Methodological Applicability of Ultra-Low Background Liquid Scintillation Counters in Low-Level Tritium Measurement

Abstract

Tritium (³H) is a low-energy β emitter commonly found in environmental water samples, and its routine monitoring requires highly sensitive techniques capable of achieving low detection limits. Liquid scintillation counting (LSC) is the standard method for low-level ³H analysis; however, quenching significantly affects detection efficiency and minimum detectable activity (MDA), and systematic evaluations across different quench levels and measurement approaches remain limited. This study evaluates quench-related uncertainties in low-level ³H measurement using two ultra-low background liquid scintillation counters, Quantulus 1220 and GCT 6220. High- and low-quench conditions were created by varying sample-to-cocktail ratios, and performance was assessed through detection efficiency, minimum detectable activity (MDA), and stability. Under the relative measurement method with limited quench variation, GCT 6220 achieved higher efficiency, lower background, and lower detection limits. Under the internal standard method with broader quench spans, Quantulus 1220 produced smoother efficiency–quench curves and more stable results. Thus, GCT 6220 is advantageous for sensitivity-demanding scenarios, while Quantulus 1220 is better suited for quench-correction applications.

1. Introduction

³H is a radioactive isotope of hydrogen with a half-life of approximately 12.3 years, a maximum β energy of 18.6 keV, and an average energy of about 5.7 keV, classified as a low-energy β nuclide [1]. Its sources include both natural production induced by cosmic rays and anthropogenic releases from nuclear facility operations, fuel reprocessing, and historical nuclear tests, allowing it to migrate within water, soil, and organisms, thereby participating in ecological cycles [2,3,4]. Due to its low energy, low environmental concentration, and ease of mixing with water, the primary exposure pathways of ³H to the public are drinking water and dietary intake through the food chain [5,6,7]. The WHO drinking-water guideline sets a guidance level of 10,000 Bq/L for tritium. However, environmental ³H concentrations are typically at the Bq/L to sub-Bq/L level, which requires analytical methods with much lower MDAs, generally in the 1–10 Bq/L range for reliable monitoring [8]. Liquid scintillation counting (LSC), as a widely used technique for measuring low-energy β nuclides, offers high counting efficiency and simple sample preparation, owing to thorough mixing of the sample and scintillation cocktail, negligible self-absorption, and nearly 4π counting geometry. It has been extensively applied to low-level ³H measurement [9]; however, results are constrained not only by instrumental background but also by factors such as chemical or color quenching, necessitating coordinated optimization in sample preparation, quench correction, and method selection.

Among commonly used ultra-low background LSC platforms, Quantulus 1220 and GCT 6220 are widely applied. The Quantulus 1220 employs massive asymmetric lead shielding combined with active protection, multiple multichannel analysis, and automatic continuous spectrum stabilization, while enabling α/β discrimination through pulse shape analysis (PSA) and pulse amplitude comparison (PAC). Its external standard source was changed from ²²⁶Ra in earlier models to ¹⁵²Eu to establish the SQP (E) quench curve [10]. The latter incorporates Guard Compensation Technology (GCT) electronic anticoincidence into conventional shielding to suppress cosmic rays and scattered background, with ≥2-inch external lead shielding and an internal ¹³³Ba external standard for energy calibration and quench curve establishment [11]. Pujol and Sánchez-Cabeza conducted optimization studies on ³H measurement conditions with Quantulus 1220, showing that by adjusting the sample-to-cocktail ratio, pulse discrimination parameters, and vial type, the detection limit could be reduced to approximately 2.2 Bq/L, highlighting the potential of Quantulus 1220 in low-level ³H measurements [12]. Ronald H. W. Edler focused on the performance characteristics of the Quantulus GCT 6220. Results showed that, owing to GCT, it can significantly reduce β-channel background, shortening the counting time required to achieve the regulatory ³H detection limit of 10 Bq/L from 32 min to 8 min, while also demonstrating good sensitivity and stability in total α/β and radon measurements [13]. Building on this, subsequent studies have extended the focus from a single platform to performance comparisons across platforms. J. L. García-Leon and colleagues systematically compared the performance of the two instruments in measuring ⁵⁵Fe and ⁶³Ni activities in low-level environmental samples, assessing the influence of different scintillation cocktails on background counts, detection efficiency, and the lower limit of detection (LLD) [14]. Xiao-gui Feng and collaborators compared the two instruments in terms of low-background count rate, counting efficiency, α/β discrimination capability, and stability when measuring ³H, ¹⁴C, ²⁴¹Am, ⁹⁰Sr, ⁹⁰Y, ⁶⁰Co, and ¹³⁷Cs, providing important insights into the fundamental performance differences between the two systems [15]. Existing research has mainly focused on multi-nuclide measurements or overall performance evaluation under different scintillation cocktail conditions. However, studies specifically comparing the relative measurement method and the internal standard method across different quench levels in low-level ³H environmental samples remain insufficiently explored.

