| Home | E-Submission/Review | Sitemap | Editorial Office |  
top_img
Korean Journal of Metals and Materials > Volume 63(3); 2025 > Article
Lee, Jyothi, Kim, Yun, and Park: Electrical Characteristics of Time-Dependent Flexible Organic Ferroelectric Field-Effect Transistors

Abstract

Organic ferroelectric field-effect transistors (Fe-FETs) are highly attractive for memory device applications because of their ability to read data without causing damage (nondestructive readout) and their excellent retention capabilities. These attributes make organic Fe-FETs an appealing alternative to their inorganic counterparts, which, while demonstrating superior memory performance for practical applications, still have significant challenges. This study investigates the impact of solvent selection and dielectric properties on the electrical characteristics, memory window, and long-term stability of flexible organic Fe-FETs. To this end, the electrical characteristics of devices fabricated using various organic solvents (dimethylformamide (DMF) and 2-butanone/ethyl methyl ketone (MEK)) were compared, and the results revealed that devices using polyvinylidene fluoride trifluoro ethylene (P(VDF-TrFE)) and Polyvinylidene fluoride hexafluoro-propylene (P(VDF-HFP)) in DMF exhibited higher drain currents with a larger memory window. This can be attributed to the relatively smoother surface morphology of ferroelectric films prepared with DMF. The time-dependent electrical properties of various flexible organic devices were investigated and the memory window of flexible Fe-FETs with a P(VDF-TrFE) dielectric layer exhibited excellent electrical stability. For the device with the P(VDF-HFP) dielectric, extracted memory windows (ΔVth) decreased from 25.1 to 19.0 V after two weeks. The electrical properties of P(VDF-TrFE) and P(VDF-HFP) dielectric layers exhibited entirely different tendencies depending on the solvent polarity, with the relatively simple molecular structure of the P(VDF-TrFE) film demonstrating higher electrical reliability. These results underscore the significant impact of solvent selection and the properties of the dielectric–semiconductor interface on the long-term reliability and operational efficiency of flexible Fe-FET devices.

1. INTRODUCTION

Over the last few decades, organic nonvolatile memory devices have attracted significant attention because of their low cost and large-area fabrication[1-6]. Nonvolatile organic memory devices also have low power consumption, and combining them with other circuit machinery can help realize nonvolatile random-access memory (RAM)[7]. In addition, such devices have been explored in attempts to develop flexible electronics. Organic materials have superior mechanical robustness and can be processed at very low annealing temperatures, which is essential for realizing mechanically flexible electronic devices[8-12]. Numerous nonvolatile organic memory devices have been reported, including resistive memory[13], floating-gate transistors[14], and ferroelectric field-effect transistor devices[15,16]. Organic ferroelectric field-effect transistors (Fe-FETs) are highly attractive because of their nondestructive data readout and good retention properties[17]. Although inorganic Fe-FETs have achieved better memory performance in practical applications, several important parameters, including the nature of the inorganic material and its high fabrication temperatures, remain challenges to implementing advanced memory devices.
In this study, flexible organic Fe-FETs were fabricated using polyvinylidene fluoride trifluoro ethylene (P(VDF-TrFE)) and Polyvinylidene fluoride hexafluoropropylene (P(VDF-HFP)) as gate dielectric materials. Polyvinylidene fluoride (PVDF) copolymers have been studied extensively because of their high retention time, improved storage capacity, and large remnant polarization (Pr)[18-20]. However, flexible organic Fe-FETs using PVDF copolymers as gate dielectric materials suffer from critical electrical performance problems.
Nonetheless, better performance has been achieved when these devices were fabricated on glass/silicon substrates. With plastic substrates the poor fabrication procedure and low water vapor transition rate can lead to Fe-FETs with poor electrical performance[21-23]. Thus, a fundamental knowledge of the electrical properties of flexible Fe-FETs is a prerequisite to their use in flexible electronic applications.
In this study, the basic electrical characteristics of flexible organic Fe-FETs fabricated using P(VDF-TrFE) and P(VDF-HFP) were investigated. The electrical characteristics of the flexible memory devices fabricated using different organic solvents (i.e., dimethylformamide (DMF) and 2-butanone/ethyl methyl ketone (MEK)) were compared. The results revealed that devices using P(VDF-TrFE) and P(VDF-HFP) in DMF exhibited higher drain currents with larger memory windows, which can be attributed to the smooth surface morphologies of the ferroelectric films from the DMF solvent compared to those from the MEK solvent. Further, the time-dependent electrical properties of the flexible devices were studied using various dielectric layers. Devices with P(VDF-TrFE) from both DMF and MEK showed an increased memory window with time. However, in the P(VDF-HFP) devices, the memory window did not significantly improve with storage time, and the HFP devices did not degrade with storage time.

