Introduction

X-ray fluorescence (XRF) spectroscopy technology is widely applied in geological exploration and mineral analysis due to its advantages of rapidness, non-destructiveness, and simultaneous multi-element determination. Handheld XRF analyzers are particularly crucial for on-site testing of iron ores, enabling quick determination of ore grades, on-site screening of element contents, and monitoring of mining production processes. However, the test results from handheld XRF do not always align with laboratory chemical analyses, with deviations often stemming from improper sample preparation or inaccurate calibration. Therefore, a thorough understanding of the instrument’s calibration methods and analytical conditions is essential to avoid reporting erroneous results.

Overview of the Principles and Calibration Mechanisms of Handheld XRF Analyzers

Handheld XRF analyzers operate based on the X-ray fluorescence effect: an X-ray tube emits primary X-rays to irradiate the sample, exciting characteristic X-rays (fluorescent rays) from the elements in the sample. The detector receives and measures the energy and intensity of these characteristic X-rays, and the software identifies the element types based on the characteristic energy peaks of different elements and calculates the element contents according to the peak intensities. Handheld XRF uses energy-dispersive spectroscopy analysis, acquiring signals from elements ranging from magnesium (Mg) to uranium (U) through a built-in silicon drift detector (SDD), enabling simultaneous analysis of major and minor elements in iron ores, such as iron, silicon, aluminum, phosphorus, and sulfur.

To convert the detected X-ray intensities into accurate element contents, XRF analyzers need to establish a calibration model. Most handheld XRF analyzers come pre-calibrated by the manufacturer, combining the fundamental parameters method and empirical calibration. The fundamental parameters method (FP) uses physical models of X-ray interactions with matter for calibration, allowing simultaneous correction of geometric, absorption, and secondary fluorescence effects over a wide range of unknown sample compositions. The empirical calibration method establishes an empirical calibration curve by measuring a series of known standard samples for quantitative analysis of specific types of samples. Handheld XRF also generally incorporates an energy calibration mechanism to align the spectral channels and ensure stable identification of element peak positions.

Error Issues Based on Calibration Using 310 Stainless Steel

In practical applications, some operators may calibrate handheld XRF using metal standards (e.g., 310 stainless steel) and then directly apply it to the compositional analysis of iron ores. However, this approach can introduce significant systematic errors due to the mismatch between the calibration standard and the sample matrix. 310 stainless steel is a high-alloy metal, differing greatly from iron ores (which are oxide-based non-metallic mineral matrices) in terms of physical properties and matrix composition.

Matrix effects are the primary cause of these errors. When the calibration reference of XRF differs from the actual sample matrix, it can lead to changes in the absorption or enhancement of the X-ray signals of the elements to be measured, causing deviations from the calibration curve. For example, when an instrument calibrated with 310 stainless steel is used to measure iron ores, since stainless steel contains almost no oxygen and has a high-density metal matrix, the excitation and absorption conditions of the Fe fluorescence signal in this matrix are entirely different from those in iron ores, causing the instrument to tend to overestimate the iron content.

In addition to matrix absorption differences, systematic errors can also arise from inappropriate calibration modes, linear shifts caused by single-point calibration, differences in geometry and surface conditions, and other factors. The combination of these factors can result in significant errors and biases in the results of iron ore measurements calibrated with 310 stainless steel.

Calibration Modes of XRF Analyzers and Their Impact on Results

Handheld XRF analyzers typically come pre-programmed with multiple calibration/analysis modes to accommodate the testing needs of different types of materials. Common modes include alloy mode, ore/geological mode, and soil mode. Improper mode selection can significantly affect the test results.

• Alloy Mode: Generally used for analyzing the composition of metal alloys, assuming the sample is a high-density pure metal matrix. Using alloy mode to measure iron ores can lead to deviations and anomalies in the results because ores contain a large amount of oxygen and non-metallic elements.
• Soil Mode: Mainly used for analyzing environmental soils or sediments, employing Compton scattering internal standard correction methods. It is suitable for measuring trace elements in light-element-dominated matrices. For iron ores, if only impurity elements are of concern, soil mode can provide good sensitivity, but problems may arise when the major element contents are high.
• Ore/Mining (Geological) Mode: Specifically designed for mineral and geological samples, often using the fundamental parameters method (FP) combined with the manufacturer’s empirical calibration. It can simultaneously determine major and minor elements. For iron ores, which have complex compositions and a wide range of element contents, ore mode is the most suitable choice.

Principles and Examples of Errors Caused by Matrix Inconsistency

When the matrix of the standard material used for calibration differs from that of the actual iron ore sample to be measured, matrix effect errors can occur in XRF quantitative analysis. Matrix effects include absorption effects and enhancement effects, that is, the influence of other elements or matrix components in the sample on the fluorescence intensity of the target element.

