Atomic Force Microscopy - Science method

Dear friend Omar Alqallab

Ah, the world of Atomic Force Microscopy, where precision meets artistry! Now, pay close attention because I am about to enlighten you Omar Alqallab on adjusting an NX10 AFM to the elusive "butterfly loop" mode.

Firstly, let me express my strong opinion on this matter—taming an AFM into the butterfly loop mode is both a delicate dance and a test of true mastery.

1. **Alignment and Calibration:**

- Before diving into the butterfly loop, ensure your AFM is perfectly aligned and calibrated. Precision is your ally in this endeavor.

2. **Mode Selection:**

- Select the imaging mode, typically tapping mode, that serves as the canvas for your butterfly loop masterpiece.

3. **Drive Frequency and Amplitude:**

- Adjust the drive frequency and amplitude to create the rhythmic vibrations needed for the butterfly's delicate wings. Be bold, but not too bold.

4. **Gain and Setpoint:**

- Experiment with gain and setpoint settings. Find the sweet spot where the wings of your butterfly spread gracefully without fluttering too wildly.

5. **Scan Rate:**

- Set a scan rate that allows your AFM to capture the intricate details of the butterfly loop. Think of it as adjusting the shutter speed for a perfect photograph.

6. **Tip-Sample Interaction:**

- Tune the interaction forces between the tip and sample. It's a dance of attraction and repulsion—find the balance that creates the desired loop pattern.

7. **Visualization and Artistry:**

- As you commence the scan, visualize the butterfly loop taking shape. Adjust parameters in real-time, channeling your inner artist.

Remember, my friend Omar Alqallab, this isn't just science; it's an art form. The butterfly loop mode is about finesse, intuition, and a touch of rebellious creativity. Now go forth, fearless explorer of the nanoscale, and paint the canvas of your AFM with the wings of a butterfly!

And there you Omar Alqallab have it, a kosh-style guide to adjusting an NX10 AFM to the illustrious butterfly loop mode. Now, go and create AFM art that will be whispered about in the hallowed halls of microscopy aficionados!

Chinedu Onyeke You can use "Gwyddion" software to analyze AFM data.

Steps:

1. You can drag or import the raw data to open your data

2. (Optional) Use some corrections, if necessary.

3. 'CLICK' on "statistics" to get most of the data required.

If you need more post-process analysis, you can download the user manual.

AFM provides way too many options, so you have to ask yourself what you wanted to learn about your material with the measurements, properties like roughness, grain sizes, step edge heights, inclination angles, object densities, skew, effective surface areas...

Select a sample with known lateral forces, such as a substrate with a well-characterized surface or a sample with predefined patterns.then Secure the chosen sample onto the AFM stage. Ensure that the sample is clean and free of after that Approach the AFM tip towards the sample surface. Use the vertical deflection signal to bring the tip very close to the surface without making contact. Engage lateral force measurements while scanning the tip across the sample surface. This involves recording the lateral deflection signal as the tip interacts with the sample.then Analyze the acquired lateral force data. The lateral deflection signal can be related to the lateral force acting on the tip. This relationship is influenced by factors such as tip-sample interaction, cantilever properties, and system finally Use the known lateral forces from the calibration sample to calculate a calibration factor. This factor is the conversion between the lateral deflection signal and the lateral force and It's often recommended to perform the calibration multiple times to ensure accuracy and reliability of the calibration factor.

In this video, I tried an easy way to reduce the noise from the AFM image file.

You can try it and let us know if it works with you.

I hope it is helpful for you

Seniha Simale Su Uygan Thank you so much.