2. EXPERIMENTAL

Bottom-gate top-contact Fe-FETs were fabricated on polyimide (PI) substrates. Before fabrication, the PI substrates were cleaned with acetone, isopropanol, and deionized water in ultrasonic water, and subsequently dried with nitrogen gas. For better absorption of the P(VDF-TrFE) and P(VDF-HFP) dielectric layers and Al gate electrode, 4 wt % of CPVP was spin coated on the PI substrate at 2000 rpm for 30 s and the coated substrate was baked at 180 °C for 1 h. A 50 nm Al gate electrode was thermally deposited onto the adhesion layer at a deposition rate of 0.6 A. Thereafter, 7 and 5 wt % of P(VDF-TrFE) and 10 wt % of P(VDF-HFP) solutions were prepared in DMF and MEK solvents. The prepared solutions were spin coated onto an Al gate electrode at 2000 rpm for 35 s to form a dielectric layer. The coated substrates were annealed at 140 °C for 2 h to enhance their ferroelectric properties. A 50-nm-thick pentacene semiconducting layer was thermally deposited onto the dielectric layers at a rate of 0.2 A/s. Finally, 50-nm-thick Au source/drain electrodes were formed via thermal evaporation. The channel lengths and widths of the fabricated devices were 100 and 800 μm, respectively. A schematic of the fabricated device is shown in Fig 1.

