (BK20170337) and the National Natural Science Foundation of China (Grant No. 51761145013) REFERENCES (1)

Sirringhaus, H. 25th Anniversary Article: Organic Field-Effect Transistors: The Path

Beyond Amorphous Silicon. Adv. Mater. 2014, 26, 1319–1335. (2)

Zaumseil, J.; Sirringhaus, H. Electron and Ambipolar Transport in Organic Field-Effect

Transistors. Chem. Rev. 2007, 107, 1296–1323. (3)

Meller, G.; Grasser, T. Organic Electronics; Springer, 2009.

(4)

Kang, M. S.; Frisbie, C. D. A Pedagogical Perspective on Ambipolar FETs.

ChemPhysChem 2013, 14, 1547–1552. (5)

Torricelli, F.; Ghittorelli, M.; Smits, E. C. P.; Roelofs, C. W. S.; Janssen, R. A. J.;

Gelinck, G. H.; Kovács-Vajna, Z. M.; Cantatore, E. Ambipolar Organic Tri-Gate Transistor for Low-Power Complementary Electronics. Adv. Mater. 2016, 28, 284–290.

ACS Paragon Plus Environment

20

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

ACS Nano

(6)

Ford, M. J.; Labram, J. G.; Wang, M.; Wang, H.; Nguyen, T.-Q.; Bazan, G. C. Carrier-

Selective Traps: A New Approach for Fabricating Circuit Elements with Ambipolar Organic Semiconductors. Adv. Electron. Mater. 2017, 3, 1600537-1–1600537-6. (7)

Phan, H.; Wang, M.; Bazan, G. C.; Nguyen, T.-Q. Electrical Instability Induced by

Electron Trapping in Low-Bandgap Donor–Acceptor Polymer Field-Effect Transistors. Adv. Mater. 2015, 27, 7004–7009. (8)

Khim, D.; Baeg, K.-J.; Caironi, M.; Liu, C.; Xu, Y.; Kim, D.-Y.; Noh, Y.-Y. Control of

Ambipolar and Unipolar Transport in Organic Transistors by Selective Inkjet-Printed Chemical Doping for High Performance Complementary Circuits. Adv. Funct. Mater. 2014, 24, 6252– 6261. (9)

Chua, L.-L.; Zaumseil, J.; Chang, J.-F.; Ou, E. C.-W.; Ho, P. K.-H.; Sirringhaus, H.;

Friend, R. H. General Observation of N-Type Field-Effect Behaviour in Organic Semiconductors. Nature 2005, 434, 194–199. (10)

Baeg, K.-J.; Khim, D.; Jung, S.-W.; Kang, M.; You, I.-K.; Kim, D.-Y.; Facchetti, A.;

Noh, Y.-Y. Remarkable Enhancement of Hole Transport in Top-Gated N-Type Polymer FieldEffect Transistors by a High-k Dielectric for Ambipolar Electronic Circuits. Adv. Mater. 2012, 24, 5433–5439. (11)

Nakano, M.; Osaka, I.; Takimiya, K. Control of Major Carriers in an Ambipolar Polymer

Semiconductor by Self-Assembled Monolayers. Adv. Mater. 2017, 29, 1602893-1–1602893-7. (12)

Cheng, X.; Noh, Y.-Y.; Wang, J.; Tello, M.; Frisch, J.; Blum, R.-P.; Vollmer, A.; Rabe, J.

P.; Koch, N.; Sirringhaus, H. Controlling Electron and Hole Charge Injection in Ambipolar Organic Field-Effect Transistors by Self-Assembled Monolayers. Adv. Funct. Mater. 2009, 19, 2407–2415.

