|
|
|
|
 |
|
Introduction
Over the past 20 years Scanning Probe Microscopes (SPM) have emerged as an essential material characterization technique in various
fields1,2,3,4,5. The importance of the SPM was evident as early as 1984 when the Nobel prize was awarded for the Scanning Tunneling
Microscope (STM) invention by IBM researchers1. Today the Atomic Force Microscopes (AFM) are the most commonly used scanning probe
technique for materials characterization2. Major advantages of AFM are that it has a combination of high resolution in three dimensions,
the sample does not have to be conductive, and there is no requirement for operation within a vacuum.
With an AFM, a large range of topographies and many types of materials can be imaged. Examples of surface features that may be
imaged include: atomic terraces, carbon nanotubes, colloidal particles, viruses, DVD textures up to micro lens textures, fractured
surfaces, and complex multi-phase polymers. In other words, AFM is capable of delivering unique 3D topography information from the
angstrom level to the micron scale with unprecedented resolution.
With an AFM, the Z-axis resolution (i.e., perpendicular to the
surface) is typically better than the resolution in the XY scan
plane of the sample surface. Under ambient conditions, the Zresolution
for most of the commercially available AFMs is on the
sub-angstrom level. Resolution in the X-Y scan place is oftern
limited by the diameter of the probe and is on the order of a
few nanometers. In the X and Y ases, AFM images are always a
convolution of the probe geometry and sample texture. However, if the probe is much smaller than the surface features, the image distortions introduced by the probe are minimal.
The extreme sensitivity of the AFM is derived from a force sensor that measures forces between the probe and target surface which are
typically less than 1 nN/nm. Most AFM’s utilize a light lever or a force sensor, as first disclosed in 1929 and then applied to the AFM in 1986. Figure 1 shows a schematic of the AFM. Recently, a new type of force sensor, based on a crystal resonator, shows promise for making the AFM much simpler to operate6,7 and it provides a very high force sensitivity required for high resolution imaging.
Table 1: Given the wide variety of applications that use particles, it makes sense that there are many different ways to analyze and characterize particles.
The following is a partial list of the material classes of particles in different environmental media, as analyized by an AFM:
1.0 Imaging in Air
1.1 Dry Powders
1.2 Evaporated Suspensions
1.3 Bio-Particles
1.4 Carbon nanotubes
1.5 TEM Samples
1.6 SEM Samples
2.0 Imaging in liquids
2.1 Bio-Particles in Buffer
2.2 Inorganic Particles 3.2 Hard Surface Materials
3.0 Imaging embedded particles
3.1 Soft polymer and Bio-materials
3.3 Membranes and defects
|
For particle characterization, there is no single instrument that
is the “right tool for every job”. In fact, more than 400 different
techniques exist for particle counting, sizing, analyzing, and
characterizing. Typically, instrumentation is chosen by engineers
and researchers through consideration of what measurements
need to be made, and in what environment the measurements
need to be made. There are two primary considerations for
selecting instrumentation: a) single particle versus ensemble,
and b) the environment the measurement is made in – air,
liquid, vacuum. Table 1 shows a comparison of the most
common material classifications and environmental media for
particle analysis. In general, morphological information, such
as shape and aspect ratio, as well as surface information, such
as texture and roughness parameters, cannot be obtained
using ensemble techniques. As presented in Figure 2, there
are three methods of characterizing nanoscale particles --
SEM, TEM, and AFM .
TEM and SEM are examples where an established singleparticle
characterization technique is combined with image
processing to measure and analyze particles. The emergence
of the application of AFM to nanoparticle characterization
has developed over the past 10 years. AFM is well suited to
individual particle characterization including: parameters,
volume, size, shape, and particle surface morphology. With
single-particle techniques, physical parameters for each
particle in a set of particles can be recorded and the data set can be processed to generate a statistical distribution (i.e. ensemble-like information) for the entire set of particles.
A major factor that contributes to microscope single particle analysis (and ultimately the accuracy of measurements) is image quality. The
AFM is similar to electron microscope techniques, SEM and TEM, where proper sample preparation is the key to measuring high quality
data. The art of sample preparation is in fact a simple procedure of critical-path steps, where every single step makes a difference.
TEM is well known for very time-consuming and complicated sample preparation8,9. SEM samples are easier to prepare, however the requirement for conductivity adds some difficulty. AFM samples do not have to be conductive, which makes sample preparation easier for the user, Figure 3. However, there are very important criteria to be met in order to do AFM imaging.