3. RESULTS AND DISCUSSION

The basic electrical characteristics of the resulting flexible Fe-FETs were analyzed to determine the effect of the solvent on the drain currents and hysteresis loops. Fig 2 shows the transfer and output characteristics of Fe-FETs fabricated with P(VDF-TrFE) films from the DMF and MEK solvents. The hysteresis loops from the transfer characteristics were extracted by supplying a bidirectional gate voltage from +30 to −30 V (transfer down) and from −30 to +30 V (transfer up), as shown in Fig 2 (a, c). The devices exhibited clockwise hysteresis loops (indicated by arrows) because of the polarization properties of the ferroelectric material[24]. The extracted memory windows (ΔVth) in P(VDF-TrFE) from the DMF and MEK solvents were 18.0 and 9.2 V, respectively. A strong hysteresis loop with a large memory window was observed in the device with the DMF film.
The memory window in the Fe-FETs is an important parameter. Fig 2 (b, d) show the output characteristics of the fabricated Fe-FETs with P(VDF-TrFE) from the DMF and MEK solvents, respectively. The extracted drain current values of devices fabricated with the DMF and MEK films were −0.27 and −0.0035 μA, respectively. Devices with the DMF film showed higher drain currents than those with the MEK film. Flexible Fe-FETs were fabricated using P(VDF-HFP) films from the DMF and MEK solvents.
Fig 3 shows the transfer and output characteristics of Fe-FETs fabricated with P(VDF-HFP) from the DMF and MEK solvents. The extracted memory window (ΔVth) and drain currents of the P(VDF-HFP) with DMF were 25.1 V and 0.025 μA and 15.0 V and 0.035 μA with MEK, respectively. HFP-based devices with large memory windows and higher drain currents were observed with the DMF film. The higher drain currents with the DMF film can be attributed to the smooth surface morphology of the ferroelectric layer, which can eventually lead to a larger pentacene grain size[25]. The smooth surface morphology of the ferroelectric layer induces a larger pentacene grain size, which ultimately affects the charge transport properties.
The time-dependent electrical characteristics were measured to investigate the stability of the organic flexible Fe-FETs. The variations in the electrical characteristics of the flexible devices were measured every 15 days for one month. Changes in the ΔVth of P(VDF-TrFE) and P(VDF-HFP) from the DMF solvent are shown in Figure 4 (a, c), and its corresponding output characteristics are depicted in Figure 5 (b, d). The initial ΔVth values of the Fe-FET with P(VDF-TrFE) and P(VDF-HFP) dielectric layers were 15.0 and 25.1 V, respectively. The device with the P(VDF-TrFE) layer showed an increase in window size with time, and the extracted ΔVth after one month was 20.4 V. For the device with the P(VDF-HFP) dielectric, ΔVth decreased from 25.1 to 19.0 V after two weeks. The obtained results demonstrate that the device with the P(VDF-TrFE) dielectric layer had excellent electrical stability with a significant improvement in the memory window. In contrast, the devices with P(VDF-HFP) exhibited the worst electrical stability and a considerably decreased memory window.
According to the output characteristics, all of the devices exhibited linear and saturation regions, as shown in Fig 4. However, a slight decrease in the drain current was observed when the devices were stored. This slight reduction in the output drain current may be related to the chemical or electronic degradation of the pentacene layer over time[26]. Despite the relatively high stability of pentacene Fe-FETs, the pentacene layer appears to be highly vulnerable to ordinary environmental conditions.
Fig 5 (a, b) show the time-dependent memory window and output characteristics of the Fe-FETs with P(VDF-TrFE) film from the MEK solvent. The initial ΔVth is 9.2 V, while after one month of storage time, the window size increased to 24.7 V. This increase in ΔVth is attributed to a threshold voltage shift. The device with P(VDF-TrFE) from the MEK solvent showed stable electrical performance with no obvious decrease in device performance. However, the slight decrease in drain currents from the output characteristics is correlated with the tendency for light-catalyzed aerial oxidation of the pentacene films[27]. The performance of Fe-FETs with P(VDF-HFP) from the MEK solvent suffered from rapid performance degradation within a short period.
These results indicate that flexible Fe-FETs with different dielectric layers have varied device stability. Devices with P(VDF-TrFE) films from both DMF and MEK solvents exhibited excellent electrical stability with significant improvements in the memory window. The increase in the memory window of the P(VDF-TrFE)-based devices is attributed to the alignment of the dipoles in the dielectric layer. In contrast, the P(VDF-HFP)-based devices showed a remarkable decrease in the memory window, with substantial degradation in electrical performance. This rapid degradation may be related to the surface energy of the gate dielectric material, i.e., pentacene aggregation occurs when a surface energy mismatch arises between the P(VDF-HFP) and pentacene semiconductors, eventually leading to a dramatic decrease in device performance. Therefore, the dielectric–semiconductor interface has a strong effect on the device stability of flexible Fe-FETs.
These results demonstrate that the device stability of the pentacene Fe-FET depends on the gate dielectric material. Few researchers have examined the effect of the gate dielectric layer on device stability, although Kumaki et al. reported an improvement in pentacene device stability after introducing an appropriate modification layer[27]. Consequently, two conclusions can be inferred based on the molecular structures. First, the polarity of the materials indicates that DMF has a higher dipole moment than MEK, with respective values of ~3.86 D and ~ 2.78 D. This results in higher electrical properties in low polarity materials, such as P(VDF-TrFE). Conversely, materials with inherently polar molecular structures, like P(VDF-HFP), exhibit better electrical performance in solvents with relatively lower polarity. Based on these findings, it can be concluded that the relatively simple P(VDF-TrFE) dielectric layer demonstrated greater electrical stability, while the more complex and highly polar P(VDF-HFP) dielectric layer was prone to rapid degradation over time.

