Low Li+ Insertion Barrier Carbon for High Energy Efficient Lithium-Ion

Dec 22, 2017 - However the key challenge in its design is the low energy efficient negative electrode which barred the realization of such research sy...
0 downloads 0 Views 6MB Size
Research Article www.acsami.org

Cite This: ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Low Li+ Insertion Barrier Carbon for High Energy Efficient LithiumIon Capacitor Wee Siang Vincent Lee,† Xiaolei Huang,† Teck Leong Tan,‡ and Jun Min Xue*,† †

Department of Materials Science and Engineering, National University of Singapore, 9 Engineering Drive 1, Singapore 117576, Singapore ‡ Institute of High Performance Computing, A*STAR, 1 Fusionopolis Way, #16-16 Connexis, Singapore 138632, Singapore S Supporting Information *

ABSTRACT: Lithium-ion capacitor (LIC) is an attractive energy-storage device (ESD) that promises high energy density at moderate power density. However, the key challenge in its design is the low energy efficient negative electrode, which barred the realization of such research system in fulfilling the current ESD technological inadequacy due to its poor overall energy efficiency. Large voltage hysteresis is the main issue behind high energy density alloying/conversiontype materials, which reduces the electrode energy efficiency. Insertion-type material though averted in most research due to the low capacity remains to be highly favorable in commercial application due to its lower voltage hysteresis. To further reduce voltage hysteresis and increase capacity, amorphous carbon with wider interlayer spacing has been demonstrated in the simulation result to significantly reduce Li+ insertion barrier. Hence, by employing such amorphous carbon, together with disordered carbon positive electrode, a high energy efficient LIC with round-trip energy efficiency of 84.3% with a maximum energy density of 133 Wh kg−1 at low power density of 210 W kg−1 can be achieved. KEYWORDS: energy efficiency, amorphous carbon, disordered carbon, voltage efficiency, practicality



INTRODUCTION Energy efficiency has long been considered as one of the key parameters that influence the capital cost per cycle,1,2 longevity of energy-storage devices (ESDs), and their operational safety. However, due to the shifting research focus toward development of high energy density ESDs, energy efficiency is often less emphasized3 compared to other more recognized and heavily compared parameters. Such phenomenon has led to numerous reports of ESDs with exceptional energy densities,4−10 which are unable to attain satisfactory energy efficiency and hence significantly reduces the objective of ESD research. To realize fruitful ESD research, greater attention has to be focused on investigating and enhancing energy efficiency of the ESD system. Lithium-ion capacitor (LIC) is a wide potential window ESD that can synergistically deliver high energy density at moderate power density11−13 due to the merger of lithium-ion battery (LIB) negative electrode with supercapacitor positive electrode. Although LIC remains as an exciting next-generation ESD, the fatal flaw in its fundamental design is the incorporation of low energy efficient LIB negative electrode (ca. 5−40%, see Supporting Information (SI) Figure S1), despite pairing with high energy efficient supercapacitor electrode. As such, numerous reported LIC systems that operate at high-voltage window of more than 3 V attained energy efficiencies of 50− © XXXX American Chemical Society

70%. Hence, even though numerous successful LIC designs with high energy densities have been reported in the past years, these devices are often impractical due to their less-than-ideal energy efficiencies (i.e., ≥80%). Therefore, this work, among few other studies, aims to realize the potential of LIC for nextgeneration technologies by designing and selecting appropriate LIB negative electrode. The key challenge that results in low energy efficient LIB negative electrode is the low voltage efficiency of the electrode materials (because energy efficiency is proportional to voltage efficiency14−16). LIB negative electrodes, in particular alloying and conversion-type materials, exhibit significant voltage hysteresis17−19 (a resultant of a set of overpotentials due to mass/charge transfer process, chemical process, etc.16,20), which drastically lowers the electrode energy efficiency. The large voltage hysteresis hence renders the high energy density alloying/conversion-type materials impractical as LIB negative electrode due to their lower voltage efficiencies. Ironically, insertion-type graphite, which is averted in most research due to its low capacity, i.e., 372 mAh g−1, remains to be the most promising LIB negative electrode21,22 due to its high energy Received: October 12, 2017 Accepted: December 22, 2017 Published: December 22, 2017 A

