Solution cathode glow discharge-atomic emission spectrometry using automated standard dilution analysis for the determination of Ca, Fe, and Zn in glucose oral solution

Abstract

Solution cathode glow discharge-atomic emission spectrometry (SCGD-AES) demonstrates significant potential as a metal element detection technique. However, its direct measurement of complex matrices still faces challenges in reliability due to matrix effects. To address this issue, standard dilution analysis (SDA) calibration emerges as an innovative strategy capable of effectively overcoming matrix interferences. This study investigated the feasibility of combining SDA with SCGD for practical complex sample analysis by employing automated SDA to collect calibration points over time and establish analytical curves. The results demonstrated that the proposed SCGD-SDA method performed effectively, showing that the relative deviations between the SCGD and inductively coupled plasma optical emission spectrometry (ICP-OES) methods were 5.2–9.4% for Ca, 3.1–7.3% for Fe, and 2.8–7.5% for Zn in five different oral glucose solutions. The detection limits of automated SDA reach ppb levels. SDA compensates for instrumental fluctuations and matrix effects, providing improved accuracy compared to external standard calibration (EC) while maintaining good agreement with ICP-OES reference values. The SCGD-SDA method establishes a simple and rapid analytical approach for trace element determination in glucose oral solutions.

---

1 Introduction

Glucose oral solution is a commonly used clinical energy supplement whose trace element content directly affects therapeutic efficacy and medication safety. Trace elements such as calcium (Ca), iron (Fe), and zinc (Zn) in the solution act as coenzyme factors and antioxidants, playing vital roles in key physiological processes including energy metabolism and immune regulation.^1–3 It is shown that controlling the concentrations of these trace elements presents a dual challenge: insufficient levels may lead to deficiency-related symptoms (e.g., iron deficiency causing anemia and zinc deficiency impairing immune function), while excessive levels could result in metal accumulation toxicity.^4–7 Notably, calcium may also inhibit iron absorption through short-term effects, further complicating formulation design.⁵ This “dose window” effect highlights the critical need for reliable trace element detection methods. Accurate quantification of trace elements in glucose oral solutions is essential not only for medication safety evaluation but also for guiding precise clinical nutritional support. Therefore, developing rapid and accurate detection methods holds significant clinical application value.

In recent years, there have been significant advancements in detection technologies for oral solution components. Liquid chromatography-mass spectrometry (LC-MS) has been widely employed for qualitative and quantitative analysis of phenolic compounds in rose oral solutions and chemical constituents in Shuanghuanglian oral solution (a common Chinese herbal medicine for cold).^8,9 Pinheiro et al. utilized ICP-OES for simultaneous determination of Cd, Co, Hg, Ni, Pb, and V in oral and parenteral drug samples.³⁷ Although capable of multi-element analysis, this method suffers from bulky instrumentation and high operational costs. In contrast, glow discharge using liquid as one or two electrodes for atomic emission spectrometry demonstrates promising potential with distinctive advantages. This technique was originally proposed by Cserfalvi et al. in 1993 and later refined by Webb et al. as solution cathode glow discharge-atomic emission spectrometry (SCGD-AES) through simplified discharge cell design for low-flow solution introduction. Compared with the traditional spectral technique, SCGD-AES eliminates the need for inert gases while enabling multi-element simultaneous detection and real-time monitoring.^35,36 This technique has been successfully applied to diverse real-world samples, including liquid matrices such as fruit juices,¹² honey,¹³ and high-salinity brines,¹⁴ as well as solid samples such as ores and biological tissues,^15–22 which presents its practical value and broad prospects in elemental analysis.

Although SCGD has made significant progress in practical sample analysis, to our knowledge, its application in medical samples for metal determination still suffers from their complex matrices. Jie Yu et al. employed SCGD-AES to measure calcium and zinc in four calcium-zinc oral solutions and three blood samples, where digestion was required to eliminate organic interference.¹⁰ However, this approach proved time-consuming, labor-intensive, and involved excessive consumption of high-purity nitric acid while generating toxic gases. Peichao Zheng et al. attempted direct dilution analysis of Ca, Fe, and Zn in gluconate oral solutions using SCGD with external calibration, yet the method still required tedious preparation of individual element gradient solutions for calibration curves and suffered from matrix-induced deviations.¹¹ To mitigate matrix effects (e.g., from fructose and glucose), nonionic surfactant (Triton X-405) was introduced, which enhanced the signals of target elements.¹³ These challenges have prompted continuous exploration of novel calibration strategies to address matrix effects effectively.

Traditional quantitative calibration methods, such as the EC method, internal standard (IS) method, and standard addition (SA) method, are widely used in quantitative analysis. However, the EC method relies on pure standards to establish a calibration curve, and the measurement errors are strongly affected by the matrix effects of complex actual samples (e.g., medical or environmental samples). The IS method reduces signal fluctuations by introducing an internal standard, but it is difficult to find an internal standard with matching physicochemical properties, and it may still be subject to matrix interference. The SA method offsets matrix effects by spiking samples stepwise; however, each sample requires individual calibration, which is time-consuming and resource-intensive.

