Molecularly Controlled Stark Effect Induces ... - ACS Publications

Subscriber access provided by YORK UNIV

Communication

Molecularly Controlled Stark Effect Induces Significant Rectification in Polycyclic Aromatic Hydrocarbon-terminated n-Alkanethiolates Soo Jin Cho, Gyu Don Kong, Sohyun Park, Jeongwoo Park, Seo Eun Byeon, Taekyeong Kim, and Hyo Jae Yoon Nano Lett., Just Accepted Manuscript • DOI: 10.1021/acs.nanolett.8b04488 • Publication Date (Web): 24 Dec 2018 Downloaded from http://pubs.acs.org on December 26, 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 24 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

Nano Letters

Molecularly Controlled Stark Effect Induces Significant Rectification in Polycyclic Aromatic Hydrocarbon-terminated n-Alkanethiolates Soo Jin Choa†, Gyu Don Konga†, Sohyun Parka, Jeongwoo Parkb, Seo Eun Byeona, Taekyeong Kimb*, Hyo Jae Yoona* aDepartment bDepartment

of Chemistry, Korea University, Seoul, 136-701, Korea

of Physics, Hankuk University of Foreign Studies, Yongin, 449-791, Korea †These authors contributed equally to this work.

Abstract Variation of electronic structure of individual molecules as a function of applied bias matters for performance of molecular and organic electronic devices. Understanding the structureelectric field relationship, however, remains a challenge because of lack of in-operando spectroscopic technique and complexity arising from ill-defined on-surface structure of molecules and organic-electrode interfaces within devices. Here, we report reliable and reproducible molecular diode can be achieved by control of conjugation length in polycyclic aromatic hydrocarbon-terminated n-alkanethiolate (denoted as SC11PAH), incorporated into liquid metal-based large-area tunnel junctions in the form of self-assembled monolayer. By taking advantage of structural simplicity and tunability of SC11PAH and high-yielding feature of the junction technique, we demonstrate the increase in conjugation length of the PAH terminal group leads to significant rectification ratio up to ~1.7 × 102 at ±740 mV. Further study suggests that the Stark shift of molecular energy resonance of PAH breaks the symmetry of energy topography across the junction and induces rectification in a temperature-independent charge transport regime.

ACS Paragon Plus Environment

Nano Letters 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 24

Introduction All molecular and organic electronic devices expose organic component to external electric field. Hence, understanding how the electronic structures of individual molecules within the organic component behave in variable applied biases is crucial for not only elucidating the mechanism of electronic function of device but also honing its performance. When a molecule is placed in an external electric field, electrostatic interaction between charge distribution of molecule and electric field takes place, and often induces shift or splitting of spectroscopic signal. This phenomenon has been long established as the Stark effect since its first discovery in 1913.1 While Stark effect has been recently investigated in some nanostructures such as quantum dots and 2D materials,2-4 little has been reported about Stark effect in the context of electronic function of molecular-scale devices, particularly in conjunction with statistically reliable experimental data. In molecular electronics, the rate of charge transport by quantum tunneling in electrode-molecule-electrode junction depends on the energy topography across the junction under applied bias.5-8 Delineating quantitatively the change of energy topography before and after applying bias is challenging because many junctions have ill-defined organic-electrode interfaces and little atomic-level information concerning conformation of molecules inside them.7,

9-11

Analytical techniques sensitive

enough characterize these in operando are rarely available and accessible. Moreover, significant junction-to-junction variation in terms of tunneling characteristics7, 12-13 or change of charge transport regime during operation14 make it difficult to single out the contribution of Stark effect to electronic function of molecular junction. Recently, a few stimulating studies have described experimentally observed tunneling characteristics in single-molecule junctions based on the Stark effect.15-16 For example, the symmetry of energy topography of symmetric Au/S(thiophene)nS/Au junction has been broken by the environmental control—exposing different surface areas for top- and


Page 3 of 24 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

Nano Letters

bottom-electrodes to an ionic liquid—leading to Stark effect-induced asymmetric features in current-voltage traces. Here, we demonstrate that the Stark effect can be modulated, molecularly, through systematic control of conjugation length of polycyclic aromatic hydrocarbon (PAH) terminated in n-alkanethiol (Figure 1a). With this approach, we reproducibly achieved rectification ratios of ~1.7 × 102 at voltage of 740 mV and ~1.5 × 102 at voltage of 1000 mV using pyrene-terminated n-undecanethiolate. By taking advantage of the systematic changes in electronic structure such as the energy gap between highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) and polarizability, induced by the control of conjugation length of PAHs, our work shows that asymmetric tunneling behavior could emerge by a molecular means. Further analysis suggests that the control of PAH structure is translated into Stark shift of PAH’s molecular resonance energy and hence its energy level alignment with respect to electrode Fermi level. As a consequence, our approach leads to remarkably high rectification ratios in a temperature-independent regime. Results and Discussion In

this

study

we

used

well-defined

large-area

junction

of

the

form,

AgTS/SC11PAH//Ga2O3/EGaIn, where AgTS is template-stripped silver having ultraflat surface (rms of ~1.2 nm);17 SC11PAH is PAH-terminated n-undecanethiolate; Ga2O3/EGaIn is eutectic gallium indium alloy covered with self-passivating oxide layer of ~0.7 nm nominal thickness;13,

