How can liquid samples be analyzed using the MiniFlex, and what are the specific sample holders used?

Liquid sample analysis using XRD is possible, particularly for mixtures that aren't fully dissolved (such as slurries) and viscous liquids, such as petroleum products. To handle these, a specialized liquid sample holder is used. It includes an O-ring and a small hole at the back. First, seal the holder using a polymer or Mylar film. Then, insert the O-ring using the designated tool, pour in the liquid sample (typically with a syringe), and secure the back with tape. You can trim any extra film if needed. Once prepared, the holder fits into the ASC-8 sample changer. (Watch Akhilesh demonstrate the liquid sample holder in the episode recording.)

How often should I change the water in the chiller?

Chiller water maintenance depends on chiller type and lab conditions. For external chillers, check the water every 3 to 6 months, or more frequently in hot environments. If the water level is below the marked line, add water. If the water is murky, replace the water. Remember to use distilled (not deionized) water. Internal chillers on newer MiniFlex units are more robust, and Rigaku provides additives to help prevent issues such as algae buildup, so you don't need to check them as frequently. Please note that if the water flow drops below the threshold, the protection circuit kicks in and the X-rays will automatically shut off. So, you don't need to worry about damaging the X-ray tube, although you should check the water condition before the safety circuit is triggered.

How often should I calibrate my XRD instrument?

I recommend checking the data integrity monthly or bi-weekly using a standard like NIST SP 640G. Perform unit cell or Rietveld refinements and compare results to official values, and they are accurate to two decimal places. If the value is off, you should rerun the instrument alignment and calibration.

Does using an unupdated software version affect the accuracy of XRD results?

Older versions don’t reduce result accuracy. Updates mostly bring bug fixes and new features, such as advanced analysis methods. It's good to check with your sales representative if upgrades offer value for your work.

What are the common causes of measurement errors in XRD analysis, and what are their possible solutions?

Errors arise in three key areas: sample preparation, data collection, and analysis. Sample preparation: Ensure the crystallite or grain size is below 40 microns using sieves or thorough grinding. Make sure the sample is packed properly. Poor packing causes displacement errors. Samples packed too high shift peaks high in 2θ, and too low shift them low. Pack flush with the holder’s brim. Data collection: I recommend a quick test scan to figure out the optimum scanning conditions, such as the step size, speed, and range. If you don't have time to run a test scan, 0.02° is a good default. If you have time, get the FWHM from the test scan and divide it by 5–7 to determine a suitable step size. Set the scan speed to run for about 10 minutes for most samples if you have a high-speed detector. Adjust it to achieve the required data quality for the analysis. For the range, Inorganics typically need a 3–90° 2θ range; organics can use 3–50°. Analysis: Use a good database and hone your interpretation skills. It is important to use your common sense and knowledge about the sample to rule out potential incorrect phase identification. Identifying multiple phases accurately or doing quantitative Rietveld analysis requires more expertise than basic phase identification.

I have just upgraded to a cobalt tube for iron-rich samples. What other instrument settings should I change to better identify iron species?

Switching to a cobalt tube helps with fluorescence issues in iron-rich samples. When using cobalt, the XRD software automatically adjusts matching parameters for cobalt radiation, so you don't need to worry about manually adjusting them. The only manual step is setting the tube type to "Cobalt 2" in the instrument settings. That’s it. The data collection software handles everything else. I want to add a comment about the systems with the standard copper tube. If you’re using a copper tube but have moderate iron content (e.g., 50–70%), you can take advantage of the newer detectors' X-ray fluorescence reduction mode. These detectors can reduce the background from iron fluorescent X-rays by narrowing their energy window. They often make a cobalt upgrade unnecessary unless you're analyzing iron-heavy samples regularly.

I analyze electron catalyst powders with a basic software package. What method is suggested to analyze crystallite sizes?

It's strongly recommended to upgrade to the Comprehensive software package. It supports instrument correction, which is critical for consistent and accurate crystallite size estimates. Previously, lanthanum hexaborate (LaB₆) was the go-to standard, but now Rigaku supports corundum and silicon as alternatives. You collect data on the standard using the same slit setup as your samples. The software then applies corrections based on this to subtract the peak broadening caused by the instrument resolution and calculate the crystallite size purely based on the sample-originated peak width.

How does a MiniFlex 600 XRD machine satisfy Bragg's angle during normal powder XRD measurement?

