Challenges in Interpreting ⁴⁰Ar/³⁹Ar Age Spectra: Clues from Hydrothermally Altered Alkali Feldspars

Abstract

Integrated ⁴⁰Ar/³⁹Ar and U-Pb geochronology, combined with microstructural analysis of Early Cretaceous volcanics from eastern China, challenge conventional interpretations of flat ⁴⁰Ar/³⁹Ar age spectra. K-feldspar sample JD-1K (122.12 ± 0.81 Ma) preserves magmatic sanidine characteristics (homogeneous composition, disordered monoclinic structure), while hydrothermally altered perthite JD-2K yields a flat plateau age of 99.83 ± 0.73 Ma (~20 Ma younger than coeval K-feldspar, biotite, and zircon samples). Microstructural analyses using energy dispersive spectroscopy (SEM−EDS), Fourier transform infrared spectroscopy (FTIR), and transmission electron microscopy (TEM) methods unequivocally demonstrate that the concordant ⁴⁰Ar/³⁹Ar age spectrum of sample JD-2K is a result of isotopic resetting during fluid-mediated recrystallization processes, rather than primary post-crystallization thermal stability. In step-heating experiments, contrasting argon release patterns correlate with microstructural heterogeneities. This study challenges the paradigm that flat ⁴⁰Ar/³⁹Ar spectra uniquely signify post-crystallization thermal histories, demonstrating that hydrothermal alteration can fully reset argon systems to produce misleadingly concordant ages. This study highlights the complexity of interpreting isotopic data in hydrothermally altered rocks, emphasizing the necessity of integrated petrological-geochemical analyses to differentiate primary magmatic signals from secondary overprints.

1. Introduction

⁴⁰Ar/³⁹Ar step-heating age spectra are constructed by plotting the apparent ages of successive ³⁹Ar gas fractions released at increasing temperatures, thereby mapping the spatial distribution of argon isotopes from mineral cores to rims. These spectra provide critical insights into post-crystallization processes, as specific morphological features (e.g., ascending, descending, saddle-shaped, hump-shaped) correlate with distinct mechanisms of argon redistribution [1,2,3,4,5,6,7,8,9,10]. Thus, precise interpretation of age spectra is fundamental for reconstructing geological histories. Complex ⁴⁰Ar/³⁹Ar age spectra require rigorous interpretation to evaluate their geological significance. In contrast, flat age spectra are typically interpreted as crystallization ages under the assumption of closed-system behavior and no significant post-crystallization thermal disturbance. However, oversimplified interpretations of flat age spectra may lead to misinterpretations, as factors controlling age spectrum morphologies are inherently complex. This underscores the need for caution even in the presence of apparently flat ⁴⁰Ar/³⁹Ar age plateaus.

Here we report ⁴⁰Ar/³⁹Ar geochronological results for two K-feldspar separates (JD-1K and JD-2K) and biotite JD-2B from coeval volcanic rocks in the Jiaolai Basin. All samples yield flat age spectra with negligible atmospheric ⁴⁰Ar contributions. Notably, however, K-feldspar JD-2K yields an age ~20 Ma younger than coexisting K-feldspar JD-1K and biotite JD-2B. This discrepancy is unexpected given the rapid cooling rates of volcanic rocks, which typically produce age congruence within analytical uncertainty. As K-feldspars are common phases in volcanic rocks and widely used in ⁴⁰Ar/³⁹Ar geochronology [11,12,13,14,15,16], this age offset necessitates detailed investigation into potential causes. This study also integrates ⁴⁰Ar/³⁹Ar and U-Pb geochronological datasets to systematically compare dating systems. Mineralogical characterizations, including chemical compositions, structural ordering, and microstructural features, were conducted for the two K-feldspar populations. We anticipate that this article will serve as a valuable reference for refining dating protocols and accurately evaluating analytical data.

2. Sample Description

The two samples, JD-1 and JD-2, were procured from the Jiaolai Basin in Shandong Province, East China. The Jiaolai Basin represents a Cretaceous extensional basin, predominantly filled with terrigenous sedimentary and volcanic rocks. The Cretaceous strata therein encompass the Lower Cretaceous Laiyang (K₁l) and Qingshan (K₁q) Groups, as well as the Upper Cretaceous Wangshi Group.