In light of the aforementioned progress and limitations, this study conducts experimental research on the direct measurement of low-level ³H in environmental samples. Using the Quantulus 1220 and GCT 6220 platforms, the relative measurement method and the internal standard method were applied under both low- and high-quench conditions. Detection efficiency (ε), minimum detectable activity (MDA), and figure of merit (FOM) were compared. The objective is to determine the applicability and trade-offs of the two methods across different quench ranges and, on this basis, to elucidate the performance differences between the platforms under identical methods and conditions, thereby providing directly applicable evidence and recommendations for subsequent method selection and platform configuration. By integrating both measurement strategies with well-defined quench ranges in real environmental matrices, this work offers a more comprehensive perspective that complements the existing studies and strengthens the practical basis for choosing appropriate measurement conditions.

2. Materials and Methods

The overall design of this study is as follows:

This study conducted experiments on two ultra-low background LSC platforms, Quantulus 1220 and GCT 6220, using identical measurement methods to verify the consistency of results and to define the parameter ranges for stable operation of each platform.

• Introduction of different quench conditions: by varying the sample-to-cocktail ratios (1:9 and 2:3), two categories of conditions—“low quench” and “high quench”—were established to simulate the differences in quenching levels found in real environmental samples and to evaluate the applicability of the methods.
• Methodological comparison: the relative measurement method and the internal standard method were employed to analyze the applicability of different approaches under varying quench conditions.
• Performance evaluation: the detection performance under different conditions was comprehensively assessed using three indicators—detection efficiency (Eff), minimum detectable activity (MDA), and figure of merit (FOM).

2.1. Instruments and Experimental Environment

This study was carried out on two ultra-low background LSC platforms: the Quantulus 1220, equipped with dual photomultiplier tubes and an external standard calibration system, using SQP (Spectral Quench Parameter of the External standard) as the quench index. This factor is the end channel of the Compton spectrum obtained with the measurement of the vial of interest while exposed to a152 Eu external source [16]; and the GCT 6220, which incorporates Guard Compensation Technology on top of conventional lead shielding, employing tSIE (transformed Spectral Index of the External Standard) as the quench index while examining the effects of the OFF, LOW, and HIGH operating modes—where OFF disables the guard detector, LOW is intended for radionuclides measured in broader low-energy windows, and HIGH provides the strongest background suppression for very low-energy emissions—on background suppression and measurement sensitivity [10]. Based on literature and preliminary experimental experience, multiple energy spectrum channel ranges were configured and compared, and the optimal configuration was determined by evaluating efficiency, detection limits, and figure of merit [13]. Detailed comparisons and trade-offs are presented in Chapter 3. All experiments were conducted in a temperature-controlled laboratory at 23 ± 1 °C with a relative humidity of approximately 50%, using a stabilized power supply to minimize interference from power fluctuations on light yield and PMT stability.

2.2. Samples and Materials

This study used two certified ³H standards with activities of 974 ± 34 Bq/g and (2.185 ± 0.076) × 10⁵ Bq/g (both with k = 2), corresponding to reference dates of 18 March 2025 and 6 August 2020, respectively. Their activities were decay-corrected to the measurement date before use. Both standard sources were stored under a constant temperature of 22 °C to maintain activity stability and traceability. Ultima Gold uLLT was selected as the scintillation cocktail, suitable for low-energy β measurement, offering high light yield and low intrinsic background [17]. Environmental sampling was conducted at six locations surrounding the nuclear power plant, covering distances of approximately 0.5–3 km. At each location, representative soil samples were collected from several depth intervals, and one plant sample from the corresponding site was also obtained. Replicate samples were collected at each depth to ensure adequate representativeness. The sampling campaign took place from 1 to 2 March 2025, and in total approximately 90 environmental samples were obtained. All samples were subsequently processed via high-temperature combustion oxidation to extract water for tritium analysis. Following the sample preparation procedure for organically bound ³H described by Zhang Qin et al., the samples were freeze-dried and homogenized, then combusted in a high-temperature oxidation furnace, with the main water-release stage occurring at approximately 850 °C. At this temperature, non-aqueous volatile organic compounds are thermally decomposed or oxidized and therefore do not enter the condensed water fraction. The collected combustion water was subsequently mixed with the scintillation cocktail for liquid scintillation counting [18]. All samples were contained in the same batch of 20 mL low-background polyethylene vials to minimize additional background interference.