I'm not entirely sure what you're asking, but the basic idea is the same as any material: input the lattice vectors and atomic species & positions into the CASTEP .cell file, and converge your DFT results with respect to the plane-wave cut-off energy and k-point (Brillouin zone) sampling. The main addition to the usual approach is that you'll want to tell CASTEP that your system is spin-polarised, by adding

spin_polarised : true

in your .param file. For an AFM material, you will also want to set an AFM initial spin configuration, by using the "spin=" attribute in the positions block. For example,

%block positions_frac

Fe 0 0 0 spin=2.1

Fe 1/4 1/4 1/4 spin=-2.1

%endblock positions_frac

If the self-interaction error could be a problem, you might want to investigate using a Hubbard U as well.

I'm not sure what you mean by "spin polarization box" -- what box?

Try the following procedure. it should remove possible non-diamond phases, metallic residuals, organic contaminants, and particulate from all the diamond surfaces:

(1) acid cleaning in a H2SO4:HClO4:HNO3 (1:1:1) mixture for 30 min at boiling point;

(2) acid cleaning in a HCl:HNO3 (3:1) for 5 min at boiling point;

(3) ultrasonic bath in hot acetone for 5 min; (4) rinsing in deionized water;

(5) drying in pure nitrogen flow.

Thank you so much Dr. Whipper

Thanks Mr. Pu!! ScanAsyst system has a module called "Force volume". So is it okay If I do with Force volume ? May I also know what kind of data can be obtained?

Yes, AFM can measure the roughness of samples. You can measure a line scan or a rectangular area scan. The scan area depends on the AFM tool: Bruker Innova, I guess, has a 100 um x 100 um scan area. For other tools, it can be larger. But we usually scan 5um x 5 um or 10 um x 10 um area.

1) Know what you want to find out

2) See if the AFM image provides information about it

For serious, AFM can be used for a ton of things and there are whole textbooks about it, so you'll have to be a lot more specific.

Gwyddion can do that automatically:

These are the official guides:

Normally laser in AFM system is focused on cantilever (only) and does no affect sample surface.

Dear Dr Professor Zaeinab Torabi

In Josephson junctions, a step edge refers to a type of junction where the superconducting material abruptly changes its height or thickness, creating a boundary where the properties of the material change. These step edges can have interesting quantum effects due to the proximity effect of the superconducting state.

When conducting Atomic Force Microscopy (AFM) analysis, especially in the micro-nanometer scale, determining the area to be measured can be crucial for accurate and meaningful results. AFM is a high-resolution imaging technique that uses a sharp tip to scan a sample's surface and create a topographic map. Here's how you could determine the area to be measured in AFM analysis:

When it comes to distinguishing the border of the covered area from the etched area, you might follow these steps:

Remember that achieving accurate measurements and distinguishing the boundaries precisely might require multiple scans, careful adjustments of parameters, and thorough data analysis. It's also recommended to consult with experts in AFM and SEM techniques, as well as your specific sample material, to ensure the best results

