12C Carbon Isotope Ratio in Carbonates

Sep 13, 2017 - This paper is the first study focused on the innovative application of 13C NMR (nuclear magnetic resonance) spectroscopy to determine t...
10 downloads 10 Views 588KB Size
Download PDF
Article Cite This: Anal. Chem. 2017, 89, 11413-11418

pubs.acs.org/ac

Determination of the 13C/12C Carbon Isotope Ratio in Carbonates and Bicarbonates by 13C NMR Spectroscopy Concetta Pironti,† Raffaele Cucciniello,*,† Federica Camin,‡ Agostino Tonon,‡ Oriana Motta,§ and Antonio Proto† †

Department of Chemistry and Biology, University of Salerno, via Giovanni Paolo II 132, 84084 Fisciano (SA), Italy Department of Food Quality and Nutrition, Research and Innovation Centre, Fondazione Edmund Mach (FEM), Via E. Mach 1, 38010 San Michele all’ Adige (TN), Italy § Department of Medicine Surgery and Dentistry “Scuola Medica Salernitana”, University of Salerno, via S. Allende 1, 84081 Baronissi (SA), Italy ‡

ABSTRACT: This paper is the first study focused on the innovative application of 13C NMR (nuclear magnetic resonance) spectroscopy to determine the bulk 13C/12C carbon isotope ratio, at natural abundance, in inorganic carbonates and bicarbonates. In the past, 13C NMR spectroscopy (irm-13C NMR) was mainly used to measure isotope ratio monitoring with the potential of conducting 13C position-specific isotope analysis of organic molecules with high precision. The reliability of the newly developed methodology for the determination of stable carbon isotope ratio was evaluated in comparison with the method chosen in the past for these measurements, i.e., isotope ratio mass spectrometry (IRMS), with very encouraging results. We determined the 13C/12C ratio of carbonates and bicarbonates (∼50−100 mg) with a precision on the order of 1‰ in the presence of a relaxation agent, such as Cr(acac)3, and CH313COONa as an internal standard. The method was first applied to soluble inorganic carbonates and bicarbonates and then extended to insoluble carbonates by converting them to Na2CO3, following a simple procedure and without observing isotopic fractionation. Here, we demonstrate that 13C NMR spectroscopy can also be successfully adopted to characterize the 13C/12C isotope ratio in inorganic carbonates and bicarbonates with applications in different fields, such as cultural heritage and geological studies.

T

he 13C/12C carbon isotope ratio (δ13C) has been recognized as a valuable chemical parameter, covering a wide range of scientific domains, such as climatology, ecology, environmental studies, biology, archeology, forensics, and food science. Stable carbon isotope analysis, in fact, provides a powerful tool to trace the source and fate of CO2 in the environment;1,2 it allows for the characterization of the geographical origin of food3 and has been recognized as a molecular marker in biological studies.4,5 Isotope ratio mass spectrometry (IRMS) represents the preferred method for analyses of the bulk 13C/12C carbon isotope ratio at natural abundance, because of the relative high accuracy (0.1 ‰) and sensitivity (up to 0.01 ‰).6,7 However, because of the increasing interest in this parameter, several alternative analytical methods to IRMS have been designed and applied in the δ13C analysis, such as mid-infrared laser spectroscopy, 8 nondispersive infrared spectrometry (NDIRS),9 and Fourier transform infrared spectrometry (FTIR),10,11 all of which offer the advantage of being less expensive and complex. Moreover, since the 1980s, SNIF 2H NMR (site-specific natural isotope fractionation deuterium nuclear magnetic resonance spectroscopy) has been used to measure the sitespecific isotope ratios of D/H at natural abundance. The © 2017 American Chemical Society

technique has been extended to a wide range of applications, such as metabolic analyses,12 climate studies,13 environmental studies,14 and food chemistry. It is also the official method applied by the OIV (Organizzazione Internazionale della Vigna e del VinoInternational Organization of Vines and Wine), as well as being adopted by the European Commission to control the addition of sugar to wine.15,16 In the last years, NMR spectroscopy has also been used to conduct 13C position-specific isotope ratio monitoring (irm-13C NMR) with a precision of better than 1 ‰.4 Bayle and coworkers applied this methodology to vanillin and identified its geographic origin on the basis of the 13C isotopic profiles.17 On the one hand, this approach offers some advantages such as the possibility to define the isotopomer composition in a target molecule, despite the determination of only bulk δ13C values for IRMS. On the other hand, irm-13C NMR is not directly linked to international standards as much as IRMS and, generally, the reduced molar fraction ( f i/Fi) of 13C is used to determine the site-specific δ13C values for each carbon in a target molecule. These values are further combined with bulk Received: June 26, 2017 Accepted: September 13, 2017 Published: September 13, 2017 11413