Mesozoic volcanic lavas preponderantly manifest within the Qingshan Group (K₁q), which is underlain by the terrestrial facies of the Laiyang (K₁l) Group and overlain by that of the Wangshi (K₁w) Group (Figure 1b). Characteristically, the Qingshan Group (K₁q) consists of a series of intermediate to acidic lavas and pyroclastic rocks interspersed with sedimentary formations. Based on comprehensive field geological surveys, the stratum of the Qingshan Group can be further subdivided, from bottom to top, into four formations, namely the Houkuang (K₁qh), Bamudi (K₁qb), Shiqianzhuang (K₁qs), and Fanggezhuang (K₁qf) Formations [17,18,19], Figure 1b. All four formations are conspicuously exposed in the eastern subregion of the Jiaolai Basin. The Houkuang, Shiqianzhuang, and Fanggezhuang Formations are principally constituted of purple andesite- and breccia-bearing tuffs, whereas the Bamudi Formation is predominantly composed of conglomerates and coarse-grained sandstones intercalated with trachyandesite layers. Some U-Pb ages of zircons derived from diverse layers within this suite of volcanic rocks, as reported by [19], suggest that the volcanic eruption transpired at approximately 119–122 Ma (Figure 1b).

We collected two samples, JD-1 and JD-2, from the lower and middle segments of the Houkuang Formation. The detailed sampling locations are illustrated in Figure 2. The lithology of sample JD-1 is rhyolite, while that of sample JD-2 is rhyodacite (Figure 2a,b). K-feldspars, namely JD-1K (depicted in Figure 2d) and JD-2K (shown in Figure 2f), were selected from samples JD-1 and JD-2, respectively. Given the pronounced age disparities between the two K-feldspars, JD-1K and JD-2K, we opted to extract zircon (JD-1Z, Figure 2c) from sample JD-1 and biotite (JD-2B, Figure 2e) from sample JD-2, with the intention of conducting U-Pb and ⁴⁰Ar/³⁹Ar dating, respectively, in order to contrast the ages obtained from different minerals using different chronometric methods.

3. Methods

Herein, we present detailed procedures for dating, along with techniques such as scanning electron microscopy—energy dispersive spectroscopy (SEM−EDS), Fourier transform infrared spectroscopy (FTIR), and transmission electron microscopy (TEM).

3.1. ⁴⁰Ar/³⁹Ar and U-Pb Analytical Processes

3.1.1. ⁴⁰Ar/³⁹Ar Dating

The K-feldspar, biotite separates, and flux monitors were individually encapsulated in aluminum foils and then placed at position H8 of the 49-2 Nuclear Reactor (49-2 NR) in Beijing, China for irradiation. To ensure homogeneous neutron distribution, the samples were continuously rotated during irradiation. Cd shielding was used to minimize undesirable nuclear interference reactions in irradiation. Parameter J, which serves to quantify the conversion of ³⁹K to ³⁹Ar, is meticulously determined through analysis of a co-irradiated fluence monitor with a precisely known age. Specifically, the J values of JD-1K, JD-2K, and JD-2B were determined to be 0.003583 ± 0.000011, 0.003559 ± 0.000011, and 0.003539 ± 0.000011, respectively. To calibrate the ages of the unknown samples, we employed the YBCs standard, which has an age of 29.475 ± 0.096 Ma (again, based on the decay constants of [20]), as per the methodology described in [21].

Following irradiation, analysis of Ar isotopes was carried out through incremental heating at the ⁴⁰Ar/³⁹Ar laboratory of the Institute of Geology and Geophysics, Chinese Academy of Sciences (IGGCAS) in Beijing. The power of the New Wave MIR10-50 laser was gradually increased from 1 W (with a power percentage of 1.4%) to 8 W (10%) until complete fusion of the samples was achieved. Each heating step lasted approximately 5 min, and during this time the gas was purified simultaneously using two SEAS NP10 Zr–Al pumps. The purified gas was then introduced into a Noblesse mass spectrometer for Ar isotopic analysis, which involved 10 cycles of peak hopping.