The main focus of this article is to present a general survey of AFM sample preparation methods for nanoparticle characterization. The
authors present a review of the available methods for AFM particle imaging and characterization in ambient conditions and in liquids. When available, references for additional information are provided. All AFM images shown is this paper have been obtained on a Light Lever or Crystal Nano-Rp™ AFM. Close contact mode is the imaging mode for all images scanned in air this paper.
Particles, Dispersion, Substrates, Adhesives
The vast variety of particles can be categorized as engineered and non-engineered environmental particles. Engineered or artificially
created particles break down to organic and inorganic categories. Those categories in turn have additional subdivisions, for example powders, suspensions and embedded particles. Different approaches can be used to prepare AFM samples10-16. Particle size, hydrophobicity, native environment and bio-compatibility are taken into consideration.
AFM particle imaging requires that:
a) The particles to be rigidly adhered to a substrate.
b) The particles to be dispersed on a the substrate.
c) The substrate roughness is less than the size of the nanoparticles .
Often an adhesive is required for affixing the nanoparticles on a substrate. There are a large number of adhesive choices of the
adhesive for small particle deposition. The most commonly used chemicals are poly-l-lysine, poly-D-lysine17, PEI (poly-ethyleneimide) or
APTES (aminopropyltriethoxy silane) to facilitate chemical bonding between particle and substrate. Functionalized surfaces might either
promote adsorption or allow covalent bonding. Sometimes hydrophobic substrates are preferable in air-AFM, this way formation of a
water monolayer can be avoided12. HPOG or spin-coating of polymers is a convenient method for creating hydrophobic substrates. If
imaging needs to be done in an aqua-solution then hydrophilic agents should be used. Particle-substrate affinity has to be stronger
than tip to particle interaction. Buffer solution chemical composition, pH, can be modified to maximize the adhesion between particles
and the substrate, when imaging in liquid10.
If the nanoparticles are not dispersed on the substrate it is not possible to characterize them. Establishing the optimal method for
dispersing nanoparticles on a substrate often requires experimentation. The challenge of dispersing nanoparticles on the substrate
is great primarily because of several competing factors. On a large scale, exposure time and dilution of the particle solution must be
considered. On a smaller scale, the interfacial free energy and electrostatic energy associated with the nanoparticles tend to cause them
to clump together or keep them far apart. On other hand, hydrophilic-hydrophobic forces interacting between particles, substrate and solution can cause agglomeration and coalescence. In many cases additives and surfactants present in particle suspension may cause various effects on dispersing of the particles, especially during and after evaporation.
In general, for AFM particle analysis the smaller the size of the particles the flatter/smoother the substrate must be. In other words, the
size of the particles should be greater than the topographical features of the substrate. The most commonly used substrates include:
glass cover slips, mica, HOPG graphite, silicon oxide wafers, and atomically flat gold. This is the reason that atomically flat substrates
are preferable for biological applications, for example imaging DNA12 and proteins. Glass, mica and silicon work very well for fine-size
features, like bio-cells, colloids, quantum dots and carbon nanotubes. Sometimes polymer membranes, filters or macromolecular gels can be used to immobilize large particles. If a sample comes in the form of a bulk material such as wood or epoxy-resin, then metal discs are normally used as a substrate. The adhesive used in this case is typically carbon tape or thermal wax.
Examples of Imaging Nanoparticles with AFM
There are numerous combinations of nanoparticles, substrates and adhesives that have been demonstrated to work with an AFM. Described here are examples of some of the most common types of applications addressed with an AFM.
1.0 Imaging in air
A substantial advantage of using an AFM for Nanoparticle characterization is that the scanning may be done in ambient conditions. Thus, most applications for characterizing Nanoparticles are performed in air.
1.1. Dry powders
A great variety of particles are produced or
distributed as dry powders. Commonly used
substrates for ultra-fine powder deposition
are glass slides, HOPG and mica. In order
to increase the adhesive properties of the
substrate, poly-L-lysine is may be deposited
on the substrate’s surface. Once a substrate
is chemically treated and dry, it is immediately
ready for powder deposition. Powder
distribution is achieved by dusting a small
amount of powder over the entire area of
the substrate, and setting it aside for a few
minutes. Then the substrate is flipped over
to remove large agglomerate of particles.
Dry adsorption works very well for super
fine powders, particle size less than 150nm,
Figure 4. Deposition rate as well as density of the deposition are tow of the challenges associated with this method.
If granular size is larger than 500nm a
different method should be used. A polished
metal disc works very well as a substrate and
thermal wax works well as anchoring medium.