18

and “//” indicates the presence of van der Waals contact (Figure 1a). This

EGaIn-based junction platform guarantees convenience in operation and fabrication, high yields of working junctions, and collection of large amount of data for statistically proven interpretation of data.7, 19 We varied the conjugation length of PAH by changing the structure from naphtyl (NAP) to phenanthrenyl (PHE), anthracenyl (ANT), pyrenyl (PYR) and benzo[a]pyrenyl (BP) (Figure 1b). For details of synthesis and characterization of HSC11PAH


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

Figure 1. a) Schematic illustration of AgTS/SC11PAH//Ga2O3/ EGaIn junction (PAH=PYR as an exemplary structure) where conjugation length of PAH is systematically varied while keeping other components identical. In our junction, the bottom electrode (AgTS) supporting the SAM is grounded. b) Control of electronic structure (HOMO-LUMO gap) with varying the conjugation length. c) Analysis of SAM with X-ray photoelectron spectroscopy (XPS): exemplary survey and high resolution C1s and S2p spectra for SC11PYR on AgTS.


Page 4 of 24

Page 5 of 24 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

Nano Letters

molecules, see Supporting Information. 1H,

13C

NMR and HRMS data confirmed all the

compounds were consistent with the desired structures. SAMs were prepared by following previously reported procedures. Briefly, freshly prepared AgTS was incubated in 3 mM toluene solution of thiol under N2 atmosphere for 3 h. After incubation, the SAM-bound AgTS was rinsed with pure toluene three times and dried in air. The structure of SAMs was analyzed with X-ray photoelectron spectroscopy (XPS). Figure 1c shows exemplary data of XPS: high resolution spectra measured over AgTS/SC11PYR SAM. Deconvolution of high resolution spectra for C1s led us to estimate the ratio of sp3 and sp2 carbons. The measured atomic ratio of these carbons was consistent with the theoretical one. Table S1 in Supporting Information summarizes the data for all the molecules. The high resolution spectrum of S2p showed spin-orbit doublets at ~163.0 eV and ~161.8 eV corresponding to S2p1/2 and S2p3/2, indicative of the presence of thiolate species, covalently linked to the Ag surface.20 The other molecules exhibited the identical signals in S2p spectra. (See the Supporting Information for all the XPS spectra and data analysis). To understand the packing structure of SAMs, we conducted further structural characterizations of SAMs with i) static and dynamic contact angle goniometries, ii) wetelectrochemical analyses (measurements of percent electrochemically active surface area (%EAS), and reductive desorption), and iii) atomic force microscopy (AFM). Static and dynamic contact angle measurements permit access to the information on the surface structure and the degree of structural disorder. In particular, measurements of dynamic contact angles (hysteresis of contact angles estimated from the difference between advancing and receding contact angles) on SAMs gives direct information about the surface landscape.21-22 As shown in Figure S27-S28 and Table S1-S2 in the Supporting Information, all the SAMs we tested showed indistinguishable static and dynamic contact angles. Figure


Nano Letters 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

S21-S22 and Table S3-S4 in the Supporting Information summarize data of %EAS and reductive desorption measurements on our SAMs. Value of %EAS is the ratio of peak reduction currents for a SAM‐bound electrode to the corresponding bare electrode. %EAS allows one to relatively compare the degree of structural defect between SAMs.23 High %EAS values indicate defective monolayers, whereas low %EAS values indicate well packed monolayers. The measured values of %EAS were similar across the SC11PAH SAMs. We also estimated surface coverage (Γ, nmol/cm2) of SAMs using reductive desorption method.24-25 (see the Supporting Information for details). The coverage our SAMs narrowly ranged from ~13 to ~16 nmol/cm2, indicating indistinguishable surface coverage of the SAMs. Using AFM, we characterized the surface topography of our SAMs. As shown in Figure S23 in the Supporting Information, we did not observe significant difference of surface roughness and topography for our SAMs. All the structural characterization data clearly indicate that the packing structure of the SC11PAH SAMs did not differ significantly. We conducted electrical characterization of SAMs following the previously reported procedures and measured current density (J, A/cm2) and rectification ratio—quotient of current density (J, A/cm2) at ±V; |r-|=|J(-V)|/|J(+V)| and |r+|=1/|r-|. Figure 2 displays histograms of log|J| at ±1.0 V, and log|r-|, and heatmaps of the corresponding log|J|-V traces. All the histograms exhibited normal distributions and fit single Gaussian curves from which values of mean, median and standard deviations in J and r- were extracted; Table 1 summarizes these data. No significant outliers were observed in the histograms, and yields of working junctions ranged from 88% to 100%. For all the molecules, the mean values of rwere indistinguishable from the median values (|r-|median), and the distributions of r- values were narrow (standard deviations of |r-|, σ|r-|= ~2). These findings indicate that our rectification data were statistically significant and reproducible. PYR- and BP-terminated


Page 6 of 24

Page 7 of 24 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

Nano Letters

Figure 2. Histograms of a) log|J| and b) log|r-| at ±1.0, and c) heatmaps of log|J|-V traces measured for SC11PAH SAMs (PAH=NAP, PHE, ANT, PYR and BP).


