Abstract
There is a growing interest in new semiconductor nanostructures for future high-density high-performance flexible electronic devices. Two-dimensional conjugated microporous polymers (2D-CMPs) are promising candidates because of their inherent optoelectronic properties. Here, we are reporting a novel donor–acceptor type 2D-CMP based on Pyrene and Isoindigo (PI) for a potential nano-scale charge-trapping memory application. We exfoliated the PI polymer into ~ 2.5 nm thick nanoparticles (NPs) and fabricated a Metal–Insulator–Semiconductor (MIS) device with PI–NPs embedded in the insulator. Conductive AFM (cAFM) is used to examine the confinement mechanism as well as the local charge injection process, where ultrathin high-κ alumina supplied the energy barrier for confining the charge carrier transport. We have achieved a reproducible on-and-off state and a wide memory window (ΔV) of 1.5 V at a relatively small reading current. The device displays a low operation voltage (V < 1 V), with good retention (104 s), and endurance (103 cycles). Furthermore, a theoretical analysis is developed to affirm the measured charge carriers’ transport and entrapment mechanisms through and within the fabricated MIS structures. The PI–NPs act as a nanoscale floating gate in the MIS-based memory with deep trapping sites for the charged carriers. Moreover, our results demonstrate that the synthesized 2D-CMP can be promising for future low-power high-density memory applications.
Similar content being viewed by others
Explore related subjects
Discover the latest articles, news and stories from top researchers in related subjects.Introduction
Organic electroactive memory materials are advantageous compared to their inorganic counterparts because they are cheaper, easier to fabricate, and can achieve flexibility, deformability, and more importantly, tunable electrical characteristics by using a molecular design approach. Recently, two-dimensional conjugated microporous polymers (2D-CMPs)1,2,3,4,5 are gaining significant attention in this field due to their extensive π-conjugation, excellent chemical and thermal stabilities, tunable layered porous structures, and functions. In fact, for over a decade, they have been utilized in successful commercial applications, such as anti-static bags, magnetic storage, capacitors, batteries, and electrostatic loudspeakers6,7,8.
Electroactive memory materials including CMPs exhibit both electrically volatile and non-volatile memory (NVM) characteristics9,10,11,,1, 10, 11. The conjugated backbone of CMPs is the primary contributor to their electrical properties. When various donors and acceptors are incorporated into the CMPs, the memory properties can be significantly influenced, which can either establish trapping sites or create charge transfer (CT) conducting channels. Making CMPs intriguing as the volatility of the memory device is determined by the stability of charge trapping. However, because of the cross-linked configuration and non-solubility in most solvents, forming thin films from CMPs is a major obstacle to their incorporation in optoelectronic devices. The electrical properties of these devices can be characterized at the nanoscale using conductive atomic force microscopy (cAFM), hence providing a useful method for studying electronic transport and conduction in such structures without the need for thin film formation.
Nanoscale memory devices can utilize nanoparticles (NPs) as isolated charge trap sites. These devices reap the benefits of conventional NVM while improving the stability of the stored information12,14,14. The enhanced stability in charge storage is mainly associated with the quantum confinement and electrical isolation of the NPs, allowing for both higher power efficiency and density. One approach to achieve this is by embedding semiconductor or metallic NPs in a floating gate layer made of dielectric oxides. In this case, the charged carriers are injected from and to the active region through direct tunneling. The trapped charge carriers in the NPs dampen out the electric field through the dielectric and shift the device's threshold voltage approximately by the number of carriers trapped within each NP. The trapped charge can also be erased by applying a sufficient voltage bias.
Several studies9,10,11,,1, 1011 have explored the hybridization of nanoscale NVMs and organic thin film devices (OTFDs). The goal is to leverage the benefits of OTFDs, such as processability, low cost, and the ability to fabricate large-area structures on flexible substrates. Unlike traditional NVMs, which rely on more rigid solid-state structures, many flexible electronic devices require optimized nonvolatile data storage. However, previous approaches for charge trapping in the OTFD have used CPs as a continuous charge-trapping layer to replace the floating gate in NVMs11. Although some studies have investigated the integration of solution-processed NPs into OTFDs to mimic the charge-trapping mechanisms used in nanoscale solid-state devices that employ NPs, the challenge arises due to incompatibility with traditional CMOS processes14,16,17,18,19,20,21,22,22. Additionally, no previous study has investigated single CP-NP-based memory. Herein, we report a nano-memory Metal–Insulator–Semiconductor (MIS) structure that utilizes an acetylene-linked Pyrene–Isoindigo (PI) 2D-CMP NPs embedded in high-κ ALD oxide on bulk Ge. Conductive atomic force microscopy (cAFM) provides insight into the structure’s charge-trapping properties using local IV characterization and current mapping. The developed PI-based structure has the potential to be applied in next-generation flexible electronics and memory devices.
