Carbon Nanotube Computer

H.-S. Philip Wong (left) & Subhasish Mitra (right) led the team of researchers that developed the carbon nanotube computer

Authors:
Max M. Shulaker¹, Gage Hills¹, Nishant Patil¹, Hai Wei¹, Hong-Yu Chen¹, 
H.-S. Philip Wong¹, and Subhasish Mitra^1,2

Affiliation:
¹Department of Electrical Engineering, Stanford University, USA
²Department of Computer Science, Stanford University, USA

The miniaturization of electronic devices has been the principal driving force behind the semiconductor industry, and has brought about major improvements in computational power and energy efficiency. Although advances with silicon-based electronics continue to be made, alternative technologies are being explored. Digital circuits based on transistors fabricated from carbon nanotubes (CNTs) have the potential to outperform silicon by improving the energy–delay product, a metric of energy efficiency, by more than an order of magnitude. Hence, CNTs are an exciting complement to existing semiconductor technologies [1,2].

Owing to substantial fundamental imperfections inherent in CNTs, however, only very basic circuit blocks have been demonstrated. Here we show how these imperfections can be overcome, and demonstrate the first computer built entirely using CNT-based transistors. The CNT computer runs an operating system that is capable of multitasking: as a demonstration, we perform counting and integer-sorting simultaneously. In addition, we implement 20 different instructions from the commercial MIPS instruction set to demonstrate the generality of our CNT computer. This experimental demonstration is the most complex carbon-based electronic system yet realized. It is a considerable advance because CNTs are prominent among a variety of emerging technologies that are being considered for the next generation of highly energy-efficient electronic systems [3, 4].

Since the initial discovery of CNTs, there have been several major milestones for CNT technologies [6]: Carbon nanotube field effect transistors (CNFETs), basic circuit elements (logic gates), a five-stage ring oscillator fabricated along a single CNT, a percolation-transport-based decoder, stand-alone circuit elements such as half-adder sum generators and D-latches, and a capacitive sensor interface circuit [7, 8, 9, 10, 11, 12, 13]. Yet there remains a serious gap between these circuit demonstrations for this emerging technology and the first computers built using silicon transistors, such as the Intel 4004 and the VAX-11 (1970s). These silicon-based computers were fundamentally different from the above-mentioned CNFET-based circuits in several key ways: they ran stored programs, they were programmable (meaning that they could execute a variety of computational tasks through proper sequencing of instructions without modifying the underlying hardware [14]) and they implemented synchronous digital systems incorporating combinational logic circuits interfaced with sequential elements such as latches and flip-flops [15].

It is well known that substantial imperfections inherent in CNT technology are the main obstacles to the demonstration of robust and complex CNFET circuits [16]. These include mis-positioned and metallic CNTs. Mis-positioned CNTs create stray conducting paths leading to incorrect logic functionality, whereas metallic CNTs have little or no bandgap, resulting in high leakage currents and incorrect logic functionality [17]. The imperfection-immune design paradigm, which combines circuit design techniques with CNT processing solutions, overcomes these problems [17, 18]. It enables us to demonstrate, for the first time, a complete CNT computer, realized entirely using CNFETs. Similar to the first silicon-based computers, our CNT computer, which is a synchronous digital system built entirely from CNFETs, runs stored programs and is programmable. Our CNT computer runs a basic operating system that performs multitasking, meaning that it can execute multiple programs concurrently (in an interleaved fashion). We demonstrate our CNT computer by concurrently executing a counting program and an integer-sorting program (coordinated by a basic multitasking operating system), and also by executing 20 different instructions from the commercial MIPS instruction set [19].

The CNT computer is a one-instruction-set computer, implementing the SUBNEG (subtract and branch if negative) instruction, inspired by early work in ref. 23. We implement the SUBNEG instruction because it is Turing complete and thus can be used to re-encode and perform any arbitrary instruction from any instruction-set architecture, albeit at the expense of execution time and memory space [21, 22]. The SUBNEG instruction is composed of three operands: two data addresses and a third partial next instruction address (the CNT computer itself completes the next instruction address, allowing for branching to different instruction addresses). The SUBNEG instruction subtracts the value of the data stored in the first data address from the value of the data stored in the second data address, and writes the result at the location of the second data address.

The next instruction address is calculated to be one of two possible branch locations, depending on whether the result of the subtraction is negative. The partial next instruction address given by the present SUBNEG instruction omits the least significant bit. The least significant bit is calculated by the CNT computer, on the basis of whether or not the result of the SUBNEG subtraction was negative. This bit, concatenated with the partial next instruction address given in the SUBNEG instruction, makes up the entire next instruction address. A diagram showing the SUBNEG implementation is shown in Fig. 1a.

Figure 1: (Click on the image to view higher resolution) SUBNEG and program implementation. (a) Flowchart showing the implementation of the SUBNEG instruction. (b) Sample program on CNT computer. Each row of the chart is a full SUBNEG instruction. It is composed of two data addresses and a partial next instruction address. The (omitted) least significant bit (LSB) of the next instruction address is calculated by the arithmetic unit of the CNT computer, and the most significant bit (MSB) of the next instruction address indicates the running program, either a counter or bubble-sort algorithm in this instance.

The CNFET computer is composed of 178 CNFETs, with each CNFET comprising ~10–200 CNTs, depending on relative sizing of the widths of the CNFETs. It follows the von Neumann architecture and the convention of most computers today, with the instruction and data memories implemented off-chip (these memories perform no other function than a single read or write in a clock cycle). The computer circuitry contains up to seven stages of cascaded logic, demonstrating our ability to cascade combinational logic stages, which is a necessity in realizing large digital systems. Transistor-level schematics of the subcomponents (D-latches and arithmetic units) are provided in ref. 24, along with a block-level diagram of the full CNT computer. Ref. 24 also provides details of the tasks performed by the CNT computer (instruction fetch, data fetch, arithmetic operation, and write-back), as well as details of its operating system.

The CNT-specific fabrication process is based on the process described in refs 18, 20, 23, 24. Importantly, the fabrication process is completely silicon-CMOS compatible owing to its low thermal budget (125 °C). We use standard cells for our subsystems, designed following the imperfection-immune methodology, which renders our circuits immune to both mis-positioned and metallic CNTs. Because this method ensures that the immunity to CNT imperfections is encapsulated entirely within standard cells, the fabrication is completely insensitive to the exact positioning of CNTs on the wafer and there is no per-unit customization, rendering our processing and design VLSI (very large-scale integration) compatible. The entire CNT computer is fabricated completely within a die on a single wafer. Each die contains five CNT computers, and each wafer contains 197 dies. There is no customization of any sort after circuit fabrication: all of the CNFETs and interconnects are predetermined during design, and there is no post-fabrication selection, configuration or fine-tuning of functional CNFETs. Just like any von Neumann computer, off-chip interconnects are used for connections to external memories. Our CNT-specific fabrication process and imperfection-immune design enables high yield and robust devices; the yield of the subsystems, such as D-latches, typically ranges from 80% to 90% (waveforms of 240 subsystems, including arithmetic logic units and D-latches, are shown in ref. 24). The primary causes of yield loss—particles resulting in broken lithography patterns, adhesion issues with metal lift-off and variations in machine etch rates—are consequences of the limitations of performing all fabrication steps in-house in an academic fabrication facility.

A scanning electron microscopy (SEM) image of a fabricated CNT computer is shown in Fig. 2. To demonstrate the working CNT computer, we perform multitasking with our basic operating system, concurrently running a counter program and an integer-sorting program (performing the bubble-sort algorithm). Although CNFET circuits promise improved speed [2, 4, 5], our computer runs at 1 kHz. This is not due to the limitations of the CNT technology or our design methodology, but instead is caused by capacitive loading introduced by the measurement setup, the 1-μm minimum lithographic feature size possible in our academic fabrication facility, and CNT density and contact resistance. The measured and expected outputs from the CNT computer (Fig. 2b) show correct operation. To demonstrate the flexibility and ability of the SUBNEG computer to implement any arbitrary instruction, we additionally perform 20 MIPS instructions (Fig. 2c) on the CNT computer. Although the CNT computer operates on single-bit data values, this is not a fundamental limitation, because any multibit computation can be performed with a single-bit computer through serial computation [20].

Figure 2: (Click on the image to view higher resolution) CNT computer results. (a) Scanning electron microscopy (SEM) image of an entire CNT computer. (b) Measured and expected output waveforms for a CNT computer, running the program shown in Fig. 1b. The exact match in logic value of the measured and expected output shows correct operation. As shown by the MSB (denoted [4]) of the next instruction address, the computer is switching between performing counting and sorting (bubble-sort algorithm). The running results of the counting and sorting are shown in the rows beneath the MSB of the next instruction address. (c) A list of the 20 MIPS instructions implemented and tested on the CNT computer.

