Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Nano Lett. Author manuscript; available in PMC 2013 July 11.
Published in final edited form as:
PMCID: PMC3491980

Voltage-gated ion transport through semiconducting conical nanopores formed by metal nanoparticle assisted plasma etching


Nanopores with conical geometries have been found to rectify ionic current in electrolytes. While nanopores in semiconducting membranes are known to modulate ionic transport through gated modification of pore surface charge, the fabrication of conical nanopores in silicon (Si) has proven challenging. Here, we report the discovery that gold (Au) nanoparticle (NP) assisted plasma etching results in the formation of conical etch profiles in Si. These conical profiles result due to enhanced Si etch rates in the vicinity of the Au NPs. We show that this process provides a convenient and versatile means to fabricate conical nanopores in Si membranes and crystals with variable pore-diameters and coneangles. We investigated ionic transport through these pores and observed that rectification ratios could be enhanced by a factor of over 100 by voltage gating alone, and that these pores could function as ionic switches with high on-off ratios of approximately 260. Further, we demonstrate voltage gated control over protein transport, which is of importance in lab-on-a-chip devices and biomolecular separations.

Keywords: nanofluidics, membranes, separations, ion channels, ionic circuits, electrochemistry

Solid-state nanopores have attracted a great deal of scientific interest due to their possible use in replicating some behaviors observed in naturally occurring trans-membrane protein channels13. In order to mimic the complex chemo-electrical control that is observed in protein channels4, 5, there is a need to incorporate mechanisms for active control of ion transport through solid state nanopores6, 7. Nanopores formed in semiconducting substrates such as silicon (Si) offer the possibility for voltage-gated control of transmembrane ionic and molecular transport8, 9. Furthermore, nanopores in Si membranes are mechanically robust, biocompatible, able to tolerate a wide variety of environmental conditions and permit facile integration with existing Si-based electronic and micromechanical devices10. When considering the effect of geometry on nanopore transport characteristics, conical nanopores have been found to exhibit unique ion transport properties as a consequence of their asymmetric geometry11. For example, the ionic rectification behavior observed in conical nanopores, due to surface charge density asymmetry12, can be utilized to develop ionic analogs of electronic devices13, 14. Further, for nanopores of conical geometry exhibit lower electrical resistance in comparison to those with cylindrical geometries of the same tip diameter, making them more suitable for molecular sensing applications, as these pores generate higher ionic currents15. Hence, conical nanopores in semiconducting substrates are attractive for a wide range of applications including the creation of synthetic analogous of biological nanochannels16, ionic logic circuitry17, 18 and actively controllable molecular separation platforms19. However, the generation of conical nanopores in single crystal Si, to enable gated, active control, is challenging using existing methodologies.

There exist a variety of methods to create nanopores. These include the widely utilized track etching method, which enables the creation of conical nanopores in polymeric thin films through ion bombardment and subsequent chemical etching12. However, this method is restricted to homogenous dielectrics and also requires the use of MeV to GeV ion sources20. Although wet chemical etching of e-beam lithography patterned Si has been used to form nanopores, it offers limited control over the pore geometry in terms of cone angles and often results in faceted pores due to the varying etch rates along the different Si crystallographic planes2124. Here, we describe the process of nanopore formation in Si substrates, that relies on an enhancement of the dry etch rate observed in the presence of Au nanoparticles (NPs)25. The process can be implemented in a maskless manner and does not require lithographic patterning. Further, the etch process allows considerable control over pore geometry and the application of a voltage to nanoporous Si membranes enables active control over small ion and protein transport.

We discovered that CF4-O2 plasma etching of single crystal Si substrates in the vicinity of Au NPs resulted in a conical etch profile (Figure 1). In order to characterize the process, we utilized silicon on insulator (SOI) wafers with different Si device layer thicknesses. We first dispersed Au NPs on the surface of the Si device layer and then plasma etched and released the Si membranes by dissolving the buried oxide layer in hydrofluoric acid (HF) (Figure 2a and details in supporting information). It is also noteworthy that this process is versatile in that Au NPs can also be dispersed on flexible Si membranes or Si powder crystals so that such conical nanopores can be formed on these substrates (Figure 2b–d). While dispersal of NPs can be used to form conical nanopores in a lithography-free manner, our process is also compatible with lithographically defined Au patterns which may be required in applications that require precise positioning or ordered arrays of nanopores. We were able to utilize both e-beam lithography to pattern individual Au discs which could be plasma etched to form precisely positioned isolated nanopores as well as nanoimprint lithography (NIL) to define arrays of Au nano-discs (Figure 2e), which could also be plasma etched to create large-area ordered nanopore arrays in Si membranes (Figure 2f).