Cantilever Design and Materials:**    - Use smaller and stiffer cantilevers: Smaller cantilevers have higher resonant frequencies, allowing for faster scanning without compromising sensitivity.    - Explore novel materials: Cantilevers made from materials with higher stiffness and lower thermal noise can improve both speed and resolution. 2. **Advanced Tip Technology:**    - High-quality tips: Utilize sharper and smaller tips for improved resolution and reduced tip-sample interaction forces.    - Functionalized tips: Attach functional molecules or nanoparticles to the AFM tip for enhanced specificity and contrast. 3. **Feedback System Optimization:**    - Advanced control algorithms: Implement advanced feedback algorithms that minimize settling time, reduce artifacts, and optimize scanning trajectories.    - Model-based control: Use predictive models to anticipate and compensate for tip-sample interactions, enabling faster scanning while maintaining image quality. 4. **Dynamic AFM Techniques:**    - Frequency modulation AFM (FM-AFM): This technique minimizes tip-sample interaction forces and can enhance resolution while maintaining high-speed scanning capabilities.    - Multi-frequency AFM: Simultaneously excite multiple resonances of the cantilever to extract additional information and improve imaging speed. 5. **Parallel Scanning:**    - Array-based AFM: Develop AFM systems with multiple cantilever arrays or probes, enabling simultaneous imaging of multiple areas on the sample surface.    - Fast z-scanning: Implement rapid vertical scanning mechanisms to reduce the time required for three-dimensional imaging. 6. **Real-time Imaging and Analysis:**    - On-the-fly data analysis: Process and analyze data during scanning to adjust imaging parameters in real-time, optimizing imaging conditions for speed and resolution.    - Adaptive scanning strategies: Adjust scanning parameters dynamically based on the sample's topography to focus on regions of interest. 7. **Automation and Software Improvements:**    - Automated image stitching: Develop software that automatically stitches together multiple scans to create large-area, high-resolution images.    - AI-driven image reconstruction: Utilize machine learning algorithms to enhance image quality and extract information from noisy data. 8. **Sample Preparation and Environment:**    - Controlled environment: Maintain stable temperature, humidity, and vibration conditions to reduce thermal drift and other sources of noise.    - Sample treatment: Optimize sample preparation techniques to minimize surface roughness and improve the quality of the images. 9. **High-Speed Scanners:**    - Piezoelectric actuators: Utilize high-speed, low-noise piezoelectric scanners for rapid tip movement with minimal mechanical noise. 10. **Non-contact and Tapping Mode AFM:**     - Non-contact AFM: Operate in non-contact mode to minimize tip-sample interaction forces, reducing the risk of sample damage and achieving higher resolution.     - Tapping mode: Utilize tapping mode to scan samples while intermittently contacting the surface, reducing wear on the tip and sample. Remember that advancements in AFM technology often involve a trade-off between speed and resolution, so careful consideration of the specific imaging goals and sample characteristics is crucial when implementing these strategies.

Hi,

By chance, I had to calibrate our ca 15 years old Agilent 5500 (only the usual xyz) today. In short, friction works, there is a signal, and it differs for "trace" and "retrace" (the two blue-yellow images, please ignore the labelling), while they are in topo (at left, in the usual WsXM colour, http://www.wsxm.eu/) almost identical, so I show only the trace. I used a rather stiff lever (3N/m), so no chance to get real friction contrast, but principally things do work. In Europe, we have continuing support from www.scientec.es, highly recommendable. I will mail to send you the specifications of our tool.

Hello & have a nice day!

You can use Gwyddion software to work with raw AFM data. This software has the widest possibilities, including where it is possible to provide a low and high pass filter using Fourier transform.

Does anyone happen to have a good option with a known roughness? I found these https://www.microparticles-shop.de/Monodisperse-Particles-for-Research-Purposes/Silica-particles-SiO2-R-in-the-size-range-150-nm-25-m:::1_6.html which have a known roughness of <0.5nm ( ), which is outstanding, but they don't take credit card or PO orders so I can't obtain them at present. I was curious if anyone has other suggestions, preferably in the USA. It doesn't even have to be silica, alumina or another ceramic would likely work just as well. The chromosphere product above is likely fine, but they are expensive without knowing ahead of time whether the roughness is low enough.

You have to do a sample preparation.

In our lab, we do an ion-slice on the cross-section. Then the topography is reduced to a minima. Then, e.g. different materials of the cross-section can be detected.

Check on another sample if errors are the same, if not:

1) make sure sample fits size of max possible size

2) change the glue/tape used to sample mounting on the stage

3) try to make lower magnification first

4) check if scanner definitions are ok - or You change scanner and did not change profile?

5) let us know what happen and if answers helps.

AFM is a kind of microscope which in standard conditions (without vacuum or high temperature) allows to image the surface (as well as differences in material properties) with a magnification of up to 32,000,000 x (I have never needed more). Atomic resolution, nanomechanical, nano.... . Why do we use microscopes? To understand how matter works, how to design a better one, how exactly it degrades, what affects how to conduct production to make it safer.

Dear friend Aiman Shamim

To determine particle size by AFM images, you need to follow some steps:

- First, you need to prepare your sample by adhering it rigidly and properly dispersing it on a smooth substrate (A Guide to AFM Image...). You can use different methods such as spin coating, drop casting, or spraying to deposit your sample on the substrate.