Experimental method
PI 2D-CMP synthesis and characterization
Donor–acceptor-based 2D CMP polymer based on pyrene and isoindigo (Fig. 1) was synthesized by following the reported procedure23. The synthesized polymer was well-characterized before testing it for further applications. In short, the Fourier-transform infrared (FT-IR) spectrum of PI showed the absence of ≡C–H and C–Br stretching bands corresponding to the monomers and the presence of alkyne –C≡C– stretching band to confirm the network formation (Fig. S1). Furthermore, molecular-level characterization is realized by solid-state NMR analysis (Fig. S2) whereas bulk microscopic characterization is achieved by scanning electron microscopy (Fig. S3). To understand the 2D nature of the polymer, Transmission Electron Microscopy (TEM) characterization is used. The images (Fig. S4) indicate the formation of stacked thin 2D-sheets to confirm the dimensionality of our material. The observation is further endorsed by the structural and morphological characterization described in our previous report23. The stacking of 2D-layers is governed by the strong van der Waals interaction originating from the N-hexyl chain and the large local dipole of isoindigo units and a strong π–π stacking interactions of pyrene units.
MIS fabrication
To start the fabrication process, a germanium (Ge) (100) wafer is used as the starting material. The wafer is first cut into 1 cm × 1 cm pieces and then subjected to sonication in de-ionized (DI) water for 15 min to etch Ge native oxide. The pieces are then rinsed sequentially in acetone, isopropanol (IPA), and DI water. A 2 nm of Al2O3 is then grown on the prepared pieces using an Oxford FlexAl Plasma Atomic Layer Deposition (ALD) system. The precursor used for the deposition of the high-k oxide was trimethyl-aluminium, and the co-reactant employed was a remote O2 plasma. Growth is conducted at 200 °C and an operating pressure of 200 mTorr, ending up with a 0.12 nm/cycle growth rate. The TMAl pulsing and purge times are 0.25–4 s, respectively.
The synthesized PI–NPs that have undergone the exfoliation process were suspended in IPA and stabilized with a concentration of 1 mg/mL. The solution was then immediately spin-coated onto the Al2O3 blocking oxide layer (BO). A volume of 20 µL was dropped on each sample while spinning at 500 RPM, raised to 1000 RPM for 45 s. To prevent any unwanted contamination from forming traps between the two alumina layers, an additional Al2O3 tunneling oxide (TO) was deposited immediately with a thickness of 2 nm, capping the PI–NPs. The thickness and roughness of the ALD films were verified using ellipsometry and AFM. Figure 2 illustrates the schematic of the fabricated structure. The PI–NPs size and dispersion along with the alumina BO/TO thicknesses were optimized and engineered for their charge trap properties.
MIS characterization
The spatial characterization of well-separated PI NPs and their size were determined using AFM for samples without an Al2O3 tunneling oxide layer. Figure 3 depicts the AFM topographic image of the PI–NPs dispersion on top of the Al2O3 and their mean radius, respectively. The scans reveal that the exfoliated particles have a density of 9 µm−2 and a mean grain size of 50 nm. The AFM scans reveal that the exfoliated particles are dispersed and well separated with a mean grain radius ranging from ~ 10.5 to 25 nm along with a thickness ranging from ~ 2.1 to 3.3 nm.
Figure 4a shows cross-sectional transmission electron microscopy (XTEM) image of the fabricated MIS stack with the embedded PI–NPs. An obvious sandwich of the Al2O3 TO (2 nm), PI–NPs (3.5 nm), and Al2O3 BO (2 nm) can be observed (Fig. 4b). The TEM images are conducted using ThemisZ S/TEM operated at 300 kV. The samples were prepared using Helios G4 PFIB, a dual beam scanning electron microscope (SEM) that is also equipped with a focused ion beam (FIB) and an Omniprobe. The sample compositional and relative thickness mapping were conducted using a oxford energy-dispersive X-ray spectrometer (EDS) as shown in Fig. 4b.