Nano Letters 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 24

Table 1. Results of Electrical Characterization for AgTS/SC11PAH//Ga2O3/EGaIn Junctions (PAH=NAP, PHE, ANT, PYR and BP).

PAH

Number of junctionsa

Number of J-V traces

Yield of working junctions (%)

BP

15

630

PYR

15

ANT

log|J(+1.0V)|mean ± σlog|J| (log|J(+1.0V)|median)

log|J(-1.0V)|mean ± σlog|J| (log|J(-1.0V)|median)

log|r-|mean ± σlog|r-| (|r-| mean ± σ|r-|)

100

-1.6 ± 0.3 (-1.7)

-0.2 ± 0.6 (-0.1)

1.7 ± 0.3 (46 ± 2)

630

100

-2.6 ± 0.2 (-2.5)

-0.4 ± 0.2 (-0.5)

2.2 ± 0.3 (148 ± 2)

8

336

100

-1.6 ± 0.4 (-1.7)

-1.9 ± 0.3 (-2.1)

-0.4 ± 0.2 (0 ± 2)

PHE

13

546

88

-2.0 ± 0.3 (-1.9)

-2.5 ± 0.6 (-2.3)

-0.3 ± 0.2 (1 ± 2)

NAP

13

546

88

-1.8 ± 0.5 (-1.7)

-1.6 ± 0.6 (-1.9)

-0.2 ± 0.2 (1 ± 2)

aTwo

to five junctions in different places were measured on a sample, and three samples were

tested.


Page 9 of 24 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

Nano Letters

n-alkanethiolates exhibited markedly high rectification ratios: |r-| at ±1.0 V for PYR and BP were 148 ± 2 and 46 ± 2, respectively. The highest r- value was recorded as 172 ± 2 for PYR at 740 mV. In contrast, PHE-, ANT- and NAP-terminated n-alkanethiolates did not rectify. To examine the origin of rectification, we first performed two separate experiments: replacement of Ga2O3/EGaIn top-contact with different electrode (gold), and junction measurements at variable temperatures. Assuming that pure Ga2O3 is n-type semiconductor with a bandgap of ~4.8 eV,26 the PYR//Ga2O3 top-interface might resemble p-n junction diode. To test this hypothesis, we replaced the Ga2O3/EGaIn top-electrode with Au cantilever using conducting probe atomic force microscope (CPAFM) while keeping other components constant.

Figure

3a

shows

the

current-voltage

(I-V)

traces

measured

from

AgTS/SC11PYR//Aucantillever junctions. Large values of r- (|r-|mean = 79 ± 2) were recorded. The rectification ratio was slightly decreased by approximately a factor of two rectification ratio as compared to that measured in EGaIn junction. The small decrease in rectification ratio is probably due to the change of work function for top-electrode: the work function of nanoscale Au tip (~4.8 eV)27 is close to the HOMO of PYR, and would result in reduced contact resistance as compared to that of EGaIn, leading to slightly reduced rectification ratio. Nonetheless, the rectification for the gold top-electrode was significant, which indicates that rectification is not limited to the Ga2O3/EGaIn top-electrode and stems from the SC11PYR molecule. Our conclusion is in line with previous results that have demonstrated that the Ga2O3 in conical EGaIn tip (that we used here) is highly defective and behaves as metal rather than wide-bandgap semiconductor, and thus, does not affect junction measurements.7, 28 Next, to investigate whether the observed rectification predominantly relies on thermally activated hopping, we performed junction measurements at variable temperatures. The rationale for this experiment is that the identical junction except the replacement of PAH


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

Figure 3. a) The plot of current (I, pA) as a function of applied voltage (0 V to ±1.0 V), measured in the CPAFM junction. The gray region represents standard deviation. Note that the opposite polarity of rectification to that in the EGaIn junction results from the difference in electrical grounding—bottom and top electrodes are grounded for the EGaIn and CPAFM junctions, respectively. Data of low temperature (from 298 K to 98 K) experiments for AgTS/SC11PYR//Ga2O3/EGaIn junction. b) J-V traces measured at ±1.0 V; c) the corresponding Arrhenius plot; d) log|r-| plot against temperature.