We have reported a CNT computer fabricated entirely from CNFETs, and have demonstrated its ability to run programs, to run a basic operating system that performs multitasking, and to execute MIPS instructions. To achieve this we used the imperfection-immune design methodology and developed robust and repeatable CNT-specific design and processing. This demonstration confirms that CNFET-based circuits are a feasible and plausible emerging technology.

Acknowledgments: We acknowledge the support of the NSF (CISE), FCRP C2S2, FCRP FENA, STARNet SONIC and the Stanford Graduate Fellowship and the Hertz Foundation Fellowship (M.M.S.). We also acknowledge Z. Bao, A. Lin, H. (D.) Lin, M. Rosenblum, and J. Zhang for their advice and collaborations.

References:
[1] Aaron D. Franklin, Mathieu Luisier, Shu-Jen Han, George Tulevski, Chris M. Breslin, Lynne Gignac, Mark S. Lundstrom, Wilfried Haensch. "Sub-10 nm carbon nanotube transistor". Nano Letters, 12, 758–762 (2012). Abstract.
[2] Lan Wei, D.J. Frank, L. Chang, H.-S.P. Wong. "A non-iterative compact model for carbon nanotube FETs incorporating source exhaustion effects" in Proc. 2009 IEEE Intl Electron Devices Meeting 917–920 (2009). Abstract.
[3] L. Chang in Short Course IEEE Intl Electron Devices Meeting (IEEE, 2012).
[4] D.E. Nikonov, & I.A. Young, "Uniform methodology for benchmarking beyond-CMOS logic devices" in Proc. 2012 IEEE Intl Electron Devices Meeting 24–25 (2012). Abstract.
[5] Jie Deng, N. Patil, Koungmin Ryu, A. Badmaev, Zhou Chongwu, S. Mitra, H.-S.P. Wong, "Carbon Nanotube Transistor Circuits: Circuit-Level Performance Benchmarking and Design Options for Living with Imperfections" in Proc. 2007 IEEE Intl Solid State Circuits Conf. 70–78 (IEEE, 2007). Abstract.
[6] Suio Iijima,  "Helical microtubules of graphitic carbon". Nature, 354, 56–58 (1991). Abstract.
[7] R. Martel, T. Schmidt, H. R. Shea, T. Hertel, Ph. Avouris. "Single-and multi-wall carbon nanotube field-effect transistors". Applied Physics Letters, 73, 2447 (1998). Abstract.
[8] Sander J. Tans, Alwin R. M. Verschueren, Cees Dekker, "Room-temperature transistor based on a single carbon nanotube". Nature, 393, 49–52 (1998). Abstract.
[9] Zhihong Chen, Joerg Appenzeller, Yu-Ming Lin, Jennifer Sippel-Oakley, Andrew G. Rinzler, Jinyao Tang, Shalom J. Wind, Paul M. Solomon, Phaedon Avouris. "An integrated logic circuit assembled on a single carbon nanotube". Science, 311, 1735 (2006). Abstract.
[10] Qing Cao, Hoon-sik Kim, Ninad Pimparkar, Jaydeep P. Kulkarni, Congjun Wang, Moonsub Shim, Kaushik Roy, Muhammad A. Alam, John A. Rogers. "Medium-scale carbon nanotube thin-film integrated circuits on flexible plastic substrates". Nature, 454, 495–500 (2008). Abstract.
[11] N. Patil, A. Lin, E.R. Myers, H.-S. P. Wong, S. Mitra, in Proc. Symp. VLSI Tech. 205–206. (2008).
[12] N. Patil, A. Lin, Jie Zhang, Hai Wei, K. Anderson, H.-S.P. Wong, S. Mitra. "Scalable carbon nanotube computational and storage circuits immune to metallic and mis-positioned carbon nanotubes". IEEE Transactions on NanoTechnology, 10, 744–750 (2011). Abstract.
[13] M. Shulaker, et al. in Proc. 2013 IEEE Intl Solid State Circuits Conf. 112–113 (IEEE, 2013).
[14] J. von Neumann, "First draft of a report on the EDVAC". IEEE Annals of the History of Computing, 15, 27–75 (1993). Abstract.
[15] Edward J. McCluskey, "Logic Design Principles with Emphasis on Testable Semicustom Circuits" (Prentice-Hall, 1986).
[16] Qing Cao, Shu-jen Han, George S. Tulevski, Yu Zhu, Darsen D. Lu, Wilfried Haensch, "Arrays of single-walled carbon nanotubes with full surface coverage for high-performance electronics". Nature Nanotechnology, 8, 180–186 (2013). Abstract.
[17] Jie Zhang, A. Lin, N. Patil, Wei Hai, Wei Lan, H.-S.P. Wong, S. Mitra."Carbon Nanotube Robust digital VLSI". IEEE Transactions on CAD, 31, 453–471 (2012) . Abstract.
[18] Nishant Patil, "Design and Fabrication of Imperfection-Immune Carbon Nanotube Digital VLSI Circuits". PhD thesis, Stanford Univ. (2010). Full Thesis.
[19] J. L. Hennessy, D. A. Patterson, "Computer Architecture: A Quantitative Approach" (Kaufmann, 1990).
[20] Albert Lin, "Carbon Nanotube Synthesis, Device Fabrication, and Circuit Design for Digital Logic Applications". PhD thesis, Stanford Univ. (2010). Link.
[21] Rolf Herken, ed. "The Universal Turing Machine: A Half-Century Survey" (Springer, 1995).
[22] Peter J. Nürnberg, Uffe K. Wiil, David L. Hicks, "A grand unified theory for structural computing". Metainformatics, 3002, 1–16 (2004). Abstract.
[23] Max M. Shulaker, Hai Wei, Nishant Patil, J. Provine, Hong-Yu Chen, H.-S. P. Wong, Subhasish Mitra, "Linear increases in carbon nanotube density through multiple transfer technique". Nano Letters, 11, 1881–1886 (2011). Abstract.
[24] Max M. Shulaker, Gage Hills, Nishant Patil, Hai Wei, Hong-Yu Chen, H.-S. Philip Wong, Subhasish Mitra. "Carbon Nanotube computer". Nature, 501, 526–530 (2013). Abstract.

Combining Infrared Spectroscopy and Scanning Tunneling Microscopy

Graduate students working on the project (from left to right): Xiaoping Hong, Giang D. Nguyen, Ivan V. Pechenezhskiy

Authors:
Ivan V. Pechenezhskiy^1,2, Xiaoping Hong¹, Giang D. Nguyen¹, Jeremy E. P. Dahl³, Robert M. K. Carlson³, Feng Wang¹, and Michael F. Crommie^1,2

Affiliation:
¹Department of Physics, University of California at Berkeley, California, USA
²Materials Sciences Division, Lawrence Berkeley National Laboratory, California, USA
³Stanford Institute for Materials and Energy Science, Stanford University, California, USA

Scanning tunneling microscopy (STM) [1] is an outstanding tool for probing electronic structures and surface morphologies at the nanoscale. In a scanning tunneling microscope, a metallic tip moves above the surface in a raster-like manner and measures variations in the quantum mechanical tunneling current that exists in a nanometer-sized vacuum gap between the tip and the surface. These days STM is routinely used to explore metallic, semiconducting, and superconducting substrates, novel two-dimensional materials, as well as atomic and molecular adsorbates on these surfaces.

While STM allows us to image surfaces with unparalleled atomic resolution, one significant drawback of STM is the lack of chemical contrast. For the purpose of chemical recognition, however, vibrational spectroscopy (specifically infrared spectroscopy) can be used because precise knowledge of molecular vibrational modes often leads to identification of the corresponding molecular structures. A significant breakthrough in probing molecular vibrations with STM was made in 1998 with the invention of STM-based inelastic electron-tunneling spectroscopy (STM-IETS) [2]. However, STM-IETS has some disadvantages, such as its relatively poor spectral resolution which is dependent on temperature and bias modulation [3]. Therefore, combining STM with infrared spectroscopy to probe molecular vibrations is still an extremely appealing idea, and a successful combination has so far eluded scientists.

There have been a number of attempts to combine light and STM for many different purposes [4]. However, in such studies, it is typically very difficult to separate useful information in measured signals from artifacts induced by the light heating up the tip. For example, one very common trick that is used to increase measurement sensitivity — light modulation in combination with phase-sensitive detection — does not work particularly well with STM since varying light intensity not only modulates the optical field at the junction but also modulates the tip-sample separation, and therefore the tunneling current. We realized that this problem can be bypassed if the laser light illumination is confined to a part of the surface away from the junction. The challenge, however, is to show that valuable information can still be obtained in this configuration. If so, all of the complex problems caused by the presence of light in the junction, such as tip thermal expansion, rectification and thermoelectric current generation [5], could be avoided.