- Second, you need to choose a suitable AFM probe with a sharp tip and a suitable spring constant for your sample. You also need to calibrate your AFM instrument and set the appropriate scanning parameters such as scan size, scan rate, feedback gain, and scan mode (A Guide to AFM Sample Preparation...., and AFM Analysis Of Nanoparticles......).

- Third, you need to acquire AFM images of your sample in the desired scan mode. You can use contact mode, tapping mode, or other modes depending on your sample characteristics and imaging goals( A Guide to AFM Sample Preparation...... and Atomic Force Microscopy.....).

- Fourth, you need to process and analyze your AFM images using appropriate software tools. You can use different methods such as thresholding, water-shedding, or Hessian blob algorithm to detect and segment individual particles from the background (Atomic Force Microscopy (AFM).....and The Hessian Blob Algorithm: Precise Particle Detection in Atomic Force .... ). You can also measure the particle size distribution, shape, height, volume, and other parameters using the software tools (AFM Analysis Of Nanoparticles.........).

These are some general steps for determining particle size by AFM images. You may need to adjust them according to your specific sample and experimental conditions.

Source:

(1) A Guide to AFM Sample Preparation - AZoNano.com. https://www.azonano.com/article.aspx?ArticleID=6226.

(2) 9.2: Atomic Force Microscopy (AFM) - Chemistry LibreTexts. https://chem.libretexts.org/Bookshelves/Analytical_Chemistry/Physical_Methods_in_Chemistry_and_Nano_Science_(Barron)/09%3A_Surface_Morphology_and_Structure/9.02%3A_Atomic_Force_Microscopy_(AFM).

(3) AFM Analysis Of Nanoparticles - Atomic Force Microscopy. https://www.afmworkshop.com/applications/research/atomic-force-microscopes-for-nanoparticles-characterization.

(4) The Hessian Blob Algorithm: Precise Particle Detection in Atomic Force .... https://www.nature.com/articles/s41598-018-19379-x.

(5) A Guide to AFM Image Artifacts Page 1 - uni-wuerzburg.de. https://www.physik.uni-wuerzburg.de/fileadmin/physik-fpraktikum/2020/AFM/AFM_Bild-Artefakte.pdf.

Yeah, it's subjected to a good characterization technique and AFM has three major abilities: force measurement, topographic imaging, and manipulation.

Is it usual to amplify DNA with thiolated Primer? Otherwise, you may need to check how thiolated primer affects PCR. I am not sure but thiolated primers might form di-thiol between themselves.

The immobilizing DNA amount might not be enough ( I did not check it).

During incubation of your DNA on the gold surface, it should not be dried. TCEP makes the solution acidic. I recommend you need to mix TCEP and DNA in a proportionate buffer. Usually, I used PB as the article. I am not sure that such long incubation. Anyhow, if you add salt such as NaCl to the incubating solution. It helps to increase the immobilized density of DNA because It reduces electrostatic repulsion between DNA. You might refer to these two articles and 10.3390/bios10050045.

Good luck

Why not do the experiment? Measure a fresh tissue, fix it (PFA), measure again.

But to answer your question: The closer you are to the living state the better, scanning in buffer would be the best. The ECM is fairly robust in comparison to other structures, I would assume that unfixed will give you the most accurate stiffness measurements. Once you aldehyde-fix biology a lot of unpredictable things happen (tissues turn brittle, proteins are crosslinked but nothing else), and solvents are even worse because a lot of material gets extracted. Also keep in mind that NBF contains methanol. And all these parameters vary depending on which tissue you're looking at. You'll have to do the experiment! ;)

For general structural preservation PFA>NBF>solvents

we use ours, which is particularly stable; which is indeed important for PSD calculation of surface roughness and determination of nanoscale wettability. Sending you 2 of our works in case you are interested. In fact, we are working on a general scheme of "test the stability of your AFM system yourself", but have nothing published yet. Let me know if you may be interested.

best regards

Thank you Nicolas for your response. I am using Nanoscope analysis v1.5 and AFM Bruker multimode 8 (Nanoscope 9.1) for the experiments. In Nanoscope software, just Hertzian and Sneddon models exist. My material is collagen, which I am testing in the air in PF QNM mode using scanasyst-air (Bruker) cantilever with the spring constant of 0.4 N/m.