To investigate the effects of quenching and mixing ratios, this study established two groups: 2 mL sample + 18 mL scintillation cocktail (low quench, 1:9) and 8 mL sample + 12 mL scintillation cocktail (high quench, 2:3) [19]. The 1:9 ratio was chosen primarily due to the limited amount of combustion-derived water, ensuring feasibility and repeatability of the experiments; moreover, this ratio has also been adopted in the literature [18], providing additional validation of its rationality.

Quenching was further characterized using instrument-specific indices, tSIE for the GCT 6220 and SQP for the Quantulus 1220, both of which decrease with increasing quench severity. In this study, the high-quench (2:3) samples showed tSIE values of approximately 310–330 and SQP values of 770–780. The low-quench (1:9) samples corresponded to higher quench-index values, with tSIE ranging approximately from 600–680 on the GCT 6220, and SQP distributed more broadly from about 760 to 840 on the Quantulus 1220. These values form the quantitative basis for the high- and low-quench classifications used throughout this work.

Blank samples were prepared using low-³H water and scintillation cocktail at the two ratios mentioned above, mixed for 1 min, and then kept in the dark for 24 h to minimize the influence of short-term optical fluctuations on counting [20]. For the relative measurement method, the standard samples were prepared by diluting 0.1 mL of 974 Bq g⁻¹ standard ³H water to volumes of 2 mL and 8 mL, respectively, and then adding scintillation cocktail to a constant total volume of 20 mL. The resulting activity concentrations were approximately 4.87 × 10⁴ Bq/L and 1.22 × 10⁴ Bq/L, which were used to calculate detection efficiency and to verify changes in MDA and FOM. For the internal standard method, high-activity ³H water (2.185 × 10⁵ Bq g⁻¹) was used as the spiking source. A minimal volume was added to enhance counting rates while minimizing changes to the quench index, thereby ensuring reliable characterization of the efficiency–quench relationship. Multiple energy spectrum channel ranges were configured based on literature and preliminary experiments, which were subsequently used in results and discussion for systematic evaluation and optimization with respect to efficiency, MDA, and FOM [16].

2.3. Methods

2.3.1. Relative Measurement

In the relative measurement method, the counting rate of a standard sample with known radioactivity is measured, and in combination with the background counting rate, the counting efficiency (E_ff) of the sample can be calculated, thereby determining the ³H activity in the test sample [21]. The calculation formula for the counting efficiency E_ff is as follows:

$$E_{ff}={\displaystyle \frac{N_s-N_b}{A_s\times V_s\times 60}}$$

(1)

where N_s is the counting rate of the standard sample (cpm); N_b is the counting rate of the background sample (cpm); A_s is the activity of the standard sample (Bq mL⁻¹); and V_s is the volume of the standard sample (mL).

After obtaining the efficiency, the net counting rate of the unknown environmental sample is substituted into the equation to determine its activity concentration:

$$A_x={\displaystyle \frac{N_x-N_b}{E_{ff}\times V}}$$

(2)

where N_x is the counting rate of the sample to be measured (cpm); N_b is the background counting rate (cpm); V is the sample volume (L); and E_ff is the counting efficiency (%). The advantage of this method lies in its simplicity of calculation, as the activity of the unknown sample can be directly obtained through calibration with the standard sample. However, its accuracy depends on the consistency of quench conditions between the test sample and the standard sample.

In practical data analysis, the choice of channel width directly affects the signal acquisition efficiency and background level of the liquid scintillation counter. If the channel width is set too narrow, part of the valid signal may be excluded; if it is too wide, additional background may be introduced, resulting in reduced detection sensitivity. Therefore, an appropriate choice of channel width is of great significance for lowering the detection limit [11]. To quantify the balance between efficiency and background under different settings, this study introduces the figure of merit (FOM) as an evaluation parameter. The magnitude of FOM can be used to determine the optimal channel width; when it reaches its maximum value, it indicates that the chosen width balances high signal collection capacity with low background level, thereby improving the detection limit of the method to some extent [22]. The formula for the figure of merit is as follows:

$$FOM={\displaystyle \frac{E_{ff}^2}{N_b}}$$

(3)

where E_ff is the technical efficiency (%); and N_b is the background counting rate (cpm).