Neutron activation produces not only ³⁹Ar from ³⁹K but also a diverse array of other Ar isotopes. For example, some ⁴⁰Ar is produced by neutron activation of ⁴⁰K; additional ³⁹Ar is produced from ⁴²Ca; and ³⁶Ar is produced from ⁴⁰Ca and ³⁵Cl. To evaluate the interfering nucleogenic reactions caused by irradiation, CaF₂ and K₂SO₄ were utilized. The interference corrections were made based on the following nucleogenic production ratios: (⁴⁰Ar/³⁹Ar)_K = 0.00020 ± 0.000013; (³⁸Ar/³⁹Ar)_K = 0.012950 ± 0.000010; (³⁹Ar/³⁷Ar)_Ca = 0.00068 ± 0.000002; (³⁸Ar/³⁷Ar)_Ca = 0.0000196 ± 0.0000008; and (³⁶Ar/³⁷Ar)_Ca = 0.000278 ± 0.000002. Mass discrimination was calibrated through the use of an online air pipette, which was inserted at regular intervals among the unknown samples. The detailed analytical blanks are presented in Supplementary Table S1. The measured isotope ratios were corrected for system blanks, mass discrimination, atmospheric contamination, and irradiation-induced mass interference, following the procedures outlined in [22]. Additionally, mass discrimination was further corrected using an atmospheric ⁴⁰Ar/³⁶Ar ratio of 298.56, as recommended by [23]. Finally, the ⁴⁰Ar/³⁹Ar ages of the samples were computed using ArArCALC software (Version 2.5.2, developed by Anthony Koppers at Oregon State University, Corvallis, OR, United States), as described in [24].

3.1.2. U-Pb Dating

Measurements of U, Th, and Pb isotopes were conducted using a Cameca IMS-1280HR SIMS at the IGGCAS. Details regarding the instrument and the analytical procedure can be found in [25]. The primary O₂^– ion beam spot has an approximate size of 20 × 30 μm. Positive secondary ions were extracted with a potential of 10 kV. In the secondary ion beam optics, a 60 eV energy window was employed in conjunction with a mass resolution of around 5400 (at 10% peak height) to distinguish Pb⁺ peaks from isobaric interferences. Pb/U calibration was performed relative to the zircon standard Plesovice (with a ²⁰⁶Pb/²³⁸U age of 337 Ma, as reported by [26]). A long-term uncertainty of 1.5% (1 s relative standard deviation (RSD)) for the ²⁰⁶Pb/²³⁸U measurements of the standard zircons was applied to the unknown samples (as described in [27]). However, the measured ²⁰⁶Pb/²³⁸U error in a specific session was typically ≤1% (1 s RSD). The measured compositions were corrected for common Pb (i.e., non-radiogenic Pb) using nonradiogenic ²⁰⁴Pb. These corrections were so small that the results were hardly affected by the choice of common Pb composition. An average of the present-day crustal composition from [28] was used for the common Pb, assuming that the common Pb mainly consists of surface contamination introduced during sample preparation. Uncertainties in the individual analyses in the data tables are reported at the 1σ level. Concordia U-Pb ages are presented with 95% confidence intervals, unless otherwise stated. To keep track of the external uncertainties in secondary ion mass spectrometry (SIMS), U-Pb zircon dating calibrated against the Plesovice standard, and an in-house zircon standard, Qinghu, was analyzed alternately as an unknown along with other unknown zircons. Five measurements of the Qinghu zircon (see Supplementary Table S2) resulted in a concordia age of 160.9 ± 2.9 Ma, which is consistent with the recommended error of 159.5 ± 0.2 Ma (as reported in [29]).

3.2. Backscattered Electron (BSE) Images and Energy-Dispersive X-Ray Spectroscopy (EDS)

JD-1K and JD-2K grains were affixed to epoxy and polished for microstructural examination at the IGGCAS. A high-resolution field emission scanning electron microscope (Zeiss Gemini 450 from Zeiss Company, Germany), operating at 15 kV with a beam current of 2 nA, was employed to capture and visualize the microstructures of JD-1K and JD-2K. To mitigate charging issues and enhance the contrast for low-density materials, an approximately 8 nm carbon coating was applied using a coater system (Leica EM ACE600 from Germany), facilitating high-resolution field emission scanning electron microscope (FESEM) imaging. To uncover the elemental distribution within the samples, energy-dispersive X-ray spectroscopy (EDS) data were gathered with Oxford Ultim Max 60 mm² EDS detectors attached to the SEM instrument at 15 kV.