DOI: 10.1021/acsami.7b15473 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces

exclusive due to the abovementioned challenge for LIC negative electrode), the LIC designed in this work was able to attain good balance between energy efficiency of 84.3% and energy density of 133 Wh kg−1, which positions it as a more appealing and practical LIC. In addition to high energy efficiency and high energy and power densities, the devised LIC was able to achieve a good cyclic performance of 81.8% capacitance retention after cycling for 5000 cycles at 5 A g−1. Thus, this work shows the possible harmonious coexistence of energy density and energy efficiency, which serves as a guideline for developing practical high-performing LIC so as to move beyond the experimental stage. To further improve the energy efficiency of LIC, it would require the efforts of cell engineer, on top of the proposed materials design.

efficiency. Insertion-type graphite typically has a significantly smaller voltage hysteresis compared to high-capacity alloying/ conversion-type materials due to its (1) minimal chemical reaction (e.g., nucleation, phase transfer, etc.),14,23 (2) more efficient mass/charge transfer mechanism,24,25 and (3) smaller structural expansion during lithiation/delithiation process. However, in addition to the low capacity, graphite possesses inherent trade-offs, such as low intercalation efficiency at high current rate26 and phase transition during lithiation/delithiation,27,28 which inevitably increase the electrode overpotential. In particular, on the basis of the simulation result, significant Li+ insertion barrier has to be overcome in graphite during lithiation process, which inevitability increases strain overpotential that ultimately increases voltage hysteresis. Hence, there is an urgent need to reduce this voltage hysteresis by reducing Li+ insertion barrier of insertion-type materials. Amorphous carbon (AMC) is an attractive insertion-type LIB negative electrode material owing to its (1) higher specific capacity (ca. 740 mAh g−1, Li2C6) compared to graphite29,30 and (2) greater resistance to structural changes during ion intercalation process.31 Most importantly, on the basis of the simulation result, it was discovered that Li+ insertion barrier for amorphous carbon can be significantly reduced due to the wider interlayer spacing (ca. 0.47 nm), which is beneficial toward reducing both insertion overpotential and strain overpotential and hence improving overall voltage efficiency of the negative electrode. To achieve a high overall LIC energy efficiency, high energy efficient positive electrode has to be employed as well. Unlike the negative electrode, which is protected by the passivation of solid electrolyte interface (SEI), positive electrode is more prone to electrolyte degradation as it is unprotected from the electrolyte. 32 This results in unavoidable co-insertion of solvated ions into the interlayer spacing. The effect of solvated ions co-insertion is apparent at high voltage, whereby the trapped solvated ions can be oxidized into gaseous products, such as CO2,11,33 which the latter can exfoliate and pulverize the electrode structure. As shown in the previous work, disordered carbon as opposed to graphitic carbon could reduce the extent of entrapped solvated ions and hence reduce the intensity of electrolyte decomposition.34 As such, disordered carbon is recommended as LIC positive electrode so as to improve the columbic efficiency of the electrode at wide voltage range. Herein, to the best of knowledge, this work is one of the first few studies to consider the LIC energy efficiency during the designing process. With the aid of simulation model, highperforming, yet high energy efficient LIC was designed with amorphous carbon negative electrode and disordered carbon positive electrode in this work due to the lowered Li+ insertion barrier of amorphous carbon. The resultant 3.5 V LIC possesses both high energy and power densities as well as can operate with a high round-trip energy efficiency of 84.3% at a high current density of 1 A g−1. The designed system was able to deliver a maximum energy density of 133 Wh kg−1 at a low power density of 210 W kg−1. Even at a high power density of 11 200 W kg−1, the device was able to deliver a respectable energy density of 42 Wh kg−1. On the basis of the previously reported high voltage window LIC (≥3.5 V), the highest energy efficiency of ca. 70% was reported with a poor energy density of ca. 80 Wh kg−1, whereas the highest energy density of ca. 220 Wh kg−1 was reported at a poor energy efficiency of ca. 55%. Compared to the previously reported works, which have either high energy density or high energy efficiency (mutually



