User login


You are here

Journal Club January 2010: MEMS-based Instrumentation for Experimental Nanomechanics

Yong Zhu's picture

Welcome to the January 2010 issue! In the issue of May 2007 , Prof. Xiaodong Li outlined the existing experimental methods for mechanical characterization of 1D nanostructures. In this issue, I will discuss along the same line but focus on experimental methods enabled by microelectromechanical systems (MEMS).

With the emergence of a broad range of 1D nanostructures (such as nanowires, nanotubes, nanobelts and etc), a considerable amount of work on their mechanical properties including elasticity, plasticity and fracture has been carried out in recent years. However, one critical issue is that the experimental data exhibit large scatter and at times, discrepancy in terms of size effects, and there exists a substantial gap between experiments and simulations. Harold Park and co-authors (MRS Bulletin 34, 178, 2009) gave a nice review on the state-of-the-art from both experimental and modeling perspectives. For elasticity, the accurate measurement of nanowire dimensions, crystalline orientation and in some cases, surface oxide thickness is very important. For plasticity, interesting issues include interaction between dislocations with grain/twin boundaries, competition between brittle and ductile fracture, phase transformation and etc. Other issues can be seen from the discussions led by Julia Greer and Wei Cai. To answer all these questions, real-time imaging in addition to quantitative stress-strain measurement becomes necessitated.

In-situ electron microscopy testing offers unprecedented opportunities to explore these issues. A number of promising MEMS-based testing stages have been developed recently for in-situ testing of nanostructures. Various designs and methods for actuation, load sensing and positioning of nanostructures on the testing stages are briefly reviewed and are open for discussion.

One method is to use an external piezo-actuator. This method takes advantage of existing piezo-actuators that can impose sub-nanometer motion resolution. Typical issues associated with piezo-actuators include hysteresis and nonlinearity in addition to alignment issues.

The other method is to employ on-chip MEMS actuators. Thermal actuators and electrostatic (comb drive) actuators are two prominent candidates due to their compatibility with common microfabrication methods. Thermal actuators are compact and stable, easily providing milli-Newton actuation force; they offer quasi-displacement control. One possible drawback is the unintentional heating of the specimen; heat sink structures have been developed to largely mitigate this problem. Comb drive actuators generate constant force for a given actuation voltage, typically with forces ranging from several micro-Newton up to one milli-Newton. They can achieve larger travel range up to tens of micrometers. The entire structure is relatively compliant, which may render it vulnerable to pull-in instability. Another potential drawback is the levitation effect that could cause out-of-plane movement of the device.

Load Sensing
Typical load sensors include flexible members that deflect under the applied load. Several mechanisms to detect such deflection in MEMS include direct observation using electron beam, piezoresistivity measurement and capacitance measurement. Observation of the load sensor using electron beam prevents the continuous, real-time observation of the specimen. Piezoresistive sensing can electronically measure the load with nano-Newton resolution, but suffer from large thermal drift. Capacitance measurement can measure the displacement with 1 nanometer resolution and the load with nano-Newton resolution; it requires a signal conditioning circuitry to eliminate the parasitic capacitance and electromagnetic noise.

Manipulation and Positioning of Specimens
Typical methods include nanomanipulation, dielectrophoresis, and direct growth. Nanomanipulators can be used inside a SEM to pick, manipulate and position these structures to the desired testing platform. Electron beam (or focused ion beam) induced deposition of carbon or platinum can be used to weld and fix the specimens. Dielectrophoresis have been used to manipulate nanostructures, but alignment is typically a problem. Direct growth would be the ideal method (though very different as of today) to prepare specimens for nanomechanical characterization. It does not involve the nano-welding step that might cause some spurious effects on the properties being measured.

All the reported MEMS-based instruments can be roughly grouped into three categories. The first one possesses only an actuator (either thermal or comb drive actuator), in which case the actuator also works as the load sensor (Naraghi and Chasiotis, JMEMS 18, 1032, 2009 ). The second type includes an external actuator and a microfabricated testing frame (Lu et al., Experimental Mechanics, 2009 ). The third one is an integrated system with on-chip actuator and load sensor (Zhang et al., J. Micromechanics and Microengineering 19, 075003, 2009 ). MEMS-based instrumentation for experimental nanomechanics has been growing rapidly in recent few years. Here I list three very recent representative articles and some earlier works can been found therein and also in the related disussion of May 2007 issue . This technique has seen promising success in mechanical testing of a variety of 1D nanostructures including carbon nanotubes, semiconductor nanowires, metallic nanowires and polymer nanofibers. The technique itself continues evolving to achieve better accuracy and more functionality with the integration of new structural design, fabrication process, signal conditioning and etc.

The same MEMS-based sensing and actuation principles can be employed to study the mechanical response of thin films and biological cells. See the two reviews on these topics for more details (Haque and Saif, Experimental Mechanics 43, 248, 2003 ; Loh et al., Experimental Mechanics 49, 105, 2009 ).


Harold S. Park's picture

Hi Yong:

Thanks for raising these interesting issues with this J-Club.  I am very interested by your comments with regards to "direct growth" to prepare specimens for nanomechanical characterization.  By that, do you mean to grow/synthesize a nanostructure directly in, say, the MEMS-based instrument in order to test it immediately?  If so, considering that growth and synthesis of nanostructures is an increasingly well-understood field these days, as is nanomechanical testing via various MEMS platforms, can you explain why direct growth is so difficult today?