3.3. Fourier Transform Infrared Spectroscopy (FTIR)

JD-1K and JD-2K were pulverized to 200 mesh to conduct the FTIR analyses at the IGGCAS. A Hyperion 2000 IR microscope equipped with a DTGS detector was interfaced with a Bruker Vertex 70 V FTIR spectrometer that incorporated a KBr beam splitter and a globar source. Spectra were recorded within the span of 400–4000 cm⁻¹, with a resolution of 4 cm⁻¹. The sample was combined with KBr in a 1:100 ratio and then examined via the transmission method after being compressed. Once the spectra had been acquired, the atmospheric compensation tool was used, and background subtraction procedures were carried out, using OPUS 9.0 software.

3.4. Transmission Electron Microscopy (TEM)

To obtain nanometer-scale morphological, structural, and chemical characteristics, Transmission Electron Microscopy (TEM) experiments were conducted. Two TEM foils of JD-1K and JD-2K were fabricated using a Thermo Fisher Scientific Helios G4Ux focused ion beam (FIB) dual beam system. During the ion milling process, an accelerating voltage ranging from 2 to 30 kV and diverse beam currents (from 20 pA to 2 nA) were employed, resulting in final foils with a thickness of approximately 150 nm. The detailed preparation protocol can be found in [30]. The microtextural and mineralogical features of these two foils were examined under a JEM Thermo Fisher Scientific Talos f200s. All TEM observations were carried out at an acceleration voltage of 200 kV. Conventional Bright-Field (BF) TEM and High-Angle Annular Dark-Field (HAADF) imaging were utilized to depict the petrographic textures of the TEM foils. Selected-Area Electron Diffraction (SAED) and High-Resolution TEM (HRTEM) imaging were employed to ascertain the mineral structures. Additionally, Energy-Dispersive (EDS) X-ray spectroscopy was applied to determine the chemical compositions of the minerals.

4. Age Results

4.1. ⁴⁰Ar/³⁹Ar Ages

Comprehensive summaries of the ⁴⁰Ar/³⁹Ar age data for K-feldspars (JD-1K and JD-2K) and biotite (JD-2B) are detailed in Supplementary Table S1, while the corresponding age spectra and inverse isochron diagrams are illustrated in Figure 3. The uncertainties associated with the plateau and inverse isochron ages are reported at the 2σ confidence level.

The K-feldspar JD-1K was extracted from sample JD-1. It exhibited a distinct and well-defined age spectrum. Thirteen consecutive laser-heating steps released >99% of total ³⁹Ar (Figure 3A), yielding a plateau age of 122.12 ± 0.81 Ma (MSWD = 1.3). This plateau age is in close agreement with the inverse isochron age (shown in Figure 3a as 122.21 ± 0.87 Ma with an MSWD of 1.4). Notably, all the selected steps contained more than 99% radiogenic ⁴⁰Ar, suggesting minimal air contamination (refer to Supplementary Table S1). Owing to the low atmospheric argon content in the sample, the data points on the inverse isochron diagram (Figure 3a) cluster closely around the X-axis (³⁹Ar/⁴⁰Ar), leading to a relatively large error in the initial argon value (⁴⁰Ar/³⁶Ar_i = 241.1 ± 122.0). Nevertheless, this substantial error does not undermine the reliability and precision of the age determination.

For the K-feldspar JD-2K sourced from JD-2, sixteen heating steps, which released more than 87% of ³⁹Ar, defined a plateau age of 99.83 ± 0.73 Ma with an MSWD of 1.1 (illustrated in Figure 3B). The reverse isochron diagram yielded a reverse isochron age of 100.35 ± 0.88 Ma (depicted in Figure 3b, with an MSWD of 0.9), consistent with the plateau age and indicating an initial ⁴⁰Ar/³⁶Ar ratio of 280.5 ± 17.7. This ratio implies that the sample did not possess significant excess argon (as shown in Figure 3b). In comparison to the Ar release pattern of JD-1K, JD-2K manifested two distinctive traits: 1) the proportion of ³⁹Ar_K released during the low-temperature stage was relatively large, reaching up to 27%; 2) JD-2K exhibited a relatively high yield of 40Ar from the atmosphere, and as the temperature increased, the fraction of atmospheric ⁴⁰Ar gradually augmented (refer to Figure 3 and Supplementary Table S1).