MATERIALS AND EXPERIMENTAL PROCEDURES

Materials. α-D-Glucose (anhydrous, 96%), sodium dodecyl sulfate (≥98.5%), graphite (Darco, 100 mesh), and urea (anhydrous, 99%) were purchased from Sigma-Aldrich. Ethanol (absolute) and methanol (absolute) were also used in the experiment as solvents. Deionized (DI) water was used in all reactions. The materials were used without any further purification. Synthesis of Amorphous Carbon (AMC). In a typical synthesis, 5 g of α-D-glucose and 25 mg of sodium dodecyl sulfate were dissolved in 50 mL of DI water by vortexing and bath-sonicating for 10 min. The resultant clear solution was later transferred into a Teflon-lined hydrothermal bomb and into a hydrothermal oven for 5 h at 180 °C. After the hydrothermal process, a black slurry was collected and washed with DI water and ethanol repeatedly. The hydrothermal product was later left to dry in an electric oven overnight. The dried product was eventually calcinated at 800 °C for 2 h at a ramping rate of 10 °C min−1 in N2 atmosphere. The calcinated product was denoted as AMC and was used for electrochemical testing without any further purification. Synthesis of Porous Disordered Carbon Sheet (PdCS). In a typical synthesis, 5 g of urea, 0.5 g of α-D-Glucose, and 25 mg of sodium dodecyl sulfate were dissolved in 3 mL of DI water, 9 mL of ethanol, and 3 mL of methanol. The mixture was bath-sonicated for 10 min to obtain a clear solution. The solution was later magnetically stirred in a hot silicone oil bath at 75 °C for 1 h to obtain a white solid. The white solid was later left to dry in an electric oven (80 °C) overnight to obtain yellowish sweet-smelling crystals. These yellowish crystals were later collected and calcinated at 350 °C for 4 h at a ramping rate of 5 °C min−1 and further calcinated at 950 °C for 10 h at a ramping rate of 5 °C min−1 in N2 environment. The final obtained sample was denoted as PdCS and used for electrochemical testing without any washing. Electrochemical Fabrication of Half-Cell, and LIC (AMC// PdCS). All half-cell devices studied for their material performance in this work were assembled using the standard industrial method, following the recommended method to best characterize the performance of the device. To prepare the negative electrode halfcell, AMC was mixed with poly(tetrafluoroethylene) (PTFE) and carbon black in a ratio of 8:1:1 with N-methyl-2-pyrrolidone (NMP). The mixture was hand-ground for at least 15 min to obtain a black slurry. The slurry was later coated onto copper foil, which served as a current collector using the thick film coater. After heating at 80 °C overnight, the sheet was pressed and punched into a 1.2 cm diameter electrode with mass loading of about 0.7 mg. To prepare the positive electrode half-cell, PdCS was mixed with poly(tetrafluoroethylene) (PTFE) and carbon black in a ratio of 8:1:1 with N-methyl-2pyrrolidone (NMP). The mixture was hand-ground for at least 15 min to obtain a black slurry. The slurry was later coated onto aluminum foil, which served as a current collector using the thick film coater. After heating at 80 °C overnight, the sheet was pressed and punched into a 1.2 cm diameter electrode with mass loading of about 1.2 mg. Individual half-cell measurement, in the form of coin cell, was conducted for both the AMC and PdCS electrodes, with lithium plate B

DOI: 10.1021/acsami.7b15473 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces

Figure 1. (a) Schematic showing the diffusion pathway of Li (purple atom) in graphite. At low concentration, Li adsorbs most stably at the hexagonal site. The transition state occurs with Li located above the bridge site connecting two carbon atoms. (b) Calculated energy barriers along the diffusion pathway for selected interlayer spacing of graphite. (c) Diffusion barrier vs interlayer spacing of graphite. as the counter and reference electrodes, in 1 M LiPF6 dissolved in 1:1:1 (volume) mixture of ethylene carbonate/diethyl carbonate/ dimethyl carbonate. LIC was assembled with the prelithiated AMC electrode (by cycling for 50 cycles at 1 A g−1) and PdCS electrode. Electrochemical Measurements. All of the electrochemical tests were performed at room temperature. Cyclic voltammetry (CV) and galvanostatic charge−discharge were studied using BioLogic Science Instruments VMP 3. The specific capacitance (C: mAh g−1; F: F g−1) was calculated using the following formula

solutions onto a lacey carbon grid. Powder X-ray diffraction (XRD) pattern was measured by a powder diffractometer (Bruker D8 Advanced Diffractometer System) with Cu Kα (1.5418 A) source. Powder Raman spectrometry was conducted for AMC and PdCS on Horiba MicroRaman HR Evolution System using an argon laser beam with an excitation wavelength of 514.5 nm. BET measurement for AMC and PdCS was performed with a Micromeritics ASAP 2020 surface area and porosity analyzer.