To address the limitations of conventional calibration methods, SDA was developed and first reported by Jones et al. in 2015 (ref. 23) as an innovative calibration strategy, which creatively combines the technical advantages of IS and SA through a unique dual-solution system. This method employs two specifically designed calibration solutions: the first containing an equal-volume mixture of the sample and standard solution with analytes and internal standards, and the second comprising an equal-volume mixture of the sample and blank matrix.^23,24 By gradient mixing these two solutions, SDA achieves continuous dilution of both analytes and internal standards while maintaining constant matrix conditions. This design not only effectively corrects for matrix effects and instrumental fluctuations but also reduces required calibration solutions to just two, significantly improving analytical efficiency.^25,26 Honestly, the implementation of automation in SDA necessitates the incorporation of control hardware such as high-pressure valves, which increases the overall cost. Nevertheless, its fundamental advantage lies in eliminating the substantial manual effort required for the preparation of multiple calibration solutions, thereby enhancing overall analytical efficiency and reducing human errors. Comprehensive methodological studies on SDA have systematically explored critical parameters including analyte-to-internal standard ratios, sample-to-standard mixing proportions, platform stabilization time, calibration interval selection criteria, and automated implementation approaches.^27–29

To date, SDA has been successfully implemented in various analytical techniques including Flame-Atomic Emission Spectrometry (F-AES),³⁰ Flame Atomic Absorption Spectrometry (FAAS),³¹ Microwave-Induced Plasma Optical Emission Spectrometry (MIP-OES),³² Inductively Coupled Plasma-Optical Emission Spectrometry (ICP-OES),^24,26–29 Inductively Coupled Plasma-Mass Spectrometry (ICP-MS),²⁵ and Raman Spectroscopy(RS),³³ and presents the ability for analysis of complex samples.

Here, a novel method of SCGD-AES combined with SDA was developed to determine Ca, Fe, and Zn in glucose oral solutions. An automated SDA system was implemented using relay control boards, high-pressure valves, and Y-shaped tubing to minimize operational errors from manual mixing. The method employed Cs and Rb as dual internal standards to establish calibration curves through dynamic gradient dilution. Detection limits and quantification limits were determined for all target elements, while spike recovery tests confirmed minimal systematic error. Comparative analysis with ICP demonstrated comparable results for Ca, Fe, and Zn quantification in glucose oral solutions, validating the method's accuracy and precision. The combination of SCGD and SDA provides a novel strategy for trace element analysis in complex-matrix samples.

2 Experimental

2.1 Experimental setup

The integrated SCGD-AES system coupled with automated SDA, as illustrated in Fig. 1, was independently developed in our laboratory. The SDA automation was achieved using a self-designed relay control board, high-pressure valve (Runze Fluid Mrv-01B), peristaltic pump (Baoding Longer, BT100-2J, China) and Y-shaped connector. The relay control board consists of a microcontroller and relays. S₁ denotes the standard solution, while S₂ represents the blank solution. The COM port (marked as C in Fig. 1) of the high-pressure valve switches between S₁ and S₂ at 90-second intervals, driven by the relay. The switching timing is controlled by the microcontroller through relay activation/deactivation. The standard solution undergoes stepwise dilution from maximum concentration to blank level in an alternating fashion. The first channel of the peristaltic pump generates the standard-plus-blank mixture. The high-pressure valve's ON port aspirates the prepared standard solution, while the OFF port aspirates the prepared blank solution. By default, the ON port remains connected to the COM port. When the relay triggers a switching signal, the ON-COM connection breaks while the OFF port connects to COM instead. The sample solution is aspirated through the peristaltic pump's second channel. Both channels converge via a Y-connector into the SCGD cell.

Fig. 1 Schematic diagram of the experimental setup.

The SCGD-AES component is similar to our previous work.^11,38 The analyte solution serves as the cathode, emerging from the top of a capillary glass tube, while a tungsten rod anode (2.4 mm diameter) is positioned above the capillary. Waste solution is evacuated through a disposal tube (2 mm ID and 4 mm OD) via the third channel of a peristaltic pump into a collection vessel. A DC high-voltage power supply (Dongwen High Voltage, China; Output: 0–1500VDC/100 mA) was connected between the solution cathode and the tungsten anode through a resistor to generate a stable microplasma.

The emitted photon flux was collimated and focused by a pair of plano-convex quartz lenses (f = 50 mm, DaHeng Optics, China) in a 1 : 1 imaging configuration, and transmitted to the entrance slit of a compact spectrometer (AvaSpec-Mini4096CL) via an optical fiber. The spectrometer continuously acquired data with an integration time of 400 ms and 5 accumulations. An industrial computer recorded the intensity–time profiles, and the data were processed using Origin software based on eqn (1) to calculate concentrations.

2.2 Solution preparation

The blank solutions were prepared using nitric acid (trace grade) and deionized water (18.25 MΩ cm⁻³) from an ultrapure water system (UIPR–II–10T, Ulupure, China). Single-element standard solutions (1000 mg L⁻¹) of Cs, Rb, Ca, Fe, and Zn were supplied by China North Weiye Measurement Group Co., Ltd (China), with Cs and Rb serving as internal standards.

Standard Solution 1 (S₁) was prepared by diluting the 1000 mg L⁻¹ stock solutions in 1% v/v HNO₃ to contain 5 mg L⁻¹ Ca, 2 mg L⁻¹ each of Fe and Zn, and 5 mg L⁻¹ Cs. Blank Solution 2 (S₂) contained 1 mg L⁻¹ Rb as the internal standard. Sample Solution 3 (S₃) was prepared with 2 mg L⁻¹ Ca and 1 mg L⁻¹ each of Fe and Zn. Solutions 1–3 were used for fundamental SDA validation experiments. For standard curve calibration and limit of detection measurements, gradient solutions were prepared containing 0.5, 5.0, 10.0, 50.0, and 150.0 mg L⁻¹ for Ca, 0.5, 1.0, 2.0, 3.0, and 5.0 mg L⁻¹ for Fe, and 0.5, 1.0, 2.0, 3.0, and 5.0 mg L⁻¹ for Zn, respectively. Solution 4 (S₁) was prepared containing 10 mg L⁻¹ each of Ca, Fe, and Zn with 5 mg L⁻¹ Cs as the internal standard. Solution 5 (S₁) was prepared with 1 mg L⁻¹ each of Ca, Fe, and Zn, also containing 5 mg L⁻¹ Cs. Solution 6 consisted of 1% v/v HNO₃ blank solution. Solutions 2, 4, 5, and 6 were employed for the determination of SDA detection limits.