DOI: 10.1021/acs.analchem.7b02473 Anal. Chem. 2017, 89, 11413−11418

Article

Analytical Chemistry δ13C values obtained by IRMS. By using an internal standard, such as dimethylsufone, Bayle and co-workers determined δ13C with high accuracy.17 The research of new independent methodologies for δ13C analysis is a powerful approach that allows one to validate the accuracy obtained by using a reference method, such as, in this case, IRMS. In recent years, NMR spectroscopy has been deeply investigated as an ideal tool to study the sequestration and storage of CO2 in geologic formations and has been identified as a promising strategy to reduce the impact of greenhouse gases on global warming.18 In this scenario, Diefenbacher and co-workers19 developed a NMR probe solution to study the reaction of carbon dioxide sequestration in a water solution, at elevated pressure and temperature. Furthermore, an in situ measurement of the development of MgCO3 under CO2 sequestration-like conditions and in the presence of Mg(OH)2, as an insoluble reactive compound, was also performed.20 13C NMR spectroscopy was also favorably used as a quantitative spectroscopic method for the determination of [CO2]/ [HCO3−] and [HCO3−]/[CO32−] ratios nondestructively.21 Moreover, solid-state 13C NMR was also used as an effective tool to quantitatively distinguish and characterize magnesium carbonate phases, such as magnesite, hydromagnesite, dypingite, and nesquehonite. Results have demonstrated that NMR spectroscopy represents a valid tool for the distinction of carbonate species with small structural differences among each other.22 Thus, considering the high potential of NMR spectroscopy, in this work, we attempt to apply this technique to a new methodology to determine the bulk δ13C values, at natural abundance, in certain inorganic carbonates and bicarbonates, using sodium acetate as an internal standard. The use of 13C NMR spectroscopy to determine the bulk 13C/12C carbon isotope ratio is unprecedented, to the best of our knowledge, which could favor its applicability in several research fields. Measurements were performed, in comparison with IRMS and NDIRS, to assess the reliability of the proposed methodology.

thermogravimetric analysis (TGA), using a Netzsch TG 209 apparatus. The analyses were performed on samples with a mass of ∼10 mg placed inside an alumina crucible. The sample temperature was then increased at a heating rate of 10 °C min−1 from room temperature up to 800 °C, under an inert atmosphere of nitrogen. 2.2. Stable Carbon Isotope Ratio Analysis. The carbon isotope ratio (δ) was expressed in ‰, relative to V-PDB (Vienna-Pee Dee Belemnite), according to the following IUPAC protocol: δ=

R sample − R standard R standard

where R is the ratio between the heavier isotope and the lighter one. 2.2.1. Isotopic Analysis of Bulk Materials by Elemental Analysis/Isotope Ratio Mass Spectrometry (EA/IRMS). A Delta Plus V isotope ratio mass spectrometer (ThermoFinnigan, Bremen, Germany) equipped with a Flash EA 1112 Elemental Analyzer (ThermoFinnigan) was used to measure δ13C. The δ13C isotopic values were calculated using two homogenized in-house protein standards, which were themselves calibrated against international reference materials: Lglutamic acid USGS 40 (International Atomic Energy Agency (IAEA), Vienna, Austria), fuel oil NBS-22 (IAEA), and sugar IAEA-CH-6 for 13C/12C. The measurement uncertainty, computed using the NORDTEST,23,24 which combines the internal reproducibility with the performances results achieved in proficiency test FIT-PTS, was 0.3 ‰. The δ13C values were reported relative to V-PDB on a scale that was normalized by assigning a value of 46.6 ‰ to LSVEC lithium carbonate (IAEA). 2.2.2. NDIRS. The samples were reduced to carbon dioxide for NDIRS (nondispersive infrared spectroscopy) analysis. A portion of 100 mg of carbonate was introduced into a 10 mL glass flask, which was evacuated, and 2.5 mL of orthophosphoric acid was added, using a syringe, to produce CO2. The CO2 gas produced was collected in a specific aluminized bag. NDIRS spectroscopy was conducted by means of a HeliFANplus analyzer (Medimar srl, Milan, Italy) that was equipped with a single-beam nondispersive infrared industrial photometer. The aluminized bags were directly connected with the inlet ports of the NDIR spectrometer for sequential measurements. The NDIRS device was interfaced to a computer system that enabled the software-guided measurement and calculation of results. All the chemicals were purchased from Sigma−Aldrich (St. Louis, MO, USA). NDIRS calibration was performed by using the international standard purchased from the International Atomic Energy Agency (IAEA) (marble, δ13C = +2.5 ± 0.1 ‰). The measurement uncertainty was 0.6‰.23,24 2.2.3. 13C NMR Spectroscopy. The quantitative NMR spectra were recorded using a Bruker 600 system, with a probe accepting 10 mm outer diameter (o.d.) tubes. The sample (0.1000 g), CH313COONa (used as an internal standard (0.0100 g)), and the accurately weighed relaxation reagent Cr(acac)3 (0.0050 g), were added to 0.5 mL of deuterated solvent (D2O) in order to lock the field to the frequency of the spectrometer. Gated decoupling techniques were applied in order to obtain quantitative results. The pulse angle was set at 90° and the pulse intervals (D) were selected (D > 3T1,max),