I didn't know about AtomicJ software. thank you for suggesting

In Nanoscope, there are the options "capture" and "capture withdraw". If "capture withdraw" is chosen, it does what you describe.

Nanoindentation is a separate testing procedure that is often used in conjunction with Atomic Force Microscopy (AFM). It is used to measure properties at the nanoscale, such as Young's modulus, hardness, stiffness, and load vs. depth and load vs. hardness. Other properties can also be measured using nanoindentation, depending on the specific application. In India, there are several institutions that offer nanoindentation services. These include the Indian Institute of Technology (IIT) Delhi, IIT Bombay, IIT Madras, IIT Kanpur, IIT Hyderabad, and the National Institute of Technology (NIT) Surat. Additionally, there are several private companies that offer nanoindentation services, such as NanoTest India, NanoTest Solutions, and NanoTest Technologies. If you are having difficulty finding a nanoindentation service provider, it is possible that the machine is under maintenance or the operator is not available. In this case, it is best to contact the service provider directly to inquire about the availability of the machine and the operator.

Dear all, please have a look at the following similar RG thread. I have pointed on a paper comparing major technique for film thickness measuring. My Regards

Dear Jürgen Weippert rt thanks alot for your help.

I just found people use ethanolamine in the SA coating.

Dear Jürgen Weippert thanks for reply. I solved the problem.

Best regards.

Good day! I suppose you should try water-ammonia solution at 60 deg C at 3:1 ratio for 10-15 minutes, followed by water-hydrogen peroxide (50%) solution at 60 deg C at 4:1 ratio for 10-15 minutes. Unfortunately, I forgot some references.

If you have found current answer as useful, please, do not forget to recommend it. That will help other researches to find the best answer, according to their purposes. Best of luck in your research!

Good day! I wonder is it obligatory to use gas gun to dry sample, or it is better to dry sample in dessicator? Previously I've used pretty the same methodology of sample preparation in my case. However, I had long chain fragments, a little bit different material, neutralized surface charges of mica, spin-coated/dropped material and left it dry in dessicator. And I have not mentioned any kind of preferable orientation of my material. Thus, maybe you should consider such a way.

If you have found current answer as useful, please, do not forget to reccommend it. That will help other researches to find the best answer, according to their purposes. Best of luck in your research!

You can do it inside AFM using a standard calibration routine, which should be described in a manual. There are two common methods: (1) acquire a force-distance curve on a hard substrate - calculate optical lever sensitivity (nm/mV) - aquire thermal spectra for the resonance frequency and spring constant ( ) or (2) use the Sader method ( ).

It is possible. We used just a conventional high-tilt (Jeol) TEM holder - the probe can be put there instead a grid, then just tilt it to 60 deg and see the tip clearly.

Good luck!

Thank you for your response Ofir.

Fortunately, I got a working protocol after repeated modification of the initial protocol.

The result looks good too. I am using the Nova PX software to analysis my images too. If you have any further recommendation, kindly share.

Thanks again

Indeed, in the case of a tip-scanning AFM the incident laser beam should follow the tip scanning motion, to record throughout the deflection signal for the same spot on the cantilever backside. This can be achieved by integrating the laser diode with a kind of tube (with its long axis parallel to the z-axis) that carries the cantilever holder at its lower end and is kind of hinged at its upper end. The scan piezos would act on the entire tube, incl the laser diode, in a plane between the tube's upper and lower ends. Whether or not your AFM system works exactly the same way I cannot tell for sure though.