Biotite JD-2B, derived from Sample JD-2, yielded a relatively flat age spectrum, with approximately 82% of the ³⁹Ar being released (as shown in Figure 3C). This led to a plateau age of 120.39 ± 0.79 Ma (with an MSWD of 1.7) and a congruent inverse isochron age of 121.25 ± 1.04 Ma (with an MSWD of 1.2, as illustrated in Figure 3c). The initial ⁴⁰Ar/³⁶Ar ratio of this sample, calculated as 269.1 ± 24.5, does not significantly deviate from the atmospheric value when considering the associated uncertainty. This suggests that no excessive ⁴⁰Ar was entrapped within the sample.

4.2. U-Pb Age

Twelve zircon grains from sample JD-1 were analyzed. Complete U-Pb data are listed in Supplementary Table S2, and concordia diagrams and weighted ages are shown in Figure 4. The selected zircons are euhedral to subhedral. In the cathodoluminescence images, the zircons show good crystal morphology and clear oscillatory zoning, as is the case for magmatic zircon (Figure 4). In addition, these zircons yield Th/U ratios of 0.95–2.29. We obtained a zircon U–Pb concordia age of 121.6 ± 1.4 Ma (MSWD = 1.9, n = 12, Figure 4a). The zircons also yield an average ²⁰⁶Pb/²³⁸U age of 121.8 ± 1.4 Ma (MSWD = 0.61, n = 12, Figure 4b), which is consistent with the concordia age.

4.3. Age Comparison Between JD-1K and JD-2K

The zircon U-Pb age of JD-1Z (121.6 ± 1.4 Ma) is consistent with the ⁴⁰Ar/³⁹Ar age of K-feldspar JD-1K (122.12 ± 0.81 Ma). In contrast, the ⁴⁰Ar/³⁹Ar ages of K-feldspar JD-2K and biotite JD-2B from sample JD-2 are 99.83 ± 0.73 Ma and 120.39 ± 0.79 Ma, respectively, indicating a discrepancy of over 20 Ma. According to a previous study [19], the eruption age of this volcanic suite is estimated to be approximately 119–122 Ma. Therefore, the ⁴⁰Ar/³⁹Ar ages of K-feldspar JD-1K from sample JD-1 and biotite from sample JD-2 provide reliable constraints on the eruption age, whereas the ⁴⁰Ar/³⁹Ar age of K-feldspar JD-2K appears significantly younger. Notably, the ⁴⁰Ar/³⁹Ar results for JD-2K exhibit both a consistent plateau age and inverse isochron age, with no evidence of significant excess argon, suggesting a plausible but potentially misleading age. Without corroborating data from other minerals or isotopic systems, it is easy to misinterpret the geological significance of this age. Both thermal events experienced by the minerals and their intrinsic structural characteristics play crucial roles in determining their ages. The subsequent section will thoroughly investigate these factors to explore possible explanations.

5. Composition and Structure of JD-1K and JD-2K

The composition, microstructure, and degree of order of the two K-feldspar samples, JD-1K and JD-2K, were characterized using optical microscopy, scanning electron microscopy-energy dispersive spectroscopy (SEM-EDS), Fourier transform infrared spectroscopy (FTIR), and transmission electron microscopy (TEM) to investigate the underlying reasons for this phenomenon.

5.1. Optical Microscopy and SEM-EDS

Optical micrographs of samples JD-1 and JD-2 are presented in Figure 5A,C. In sample JD-1, the minerals display a pristine polished surface, and notably, there are neither significant internal fractures nor impurities. Conversely, for sample JD-2, the mineral edges appear damaged, and substantial internal impurities are evident.

Backscattered electron (BSE) images and energy-dispersive X-ray spectroscopy (EDS) spectra of JD-1K and JD-2K are illustrated in Figure 5. BSE and EDS analyses indicated that JD-1K represents a pure K-feldspar, exhibiting homogeneity in both chemical composition and microstructure (Figure 5B,a,b). In stark contrast, the composition and structure of JD-2K prove to be far more intricate. BSE images of JD-2K reveal that the interior of the mineral lacks uniformity and is partitioned into distinct dark and bright zones (Figure 5D). Via EDS analysis, it was determined that the principal constituent of the bright area is K-feldspar, while the dominant component of the dark area is albite (Figure 5c,d). K-feldspar serves as the main crystal phase, with albite assuming an irregular dendritic or reticular form and dispersing along the edges or cracks of the K-feldspar. Consequently, the JD-2K feldspar manifests characteristics suggestive of perthite.