Asylum Research MFP-3D AFM equipped with the (908.036) probe holder is used for both topography and electrical measurements. The probe holder allows current ranges of 1pA to 20nA, and voltage sweeps of ± 10V with a sensitivity of 2nA/V. Imaging in both contact and tapping mode was conducted with a (PPP-NCSTAu) gold-coated silicon probe. A diamond pen was used to scratch the back side of the samples and silver paste was applied before placing the sample on a metallic disk, assuring ohmic contact. The conductive probe was kept at a virtual ground while the Ge substrate’s back was biased. Once an isolated particle was located using topography and current mapping, the probe was situated on it, and potential sweep sequences are applied (refer to Fig. 5).
Results
We conducted a series of current scans using the cAFM, operating in contact mode, with dimensions of 2.5 × 2.5 µm2. The scans were carried out as follows: a read scan was performed at 0.5 V (Fig. 6a), a write scan was done at 5 V (Fig. 6b), followed by another read scan at 0.5 V (Fig. 6c), an erase scan was executed at − 8 V (Fig. 6d), and lastly, a final read scan was carried out at 0.5 V (Fig. 6e). It was observed that when the pristine device was subjected to the low-read voltage (0.5 V), no currents were detected. However, under a higher write voltage (5 V), current readings can be detected. The current readings, which were produced by the write scan, were also visible during the subsequent low-read voltage scan (0.5 V). These same currents were seldom visible after any erase scan (− 8 V) or during the succeeding read scan (0.5 V), as depicted in Fig. 6.
We also demonstrate the charge storage component of our PI–NPs in MIS devices by conducting electrical measurements in the form of voltage sweeps and collecting the current response at predetermined locations. For clarification, MISA is used to refer to the PI–NPs-based MIS structure of Au-Tip/Al2O3/PI–NPs/Al2O3/Ge, while MISB refers to the control MIS structure of Au-Tip/Al2O3/Ge. Our findings, depicted in Fig. 7a and b, reveal distinct behaviors for the two stacks. Specifically, we observed current–voltage (I–V) characteristics for points A and B of the MISA stack, as well as for points C and D of the MISB stack. To ensure the reproducibility of the electrical behavior of MISA and MISB, we performed I–V characteristic measurements at various locations, which demonstrated similar behavior.
By referring to Fig. 7a, it can be noted that the turn-on voltage of MISA at NPs A and B is much lower (approximately 0.9 V) when compared to that of MISB at oxides C and D (approximately 3.2 V) in Fig. 7b. At points C and D, in the absence of NPs, there is a distinct change in the I–V response at low bias voltages, where no current is detected as in the MISA stack. Nevertheless, as the bias voltage increases, a sudden rise in current response can be observed, signifying a breakdown in the oxide film. This is indicative of the creation of a conducting filament throughout the oxide layers under the first potential sweep. Nonetheless, the setting and resetting mechanisms of channels within dielectrics, such as Al2O3, have been previously studied and are not the emphasis of the present work24,25.
The current responses after the first voltage sweep are consistently smaller compared to that of the latter, regardless of the reading voltage. This reduction in current response throughout the subsequent sweeps occurs because of carrier injection and confinement within the NP during the first sweep. As illustrated in Fig. 7a, a voltage shift (∆V) around 1 V is obtained between the first and second sweep at 200 pA reading current for both NPs. The window size is sufficient to differentiate between the two on and off states and, as a result, demonstrates the memory effect. The electron charge trapping behavior is experimentally verified by the positive voltage shift (∆V) in the I–V characteristics conducted on the MIS device structure, therefore confirming that the 2D-CMP behaves as an n-type semiconductor.
In Fig. 7, during the initial voltage sweep, MISA is configured to a low state, leading to subsequently lower current responses, while MISB is set to a high state. MISB functions as a resistive switching memory by altering the resistance of Al2O3, whereas MISA operates as a charge trapping memory, trapping, and de-trapped charge carriers within the PI–NP. Charge trapping memory typically exhibits slower operational speeds and a smaller memory window when compared to resistive switching memory. However, it often exhibits higher endurance, improved retention along with seamless integration with conventional CMOS technology26,27.