(1)

RESULTS AND DISCUSSION Material synthesis of both negative and positive electrodes was based on a simple pyrolysis route, with low-cost, nontoxic, and environmentally friendly precursors. Amorphous carbon (AMC) was prepared by pyrolyzing dewatered glucose (via hydrothermal at 180 °C) at 800 °C in nitrogen environment, whereas disordered porous carbon sheets (PdCS) were prepared via simple pyrolysis of dewatered glucose/urea at 950 °C. In this work, LIC was assembled with AMC as negative electrode and PdCS as positive electrode, and such assembly was denoted as AMC//PdCS. From first-principles calculations, it is demonstrated that the diffusion barrier of Li+ decreases with the interlayer distance of graphite (Figure 1). In normal (crystalline) graphite, the interlayer distance is small (calculated d = 3.67 Å), which translates to a high diffusion barrier of 0.62 eV (Figure 1b). The high energy barrier leads to slow diffusion of Li+ in graphite, which reduces the transfer kinetics process and hence results in lower energy efficiency due to the increased voltage hysteresis. Increasing the interlayer distance, d, decreases the diffusion barrier, as shown in Figure 1c. For amorphous carbon with d ∼ 4.7 Å, the diffusion barrier is decreased to 0.07 eV, thus explaining why a higher energy efficiency could be achieved by

C = I Δt/3.6



(2)

F = I Δt /ΔV −1

where I is the current density (A g ), Δt is the discharge time (s), and ΔV is the potential difference (V). On the basis of the total mass of the active materials from the positive and negative electrodes, energy density (E) and power density (P) were calculated using the following formula

E=

∫t

t2

V · I dt /3.6

(3)

1

(4)

P = 3600E /Δt −1

where V is the voltage range (V), I is the current density (A g ), t is the time (s), t1 is the time at which the cell is fully discharged, t2 is the time at which the cell is fully charged, and Δt is the discharge time (s). All measurements were conducted at room temperature (25 °C). Energy efficiency is calculated on the basis of the discharging area and charging area from galvanostatic charge/discharge with resolution of 0.1 s. Material Characterizations. SEM images of AMC and PdCS were recorded on ZEISS SEM Supra 40 (5 kV). SEM samples were prepared by dripping the sample solutions onto silicon substrate. TEM was done on JEOL-3010 (300 kV acceleration voltage) for both AMC and PdCS. TEM samples were prepared by dripping the sample C

DOI: 10.1021/acsami.7b15473 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces

Figure 2. Material characterization of AMC. (a) SEM image, (b) low-resolution TEM image (inset shows the SAED pattern), (c) high-resolution TEM image showing the interlayer spacing (inset shows a magnified portion of the interlayer spacing), (d) Raman spectra of AMC from 1000 to 3000 cm−1 (inset shows the magnified portion between 2200 and 3000 for the determination of 2D band), and (e) high-resolution C 1s XPS image of AMC (inset shows the O 1s XPS image).