Page 10 of 24

Page 11 of 24 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

Nano Letters

with ferrocene (Fc)—AgTS/SC11Fc//Ga2O3/EGaIn junction—has shown rectification of the identical magnitude (|r-| = ~1.0×102) and polarity.14 The rectification of SC11Fc is explained by asymmetrically occurring thermally activated hopping at –V and not +V. More specifically, there is a transition of charge transport regime from tunneling to hopping at the energy offset between the HOMO of Fc (EHOMO = ~5.0. eV at 0 V) and the Fermi level (EF) of Ga2O3/EGaIn (~4.3 eV at 0 V) at –V while pure tunneling dominates transport at +V. The presence of asymmetrically occurring electron hopping has been experimentally confirmed by junction measurements in liquid nitrogen temperature. Thus, to test if such a hopping is responsible for our rectification, we constructed ‘untethered’ EGaIn junction (that having the same junction structure with the conventional EGaIn junction but in untethered form)29 and brought it into cryogenic probe station. Under vacuum condition (~1 × 10-4 Torr) temperature was lowered using liquid nitrogen, and J-V curves for SC11PAH SAMs were recorded at the range of temperature from 298 K to 98 K. (See the Supporting Information for detailed data of low-temperature experiments). Rather surprisingly, the values of J(+V), J(-V) and r- did not significantly change over the range of temperature (Figure 3b-d). This finding indicates that despite structural similarity between the PAH- and the Fc-based junctions (T=PYR vs. Fc in the AgTS/SC11T//Ga2O3/EGaIn junction; Figure S14 in the Supporting Information), the rectification of SC11PYR and SC11BP depends on charge transport in temperatureindependent regime. To understand the rectification of the SC11PAH junctions more in detail, we considered a Landauer expression (Eq. 1) for which 2𝑒 𝑒𝑉/2

𝐼 = ℎ ∫ ―𝑒𝑉/2𝑇(𝐸, 𝜀, 𝛤) 𝑑𝐸

(1)

we assume off-resonant tunneling, zero temperature, and no significant differences in lateral interaction inside monolayer across different PAH groups. In Eq. 1, e is the fundamental unit


Nano Letters 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 24

of charge, h is Planck’s constant, and V is the applied voltage. By considering a transport via single electronic level, the transmission function T(E, ε, Γ) is assumed to be a single Lorentzian, as shown in Eq. 2. It depends on the energy (E), the molecular 𝛤2/4

𝑇(𝐸) = (𝐸 ― 𝜀 ― 𝛼𝑒𝑉)2 + 𝛤2/4

(2)

energy level (ε) and the molecule-electrode coupling strength (Γ).30 In the transmission function, α is the dimensionless constant controlling the shift of the molecular level with respect to the electrode chemical potential under an external bias. We approximate this change by the first order variation (linear Stark shift) upon application of a bias. For α ≈ 0, the molecular energy level does not shift relative to the electrode chemical potential by the applied bias, and there is no rectification (Figure 4a). In case of α < 0, the molecular energy level tends to be pinned to the EGaIn top-electrode chemical potential and yields rectification (Figure 4b). When the Ga2O3/EGaIn tip is biased negatively relative to the AgTS substrate, the molecular energy level (HOMO of PAH) moves up, and consequently a larger fraction of the HOMO resonance peak enters the bias window, while at positive tip bias, the portion of HOMO peak inside the bias window is significantly decreased, resulting in a low current. Given that all the tested molecules possess the alkyl backbone of same length (nundecanethiolate) with large HOMO and LUMO gap (~8 eV)31 and the n-alkyl chain electronically isolates the PAH from the EF of silver bottom-electrode to similar extent, we take the molecular resonance energy in the transmission function from the frontier orbital of PAH. Indeed, DFT calculation of HSC11PYR molecule showed that HOMO and LUMO are positioned on the PYR group (Figure S29 in the Supporting Information). With the equations 1 and 2 we simulated I-V traces and fitted experimental data. Our focus was on fitting of rectification behavior rather than the absolute values of currents. From fitting values of α, ε and Γ were extracted. We initially took the transmission function modeled with a single


Page 13 of 24 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

Nano Letters

Figure 4. a) Energy level diagram illustrating the mechanism of rectification induced by Stark shift for a HOMO conducting molecular junction. b) Comparison of measured I-V trace for SC11PYR SAM with simulated one obtained by the Laudauer formula (Eq. 1). c) Comparison of measured |r-|-|V| for SC11PYR SAM with simulated one from the data (b). d) Comparison of HOMO and LUMO energy levels estimated by UPS with those extracted from the simulation.