Figure 1
Schematic illustration and corresponding SEM images of conical nanopore formation during plasma etching of a Si substrate in the vicinity of an Au NP
Figure 2
Versatility of conical nanopore formation in different Si substrates

In order to optimize the etching conditions, we first explored the effects of varying CF4/O2 ratio on bare Si wafers and observed that a ratio of 25 sccm CF4 and 4 sccm O2 resulted in optimum etching (details in supporting information). Further, we estimate that the Au NPs enhance the etch rate of Si by over an order of magnitude, with measured increases in Si etch rate from ~110 nm/min to as high as ~4000 nm/min, and the etch rate was sensitive to the type of etcher used and the density of the nanoparticles. It is noteworthy that enhancements in wet etching rates of Si have been observed using a metal assisted chemical etching (MaCE) process26, 27. In the MaCE process, the Si substrate with a noble metal mask is wet etched with hydrofluoric acid (HF) and an oxidative agent, typically hydrogen peroxide (H2O2). During etching, the Si beneath the noble metal is etched much faster than the uncovered Si. The enhanced etching of Si at the Si/metal interface has been attributed to preferential reduction of the oxidant at the surface of the noble metal with associated hole injection into the Si26, 28.

In contrast, our mechanism for the formation of conical etch-profiles is based on the accelerated dry etching of Si in the vicinity of Au NPs25. It has been shown that the presence of Au or its oxide increases the etch rate of Si in CF4-O2 plasma29; this acceleration has been attributed to a metal-catalyzed increase in the concentration of reactive fluorine radicals (F*)30, 31. We developed a theoretical model (details in supporting information) that captures the effects of an enhanced etch rate in the vicinity of the Au NP and predicts a conical etch profile (Figure 3a). From the model, the cone half-angle (θ) can be related to the etch rates and etch anisotropy (k) as,

(Equation 1)
Figure 3
Model and experimental control of nanopore cone angle and size

Experimentally, we varied the etch power and found that higher RF powers generated wider cone angles (Figure 3b; results in supporting information). We observed that the conical tip radii of nanopores formed using Au NPs of 50 nm diameter were smaller than those formed using 100 nm particles and that the radii increased linearly with time (Figure 3c). The smallest pore diameter that we have reproducibly achieved is around 20 nm. Since the etch rate of Si is rapid, reproducible formation of even smaller pores would require further optimization in terms of tuning the etch rate as well as the Si device layer thickness.

We characterized the NP dispersed Si surface after plasma etching using Fourier Transform Infra-Red (FTIR) spectroscopy, X-Ray Photoelectron Spectroscopy (XPS) and contact angle measurements (details in supporting information). Our characterization studies suggest that the Si surface is coated with a dielectric layer, about 1.5 nm thick, composed largely of oxidized silicon. Such oxidized Si surfaces (pI ~ 2) are known to exhibit a negative surface charge when in contact with aqueous electrolyte solutions of pH > 3 due to de-protonation of the Si-OH surface groups (pKa ~ 6.9)32, 33.