How can one still do optical spectroscopy with an STM without direct illumination of the junction? The main idea behind our new spectroscopy technique is to use an STM tip as an extremely sensitive detector that measures surface expansion in the direction perpendicular to the surface. As the light frequency is tuned to a particular molecular vibrational resonance, the molecules absorb more light. The absorbed energy dissipates into the substrate and leads to substrate expansion. This expansion results in an increased tunneling current since the tip-sample separation decreases. In the actual setup, a feedback control system is used to keep the tunneling current constant by moving the tip away from the surface when the surface expands, and we measure the distance by which the tip retracts. Accordingly, when the substrate contracts, the tip moves closer to the surface to maintain the tunneling current. Recorded traces of the tip motion plotted versus laser light frequency immediately yield molecular absorption spectra. Attempts to make similar measurements have been made previously [6] but the recent progress, including demonstration of high spectral sensitivity in our measurements, only recently became possible thanks to our advances in designing and building a state-of-the-art tunable infrared laser with a stable power output (shown in Figure 1) [7].

Figure 1: IRSTM setup. An ultra-high vacuum chamber with a home-made STM is on the left and our home-made infrared laser based on an optic parametric oscillator is on the right.

In order to demonstrate the performance of the technique, referred to as infrared scanning tunneling microscopy (IRSTM) [8], we prepared samples of two different isomers of tetramantane by depositing these molecules on a Au(111) surface. These molecules, with chemical formula C₂₂H₂₈, belong to a class of molecules called diamondoids (as their structures closely-resemble that of diamond) [9]. When deposited on the gold surface, both isomers, [121]tetramantane and [123]tetramantane, form very similar single-layered close-packed assemblies as shown in Figure 2 (a, c). To obtain infrared spectra of these self-assembled structures, a beam of light from our tunable infrared laser was focused about 1 mm away from the tip-sample junction to a spot size of 1.2 mm in diameter. The spectra were recorded by tracing the motion of the tip while sweeping the laser frequency as described above. The measured IRSTM spectra for [121]tetramantane and [123]tetramantane are shown in Figure 2 (b, d). Several peaks corresponding to different modes of tetramantane CH-stretch vibrations are seen in the spectra and the spectra of the two isomers are clearly different. To compare with the STM-IETS technique, in Figure 2(b) the blue line shows an STM-IETS spectrum that has been previously obtained on a [121]tetramantane molecule [10]. In that work, STM-IETS resolution was not nearly enough to resolve any specific CH-stretch modes that exist in this frequency range. In contrast, Figure 2 demonstrates the remarkable chemical sensitivity of our new IRSTM technique.

Figure 2: (a) STM image of [121]tetramantane molecules on Au(111). (b) IRSTM spectrum (black line) of [121]tetramantane on Au(111). The blue line (with a single broad peak) shows an STM-IETS spectrum of [121]tetramantane on Au(111) from Ref. [10]. (c) STM image of [123]tetramantane molecules on Au(111). (d) IRSTM spectrum of [123]tetramantane on Au(111). Vibrational peaks for [123]tetramantane are seen to differ significantly from vibrational peaks for [121]tetramantane.

Vibrational spectra of molecules on surfaces can be used to study molecule-molecule and molecule-substrate interactions. Using IRSTM we have been able to observe the notable influence of molecule-molecule interactions between tetramantane molecules on their vibrational spectra. To investigate molecule-molecule and molecule-substrate interactions in detail, IRSTM measurements were also performed on monolayers formed by the simplest diamondoid molecule, which is called adamantane. These measurements were supported by first-principle calculations and new interesting insights have been reported in another recent study [11].

While IRSTM allows atomic-scale imaging and the probing of vibrational modes of molecules in the same setup, the phenomenal spatial resolution inherent to conventional STM measurements cannot (yet) be attributed to the IRSTM vibrational spectra. The measured signal in IRSTM comes from excitations of all molecules that are illuminated by the laser beam, i.e. from roughly a trillion molecules in our measurements. Though extension of IRSTM to the single-molecule limit will not be trivial, and will require some bright ideas, there is a lot of room for improvement in the current technique. In addition, the indirect measurement of absorbed heat in a scanning probe microscopy setup could be applied to a number of other important technological applications.

References:
[1] Gerd Binnig and Heinrich Rohrer, “Scanning tunneling microscopy — from birth to adolescence”, Reviews of Modern Physics, 59, 615 (1987). Abstract.
[2] B. C. Stipe, M. A. Rezaei, W. Ho, “Single-Molecule Vibrational Spectroscopy and Microscopy”, Science, 280, 1732 (1998). Abstract.
[3] L. J. Lauhon and W. Ho, “Effects of temperature and other experimental variables on single molecule vibrational spectroscopy with the scanning tunneling microscope”, Review of Scientific Instruments, 72, 216 (2001). Abstract.
[4] Stefan Grafström, “Photoassisted scanning tunneling microscopy”, Journal of Applied Physics, 91, 1717 (2002). Abstract.
[5] M. Völcker, W. Krieger, T. Suzuki, and H. Walther, “Laser‐assisted scanning tunneling microscopy”, Journal of Vacuum Science & Technology, B 9, 541 (1991). Abstract.
[6] D. A. Smith and R. W. Owens, “Laser-assisted scanning tunnelling microscope detection of a molecular adsorbate”, Applied Physics Letters, 76, 3825 (2000). Abstract.
[7] Xiaoping Hong, Xinglai Shen, Mali Gong, Feng Wang, “Broadly tunable mode-hop-free mid-infrared light source with MgO:PPLN continuous-wave optical parametric oscillator”, Optics Letters, 37, 4982 (2012). Abstract.
[8] Ivan V. Pechenezhskiy, Xiaoping Hong, Giang D. Nguyen, Jeremy E. P. Dahl, Robert M. K. Carlson, Feng Wang, Michael F. Crommie, “Infrared Spectroscopy of Molecular Submonolayers on Surfaces by Infrared Scanning Tunneling Microscopy: Tetramantane on Au(111)”, Physical Review Letters, 111, 126101 (2013). Abstract.
[9] J. E. Dahl, S. G. Liu, and R. M. K. Carlson, “Isolation and Structure of Higher Diamondoids, Nanometer-Sized Diamond Molecules”, Science, 299, 96 (2003). Abstract.
[10] Yayu Wang, Emmanouil Kioupakis, Xinghua Lu, Daniel Wegner, Ryan Yamachika, Jeremy E. Dahl, Robert M. K. Carlson, Steven G. Louie, Michael F. Crommie, “Spatially resolved electronic and vibronic properties of single diamondoid molecules”, Nature Materials, 7, 38 (2008). Abstract.
[11] Yuki Sakai, Giang D. Nguyen, Rodrigo B. Capaz, Sinisa Coh, Ivan V. Pechenezhskiy, Xiaoping Hong, Feng Wang, Michael F. Crommie, Susumu Saito, Steven G. Louie, Marvin L. Cohen, “Intermolecular interactions and substrate effects for an adamantane monolayer on the Au(111) surface”, arXiv:1309.5090. 

Nanoscale Fourier-Transform Magnetic Resonance Imaging

(Clockwise from Top Left) John M. Nichol, Tyler R. Naibert, Raffi Budakian, Lincoln J. Lauhon

Authors:
John M. Nichol¹, Tyler R. Naibert¹, Eric R. Hemesath², Lincoln J. Lauhon², Raffi Budakian¹

Affiliation:
¹Dept of Physics, University of Illinois at Urbana-Champaign, USA
²Dept of Materials Science and Engineering, Northwestern University, USA

Magnetic resonance imaging (MRI) has had a profound impact on biology and medicine [1]. Key to its success has been the unique ability to combine imaging with nuclear magnetic resonance spectroscopy—a capability that has led to a host of powerful modalities for imaging. Common examples include spin-relaxation weighted imaging [2], chemical shift imaging [2], and functional MRI [3]. These and most other modern MRI techniques involve applying a combination of sophisticated radiofrequency and static magnetic field pluses to image the sample. These “pulsed” magnetic resonance methods [4] enable highly-efficient imaging by acquiring data from the entire sample at all times.

The spatial resolution of inductive MRI remains limited to millimeter lengths scales in common practice and to a few micrometers in the highest-resolution experimental instruments [5]. Although it remains a significant challenge, there is considerable interest to extend these powerful spectroscopic and imaging capabilities to the nanometer scale as the capability to perform nanoscale MRI would revolutionize biology and medicine. Promising work towards this goal includes force-detected magnetic resonance [6], which has been used to perform three-dimensional imaging of single tobacco mosaic virus particles with 5 nm resolution [7], and nitrogen-vacancy-based magnetic resonance [8, 9], which has been used to detect proton resonance in volumes as small as (5 nm)³ [10, 11]. The difficulties associated with the detection of nanometer-size volumes of nuclear spins, however, have required techniques such as these -- that are strikingly different from conventional inductive MRI. Moreover, it remains difficult to apply classic pulsed magnetic resonance techniques to nanometer-size samples.