5.2. FTIR

The infrared absorption spectroscopy approach represents the most sensitive means of ascertaining the degree of order of K-feldspar. It can accurately detect any fluctuations in the degree of order of K-feldspar, remaining unaffected by its symmetry characteristics [31,32,33,34,35]. Fourier transform infrared spectroscopy (FTIR) analyses were implemented to explore the degree of order and triclinicity of JD-1K and JD-2K, which has the potential to furnish significant evidence concerning the origin of rocks, as documented in prior research [31,36,37,38,39]. The detailed raw data are presented in Supplementary Table S3.

The formula for computing the degree of order is θ = 0.05 (Δv − 90), where θ designates the degree of order and Δv stands for the difference between the wavenumber values of the two maximum absorption bands lying within the ranges of 500–550 cm⁻¹ and 600–650 cm⁻¹, which can be directly retrieved from the infrared spectrum [31,35,40]. For JD-1K, the wavenumbers of the two maximum absorption bands within the aforesaid ranges are 546.720 and 637.734 cm⁻¹ respectively, while for JD-2K, they are 538.671 and 645.096 cm⁻¹, respectively (as shown in Figure 6). Accordingly, the degree of order (θ) of JD-1K is 0.05, whereas that of JD-2K is 0.82. We derive the triclinicities of the minerals based on the degree of order. The formula is Δ_{triclinicities} = 2 × (θ_{degree of order} − 0.5). The triclinicity of JD-1K is negative, and that of JD-2K is 0.64.

5.3. TEM

BF, HAADF, SAED, HRTEM, and EDS images of the JD-1K and JD-2K obtained through TEM analysis are presented in Figure 7. EDS elemental mapping of K, Al, Si, and O in JD-1K demonstrates its homogeneity at the nanoscale (Figure 7B). As stated in previous studies [41,42,43], SAED images possess the ability to reflect the eigenstate of matter. The SAED pattern of sample JD-1K exhibits an array of diffraction spots, signifying that this potassium feldspar has a single crystal phase (Figure 7D,E). The observation of a rippled morphology within JD-1K, as revealed by high-resolution imaging, can indeed indicate strain accumulation linked to rapid cooling (Figure 7F). For sample JD-2K, the coexisting regions of albite and potassium feldspar were chosen to prepare the TEM foil for energy spectrum analysis. Elemental mapping of Na and K indicates that the upper portion of the selected area is albite and the lower portion is potassium feldspar (Figure 7H,I). The electron diffraction pattern of albite appears as a concentric halo ring, suggesting that it is amorphous; nevertheless, some cryptocrystals are also detected within the amorphous albite (Figure 7K, k). The electron diffraction pattern of potassium feldspar shows an array of diffraction spots, indicating that it is a single crystal, similar to JD-1K (Figure 7L).

6. Discussion

6.1. Different Genesis of K-Feldspars JD-1K and JD-2K

On the basis of the infrared absorption spectrum, it was ascertained that the degree of order (θ) and triclinicity (Δ) of JD-1K amounted to 0.05 and −0.9 respectively, whereas those of JD-2K were 0.82 and 0.64, respectively. SEM and TEM experiments revealed that K, Na, Al, Si, and O were uniformly dispersed within the JD-1K crystal. In contrast, the distribution of K and Na in JD-2K was non-uniform, with small quantities of albite intermingled with potassium feldspar. These findings indicated that JD-1K constituted a highly disordered, homogeneous, and monoclinic sanidine, while JD-2K represented a relatively ordered and triclinic perthite.

