• Home
  • Sitemap
  • Contact us
Article View

Research Paper

Applied Science and Convergence Technology 2023; 32(4): 89-92

Published online July 30, 2023

https://doi.org/10.5757/ASCT.2023.32.4.89

Copyright © The Korean Vacuum Society.

Solid to Quasi-Liquid Phase Transition of Submonolayer Pentacene

Eunji Sima , b , c , Mihyun Yanga , Ha Eun Choa , d , Somang Kooa , Seong Chu Limb , c , ∗ , and Kyuwook Ihma , ∗

aBeamline Research Division, Pohang Accelerator Laboratory, Pohang 37673, Republic of Korea
bDepartment of Energy Science, Sungkyunkwan University, Suwon 16419, Republic of Korea
cDepartment of Smart Fabrication Technology, Sungkyunkwan University, Suwon 16419, Republic of Korea
dDepartment of Physics, University of Ulsan, Ulsan 44610, Republic of Korea

Correspondence to:seonglim@skku.edu, johnet97@postech.ac.kr

Received: May 2, 2023; Revised: June 29, 2023; Accepted: July 13, 2023

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (https://creativecommons.org/licenses/by-nc-nd/4.0/) which permits non-commercial use, distribution and reproduction in any medium without alteration, provided that the original work is properly cited.

Based on surface charge states, the thermodynamics of pentacene molecules exhibit intriguing structural transitions with various temperatures. On an SiO2 surface, the mean tilt angle decreases with increasing temperature, whereas three different structural regimes form on the Au surface. To understand these contrasting temperature dependencies of molecular orientation on the two surfaces, a model wherein the tilt angle α has two stable states is proposed in this study. Consequently, the pentacene crystals undergoes a phase transition at a relatively low temperature of 350 K, thus indicating a hidden factor governing the morphology of molecular crystals, and providing insights into the thermal behavior of physisorbed organic molecules.

Keywords: Surface melting, Phase transition, Organic crystal, Pentacene

Physically adsorbed organic films have attracted considerable attention owing to their potential applications as inhibitor layers for area-selective deposition (ASD) [15] and electronic devices [6,7]. Intermolecular and molecule-substrate interactions are crucial in determining the degree of crystallinity of organic films, which is directly related to the area selectivity in ASD or charge mobility in electronic devices [8,9]. However, because these phenomena are based on weak van der Waals interactions, the crystallinity of molecular film formed on the surface is affected by small external stimuli and sensitive changes. Particularly, the surface charge density is a sensitive determinant of the crystallinity of a molecular film, and its understanding is crucial for improving the selectivity of area-selective film formation.

Pentacene is a model system for organic crystal films based on van der Waals interactions. Depending on their substrate, pentacene molecular films exhibit dramatic variations in their structural ordering. The charge-carrier mobility is directly dependent on the crystallinity of the grown organic film, whose reported value is in the range of 0.1–58.0 cm2V–1 s–1 [10]. Studies on the structural ordering of pentacene films under various conditions have demonstrated that the initial arrangement of pentacene molecules on the surface acts as a nucleation seed. [1113]. On the crystal structure surface, the pentacene molecule exhibits two orientations, i.e., standing-up or lying-down, which are observed when the angle α between the longitudinal axis of pentacene molecule and surface is greater than 70° or less than 10°, respectively [3].

In the early stage of pentacene growth, the standing-up orientation is important for obtaining highly ordered crystalline structures, owing to the lower surface energy of (001) crystal plane compared with that of the other planes [14,15]. Therefore, the growth directions of (122), (011), and (121) planes can be varied to (001) [13].

On the substrate, pentacene molecules interact with surface electrons and neighboring molecules with energies of Ems and Emm, respectively.

At a given temperature during the early stage of crystallization, the energy difference between Ems and Emm is crucial in determining the orientation α of the pentacene molecules [3,12,16]. However, as the substrate temperature increases, the variation in average orientation ⟨α⟩ is independent of the type of growth substrate [7,13,17].

In this study, the near-edge X-ray absorption fine structure (NEXAFS) spectroscopy was used to analyze the temperature dependence of low-coverage pentacene molecules on silicon dioxide (SiO2) and gold (Au) surfaces. To explain the observed variations in α, a thermodynamic model was developed for the two orientations (standing-up and lying-down) of pentacene. Using this model, the intermolecular interactions were abruptly weakened and induced by a phase transition of the pentacene crystal from solid to quasi-liquid at 350 K.