ACS Paragon Plus Environment

21

ACS Nano 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

(13)

Page 22 of 25

Xu, Y.; Sun, H.; Shin, E.-Y.; Lin, Y.-F.; Li, W.; Noh, Y.-Y. Planar-Processed Polymer

Transistors. Adv. Mater. 2016, 28, 8531–8537. (14)

Ford, M. J.; Wang, M.; Phan, H.; Nguyen, T.-Q.; Bazan, G. C. Fullerene Additives

Convert Ambipolar Transport to P-Type Transport While Improving the Operational Stability of Organic Thin Film Transistors. Adv. Funct. Mater. 2016, 26, 4472–4480. (15)

Olivier, Y.; Niedzialek, D.; Lemaur, V.; Pisula, W.; Müllen, K.; Koldemir, U.; Reynolds,

J. R.; Lazzaroni, R.; Cornil, J.; Beljonne, D. 25th Anniversary Article: High-Mobility Hole and Electron Transport Conjugated Polymers: How Structure Defines Function. Adv. Mater. 2014, 26, 2119–2136. (16)

Zhang, M.; Tsao, H. N.; Pisula, W.; Yang, C.; Mishra, A. K.; Müllen, K. Field-Effect

Transistors Based on a Benzothiadiazole−Cyclopentadithiophene Copolymer. J. Am. Chem. Soc. 2007, 129, 3472–3473. (17)

Nketia-Yawson, B.; Lee, H.-S.; Seo, D.; Yoon, Y.; Park, W.-T.; Kwak, K.; Son, H. J.;

Kim, B.; Noh, Y.-Y. A Highly Planar Fluorinated Benzothiadiazole-Based Conjugated Polymer for High-Performance Organic Thin-Film Transistors. Adv. Mater. 2015, 27, 3045–3052. (18)

Venkateshvaran, D.; Nikolka, M.; Sadhanala, A.; Lemaur, V.; Zelazny, M.; Kepa, M.;

Hurhangee, M.; Kronemeijer, A. J.; Pecunia, V.; Nasrallah, I.; Romanov, I.; Broch, K.; McCulloch, I.; Emin, D.; Olivier, Y.; Cornil, J.; Beljonne, D. Sirringhaus, H. Approaching Disorder-Free Transport in High-Mobility Conjugated Polymers. Nature 2014, 515, 384–388. (19)

Sun, B.; Hong, W.; Yan, Z.; Aziz, H.; Li, Y. Record High Electron Mobility of 6.3

Cm2V−1s−1 Achieved for Polymer Semiconductors Using a New Building Block. Adv. Mater. 2014, 26, 2636–2642.

ACS Paragon Plus Environment

22

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

ACS Nano

(20)

Ying, L.; Hsu, B. B. Y.; Zhan, H.; Welch, G. C.; Zalar, P.; Perez, L. A.; Kramer, E. J.;

Nguyen,

T.-Q.;

Heeger,

A.

J.;

Wong,

W.-Y.;

Bazan,

G.

C.

Regioregular

Pyridal[2,1,3]Thiadiazole π-Conjugated Copolymers. J. Am. Chem. Soc. 2011, 133, 18538– 18541. (21)

Tseng, H.-R.; Ying, L.; Hsu, B. B. Y.; Perez, L. A.; Takacs, C. J.; Bazan, G. C.; Heeger,

A. J. High Mobility Field Effect Transistors Based on Macroscopically Oriented Regioregular Copolymers. Nano Lett. 2012, 12, 6353–6357. (22)

Tseng, H.-R.; Phan, H.; Luo, C.; Wang, M.; Perez, L. A.; Patel, S. N.; Ying, L.; Kramer,

E. J.; Nguyen, T.-Q.; Bazan, G. C.; Heeger, A. J. High-Mobility Field-Effect Transistors Fabricated with Macroscopic Aligned Semiconducting Polymers. Adv. Mater. 2014, 26, 2993– 2998. (23)

Ford, M. J.; Wang, M.; Patel, S. N.; Phan, H.; Segalman, R. A.; Nguyen, T.-Q.; Bazan, G.