Five commercially available brands of gluconate oral solutions were purchased from pharmacies and designated as Oral Solution A through E. Samples A and B are composite calcium–iron–zinc supplements, while C, D, and E are single-element supplements, specifically containing Ca, Fe, and Zn, respectively. They were selected to validate the method's applicability across samples with varying compositional profiles. Each oral solution was diluted 100-fold and 200-fold with 1% (v/v) HNO₃, with five replicate preparations made for each dilution level to serve as samples for SDA, EC, and ICP-OES measurements. Additionally, a separate series of Oral Solution A samples was prepared with dilution factors ranging from 100- to 200-fold at 20-fold intervals.

2.3 SDA theory

SDA requires three primary solutions: S₁ (standard containing analytes and primary internal standard IS₁), S₂ (blank solution with secondary internal standard IS₂), and S₃ (sample solution). The experimental mixing system utilizes two key solutions—S_a (1 : 1 mixture of S₁ and S₃) and S_b (1 : 1 mixture of S₂ and S₃)—prepared automatically. As described in recent studies,²⁸ the analyte concentration is given by eqn (1). A linear response curve was established between the analyte signal and the primary internal standard IS₁ (Cs) to derive Slope₁ and Intercept₁, while Intercept₂ was determined from the linear correlation between IS₁ (Cs) and secondary internal standard IS₂ (Rb) signals. C_a^sam denotes the analyte concentration in the sample, while C_a^std represents the analyte concentration in standard solution S₁.

The identical matrix composition of solutions S_a and S_b ensures constant matrix matching throughout the dilution process. When mixing S_a and S_b (containing the target analyte and both internal standards, IS₁: Cs and IS₂: Rb), the system automatically generates continuous calibration points via dynamic dilution. During this process, the instrument monitors in real-time the response signals of the analyte (Ca) and two internal standards (Cs and Rb). The data points from either the rising or falling segment of each complete cycle were linearly fitted (Fig. 2). The resulting intercept and slope were then substituted into eqn (1) to calculate the concentration.^25,27–29Fig. 2 clearly demonstrates that the signal intensities of both the analyte (Ca) and IS₁ (Cs) progressively decrease with increasing IS₂ (Rb) concentration, reflecting gradual dilution of the standard by the blank matrix. Simultaneously, the sample analyte intensity shows synchronous variation with IS₁ (Cs) signals, confirming the linear dynamic characteristics of the dilution process. This real-time signal tracking strategy eliminates the need for preparing traditional discrete calibration points, significantly improving both calibration efficiency and reliability.

Fig. 2 Calibration curves of analyte-IS₁ and IS₁-IS₂, showing representative plots of the analyte-IS₁ calibration relationship. Here, Cs serves as IS₁, Rb as IS₂, and Ca as the target analyte.

2.4 Automation of standard dilution analysis

In the automated SDA system, a four-channel peristaltic pump was employed to pump the solutions, integrated with a relay control board, high-pressure valve, and Y-connector to form a complete fluidic control system. This technology's innovation lies in its ability to alternately mix sample solution with standard and blank solutions, thereby generating multiple calibration points for curve construction. The workflow operates as follows: The first channel of the peristaltic pump aspirates either standard or blank solution, with the relay board precisely controlling the high-pressure valves switching timing to achieve online gradient dilution from maximum concentration to blank. The second channel delivers the sample solution at a constant flow rate, with both streams undergoing dynamic mixing at the Y-connector. The flow rate of the peristaltic pump was maintained at 1.3 mL min⁻¹ per channel—excessive rates would increase mixing curve steepness and reduce usable calibration points. The high-pressure valve's design prevents bubble formation during solution switching, ensuring fluid continuity. Optimal switching duration is critical: insufficient time causes inadequate diffusion, while excessive duration prolongs measurement cycles and reduces efficiency. During SDA measurements, the spectrometer's integration time and signal averaging parameters are particularly critical. These settings directly determine the temporal resolution between successive calibration points while ensuring reliable signal acquisition throughout the blank-mediated standard solution dilution process. All instrument parameters employed in this study are detailed in Table 1. The analytes and internal standard species with their respective emission wavelengths were Cs (852.06 nm), Rb (780.21 nm), Ca (422.63 nm), Fe (248.30 nm), and Zn (213.79 nm).

Table 1 Instrumental parameters employed in this study

Instrument parameter	Operating conditions
Current of the SCGD source	75 mA
Flow rate of the samples	2.6 mL min^−1
Inter-electrode distance	2.4 mm
Integration time of the spectrometer	400 ms
Accumulation times of the spectrometer	5
Relay switching cycle	90 s

---

3 Results and discussion

3.1 Optimization of experimental parameters

The discharge current is a critical parameter affecting the excitation efficiency in SCGD. Experimental results demonstrate that current optimization must balance excitation efficiency with system stability. Insufficient current leads to inadequate plasma energy, significantly reducing the emission intensities of target elements (Cs, Rb, Ca, Fe, and Zn). Conversely, excessive current induces localized overheating, manifested by anode tungsten rod reddening, which accelerates electrode degradation and may trigger discharge interruptions.

As shown in Fig. 3, systematic evaluation across 50–80 mA (5 mA increments) revealed distinct trends: the emission intensities of Ca, Fe, and Zn increased linearly within 50–75 mA, followed by an anomalous surge at 80 mA. In contrast, Cs and Rb intensities peaked at 75 mA before declining. Notably, discharge stability deteriorated markedly at 80 mA, whereas 75 mA provided optimal performance with relative standard deviations (RSDs) below 1% for all elements. Based on combined intensity and RSD metrics, 75 mA was identified as the optimized operational current for the SCGD system.