In order to explore the use of these conical nanopores in the creation of bio-mimetic ion channels, we measured the transmembrane current across a single plasma etched nanopore (Figure 4). The nanopore was fabricated in a 20 μm thick Si device layer with a pore base and tip diameter of approximately 2.5 μm and 43 nm respectively (Figure 4a, see supporting information for fabrication details). We attribute the origin of the observed rectification to the conical asymmetric geometry and the negative charge present on the pore surface13, 3436 (Figure 4b). In addition, we observed that the rectification ratio (f) decreased from 3.9 to 2.3 when the KCl electrolyte concentration increased from 20 mM to 100 mM (Figure 4c). We rationalize this observation by noting that theoretical predictions based on Poisson-Nernst-Planck equations suggest that rectification occurs when the electrical double layer widths become comparable to the charged conical nanopore dimensions3739. As the concentration of the electrolyte and consequently ionic strength increases, the double layer width decreases relative to the pore size and hence the pore becomes a less effective rectifier36, 37. Further, at a fixed electrolyte concentration of 20 mM KCl, we observed that the rectification ratio also varied with pH (Figure 4d). Significantly, the rectification ratio was found to decrease from f = 4.2 at pH 8 to f = 1.1 at pH 2, indicating that the surface charge of the pore walls play a major role in the rectification behavior. The low degree of rectification at low pH is indicative of the neutralization of the negative surface charges present on the pore wall, due to protonation. Such a pH sensitive behavior of Electrolyte Insulator Semiconductor (EIS) systems consisting of Si, oxidized silicon and aqueous electrolyte have been well previously described in the context of Ion Selective Field Effect Transistors (ISFETs)33, 40.

Figure 4
Ion-current rectification observed in an individual conical nanopore

One significant advantage of semiconducting conical nanopores is that the space charge at the semiconductor interface can be manipulated by the application of a voltage. We measured the I-V characteristics of transmembrane ionic transport in 20 mM KCl and observed a variation in the ionic current as a function of the voltage applied to the Si membrane (Figure 5). Using a transistor analogy, we plotted the ionic current (ID) as a function of drain voltage (VD) for a fixed gate voltage (VG) (Figure 5c). In contrast to manipulating transmembrane ion transport by varying the size, shape or surface charge of the pore using various strategies such as thiolated DNA molecules41, polyelectrolyte deposition42 and pH responsive polymer brushes43, 44, we observe that active control can be achieved by voltage gating. We found the variation to be more pronounced at negative VG. For example, at VG = −2 V, ID was increased by a factor of six as compared to that when no gate voltage was applied, while there was a negligible effect at positive voltages, as expected. We rationalize this result by noting that increasing the negative gate bias on the Si enhances the negative space charge within the semiconducting Si surface (as verified by Kelvin probe microscopy; details in supporting information) resulting in higher transport rate of K+ ions through the more negatively charged pore.

Figure 5
Voltage-gated switching of ionic current through an individual conical nanopore

We note that this non-linear I-V behavior can be used to enable ionic switching (ON/OFF) devices (Figure 5d) of relevance to ionic logic circuitry. Here, a square wave pulse of amplitude −2 V to +2 V was applied to the gate electrode and the transmembrane current was measured. The nanopore device was found to continuously switch between ON and OFF states at negative gate bias (−2 V) and positive gate bias (+2 V) respectively (Figure 5e). We attribute the high noise in the ON state of the device to the combined effects of fluctuations in the pore surface charge, dynamic electrowetting phenomena in the porewall and the thermal fluctuations in the conductivity of the salt solution4549. We estimate that the rectification ratios in our devices could be readily varied across two orders of magnitude by voltage gating alone and the ionic switching measurement revealed an ionic current on/off ratio of approximately 260. Thus, analogous to biological membranes, active control over transmembrane ion transport can be achieved by the application of electric potential to Si nanoporous membrane.

In addition to gate modulated rectification properties of small ions, we observed that active control over permeation of charged biomolecules can also be achieved through semiconducting nanoporous membranes, which is important in lab-on-a-chip devices and biomolecular separations50. We studied the time dependent permeation of fluorescently labeled Bovine Serum Albumin (BSA) molecules through an approximately 4 mm diameter nanoporous region of the membrane with and without the application of a voltage (Figure 6a) (see supporting information for details). Importantly, as shown in the plot of fluorescence intensity with time (Figure 6b), we observed significantly enhanced transport at +0.3 V bias as compared to that at 0 V and −0.3 V. We attribute this change to electrostatic effects of the negatively charged BSA (pI 4.7), and enhanced diffusion at positive voltages. In order to estimate the significance of this difference, we developed a transport model (details in supporting information) which suggests that this difference is equivalent to an increase in the effective diffusion coefficient of BSA by over an order of magnitude when either +0.3 V or −0.3 V is applied.