Along with obtaining the FOM, the minimum detectable activity (MDA) can be further estimated by combining the total background counts with the measurement time [23]. The calculation formula is as follows:

$$MDA={\displaystyle \frac{2.71+\sqrt{N_b\times T}}{60\times E_{ff}\times V\times T}}$$

(4)

where N_b is the background counting rate (cpm); T is the measurement time (min); E_ff is the counting efficiency (%); and V is the volume of the sample to be measured (L). The constant 2.71 originates from Currie’s detection-limit theory, representing the Poisson-based decision level under low-background conditions. In all measurements performed in this study, including background samples, standards, and environmental samples, a total counting time of T = 1000 min was used. This fixed measurement duration ensures consistency in the evaluation of MDA and FOM across all experiments.

2.3.2. Internal Standard Method

In the internal standard method, the determination of detection efficiency does not rely on external standard samples but is achieved through spiking experiments. Specifically, a small amount (1 μL) of high-activity standard ³H water is added to the sample under test to produce a quantifiable increase in counting rate while keeping the quench conditions as unchanged as possible [24]. The efficiency can be calculated using the following equation:

$$E_{ff}={\displaystyle \frac{N_{cs}-N_c}{60\times \mathrm{V}\times A_{CS}}}$$

(5)

where N_c is the counting rate of the original sample (cpm); N_cs is the counting rate after spiking (cpm); A_cs is the activity concentration of the added standard ³H water (Bq mL⁻¹); and V is the total volume (mL). Compared with the relative measurement method, this approach can effectively eliminate the influence of quench differences among samples on efficiency.

The obtained efficiency can then be used to calculate the activity concentration of the test sample. Since the calculation method for unknown samples is identical to that of the relative measurement method, it will not be repeated here.

2.3.3. Uncertainty Evaluation

Each sample was measured repeatedly, and the uncertainty was assessed using a Type A evaluation based on the statistical variation in the replicate results. For a set of measurements $x_1,x_2,\dots ,x_n$, the mean and standard deviation were calculated as:

$$\overline{x}={\displaystyle \frac{1}{n}}\sum _{i=1}^n{x}_i$$

(6)

$$s=\sqrt{{\displaystyle \frac{1}{n-1}}\sum _{i=1}^n{\left(x_i-\overline{x}\right)}^2}$$

(7)

The standard uncertainty was calculated as:

$$u={\displaystyle \frac{\mathrm{s}}{\sqrt{n}}}$$

(8)

and the expanded uncertainty was obtained as:

$$U=k\times u$$

(9)

where $x_i$ represents each individual measurement, $\overline{x}$ is the mean value, $n$ is the number of replicates, $s$ is the standard deviation, $u$ is the standard uncertainty, and $U$ is the expanded uncertainty.

A coverage factor of $k=2$ corresponds to a confidence level of approximately 95%.

In this study, each sample—whether background, standard, or environmental—was measured in ten consecutive runs, and all reported measurement results represent the arithmetic mean of these ten replicates. This repeated-measurement design reduces statistical fluctuations and provides a more reliable basis for uncertainty evaluation.

3. Results and Discussion

3.1. Region of Interest (ROI)

In liquid scintillation counting experiments, the setting of energy regions or channels plays a critical role in balancing detection efficiency, MDA, and FOM. To ensure comparability in subsequent methodological comparisons, this section pre-optimizes the counting windows and related parameters of the GCT 6220 and Quantulus 1220 under both high- and low-quench sample conditions, and determines the unified settings for subsequent experiments based on the observations from figures and Supplementary Materials.

To provide basic context for the following methodological comparison, the environmental samples used in this study (Tables S5–S7 in the Supporting Information) exhibited tritium activities of approximately 5–60 Bq/L, which fall within typical background levels around a nuclear power plant and are far below regulatory limits such as the WHO guideline of 10,000 Bq/L.