The origin of sanidine JD-1K was comparatively straightforward. Magma laden with a solid solution of KAlSi₃O₈ (Or) and NaAlSi₃O₈ (Ab), which were miscible in arbitrary proportions, breached the overlying surrounding rock and ascended. These solid solutions underwent rapid cooling as the magma erupted to the surface, giving rise to the high-temperature sanidine JD-1K. The disordered arrangement of Al and Si atoms within the crystal lattice and the manifestation of ripple strain in HRTEM images both corroborated the occurrence of rapid cooling. Moreover, JD-1K maintained its original composition and structure unaltered subsequent to its formation. In comparison, the formation of JD-2K involved a polyphase geological evolution encompassing magmatic crystallization and hydrothermal alteration, as indicated by textural evidence. If the observed features resulted from exsolution, the lamellar structures formed by Na-K interdiffusion within the continuous framework of aluminosilicate tetrahedra (i.e., single crystals) generated during magmatic crystallization would have maintained their solid-state characteristics. Such exsolution textures are typically constrained to develop exclusively along specific crystallographic orientations with highly regular boundary morphologies. However, the irregular albite veins embedded within the K-feldspar matrix of sample JD-2K should be attributed to dissolution–reprecipitation processes rather than exsolution structures. This interpretation is fully consistent with the findings of Parsons’ research team on analogous systems. Furthermore, petrographic evidence for intense metasomatic alteration of the parental alkali feldspar, as clearly demonstrated in Figure 5C, provides robust support for this conclusion.

6.2. Differential Gas Release Patterns During Step-Heating

Based on the acquired age plateau and inverse isochron spectra, the argon gases in samples JD-1K and JD-2K exhibited distinct release characteristics. Firstly, the ³⁹Ar_K in JD-1K was predominantly released at medium to high temperatures, whereas JD-2K witnessed the greatest accumulation of ³⁹Ar_K during the initial low-temperature stage (Figure 8a,c). ³⁹Ar_K represents the ³⁹Ar generated from ³⁹K upon neutron irradiation and is released through step heating. The inhomogeneous distribution of potassium within the mineral and the presence of microstructures constitute the two primary factors responsible for the substantial release of ³⁹Ar_K at low temperatures. If this phenomenon were attributable to the potassium distribution, the potassium content at the grain edge should be significantly higher than that at the core, leading to a large proportion of ³⁹Ar_K in the first step. However, according to the compositional analysis of JD-2K, the potassium content at the grain edge is not elevated, and albite frequently develops at the edge (Figure 5). Consequently, the most probable cause for the significant release of ³⁹Ar_K in JD-2K during the low-temperature stage is that the complex microstructures result in a diminished argon retention capacity.

Secondly, the proportion of radiogenic ⁴⁰Ar, namely ⁴⁰Ar_R, in JD-1K exceeds 99% in nearly all heating steps, while the proportion of atmospheric ⁴⁰Ar, or ⁴⁰Ar_A, in JD-2K is relatively high (Figure 8b,d). To mitigate the impact of airborne argon, we eliminated the atmospheric argon adhering to the sample surface prior to analysis. JD-1K, possessing a homogeneous composition and structure, demonstrated a consistently high and stable proportion of ⁴⁰Ar_R throughout the heating process (Figure 8b). In contrast, as the temperature increased, the proportion of ⁴⁰Ar_R in JD-2K decreased from 99% to 70%, and the proportion of ⁴⁰Ar_A increased from 1% to 30% (Figure 8d). One possible explanation for this phenomenon is that a substantial amount of ⁴⁰Ar_A was entrapped within the intricate microstructures, and this argon could not be removed during the initial degassing procedure and was gradually released as the temperature rose. Undoubtedly, microstructures play a crucial role in the release of argon within minerals. Therefore, samples that exhibit significant argon release at low temperature and a high percentage of ⁴⁰Ar_R at high temperature warrant particular attention.

6.3. The Relationship Between Hydrothermal Processes and Age Spectra

The notion that potassium feldspar exhibits a notably low Ar closure temperature has gained extensive acceptance, typically ranging from 150 to 300 °C, as reported by [22,44,45]. Consequently, potassium feldspar is far more vulnerable to perturbations induced by thermal events, which can trigger opening of the Ar closed system and subsequently impact the configuration of the age spectrum.