Figure 6
Voltage control of protein transport

In summary, we have uncovered a convenient and versatile process to form conical nanopores in semiconducting, single crystal Si substrates. The process is simple, can be utilized in a maskless, lithography free manner which increases its accessibility. Additionally, semiconducting nanopores allow facile control over transmembrane ionic and molecular transport and it is conceivable that the incorporation of gate dielectrics could enable ionic field effect transistors. Since the process is versatile and compatible with MEMS / CMOS processes, these voltage gated ionic devices could be integrated with electronic and micromechanical structures.

Supplementary Material



This work was supported by the NSF CMMI 0854881 and the NIH Director's New Innovator Award Program, part of the NIH Roadmap for Medical Research, through grant number DP2-OD004346-01. Information about the NIH Roadmap can be found at We are grateful to George P. Watson and Joseph Palmer of PRISM (Princeton Institute for the Science and Technology of Materials) clean room facility for their help with device fabrication and valuable discussions. We acknowledge technical assistance and suggestions from Gary Johns, Nina Markovic, Michael Barclay, Howard Fairbrother and Joyce Breger. We also acknowledge use of the Surface Analysis lab at JHU (Department of Materials Science and Engineering).


1. Dekker C. Nat. Nanotechnol. 2007;2(4):209–215. [PubMed]
2. Hou X, Guo W, Jiang L. Chem. Soc. Rev. 2011;40(5):2385–2401. [PubMed]
3. Jovanovic-Talisman T, Tetenbaum-Novatt J, McKenney AS, Zilman A, Peters R, Rout MP, Chait BT. Nature. 2009;457(7232):1023–1027. [PMC free article] [PubMed]
4. Majd S, Yusko EC, Billeh YN, Macrae MX, Yang J, Mayer M. Curr. Opin. Biotechnol. 2010;21(4):439–476. [PMC free article] [PubMed]
5. Sigworth FJ. Nature. 2003;423(6935):21–22. [PubMed]
6. Karnik R, Castelino K, Fan R, Yang P, Majumdar A. Nano Lett. 2005;5(9):1638–1642. [PubMed]
7. Siwy ZS, Howorka S. Chem. Soc. Rev. 2010;39(3):1115–1132. [PubMed]
8. Gracheva ME, Vidal J, Leburton J-P. Nano Lett. 2007;7(6):1717–1722. [PMC free article] [PubMed]
9. Vidal J, Gracheva ME, Leburton JP. Nanoscale. Res. Lett. 2007;2(2):61–68.
10. van den Berg A, Wessling M. Nature. 2007;445(7129):726–726. [PubMed]
11. Sexton LT, Horne LP, Martin CR. Mol. BioSyst. 2007;3(10):667–685. [PubMed]
12. Apel PY, Korchev YE, Siwy Z, Spohr R, Yoshida M. Nucl. Instrum. Methods Phys. Res. Sect. B. 2001;184(3):337–346.
13. Zhou K, Perry JM, Jacobson SC. Annu. Rev. Anal. Chem. 2011;4:321–341. [PubMed]
14. Guan W, Fan R, Reed MA. Nat. Commun. 2011;2:506. [PubMed]
15. Mara A, Siwy Z, Trautmann C, Wan J, Kamme F. Nano Lett. 2004;4(3):497–501.
16. Martin CR, Siwy ZS. Science. 2007;317(5836):331–332. [PubMed]
17. Zhong C, Deng Y, Roudsari AF, Kapetanovic A, Anantram MP, Rolandi M. Nat. Commun. 2011;2:476. [PubMed]