For a more comprehensive investigation of the shift in the current response of the MISA stack throughout the initial and subsequent voltage sweeps, a series of individual potential scans can be used to read, write, and erase the charge stored in the PI–NP. These sweeps function as both charging (writing), charge detection (reading), and discharging (erasing) of the PI–NP. The entire process is depicted in Fig. 8, where a memory window (∆V = 1.46 V at 200 pA) is apparent between the initial and subsequent sweeps. Initially, no carriers are confined within the NP, so the first current response (W1), to a gate voltage sweep (Vg) of 4 V, displays the charge transport and confinement within the NP (writing). Carriers tunnel across the Al2O3 TO only to get trapped in the NP due to the BO. On the other hand, the second response (R1) to the same gate sweep indicates a reduction in electric current due to the field screening introduced by the confined negative charge during the initial sweeps (reading). The lower current observed after the subsequent sweep further verifies the storage of charged carriers in the NP. The third response (E1) with a sweep of -6 V with opposite polarity is applied demonstrating the removal of stored charge (erasing) indicating it was sufficient to eject all confined charge. Charge ejection is verified with a follow-up potential scan of 5 V as seen in the fourth response (W2), where it almost coincides with those of the initial response. Further voltage sweeps overlap with the initial response (W1), signaling the reproducibility of the memory effect, as indicated by the six (Write/Read/Erase) sets of sweeps in Fig. 8. The I–Vs behavior presented illustrates a higher current flow under positive bias, which can be attributed to the improved tunneling due to the electric field enhancement at the nanoscale conductive probe and NP, narrowing the barrier width at the interface.
As illustrated by Figs. 2, 4 and 5 nm thick Al2O3 layers serve as an effective TO for the injected charge in the PI–NP sandwich, and act as an efficient BO for confining the charge. Nonetheless, there is a small contribution of Al2O3 to charge trapping in the system. Given the reported interface trap density within ultrathin Al2O3 thin films, which is approximately 1011–1012 cm−228.
While Fig. 8 only illustrates the initial six write/erase (W/E) cycles, we also assessed the endurance characteristics of the MISA stack up to 210 cycles. The endurance measurement starts with the application of a W/E voltage pulse at ± 4 V for a given cycle number, followed by voltage sweeps from 0 to 4 V. Subsequently, Fig. 9a displays the I-V characteristics of the MISA stack throughout the endurance test, following 1, 110, 170, and 210 cycles. Additionally, Fig. 9b presents the current response recorded after a given cycle number, measured at 0.8 V, and extracted from 22 I–V sweeps. Even when extrapolated to 103 W/E cycles, there is only a marginal 25% decrease in the memory on/off ratio. It is worth noting that while the reduction in discharged current remains minimal, the programmed current exhibits more pronounced increases as each erase process removes less charge compared to the write process therefore blocking less carriers, aligning with the asymmetrical nature of the IV curves. Additionally, it is plausible that local defect sites within the oxides facilitate increased electron tunneling, particularly under a high number of cycles.
We also evaluated the retention characteristics of the MISA device, as depicted in Fig. 9c. To monitor charge leakage, the retention measurement starts with a write voltage pulse at 4 V/100 ms, followed by the measurement of the current response using a read voltage pulse of 0.8 V/10 ms after a specified retention time interval, employing a custom pulse train. After ~ 1300 s, there is an observable increase in current response attributed to the reduction in trapped electric charge and, consequently, the diminishing shielding effect. Nonetheless, even after ~ 6000 s, the stored charges retained roughly 85% of their initial pristine value. The device exhibited good charge storage retention properties, characterized by a high on/off ratio memory, with a retention duration exceeding 104 s.
Figure 10 illustrates various write/read voltage biases when utilized to execute similar IV sweeps on a different PI–NP. By applying two successive gate voltage sweeps (Vg) of 3, 3.5, 4, 4.5, and 5 V, resulting in a broad memory window up to (∆V = 1 V) at (Vg = 5 V at 200 pA).