The sample preparation and spectroscopic measurements were performed at the 4D NEXAFS beamline of the Pohang Accelerator Laboratory. Si with a thermal oxide layer (100 nm) and polycrystalline Au of dimensions 5 × 10 cm2 were simultaneously mounted on a grounded molybdenum holder, which was thermally linked to a resistive heater wrapped in phosphor-boron-nitride. For degassing, the substrates were annealed in an ultrahigh vacuum chamber (base pressure: 4 × 10−10 Torr) at 670 K for 48 h. Prior to sample heating, surface contaminants were cleaned using ion sputtering for 10 min (Ar+, 3 KeV), and the cleanliness of the substrate surfaces was confirmed by using X-ray photoemission spectroscopy (XPS).

Pentacene (Aldrich, purified by gradient sublimation) was thermally evaporated onto the SiO2 and Au surfaces at room temperature (RT) and deposition rate of 0.05 Ås−1. The deposition rate and thickness of the pentacene layer were monitored using a crystal-oscillator thickness monitor (Inficon XTC). The temperature of the sample was regulated within ±0.1 K. The partial electron yield (PEY)-mode NEXAFS was measured by counting the carbon KLL Auger electrons.

The temperature-dependent desorption patterns of 100 nm pentacene films grown on SiO2 and Au were observed using optical microscopy. As the temperature increased, the inhomogeneous and homogenous desorption of pentacene film was observed on SiO2 and Au surfaces, respectively (Fig. 1). In the former case, the desorption was initiated by the defect sites denoted by the red arrows in Fig. 1b, which continued to evolve gradually around the defect centers [Figs. 1(c) and 1(d)]. Therefore, the intermolecular interaction energy (Emm) or cohesive energy of neighboring molecules in the film is greater than the molecule-substrate interaction energy (Ems). Because the molecules at the defect site are outside their normal positions, their binding energies with neighboring molecules is considerably weakened. In the absence of pentacene, the molecules were desorbed on the substrate; hence, the interaction energy EmmD between the molecules at the defect site can be described as Ems < EmmD < Emm [14,15]. If the thermal energy falls within the range of EmmDEmm, desorption initiates from the defect center owing to the unrestricted mobility of molecules at the defective site, which are detached from the substrate and neighboring molecules. Consequently, the desorption process exhibits inhomogeneity, similar to the observed behavior of pentacene on SiO2.

Figure 1. Optical images of inhomogeneous desorption of pentacene molecules on SiO2 surface. Pentacene thin film (100 Å) on SiO2 surface at: (a) room temperature (RT), (b)–(d) after annealing at 570 K for (b) 0 s, (c) 10 s (c), and (d) 20 s. Homogeneous desorption of pentacene molecules on Au surface. Pentacene thin film (100 Å) on Au at: (e) room temperature, and (f) 640 K after complete desorption of molecules for 60 s.


Compared with the SiO2 surface, the Au surface contains a larger number of crystal defects; hence, the pentacene molecules exhibit homogeneous desorption from the Au surface [13]. This can be attributed to the interaction energies of pentacene molecules formed on Au, which are in the order EmmD < Emm < Ems. Herein, desorption occurred when the thermal energy of molecule was greater than its binding energy with the substrate, thus implying that thermal desorption cannot discern the intermolecular interaction energy at the defective sites [Figs. 1(e) and 1(f)].

To analyze the molecular orientation of sub-monolayer, a 7 Å-thick layer (approximately 0.5 molecular layers) of pentacene molecules was simultaneously deposited on cleaned SiO2 and Au. Furthermore, NEXAFS measurements were performed in vacuum [14,15].

The NEXAFS spectra of an as-prepared sample were recorded at incident angles θi of 20° (red) and 90° (blue) [Figs. 2(a) and 2(d)] [18, 19]. The overall line shape of the NEXAFS spectra of pentacene on the substrates is similar to that of gas-phase pentacene, thus indicating that pentacene molecules do not chemically bond with SiO2 on the Au surface [20]. In the Fig. 2(a), the peaks π1* and π2* at 284.7 and 287 eV, respectively, are attributed to the resonant electrons generated during the relaxation process of excited electron from C 1s to π* antibonding orbital. Furthermore, the broad features above 290 eV originate from σ* resonances [16].

Figure 2. (a) NEXAFS spectra of a pentacene molecules (7 Å) grown on an SiO2 surface at photon incident angles of 20∘ (red line) and 90∘ (blue line). The inset of (a) shows experimental geometry of polarization-dependent NEXAFS spectroscopy with photon incident angle θi when the molecule has tilt angles α and β with the surface. Enlarged NEXAFS spectrum in the 280–300 eV region of pentacene on SiO2 surface at: (b) RT and (c) 410 K. (d)–(f) NEXAFS spectra of pentacene (7 Å) on the Au surface at: (d) RT, (e) 370 K, and (f) 420 K; here, the photon incident angles are 20∘ (red line) and 90∘ (blue line).