In a recent proof-of-concept experimental work [12], we demonstrate a new technique, which allows us to perform pulsed nuclear magnetic resonance imaging and spectroscopy with nanometer-scale spatial resolution. Two unique components central to this work are (1) the ability to generate intense time-dependent magnetic fields on the nanometer scale, and (2) the development of a novel spin manipulation protocol, which allows us to encode the quantum spin noise in nanometer-scale ensembles of nuclear spins.

In particular, we perform nanometer-scale solid-state Fourier-transform MRI with roughly 10-nm spatial resolution. Fourier-transform imaging [13, 14], a pulsed magnetic resonance technique that relies on coherent manipulation of spins in the sample, is the most common method of MRI because it is highly efficient [15]. We use a nanometer-scale metal wire, or constriction, to generate intense static and radiofrequency magnetic field gradient pulses, which create temporal correlations in the statistical spin fluctuations in the sample. The correlations are recorded for a set of pulse configurations and Fourier-transformed to give the spin density. The sample used in this study is a nanometer-sized volume of polystyrene, a solid organic material containing a high proton density (Fig. 1).

Figure 1. Schematic of the experimental apparatus. A silicon nanowire coated with polystyrene is positioned near a constriction in a lithographically fabricated Ag wire. Electric current through the constriction generates static and radiofrequency magnetic field pulses, which are used to image protons in the polystyrene coating.

The magnetic resonance sensor we use is an ultra-sensitive silicon-nanowire mechanical oscillator [16], and the sample is mounted on the tip of the nanowire (Fig. 1). In addition to providing pulses for magnetic resonance, the metal constriction also produces a magnetic field gradient that oscillates at the silicon nanowire mechanical resonance frequency. The force of interaction between the protons in the sample and the magnetic field gradient from the constriction induces an angstrom-scale vibration of the nanowire, which we measure using a fiber-optic interferometer [17]. The silicon nanowire is an exceptionally sensitive mechanical oscillator and is well-suited to detecting the minute forces originating from small ensembles of nuclear spins [16].

Looking forward, our technique establishes a paradigm by which all other pulsed magnetic resonance techniques can be implemented for nanoscale imaging and spectroscopy. Constrictions capable of producing two orthogonal static gradients could be fabricated, and such devices would enable full three-dimensional imaging. A small coil near the sample could also provide a uniform radiofrequency field throughout the sample, which would enable the use of solid-state nuclear magnetic resonance spectroscopy techniques. More generally, our approach serves as a model for leveraging these and other sophisticated pulsed magnetic resonance tools to aid nanoscale MRI in its progress toward atomic-scale imaging.

References:
[1] R. R. Ernst, G. Bodenhausen, and A. Wokaun, "Principles of Nuclear Magnetic Resonance in One and Two Dimensions" (Oxford University Press, New York, 1997).
[2] D. D. Stark and W. G. Bradley, Magnetic Resonance Imaging (C.V. Mosby Co., St. Louis, 1988).
[3] Seiji Ogawa, Tso-Ming Lee, Asha S. Nayak, Paul Glynn, "Oxygenation-sensitive contrast in magnetic resonance image of rodent brain at high magnetic fields", Magnetic Resonance in Medicine, 14, 68 (1990). Abstract.
[4] R. R. Ernst and W. A. Anderson, "Application of Fourier Transform Spectroscopy to Magnetic Resonance", Review of Scientific Instruments, 37, 93 (1966). Abstract.
[5] L. Ciobanu, D. A. Seeber, and C. H. Pennington, "3D MR microscopy with resolution 3.7μm by 3.7μm by 3.7μm", Journal of Magnetic Resonance, 158, 178 (2002). Abstract.
[6] J. A. Sidles, J. L. Garbini, K. J. Bruland, D. Rugar, O. Zuger, S. Hoen, C. S. Yannoni, "Magnetic resonance force microscopy", Review of Modern Physics, 67, 249 (1995). Abstract.
[7] C. L. Degen, M. Poggio, H. J. Mamin, C. T. Rettner, and D. Rugar, "Nanoscale magnetic resonance imaging", Proceedings of the National Academy of Sciences of USA, 106, 1313 (2009). Article.
[8] C. L. Degen, "Scanning magnetic field microscope with a diamond single-spin sensor", Applied Physics Letters, 92, 243111 (2008). Abstract.
[9] J. M. Taylor, P. Cappellaro, L. Childress, L. Jiang, D. Budker, P. R. Hemmer, A. Yacoby, R. Walsworth, M. D. Lukin, "High-sensitivity diamond magnetometer with nanoscale resolution", Nature Physics, 4, 810 (2008). Abstract.
[10] H. J. Mamin, M. Kim, M. H. Sherwood, C. T. Rettner, K. Ohno, D. D. Awschalom, D. Rugar, "Nanoscale Nuclear Magnetic Resonance with a Nitrogen-Vacancy Spin Sensor", Science 339, 557 (2013). Abstract.
[11] T. Staudacher, F. Shi, S. Pezzagna, J. Meijer, J. Du, C. A. Meriles, F. Reinhard, and J. Wrachtrup, "Nuclear Magnetic Resonance Spectroscopy on a (5-Nanometer)³ Sample Volume", Science 339, 561 (2013). Abstract.
[12] John M. Nichol, Tyler R. Naibert, Eric R. Hemesath, Lincoln J. Lauhon, Raffi Budakian, "Nanoscale Fourier-Transform Magnetic Resonance Imaging", Physical Review X, 3, 031016 (2013). Abstract.
[13] Anil Kumar, Dieter Welti, Richard R Ernst, "NMR Fourier zeugmatography", Journal of Magnetic Resonance, 18, 69 (1975). Abstract.
[14] D. I. Hoult, "Rotating frame zeugmatography", Journal of Magnetic Resonance, 33, 183 (1979). Abstract.
[15] P. Brunner and R. R. Ernst, "Sensitivity and performance time in NMR imaging", Journal of Magnetic Resonance, 33, 83 (1979). Abstract.
[16] John M. Nichol, Eric R. Hemesath, Lincoln J. Lauhon, Raffi Budakian, "Nanomechanical detection of nuclear magnetic resonance using a silicon nanowire oscillator", Physical Review B, 85, 054414 (2012). Abstract.
[17] John M. Nichol, Eric R. Hemesath, Lincoln J. Lauhon, Raffi Budakian, "Displacement detection of silicon nanowires by polarization-enhanced fiber-optic interferometry", Applied Physics Letters, 93, 193110 (2008). Abstract.

A Current-Driven Single-Atom Memory

Top row: (left to right): Christian Schirm, Manuel Matt, Fabian Pauly. Bottom row (left to right): Elke Scheer, Juan Carlos Cuevas, Peter Nielaba.

Authors: Christian Schirm¹, Manuel Matt¹, Fabian Pauly¹, Juan Carlos Cuevas², Peter Nielaba¹, Elke Scheer¹

Affiliation:
¹Department of Physics, University of Konstanz, Germany.
²Departamento de Física Teórica de la Materia Condensada and Condensed Matter Physics Center (IFIMAC), Universidad Autónoma de Madrid, Spain.

Building functional devices at the atomic scale is a central vision of nanotechnology. Recently we demonstrated a two-state electrical switch with basically a single atom acting as moving part [1]. The switch consists of an aluminum atomic contact that is connected to electrodes at each side. Applying positive or negative currents above a certain threshold will change the state. Lower currents can be used to read out the switching status by simply measuring the resistance. This qualifies the device for future potential applications in high-density solid-state memories.

The single-atom switch was analyzed in close collaboration between experimental and theoretical physicists. Quantum mechanical transport channels of the electrical conductance served as an important link between experiment and theory. (The conductance is the inverse of the electrical resistance.) On the one hand these channels could be determined experimentally by measuring non-linear current-voltage characteristics in the superconducting state at very low temperatures [2]. On the other hand the channels could be calculated through a combination of molecular dynamics simulations, electronic structure theory, and transport theory expressed in terms of non-equilibrium Green's functions [4]. The theory enabled us to identify atomic configurations for both states of the experimentally realized switches. The comparison showed that in some cases the observed switching effect can be attributed to the geometrical rearrangement of a single atom.

Fig 1: Aluminum bridge with a 100 nm constriction, fabricated by electron beam lithography (see left panel). Due to the use of an elastic substrate, the aluminum structure can be stretched by carefully bending the sample. Since we measure the conductance continuously, we can stop the breaking procedure at a constriction narrowed to a single atom (see sketch in the right panel).

In the experiment we created an atomic contact of aluminum by the mechanically controllable break-junction technique using a lithographically fabricated bridge with a 100 nm wide constriction [3]. Fig. 1 shows a scanning electron micrograph (in false colors) of such an aluminum sample on an elastic substrate and illustrates the breaking process to reach an atomic-sized contact. Next, as shown in the graph in Fig. 2, we applied a slowly increasing electrical current and measured the conductance simultaneously. At some threshold current an electromigration process takes place, which can be observed as a jump in the conductance. The current is now decreased until the next jump is detected, then increased and so on. After some rearrangements the system may reversibly switch between two conductance states. A comparison of the conduction channel signatures verifies that these states remain exactly the same.