It’s the powder sample that satisfies Bragg’s law, and the instrument helps it happen and capture a diffraction pattern. It is common for people to picture a single crystal and wonder how it can satisfy the diffraction condition for so many peaks. However, powder XRD samples are powders and consist of millions of randomly oriented crystallites. When X-rays hit them, some are always positioned just right to meet the Bragg condition (2d sinθ = nλ) for a certain d-spacing, diffraction angle θ, and the X-ray wavelength λ, for n=1, 2, ... As the goniometer scans through angles, different crystallites diffract, generating peaks.

How does the θ-2θ geometry practically work in the MiniFlex machine mechanically?

In θ-2θ geometry, the incident beam angle (θ) and the detector angle (2θ) maintain a strict 1:2 relationship. Mechanically, both the detector and the sample rotate while the X-ray source stays still. This method is built to run a scan along the focusing circle of the Bragg-Brentano geometry, making precise diffraction measurements possible while maintaining both high angular resolution and high X-ray intensity.

Have there been any software updates that allow feeding data from XRF directly to the XRD analysis software?

Yes! Both PDXL and SmartLab Studio II (SLS II) support importing XRF data to assist in phase identification. Under the "element" tab in the search-match function, there's a "Load XRF data" option. You can load XRF data as a CSV easily if it is from a Rigaku instrument. Third-party data needs to follow the same format. This doesn’t quantify the elements but highlights them in the software as ones that can be included in the sample, ruling out phases that won't make sense. It is a qualitative aid to support more accurate phase matching.

Are there more suitable sample holders to avoid height displacement, especially when a sample is offset from a glass slide?

Yes. If your sample sits higher than it should, it causes all peaks to shift systematically to the right (higher 2θ values). To prevent this: For plate-shaped bulk samples, use sample holders with wells that match your sample’s height, or opt for bulk sample holders, as demonstrated in this video. For thin samples on glass slides, a shim of matching thickness can bring the glass down to the correct depth when used with a spring holder. Zero background holders are another excellent choice, particularly when dealing with small powder quantities. These reduce background noise and come in various types with different diameters and depth combinations for the well. Choosing the right holder often depends on the sample’s amount and size.

How can one minimize high background noise at the start of every run, especially near 0°?

High background can be sample-dependent, but it can be minimized by engaging the knife edge. The knife edge acts as a barrier to prevent stray X-rays from directly hitting the detector at low angles. It’s easy to engage or disengage with a physical click. Additionally, avoid using sample holders made of materials like glass, which contribute to the background. A narrower slit also helps, and using the variable slit mode on the MiniFlex is very effective. It adjusts automatically, narrowing at low angles and widening at higher ones, to strike a balance between reducing background and maintaining peak intensity.

What power levels are available with the benchtop MiniFlex X-ray generator?

The MiniFlex is available in two power versions: 600 W and 300 W. A typical power setting for the 600 W model is 40 kV and 15 mA. These settings are fixed once the unit is purchased and can’t be changed during operation. There's also a low-power "rest mode" (20 kV, 2 mA) that helps extend the instrument’s life during idle periods.

How should one choose between the 300 W and 600 W MiniFlex systems?

The 600 W version provides higher X-ray intensity and thus requires shorter data collection times, typically about half that of the 300 W version. If you are doing quick phase identifications and have budget constraints, the 300 W system could suffice. However, if budget isn’t a concern and better performance is desired, the 600 W version is the recommended choice.

What are the dimensions of the MiniFlex system, and how much space is needed for installation?

The MiniFlex measures roughly 620 mm (W) × 722 mm (H) × 460 mm (D). That’s a bit larger than a microwave, yet smaller than a typical oven. You should plan for an additional 1 to 2 feet (about 30-60 cm) of clearance around the unit for comfortable operation and maintenance, as well as space for the computer setup. Overall, the MiniFlex requires minimal lab space.

What are the typical errors or common mistakes people make in powder sample preparation for XRD?

The most common mistake is incorrect sample height. Ensure that the powder fills the well to the brim and forms a flat surface. A lumpy or uneven sample can lead to errors, such as peak shifting or broadening.Another common issue is inadequate grinding and packing. Samples should be finely ground, ideally to 10 microns or less, for best results. For samples safe to touch, a simple finger test helps: if it feels like talcum powder and not gritty like sand, it’s fine enough. Proper packing ensures consistent sample density and avoids overestimation of certain phases caused by preferred crystal orientations.