Sanidine JD-1K, a high-temperature feldspar formed during rapid cooling in volcanic rocks, encompasses homogeneous chemical composition and microstructures. Given that this sample has not been interfered with by external factors, its age spectrum presents an extremely flat form and accurately indicates the time when the volcanic eruption took place. In contrast, perthite JD-2K exhibits distinct alteration characteristics based on the above experimental results. However, JD-2K yielded a flat age spectrum throughout the low-temperature to high-temperature step sequence, remarkably similar to the spectra produced by an undisturbed crystal following initial crystallization and rapid cooling. Generally, minerals that have undergone hydrothermal alteration after their initial crystallization tends to display their irregular age spectrum morphology with an undulating outline. Argon situated at the edge of the crystal is initially lost and then progressively permeates towards the core. During an incremental heating experiment, such disparities are manifested by variations in the ⁴⁰Ar*/³⁹Ar_K ratio computed from the successively released gas fractions, culminating in a non-flat age spectrum. In the ladder-like age spectra, the ⁴⁰Ar/³⁹Ar age for the gas released in the first step of the experiment signified the time of the thermal event, while the plateau age measured based on the gas released at the highest temperature furnished a minimum age for the original crystallization. Through systematic experimental investigations and theoretical modeling, ref. [46] clearly demonstrated that fluid-induced recrystallization can reset mineral isotope systematics via a dissolution-reprecipitation mechanism. This process involves the removal of indigenous argon and the establishment of new isotope systematics in the neoformed mineral phases, thereby providing a critical theoretical framework for interpreting Ar/Ar age anomalies in hydrothermally altered rocks. In most instances, the thermal histories endured by minerals can be deciphered through a poorly defined plateau that was obtained by step-heating analysis. The hydrothermal K-feldspar samples generally yielded disturbed ⁴⁰Ar/³⁹Ar age spectra. The age spectra either show significantly younger ages in the low-temperature stage or have the characteristic of large age errors in the high-temperature stage [47,48]. However, the JD-2K sample exhibits mineralogical traits modified by hydrothermal processes, yet concurrently presents a flat age spectrum suggestive of a closed system. This phenomenon is not commonly observed in geological systems. In accordance with the findings of [49,50], the argon previously stored within the JD-2K feldspar was entirely eliminated by post-magmatic hydrothermal fluids that induced full permeability in the isotope system. Subsequent cooling led to the re-establishment of a closed isotopic system, thereby resetting the geochronological clock.

Consequently, hydrothermally altered minerals may exhibit not only complex morphologies in their Ar age spectra, but also highly flattened profiles analogous to those presented in this study. To avert misinterpretations of age significance, it is imperative to employ certain tools to analyze composition, microstructure, and other characteristics prior to dating. This will enable a more accurate determination of the meaning of the obtained age. It should be emphasized that this work offers a practical illustration for discussions concerning the factors governing the reliability of ⁴⁰Ar/³⁹Ar dating using alkali feldspar and the importance of comprehending Ar release mechanisms within different feldspars. Additional experimental efforts are requisite to fathom the relationships among microstructure, alteration processes, and age spectra.

7. Conclusions

(1)

The ⁴⁰Ar/³⁹Ar ages obtained for the K-feldspar JD-1K and the biotite JD-2B, along with the U-Pb age of the zircon JD-1Z, exhibit consistence within the error margins. In contrast, the ⁴⁰Ar/³⁹Ar age spectrum of the JD-2K sample displays a flat profile, and its determined age is more than 20 Ma younger than the aforementioned samples.

(2)

Detailed petrological analyses reveal that sample JD-1K exhibits an exceptionally uniform microstructure and chemical composition, consistent with magmatic crystallization under closed-system conditions. In marked contrast, JD-2K displays unambiguous evidence of hydrothermal alteration, including dissolution–reprecipitation textures, which have resulted in a complex microstructural hierarchy.

(3)

The ⁴⁰Ar/³⁹Ar age data for sample JD-2K highlight a critical limitation in interpreting flattened age spectra: such profiles should not be interpreted as definitive evidence of closed isotope systematics since crystallization.

(4)

Hydrothermally induced recrystallization can generate apparently undisturbed ⁴⁰Ar/³⁹Ar age plateaus even in samples undergoing significant alteration. This finding provides evidence that flat ⁴⁰Ar/³⁹Ar age spectra may result from fluid-assisted isotopic resetting rather than primary magmatic processes.

(5)

During step-heating experiments, it is necessary to pay close attention to the compositional ratios of radiogenic ⁴⁰Ar, ³⁹Ar_K generated by irradiation, and ⁴⁰Ar derived from the atmosphere released from different stages. Different release patterns of Ar can serve as indicators for evaluating the characteristics of the microstructures within samples.

(6)

Moreover, prudence must be exercised when deciphering the geological implications of ⁴⁰Ar/³⁹Ar age spectra. Comprehensive mineralogical analysis and cross-verification using multiple dating techniques are of paramount importance.