While cAFM is a powerful tool to investigate the local electrical properties of materials at the nanoscale, it does have its limitations when it comes to current–voltage I-V and ultra-fast pulsed I-V characterization and lacks the capability for capacitance–voltage (C–V) measurements. The technique provides qualitative data about the electrical properties of the sample, but getting accurate quantitative data can be challenging due to the complexities of the electrical contact between the AFM tip and the sample. During measurements, the elevated local electric fields at the AFM tip can introduce additional stress to the device in addition to rapid and sustained current measurements can lead to rapid degradation of the AFM tip. Moreover, the cAFM technique exhibits a low signal-to-noise ratio, especially at low conductivities, and faces limitations in lateral resolution due to the dispersion of current within the material.
Previous work utilizing cAFM in similar memory structures usually employ resistive switching29,31,31. Resistive switching provides a larger memory window but usually requires significantly higher operation voltages and suffers from endurance issues. The few studies making use of charge trapping are limited but show similar memory window, retention, and endurance to our structure15,32, 33.
Theoretical analysis and discussion
To affirm the charge confinement and transport within the produced MIS structures, a theoretical analysis (TA) is developed. The conducted TA starts with the calculation of charge capacitance (Cq) and trapped charge density of (Nt) and is analyzed through the charge-trapping mechanism34. The unknown dielectric constant value of PI is assumed to range from 1 to 3 based on previous work on polymers with similar chemical structure35. The total injected charge (Qt) is related to the threshold voltage shift (∆Vth) by the following equation:
The fabricated MIS structure (Fig. 7a) has a charge capacitance (Cq) that can be expressed as two capacitors in series as depicted in Fig. 11:
where ε0, εAl2O3, and εPI are the dielectric constant of free space, Al2O3 and PI, respectively. While tAl2O3 and tPI are the thicknesses of Al2O3 and PI–NP, respectively. Trapped charge density relies on various factors including the thickness of the oxide under the AFM probe, the linear dimension of the quantum well, and the layers’ relative permittivity. Therefore, the charge storage capacitance (Cq) can be estimated for an εPI ranging from 1 to 3 as illustrated in Fig. 11 showing a maximum capacitance of 9.52 pF/m2 with an εPI of 3.0.
The trapped charge density (Nt) can now be calculated from the charge capacitance (Cq) for various εPI values using the following equation36,37:
where ∆Vth is the change in threshold voltage of the on-and-off states of the memory and q is the elementary charge.
Figure 12a shows the trapped charge density (Nt) for different εPI as a function of voltage shift (∆Vth) which consent to the kinetics model of the charge transport where ∆Vth and Nt are linearly proportional. The same is reflected in Fig. 12b with a gate voltage (Vg) variation.
Finally, the electric field across the tunneling oxide is calculated to investigate the charge emission mechanism. The electric field across the device is extracted by applying Gauss’s law to simple MIS devices38. The electric field across the tunneling oxide (E) is:
The electric field (E) under various gate voltages (VG) is shown in Fig. 13, which reflects the enhancement of the electric field across the tunnel oxide due to the nanoscale size of the AFM probe. The initially higher electric field across the tunnel oxide conveys that the charge-trapping mechanism is based on Fowler–Nordhiem (F–N) tunnelling. This can be confirmed by plotting the natural log of (∆Vth) divided by the squared inverse of the electric field (E) in Fig. 13b for εPI of 3.0. The linear trend confirms that the dominant charge emission mechanism is based on F-N tunneling.
Figure 14 demonstrates the F–N tunneling mechanism employing the energy band diagram of the MISA structure. During the writing operation shown in Fig. 14b, upon applying a sufficiently positive bias (Vg) to the substrate, a very high electric field forms beneath the cAFM Au-coated tip, leading to an elevated Au-tip energy band. Electrons within the Au-tip acquire sufficient energy to be easily injected into the conduction band of PI–NP through the blocking oxide (BO) and become confined into states residing within the PI bandgap. Simultaneously a smaller number of the holes within the bulk can acquire adequate energy to directly tunnel from the valence band of Ge through the tunnel oxide (TO) and become confined also. This leads to the trapping of surplus of electrons within the PI–NP and interface states. Upon the application of subsequent positive bias as depicted in Fig. 14c, electrons encounter a shielding effect and are screened by the trapped charge, and hence, fewer electrons can tunnel causing a rightward shift in the memory's I–V characteristic. Conversely, during the erase operation as illustrated in Fig. 14d, a negative bias (− Vg) is applied to the substrate causing an elevation of the Ge energy band. This promotes the tunneling of the stored electrons in the PI–NP back, whereas holes confined within the PI–NP can easily tunnel back into the bulk. The absence of trapped electrons results in the I-V characteristic of the memory shifting back to its initial state. This description aligns with our observations of the asymmetrical I-V curves under forward and reverse bias conditions which can be attributed to variations in conductance and valence band barrier heights at the interface.