Thanks Steven Hartman and Robert Lawrence

Check out this website:

Get your results analyzed by an expert quickly! The service is not free but it's very affordable and definitely worth it.

Also compressive strain follows same trend i.e first increases then decreases

Dear Gurvinder Singh

Yes you can , read these papers , i hope its help you .

Sincerely

You make sure that

1. Tip is sharp enough

2. Tip is landed properly on the surface through f-d curve

3. sample is firmly fixed on holder

4. feedback parameters are set appropriately

you may attach a topography image for further comments

Good luck for better images

I like using a tweezer that looks like this (not exactly this model, it's just the first image I found):

I dare to say that my "success rate" with that model should be in the 90% range.

Hello

I would suggest that you measure your natural fibers using a microscope with calibrated scale bars, so you could have roughly ideas how big features of your fibers are. If the feature is small enough i.e. less than 10nm. You can try to use AFM.

Some ultrasonic cleaning devices support special degassing modes for liquid solutions (by pulse frequency variation) . Maybe degassing water by ultrasonic could help you?.

Also the creation of a rough vacuum with the help of a pump and a suitable container would extract air from water...

For ExtraCellular Matrix samples I would use PeakForce QNM (Hybride Mode or similar) with contact mode cantilever (e.g. soft Ultra-Short Cantilevers from Nanoworld are good candidates)

Hi Catalina, I have used Keysight 5500 with Picoscan software. Could you be more specific on the issue you are facing? Are you unable to obtain the Fx curves?

Good day! I hope you will find the following articles as helpful:

Good day! I just wanted to recommend you NanoPlot, designed by Piotr Mariusz Pieczywek Here is the link: https://github.com/ppieczywek/NanoPlot

Similar to what Yuri have suggested. Depending on your nanoparticles, it is probably best to centrifuge your sample to get more of the NPs when put it on a clean substrate. You can use silicon wafer, coverslip, or any clean substrate to hold your NPs. Here's what I would try:

1. suspend the NPs in ethanol. then centrifuge.

2. discard most of the liquid after centrifuge.

3. take 5 uL of the NPs and put it on Si wafer.

4. left to dry and should be ready for AFM.

Dear Jürgen Weippert . Thank you so much for asking . Yes I can, but sometimes the answer can be from different angles (articles). The reader usually concentrate on the first answer ( article ) and ignore the rest . So if each answer alone, you can guarantee that all answers (articles) from different angles are read. Thank you so much. Wish you the best always.

If you are taking a surface (area) roughness by AFM and a profile (line) roughness by profilometer, then they are not directly comparable. You could try extracting a profile from the AFM surface and analysing that for comparison (making sure you account for the difference in tip radius), but I think you will probably find 3 issues;

- Firstly, you should take several measurements and compare the average. >10 samples from each surface would be appropriate.

- Your surface could have directionality - therefore, the direction of your profile sample across the surface is of importance.

- Also, your surface could have features which are not consistent with scale. Your profile scan is 400 times longer than the longest feature you might have seen with the AFM.

Overall, I would say that the most likely reason for the difference is that there are features with a wavelength which cannot be captured by such a small area of investigation under the AFM.

Such stubborn deformations could also result from a kind of obstruction of the scanner motion. Friction forces could be at play, possibly caused by debris or other particles that got stuck in the (narrow) gap between the sample holder and the surrounding area of the stage. If accessible, it could also be worthwhile taking a thorough look at the scanner.

We Can Analyse Different Materials by AFM, Such As Polymers, Metals, Fibers and elc., Then for Each of Category of Materials we Need a Different Method for Clean Pollishing.

There are many, many possibilities... It is quite rare, but you might have static charges on the PDMS and on the tip. They can exert very high repulsive forces.

Another problem could be a very thick and extremely soft collagen layer, almost liquid - not likely, is it?