and hence reducing strain overpotential (increased structure stability) and (2) the enhanced removal of any co-intercalated solvated ions out of the structure (due to the breakdown of SEI layer). To further verify the disorder degree of the synthesized AMC, Raman spectroscopy was conducted, as shown in Figure 2d. The Raman spectrum of AMC revealed two modes, a D peak around 1345 cm−1 and a G peak around 1583 cm−1, due to the breathing modes of rings and the relative motion of sp2 carbon atoms, respectively. ID/IG of AMC was calculated to be ca. 1.01 on the basis of the peak height ratio, which may suggest the disordered state of the sample. The broadening of D band, as shown in Figure 2d, indicates a distribution of clusters with different dimensions and the presence of ring orders other than six,36 which further suggests the disordered nature of AMC. At higher Raman shift (Figure 2d inset), a small signal can be observed at ca. 2900 cm−1, which indicates the presence of insignificant graphitic ordering.37 Thus, on the basis of the collective results from TEM and Raman spectrum, AMC possesses a disordered nature with slight graphitic character. Xray photoelectron spectroscopy (XPS) was conducted for AMC to determine the oxygen functionalities in the sample. AMC shows a typical reduced graphene oxide high-resolution C 1s XPS image (77% sp2 and 13% sp3 C), which is due to the ion bombardment on the sample during the XPS measurement, resulting in the formation of more stable sp2 C from the metastable sp3 C.38 On the basis of the high-resolution C 1s XPS image, AMC contained small proportion of C−OH (6.9%) and C−O (3.4%) functional groups, which were further supported by the high-resolution O 1s XPS image (Figure 2e inset). Thus, the low oxygen atomic concentration (ca. 10%) in the synthesized AMC, as determined by XPS, is essential for constructing an efficient conduction channel and also a lower extent of solvated ion adsorption on the plane. The half-cell assembly of AMC was studied using Li plate as both counter and reference electrodes, and its electrochemical

replacing normal graphite with expanded ones. Structure wise, Li becomes asymmetrically located between the carbon layers when d > 4.4 Å (see SI Figure S2). Hence, on the basis of the simulation result, energy barrier can be greatly reduced when interlayer spacing is at 0.47 nm (greater than graphite), which hints the importance of amorphous carbon in reducing Li+ insertion barrier (strain overpotential and transfer kinetic overpotential). Material Characterization and Electrochemical Performance of AMC Negative Electrode. Morphology of AMC was first studied with scanning electron microscopy (SEM), and the corresponding image is shown in Figure 2a. Necking of particles can be clearly observed with the SEM image (Figure 2a), and such necking among particles can provide a continuous conduction channel, which is beneficial for charge transport. Furthermore, the observed smooth carbon surface is advantageous for the formation of a more efficient solid electrolyte interface (SEI) layer,35 which can provide greater stability during lithiation/delithiation. The necking (interconnection) of nanoparticles is further confirmed when the sample was investigated under low-resolution transmission electron microscopy (TEM), as shown in Figure 2b. The selected area electron diffraction (SAED) pattern of AMC revealed diffused rings, which suggests the disordered state of the synthesized carbon. Under close examination with highresolution TEM, as shown in Figure 2c, relatively well-oriented expanded interlayer spacing with defects can be clearly observed, which hints that the synthesized carbon possesses a disordered structure with slight graphitic character with the interlayer spacing shown in the inset of Figure 2c. The wider interlayer spacing was estimated to be ca. 0.4 nm based on Xray diffraction (XRD), as shown in Figure S3, which is larger than the 0.34 nm graphitic interlayer spacing. Such widened interlayer spacing, which is typical of disordered carbon, is beneficial for (1) the reduction of Li+ insertion barrier energy D

DOI: 10.1021/acsami.7b15473 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces

Figure 3. Electrochemical performance of AMC negative electrode half-cell with a potential window of 0.01−3 V. (a) First three CV curves of AMC conducted at a scan rate of 1 mV s−1, (b) rate performance of AMC with various current densities (0.1−5 A g−1), (c) first three galvanostatic charge/ discharge profile at 0.1 A g−1, (d) energy efficiency with different current densities (0.5, 1, 2, and 5 A g−1) (inset shows voltage hysteresis), and (e) cyclic stability test conducted at 1 A g−1 for 200 cycles and its corresponding Coulombic efficiencies.