Because the π* orbital is strongly localized perpendicular to the plane of the carbon ring of pentacene molecule, the π* resonance peak strongly depends on the polarization of electric field. Therefore, the angle-dependent NEXAFS spectra can be used to probe the orientation of pentacene molecules [Fig. 2(a)]. The peak intensity Iπ* depends on the molecular tilt angle ⟨α⟩ and photon incident angle θ as Iπ* = P/3[1+(3 cos2 θ−1)(3 cos2(π/2− α))−1/2], where P is the polarization ratio of the incident photon and β is the azimuthal rotation about the longitudinal axis of molecule [Fig. 2(a)] [21]. The p-value of the used beamline was 0.8 [Fig. 2(a)].

To evaluate the average ⟨α⟩ values at each temperature, the NEXAFS spectra at θi = 20, 45, 70, and 90° of photon incidence were acquired. At RT, pentacene on SiO2 surface exhibited strong angular dichroism of the π* peak at 287 eV [Fig. 2(b)]. Based on the relationship Iπ* (θ, α), ⟨α⟩ of pentacene molecule on SiO2 surface at RT was estimated to be 72°, which is close to the tilt angle of crystal phase on SiO2 in the stand-up orientation [16]. As the temperature increased to above 410 K, the strong dichroic response nearly disappeared [Fig. 2(c)], thus implying that the initial molecular ordering vanishes. This agrees with previous reports on the broadening of X-ray diffraction patterns of pentacene films grown at elevated temperatures [17,22].

Figures 2(d)-(f) show three different dichroic responses observed on the Au surface. Compared with the pentacene molecule on Si surface, the π1* transition peak located at 284.7 eV disappeared for the Au surface [Fig. 2(d)]. This occurs because of the spatial distribution of the π1* orbital of pentacene, which predominantly surrounds the outer carbon atoms of the molecule, thus leading to its direct interaction with the valence electrons of the surface Au atoms. As the π1* orbital interacts with Au, the electron orbital of pentacene molecule localizes, thereby exhibiting an electronic structure similar to that of poly(p-phenylene) [2325].

At RT, the intensity of π2* peak observed at θi = 20° was greater than that at θi = 90° [Fig. 2(d)], thus corresponding to average ⟨α⟩ 19 ± 2° for the angle induced by β and indicating the formation of an adlayer. Pentacene molecules in the first monolayer lie flat (α ≈ β ≈ 0°) on the Au surface, whereas those in the adlayer are slightly rotated (α ≈ 0° and β ≈ 20°) [13]. As the temperature of the substrate increased, the tilt angle of the adlayer approached 45° [Fig. 2(e)], which is confirmed by the previous NEXAFS results [13]. When the temperature is increased to 420 K, the π2* peak at θi = 20° is greater than the intensity measured at θi = 90°, indicating the decrement in molecular tilt angel (estimated ⟨α⟩ = 89 ± 3°) [Fig. 2(f)].

To describe the thermodynamic behavior of ⟨α⟩ pentacene molecules on SiO2 and Au surfaces, a model with the following two stable states of α was considered: standing-up (α = 74°) with E = Est and lyingdown (α = 0°) with E = Ely (Fig. 3) [12]. The energies of surface molecules in these two geometries can be described as Est = E0EmmE1 + fkBT/2 and Ely = E0EmsE2 + fkBT /2, respectively, where E0 is the energy of molecule in the gas phase. For molecules in the standing-up geometry, Emm and E1 are the interaction energies with the neighboring molecules and surface atoms, respectively; Ems and E2 are the interaction energies with the substrate and two molecules in the lying-down orientation, respectively; f is the degree of freedom of surface pentacene molecule; kB is the Boltzmann constant; and T is the absolute temperature. Because E1 and E2 are infinitesimally less than Emm and Ems [12,14,15], they are neglected. In this model, the partition function can be written as follows.

Figure 3. Thermodynamic behavior of average tilt angles ⟨α⟩ for various values of EmsEmm. EmsEmm < 0: pentacene molecules on SiO2 surface; EmsEmm > 0: pentacene molecules on Au surface. Illustration of possible intermolecular interactions Emm and molecule-substrate interaction Ems.

Ζ=eE0 κBT+f2 eEmsκBT+eEmmκBT .