Fig 2: Applying a careful electromigration protocol when an atom-size contact has formed, a two-state atomic switch may develop. For this the current through the contact is increased slowly with time. When a jump in the conductance is detected, the current is reversed. After some time the system may switch reversibly between two conductance states. The inset shows the rectangular conductance-current hysteresis of a successfully generated two-level switch. The unit of conductance is G₀=2e²/h, also called the “quantum of conductance”.

In molecular dynamics simulations we performed a stretching procedure similar to the preparation of the contact in the experiment. The simulations generate realistic contact geometries and transport data as we verified by comparing experimental and theoretical conductance histograms [4,5]. Fig. 3 shows two atomic geometries as obtained in a stretching process. The calculated total conductance and individual conductance channels were in good agreement with several experimentally realized switches.

Fig 3: Molecular dynamics simulation of geometries corresponding to conductance values and channel signatures found in the experiment. The primary difference between both geometries is a relocation of the central atom and the breaking of two bonds to move from state “1” to state “0” (symbolized by the red scissors). The conductance and the individual channels are given in units of G₀.

The project demonstrates how theory and experiment can work hand in hand to advance nanotechnology. The atomic switch fulfills several technological requirements. For instance, it is relatively easy to fabricate due to the two-terminal configuration. Our work can be considered as a feasibility study of a one-bit atomic-size memory [6]. Potentially, an array of such switches can be implemented in a cross-bar architecture of solid-state memories when further technological issues, such as the stable operation at room temperature, can be solved.

References:
[1] C. Schirm, M. Matt, F. Pauly, J. C. Cuevas, P. Nielaba, E. Scheer, "A current-driven single-atom memory", Nature Nanotechnology 8, 645–648 (2013). Abstract. 
[2] E. Scheer, P. Joyez, D. Esteve, C. Urbina, M. H. Devoret, “Conduction channel transmissions of atomic-size aluminum contacts”, Physical Review Letters, 78, 3535–3538 (1997). Abstract. 
[3] J. M. van Ruitenbeek, A. Alvarez, I. Piñeyro, C. Grahmann, P. Joyez, M. H. Devoret, D. Esteve, C. Urbina, "Adjustable nanofabricated atomic size contacts", Review of Scientific Instruments, 67, 108 (1996). Abstract. 
[4] F. Pauly, M. Dreher, J. K. Viljas, M. Häfner, J. C. Cuevas, P. Nielaba, "Theoretical analysis of the conductance histograms and structural properties of Ag, Pt, and Ni nanocontacts", Physical Review B, 74, 235106 (2006). Abstract. 
[5] See supplementary information in reference [1].
[6] Sense Jan van der Molen, "Single-atom switches: Toggled with electrical current", Nature Nanotechnology, 8, 622 (2013). Abstract. 

Coherently Manipulating Nanomechanical Oscillators

[From Left to Right] Hajime Okamoto, Imran Mahboob, Hiroshi Yamaguchi

Authors: Hajime Okamoto, Imran Mahboob, and Hiroshi Yamaguchi

Affiliation: NTT Basic Research Laboratories, Atsugi-shi, Kanagawa, Japan

Semiconductor nanomechanical oscillators enable the pursuit of new physical phenomenon that can only be observed through the tiny mechanical displacement as well as enabling the development of nanoscience and nanotechnology. These systems with sharp mechanical resonances can be manipulated by electrical or optical means permitting a wide range of applications such as metrology and information processing [1].

Coupling two such nanomechanical oscillators has recently emerged as a subject of interest. This is because the sympathetic oscillation dynamics in the coupled system expand the potential applications of nanomechanical objects such as highly precise sensors, high-Q band-pass filters, signal amplifier, and even logic gates [2-5]. However, an obstacle to the further development of this architecture arises from the usually weak elastic (or electric) coupling between the nanomechanical components. This limits the ability to coherently transfer the vibration energy between the oscillators within the oscillator’s ring-down time. In other words, the coherent manipulation of the vibrational states, e.g., Rabi cycle and Ramsey fringe as in two-level quantum systems [6], is highly challenging in this coupled phonon system.

In our recent publication [7], we demonstrated coherent coupling of nanomechanical oscillators. This was realized in two geometrically interconnected GaAs doubly-clamped beams [Fig. 1(a)], where the frequency of beam R is higher than that of beam L as shown in Fig. 1(b). Because of the frequency mismatch and the weak mechanical coupling, coherent energy exchange between the two beams is not possible by the geometric interconnection alone.

Figure 1. (a) Schematic drawing of the sample and the piezoelectric effect. The piezoelectric effect enables harmonic driving, pumping, and detection of the mechanical motion via gate voltage. (b) Schematic drawing of the mechanical resonance for beam L and beam R. The frequency of beam R is higher than that of beam L (293.93 kHz) by 440 Hz. The coherent energy exchange between the two beams is achieved by applying the pumping voltage to beam L with the frequency difference between the two beams. (c) Schematic of the pumping protocol in a mass and spring model.

Coherent energy exchange can be achieved by dynamically coupling the mechanical oscillations of the two beams. This is realized by periodically modulating the spring constant of one beam at the frequency difference between the two beams [Fig. 1(c)]. This periodic modulation, namely pumping, can be induced by applying gate voltage via the piezoelectric effect in this sample [Fig. 1(a)]. This pumping enables strong vibrational coupling, leading to the cyclic (Rabi) oscillations between the two vibrational states (the beam-L state and the beam-R state) on the Bloch sphere (Fig. 2). The Rabi cycle period, i.e., the coupling strength, is fully adjustable by changing the pump amplitude via the gate voltage. As a result, the vibration energy can be quickly transferred from one beam to the other enabling the vibration of the original beam to be switched off on a time-scale orders of magnitude shorter than its ring-down time. This quick energy transfer to the adjacent oscillator opens up the prospect of high-speed repetitive operations for sensors and logics using nanomechanical systems [8,9].

Figure 2. Schematic drawing showing the evolution of the coupled mechanical oscillators on the Bloch sphere when undergoing a Rabi cycle.

In terms of phonon populations, the present system with sub-megahertz frequency is still in the classical regime with large mode occupation. However, the above technique could also be extended to gigahertz frequency mechanical oscillators in which the average phonon number in the mechanical modes falls below one at cryogenic temperatures [10]. This in turn leads to the exciting possibility of phononic quantum bits and entanglement between distinct macroscopic mechanical elements [11].

References:
[1] A. N. Cleland, “Foundations of Nanomechanics” (Springer-Verlag Berlin Heidelberg New York, 2003). [2] Matthew Spletzer, Arvind Raman, Alexander Q. Wu, Xianfan Xu, Ron Reifenberger, “Ultrasensitive mass sensing using mode localization in coupled microcantilevers”, Applied Physics Letters, 88, 254102 (2006). Abstract.
[3] F. D. Bannon, J. R. Clark, and C. T.-C. Nguyen, “High-Q HF microelectromechanical filters”, IEEE J. Solid-State Circuits 35, 512-526 (2000). Abstract.
[4] R. B. Karabalin, Ron Lifshitz, M. C. Cross, M. H. Matheny, S. C. Masmanidis, and M. L. Roukes, “Signal amplification by sensitive control of bifurcation topology”, Physical Review Letters, 106, 094102 (2011). Abstract.
[5] Sotiris C. Masmanidis, Rassul B. Karabalin, Iwijn De Vlaminck, Gustaaf Borghs, Mark R. Freeman, Michael L. Roukes, “Multifunctional nanomechanical systems via tunably coupled piezoelectric actuation”, Science, 317, 780-783 (2007). Abstract.
[6] J. R. Petta, A. C. Johnson, J. M. Taylor, E. A. Laird, A. Yacoby, M. D. Lukin, C. M. Marcus, M. P. Hanson, A. C. Gossard, “Coherent manipulation of coupled electron spins in semiconductor quantum dots”, Science, 309, 2180-2184 (2005). Abstract.
[7] Hajime Okamoto, Adrien Gourgout, Chia-Yuan Chang, Koji Onomitsu, Imran Mahboob, Edward Yi Chang, Hiroshi Yamaguchi, “Coherent phonon manipulation in coupled mechanical resonators”, Nature Physics, 9, 480-484 (2013). Abstract.
[8] Hiroshi Yamaguchi, Hajime Okamoto, and Imran Mahboob, “Coherent control of micro/nanomechanical oscillation using parametric mode mixing”, Applied Physics Express, 5, 014001 (2012). Abstract.
[9] I. Mahboob, E. Flurin, K. Nishiguchi, A. Fujiwara, H. Yamaguchi, “Interconnect-free parallel logic circuits in a single mechanical resonator”, Nature Communications, 2, 198 (2011). Abstract.
[10] A. D. O’Connell, M. Hofheinz, M. Ansmann, Radoslaw C. Bialczak, M. Lenander, Erik Lucero, M. Neeley, D. Sank, H. Wang, M. Weides, J. Wenner, John M. Martinis, A. N. Cleland, “Quantum ground state and single-phonon control of a mechanical resonator”, Nature, 464, 697-703 (2010). Abstract.
[11] Simon Rips and Michael J. Hartmann, “Quantum information processing with nanomechanical qubits”, Physical Review Letters, 110, 120503 (2013). Abstract.