Fig. 3 Effect of discharge current on excitation intensity and plasma stability.

The solution flow rate was systematically investigated as a critical parameter influencing plasma performance under optimized 75 mA discharge current. Experimental results demonstrate the necessity to balance competing effects: excessive flow rates induce strong cooling that reduces plasma temperature and excitation efficiency, while insufficient flow causes solution depletion at the anode interface, compromising discharge stability.

As shown in Fig. 4, comprehensive evaluation across 1.9–3.0 mL min⁻¹ revealed optimal performance at 2.6 mL min⁻¹, where all elements (Cs, Rb, Ca, Fe, and Zn) achieved maximum emission intensities with excellent reproducibility (RSD <1.2%). This flow rate effectively prevents the discharge interruptions observed at lower rates (1.9 mL min⁻¹) while avoiding the excitation suppression caused by excessive cooling at higher flows. Therefore, 2.6 mL min⁻¹ was established as the optimal flow rate of samples for subsequent experiments.

Fig. 4 Effect of peristaltic pump flow rate on excitation intensity and plasma stability.

3.2 Basic examples of SDA

In the analytical method, Cs and Rb were employed as the first and second internal standards, respectively, with Cs present in the first standard solution and Rb in the blank solution. The standard solution also contained Ca (5 mg L⁻¹), Fe (2 mg L⁻¹), and Zn (2 mg L⁻¹), while the sample solution was prepared with Ca (2 mg L⁻¹), Fe, and Zn at 1 mg L⁻¹ each. Fig. 5 displays the temporal spectral intensity profiles of the two internal standards and three analytes.

Fig. 5 Time-resolved intensity profiles of three analytes (Ca, Fe, and Zn) and two internal standards (Cs and Rb). The black curve represents the first internal standard present in the standard solution, while the red curve corresponds to the second internal standard in the blank solution. The blue, green, and purple curves depict the measured elements existing in both standard and sample solutions.

The first internal standard, Cs, exhibited a consistent trend with the target elements (Ca, Fe, and Zn), whereas the second internal standard, Rb, showed an inverse correlation, decreasing as the Cs signal intensity increased. The high-pressure valve switched channels every 90 s, causing the signal intensity trajectory to transition from maximum to minimum (or vice versa) at 90 s intervals. Notably, two valve switches were required to complete one full cycle of signal variation. Each transition from minimum to maximum (or the reverse) generated multiple calibration points for calculating the sample solution concentration. When the solution passed through the high-pressure valve and peristaltic pump into the Y-tube, dilution of the standard by the blank occurred via diffusion, resulting in a continuous sloping transition rather than abrupt jumps in the signal waveform, as illustrated in Fig. 5.

Fig. 6 demonstrates representative response curves required for SDA calculations, including analyte-to-primary internal standard ratios and first-to-second internal standard ratios (bottom right). These four plots, combined with the known concentrations of three analytes in the standard solution, constitute all necessary components for calculating analyte concentrations in samples. The SDA calculations were performed using eqn (1).

Fig. 6 Parameters required for analyte concentration calculation via SDA, including the intercept and slope of calibration curves between analytes and the primary internal standard (Cs), as well as the y-axis intercept between the primary (Cs) and secondary (Rb) internal standards.

Table 2 presents the results of six replicate measurements of a standard sample solution for Ca, Fe, and Zn. The results showed concentrations of 2.08 mg L⁻¹ (Ca), 1.07 mg L⁻¹ (Fe), and 0.97 mg L⁻¹ (Zn), with less than 10% error compared to the true values.

Table 2 Concentration measurements of standard sample solutions

Element	Reference (mg L^−1)	Average (n = 6) (mg L^−1)	Relative error (%)	RSD (%)
Ca	2.00	2.08 ± 0.02	4	3.1
Fe	1.00	1.07 ± 0.02	7	2.0
Zn	1.00	0.97 ± 0.01	3	2.4

---

3.3 Limits of detection and quantification

3.3.1 Detection and quantification limits by the external calibration (EC) method. According to IUPAC recommendations,³⁴ the limit of detection (LOD) for EC was calculated as three times the standard deviation of ten blank measurements (S_blank, n = 10) divided by the calibration curve slope (k), expressed as LOD = 3 × S_blank/k. Similarly, the limit of quantification (LOQ) was determined as LOQ = 10 × S_blank/k. The blank solution standard deviation was obtained by measuring an impurity-free nitric acid solution at pH 1.0. Calibration curves were established by linear regression of spectral intensities from standard solutions with varying concentrations of Zn (0.5–5.0 mg L⁻¹), Fe (0.5–5.0 mg L⁻¹), and Ca (0.5–150.0 mg L⁻¹). The derived parameters and calculated results are summarized in Table 3. The LODs for Ca, Fe, and Zn were 0.037, 0.138, and 0.031 mg L⁻¹, respectively, with correlation coefficients (R²) of 0.999, 0.999, and 0.993.

Table 3 Detection limit and quantitation limit by the standard curve calibration method (EC)

Element	Linear range	Calibration equation	R ^2	RSD (%)	LOD (mg L^−1)	LOQ (mg L^−1)
Ca	0.5–150.0 mg L^−1	y = 677.75x + 1055.97	0.999	0.73	0.037	0.123
Fe	0.5–5.0 mg L^−1	y = 289.92x + 3213.39	0.999	0.6	0.138	0.46
Zn	0.5–5.0 mg L^−1	y = 1356.0x + 3472.34	0.993	1.14	0.031	0.103

---

3.3.2 Detection limit and quantitation limit of SDA. To ascertain the detection limit for the element of interest, a procedural blank was subjected to the standard dilution analysis (SDA) procedure as if it were an analytical sample. The LOD and LOQ of SDA were calculated as 3 times (for LOD) and 10 times (for LOQ) the standard deviation of the analyte concentration in blanks (n = 10), i.e., LOD = 3 × S_blank and LOQ = 10 × S_blank. The standards used for LOD determination included Solutions 2, 4, 5, and 6 as previously discussed. Solutions 4 and 5 contained three analytes along with Cs as the internal standard, with analyte concentrations of 10 mg L⁻¹ (Solution 4) and 1 mg L⁻¹ (Solution 5); both concentrations were employed for LOD calculation. Solution 2 contained the internal standard Rb, while Solution 6 was a blank background matrix (1% v/v HNO₃). The calculated LOD and LOQ values for each element are summarized in Table 4.