A piece of wax is placed on the metal disk
and is warmed up on a heating element until
the wax softens, at approximately at 60-70
degrees C. After the wax softens, there is
a visible liquid interface or its surface. After
seeing the liquid interface, remove the metal
disc from the heater, wait until the surface
just starts to solidify, and sprinkle some
powder over the sample area. The sample is
ready for AFM imaging when the thermal wax
becomes solid and the metal disk is at room temperature, typically after 10-15 minutes. Experimentation is often required to obtain the optimal particle surface density. The depth
of the embedding depends on particle weight and size as well as the temperature of the thermal wax, Figure 5-A and 5-B, and is very difficult to control with this method. If this effect is undesirable, droplet deposition should be used. In the case of the wet method, the amount of the particle embedded into the anchoring media is negligible.
A novel approach to creating nanoparticles is laser ablation. It is demonstrated that laser ablation may be used for depositing
nanosize metal clusters on substrates18. In
this case there is no additional treatment or
preparation required. This method produces a
wide distribution of particles size and shapes, depending on the conditions of the irradiating laser, Figure 5-C.
1.2. Evaporated Suspensions
Droplet-evaporation or adsorption methods
are used for preparing AFM samples from
liquid suspensions13, Figure 6. A droplet of
liquid is deposited on freshly cleaved mica or a
poly-l-lysine covered slide. The droplet is then
carefully washed after allowing the sample to
sit for about 10 minutes. To dry the sample before scanning, either leave it overnight in a dust protected environment or use a furnace/
heater to accelerate the drying process.
Certain kinds of particles, quantum dots for
example, come in a toluene solution. If the
solution or suspension comes in any nonaqua form, it is very important to choose the substrate accordingly. Glass or silicon work well for toluene, see Figure 6-C.
1.3. Bio-particles
Using the correct sample preparation
techniques for life-science and biological
applications is extremely critical because an
immobilized specimen can degrade during
sample preparation or even during imaging. Requirements for substrate flatness, chemical compatibility and reagent purity are rigorous.
Also, surface charges, surface energy and hydrophobicity play a very important role in selecting the optimal sample preparation
methodology. There are several review papers written on biological sample preparation11, 12. There are the major methods: absorption,
replication, and mechanical trapping. Sometimes additional fixation is necessary and several methods can be combined to achieve the
desired results, Figure 7.
Physorption (physical absorption) or non-covalent methods, for example aerosol-spray deposition, immersion and droplet-evaporation,
is achieved by adsorbing biological cells on highly negatively charged mica. Additional chemical treatment, such as fuctionalizing by salinization, can be used to facilitate stronger bonding on the surface of the biological specimen. Tight affinity to the substrate is
the mandatory requirement for successful AFM imaging. The downside of the non-covalent absorption method is that it could cause undesired re-arrangements of the bacterial cells. If displacement or distortions are critical or if the molecular object has to be integrated into a complex molecular assembly, then covalent methodologies should be used.
Sometimes fixing with glutaraldehyde is
necessary to minimize tip-object interaction
and to prevent possible damage of the
biological sample. Studies show that
mechanical trapping of biological objects in
a membrane filter appears to be the most
reliable method to measure actual surface topography11,20. It is beyond the scope of this paper to describe all existing techniques.
In the case of many pharmaceutical applications, particles come either as
dry powders or liquid suspensions, Figure 8. If this is the case, sample preparation for AFM is the same as for inorganic powders and suspensions as described in Powder/Suspensions, sections, 1.1 and 1.2.
1.4. Carbon-nanotubes
Carbon nanotubes, nanowires and whiskers are a subset of nanoparticles.
These particles are normally produced in large quantities as powders or are
grown directly on a substrate. Arc-discharge, laser ablation and chemical
vapor deposition (CVD) methods have been successful in making carbon
fibers, filaments, and nanotube materials. These methods are well described by H.Dai21. Typically one of two methods is used for preparing nanotube samples for AFM imaging: catalyst growth or deposition. Catalyst growth is the best method for creating a clean sample for studying the unique properties of single-wall nanotubes, Figure 9.
When preparing carbon nanotube samples for AFM imaging with deposition, it is important to use a dispersant. Very diluted dispersant suspensions of carbon nanotubes are spin coated on a silicon wafer, rinsed thoroughly with water, then dried in air. Any commercial spin-coater may be used.