Consequently, it can be inferred that a high density of trapping states exists in PI–NP with increased defect density, leading to an expanded memory window. However, achieving precise control over the defect density in our current device setup poses a significant challenge due to its 2D and porous nature. In future research endeavors, additional efforts will be required to enhance the precision of defect density manipulation in PI–NP. From an electron theory perspective, a low-lying LUMO, indicated by high electron affinity from Density Functional Theory (DFT) calculations on PI23, can serve as a charge-trapping center which in turn allow the capture of charged carriers from the Fermi energy level and tunneling from either the conductive probe or semiconductive substrate39.
Conclusions
In summary, the fabricated MIS structure utilizes an acetylene-linked Pyrene-Isoindigo (PI) 2D-CMP NPs embedded in high-κ ALD oxide on bulk Ge. Conductive atomic force microscopy (cAFM) is used to provide insight into the structure’s charge-trapping properties using local IV characterization and current mapping. The device displays a wide memory window (∆V) up to 1.5 V at relatively small reading voltages along with reproducible on-and-off states. Furthermore, a theoretical analysis is developed to affirm the measured charge carriers’ transport and entrapment through and within the fabricated MIS structures. The developed PI-based structure has the potential to be applied in next-generation low-power high-density memory devices.
Data availability
The data that support the findings of this work are available from the corresponding authors upon reasonable request.
References
Lee, J.-S.M. & Cooper, A. I. Advances in conjugated microporous polymers. Chem. Rev. 120(4), 2171–2214. https://doi.org/10.1021/acs.chemrev.9b00399 (2020).
Chaoui, N., Trunk, M., Dawson, R., Schmidt, J. & Thomas, A. Trends and challenges for microporous polymers. Chem. Soc. Rev. 46(11), 3302–3321. https://doi.org/10.1039/C7CS00071E (2017).
Geng, K., Arumugam, V., Xu, H., Gao, Y. & Jiang, D. Covalent organic frameworks: Polymer chemistry and functional design. Prog. Polym. Sci. 108, 101288. https://doi.org/10.1016/j.progpolymsci.2020.101288 (2020).
Martínez-Abadía, M. & Mateo-Alonso, A. Structural approaches to control interlayer interactions in 2D covalent organic frameworks. Adv. Mater. 32(40), 2002366. https://doi.org/10.1002/adma.202002366 (2020).
Evans, A. M. et al. Two-dimensional polymers and polymerizations. Chem. Rev. 122(1), 442–564. https://doi.org/10.1021/acs.chemrev.0c01184 (2021).
Cooper, A. I. Conjugated microporous polymers. Adv. Mater. 21(12), 1291–1295. https://doi.org/10.1002/adma.200801971 (2009).
Xu, Y., Jin, S., Xu, H., Nagai, A. & Jiang, D. Conjugated microporous polymers: Design, synthesis and application. Chem. Soc. Rev. 42(20), 8012. https://doi.org/10.1039/C3CS60160A (2013).
Tan, J., Chen, W.-J. & Guo, J. Conjugated microporous polymers with distinctive π -electronic properties exhibiting enhanced optical applications. Chin. Chem. Lett. 27(8), 1405–1411. https://doi.org/10.1016/j.cclet.2016.06.050 (2016).
Cheng, Z. et al. Rapid metal-free synthesis of pyridyl-functionalized conjugated microporous polymers for visible-light-driven water splitting. Polym. Chem. 11(20), 3393–3397. https://doi.org/10.1039/D0PY00249F (2020).
Platzer, N. Encyclopedia of Polymer Science and Engineering, H. F. Mark, N. M. Bikales, C. G. Overberger, and G. Menges, Wiley-Interscience, New York, 1985, 720 Pp. Journal of Polymer Science Part C: Polymer Letters, 24 (7), 359–360. https://doi.org/10.1002/pol.1986.140240720 (1986).
Baeg, K.-J. et al. Organic non-volatile memory based on pentacene field-effect transistors using a polymeric gate electret. Adv. Mater. 18(23), 3179–3183. https://doi.org/10.1002/adma.200601434 (2006).