Nano Letters 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

Lorentzian peaked at orbital energy of PAH, the closest to the EF of top-electrode. Ultraviolet photoelectron spectroscopy (UPS) analysis (Figure S1 in the Supporting Information) showed that HOMO was in proximity to the EF of Ga2O3/EGaIn for the PYR and BP molecules while LUMO for the NAP, PHE and ANT molecules. Figure 4b shows an exemplary fitting result for PYR. The experimentally measured asymmetric feature in I-V curves was roughly fit to the simulated one except current behavior at a high bias regime at +V, as pointed out in Figure 4b. The increase in current in the high bias regime at +V resulted in the regression of |r-| in the high bias regime as shown in Figure 4c. However, the single Lorentzian model failed to account for this, which led us to conclude that not only HOMO but also LUMO should be involved in charge tunneling. Thus, we repeated the simulation and fitting using a transmission equation corrected with double Lorentzian model (see the Supporting Information). Finally, we were able to achieve the better fitting for all the molecules, as shown in Figure S5-7 in the Supporting Information. Table 2 summarizes the fitting data based on the double Lorentzian model. The rectification of SC11PAH relied on the Stark shift parameter (α), as we hypothesized above. Values of ΓR (corresponding to the Ga2O3//PAH top interface) did not significantly differ across the tested molecules. It is noteworthy that further simulation study indicates varying Γ values while keeping α=0 did not cause significant rectification (Figure S8 in the Supporting Information). These findings led us to attribute the high rectification ratios for PYR and BP to the shift of molecular resonance energy. The molecular resonance energies extracted from junction measurements were largely perturbed, as compared to those of the identical SAMs characterized by UPS, leading to significant reduction of the HOMO-LUMO gap (Figure 4d). This gap reduction in charge transport could be explained by the image charge effect, frequently observed in molecular tunnel junctions.32-34 The still-existing some deviations of the modeling data from experimental results in Figure S5-6 of the Supporting


Page 14 of 24

Page 15 of 24 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

Nano Letters

Table 2. Summary of Data Fitting Using Double Lorentizan Model. PAH

α

ΕF-εHOMO (eV)a

ΕF-εLUMO (eV)b

ΕF-εHOMO (eV)c

ΕF-εLUMO (eV)c

ΓR (eV)c

ΓL (eV)d

BP

-0.40

1.2

1.9

0.65

1.25

0.035

0.003

PYR

-0.40

1.4

2.1

0.7

1.35

0.050

0.003

ANT

-0.080

1.9

1.3

1.0

0.9

0.020

0.003

PHE

-0.060

2.4

1.7

1.3

1.2

0.025

0.003

NAP

-0.030

2.1

2.1

1.3

1.25

0.025

0.003

aMeasured

by ultraviolet photoelectron spectroscopy (UPS). bEstimated by adding the optical gap to the HOMO energy measured by UPS. cObtained from fitting of experimentally measured J-V traces with the Landauer formula (Eq. 1) and double Lorentzian model-based transmission function (see the Supporting Information). dAll the molecules have the same S-Ag interface, and thus the identical literature value of ΓL35 was used for the sake of simplicity.


Nano Letters 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

Information are probably due to several factors. First of all, the transmission functions for the real molecular junctions do not have the form of the perfect Lorentzian as assumed in the modeling.36-37

Secondly, dependence of molecular resonance energy on the applied bias

might not be purely linear for the Stark shift.38 In addition, the applied bias can have an impact on the molecular orbital shape, resulting in the change of the coupling strengths of the molecular levels to the electrodes.39-40 We further conducted two separate control experiments to confirm the validity of our simulations. First, to test if the polarity of α values in our simulations above was relevant to rationalizing our experimental data, we repeated I-V simulations with the α values of same magnitude and opposite polarity. As shown in Figure S9 in the Supporting Information, we found that the change of polarity of α values reversed the direction of rectification and did not fit our experimental results. Second, to test if the double Lorentzian model based on HOMO and HOMO-1 of PAH, rather than the HOMO and LUMO could fit the experimental data, we repeated I-V simulations with the HOMO and HOMO-1 estimated from the DFT calculation (Figure S10 in the Supporting Information). The simulation results indicate that the inclusion of HOMO and HOMO-1 into the transmission function did not yield significant rectification and poorly fit the experimental data (see Figure S9 – S10 in the Supporting Information). The Stark shift is equivalent to the change of energy offset of ε with respect to EF upon application of a bias, as described in the transmission function. From a tunneling barrier point of view, relative comparison of barrier’s height in the absence and presence of external electric field should be translated into the degree of asymmetry in junction (here, defined quantitatively as α). Therefore, we further evaluated the Stark shift of molecular energy resonance by analyzing the measured J-V traces with transition voltage spectroscopy (TVS).41 When the transmission function can be well described by a Lorentzian form, TVS is a useful spectroscopic tool in molecular electronics and allows one to probe effective molecular


Page 16 of 24

Page 17 of 24 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

Nano Letters

Figure 5. Correlation between the Stark shift coefficient (|α|) with a) |ε-EF|/Vt values obtained from the experimental (UPS) and fitting data; and b) polarizability of model structure of PAHs. Values of Vt were measured at –V, and polarizability values were obtained from the literature.42