1.5. TEM-samples
Imaging of samples prepared for TEM analysis is very simple and straightforward with AFM. There is no need for additional surface
treatment or even a sample fixture, Figure 10. The 3 mm disc typically used for TEM analysis must be firmly fixed in the AFM sample
stage. Usually double sticky tape or carbon tape works well to secure the disc. The perforation in the TEM sample can easily be located
with the optical microscope in the AFM stage. Once the perforation is located, the AFM probe can then be precisely positioned over the electron transparent area or father away from the perforated area for AFM scanning. It is recommended that vacuum tweezers be used for handling TEM samples.
1.6. SEM-samples
Samples made for SEM imaging are may be directly imaged in an AFM, Figure 11. Sample preparation procedures can be simplified for
AFM scanning because the sample does not require a conductive coating. When compared to an SEM, traditional considerations for good material contrast are not important. For example, in an
AFM precipitated particles on a flat substrate will always appear with good contrast regardless of material choice for the particle and substrate8,9.
2.0. Imaging in liquid
AFM is an essential tool to identify topographical features of
particles submerged in liquid. The range of applications include soft polymers, bio-particles (cells, membranes, viruses) and a variety of inorganic particles.
2.1. Bio-particles in buffer
An important advantage of AFM over other microscopy techniques is the ability of an AFM to image biological samples in a native
aqueous environment. AFM offers the possibility of in-vivo monitoring of the dynamics of biological changes in living cells, viruses and
micromolecular crystals17, 19, 20, 22-26. Imaging in liquid with an AFM requires a stable immobilization of biological objects. Absorption on
a polycationic treated surface or on an agarose coating provide stable fixation for experiments in liquid19. Absorbing specimens directly
from the buffer solution can be controlled by the electrolyte concentration and pH of the buffer solution17,19. Glutaraldehyde fixing is
necessary for certain applications in a bio-AFM sample. The fixing agent is applied after absorption. In fact, fixation destroys molecular functionality and could affect true structure. Hydrophobic substrates and bio-incompatible agents are not recommended for solutionbased measurements and should be avoided.
Both contact or close contact imaging modes can be used in liquid. Close contact mode is the mode of choice for imaging soft samples
in air, however in liquids it may not be the best technique19. In fact, the contact mode is the optimum mode for imaging samples with large contact area, such as purple membrane. In the case of samples with individual particles attached to the substrate, other imaging modes are used
2.2. Inorganic particles in intermittent medium
Both the particle and adhesive holding the particles in place during imaging should be un-dissolvable when imaging in liquid for obvious
reasons. It is also very desirable if both adhesive and particles are hydrophilic if imaging is done in an aqueous solution.
3.0. Imaging of embedded particles
Nanoparticles that are imbedded in surfaces can be visualized using physical measurement techniques such as vibrating phase and LFM27. Certainly, if particles largely extend out of the surface, then traditional methods for topographic imaging will work.
3.1. Soft and bio-materials
Often it is desirable to image particles embedded in a solid medium, bio-tissue, or polymer thin film. In the case of relatively soft material
like organic tissue or soft polymers it is important to cross section the specimen and make a very smooth, clean cut. A microtone is
typically used to produce 0.5 micron or thinner slices that are suitable for AFM imaging, Figure 13-A. The microtone slices must be
firmly fixed on a glass substrate before AFM scanning. Chemical etching of semi-thin sections of an epoxy-resin embedded specimen is a very good technique for visualizing embedded particles with an AFM. Being able to scan with an AFM in this case depends on sample preparation before the particles are embedded in the substrate.
Thin films can be spin coated on silicon or a
glass substrate and the examined with an
AFM with no additional treatment, Figure 13-
B. If a composite material is the subject of investigation and the constituent component have dramatically different elastic, adhesive
or frictional properties, then material contrast
using vibrating phase modes can be achieved.
Vibrating phase imaging can provide unique
information about local materials distribution
for composite/organic thin film, Figure 13-C. Phase-lag data are obtained simultaneously with topography data in AFM2, Figure 12.
3.2. Hard materials
If a matrix material is rather hard, for example metal or ceramic, the standard polishing/ etching technique employed in SEM may be
used for AFM scanning, Figure 14.
There are numerous combinations of
nanoparticles, substrates and adhesives that
have been demonstrated to work with an AFM.
Described here are examples of some of the most common types of applications addressed with an AFM.
3.3. Membranes and defects
Porous materials and materials with cavities/
voids are considered the same as particle specimens, Figure 15. In the case of membrane defect visualization, there is no extra sample preparation required. For bulk materials, AFM sample preparation can be done as described in earlier section
Conclusions
Sample preparation techniques for powders, colloids/suspentions, bio-objects and embedded particles are outlined in this paper.