18. Mafe S, Manzanares JA, Ramirez P. J. Phy. Chem. C. 2010;114(49):21287–21290.
19. Kovarik ML, Jacobson SC. Anal. Chem. 2009;81(17):7133–7140. [PubMed]
20. Spohr R. Radiat. Meas. 2005;40(2–6):191–202.
21. Storm AJ, Chen JH, Ling XS, Zandbergen HW, Dekker C. Nat. Mater. 2003;2(8):537–540. [PubMed]
22. Striemer CC, Gaborski TR, McGrath JL, Fauchet PM. Nature. 2007;445(7129):749–753. [PubMed]
23. Park SR, Peng H, Ling XS. Small. 2007;3(1):116–119. [PubMed]
24. Bean KE. IEEE Trans. Electron. Devices. 1978;ED-25(10):1185–1193.
25. James T, Cho JH, Fernandes R, Randhawa JS, Gracias DH. Anal. Bioanal. Chem. 2010;398(7–8):2949–2954. [PubMed]
26. Huang Z, Geyer N, Werner P, de Boor J, Gösele U. Adv. Mater. 2011;23(2):285–308. [PubMed]
27. Rykaczewski K, Hildreth OJ, Wong CP, Fedorov AG, Scott JHJ. Nano Lett. 2011;11(6):2369–2374. [PubMed]
28. Tsujino K, Matsumura M. Adv. Mater. 2005;17(8):1045–1047.
29. Kataoka Y, Shinmura T, Kanoh M. J. Vac. Sci. Technol. A. 2000;18(2):388–392.
30. Fedynyshyn TH, Grynkewich GW, Chen BA, Ma TP. J. Electrochem. Soc. 1989;136(6):1799–1804.
31. Fedynyshyn TH, Grynkewich GW, Dumas RH. J. Electrochem. Soc. 1988;135(1):268–269.
32. Bousse L, De Rooij NF, Bergveld P. IEEE Trans. Electron. Devices. 1983;ED-30(10):1263–1270.
33. van Hal REG, Eijkel JCT, Bergveld P. Adv. Colloid Interface Sci. 1996;69(1–3):31–62.
34. Siwy ZS. Adv. Funct. Mater. 2006;16(6):735–746.
35. Siwy Z, Heins E, Harrell CC, Kohli P, Martin CR. J. Am. Chem. Soc. 2004;126(35):10850–10851. [PubMed]
36. Liu Q, Wang Y, Guo W, Ji H, Xue J, Ouyang Q. Phys. Rev. E. 2007;75:051201. [PubMed]
37. Daiguji H, Yang P, Majumdar A. Nano Lett. 2004;4(1):137–142.
38. Cervera J, Schiedt B, Ramírez P. Europhys. Lett. 2005;71(1):35–41.
39. Ai Y, Zhang M, Joo SW, Cheney MA, Qian S. J. Phys. Chem. C. 2010;114(9):3883–3890.
40. Fung CD, Cheung PW, Ko WH. IEEE Trans. Electron. Devices. 1986;ED-33(1):8–18.
41. Harrell CC, Kohli P, Siwy Z, Martin CR. J. Am. Chem. Soc. 2004;126(48):15646–15647. [PubMed]
42. Ali M, Yameen B, Cervera J, Ramírez P, Neumann R, Ensinger W, Knoll W, Azzaroni O. J. Am. Chem. Soc. 2010;132(24):8338–8348. [PubMed]
43. Yameen B, Ali M, Neumann R, Ensinger W, Knoll W, Azzaroni O. J. Am. Chem. Soc. 2009;131(6):2070–2071. [PubMed]
44. Yameen B, Ali M, Neumann R, Ensinger W, Knoll W, Azzaroni O. Nano Lett. 2009;9(7):2788–2793. [PubMed]
45. Smeets RMM, Dekker NH, Dekker C. Nanotechnol. 2009;20:9. [PubMed]
46. Smeets RMM, Keyser UF, Dekker NH, Dekker C. Proc. Natl. Acad. Sci. U.S. A. 2008;105(2):417–421. [PubMed]
47. Wang L, Sun L, Wang C, Chen L, Cao L, Hu G, Xue J, Wang Y. J. Phys. Chem. C. 2011;115(46):22736–22741.
48. Hoogerheide DP, Garaj S, Golovchenko JA. Phys. Rev. Lett. 2009;102:25. [PMC free article] [PubMed]
49. Smirnov SN, Vlassiouk IV, Lavrik NV. ACS Nano. 2011;5(9):7453–7461. [PubMed]
50. Karnik R, Castelino K, Majumdar A. Appl. Phys. Lett. 2006;88(12):123114/3.