C. High Mobility Organic Field-Effect Transistors from Majority Insulator Blends. Chem. Mater. 2016, 28, 1256–1260. (24)

Collins, B. A.; Tumbleston, J. R.; Ade, H. Miscibility, Crystallinity, and Phase

Development in P3HT/PCBM Solar Cells: Toward an Enlightened Understanding of Device Morphology and Stability. J. Phys. Chem. Lett. 2011, 2, 3135–3145. (25)

Domanski, A. L.; Lieberwirth, I.; Sengupta, E.; Landfester, K.; Butt, H.-J.; Berger, R.;

Rauh, J.; Dyakonov, V.; Deibel, C. Effect of Morphological Changes on Presence of Trap States in P3HT:PCBM Solar Cells Studied by Cross-Sectional Energy Filtered TEM and Thermally Stimulated Current Measurements. J. Phys. Chem. C 2013, 117, 23495–23499. (26)

Bartelt, J. A.; Beiley, Z. M.; Hoke, E. T.; Mateker, W. R.; Douglas, J. D.; Collins, B. A.;

Tumbleston, J. R.; Graham, K. R.; Amassian, A.; Ade, H.; Fréchet, J. M. J.; Toney, M. F.;

ACS Paragon Plus Environment

23

ACS Nano 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 25

McGehee, M. J. The Importance of Fullerene Percolation in the Mixed Regions of Polymer– Fullerene Bulk Heterojunction Solar Cells. Adv. Energy Mater. 2013, 3, 364–374. (27)

Salleo, A.; Kline, R. J.; DeLongchamp, D. M.; Chabinyc, M. L. Microstructural

Characterization and Charge Transport in Thin Films of Conjugated Polymers. Adv. Mater. 2010, 22, 3812–3838. (28)

Roehling, J. D.; Batenburg, K. J.; Swain, F. B.; Moulé, A. J.; Arslan, I. Three-

Dimensional Concentration Mapping of Organic Blends. Adv. Funct. Mater. 2013, 23, 2115– 2122. (29)

Roehling, J. D.; Baran, D.; Sit, J.; Kassar, T.; Ameri, T.; Unruh, T.; Brabec, C. J.; Moulé,

A. J. Nanoscale Morphology of PTB7 Based Organic Photovoltaics as a Function of Fullerene Size. Sci. Rep. 2016, 6, srep30915. (30)

Liedtke, M.; Sperlich, A.; Kraus, H.; Baumann, A.; Deibel, C.; Wirix, M. J. M.; Loos, J.;

Cardona, C. M.; Dyakonov, V. Triplet Exciton Generation in Bulk-Heterojunction Solar Cells Based on Endohedral Fullerenes. J. Am. Chem. Soc. 2011, 133, 9088–9094. (31)

Ross, R. B.; Cardona, C. M.; Guldi, D. M.; Sankaranarayanan, S. G.; Reese, M. O.;

Kopidakis, N.; Peet, J.; Walker, B.; Bazan, G. C.; Van Keuren, E.; Holloway, E. C.; Drees, M. Endohedral Fullerenes for Organic Photovoltaic Devices. Nat. Mater. 2009, 8, 208–212. (32)

Wang, M.; Ford, M.; Phan, H.; Coughlin, J.; Nguyen, T.-Q.; Bazan, G. C. Fluorine

Substitution Influence on Benzo[2,1,3]Thiadiazole Based Polymers for Field-Effect Transistor Applications. Chem. Commun. 2016, 52, 3207–3210. (33)

Szendrei, K.; Jarzab, D.; Chen, Z.; Facchetti, A.; Loi, M. A. Ambipolar All-Polymer Bulk

Heterojunction Field-Effect Transistors. J. Mater. Chem. 2010, 20, 1317–1321.

ACS Paragon Plus Environment

24

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

ACS Nano

(34)

Larson, B. W.; Whitaker, J. B.; Wang, X.-B.; Popov, A. A.; Rumbles, G.; Kopidakis, N.;

Strauss, S. H.; Boltalina, O. V. Electron Affinity of Phenyl–C61–Butyric Acid Methyl Ester (PCBM). J. Phys. Chem. C 2013, 117, 14958–14964.

Insert Table of Contents Graphic and Synopsis Here

Solution processable electron-accepting additives, like PC61BM and PDI, can tune ambipolar transport to unipolar p-type transport. However, the tolerance to changes in blend composition depends on the morphological arrangement of the additive. For effective unipolarization, a morphology must exist such that the acceptor cannot support electron (or hole) conduction while the ambipolar semiconductor must still be able to efficiently transport charge across the channel.

ACS Paragon Plus Environment

25