Gelinck, G. Trapped fast at the gate. Nature 445(7125), 268–268. https://doi.org/10.1038/445268a (2007).
Leong, W. L. et al. Non-volatile organic memory applications enabled by in situ synthesis of gold nanoparticles in a self-assembled block copolymer. Adv. Mater. 20(12), 2325–2331. https://doi.org/10.1002/adma.200702567 (2008).
Kim, S.-J. & Lee, J.-S. Flexible organic transistor memory devices. Nano Lett. 10(8), 2884–2890. https://doi.org/10.1021/nl1009662 (2010).
Rezk, A., Abbas, Y., Saadat, I., Nayfeh, A. & Rezeq, M. Charging and discharging characteristics of a single gold nanoparticle embedded in Al2o3 thin films. Appl. Phys. Lett. 116(22), 223501. https://doi.org/10.1063/5.0004000 (2020).
Tiwari, S. et al. A silicon nanocrystals based memory. Appl. Phys. Lett. 68(10), 1377–1379. https://doi.org/10.1063/1.116085 (1996).
Wakai, H., Kobayashi, M., Kumise, T., Yamaguchi, M., Nakanishi, T. & Tanaka, H. Ultra High Density HFO2 Nanodots Memory for Scaling. In 2006 21st IEEE Non-Volatile Semiconductor Memory Workshop. https://doi.org/10.1109/.2006.1629493
Lee, C., Gorur-Seetharam, A. & Kan, E. C. Operational and reliability comparison of discrete-storage nonvolatile memories: Advantages of single- and double-layer metal nanocrystals. IEEE Int. Electron Devices Meet. https://doi.org/10.1109/IEDM.2003.1269344 (2003).
Yamaguchi, Y. Design of novel σ*-π* conjugated polysilanes. Synth. Met. 82(2), 149–153. https://doi.org/10.1016/S0379-6779(96)03782-4 (1996).
Xu, Y. et al. Electroluminescent properties of a novel σ*–π* conjugated polymer, Poly[1,1-(2,3,4,5-Tetraphenylsilole)]. Jpn. J. Appl. Phys. 38(12R), 6915. https://doi.org/10.1143/JJAP.38.6915 (1999).
Sze, S. M., Ng, K. K. & Li, Y. In Physics of Semiconductor Devices 125–197 (Wiley, 2021).
List of Errata to Volume 1 of Instabilities in Silicon Devices Silicon Passivation and Related Instabilities. New Insulators, Devices and Radiation Effects, 935–938. https://doi.org/10.1016/s1874-5903(99)80022-6 (1999).
Ranjeesh, K. C. et al. A rational design of isoindigo-based conjugated microporous n-type semiconductors for high electron mobility and conductivity. Adv. Sci. https://doi.org/10.1002/advs.202303562 (2023).
Vishwanath, S. K., Woo, H. & Jeon, S. Enhancement of resistive switching properties in Al2O3 bilayer-based atomic switches: Multilevel resistive switching. Nanotechnology 29(23), 235202. https://doi.org/10.1088/1361-6528/aab6a3 (2018).
Molina-Reyes, J. & Hernandez-Martinez, L. Understanding the resistive switching phenomena of stacked Al/Al2O3/al thin films from the dynamics of conductive filaments. Complexity 2017, 1–10. https://doi.org/10.1155/2017/8263904 (2017).
Chen, A. A review of emerging non-volatile memory (NVM) technologies and applications. Solid-State Electron. 125, 25–38. https://doi.org/10.1016/j.sse.2016.07.006 (2016).
Scott, J. C. & Bozano, L. D. Nonvolatile memory elements based on organic materials. Adv. Mater. 19, 1452–1463. https://doi.org/10.1002/adma.200602564 (2007).
Rahman, Md. M., Shin, K.-Y. & Kim, T.-W. Characterization of electrical traps formed in Al2o3 under Various ALD Conditions. Materials 13(24), 5809. https://doi.org/10.3390/ma13245809 (2020).
Nhu Hoang Tran, T. et al. C-AFM study on multi—Resistive switching modes observed in metal–organic frameworks thin films. Org. Electron. 93, 106136. https://doi.org/10.1016/j.orgel.2021.106136 (2021).
Liang, J. et al. Forming-free resistive switching memory based on lifepo4nano-particle embedded in conjugated polymer. Semicond. Sci. Technol. 29, 115029. https://doi.org/10.1088/0268-1242/29/11/115029 (2014).