3.1.1. Optimization Results of GCT 6220

As shown in Figure 1, under both high- and low-quench conditions, the experimental results of the GCT 6220 exhibit consistent patterns: performance differences across energy regions are significant, with the 2–18.6 keV region performing best in the HIGH mode. This superiority is reflected not only in higher counting efficiency and lower detection limits but also in the substantially reduced background counting rate. As the GCT mode shifts from OFF to LOW to HIGH, the background noise level decreases markedly: in the OFF mode, background levels are considerably elevated, reducing effective net counts and thereby limiting detection sensitivity; in the LOW mode, background is significantly reduced, leading to notable performance improvement; in the HIGH mode, the combined benefits of increased efficiency and greatly reduced background result in fully optimized sensitivity and stability, making the overall measurement capability the most superior, consistent with findings from previous studies [25].

Therefore, based on a comprehensive comparison of efficiency, detection limit, and background level under different quench conditions, the 2–18.6 keV region in the HIGH mode should be regarded as the optimal detection configuration, making it more suitable for precise measurement of low-level ³H samples.

3.1.2. Optimization Results of Quantulus 1220

As shown in Figure 2, the overall detection efficiency of the Quantulus 1220 remains stable across different channel ranges, but indicators such as background counting rate, figure of merit, and detection limit still exhibit significant variations. As the channel range is progressively narrowed, background noise decreases from approximately 2 cpm to 1.4–1.6 cpm, accompanied by a reduction in detection limit and an improvement in the figure of merit, indicating that channel selection directly affects the separation between signal and background.

A comprehensive comparison shows that the 60–220 channel range exhibits the most balanced performance, providing both a high figure of merit and a low detection limit, while avoiding the additional background introduced by wider channel ranges. Therefore, under different quench conditions, the 60–220 channel range can be regarded as the optimal configuration for the Quantulus 1220, making it more suitable for precise measurement of low-level ³H samples.

3.2. Evaluation of the Relative Measurement Method

Under high-quench conditions, the energy transfer process within the sample is significantly suppressed, reducing the number of scintillation photons generated per radioactive decay, weakening the light signal, and thereby increasing statistical uncertainty. In such cases, the detection system must possess high detection efficiency to compensate for photon loss and rely on a low detection limit to distinguish weak signals from background noise. Figure 3 presents the results of high-quench samples measured using the relative measurement method. Both platforms completed the measurements under their respective parameter settings, but differences were observed in the combined performance of efficiency, background, and detection limit. Within the parameter range covered in this study, the GCT 6220 demonstrated a lower minimum detectable activity and a higher figure of merit, aligning more closely with the detection requirements under high-quench conditions.

In this batch of low-quench samples, although the overall light output was higher, the broader distribution of quench indices resulted in noticeable fluctuations in efficiency and detection limits, restricting the applicability of the relative method [9]. Therefore, this study turned to the internal standard method for analyzing such samples.

To evaluate the comparability of results between different instruments, a correlation analysis was conducted on measurements of the same batch of high-quench samples using the Quantulus 1220 and GCT 6220. The data are provided in Tables S5 and S6 in the Supporting Information, and the correlation analysis results are shown in Figure 4. The results show a high degree of consistency between the two (R² = 0.91). This indicates that when quench levels do not vary significantly, a linear relationship exists between the measurements of the two instruments, suggesting the feasibility of cross-calibration. This finding suggests that in multi-instrument collaborations or interlaboratory comparisons, a linear calibration strategy can be used to harmonize data, thereby enhancing the comparability and generalizability of the results.

3.3. Evaluation of the Internal Standard Method

Under high-quench conditions (Figure 5), although the overall efficiency of the GCT 6220 is higher than that of the Quantulus 1220, the efficiency–quench index relationships of both instruments exhibit an approximately flat trend, failing to display the expected gradual increase with quench severity under ideal conditions. This phenomenon may be attributed to the narrow range of quench variation in this batch of samples (SQP values varied only from ~440 to 460 on the Quantulus 1220 and ~520 to 540 on the GCT 6220): the differences in quench levels were insufficient to produce a distinct efficiency gradient, resulting in a nearly flat fitted curve. This implies that under the present experimental conditions, the internal standard method did not achieve the expected correction effect [19]. It should be noted that this does not indicate a general failure of the internal standard method under high-quench conditions but rather suggests that its effectiveness may be influenced by the distribution range of quench levels in the samples. Future tests conducted over a wider range of quench conditions may more clearly reveal the expected trend of increasing efficiency with decreasing quench severity (i.e., with increasing quench index), thereby allowing a more comprehensive evaluation of the internal standard method under such conditions.