Table 4 Detection limits and quantification limits of SDA. The detection limits for Ca, Fe, and Zn were determined by the SDA method using a blank background solution as the sample, with the LOD defined as three times the standard deviation of 10 replicate measurements

Element	1 mg L^−1 standard		10 mg L^−1 standard
	LOD	LOQ	LOD	LOQ
Ca	0.07	0.23	0.84	2.80
Fe	0.12	0.40	1.05	3.5
Zn	0.05	0.17	0.42	1.40

---

The results showed that the LODs of the tested elements were generally similar, with Fe being slightly higher compared to EC. When using the 1 mg L⁻¹ standard, the SDA detection limits ranged from 0.05 to 0.12 mg L⁻¹. In contrast, when using the 10 mg L⁻¹ standard, the detection limits ranged from 0.42 to 1.05 mg L⁻¹, differing by an order of magnitude. As described by Sloop et al.,²⁴ the LOD varies when using different standard concentrations because higher concentrations increase measurement uncertainty near the zero-concentration region. Since the only variable parameter in these two SDA measurements was the standard concentration, the calibration points near the detection limit under low-concentration conditions were less affected by interference signals. Consequently, using a lower standard concentration provided more low-concentration data points during calibration, leading to a slightly lower measured LOD.

A comparison of Tables 3 and 4 reveals that at low standard concentrations, the detection limits of SDA and EC fall within the same order of magnitude. However, when using higher-concentration standard solutions, the SDA detection limits become significantly higher than those of EC. As discussed earlier, deviations near the zero point may introduce additional errors into the measurement results through their influence on the estimated slope and intercept of the calibration curve, potentially leading to increased uncertainty.

3.4 Automated SDA measurement of oral liquid samples

Five commercially available gluconate oral solutions, designated as Oral Solution A through E, were selected for analysis. Solutions A and B contained Ca, Fe, and Zn, while Solutions C, D, and E contained individual elements of Ca, Fe, and Zn, respectively. All samples were diluted 100- and 200-fold with 1% (v/v) HNO₃ prior to analysis. The elemental contents of Ca, Fe, and Zn were determined using three analytical techniques (ICP, EC, and SDA), with each sample measured five times. The comparative results obtained from these different measurement approaches are presented in Table 5.

Table 5 Analysis of Oral Solutions A, B, C, D, and E using ICP, EC, and SDA methods following 100-fold and 200-fold dilution

Sample	Ca			Fe			Zn
	ICP-OES mg L^−1	SCGD-AES mg L^−1		ICP-OES mg L^−1	SCGD-AES mg L^−1		ICP-OES mg L^−1	SCGD-AES mg L^−1
	±SD	EC	SDA		EC	SDA		EC	SDA
A:100-fold dil	80 ± 0.18	72.4 ± 0.72	73.1 ± 0.21	1.71 ± 0.01	1.93 ± 0.03	1.80 ± 0.02	1.44 ± 0.01	1.38 ± 0.06	1.48 ± 0.04
		9.5%(RE)	8.6%(RE)		12.9%	5.3%		4.2%	2.8%
A:200-fold dil	39.1 ± 0.16	37.5 ± 0.65	41.6 ± 0.23	0.82 ± 0.01	0.90 ± 0.02	0.88 ± 0.02	0.67 ± 0.01	0.74 ± 0.02	0.72 ± 0.02
		4.1%	6.4%		9.8%	7.3%		10.4%	7.5%
B:100-fold dil	130 ± 0.31	124 ± 0.79	119 ± 0.67	2.86 ± 0.02	2.59 ± 0.04	2.77 ± 0.03	2.35 ± 0.02	2.16 ± 0.04	2.26 ± 0.03
		4.6%	8.5%		9.4%	3.1%		8.1%	3.8%
B:200-fold dil	66 ± 0.29	64.9 ± 0.68	62.3 ± 0.75	1.45 ± 0.01	1.57 ± 0.02	1.54 ± 0.02	1.19 ± 0.01	1.10 ± 0.03	1.11 ± 0.02
		1.7%	5.6%		8.3%	6.2%		7.6%	6.7%
C:100-fold dil	98.5 ± 0.23	92.9 ± 0.95	89.2 ± 0.99	—	—	—	—	—	—
		5.7%	9.4%
C:200-fold dil	50.2 ± 0.18	48.7 ± 0.89	47.6 ± 0.77	—	—	—	—	—	—
		3.0%	5.2%
D:100-fold dil	—	—	—	4.5 ± 0.01	4.08 ± 0.04	4.23 ± 0.03	—	—	—
					9.3%	6.0%
D:200-fold dil	—	—	—	2.29 ± 0.01	2.09 ± 0.03	2.17 ± 0.02	—	—	—
					8.7%	5.2%
E:100-fold dil	—	—	—	—	—	—	5.08 ± 0.01	4.67 ± 0.04	4.91 ± 0.03
								8.1%	3.3%
E:200-fold dil	—	—	—	—	—	—	2.62 ± 0.01	2.44 ± 0.03	2.51 ± 0.02
								6.9%	4.2%

---

Based on the determination results obtained by ICP-OES, Table 5 presents the relative errors of the analytical results derived from both the external calibration (EC) and standard dilution analysis (SDA) methods. The results demonstrate that SDA exhibited relative errors of 5.2–9.4%, 3.1–7.3%, and 2.8–7.5% for Ca, Fe, and Zn measurements respectively, while EC showed relative errors of 1.7–9.5%, 8.3–12.9%, and 4.2–10.4% for the same elements. The experimental data indicate improved accuracy of SDA over EC. For SDA, the error rates remained comparable between 100-fold and 200-fold dilutions, whereas EC demonstrated significantly higher errors at 100-fold dilution compared to 200-fold dilution. This discrepancy may be attributed to organic components in the gluconate oral solutions, where higher dilution factors potentially reduce matrix effects.

