Abstract
Quantifying the decomposition of soil organic carbon (SOC) fractions under climate change is essential to predict carbon-climate feedbacks. The accuracy and utility of a combustion module coupled with cavity ring-down spectroscopy (CM-CRDS) system were assessed for simultaneously determining SOC and δ13C-SOC. Using a range of standard materials as well as soil samples, we compared the results of the CM-CRDS system with those from other systems for determining C content and δ13C value. The CM-CRDS system can determine a vast range of δ13C values from −7.639‰ to −34.318‰. The δ13C values measured at C content > 0.2 mg C, corresponding to 1000 ppmv of CO2, were relatively stable. However, below a content of 0.2 mg C, the δ13C values appeared unsteady and seemed to be affected by background signal. We found that, with the increase of C content, the recovery rates (RRs) for soil samples also increased. On the contrary, the RRs for inorganic materials were much lower than organic material and soil samples. Overall, the CM-CRDS system provides a valid alternative method to determine SOC and δ13C-SOC for a sample simultaneously.
1. Introduction
Quantifying soil organic carbon (SOC) decomposition under climate change is essential to predict carbon-climate feedbacks and global C cycle [1, 2]. According to turnover times, SOC stocks can be divided into three fractions: annual cycling (aSOC), decadal cycling (dSOC), and millennial cycling, comprising of 0–5%, 60–85%, and 10–40% of total SOC stocks, respectively [3]. The approaches for investigating SOC fraction decomposition include improved measurement systems and labeling experiments using carbon isotope tracers (13C or 14C). Soils enriched with 14C (“bomb carbon”) [4] or 13C (natural 13C tracer) [5] are used to investigate how climate change affects SOC decomposition in terrestrial ecosystems. Natural 13C tracer studies take advantage of the differences of δ13C value between soil and plant, or soils had experienced a period of C3-C4 (or vice versa) vegetation switch (e.g., continuous paddy (C3 ≈ −26‰) cropping of fields previously dominated by C input from C4 vegetation ≈ −14‰).
Previous studies have tested δ13C-SOC by elemental analyzer coupled with isotope ratio mass spectrometry (IRMS) [6], gas chromatography (GC) coupled with IRMS [6], or nuclear magnetic resonance (NMR) spectrometry [7]. SOC is traditionally measured using either an elemental analyzer [8] or by the chromic acid titration method [9]. CM-CRDS system has been recently developed for isotope test in soil, plant, rock, and so forth. Onac et al. measured guano samples to explore guano-derived δ13C-based paleohydroclimate record [10]. Li et al. analyzed leaf stable C isotope composition by the CM-CRDS system with 0.2–0.3‰ precision [11]. Petrillo et al. measured the total C content and the δ13C isotopic ratio of wood powder for decomposition of coarse woody debris about different elevation [12]. Qiu et al. used CM-CRDS to measure the δ13C isotope in C3 and C4 plant for calculating plant water use efficiency [13]. Rossier et al. used CM-CRDS to measure the dried residue of wine to assess the production method of sparkling wine [14]. Burud et al. analyzed soil C content and δ13C signature for calculating biochar mass percentage [15]. Shi et al. discussed the effect of doubled CO2 concentration on the accumulation of photosynthate in Lycium barbarum by CM-CRDS [16]. Hayes et al. tested total C and δ13C of mulch films to ensure the effect of weathering conditions on the physicochemical properties of biodegradable plastic mulches [17]. To date, few studies have reported the application of CM-CRDS system for simultaneous measurement of SOC and δ13C-SOC in soil samples.
In this study, combustion module (CM) coupled with CRDS system was explored as an alternative method for making bulk measurements of SOC and δ13C-SOC. CM-CRDS system provides a method that combines information about the quantity and isotopic signature of SOC for a sample in one measure. With this method all organic carbon in samples is converted into CO2 by 1600–1800°C combustion, avoiding 12C versus 13C fractionation effects due to derivatization processes. We tested the CO2 conversion efficiency of the combustion using different organic standards with different molecular structure and stability. The recovery rates for standard substances were also determined. Finally, the method was applied to natural samples for SOC and δ13C-SOC, obtained with CM-CRDS system, and was compared with the results from other techniques.
2. Methods and Materials
2.1. Soil Sampling and Preparation
The surface soil samples (0–20 cm) that had experienced a period of C3-C4 (or vice versa) vegetation switch were collected (Table 1). The samples were passed through a 2 mm sieve, thoroughly homogenized, and air-dried in the field. Visible roots and stones were carefully removed and milled for 5 min prior to the measure.
2.2. CM-CRDS System
A combustion module (CM) coupled to CRDS system (G2131-i Analyzer, Picarro Inc., USA) was used for SOC and δ13C-SOC (Figure 1). The system was controlled by Picarro G2000 iCO2 host software. Samples in tin capsules were loaded by an autosampler (Costech, USA) into the combustion module (980°C). The CO2 collected by Picarro Liaison™ A0301 interface is inputted into CRDS for analysis.
Acetanilide (C8H9NO, C 71.09%, N 10.36%) and atropine (C17H23NO3, C 70.56%, N 4.84%) were from Costech Analytical Technologies Inc., USA, and used as standards of carbon content. Glycine (C2H5NO2, δ13C = −33.3‰) used as an isotopic reference material was purchased from the National Institute of Standards and Technology (Gaithersburg, MD, USA). Calcium carbonate (CaCO3, ≥99.9% p.a.) and sodium bicarbonate (NaHCO3, ≥99.5% p.a.) were purchased from Kermel Chemical Reagents, China.