Understanding Nanoparticle Interaction with Cell Membranes

(left) Sabina Tatur, 
(right) Marco Maccarini











Authors: Sabina Tatur¹, Marco Maccarini²

Affiliation: 
¹Department of Physics, University of Illinois at Chicago, USA 
²CEA Grenoble, France.

Although metallic nanoparticles – small entities with dimensions of just a few nanometers – have already been used for the fabrication of lustre decorations in medieval ceramics from Mesopotamia [1], it took another millennium until Michael Faraday provided a first scientific description of their optical properties in his 1847 lecture [2]. But only the emergence of modern experimental techniques - developed in the past thirty to forty years - triggered a boost in the interdisciplinary area of research, development and industrial activity of nanotechnology and nanoscience [3]. Today, metallic nanoparticles, made of noble metals, magnetic or semiconducting materials are added to many consumer goods to revolutionize the product with unique properties.

The skyrocketing progress in nanotechnology in recent years promises a great benefit for our civilization. Yet, there is, at the same time, an increasing concern about their potential harmful side effects on health and environment, and too little research has been conducted in this field so far. One might suppose that the principal determinants of hazardous materials are the surface properties and the surface area to which organisms are exposed. However, a general extrapolation from known hazardous or non-hazardous materials is impracticable since the physicochemical properties of many nanomaterials and nanoparticles deviate considerably from those of bulk or molecular materials.

The first contact that nanomaterials have with any living organism, is always the plasma membrane, a ~5 nm thick lipid bilayer surrounding each animal or plant cell [4]. Lipid membranes are therefore biologically relevant models to study nanomaterial-cell interactions. Understanding the interaction between nanoparticles and cell membranes has two-fold importance in terms of the evaluation of nanosafety as well as the effectiveness of nanoparticles in medical applications.

In this context, we recently studied the interaction of gold nanoparticles (AuNPs) with model cell membranes and focused our research on the effect of surface charge on AuNP destiny and membrane integrity. Previous research in this field showed that the degree of membrane disruption by AuNPs depend on their concentration, size, surface modification and charge, as well as on the phase state of the lipid bilayer (gel or liquid phase) [5-9]. These studies were, however, only symptomatic and missing the attempt to understand the link between surface characteristics of cell membrane and AuNPs to the final destination of the AuNPs.

To provide insight into the fate of AuNPs and their impact on cell membrane integrity, we analyzed the interaction of cationic ammonium AuNPs and anionic carboxy AuNPs with model membranes of only one lipid component. The single component membranes allowed us to focus on the interaction of the AuNPs with the lipid phases, avoiding complications arising from changes in structure and organization of the lipid membrane as a result of additional components in the membrane. To further bypass the electrostatic interactions between the solid substrate, onto which the lipid membrane was deposited, and the lipid bilayer itself, we prepared double lipid bilayers, in which a second lipid bilayer floats on top of the first one. This kind of membrane system gave us access to a highly hydrated, fluctuating bilayer with dynamic properties comparable to natural cell membranes [10] and was, therefore, our model of choice for the examination of the structural integrity of the model membrane with insight into the fate of the AuNPs by neutron reflectometry.

Figure 1a: The sketch shows structural details of the lipid double bilayer after addition of 1 mg/ml (left) and 0.1 mg/ml (right) of cationic AuNPs. The addition of the higher concentration causes a disruption of the membrane, whereas the addition of the lower one causes the AuNPs to incorporate into the hydrophobic moiety of the membrane.

We found that the AuNPs functionalized with cationic head groups penetrated into the center of the lipid bilayers and caused membrane disruption (see Figure 1a) [11]. In contrast, the AuNPs functionalized with anionic head groups did not permeate through the membrane but rather impeded the disruption of the lipid bilayer at alkaline pH (see Figure 1b).

Figure 1b: The sketch shows structural details of the lipid double bilayer after the addition of 0.01 mg/ml of anionic AuNPs. At alkaline pH, the NPs shield the membrane from disintegration that would usually happen under these conditions.

We are now working on understanding the mechanism of this interaction and hope that our approach can provide a strategy for a prospective nanoparticle risk assessement based on a surface charge evaluation and can contribute to nano-safety considerations during their design.

By understanding how nanoparticles affect our environment and taking precautions accordingly, not only will we revolutionize our products but we will be able to benefit from them yet another millennium.



References:
[1] Philippe Colomban, “The use of metal nanoparticles to produce yellow, red and iridescent colour, from bronze age to present times in lustre pottery and glass: solid state chemistry, spectroscopy and nanostructure”, Journal of Nano Research, 8, 109-132 (2009). Abstract.
[2] Michael Faraday, “The Bakerian lecture: experimental relations of gold (and other metals) to light”, Philosophical Transactions of the Royal Society of London, 147, 145-181 (1847). Full Text.
[3] Chris Toumey, “The man who understood the Feynman machine”, Nature Nanotechnology, 2, 9-10 (2007). Abstract.
[4] Jeremy M Berg, John L Tymoczko, and Lubert Stryer, “Biochemistry”, W.H.Freeman and Company, 5th edition (2002). [5] Ralph A. Sperling, Pilar Rivera Gil, Feng Zhang, Marco Zanella, Wolfgang J. Parak, “Biological applications of gold nanoparticles”, Chemical Society Reviews, 37, 1896-1908 (2008). Abstract.
[6] Alaaldin M. Alkilany, Pratik K. Nagaria, Cole R. Hexel, Timothy J. Shaw, Catherine J. Murphy, Michael D. Wyatt, “Cellular Uptake and Cytotoxicity of Gold Nanorods: Molecular Origin of Cytotoxicity and Surface Effects”, small, 5, 701-708 (2009). Abstract.
[7] Nikolai Khlebtsov, Lev Dykman, “Biodistribution and toxicity of engineered gold nanoparticles: a review of in vitro and in vivo studies”, Chemical Society Reviews, 40, 1647-1671 (2011). Abstract.
[8] Pakatip Ruenraroengsak, Pavel Novak, Deborah Berhanu, Andrew J. Thorley, Eugenia Valsami-Jones, Julia Gorelik, Yuri E. Korchev, Teresa D. Tetley, “Respiratory epithelial cytotoxicity and membrane damage (holes) caused by amine-modified nanoparticles”, Nanotoxicology, 6, 94-108 (2012). Abstract.
[9] Atsushi Hirano, Hiroki Yoshikawa, Shuhei Matsushita, Yoichi Yamada, and Kentaro Shiraki, “Adsorption and Disruption of Lipid Bilayers by Nanoscale Protein Aggregates”, Langmuir, 28, 3887-3895 (2012). Abstract.
[10] Giovanna Fragneto, Thierry Charitat, Jean Daillant, “Floating lipid bilayers: models for physics and biology”, European Biophysics Journal, 41, 863-874 (2012). Abstract.
[11] Sabina Tatur, Marco Maccarini, Robert Barker, Andrew Nelson, Giovanna Fragneto, “Effect of functionalized gold nanoparticles on floating lipid bilayers”, Langmuir, 29, 6606-6614 (2013). Abstract.

Mesoscopic Interference and Light Trapping in Semiconductor Nanowire Mats

[From left to right clockwise] Claire Blejean, Otto Muskens, Tom Strudley, Tilman Zehender, Erik Bakkers

Authors: Tom Strudley¹, Claire Blejean¹, Tilman Zehender², Erik P.A.M. Bakkers^2,3, Otto L. Muskens¹,

Affiliation: 
¹Faculty of Physical and Applied Sciences, University of Southampton, UK
²Dept of Applied Physics, Eindhoven University of Technology, Netherlands
³Kavli Institute of Nanoscience, Delft University of Technology, Netherlands.

For many years the field of mesoscopic physics, defined as the interface between the microscopic and macroscopic worlds, has been the domain of solid state electronics. Amongst the most prominent of such mesoscopic effects are those of universal conductance fluctuations [1] and Anderson localization [2]. The case of localization, in which transport is completely halted due to interference effects in the presence of sufficiently strong disorder, has been of considerable interest over the years. Indeed it was one of the two pieces of work for which Anderson was jointly awarded a Nobel prize in 1977, and its complexity has prompted many to quote Anderson’s Nobel lecture: “Very few believed it at the time…among those who failed to fully understand it at first was certainly its author.” [3]

Past 2Physics articles by Erik Bakkers:
May 20, 2012: "Signatures of Majorana Fermions in Hybrid Superconductor-Semiconductor Nanowire Devices" by Vincent Mourik, Kun Zuo, Sergey Frolov, Sébastien Plissard, Erik Bakkers, Leo Kouwenhoven.