Fig. 7 presents the relationship between spectral intensity and dilution factor for each element. The linearity is affected by matrix effects. However, SDA employs online mixing of multiple elements to accurately correct for these matrix effects, resulting in measurements with smaller relative errors. An exception occurs in the measurement of high-concentration calcium, where SDA does not demonstrate improved performance. This may be attributed to the significant concentration disparity between the prepared standard solution (10 mg L⁻¹ Ca) and the oral solution (up to 130 mg L⁻¹ Ca), leading to reduced accuracy. It should be noted that excessively high concentrations of target elements in standard solutions should be avoided, as discussed previously, since higher concentrations introduce greater interference signals that adversely affect detection limits. Furthermore, for CCD array detectors, extremely high concentrations of Ca may also cause signal saturation and the ‘blooming’ effect, which could potentially impact the accurate measurement of other elements.

Fig. 7 Spectral intensities of Ca, Fe, and Zn in Oral Solution B measured across 100- to 200-fold dilution range at 20-fold intervals.

The primary objective of this work was to develop a reliable method for measuring Ca, Fe, and Zn in oral solutions by coupling SCGD with automated SDA technology. Consequently, comprehensive comparisons with traditional calibration approaches (e.g., internal standardization [IS] and standard addition [SA]) were not performed. Previous studies have demonstrated that SDA achieves comparable or superior performance to conventional SA methods while outperforming external calibration and IS techniques.^25–27 As anticipated, our results confirmed SA's improved accuracy over EC. By combining the advantages of both IS and SA while incorporating numerous calibration points to compensate for signal fluctuations and matrix effects, SDA delivered enhanced measurement accuracy.

To evaluate the accuracy of the SCGD-SDA method, a t-test was performed to compare the results obtained by SCGD with those from the comparative ICP-OES method. The analysis indicated that there was no significant difference between the results of the SCGD-SDA method and the ICP-OES results (p = 0.161 > 0.05). In contrast, a significant difference was observed between the SCGD-EC results and the comparative method (p = 0.003 < 0.05). This statistical outcome further confirms that the SDA calibration strategy effectively overcomes complex matrix interferences, enabling the SCGD-AES technique to achieve accuracy comparable to that of the standard ICP-OES method, thus making it suitable for reliable analysis of practical samples such as glucose oral solutions.

3.5 SDA accuracy

Since the initial calcium concentration in the oral solutions led to saturation of the signal in the spectrometer, all spike recovery tests were performed after appropriate dilution of the samples. Prior to analyzing the spiked sample matrices, the five test glucose oral solutions, diluted 100-fold, were first subjected to baseline determination of the unspiked matrices to quantify the intrinsic analyte content (the samples were identical to those used in the automated standard dilution analysis). Subsequently, Ca, Fe, and Zn (5 mg L⁻¹ each) were added to the solutions. For accurate recovery calculations, the background concentrations were subtracted from the spiked sample results. As shown in Table 6, the analysis of five oral solution matrices using both SDA and EC methods (with quintuplicate measurements per sample) demonstrated recoveries of 85.4–105.3% (RSD: 1.6–3.1%) for EC and 87.4–102.9% (RSD: 1.4–2.6%) for SDA. Comparative results indicated improved performance of SDA in both recovery rates and precision.

Table 6 Recovery rates and precision percentages obtained from analysis of five spiked oral solution matrices. Average recovery percentages and % RSD values were determined from five independent measurements (n = 5) for each sample

Analytes	Sample (% recovery ± % RSD for n = 5)
	Ca		Fe		Zn
	EC	SDA	EC	SDA	EC	SDA
Oral A:Dil 100	88.5 ± 2.9	89.7 ± 1.4	87.6 ± 2.6	91.2 ± 2.2	105.3 ± 1.6	101.3 ± 1.5
Oral B:Dil 100	85.4 ± 3.1	88.9 ± 2.2	89.2 ± 3.0	92.2 ± 2.0	103.9 ± 1.9	98.2 ± 1.7
Oral C:Dil 100	86.8 ± 2.4	87.4 ± 1.9	—	—	—	—
Oral D:Dil100	—	—	91.3 ± 2.8	94.3 ± 2.6	—
Oral E:Dil 100	—	—	—	—	98.7 ± 1.8	102.9 ± 2.3

---

4 Conclusions

This study developed a novel analytical method coupling SCGD-AES with SDA. The innovative integration of relay control boards, high-pressure valves, and Y-shaped tubing achieved fully automated SDA control. Using Cs and Rb as a dual internal standard system for calibration curve construction enabled simultaneous correction of both sample matrix effects and instrumental fluctuations from peristaltic pumps, high-voltage power supplies, and spectrometers during the rapid determination of Ca, Fe, and Zn in glucose-based oral solutions. The method demonstrated detection limits of 0.07 mg L⁻¹, 0.12 mg L⁻¹, and 0.05 mg L⁻¹ for Ca, Fe, and Zn, respectively. Spike recovery tests across five commercial oral solutions yielded average recoveries of 87–103% with relative standard deviations (RSDs) of 1.4–2.6%. Compared with ICP-OES methods, the relative measurement errors were 5.2–9.4% for Ca, 3.1–7.3% for Fe, and 2.8–7.5% for Zn, showing excellent agreement with ICP-AES results and significant improvement over conventional external calibration (EC). The SCGD-SDA technique offers distinct advantages including operational simplicity, rapid analysis, cost-effectiveness, elimination of inert gas requirements, providing a new strategy for SCGD analysis of complex matrix samples.