in Figure 3c. An initial irreversible capacity loss of 859 mAh g−1 was calculated, which can be attributed to the formation of SEI, and such SEI formation is sensitive to the surface area39,40 or the so-called active surface area41 of the sample. On the basis of Brunauer−Emmett−Teller (BET) N2 adsorption result, AMC possessed a BET surface area of 243 m2 g−1 (Figure S4), which is mainly responsible for the large capacity loss due to SEI layer formation. However, the subsequent cycles gave a Coulombic efficiency of ca. 100% with a reversible discharge capacity of 600 mAh g−1 after three charge/discharge cycles. Two distinct segments can be observed in the galvanostatic charge/discharge profiles, which are, namely, the sloping region at higherpotential region (≥0.9 V) and the plateau region at lowerpotential region (≤0.9 V).42 The sloping region is due to the intercalation of Li+ between turbostratically disordered graphene sheets, whereas the plateau region is due to the formation of small Li metal clusters inside the hard carbon nanopores.43 Energy efficiency of AMC negative electrode was studied by varying the current density, as shown in Figure 3d. AMC was able to achieve energy efficiencies of 54.7, 51.9, 47.8, and 38.2% at 0.5, 1, 2, and 5 A g−1, respectively, compared to the corresponding graphite energy efficiencies of 43.8, 46.7,

performance (conducted under ambient condition) is summarized in Figure 3. The sample was cycled with a potential window of 0.01−3 V, and its cyclic voltammetry (CV) curves are shown in Figure 3a. AMC revealed a cathodic peak, which is characteristic of Li+ insertion into disordered carbon, which is in contrast to the prominent redox peaks in graphite CV curve. The cathodic peak in the first cycle (ca. 0.2 V) is attributed to the decomposition of electrolyte to form an SEI layer on the negative electrode. Rate performance of AMC was studied by increasing current densities from 0.1 to 5 A g−1, as shown in Figure 3b. AMC was able to deliver specific capacities of 579, 469, 390, 327, 276, and 202 mAh g−1 at current densities of 0.1, 0.2, 0.5, 1, 2, and 5 A g−1, respectively. With a 50 times increment in current density (0.1−5 A g−1), an initial capacity retention of 34.9% can be achieved, which indicates the moderate kinetic capability of the negative electrode due to its interconnected conduction pathway. The electrode was also able to return to a specific capacity of 347 mAh g−1 at a current density of 1 A g−1, which indicates excellent reversibility. As disordered carbon is well known to exhibit large initial irreversibility due to the formation of SEI layer, the first three galvanostatic charge/discharge profiles of AMC are presented E

DOI: 10.1021/acsami.7b15473 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces

Figure 4. Material characterization of PdCS. (a) SEM image, (b) low-resolution TEM image (inset shows the SAED pattern), (c) high-resolution TEM image showing the interlayer spacing (inset shows a magnified portion of interlayer spacing), (d) Raman spectra of PdCS from 1000 to 3000 cm−1 (inset shows the magnified portion between 2200 and 3000 for the determination of 2D band), and (e) high-resolution C 1s XPS image of PdCS (inset shows the O 1s XPS image).

was also conducted, and the low-resolution TEM image shown in Figure 4b indicates a high density of nanosheets interconnecting one another to form the porous structure, which further supports the ultrathin sheetlike structure of PdCS, as observed in the SEM. The weak diffusive rings in the SAED pattern (Figure 4b inset) suggests the disordered nature of PdCS. Raman spectroscopy was also conducted to further verify the disordered nature of the sample. As shown in Figure 4d, weak D and G peaks can be observed at 1340 and 1581 cm−1 in the Raman spectrum. Even though ID/IG does not give a clear indication of the disorder degree of the sample, a low 0.98 ratio may hint the disordered state of the sample.36 In addition, the broadening of D band is indicative of increasing disorder in the structure.36 At higher Raman shift (Figure 4d inset), no visible peak can be observed, which further indicates the disordered nature of PdCS. As it is commonly consented that a low oxygen functionality in electrode is important to reduce the oxidation of electrolyte,44,45 dehydration of organic precursor is one of the strategies to decrease the number of functional groups.46 Thus, to investigate the oxygen functionalities in PdCS so as to study its stability as supercapacitor electrode, XPS was conducted. A high-resolution C 1s spectrum is shown in Figure 4e, which revealed the presence of C−OH and CO functionalities, which is in good agreement with the high-resolution O 1s spectrum (Figure 4e inset). On the basis of the XPS result, PdCS possesses a low oxygen functionality of 2.8%, which is important for minimizing electrolyte oxidation. Another important aspect of supercapacitor electrode is to possess high surface area with nanopores