Nano Letters 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

resonance energy during operation through measurement of transition voltage (Vt), the minimum voltage obtained from Fowler-Nordheim plot. The Vt is proportional to |ε-EF| provided that |ε-EF|>>Vt where the proportionality constant is related to the asymmetry factor α. Therefore, the relative ratio of |ε-EF|/e to Vt could be an indication of symmetry/asymmetry of junction; a symmetric junction exhibits |ε-EF|/e ≈ Vt.43-44 Table S5 in the Supporting Information summarizes data of TVS analysis for our junctions. Plot of |ε-EF|/Vt values estimated from experimental (UPS) and fitting data as a function of α values in Figure 5a shows that the increase ratio (a factor of ~4.2 and ~2.5 for experimental and fitting data, respectively) of |ε-EF|/Vt from non-rectifying to rectifying PAHs is roughly correlated with that (a factor of ~5.0) of α. This indicates that the molecular energy level for the molecule with a large α is shifted to the substrate (AgTS) chemical potential by the finite bias voltage, resulting in the small Vt and large |ε-EF|/Vt values. Although little has been unveiled about what types of chemical features in organic molecules are associated with the Stark shift, another work3 involving Stark effect in different nanomaterials (e.g., quantum dots) led us to take into account polarizability of materials. The quantum-confined Stark effect in single CdSe quantum dot (with ~4 nm radius) has been characterized by highly polarizable excited state resulting from photoexcitation. Considering this stimulating work, intuitively, more polarizable PAH would undergo more charge redistribution upon application of a bias, stronger electrostatic interaction with the electrode being in contact, and hence larger Stark shift.45 Indeed, the plot of |α| against polarizability for the model structure of PAHs (Figure 5b) showed there is approximate correlation between the two values, suggesting the role of polarizability in the Stark shift-induced rectification. Further study is needed to understand detailed role of polarizability and others in the Stark shift. There was slight decrease by a factor of three in rectification ratio when the conjugation length was further increased from PYR to BP (Figure 2 and Table 1). We believe


Page 18 of 24

Page 19 of 24 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

Nano Letters

that this is probably due to changes in the degree of asymmetric charge distribution of orbital energy in biases of opposite polarities,46 and/or the distance between the PAH terminal group and the EGaIn top-electrode.47 Interestingly, the significant rectification occurred when a phenyl ring was added to the ANT to form the PYR. This could be rationalized with the change of electronic structure resulting from the structural modification. Our UPS data in Figure S1 in the Supporting Information indicated that the energy offset of HOMO with respect to the EF of EGaIn was reduced by ~36% upon the structural change from ANT (|εHOMO- ΕF|=~1.9) to PYR (~1.4). The reduced energy offset yielded the more accessible molecular orbital energy state and hence the remarkable rectification. Note that another rectifying PAH, the BP molecule, also showed the decreased energy offset value (~1.2) as compared to non-rectifying ANT molecule, whereas the other non-rectifying PAHs, the PHE and NAP molecules showed relatively large energy offset values of ~2.4 and ~2.1, respectively. Recently, a stimulating work by Xie et al.48 provided insight into Stark effect as a possible origin for achieving large rectification ratio. Namely, this work theoretically showed that Stark effect can induce significant rectification ratio up to ~500 in n-alkanethiolates at ±1.5 V. However, experiments showed the modest rectification ratio of ~1.5 once CPAFM was used to measure tunneling currents for n-alkanethiolates of different lengths (SCn where n=7, 8, 9, 10, 12). Such a marked difference in experimental and calculation results was attributed to non-ideal molecule-electrode contact in reality that screened the Stark effect. In our work, we demonstrated that the Stark effect-induced significant rectification up to ~170 at ±740 mV could be achieved by a chemical means that controls the structure of polycyclic aromatic hydrocarbons (PAHs) terminated in n-alkanethiolates. Conclusion


Nano Letters 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

In summary, we have demonstrated a chemical means with which to control the Stark shift of molecular energy resonance and create robust hydrocarbon-based molecular diodes with significant rectification ratios in a temperature-independent transport regime. The selfreferencing feature of rectification permits reliable access to manifestation of Stark effect in large-area molecular tunnel junctions. Indeed, we have drawn statistically meaningful inference about relationship of chemical structure with rectification and hence Stark effect, by taking advantage of the fine tunability of structurally simple PAHs to modulate conjugation length. Given that singling out factor(s) associated with Stark effect in many cases of molecular-scale devices is difficult to achieve, our results reported herein delineate the underlying origin and mechanism of Stark effect in molecular junction. We envisage that our approach can be implemented in other systems beyond PAHs not only to elucidate the origin of rectification but also to design other functional molecular-scale devices through control of Stark effect by chemical means.

ASSOCIATED CONTENT Supporting Information The Supporting Information is available free of charge on the ACS Publications website. Detailed experimental procedures, data of junction measurements, minor discussions and supplementary Figures and Tables (PDF)

AUTHOR INFORMATION Corresponding Author *[email protected] (T.K.); [email protected] (H.J.Y.) Notes The authors declare no competing financial interests.