Yong Zhu's picture

Hi Harold,

Thanks for raising the question – it’s an important issue related to nanomechanical testing. You are absolutely right that nanostructure synthesis is relatively well understood and controlled nowadays. A large number of synthesis methods have been developed in the last decade such as chemical vapor deposition (CVD), arc discharge and laser ablation for carbon nanotubes, CVD for semiconductor nanowires, electrodeposition and solution-phase method for metallic nanowires, and electrospinning for polymer nanofibers. But each synthesis process requires specific setup, environment, temperature range and etc., which are not easily met with the MEMS devices. For instance, the arc discharge method (producing carbon nanotubes of the highest quality known) involves a graphite cathode and extremely high temperature. Furthermore, the nanostructures must be positioned and well aligned between two terminals in the MEMS device, which poses more challenges. As an example, even the well-developed electrospinning technique is very difficult to deposit a segment of polymer nanofiber aligned on the MEMS device.

Typically MEMS devices are made of either polysilicon or single-crystalline silicon (SCS) that limit the synthesis temperature to ~800 K (recrystallization temperature of polysilicon). CVD is a possible method to grow carbon nanotubes on the both polysilcion and SCS devices. For SCS device, silicon nanowires have been synthesized between two <111> sidewalls in the epitaxial manner. I list two papers below on CVD growth of carbon nanotubes and epitaxial growth of silicon nanowires, respectively. With the advancement of nanostructure synthesis, hopefully other types of semiconductor nanowires and metallic nanowires can be directly synthesized on the MEMS instruments soon.

Kong J, Soh HT, Cassell AM, Quate CF, Dai HJ (1998) Synthesis of individual single-walled carbon nanotubes on patterned silicon wafers. Nature 395:878.
He RR, Gao D, Fan R, Hochbaum AI, Carraro C, Maboudian R, Yang PD (2005) Si nanowire bridges in microtrenches: Integration of growth into device fabrication. Adv Mater 17:2098.

Hope the above comments are helpful.



Xiaodong Li's picture


Thank you very much for leading this theme which is timely and of great interest to the community. The MEMS-based instruments and studies have opened up fresh opportunities to probe the mechanical behavior of nanostrucures and unveil the physical mechanisms that govern the size-effects. The papers you listed are excellent, which I will use for my graduate course lectures (I teach nanomaterials this semster). I very much like the challenges you proposed. For instance, can one integrate the MEMS-based loading frame with an AFM to perform 3D loading with both the MEMS loading frame and AFM tip? Can one use the AFM to calibrate the MEMS laoding frame? It seems there are still a lot to explore down the road. 

Bionanomaterials are difficult to handle. It may be great to have a nanomanipulator integrated with the MEMS to study the mechanical properties of bionanomaterials.

Again, thank you and hope to see more discussions. 


Yong Zhu's picture

Hi Xiaodong,

Thank you for raising an excellent point. So far most MEMS-based instruments are designed for tensile testing of nanostructures. I think they can also be used for compression, bending and possibly torsion tests. In a more general sense, almost all the nanomechanical testing methods (except nanoindentation, I guess) can only deals with a simple loading. Combined loading with MEMS and AFM tips can provide more sophisticated stress state, which can be important for understanding certain mechanical behaviors, and also of practical importance. I don’t know if the combined loading is often encountered in simulations.

Mechanical testing of biomaterials with MEMS is another interesting direction. Most MEMS-based instruments are integrated inside either SEM or TEM for in-situ testing. SEM and TEM have stringent requirement on vacuum, which is often incompatible with biomaterials. AFM, in this case, can provide a viable means for in-situ imaging. I have almost no experience with biomaterials but tend to agree that bionanomaterials are difficult to handle. What do you think are the major challenges in handling the biomaterials?


Xiaodong Li's picture

Thanks Yong. I agree that vacuum and electron beam may alter biomaterials in SEM and TEM. AFM does not have this problem. To perform AFM imaging remains challenges. Not like a solid rigid sample, a soft biomaterial sample may move according to AFM sacnning. Tapping mode is essential in some cases. The following paper is about the nanomechanical testing using a nano bionix tensile teter and AFM imaging on collagen fiber samples. 

Xinnan Wang, Yongda Yan, Michael J. Yost, Shen Dong, and Xiaodong Li,
"Nanomechanical Characterization of Micro/nanofiber Reinforced Type I
Collagens," Journal of Biomedical Materials Research Part A, 83 (2007)

Yong Zhu's picture

Speaking of biological structures, I recall a very interesting paper by Prof. Ballarini and co-workers on collagen fibrils. They used a MEMS device to directly measure the tensile strength, stiffness and fatigue behaviour of nanoscale fibres.

Eppell SJ, Smith BN, Kahn H, Ballarini R, Nano measurements with micro-devices: mechanical properties of hydrated collagen fibrils, JOURNAL OF THE ROYAL SOCIETY INTERFACE 3, 117-121 (2006).

Subscribe to Comments for "Journal Club January 2010: MEMS-based Instrumentation for Experimental Nanomechanics"

Recent comments

More comments


Subscribe to Syndicate