Conflicts of interest

There are no conflicts of interest to declare.

Data availability

All data supporting the findings of this study are included within the manuscript.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (62575044 and 32171627), Distinguished Young Scientists Program of Beijing Natural Science Foundation (JQ19023), Science and Technology Research Program of Chongqing Municipal Education Commission (KJZD-M202200602), Natural Science Foundation of Chongqing (CSTB2024NSCQ-LZX0078) and Chongqing Entrepreneurship and Innovation Support Program for Overseas Students Returning to China (RC2022-56).

References

1. P. Singh and S. Prasad, A review on iron, zinc and calcium biological significance and factors affecting their absorption and bioavailability, J. Food Compos. Anal., 2023, 123, 105529,  DOI:10.1016/j.jfca.2023.105529.
2. F. S. Al-Fartusie and S. N. Mohssan, Essential trace elements and their vital roles in human body, Indian J Adv Chem Sci, 2017, 5(3), 127–136,  DOI:10.22607/IJACS.2017.503003.
3. N. Abbaspour, R. Hurrell and R. Kelishadi, Review on iron and its importance for human health, J. Res. Med. Sci., 2014, 19(2), 164.
4. N. Roohani, R. Hurrell and R. Kelishadi, et al., Zinc and its importance for human health: An integrative review, J. Res. Med. Sci., 2013, 18(2), 144.
5. L. M. Plum, L. Rink and H. Haase, The essential toxin: impact of zinc on human health, Int. Res. J. Publ. Environ. Health, 2010, 7(4), 1342–1365,  DOI:10.3390/ijerph7041342.
6. B. Lönnerdal, Calcium and iron absorption—mechanisms and public health relevance, Int. J. Vitam. Nutr. Res., 2010, 80(4), 293,  DOI:10.1024/0300-9831/a000036.
7. S. J. Fairweather-Tait, Iron–zinc and calcium–Fe interactions in relation to Zn and Fe absorption, Proc. Nutr. Soc., 1995, 54(2), 465–473,  DOI:10.1079/PNS19950015.
8. D. Nijat, R. Abdulla and G. Liu, et al., Identification and quantification of Meiguihua oral solution using liquid chromatography combined with hybrid quadrupole-orbitrap and triple quadrupole mass spectrometers, J. Chromatogr. B, 2020, 1139, 121992,  DOI:10.1016/j.jchromb.2020.121992.
9. F. X. Zhang, M. Li and Z. H. Yao, et al., A target and nontarget strategy for identification or characterization of the chemical ingredients in Chinese herb preparation Shuang-Huang-Lian oral liquid by ultra-performance liquid chromatography–quadrupole time-of-flight mass spectrometry, Biomed. Chromatogr., 2018, 32(3), e4110,  DOI:10.1002/bmc.4110.
10. J. Yu, X. Zhang and Q. Lu, et al., Determination of calcium and zinc in gluconates oral solution and blood samples by liquid cathode glow discharge-atomic emission spectrometry, Talanta, 2017, 175, 150–157,  DOI:10.1016/j.talanta.2017.07.040.
11. P. Zheng, N. Yan and J. Wang, et al., Rapid Determination of Calcium, Iron, and Zinc in Gluconate Oral Solutions by Solution Cathode Glow Discharge–Atomic Emission Spectroscopy (SCGD-AES), Anal. Lett., 2025, 58(2), 329–340,  DOI:10.1080/00032719.2024.2324140.
12. M. Gorska and P. Pohl, Simplified and rapid determination of Ca, K, Mg, and Na in fruit juices by flowing liquid cathode atmospheric glow discharge optical emission spectrometry, J. Anal. At. Spectrom., 2021, 36(7), 1455–1465,  10.1039/D1JA00127B.
13. K. Greda, P. Jamroz and A. Dzimitrowicz, et al., Direct elemental analysis of honeys by atmospheric pressure glow discharge generated in contact with a flowing liquid cathode, J. Anal. At. Spectrom., 2015, 30(1), 154–161,  10.1039/C4JA00261J.
14. C. Yang, L. Wang and Z. Zhu, et al., Evaluation of flow injection-solution cathode glow discharge-atomic emission spectrometry for the determination of major elements in brines, Talanta, 2016, 155, 314–320,  DOI:10.1016/j.talanta.2016.04.060.
15. J. Yu, S. Yang and Q. Lu, et al., Evaluation of liquid cathode glow discharge-atomic emission spectrometry for determination of copper and lead in ores samples, Talanta, 2017, 164, 216–221,  DOI:10.1016/j.talanta.2016.11.015.
16. Z. Zhang, Z. Wang and Q. Li, et al., Determination of trace heavy metals in environmental and biological samples by solution cathode glow discharge-atomic emission spectrometry and addition of ionic surfactants for improved sensitivity, Talanta, 2014, 119, 613–619,  DOI:10.1016/j.talanta.2013.11.010.
17. K. Gręda, P. Jamróz and P. Pohl, The improvement of the analytical performance of direct current atmospheric pressure glow discharge generated in contact with the small-sized liquid cathode after the addition of non-ionic surfactants to electrolyte solutions, Talanta, 2013, 108, 74–82,  DOI:10.1016/j.talanta.2013.02.049.
18. R. Shekhar, Improvement of sensitivity of electrolyte cathode discharge atomic emission spectrometry (ELCAD-AES) for mercury using acetic acid medium, Talanta, 2012, 93, 32–36,  DOI:10.1016/j.talanta.2012.02.004.