Page 20 of 24

Page 21 of 24 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

Nano Letters

Contributions H. J. Yoon conceived the key idea, and S. J. Cho, G. D. Kong and H. J. Yoon designed the overall project. S. J. Cho, G. D. Kong and S. E. Byeon are responsible for synthesis, characterization and junction measurements. J. Park, G. D. Kong and S. Park performed CPAFM measurements. G. D. Kong, H. J. Yoon and T. K. Kim computed I-V curves. G. D. Kong, S. J. Cho, T. Kim and H. J. Yoon wrote the manuscript. ACKNOWLEDGMENT This research was supported by NRF of Korea (NRF-2017M3A7B8064518 for H.J.Y.; NRF2016R1D1A1B03931148 for T. K.). H.J.Y. also acknowledges support by the Future Research Grant of Korea University.

References (1)

Stark, J., Nature 1913, 92, 401.

(2)

Sie, E. J.; McIver, J. W.; Lee, Y.-H.; Fu, L.; Kong, J.; Gedik, N., Nat. Mater. 2015, 14,

290-294. (3)

Empedocles, S. A.; Bawendi, M. G., Science 1997, 278, 2114-2117.

(4)

Liu, Y.; Qiu, Z.; Carvalho, A.; Bao, Y.; Xu, H.; Tan, S. J. R.; Liu, W.; Neto, A. H. C.;

Loh, K. P.; Lu, J., Nano Lett. 2017, 17, 1970-1977. (5)

Akkerman, H. B.; Naber, R. C. G.; Jongbloed, B.; Hal, P. A. v.; Blom, P. W. M.;

Leeuw, D. M. d.; Boer, B. d., Proc. Natl. Acad. Sci. U. S. A. 2007, 104, 11161-11166. (6)

Sayed, S. Y.; Fereiro, J. A.; Yan, H. J.; McCreery, R. L.; Bergren, A. J., Proc. Natl.

Acad. Sci. U. S. A. 2012, 109, 11498-11503. (7)

Simeone, F. C.; Yoon, H. J.; Thuo, M. M.; Barber, J. R.; Smith, B.; Whitesides, G. M.,

J. Am. Chem. Soc. 2013, 135, 18131-18144. (8)

Xiang, D.; Wang, X. L.; Jia, C. C.; Lee, T.; Guo, X. F., Chem. Rev. 2016, 116, 4318-

4440. (9)

Weiss, E. A.; Chiechi, R. C.; Kaufman, G. K.; Kriebel, J. K.; Li, Z. F.; Duati, M.;


Nano Letters 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

Rampi, M. A.; Whitesides, G. M., J. Am. Chem. Soc. 2007, 129, 4336-4349. (10)

Song, P.; Yuan, L.; Roemer, M.; Jiang, L.; Nijhuis, C. A., J. Am. Chem. Soc. 2016,

138, 5769-5772. (11)

Thompson, D.; Nijhuis, C. A., Acc. Chem. Res. 2016, 49, 2061-2069.

(12)

Wan, A.; Jiang, L.; Sangeeth, C. S. S.; Nijhuis, C. A., Adv. Funct. Mater. 2014, 24,

4442-4456. (13)

Chiechi, R. C.; Weiss, E. A.; Dickey, M. D.; Whitesides, G. M., Angew. Chem., Int.

Ed. 2008, 47, 142-144. (14)

Nijhuis, C. A.; Reus, W. F.; Barber, J. R.; Dickey, M. D.; Whitesides, G. M., Nano

Lett. 2010, 10, 3611-3619. (15)

Li, Y.; Zolotavin, P.; Doak, P.; Kronik, L.; Neaton, J. B.; Natelson, D., Nano Lett.

2016, 16, 1104–1109. (16)

Capozzi, B.; Xia, J. L.; Adak, O.; Dell, E. J.; Liu, Z. F.; Taylor, J. C.; Neaton, J. B.;

Campos, L. M.; Venkataraman, L., Nat. Nanotechnol. 2015, 10, 522-527. (17)

Weiss, E. A.; Kaufman, G. K.; Kriebel, J. K.; Li, Z.; Schalek, R.; Whitesides, G. M.,

Langmuir 2007, 23, 9686-9694. (18)

Cademartiri, L.; Thuo, M. M.; Nijhuis, C. A.; Reus, W. F.; Tricard, S.; Barber, J. R.;

Sodhi, R. N. S.; Brodersen, P.; Kim, C.; Chiechi, R. C.; Whitesides, G. M., J. Phys. Chem. C 2012, 116, 10848−10860. (19)

Yoon, H. J.; Liao, K. C.; Lockett, M. R.; Kwok, S. W.; Baghbanzadeh, M.;

Whitesides, G. M., J. Am. Chem. Soc. 2014, 136, 17155-17162. (20)

Kong, G. D.; Yoon, H. J., J. Electrochem. Soc. 2016, 163, G115-G121.

(21)

Quéré, D., Annu. Rev. Mater. Res. 2008, 38, 71-99.