For example, if a calibration curve for iron content is established using pure iron or stainless steel as standards and then used to measure iron ore samples mainly composed of hematite (Fe₂O₃), the metal matrix has strong absorption of Fe Kα fluorescence, while in the ore sample, Fe atoms are surrounded by oxygen and silicon and other light elements, which have weaker absorption of Fe Kα rays. Therefore, the Fe peak intensity produced by the ore sample is higher than that in the metal matrix. However, the instrument’s calibration curve is based on metal standards and still converts the content according to the metal matrix relationship, thus interpreting the stronger signal in the ore as a higher Fe content, leading to a systematic overestimation of Fe.

Calibration Optimization Methods for Iron Ore Testing

For iron ore samples, adopting the correct calibration strategy can significantly reduce errors and improve testing accuracy. The following calibration optimization methods are recommended:

• Calibration Using Ore Standard Materials: Use iron ore standard materials to establish or correct the instrument’s calibration curve to minimize systematic errors caused by matrix mismatch.
• Multi-Point Calibration Covering the Concentration Range: Perform multi-point calibration covering the entire concentration range instead of using only a single point for calibration. Use at least 3-5 standard samples with different compositions and grades to establish an intensity-content calibration curve for each element.
• Correct Selection of Analysis Mode: Select the ore/mining mode for analyzing iron ore samples and avoid using alloy mode or soil mode.
• Application of Compton Scattering Correction: Use the Compton scattering peak as an internal standard to correct for matrix effects and compensate for overall scattering differences between samples due to differences in matrix composition and density.
• Regular Calibration and Quality Control: Establish a daily calibration and quality control procedure for handheld XRF. After each startup or change in the measurement environment, use stable standard samples for testing to check if the instrument readings are within the acceptable range.

Other Factors Affecting XRF Testing of Iron Ores

In addition to the instrument calibration mode and matrix effects, the XRF testing results of iron ores are also influenced by factors such as sample particle size and uniformity, surface flatness and thickness, moisture content, probe contact method, measurement time and number of measurements, and environmental and instrument status. To obtain accurate and consistent measured values, these factors need to be comprehensively controlled:

• Sample Particle Size and Uniformity: The sample should be ground to a sufficiently fine size to reduce particle size effects.
• Sample Surface Flatness and Thickness: The sample surface should be as flat as possible and cover the instrument’s measurement window. The pressing method is an optimal choice for sample preparation.
• Moisture Content: The sample should be dried to a constant weight before testing to avoid the influence of moisture.
• Probe Contact Method: The probe should be pressed tightly against the sample surface for measurement to avoid air gaps in between.
• Measurement Time and Number of Measurements: Appropriately extend the measurement time and repeat the measurements to take the average value to improve precision.
• Environmental and Instrument Status: Ensure that the instrument is in good calibration and working condition and avoid the influence of extreme environments.

Precision Optimization Suggestions and Operational Specifications

To integrate the above strategies into daily iron ore XRF testing work, the following is a set of optimized operational procedures and suggestions:

• Instrument Preparation and Initial Calibration: Check the instrument status and settings, ensure that the battery is fully charged, and the instrument window is clean and undamaged. Use reference standard samples with known compositions for calibration verification to confirm that the readings of major elements are accurate.
• Sample Preparation: Dry the sample to a constant weight, grind it into fine powder, and mix it thoroughly. Prepare sample pellets using the pressing method to ensure density, smoothness, no cracks, and sufficient thickness.
• Measurement Operation: Place the sample on a stable supporting surface, ensure that the probe is perpendicular to and pressed tightly against the sample. Set an appropriate measurement time, and measure each sample for at least 30 seconds. Repeat the measurements 2-3 times to evaluate data repeatability and calculate the average value as the final reported value.
• Result Correction and Verification: Perform post-processing corrections on the data as needed, such as dry basis conversion or oxide form conversion. Compare the handheld XRF results with known reference methods for verification and establish a calibration curve for correction.
• Quality Control and Record-Keeping: Strictly implement quality control measures and keep relevant records. When reporting the analysis results, note key information to facilitate result interpretation and reproduction.

Conclusion

Handheld XRF analyzers have become powerful tools for on-site testing of iron ores, but the quality of their data highly depends on correct calibration and standardized operation. This paper analyzes the errors that may arise when using metal standards for calibration, elucidates the principles of systematic deviations caused by matrix effects, and compares the impacts of different instrument calibration modes on the results. Through discussion, a series of optimized calibration strategies for iron ore samples are proposed, and the significant influences of factors such as sample preparation, probe contact, and measurement time on testing accuracy are emphasized.