4. CONCLUSIONS

Flexible organic Fe-FETs were fabricated using P(VDF-TrFE) and P(VDF-HFP) gate dielectric materials, and their electrical characteristics were investigated. The effects of the solvents on the electrical performance of the flexible Fe-FETs were compared. The solvent effect is an important parameter that determines the physical and chemical properties of the thin films. Two organic solvents, DMF and MEK, were investigated in this study. The obtained results demonstrated that the devices with P(VDF-TrFE) and P(VDF-HFP) films from the DMF solvent exhibited larger hysteresis loops with higher output drain currents. The calculated memory window (ΔVth) and drain currents of Fe-FETs with P(VDF-TrFE) and P(VDF-HFP) films from the DMF solvent were 18.0 V and −0.257 μA and 25.1 V and −0.025 μA, respectively. The ΔVth and drain currents of Fe-FETs with the MEK films were 9.2 V and −0.0035 μA with P(VDF-TrFE) and 15.0 V and −0.035 μA with P(VDF-HFP), respectively. The larger hysteresis loop and higher drain currents of both dielectric materials from the DMF solvent might be related to their low surface roughness.
In addition, the time-dependent electrical characteristics of the flexible organic Fe-FETs were measured using P(VDF-TrFE) and P(VDF-HFP) films from both DMF and MEK solvents. Devices with different dielectric layers showed varied device stabilities; for example, the P(VDF-TrFE)-based devices showed stable electrical properties with a significant improvement in the hysteresis loop with storage time. The time dependent ΔVth of devices with P(VDF-TrFE) from the DMF and MEK solvents were 15.0 V (initial) and 20.4 V (after storage) and 9.2 V (initial) and 24.7 V (after storage). In the P(VDF-HFP) based devices, the memory window decreased from 25.1 V to 19.0 V after two weeks with DMF films, and the devices degraded within a short period of time with the MEK films. This dramatic degradation of the P(VDF-HFP)-based devices may be related to the surface energy mismatch between the dielectric and semiconductor layers. Hence, the dielectric–semiconductor interface has a strong influence on the device stability of flexible Fe-FETs.

Notes

ACKNOWLEDGEMENT

This research was supported by Hallym Research Fund (HRF-202312-005).

Fig. 1.
Schematic cross-section of the flexible organic Fe-FET.
kjmm-2025-63-3-203f1.jpg
Fig. 2.
Transfer and output characteristics of P(VDF-TrFE) prepared with the (a, b) DMF and (c, d) MEK solvents
kjmm-2025-63-3-203f2.jpg
Fig. 3.
Transfer and output characteristics of P(VDF-HFP) prepared with DMF (a,c) and MEK (b,d) solvents.
kjmm-2025-63-3-203f3.jpg
Fig. 4.
Time-dependent memory window (ΔVth) of flexible organic Fe-FETs with P(VDF-TrFE) from the DMF solvent. (b, d) Corresponding output characteristics.
kjmm-2025-63-3-203f4.jpg
Fig. 5.
(a, b) Time-dependent memory window (ΔVth) of flexible organic Fe-FETs with P(VDF-TrFE) from the MEK solvent and (b) its corresponding output characteristics. (c, d) Changes in transfer and output characteristics of P(VDF-HFP) with time.
kjmm-2025-63-3-203f5.jpg