(22)

Folkers, J. P.; Laibinis, P. E.; Whitesides, G. M., J. Adhesion Sci. Technol. 1992, 6,

1397-1410. (23)

Schoenfisch, M. H.; Pemberton, J. E., J. Am. Chem. Soc. 1998, 120, 4502–4513.

(24)

Widrig, C. A.; Chung, C.; Porter, M. D., J. Electroanal. Chem. 1991, 310, 335-359.

(25)

Walczak, M. M.; Popenoe, D. D.; Deinhammer, R. S.; Lamp, B. D.; Chung, C.;

Porter, M. D., Langmuir 1991, 7, 2687–2693. (26)

Mastroa, M. A.; Kuramata, A.; Calkins, J.; Kim, J.; Renf, F.; Pearton, S. J., ECS J.

Solid State Sci. Technol. 2017, 6, P356-P359. (27)

Yu, Y.-J.; Zhao, Y.; Ryu, S.; Brus, L. E.; Kim, K. S.; Kim, P., Nano Lett. 2009, 9,

3430-3434.


Page 22 of 24

Page 23 of 24 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

Nano Letters

(28)

Reus, W. F.; Thuo, M. M.; Shapiro, N. D.; Nijhuis, C. A.; Whitesides, G. M., ACS

Nano 2012, 6, 4806-4822. (29)

Byeon, S. E.; Kim, M.; Yoon, H. J., ACS Appl. Mater. Interfaces 2017, 9, 40556-

40563. (30)

Paulsson, M.; Datta, S., Phys. Rev. B 2003, 67, 241403.

(31)

Holmlin, R.; Haag, R.; Chabinyc, M. L.; Ismagilov, R. F.; Cohen, A. E.; Terfort, A.;

Rampi, M. A.; Whitesides, G. M., J. Am. Chem. Soc. 2001, 123, 5075-5085. (32)

Kubatkin, S.; Danilov, A.; Hjort, M.; Cornil, J.; Brédas, J.-L.; Stuhr-Hansen, N.;

Hedegård, P.; Bjørnholm, T., Nature 2003, 425, 698-701. (33)

Perrin, M. L.; Verzijl, C. J. O.; Martin, C. A.; Shaikh, A. J.; Eelkema, R.; Esch, J. H.

v.; Ruitenbeek, J. M. v.; Thijssen, J. M.; Zant, H. S. J. v. d.; Dulić, D., Nat. Nanotechnol. 2013, 8, 282-287. (34)

Kaasbjerg, K.; Flensberg, K., Nano Lett. 2008, 8, 3809–3814.

(35)

Garrigues, A. R.; Yuan, L.; Wang, L.; Mucciolo, E. R.; Thompon, D.; Barco, E. d.;

Nijhuis, C. A., Sci. Rep. 2016, 6, 26517. (36)

Kim, T.; Darancet, P.; Widawsky, J. R.; Kotiuga, M.; Quek, S. Y.; Neaton, J. B.;

Venkataraman, L., Nano Lett. 2014, 14, 794–798. (37)

Quek, S. Y.; Choi, H. J.; Louie, S. G.; Neaton, J. B., Nano Lett. 2009, 9, 3949–3953.

(38)

Darancet, P.; Widawsky, J. R.; Choi, H. J.; Venkataraman, L.; Neaton, J. B., Nano

Lett. 2012, 12, 6250–6254. (39)

Capozzi, B.; Low, J. Z.; Xia, J.; Liu, Z.-F.; Neaton, J. B.; Campos, L. M.;

Venkataraman, L., Nano Lett. 2016, 16, 3949–3954. (40)

Liu, Z.-F.; Neaton, J. B., J. Phys. Chem. C 2017, 121, 21136–21144.

(41)

Beebe, J. M.; Kim, B.; Frisbie, C. D.; Kushmerick, J. G., ACS Nano 2008, 2, 827-

832. (42)

Malloci, G.; Joblin, C.; Mulas, G., Chem. Phys. 2007, 332, 353–359.

(43)

Chen, J.; Markussen, T.; Thygesen, K. S., Phys. Rev. B 2010, 82, 121412.

(44)

Markussen, T.; Chen, J.; Thygesen, K. S., Phys. Rev. B 2011, 83, 155407.

(45)

Scott, R.; Achtstein, A. W.; Prudnikau, A. V.; Antanovich, A.; Siebbeles, L. D. A.;

Artemyev, M.; Woggon, U., Nano Lett. 2016, 16, 6576–6583. (46)

Ding, W.; Negre, C. F. A.; Vogt, L.; Batista, V. S., J. Chem. Theory Comput. 2014,

10, 3393−3400. (47)

Taylor, J.; Brandbyge, M.; Stokbro, K., Phys. Rev. Lett. 2002, 89, 138301.


Nano Letters 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

(48)

Xie, Z.; Bâldea, I.; Frisbie, C. D., Chem. Sci. 2018, 9, 4456-4467.

ToC Graphic


Page 24 of 24

Molecularly Controlled Stark Effect Induces ... - ACS Publications

Recommend Documents