Thereafter, the averaged tilt angle of ⟨α⟩ can be obtained as

α=1ΖiαieEiκBT =αlyeΔE κB T +αsteΔE κB T +1 ,

where ΔE = EmsEmm. The curves were calculated for several selected values of ΔE (Fig. 3). Herein, two molecular geometries αst = 74° and αly = 0° were used for the two orientations of pentacene molecules. At sufficiently low system temperature, i.e., kBT ≪ |ΔE|, if ΔE < 0 similar to pentacene on SiO2 surface, ⟨α⟩ exhibits a standingup geometry. Contrastingly, if ΔE < 0 as on the Au surface, ⟨α⟩ exhibits a lying-down geometry in the low energy regime (Fig. 3). When the temperature of the system increases such that kBT ≫ |ΔE|, ⟨α⟩ for the two substrates approaches 37°, thus corresponding to (αst +αly)/2. Therefore, for sufficiently low or high temperature limits, the model describes the thermal behavior of ⟨α⟩ on the insulating and metal surfaces, as observed using NEXAFS spectroscopy (Fig. 2) [13,17].

Detailed thermal behavior of ⟨α⟩ were acquired using the angular dichroism of the NEXAFS spectra of pentacene molecules at 7 Å on the SiO2 and Au surfaces [Figs. 4(a) and 4(b)].

Figure 4. Tilt angles acquired from the polarization-dependent partial electron yield (PEY)–NEXAFS spectra of a 7 Å-thick pentacene layer on (a) SiO2 (circles) and (b) Au (squares) surfaces as a function of temperature; blue line: best fit of calculated curves ⟨α⟩ of pentacene molecules on SiO2 and Au surfaces with parameters of αst = 74° and αly = 0°, respectively. (c) ΔE values used to obtain the best fitted curves represented by blue lines in (a) and (b).


The experimental results differed significantly from the smooth transitions predicted by calculations (Fig. 3), thus indicating the absence of an unknown geometric state of α because the addition of new states in the calculation does not cause the abrupt transition observed in the experimental results. However, the abrupt variation in ΔE at a specific temperature explains the experimental results well.

The best fit of the temperature behavior of ⟨α⟩ for the two substrates is indicated by the solid blue lines [Figs. 4(a) and 4(b)]. Notably, the ΔE values used for this process decreased sharply to zero at approximately 350 K. Because the ΔE value represents the energy difference between the two possible geometries of pentacene molecules on each substrate, the removed energy difference, i.e., ΔE ≈ 0 eV, indicates that molecular melting occurs near 350 K. Furthermore, ΔE of pentacene molecules on the SiO2 surface was ≈ 1.2 eV, which commensurates with the previously reported value of 1.0 eV [12,14,15].

On the Au surface, excluding the contribution of β near RT, ⟨α⟩ was estimated to be approximately 0° [Fig. 4(b)] [13]. As the temperature increased, ⟨α⟩ increased until 380 K and sharply decreased to approximately zero [Fig. 4(b)] owing to the desorption of pentacene adlayer molecules, which is consistent with the findings of a previous report [26]. Above 380 K, the carbon 1s intensity decreased, thus facilitating the desorption of molecule.

After the desorption of adlayer molecules, the NEXAFS spectrum revealed the information regarding the direct binding of pentacene molecules to the Au surface. The desorption of a molecule from the adlayer exposes a flat-lying molecule in the Au monolayer with β = 0. The binding energy of a pentacene molecule bonded directly to the Au surface is 1.82 eV, and such molecules exhibit no desorption in the temperature range of this experiment [26]. Similar to the SiO2 surface, the fitted curve was derived by adjusting the value of ΔE [Figs. 4(b) and 4(c)], thereby revealing information about the pentacene molecules in the adlayer formed on Au substrate, which melted near 350 K.

This study investigated the thermal behaviors of pentacene molecules on SiO2 and Au surfaces using soft X-ray absorption spectroscopy. The initial configuration of these molecules can be explained using two main interactions: intermolecular and molecular-surface. Because the SiO2 and Au surfaces have significant variations in valence electron densities, they exhibit opposite initial molecular arrangements. Furthermore, a model with two energy states was proposed to adequately describe the thermodynamic behavior of pentacene molecules on the two surfaces by considering an additional phase transition. These results describe the morphologies of pentacene observed at various temperatures. The energy difference ΔE between the two geometries of pentacene molecules was removed at 350 K, thus implying the occurrence of a phase transition from solid to quasi-liquid. These findings provide deeper insights into the initial molecular arrangement on surfaces and provide a strategy for growing high-quality molecular crystals based on the surface charge states.

This study was supported by the National Research Foundation of Korea (NRF) (Grant No. NRF-2018R1D1A1B07043155, 2022R1F1A1 068192) from the Ministry of Science, ICT, and Future Planning, Korea. The experiments at the 4D beamline of the PLS–II were supported by MSIP, R. O., Korea.