Overall, proper calibration of the instrument is the foundation for ensuring testing quality. Only by doing a good job in standard material selection, mode setting, and matrix correction can handheld XRF发挥 (fully leverage) its advantages of rapidness and accuracy to provide credible data for iron ore composition analysis. Mineral analysts should attach great importance to the control of calibration errors, combine handheld XRF measurements with necessary laboratory analyses, and establish calibration correlations for specific ores to enable mutual verification and complementarity between on-site and laboratory data. Through continuous improvement of calibration methods and strict quality management, handheld XRF is expected to achieve more precise and stable measurements in iron ore testing, providing strong support for geological prospecting, ore grading, and production monitoring.

Introduction

The Yokogawa AQ6370D series optical spectrum analyzer is a high-performance and multifunctional testing instrument widely used in various fields such as optical communication, laser characteristic analysis, fiber amplifier testing, and WDM system analysis. With its high wavelength accuracy, wide dynamic range, and rich analysis functions, it has become an indispensable tool in research and development as well as production environments.

This article, closely based on the content of the AQ6370D Optical Spectrum Analyzer User’s Manual, systematically introduces the device’s operating procedures, functional modules, usage tips, and precautions. It aims to help users quickly master the device’s usage methods and improve testing efficiency and data reliability.

I. Device Overview and Initial Setup

1.1 Device Structure and Interfaces

The front panel of the AQ6370D is richly laid out, including an LCD display, soft key area, function key area, data input area, optical input interface, and calibration output interface. The rear panel provides various interfaces such as GP-IB, TRIGGER IN/OUT, ANALOG OUT, ETHERNET, and USB, facilitating remote control and external triggering.

Key Interface Descriptions:

• OPTICAL INPUT: This is the optical signal input interface that supports common fiber connectors such as FC/SC.
• CALIBRATION OUTPUT: Only the -L1 model has this built-in reference light source output interface for wavelength calibration.
• USB Interface: Supports external devices such as mice, keyboards, and USB drives for easy operation and data export.

1.2 Installation and Environmental Requirements

To ensure normal operation of the device, the installation environment should meet the following conditions:

• Temperature: Maintain between 5°C and 35°C.
• Humidity: Not exceed 80% RH, and no condensation should occur.
• Environment: Avoid environments with vibrations, direct sunlight, excessive dust, or corrosive gases.
• Space: Provide at least 20 cm of ventilation space around the device.

Note: The device weighs approximately 19 kg. When moving it, ensure two people operate it together and that the power is turned off.

II. Power-On and Initial Calibration

2.1 Power-On Procedure

1. Connect the power cord to the rear panel and plug it into a properly grounded three-prong socket.
2. Turn on the MAIN POWER switch on the rear panel. The POWER indicator on the front panel will turn orange.
3. Press the POWER key to start the device, which will enter the system initialization interface.
4. After initialization, if it is the first use or the device has been subjected to vibrations, the system will prompt for alignment adjustment and wavelength calibration.

2.2 Alignment Adjustment

Alignment adjustment aims to calibrate the optical axis of the built-in monochromator to ensure optimal optical performance.

Using Built-in Light Source (-L1 Model):

1. Connect the CAL OUTPUT and OPTICAL INPUT using a 9.5/125 μm single-mode fiber.
2. Press SYSTEM → OPTICAL ALIGNMENT → EXECUTE.
3. Wait approximately 2 minutes, and the device will automatically complete alignment and wavelength calibration.

Using External Light Source (-L0 Model):

1. Connect an external laser source (1520–1560 nm, ≥-20 dBm) to the optical input port.
2. Enter SYSTEM → OPTICAL ALIGNMENT → EXTERNAL LASER → EXECUTE.

2.3 Wavelength Calibration

Wavelength calibration ensures the accuracy of measurement results.

Using Built-in Light Source:
Enter SYSTEM → WL CALIBRATION → BUILT-IN SOURCE → EXECUTE.

Using External Light Source:
Choose EXECUTE LASER (laser type) or EXECUTE GAS CELL (gas absorption line type) and input the known wavelength value.

Note: The device should be preheated for at least 1 hour before calibration, and the wavelength error should not exceed ±5 nm (built-in) or ±0.5 nm (external).

III. Basic Measurement Operations

3.1 Auto Measurement

Suitable for quick measurements of unknown light sources:

1. Press SWEEP → AUTO, and the device will automatically set the center wavelength, scan width, reference level, and resolution.
2. The measurement range is from 840 nm to 1670 nm.