Due to the strong analogy between electron and wave transport, such phenomena should also occur for wave transport in random media. However this presents a significant challenge, as mesoscopic effects occur when the probability of a diffusion scattering back to the same point and interfering with itself becomes significant. Intuitively this break down of the traditional ‘random walk’ description requires scattering mean free paths significantly lower than the wavelength. As a consequence, demonstrations of Anderson localization of light in three dimensions have proved difficult, with most observations being for other systems such as microwaves in a waveguide [4], acoustic waves [5] and even cold atoms [6]. A couple of pioneering studies have reported localization of light in three dimensional media [7,8], however the interpretation of their results is complicated by absorption and fluorescence, which can mimic certain localization effects.

Figure 1: (left) The process of Vapour-liquid-solid growth starting from a gold nanoparticle catalyst is used to grow nanowires on a substrate. A lateral growth step increases the diameter of the wires, without affecting their length. (right) A typical high-density nanowire mat used in our experiments.

In our recent work published in Nature Photonics [9], we presented experimental results obtained using densely packed disordered mats of semiconductor nanowires. We demonstred strong mesoscopic effects using visible light in these layers. The nanowire mats were fabricated at the University of Eindhoven using the method of metallo-organic vapour phase epitaxy (MOVPE), see Fig. 1. By using a recipy of alternating cycles of vapour-liquid-solid (VLS) nanowire growth, followed by lateral growth to increase the wire diameter, precise control could be achieved over the length and diameters of the wire. A very high nanowire density of up to 50% area percentage was necessary to achieve the strong scattering strength required for observing the mesoscopic effects. These nanowire mats are remarkable as they have optical mean free paths as low as 0.2 micrometres, which along with their low intrinsic absorption make them ideal candidates for exploring Anderson localization.

Figure 2 [click on image to see higher resolution]: (left) typical intensity images of light transmitted through a nanowire mat, for a tight laser focus ('in focus') and for a laser illumination out of focus. Total image size ~20 μm. (middle) Intensity distribution normalized to ensemble average, showing the speckle fluctuations in space at the exit plane of the nanowire mat. (right) the total intensity fluctuations are enhanced for a tigthly focused laser, see Ref. [9].

By examining the intensity statistics of the transmitted light we found that it exhibited the large intensity fluctuations and long range correlations (both spatial and spectral) typical of mesoscopic interference (Fig. 2). Fitting these fluctuations with predictions from theory, we found an average of only 4 independent transmission channels through our sample, which is several orders of magnitude lower than previously reported values. These measurements unambiguously show for the first time that strong mesoscopic interference corrections can be achieved in three-dimensional nanomaterials and at optical wavelengths.

Mesoscopic transport corrections are a precursor of the strong or Anderson localization transition, where transport of light is halted by self-interference of many light paths returning to the same position in the medium. We are now in the exciting position where we can probe in detail the mesoscopic physics of light, as well as tuning our nanowire growth parameters further in a bid to observe localization itself. For practical applications, semiconductor nanowires have great potential in solar cell and light generation. A deep understanding of light transport in such materials is of great importance for further optimizing these applications. Ultimately, it may be possible for mesoscopic effects to be harnessed and turned into a new design tool for maximizing performance of real-world devices.

References:
[1] P.A. Lee and A. Douglas Stone, “Universal Conductance Fluctuations in Metals”, Physical Review Letters, 55, 1622-1625 (1985). Abstract.
[2] P.W. Anderson, “Absence of diffusion in certain random lattices”, Physical Review, 109, 1492-1505 (1958). Abstract.
[3] Philip W. Anderson, Nobel lecture (1977).
[4] A.A. Chabanov, M. Stoytchev, A.Z. Genack, “Statistical Signatures of Photon Localization”, Nature 404, 850-853 (2000). Abstract.
[5] Hefei Hu, A. Strybulevych, J. H. Page, S. E. Skipetrov, B. A. van Tiggelen, “Localization of Ultrasound in a Three-Dimensional Elastic Network”, Nature Physics, 4, 945-948 (2008). Abstract.
[6] Juliette Billy, Vincent Josse, Zhanchun Zuo, Alain Bernard, Ben Hambrecht, Pierre Lugan, David Clément, Laurent Sanchez-Palencia, Philippe Bouyer, Alain Aspect, “Direct Observation of Anderson localization of matter waves in a controlled disorder”, Nature, 453, 891-894 (2008). Abstract. 2Physics Article.
[7] Diederik S. Wiersma, Paolo Bartolini, Ad Lagendijk, Roberto Righini, “Localization of light in a disordered medium”, Nature 390, 671-673 (1997). Abstract.
[8] Martin Störzer, Peter Gross, Christof M. Aegerter, and Georg Maret, “Observation of the Critical Regime Near Anderson Localization of Light”, Physical Review Letters, 96, 063904 (2006). Abstract.
[9] Tom Strudley, Tilman Zehender, Claire Blejean, Erik P. A. M. Bakkers, Otto L. Muskens, “Mesoscopic Light Transport by Very Strong Collective Multiple Scattering in Nanowire Mats”, Nature Photonics, 7, 413-418 (2013). Abstract.

Piezotronic Nanowire “Taxel” Gives Active/Adaptive Sense of Touch

[From Left to Right] Xiaonan Wen, Zhong Lin Wang, Wenzhuo Wu


Authors: Wenzhuo Wu, Xiaonan Wen and Zhong Lin Wang

Affiliation: 
School of Materials Science and Engineering, Georgia Institute of Technology, Atlanta, USA.


Emulation of human senses via electronic means has long been a grand challenge in research of artificial intelligence as well as prosthetics, and is of pivotal importance for developing intelligently accessible and natural interfaces between human/environment and machine. Unlike other senses (seeing, hearing, smelling and tasting), capability of skin for touch sensing remains stubbornly difficult to be mimicked, which necessitates the development of large-scale pressure sensor arrays with high spatial-resolution, high-sensitivity and fast response.

Implementations of pressure sensor arrays have been previously reported using assembled nanomaterials or microstructured rubber layers based on the change in capacitance or resistance upon pressure, which have been applied to map strain distribution in a matrix format with resolution in the millimeter-scale [1,2]. In those demonstrations, electronic components like traditional field-effect-transistors (FETs) act as read-out elements for detecting pressure-induced property change in the pressure-sensitive media. Intensive efforts have been devoted to minimize the effect of substrate strain on performance of these electronic components while preserving the deformability of the substrate. This scheme of pressure sensing, whereas, not only requires complicated system integration of heterogeneous components but also lacks proficiency in directly interfacing electronics with mechanical actions in an active way that mechanical straining can be utilized to generate new electronic control/feedback. Therefore, new approaches have to be developed for better mimicking or even competing with the tactile sensing capabilities of human skins.

A direct control over the operation of electronic devices by mechanical action could be advantageous for implementing tactile sensing and realizing direct interfacing between machines and human/ambient. Piezoelectric materials are ideal candidates for this purpose, which produce electrical potential upon variations of applied pressure/stress due to the linear electromechanical coupling/interaction between mechanical and electrical state in materials lacking inversion symmetry. The most well-known piezoelectric material is perovskite-structured Pb(Zr_xTi_1-x)O₃ (PZT), which has been widely used for electromechanical sensing, actuating and energy harvesting. PZT, however, is electrically-insulating and hence less useful for building electronic devices. In addition, the extremely brittle nature of ceramic PZT films and incorporation of lead impose issues such as reliability, durability and safety for long term sustainable operations and hinder its applications in areas such as biomedical devices.

Figure 1: [click on image to view higher resolution] Piezopotential in Wurtzite crystal. (a) Atomic model of the Wurtzite-structured ZnO. (b) Distribution of piezopotential along a ZnO NW under axial strain calculated by numerical methods. The growth direction of the NW is along c-axis. The color gradient represents the distribution of piezopotential in which red indicates positive piezopotential and blue indicates negative piezopotential.

Due to the polarization of ions in a crystal that has non-center symmetry, a piezoelectric potential (piezopotential) is created in the crystal by applying a stress [3] (Figure 1). For piezoelectric semiconductor materials such as ZnO, GaN, InN and CdS, the effect of piezopotential on the transport behavior of charge carriers is significant due to their multiple functionalities of piezoelectricity, semiconductor and photon excitation. One of the important effects of piezopotential on p-n junction and metal-semiconductor contact is to tune/control the charge transport across the interface, which is the basis of piezotronic effect (Figure 2) [3,4]. Electronics fabricated by using inner-crystal piezopotential as a “gate” voltage to tune/control the charge transport behavior is called piezotronic device. Our group initiated the research in piezotronics back to 2007 [5]. Since then we have investigated the piezotronic effect for realizing novel applications such as strain-gated piezotronic logic nanodevices [6] and piezotronic strain memory device [7]. In contrast to conventional CMOS units, the strain-gated electronics is driven by mechanical agitation. A brief comparison between piezotronic transistor and field effect transistor (FET) is shown in Figure 3.