19. Z. Wang, A. J. Schwartz and S. J. Ray, et al., Determination of trace sodium, lithium, magnesium, and potassium impurities in colloidal silica by slurry introduction into an atmospheric-pressure solution-cathode glow discharge and atomic emission spectrometry, J. Anal. At. Spectrom., 2013, 28(2), 234–240,  10.1039/C2JA30253E.
20. Z. Wang, R. Gai and L. Zhou, et al., Design modification of a solution-cathode glow discharge-atomic emission spectrometer for the determination of trace metals in titanium dioxide, J. Anal. At. Spectrom., 2014, 29(11), 2042–2049,  10.1039/C4JA00173G.
21. R. Manjusha, M. A. Reddy and R. Shekhar, et al., Determination of cadmium in Zircaloys by electrolyte cathode discharge atomic emission spectrometry (ELCAD-AES), Anal. Methods, 2014, 6(24), 9850–9856,  10.1039/C4AY01242A.
22. R. Manjusha, M. A. Reddy and R. Shekhar, et al., Determination of major to trace level elements in Zircaloys by electrolyte cathode discharge atomic emission spectrometry using formic acid, J. Anal. At. Spectrom., 2013, 28(12), 1932–1939,  10.1039/C3JA50202C.
23. W. B. Jones, G. L. Donati and Jr C. P. Calloway, et al., Standard dilution analysis, Anal. Chem., 2015, 87(4), 2321–2327,  DOI:10.1021/ac504152x.
24. J. T. Sloop, H. J. B. Bonilla and T. Harville, et al., Automated matrix-matching calibration using standard dilution analysis with two internal standards and a simple three-port mixing chamber, Talanta, 2019, 205, 120160,  DOI:10.1016/j.talanta.2019.120160.
25. E. Mavrakis and S. A. Pergantis, Chip-based microfluidics on-line with inductively coupled plasma-mass spectrometry for standard dilution analysis, Anal. Chim. Acta, 2021, 1179, 338830,  DOI:10.1016/j.aca.2021.338830.
26. V. C. P. Marrocos, J. R. de Souza and T. D. Saint Pierre, Standard dilution analysis (SDA) as a powerful tool for elemental determination in biodiesel by inductively coupled plasma optical emission spectrometry (ICP OES), J. Anal. At. Spectrom., 2023, 38(12), 2547–2553,  10.1039/D3JA00192J.
27. J. A. Carter, P. J. Gray and T. I. Todorov, Exploring the working range of automated standard dilution analysis of nutrient elements in foods by inductively coupled plasma optical emission spectrometry, J. Anal. At. Spectrom., 2023, 38(4), 818–826,  10.1039/D2JA00346E.
28. W. B. Jones, G. L. Donati and C. P. Calloway, et al., Automated standard dilution analysis, J. Anal. At. Spectrom., 2020, 35(1), 178–187,  10.1039/C9JA00339H.
29. J. R. Ingham, L. T. Minas and G. L. Donati, et al., Standard dilution analysis using an automatic sampler and a peristaltic pump stop- step for ICP-OES determinations, Microchem. J., 2023, 190, 108603,  DOI:10.1016/j.microc.2023.108603.
30. F. M. Fortunato, M. A. Bechlin and J. A. G. Neto, et al., Standard dilution analysis in flow system: sodium determination by flame atomic emission spectrometry, Microchem. J., 2016, 124, 662–667,  DOI:10.1016/j.microc.2015.10.019.
31. F. M. Fortunato, M. A. Bechlin and J. A. G. Neto, et al., Internal standard addition calibration: determination of calcium and magnesium by atomic absorption spectrometry, Microchem. J., 2015, 122, 63–69,  DOI:10.1016/j.microc.2015.04.009.
32. A. G. Althoff, C. B. Williams and T. McSweeney, et al., Microwave-induced plasma optical emission spectrometry (MIP OES) and standard dilution analysis to determine trace elements in pharmaceutical samples, Appl. Spectrosc., 2017, 71(12), 2692–2698,  DOI:10.1177/0003702817721750.
33. F. M. Fortunato, A. L. Vieira and J. A. G. Neto, et al., Expanding the potentialities of standard dilution analysis: Determination of ethanol in gasoline by Raman spectroscopy, Microchem. J., 2017, 133, 76–80,  DOI:10.1016/j.microc.2017.03.015.
34. G. L. Long and J. D. Winefordner, Limit of detection. A closer look at the IUPAC definition, Anal. Chem., 1983, 55(7), 712A–724A,  DOI:10.1021/ac00258a001.
35. T. Cserfalvi, P. Mezei and P. Apai, Emission studies on a glow discharge in atmospheric pressure air using water as a cathode, J. Phys. D: Appl. Phys., 1993, 26(12), 2184,  DOI:10.1088/0022-3727/26/12/015.
36. M. R. Webb, F. J. Andrade and G. Gamez, et al., Spectroscopic and electrical studies of a solution-cathode glow discharge, J. Anal. At. Spectrom., 2005, 20(11), 1218–1225,  10.1039/B503961D.
37. F. C. Pinheiro, et al., Dispersive liquid-liquid microextraction based on deep eutectic solvent for elemental impurities determination in oral and parenteral drugs by inductively coupled plasma optical emission spectrometry, Anal. Chim. Acta, 2021, 1185, 339052,  DOI:10.1016/j.aca.2021.339052.
38. P. Zheng, Q. Hu and H. Zhang, et al., Elemental Analysis of Environmental Waters by Solution Cathode Glow Discharge—Atomic Emission Spectrometry (SCGD-AES) with a Multifunctional Injection System, Anal. Lett., 2022, 55(14), 2273–2285,  DOI:10.1080/00032719.2022.2053146.

---