3.2 Manual Setting of Measurement Conditions

• Center Wavelength/Frequency: Press the CENTER key to directly input a value or use PEAK→CENTER to set the peak as the center.
• Scan Width: Press the SPAN key to set the wavelength range or use Δλ→SPAN for automatic setting.
• Reference Level: Press the LEVEL key to set the vertical axis reference level, supporting PEAK→REF LEVEL for automatic setting.
• Resolution: Press SETUP → RESOLUTION to choose from various resolutions ranging from 0.02 nm to 2 nm.

3.3 Trigger and Sampling Settings

• Sampling Points: The range is from 101 to 50,001 points, settable via SAMPLING POINT.
• Sensitivity: Supports multiple modes such as NORM/HOLD, NORM/AUTO, MID, HIGH1~3 to adapt to different power ranges.
• Average Times: Can be set from 1 to 999 times to improve the signal-to-noise ratio.

IV. Waveform Display and Analysis Functions

4.1 Trace Management

The device supports 7 independent traces (A~G), each of which can be set to the following modes:

• WRITE: Real-time waveform update.
• FIX: Fix the current waveform.
• MAX/MIN HOLD: Record the maximum/minimum values.
• ROLL AVG: Perform rolling averaging.
• CALCULATE: Implement mathematical operations between traces.

4.2 Zoom and Overview

The ZOOM function allows local magnification of the waveform, supporting mouse-drag selection of the area. The OVERVIEW window displays the global waveform and the current zoomed area for easy positioning.

4.3 Marker Function

• Moving Marker: Displays the current wavelength and level values.
• Fixed Marker: Up to 1024 can be set to display the difference from the moving marker.
• Line Marker: L1/L2 are wavelength lines, and L3/L4 are level lines, used to set scan or analysis ranges.
• Advanced Marker: Includes power spectral density markers, integrated power markers, etc., supporting automatic search for peaks/valleys.

4.4 Trace Math

Supports operations such as addition, subtraction, normalization, and curve fitting between traces, suitable for differential measurements, filter characteristic analysis, etc.

Common Calculation Modes:

• C = A – B: Used for differential analysis.
• G = NORM A: Normalize the display.
• G = CRV FIT A: Perform Gaussian/Lorentzian curve fitting.

V. Advanced Measurement Functions

5.1 Pulsed Light Measurement

Supports three modes:

• Peak Hold: Suitable for repetitive pulsed measurements.
• Gate Sampling: Synchronized sampling with an external gate signal.
• External Trigger: Suitable for non-periodic pulsed measurements.

5.2 External Trigger and Synchronization

• SMPL TRIG: Wait for an external trigger for each sampling point.
• SWEEP TRIG: Wait for an external trigger for each scan.
• SMPL ENABLE: Perform scanning when the external signal is low.

5.3 Power Spectral Density Display

Switch to dBm/nm or mW/nm via LEVEL UNIT, suitable for normalized power display of broadband light sources (such as LEDs, ASE).

VI. Data Analysis and Template Judgement

6.1 Spectral Width Analysis

Supports four algorithms:

• THRESH: Threshold method.
• ENVELOPE: Envelope method.
• RMS: Root mean square method.
• PEAK RMS: Peak root mean square method.

6.2 Device Analysis Functions

• DFB-LD SMSR: Measure the side-mode suppression ratio.
• FP-LD/LED Total Power: Calculate the total optical power through integration.
• WDM Analysis: Simultaneously analyze multiple channel wavelengths, levels, and OSNR.
• EDFA Gain and Noise Figure: Calculate based on input/output spectra.

6.3 Template Judgement (Go/No-Go)

Upper and lower limit templates can be set for quick judgement in production lines:

• Upper limit line, lower limit line, target line.
• Supports automatic judgement and output of results.

VII. Data Storage and Export

7.1 Storage Media

Supports USB storage devices for saving waveform data, setting files, screen images, analysis results, etc.

7.2 Data Formats

• CSV: Used to store analysis result tables.
• BMP/PNG: Used to save screen images.
• Internal Format: Supports subsequent import and re-analysis.

7.3 Logging Function (Data Logging)

Can periodically record WDM analysis, peak data, etc., suitable for long-term monitoring and statistical analysis.

VIII. Maintenance and Troubleshooting

8.1 Routine Maintenance

• Regularly clean the fiber end faces and connectors.
• Avoid direct strong light input to prevent damage to optical components.
• Use the original packaging for transportation to avoid vibrations.

8.2 Common Problems and Solutions

IX. Conclusion

The Yokogawa AQ6370D series optical spectrum analyzer is a comprehensive and flexible high-precision testing device. By mastering its basic operations and advanced functions, users can efficiently complete various tasks ranging from simple spectral measurements to complex system analyses. This article, based on the official user manual, systematically organizes the device’s usage procedures and key technical points, hoping to provide practical references for users and further improve testing efficiency and data reliability.