Ouyang, J. Two-terminal resistive switching memory devices with a polymer film embedded with nanoparticles. J. Mater. Chem. C 3, 7243–7261. https://doi.org/10.1039/c5tc01668a (2015).
Abbas, Y., Rezeq, M., Nayfeh, A. & Saadat, I. Size dependence of charge retention in gold-nanoparticles sandwiched between thin layers of titanium oxide and silicon oxide. Appl. Phys. Lett. https://doi.org/10.1063/5.0063515 (2021).
Xu, J., Xu, J., Zhang, P., Li, W. & Chen, K. Nanoscale quantification of charge injection and transportation process in Si-nanocrystal based sandwiched structure. Nanoscale 5, 9971. https://doi.org/10.1039/c3nr03306f (2013).
Lue, H.-T. et al. A study of gate-sensing and channel-sensing (GSCS) transient analysis method—Part I: Fundamental theory and applications to study of the trapped charge vertical location and capture efficiency of SONOS-type devices. IEEE Trans. Electron Devices 55(8), 2218–2228. https://doi.org/10.1109/TED.2008.925926 (2008).
Ahmad, Z. Polymer dielectric materials. Dielectr. Mater. https://doi.org/10.5772/50638 (2012).
El-Atab, N. et al. Enhanced memory effect via quantum confinement in 16 Nm inn nanoparticles embedded in zno charge trapping layer. Appl. Phys. Lett. 104(25), 253106. https://doi.org/10.1063/1.4885397 (2014).
El-Atab, N., Ozcan, A., Alkis, S., Okyay, A. K. & Nayfeh, A. Low power zinc-oxide based charge trapping memory with embedded silicon nanoparticles via Poole-Frenkel hole emission. Appl. Phys. Lett. 104(1), 013112. https://doi.org/10.1063/1.4861590 (2014).
Nayfeh, A. & El-Atab, N. Nanomaterials-Based Charge Trap- ping Memory Devices (Elsevier, 2020).
Choi, J.-K., Jang, S., Kim, K.-J., Sohn, H. & Jeong, H.-D. Observation of negative charge trapping and investigation of its physicochemical origin in newly synthesized poly(tetraphenyl)silole siloxane thin films. J. Am. Chem. Soc. 133(20), 7764–7785. https://doi.org/10.1021/ja1108112 (2011).
Acknowledgements
This work is supported by Khalifa University and ASPIRE, the technology program management pillar of Abu Dhabi’s Advanced Technology Research Council (ATRC) via the ASPIRE VRI (Virtual Research Institute) Award (VRI20-07). D.S. acknowledges the financial support from the Khalifa University Competitive Internal Research Award (CIRA-2022). M.H.A., D.K., and N.E.A. acknowledge financial support from the Semiconductor Initiative, King Abdullah University of Science and Technology, Saudi Arabia (KAUST Research Funding (KRF) under Award No. ORA-2022-5314). D.S. thanks Dr. Gisha E. L., Khalifa University, UAE, and Dr. Tina, S., University of Nova Gorica, Slovenia for their help in solid-state NMR and BET analysis. B.R.K. thanks the University Research Board (URB) of the American University of Beirut (AUB) for the financial support. M.H.A., D.K., and N.E.A. thank King Abdullah University of Science and Technology the Electron Microscopy Core Labs for assisting with the TEM data and analysis.
Author information
Authors and Affiliations
Contributions
A.R. fabricated and characterized the MIS structure while supervised by A.N.; K.C.R. synthesised and characterized the PI polymer while S.G. conducted the microscopic characterizations (SEM and TEM) of the synthesised polymer (SI). Both were supervised by D.S.; M.H.A. conducted the theoretical analysis while M.B.H., handled the XTEM scans and D.K. helped with TEM analysis. All supervised by N.E.A.; A.M. and B.R.K. provided the initial monomer. All authors contributed to manuscript writing and editing.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher's note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Rezk, A., Ansari, M.H.R., Ranjeesh, K.C. et al. Nano-scale charge trapping memory based on two-dimensional conjugated microporous polymer. Sci Rep 13, 18845 (2023). https://doi.org/10.1038/s41598-023-44232-1
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598-023-44232-1
- Springer Nature Limited