2. MATERIALS AND METHODS 2.1. Materials and Sample Preparation. Sodium carbonate (Na2CO3), potassium carbonate (K2CO3), cesium carbonate (Cs2CO3), calcium carbonate (CaCO3), ammonium carbonate ((NH 4 ) 2 CO 3 ), sodium hydrogen carbonate (NaHCO3), potassium hydrogen carbonate (KHCO3), ammonium hydrogen carbonate (NH5CO3), chromium acetylacetonate (Cr(acac)3), and sodium acetate (CH313COONa) were purchased from Sigma−Aldrich (St. Louis, MO, USA). Oxalic acid (international standard) was purchased from the International Atomic Energy Agency (IAEA) and had the following specifications: IAEA-C8, oxalic acid, δ13C = −18.3 ± 0.2. CaCO3 was converted in water-soluble Na2CO3 for 13C NMR analysis, according to the following procedure: a portion of 2 g of CaCO3 was introduced into a 10 mL glass flask, which was evacuated, and 5 mL of orthophosphoric acid was added, using a syringe, to produce CO2. The obtained CO2 gas was collected in an impinger, filled with 20 mL of a saturated solution of NaCl and 50 mL of a NH3 solution at 30%−33% (w/w); then, it was quantitatively converted in NaHCO3 and Na2CO3. NaHCO3 and Na2CO3 were removed by filtration, washed with distilled water, and dried under vacuum at 300 °C for 2 h to convert NaHCO3 in Na2CO3, following the wellknown Solvay method. Na 2CO3 was characterized via 11414

DOI: 10.1021/acs.analchem.7b02473 Anal. Chem. 2017, 89, 11413−11418

Article

Analytical Chemistry based on the longitudinal relaxation times (T1 ), first determined by the inversion recovery method. Typical values of the experimental parameters were as follows: spectrum width (SW), 3000 MHz; pulse width (PW, for a pulse angle of 90°), 12 μs; memory size (SI), 32K; delay time, 150 s; temperature, 300 K; zero filling (Z), 32K; number of transients (NS), 70; and number of experiments per sample (NE), 4. The analysis time changes in the range of 1−2 h, based on the chemical difference of the compounds investigated.

3. RESULTS AND DISCUSSION 3.1. Method Optimization for 13C NMR Analysis. The first part of our work was dedicated to optimizing the spectral parameters for 13C NMR analysis, which was necessary to enhance the precision and accuracy in determining the bulk δ13C values. Indeed, we analyzed the influence of the relaxation time (T1), the presence of the relaxation agent, the pulse intervals (D1), and the use of an internal standard to determine the δ13C in several carbonates and bicarbonates, comparing the results with those previously obtained by means of IRMS. Carbon isotopic ratio analyses were performed on 0.1 g of sample (sodium carbonate, potassium carbonate, sodium hydrogen carbonate, and ammonium hydrogen carbonate) dissolved in 1 mL of D2O. The longitudinal relaxation times of carbonates and bicarbonates were obtained by the inversion− recovery sequence, a simple two-pulse sequence that creates the initial population disturbance by inverting the spin populations through the application of a 180° pulse. As shown in Table 1, we observed that the relaxation times ranged from 16.050 s (in NaHCO3) to 47.944 s (in K2CO3).