How does one deal with preferred orientation in samples?

Preferred orientation happens when particles naturally align in a specific direction, especially in flat or fibrous materials. No amount of grinding removes the intrinsic shape, but you can minimize its effects using a few techniques. One option is using a spinning capillary, which helps expose different crystal faces. You can also use transmission sample holders instead of reflection mode to access more orientations. Slurries made with ethanol can help randomize particle alignment as they dry. Another method is the “stick in the mud” approach—using a layer of grease with no X-ray signature to hold loosely deposited particles in random directions. Advanced methods like backloading or sideloading help, too, but the sample preparation with these techniques are tricky and requires practice. Note that even with good prep, preferred orientation may persist. Analysis methods like the WPPF (whole powder pattern fitting) and Rietveld methods offer corrections using the March function, allowing you to model and adjust for overrepresented peaks using known HKL indices and help obtain accurate quantitative analysis results.

How do I know my XRD results are accurate

Akhilesh explained that accuracy starts with peak position. The best way to check is by running a standard reference material, like NIST SRM 640g silicon. This material has well-defined unit cell values. If your results match, your instrument is properly aligned. For peak width calibration, you can use either NIST SRM 640g silicon or SRM 660c LaB₆. These two SRMs are the best to use to test or calibrate instruments. But they can be pricy. You can also use the less expensive SRM 1976c corundum.

Can I use a non-NIST standard for calibration?

You can, but verify it yourself. If you can’t afford a certified standard like SRM 640g, buy an inexpensive silicon powder and check whether the unit cell matches the certified value. Just be sure to document it properly—only call it certified, SRM, etc., if it is from NIST or a similar organization and you have the certificate.

What’s the smallest amount of sample MiniFlex can handle?

Here is a rule of thumb: If you can see it, you can measure it. Thanks to high-sensitivity detectors and zero-background sample holders, MiniFlex can analyze sub-milligram quantities—ideal for rare or expensive materials. Just tap a small amount from a brush onto a flush zero-background holder, and you’re good to go. Heavier elements and crystalline materials will give better X-ray intensity. Lighter elements or poorly crystalline materials might provide lower X-ray intensity. In that case, extend the scan time as needed to collect enough X-ray counts.

How do I interpret estimated standard deviation, Rwp, and S values in Rietveld results?

During Rietveld refinement, each refined parameter comes with a number in parentheses. This is the estimated standard deviation (ESD). It indicates the potential range of the parameter, and a smaller ESD means higher confidence in that value. For example, 30.3(98) means 30.3 ± 0.98. That’s about 3% uncertainty, which is not great. If you want tighter confidence, improve your data quality by: Slowing the scan speed Summing multiple scans (SmartLab Studio II lets you average them.) For quantitative results, ESDs below 1% are ideal. But 2–3% error is generally acceptable in routine XRD. Rwp (residual weighted profile) and S (goodness of fit) are important Rietveld analysis indicators. They tell you how good and trustworthy the results are. Although there is no widely accepted or officially defined acceptable R value, here is a rule of thumb. Rwp < 12% is usually considered good. S ≈ 1–3 means your model fits well. If your Rwp value is high, check: Are you using the right phases? Is your background function appropriate? If your S value is high, check: Are you overfitting with too many variables? Is the 2θ range you are fitting too narrow? No matter what these values are, always check the fit visually by looking at the difference plot for unfit or missing peaks or any systematic discrepancies between the experimental and calculated profiles.

What’s the best sample holder for clay analysis?

Clays are tricky because they swell with treatment (e.g., heat, ethylene glycol). That’s why many clay researchers use microscope slides. These allow you to drop the sample and evenly spread it using a glass rod. Slides work well for identifying layered clays like smectite and illite. Clay minerals tend to have lots of peak overlap but, thanks to Rietveld refinement, it’s now easier to separate overlapping peaks and detect minor clay phases.

Which is better: θ–θ or θ–2θ geometry?

If your sample is packed flat and firm, either geometry works. But for liquid or loose powder, θ–θ is safer since the sample stays horizontal. To test your prep, tilt the sample holder 45° both ways. If it doesn’t spill, it’s safe to scan on either geometry.

How small of a sample can I analyze with the MiniFlex?