2.3. Comparison of Measurement Systems
Elemental analyser (Vario ELIII, Germany) was used to measure the SOC. Standard soil (carbon content = 2.01%) was procured from Starplex Scientific Inc. (Etobicoke, Canada). Acetanilide (C8H9NO, C 71.09%, N 10.36%,) and atropine (C17H23NO3, C 70.56%, N 4.84%,) were obtained from Costech Analytical Technologies Inc. (Valencia, California, USA) and used as standards of carbon content.
The elemental analyser (Flash EA 1112) coupled online via a ConFlo III interface with a Deltaplus XP isotope ratio mass spectrometer (EA-IRMS, Thermo Finnigan, USA) was used to determine δ13C-SOC. Urea (δ13C vsPDB = −45.380‰) and CO2 (δ13C = −29.523 ± 0.181‰) were used as working reference standards.
2.4. Data Analysis
Recovery rates (RRs) were determined as follows: where is the measured content of a sample by CRDS and is the concentration obtained with the comparison system (EA).
For isotopic measurements, the stable carbon isotope ratios are reported in delta notation expressed in per mil [19]: where is the 13C/12C ratio of the sample and is the ratio of the international VPDB standard.
3. Results and Discussion
3.1. Standard Curve
We evaluated the linearity of the CM-CRDS method by plotting the measured 12CO2 concentration against the carbon content (0.15–1.07 mg C) of acetanilide standard (Figure 2). We found a linearity between C content and 12CO2 mean concentration with , . It is indicated that the CM-CRDS system has an acceptable CO2 conversion efficiency for detecting the C content.
3.2. Quantification of Soil Organic Carbon
Four standards can be divided into three sorts, including organic (atropine), inorganic (NaHCO3 and CaCO3), and soil reference. Aliquots were used for analysis by CM-CRDS and EA. We investigated the linearity of the CM-CRDS system by comparing the measured values of different contents from comparison method. For different standards, we found the same 12CO2 peak areas at the equal C content with correlations constant of R2 > 0.99 ( for atropine, for NaHCO3, and for CaCO3). The slopes ranged from 0.966 to 1.0434 (Figure 3). The average recovery rates (RRs) for atropine, NaHCO3, and CaCO3 were 100.31%, 98.17%, and 95.29%, respectively (Table 2). We also applied CM-CRDS to reference soil. For the comparison of SOC content measured with CM-CRDS with those obtained with the EA system, we found a correlation with and a slope of 1.0434 (Figure 3). According to the values of R2 and RRs from the three sorts of standards, we found that CM-CRDS system can give us satisfactory results as comparing method.
With the increase of C content, the higher RRs for soil samples were found (Table 2). On the contrary, the RRs for inorganic materials were much lower than organic material and soil samples. It indicated that EA system is limited in comparison with the CM-CRDS system for organic and soil samples.
3.3. Isotope Values of Standard Materials and Soil Samples
The different standard materials were used to express that the CM-CRDS system can determine a vast range of δ13C values (Figure 4). We found a linear regression with a slope of 1.0603 and a . The heaviest δ13C value was determined for NaHCO3 (−7.639‰) and the lightest for acetanilide (−34.318‰). Balslev-Clausen et al. found the similar degree of correlation when they compared the δ13C values measured with CM-CRDS with values determined by a continuous flow- (CF-) IRMS system for rock samples [18].
We also found C content-dependent variations in the δ13C values for all standards materials within a C content range of 0.2–1.1 mg C (Figure 5). Below a content of 0.2 mg C, the δ13C values appeared unsteady and seemed to be affected by background signal. However, the δ13C values measured at contents > 0.2 mg C, which corresponded to 1000 ppmv of CO2, were relatively stable. Moni and Rasse recently detected C content and δ13C signature in vegetation samples by CM-CRDS [20]. However, 13C-labeled CO2 was used to simulate a leak from geologically stored CO2, which cannot represent the situation of samples in the natural 13C abundance. Ignoring the molecular structure and complexity of a sample, the combustion module (CM) facilitates the conversion of hydrocarbons to CO2 and water. Thus, heterogeneous soils can also be analyzed with this system. This verifies the isotope analysis of rock [18] and volatile liquid [21], geologically stored CO2 [20], amino acids [22], and insects [23].
To summarize, even small changes in soil CO2 released from SOC decomposition can significantly affect atmospheric CO2 concentration and global C cycle. Thus, the interest in estimating the contribution of different fractions of SOC to soil C respired under climate change in the terrestrial ecosystem has increased in recent years. The CM-CRDS system can be used to determine SOC and δ13C-SOC simultaneously. In our investigation, the CM-CRDS system was characterized by a high signal to noise and more accurate measurements. To ensure the accuracy for a wide range of CO2 concentration and reduce transient concentration response, the instrument needs to be recalibrated, and the processing software should be upgraded. Compared with EA and EA/IRMS systems, the CM-CRDS system has an equal or superior reproducibility, memory, and drift.
Conflicts of Interest
The authors declare they have no competing financial interests.
Authors’ Contributions
Dan Liu and Zhiguo Yu contributed equally to this work.
Acknowledgments
This work was supported by the National Natural Science Foundation of China (31770529, 41601090, and 41301248); Scientific and Technological Research Program of Chongqing Municipal Education Commission (KJ1601016 and KJ1733441); Chunhui Project from Education Ministry of China (Z2015133); Natural Science Foundation of Jiangsu Province under Grant BK20160950; Chongqing Municipal Key Laboratory of Institutions of Higher Education (WEPKL2016ZD-01 and WEPKL 2016ZZ-01).