Figure 2: [click on image to view higher resolution] Schematic of energy diagram illustrating the effect of piezopotential on modulating the metal-semiconductor contact and p-n junctions. (a) With compressive strain applied, the negative piezoelectric polarization ionic charges induced near the interface (symbols with “-”) increases the local Schottky barrier height (SBH). (b) With tensile strain applied, the positive piezoelectric polarization ionic charges induced near the interface (symbols with “+”) decreases the local SBH. (c) and (d) With strain applied, the piezoelectric polarization ionic charges are induced near the junction interface.

Figure 3: [click on image to view higher resolution] Comparison between traditional FET and Strain-gated piezotronic transistor (SGT).

Designing, fabricating and integrating arrays of nanodevices into a functional system is the key for transferring nano-scale science into applicable nanotechnology. In our latest work [8], we report the first and by far the largest 3D array integration of vertical NW piezotronic transistors circuitry (92 x 92 taxels with 234 taxels per inch) as active/adaptive taxel-addressable pressure-sensor matrix for tactile imaging. Here “taxel” is short for tactile pixel, which is the unit in our array device. We combined the top-down microfabrication processes for fabricating device structure with the bottom-up synthesis of ZnO NWs at low-temperature (85^oC).

The taxel area density of SGVPT (strain-gated vertical piezotronic transistor) array is 8464/cm², which is higher than the number of tactile sensors in recent reports (~ 6-27/cm²) [1,2,9] and mechanoreceptors embedded in the human fingertip skins (~ 240/cm²) [10]. The fabricated sensors are capable of mapping spatial profiles of small pressure changes (< 10 kPa). The spatial resolution with taxel dimension (20 x 20 μm²) and pitch size (100 μm) as well as the tactile sensitivity (2.1 μs∙kPa^-1) have been demonstrated, enabling a 15-to-25-fold increase in number of taxels and 300-to-1000-fold increase in taxel area density compared to recent reports.

Each taxel in the array device is made of a strain-gated vertical piezotronic transistor (SGVPT) based on vertically-aligned n-ZnO nanowires (NWs) (Figure 4a, d). SGVPT operates based on modulation of local contact characteristics and charge carrier transport by strain-induced ionic polarization charges at the interface of metal-semiconductor contact, which is the fundamental of piezotronics. Using the piezoelectric polarization charges created at metal-semiconductor interface under strain to gate/modulate transport process of local charge carriers, piezotronic effect has been applied to design independently addressable two-terminal transistor arrays, which convert mechanical stimuli applied on the devices into local electronic controlling signals (Figure 4b-c). Figure 4c shows the equivalent circuit to present the operation scheme of SGVPT array.

Figure 4: [click on image to view higher resolution] (a) Comparison between three-terminal voltage-gated NW FET (left) and two-terminal strain-gated vertical piezotronic transistor (right). (b) Schematic illustration of a 3D SGVPT array with taxel density of 92 × 92 and scheme for spatial profile imaging of local stress. (c) Equivalent circuit diagram of the 3D SGVPT array. (d) SEM image of SGVPT array taken after etching-back the SU 8 layer and exposing top portions (~ 20 μm) of the ZnO NWs. Inset, 30^o-tilt view of the exposed ZnO NWs for single taxel. (e) Optical image of the transparent 3D SGVPT array on flexible substrate. (f) Current responses for single taxel under different pressures, presenting the gate modulation effect of applied pressure on the electrical characteristics of SGVPT.

The elimination of wrap gate offers a new approach for 3D structuring. The local contact profile and carrier transport across the Schottky barriers, formed between ZnO NW and metal electrodes, can be effectively controlled by the polarization-charge-induced potential. Electrical characteristics of the two-terminal SGVPT are therefore directly modulated by external mechanical actions induced strain, which essentially functions as a gate signal for controlling carrier transport in SGVPT. The modulation effect of applied pressure is shown from the plot of current variations against pressure changes (Figure 4f). Cross-bar electrodes have been configured for multiplexed data acquisition and the spatial profiles of applied stress can be recorded and imaged. The output signal is current response so that it is easy to integrate SGVPT array with back-end interface circuits for fast data processing, recording and transmission.

Three unique capabilities of SGVPT array device, which are not readily available in previous reports, have been demonstrated in this work:
1. The capability of using SGVPT array for multi-dimensional signature recording, which not only records the calligraphy or signature patterns, when people write, but also registers the corresponding pressure/force applied at each location/pixel by the person and the speed or writing. This augmented capability can essentially provide means for realizing personal signature recognition with unique identity and enhanced security.
2. The shape-adaptive sensing capability of SGVPT array, which provides means for detecting the shape change of device in situ in real-time and feeding-back the sensed changes in shape for calibration of other functionalities as well as corresponding control/response performed by the system. The real time detection of shape changes caused by stretching or twisting is a desirable feature for sensors embedded in an artificial tissue or prosthetic device. This unique capability enables the potential integration of SGVPT device for applications in artificial/prosthetic skin in smart biomedical treatments and intelligent robotics.
3. Additionally, the SGVPT devices can also function as self-powered active tactile sensors by converting mechanical stimulations into electrical signals utilizing the piezopotential without applied bias, which emulates the physiological operations of mechanoreceptors in biological entities, such as human hair follicles and hair cells in the cochlea.

Table 1: Comparison between piezoresistive effect and piezotronic effect.

Our approach is based on a completely different mechanism from the traditional designs based on piezoresistive effect. A comparison between piezotronic effect and commonly recognized piezoresistive effect is listed in Table 1. This technology eliminates the wrap-gate electrode for fabricating 3D vertical NW based transistor, taking advantage of piezotronic effect. The structural transformation from three-terminal configuration into two-terminal configuration significantly simplifies the layout design and circuitry fabrication while maintains effective control over individual devices. The approach presented has the potential to be integrated with silicon-based CMOS technology for achieving augmented functionalities in smart systems in the era of “More Than Moore”, such as artificial skin, personal electronics and potential integration with compliant energy harvesting modules for self-powered flexible functional systems [11]. The scalability of this platform in integrating in-place synthesized single-crystalline NWs in controllable manners together with its demonstrated compatibility with state-of-the-art microfabrication techniques enables future large-scale implementation of wurtzite NWs for practical applications in human-electronics interfacing, smart skin and micro/nano-electromechanical systems.

References:
[1] Kuniharu Takei, Toshitake Takahashi, Johnny C. Ho, Hyunhyub Ko, Andrew G. Gillies, Paul W. Leu, Ronald S. Fearing & Ali Javey. "Nanowire active-matrix circuitry for low-voltage macroscale artificial skin". Nature Materials, 9: 821-826 (2010). Abstract.
[2] Stefan C. B. Mannsfeld, Benjamin C-K. Tee, Randall M. Stoltenberg, Christopher V. H-H. Chen, Soumendra Barman, Beinn V. O. Muir, Anatoliy N. Sokolov, Colin Reese & Zhenan Bao. "Highly sensitive flexible pressure sensors with microstructured rubber dielectric layers". Nature Materials, 9: 859-864 (2010). Abstract.
[3] Zhong Lin Wang. "Progress in piezotronics and piezo-phototronics". Advanced Materials, 24: 4632-4646 (2012). Abstract.
[4] Zhong Lin Wang. "Piezotronics and piezo-phototronics" (Springer, 2013).
[5] Zhong Lin Wang. "Nanopiezotronics". Advanced Materials, 19: 889-892 (2007). Abstract.
[6] Wenzhuo Wu, Yaguang Wei, Zhong Lin Wang. "Strain-gated piezotronic logic nanodevices". Advanced Materials, 22: 4711-4715 (2010). Abstract.
[7] Wenzhuo Wu and Zhong Lin Wang. "Piezotronic nanowire-based resistive switches as programmable electromechanical memories". Nano Letters, 11: 2779-2785 (2011). Abstract.
[8] Wenzhuo Wu, Xiaonan Wen, Zhong Lin Wang. "Taxel-addressable matrix of vertical-nanowire piezotronic transistors for active/adaptive tactile imaging". Science, 340: 952-957 (2013). Abstract.
[9] Tsuyoshi Sekitani, Tomoyuki Yokota, Ute Zschieschang, Hagen Klauk, Siegfried Bauer, Ken Takeuchi, Makoto Takamiya, Takayasu Sakurai, Takao Someya. "Organic nonvolatile memory transistors for flexible sensor arrays". Science, 326: 1516-1519 (2009). Abstract.
[10] R.S. Johansson, A.B. Vallbo. "Tactile sensibility in the human hand - relative and absolute densities of 4 types of mechanoreceptive units in glabrous skin". Journal of Physiology, 286: 283-300 (1979). Article.
[11] Zhong Lin Wang, Wenzhuo Wu. "Nanotechnology-enabled energy harvesting for self-powered micro-/nanosystems". Angewandte Chemie International Edition, 51: 11700-11721 (2012). Abstract.