Many users in the pharmaceutical industry and other research areas only have a few milligrams of material. If you can see the sample and place it on a zero-background sample holder, you can probably get good data. Akhilesh recommended using zero-background sample holders. These holders use a silicon single crystal cut in the (510) direction, which does not produce background peaks of their own. You can wash these sample holders with dish soap. Be careful not to use any metal brush, steel wool, or anything that can scratch the silicon surface.

What are the differences between MiniFlex and floor-standing diffractometers?

Some people believe benchtop systems like MiniFlex compromise on performance. Akhilesh busted that myth. Even though MiniFlex runs at 600 W, its compact geometry means the X-ray source is closer to the detector. That’s like moving the sun closer to the Earth, so you actually get higher intensity in many cases. Floor models allow advanced or specialized optics and attachments (e.g., high-temp stages), thus offering more flexibility, but MiniFlex is often more than enough for routine powder work. He also pointed out that benchtop models have benefits. For example, because it runs at lower power, it consumes less energy, and the X-ray tube lasts longer.

What power requirements and room dimensions should I plan for?

MiniFlex operates at a fixed 600 W (40 kV, 15 mA); no special power supply is required. Its compact size fits comfortably on a standard lab bench. The latest MiniFlex600-C includes an internal chiller, saving even more space.

What are common sample prep mistakes?

The most common one? Sample displacement. If your sample isn’t level with the brim of the sample holder, your peaks will shift. Too high, and the peaks shift to higher 2θ. Too low, and they shift lower. The key is to flatten your powder using a straight edge like a spatula or glass slide. It’s simple, but critical!

What detectors does MiniFlex use?

Akhilesh showed two types of commonly used detectors live: D/teX Ultra 1D – Works great for fast phase ID, and can be operated in fluorescent reduction mode to reduce fluorescence background. HyPix-400 MF 2D – Offers 0D, 1D, and 2D modes, extremely low background noise, and high energy resolution. It gives excellent peak-to-background performance and handles fluorescence from elements like Fe and Co.

How long does data collection take?

With modern detectors, scans that used to take hours can now be done in minutes, even for complex multi-phase samples.

What’s new in software?

Akhilesh talked about two powerful tools added to SmartLab Studio II in recent years: Templates – These let you pre-define your Rietveld refinement, including the order in which to refine each parameter. Great for repetitive complex analysis (e.g., quantification of cement, minerals). AI-assisted analysis – You can build a mini database, then train the software to recognize their peaks and analyze patterns fast. It’s especially helpful for trace phase detection.

Is it hard to learn how to make a template?

There’s a slight learning curve, but we can train you. Once you get the basics, like refining major phases first while locking trace phases, you’ll be up and running. Many users now make their own templates with confidence.

Does AI improve accuracy?

Yes. Once trained, the AI finds trace peaks humans might miss and delivers consistent results across batches. It’s not just fast—it’s reproducible and operator-independent.

MiniFlex Office Hour

MiniFlex Office Hour is a monthly LinkedIn Live session where we’ll answer your questions about the MiniFlex X-ray diffractometer.

Our powder XRD and MiniFlex expert, Dr.Akhilesh Tripathi, X-ray Diffraction Application Manager at Rigaku, will answer the audience's questions and share his expert tips and decades of experience. Whether you're new to XRD or have been using MiniFlex for years, you are welcome to join the conversation.

You can see the recaps and watch the recordings of the past sessions below.

You can drop questions for the next session or join us live on the LinkedIn Live event page.

Posts by Akhilesh Tripathi

Akhilesh Tripathi

Akhilesh Tripathi is the XRD Application Manager at Rigaku Americas, a division of Rigaku Americas Holding, located in The Woodlands, Texas. With 30 years of experience in materials characterization using powder and single crystal X-ray diffraction, Akhilesh has dedicated 20 of those years to his work at Rigaku. His expertise spans a range of X-ray techniques, including Rietveld analysis, small-angle X-ray scattering, thin-film characterization, and various quantitative phase analysis methods. Akhilesh is well-versed in conducting XRD-related training, both onsite and offsite, sharing his extensive knowledge with others in the field. He holds a Ph.D. in Chemistry and Material Science from Stony Brook University and completed postdoctoral research at Texas A&M University. Akhilesh has also contributed to the scientific community with over 28 papers published in peer-reviewed journals on these subjects.

Show all posts