  1. N. E. Richey, C. Paula, and S. F. Bent, J. Chem. Phys. 152, 040902 (2020).
    Pubmed CrossRef
  2. Y. S. Yang, C. Beekman, W. Siemons, C. M. Schlepütz, N. Senabulya, R. Clarke, and H. M. Christen, APL Mater. 4, 036106 (2016).
    CrossRef
  3. G. E. Thayer, J. T. Sadowski, F. M. Heringdorf, T. Sakurai, and R. M. Tromp, Phys. Rev. Lett. 95, 256106 (2005).
    Pubmed CrossRef
  4. L. Miozzo, A. Yassar, and G. Horowitz, J. Mater. Chem. 20, 2513 (2010).
    CrossRef
  5. A. Cabrero-Vilatela, J. A. Alexander-Webber, A. A. Sagade, A. I. Aria, P. Braeuniger-Weimer, M. -B. Martin, R. S. Weatherup, and S. Hoffman, Nanotechnology 28, 485201 (2017).
    Pubmed CrossRef
  6. M. Halik, et al, Nature 431, 963 (2004).
    Pubmed CrossRef
  7. J. Wu, M. Agrawal, H. A. Becerril, Z. Bao, Z. Liu, Y. Chen, and P. Peumans, ACS Nano 4, 43 (2010).
    Pubmed CrossRef
  8. C. A. Schoenbaum, D. K. Schwartz, and J. W. Medlin, Acc. Chem. Res. 47, 1438 (2014).
    Pubmed CrossRef
  9. H.-L. Cheong, Y.-S. Mai, W.-Y. Chou, L.-R. Chang, and X.-W. Liang, Adv. Funct. Mater. 17, 3639 (2007).
    CrossRef
  10. O. D. Jurchescu, J. Baas, and T. T. M. Palstra, App. Phys. Lett. 84, 3061 (2004).
    CrossRef
  11. R. Lassnig, M. Hollerer, B. Striedinger, A. Fian, B. Stadlober, and A. Winkler, Org. Electron. 26, 420 (2015).
    Pubmed KoreaMed CrossRef
  12. D. Choudhary, P. Clancy, R. Shetty, and F. Escobedo, Adv. Funct. Mater. 16, 1768 (2006).
    CrossRef
  13. D. Käfer, L. Ruppel, and G. Witte, Phys. Rev. B 75, 085309 (2007).
    CrossRef
  14. J. E. Northrup, M. L. Tiago, and S. G. Louie, Phys. Rev. B 66, 121404 (2002).
    CrossRef
  15. S. Verlaak, S. Steudel, P. Heremans, D. Janssen, and M. S. Deleuze, Phys. Rev. B 68, 195409 (2003).
    CrossRef
  16. K. Ihm, B. Kim, T.-H. Kang, K.-J. Kim, M. H. Joo, T. H. Kim, S. S. Yoon, and S. Chung, App. Phys. Lett. 89, 033504 (2006).
    CrossRef
  17. J. Lee, J. H. Kim, and S. Im, J. App. Phys. 95, 3733 (2004).
    CrossRef
  18. O. McDonald, A. A. Cafolla, D. Carty, G. Sheerin, and G. Hughes, Surf. Sci. 600, 3217 (2006).
    CrossRef
  19. F.-J. M. Heringdorf, M. C. Reuter, and R. M. Tromp, Nature 412, 517 (2001).
    Pubmed CrossRef
  20. M. Alagia, C. Baldacchini, M. G. Betti, F. Bussolotti, V. Carravetta, U. Ekström, C. Mariani, and S. Stranges, J. Chem. Phys. 122, 124305 (2005).
    Pubmed CrossRef
  21. J. Stöhr, NEXAFS Spectroscopy (Springer Berlin, 1992).
    CrossRef
  22. C. Kim, K. Bang, I. An, C. J. Kang, Y. S. Kim, and D. Jeon, Curr. App. Phys. 6, 925 (2006).
    CrossRef
  23. T. Yokoyama, K. Seki, I. Morisada, K. Edamatsu, and T. Ohta, Phys. Scr. 41, 189 (1990).
    CrossRef
  24. K. Lee and J. Yu, Surf. Sci. 589, 8 (2005).
    CrossRef
  25. K. Ihm, S. Chung, T.-H. Kang, and S.-W. Cheong, App. Phys. Lett. 93, 141906 (2008).
    CrossRef
  26. C. B. France, P. G. Schroeder, J. C. Forsythe, and B. A. Parkinson, Langmuir 19, 1274 (2003).
    CrossRef

Share this article on :

Stats or metrics