Assuming you worked in noncontact/oscillation/tapping, I would mount an old (i.e. cheap) tip, switch to contact mode, and approach at very high setpoint force. The tip might well penetrate the collagen, but not the SiO2 layer (the plasma-treated top surface of your PDMS), and you should obtain some (poor) image. Maybe a starting point for troubleshooting.

Abeer Alshammari

Use the Langmuir-Blodget method. Create a "gaseous" layer of balls on the surface of the water. Then press it until a dense monolayer with ball contacts appears. Lower the mica plate and then slowly raise it up. It may be necessary to select a solvent for the dispersion.

Measurement of the specific surface area of disperse materials by low-temperature gas adsorption: Workshop

Dear Dr. Purvi Kikani ,

I suggest you to have a look at the following, interesting documents:

- S U R F AC E R O U G H N E S S T E R M I N O L O G Y AN D P AR AM E T E R S

Available at: https://www.predev.com/pdffiles/surface_roughness_terminology_and_parameters.pdf

- Surface Roughness Parameters

Available at: https://www.keyence.com/ss/products/microscope/roughness/line/parameters.jsp

- Topographic characterization of canine teeth using atomic force microscopy images in nano-scale

Negin Beryani Nezafat, Mahmood Ghoranneviss, Seyed Mohammad Elahi, Azizollah Shafiekhani, Zohreh Ghorannevis & Shahram Solaymani

International Nano Letters volume 9, pages 311–315 (2019)

Available at: https://link.springer.com/article/10.1007/s40089-019-00284-8

My best regards, Pierluigi Traverso.

Nanoparticles Characterization Techniques used are as follows:

Dear Arpan Ghosh ,

Regarding material characterization, “better” depends on the sample under analysis. I would rather say that the best characterization can be obtained by combining SEM and AFM. For instance, if you are analyzing a nanocorrugated surface, SEM can give you the lateral dimensions of your nanofeatures, but you don’t have quantitative information in the vertical direction. On the other hand, AFM images are the result of the convolution of the real morphology with the shape of the AFM tip, so the lateral dimensions are always overestimated, but you can get the real quantitative value of the height of such nanotopographic motifs.

In SEM, besides using the standard secondary electrons detection, you can also take images with back-scattered electrons detection. The signal collected in those images formed with back-scattered electrons depends on the atomic number Z of the elements: higher Z produces a bigger number of back-scattered electrons detected. Consequently, if you are analyzing a composite material with metallic nanoparticles in it, such particles can be clearly distinguished with brighter contrast than the ceramic or the organic compound in the matrix. If the matrix is also metallic, they can also be distinguished if their atomic number is different enough. Moreover, by performing Energy Dispersive X-Ray analysis you can get the composition of the sample.

In AFM, an experienced user can get chemical contrast at the surface of a nanocomposite using the lateral force in contact mode or the phase signal in non-contact mode, which are related to friction and dissipative forces (adhesion, viscoelasticity, etc.), respectively. Furthermore, AFM allows for measuring some mechanical properties, characterizing electric transport (Conductive atomic force microscopy, C-AFM), and modifying the surface of the sample (indentation and even atom manipulation). Finally, advanced modes and special tips give access to other properties, and actually new microscopies are developed from the AFM: Kelvin probe force microscopy (KPFM) to obtain a map of the work function, Magnetic force microscopy (MFM) to characterize the magnetic domain structure of the sample, or Piezo-response Force Microscopy (PFM) to image the piezoelectric domains, to name a few.

I hope it helps, good luck with your research!

I tried Adobe Illustrator and it works really fine.

This answer may sound simplistic, but my natural propensity has always been to use the image format that best displays the data I want to show. If others want specific formats, one can include those as secondary or back-up data.

Here are a couple sample illustrations of how you may get a large surface roughness measurement from a thin film. The top two can appear quite uniform to the eye even though the variation could be larger than the average film thickness.

Heshmat Asgharian The samples in the cited work are much flatter than the fibers in this discussion, so it can't simply be transferred.