Examples of procedures along with AFM images illustrate material science, pharmaceutical or biological applications. AFM sample
preparation is derived from traditional optical/SEM/TEM methods,and is much simpler and less time consuming. Nevertheless, the
importance of adequate surface treatment can not be overemphasized.
Speeding up the process of obtaining vital data is critical for many applications and definably makes AFM more attractive for individual
particle imaging. In general, AFM individual particle size characterization is both cost and time effective, see Figure below. AFM resolution
is greater or comparable to traditional techniques. The main advantage of AFM for particle characterization is unambiguous morphology
determination along with direct measurements of volume and 3D display.
Advantages of using an AFM for particle analysis are:
• Faster than SEM and TEM
• Instrumentation is more affordable
• Direct three dimensional map
• Works on many types of nanoparticles
Acknowledgments
The authors would like to thank Prof. P. Collins (UCI), Prof A. McPherson (UCI), Dr. Yu. Kuznetzov (UCI), Dr. M. Hines (Evidenttechnology,
Inc.), and Dr. J. Baldeschwieler (CalTech) for fruitful discussions.
References:
1. G. Binning, C.F. Quate, Ch. Gerber, E. Weibel, Phys.Rev.Lett. 49, 57-61 (1982)
2. G. Binning, C.F. Quate and Ch. Gerber, Phys.Rev.Lett. 56, 930-933 (1986)
3. S. Dror, Scanning force microscopy: with applications to electric, magnetic and atomic forces, Oxford
University Press, 1994
4. D.Bonnell, Scanning Probe Microscopy and spectroscopy, Wiley-VCH, 2001
5. B.P. Jena and J.K.H. Horber, Atomic force microscopy in cell biology, Academic Press, 2002
6. Zh. Peng and P. West, Appl.Phys.Lett. 86, 014107 (2005)
7. P. West, Zh.Peng, N.Starostina, American Laboratory, 23-24, april 2005
8. F. Stenley, Scanning and transmission electron microscopy, W.H.Freeman, 1993
9. P. Goodhew, J. Humpreys, R. Beanland, Electron microscopy and analysis, Taylor & Francis, 2001
10. D. Muller, M. Amrein, A. Engel, Jour.Struct.Biology 119,172-188 (1997)
11. V. Vadillo-Rodrigez, H. Busscher, W. Norde, J. Vries, R. Dijkstra, I. Stokroos and H. Mei, Applied and Enviromental
Microbiolgy, 5441-5446, Sept.2004
12. K. Kirat, I. Burton, V. Dupres and Y. Duprene, Journal of Microscopy, 218,199-207 Pt3 (2005)
13. J. Vasenka, S. Manne, R. Gison, Th. Marsh, E. Henderson, Biophysical Journal 65, 992 (1993)
14. U. Swartz, H. Haefke, Th. Jung, E. Meyer, H-J Gunttherodt, Ultramicroscopy 41,435-439 (1992)
15. M. Van Cleef, S.Holt, G. Watson, S. Myhra, Journal of Microscopy, 181,2-9, Pt1 (1996)
16. C. Ritter, M. Heyde, U. Swartz, K. Rademann, Langmuir, 18,7798-7803, 2002
17. A. Engel and D. Muller, Nature Structural Biology 7, 715-718 (2000)
18. A. Morales, Ch.Lieber, Science, 279, 208-211 (1998)
19. P. Wagner, FEBS Letters 430, 112-115 (1998)
20. A. Bolshakova, O. Kiselyova, I. Yaminsky, Biotechnol.Prog 20,1615-1622 (2004)
21. H. Dai, Carbon Nanotubes, Topics Appl.Phys. 80, 29-53 (2001)
22. F. Moreno-Herrero, J. Colchero, J. Gomez-Herrero and A. Baro, Phys.Rev.E 69, 031915 (2004)
23. Tomaselli KJ, Damsky CH, Reichardt LF, J Cell Biol,105:2347 (1987)
24. A. McPherson, Yu. Kuznetsov, A. Malkin, M. Plomp, Jour.Struct.Biology, 142,32-46 (2003)
25. Yu. Kuznetsov, A. Low, H. Fan, A. McPherson, Virology, 323, 189-196, 2004
26. Yu. Kuznetsov, J. Gurnon, J. Van Etten, A. McPherson, Jour.Struct.Biology, 149, 256-263, 2005
27. P. West, N. Starostina, Advanced Materials & Processes, 35-37, February 2004
|
|
|
|
|
|
|
© Copyright 2002 -2006 Pacific Nanotechnology, Inc. All Rights Reserved.
No part of this site can be copied without prior agreement with Pacific Nanotechnology. |
|
|