Under low-quench conditions, the efficiency–quench index relationships of the two platforms exhibit different characteristics. The GCT 6220 shows overall higher detection efficiency, but its curve fluctuates within certain quench intervals, indicating that its results are more strongly affected by variations in quench. In contrast, the Quantulus 1220 exhibits slightly lower efficiency values, yet its curve varies more smoothly with the quench index, displaying a continuous upward trend. Although regression curves are not shown in Figure 5, a polynomial regression analysis was performed to quantitatively assess the stability of the efficiency–quench relationships. The resulting goodness-of-fit values support the visual trend: the Quantulus 1220 exhibits a substantially higher R² (0.913) than the GCT 6220 (0.622), indicating a more consistent and stable dependence of efficiency on the quench index under low-quench conditions. In the application of the internal standard method, a stable and gradually increasing efficiency–quench relationship is crucial for establishing a reliable calibration curve. Accordingly, under low-quench conditions, the Quantulus 1220 is more favorable for conducting internal standard measurements. For reference, the low-quench samples measured on the GCT 6220 corresponded to tSIE values of approximately 600–680, although these data were not used for internal-standard calibration. The relevant measurement results are provided in Table S7 in the Supporting Information.

3.4. Summary and Discussion

In the relative measurement method, high-quench samples, due to reduced light yield, impose stricter requirements on efficiency and detection limits. Within the parameter range of this study, the GCT 6220 better meets these demands in terms of the combined performance of efficiency, background, and detection limit. For low-quench samples, where signal levels are sufficient, dependence on noise suppression and spectral stability is greater, and under such conditions the Quantulus 1220 exhibits better operability and stability. The results of the internal standard method further reveal that its applicability depends not only on instrument performance but also on the distribution range of sample quench indices. In this batch, the quench range of high-quench samples was relatively narrow, resulting in a nearly flat efficiency–quench relationship and thus limited calibration gains. The broader distribution of low-quench samples was more favorable for establishing a smooth and monotonic efficiency–quench relationship, making the results of the Quantulus 1220 more consistent with the requirements of the internal standard method.

Overall, the range of sample quench distribution is the primary factor determining the applicability of a method. Although the relative measurement method can be directly applied under different quench conditions, its accuracy depends on the balance between efficiency and background; as quench differences among samples increase, the uncertainty of results also rises. Whether the internal standard method can realize its correction advantage depends on the availability of quench points with sufficient range and balanced distribution to support a stable efficiency–quench curve. Therefore, the two methods are complementary under different scenarios: when quench distribution is narrow and the goal is to lower the detection limit, the relative measurement method is preferred; when the quench distribution is broader and the goal is to enhance consistency, the internal standard method is prioritized. On this basis, further comparison of platform-specific performance under given methods and conditions can yield lower detection limits, higher figures of merit, and improved result consistency.

4. Conclusions

To achieve precise measurement of low-level ³H in environmental samples, this study conducted systematic comparisons and quantitative evaluations within well-defined high- and low-quench ranges, employing both the relative measurement method and the internal standard method on the Quantulus 1220 and GCT 6220 platforms. The following conclusions apply only to the quench ranges, methodological settings, and platform parameters specified in this study.

The conclusions are as follows:

• Under high-quench conditions using the relative measurement method, the GCT 6220 platform outperformed the Quantulus 1220 in terms of combined indicators of efficiency, background, and detection limit, characterized by a lower minimum detectable activity and a higher figure of merit.
• Under high-quench conditions using the relative measurement method, the measurements from the two platforms showed linear consistency (R² = 0.90), suggesting the potential for aligning results through a linear model, though its robustness requires further verification across broader quench ranges and additional parameter settings.
• Under low-quench conditions using the internal standard method, the efficiency–quench relationship of the Quantulus 1220 platform was smoother, with more stable results and smaller deviations, making it more suitable for applications that rely on quench correction.

Accordingly, in practical applications, the method–platform combination should be selected based on the quench distribution of the samples: when the quench distribution range is relatively narrow, the relative measurement method in combination with the GCT 6220 should be used; when the quench distribution is broader, the internal standard method with the Quantulus 1220 is preferable; when the quench distribution is unclear, a preliminary quench assessment with a small number of samples may be conducted to determine the appropriate combination.