REFERENCES

1. M. Lee and B. Jeong, Adv. Physics Res. 1, 2400087 (2024).

2. P. S. Lekshmi Priya, B. Swain, S. Rajput, S. Behera, and S. Parida, Condensed Matter. 8, 105 (2023).
crossref
3. A. Laudari, J. Barron, A. Pickett, and S. Guha, ACS Appl. Mater. Interfaces. 12, 26757 (2020).
crossref
4. J. Perelaer, P. J. Smith, D. Mager, D. Soltman, S. K. Volkman, V. Subreamanian, J.G. Korvink, and U.S. Schubert, J. Mater. Chem. 20, 8446 (2010).
crossref
5. B.Y. Ahn, E.B. Duoss, M.J. Motala, X. Guo, S.I. Park, Y. Xiong, J. Yoon, R.G. Nuzzo, J.A.. Rogers, and J.A. Lewis, Science. 323, 1590 (2009).
crossref
6. J.-H. Kim, J.-H. Bae, and M.-H. Kim, Journal of Nanoscience and Nanotechnology. 17, 4149 (2017).
crossref
7. L. Petti, N. Münzenrieder, C. Vogt, H. Faber, L. Büthe, G. Cantarella, F. Bottacchi, T.D. Anthopoulos, and G. Tröster, Applied Physics Reviews. 3, 021303 (2016).
crossref
8. W.-S. Choi, Y. J. Lee, and Y. J. Kim, Korean J. Met. Mater. 61, 589 (2023).
crossref pdf
9. M.A. Khan, U.S. Bhansali, and H.N. Alshareef, Advanced Materials. 24, 2165 (2012).
crossref
10. T. Sekitani, K. Zaitsu, Y. Noguchi, K. Ishibe, M. Takamiya, T. Sakurai, and T. Someya, IEEE Transactions on Electron devices. 56, 1027 (2009).
crossref
11. D.-K. Kim, G.-J. Lee, J.-H. Lee, M.-H. Kim, and J.-H. Bae, Japanese Journal of Applied Physics. 57, 05GC03 (2018).
crossref pdf
12. Y.U Kang and J. W. Lee, Korean J. Met. Mater. 62, 239 (2024).
crossref pdf
13. J.C. Scott and L.D. Bozano, Advanced Materials. 19, 1452 (2007).
crossref
14. T. Sekitani, T. Yokota, U. Zschieschang, H. Klauk, S. Bauer, K. Takeuchi, M. Takamiya, and T. Sakurai T. Someya, Science. 326, 1516 (2009).
crossref
15. I. Katsouras, D. Zhao, M.-J. Spijkman, M. Li, P.W.M. Blom, D.M. de Leeuw, and K. Asadi, Scientific Reports. 5, 1 (2015).

16. R.C.G. Naber, K. Asadi, P.W.M. Blom, D.M. de Leeuw, and B. de Boer, Adv. Mater. 22, 933 (2010).
crossref
17. H. Wang, Y.Q. Peng, Z.Y. Ji, M. Liu, L.W. Shang, and X.H. Liu, Chinese Science Bulletin. 56, 1325 (2011).
crossref
18. R.C.G. Naber, B.de Boer, P.W.M. Blom, and D. M. de Leeuw, Applied Physics Letters. 87, 203509 (2005).

19. Y.J. Park, S.J. Kang, B. Lotz, M. Brinkmann, A. Thierry, K.J. Kim, and C. Park, Macromolecules. 4, 8648 (2008).

20. J.J. Brondijk, K. Asadi, P.W.M. Blom, and D.M. de Leeuw, Journal of Polymer Science Part B: Polymer Physics. 50, 47 (2012).
crossref
21. S.-J. Kim and J.-S. Lee, Nano Letters. 10, 2884 (2010).
crossref
22. S.-T. Han, Y. Zhou, and V.A.L. Roy, Advanced Materials. 25, 5425 (2013).
crossref
23. T. Cramer, L. Travaglini, S. Lai, L. Patruno, S. de Miranda, A. Bonfiglio, P. Cosseddu, and B. Fraboni, Scientific Reports. 6, 38203 (2016).

24. T. Xu, L. Xiang, M. Xu, W. Xie, and W. Wang, Scientific Reports. 7, 8890 (2017).

25. D.K. Kim, H. Lee, X. Zhang, J.-H. Bae, and J. Park, Micromachines. 10, 727 (2019).
crossref
26. J. Lohani, P. Saho, U. Kumar, V.R. Balakrishnan, and P.K. Basu, Pramana. 71, 579 (2008).
crossref pdf
27. C.-a. Di, G. Yu, Y. Liu, Y. Guo, X. Sun, J. Zheng, Y. Wen, Y. Wang, W. Wu, and D. Zhu, Phys. Chem. Chem. Phys. 11, 7268 (2009).
crossref
TOOLS
PDF Links  PDF Links
PubReader  PubReader
ePub Link  ePub Link
Full text via DOI  Full text via DOI
Download Citation  Download Citation
  Print
Share:      
METRICS
0
Crossref
0
Scopus
284
View
13
Download
Related article
Editorial Office
The Korean Institute of Metals and Materials
6th Fl., Seocho-daero 56-gil 38, Seocho-gu, Seoul 06633, Korea
TEL: +82-2-557-1071   FAX: +82-2-557-1080   E-mail: metal@kim.or.kr
About |  Browse Articles |  Current Issue |  For Authors and Reviewers
Copyright © The Korean Institute of Metals and Materials.                 Developed in M2PI