Figure 1. 13C NMR spectra of sodium carbonate, Na2CO3 (peak denoted by a solid triangle, ▲) in the presence of CH313CO2Na (peak denoted by a solid circle, ●), with different recovery delays: (a) delay was set equal to T1,max (13C); (b) delay was set to >3T1,max (13C); and (c) delay was set to >5T1,max (13C).

determination of vanillin with high accuracy was obtained by using dimethylsufone as a standard.13 Generally, the site-specific δ13C determination by NMR spectroscopy was obtained by combining the bulk δ13C values obtained by IRMS with the molar fraction (f i/Fi) of 13C; however, determining the bulk δ13C values using NMR with high accuracy becomes interesting when using an internal standard, such as sodium acetate. To the best of our knowledge, this work is the first example of the application of NMR spectroscopy to determine the bulk isotope carbon composition, at natural abundance. Sodium acetate (CH313CO2Na) was determined to be a suitable internal standard, because of the associated relaxation time (36.16 s) and chemical shifts (181.5 ppm), which were in the same range of the carbonates and bicarbonates investigated in this work. Furthermore, by using CH313CO2Na, only the enriched carbonyl carbon was detected. The sample was prepared by dissolving CH313CO2Na, the sample, and Cr(acac)3 in D2O. Isotope carbon composition of the bulk sample was evaluated following the equations step by step and using linear calibration. First of all, it was necessary to integrate the spectra: the intensity of 13C was normalized on carbonyl carbon of sodium acetate (internal standard) such as 1 and the signal intensities of the sample were compared to the standard signal. 13 CR is the molar carbon of the internal standard, 13Csample is the molar 13C of the sample, obtained as the product of the molar carbon of the internal standard and the signal intensities of the sample (Isample). These are described in the following equations:

Table 1. Longitudinal Relaxation Times of Carbonates and Bicarbonates Obtained by the Inversion-Recovery Sequence at 600 MHz sample

chemical shift (ppm)

T1 (13C) (s)

Cs2CO3 NaHCO3 K2CO3 NH5CO3 CH313CO2Na

168.15 161.20 168.20 160.94 181.50

42.855 16.050 47.944 23.845 36.160

We evaluated the influence of the paramagnetic relaxing complex, such as Cr(acac)3, that allows one to reduce the pulse interval values from 10T1 to 3T1 and to shorten the analysis time from 2−4 h, which is generally necessary, to 1−2 h.25 The spectra obtained for sodium carbonate, using different values of pulse intervals (D1), are reported in Figure 1. Previously published reports adopted D1 = 10T1 for 13C NMR data acquisition;4,25,26 however, in our experimental conditions, we found that D1= 3T1 was the best compromise for this quantitative analysis. In fact, as Figure 1 clearly suggests, the recovery delay influenced the quantitative analysis. We set a delay of >3T1,max (13C) (spectrum a) and equal to T1,max (13C) (spectrum b) and clearly demonstrated that this value affected the determination of 13C/12C for sodium carbonate. With regard to the δ13C of Na2CO3 obtained by IRMS, we optimized the integral value (D1 > 3T1,max) until we obtained a comparable value. Once the spectral parameters were optimized, we evaluated the possibility of using an internal standard, such as sodium acetate. As has been recently reported in the literature, the δ13C

13

CR =

mass of sodium acetate (g) molar mass of sodium acetate

13

Csample = 13CR × Isample

13 Csample C = 12 C Ctot − 13Csample 13

Therefore, once the ratio 13C/12C was evaluated, the δ13C of the sample was calculated based on the linear calibration obtained by using two carbonates (Cs2CO3 and K2CO3) and two bicarbonates (NaHCO3 and NH5CO3). Figure 2 shows the 11415

DOI: 10.1021/acs.analchem.7b02473 Anal. Chem. 2017, 89, 11413−11418

Article

Analytical Chemistry

reliability of the innovative methodology was compared to IRMS and NDIRS. As highlighted in Table 3, δ13C values obtained via IRMS, 13 C NMR, and NDIRS were very close to each other, bringing about a very good agreement among techniques. A precision value of 1 ‰ was obtained with 13C NMR, which is consistent with data obtained by 13C position-specific isotope ratio monitoring.4 These results make us confident of the fact that the 13C NMR methodology could be a valuable alternative to IRMS and NDIRS for the determination of carbon isotopic composition in bulk samples, at natural abundance. Therefore, we extended NMR methodology to the analysis of isotopic composition of insoluble calcium carbonates, which are of considerable interest, because of their importance in several applications, such as environmental studies and in the field of cultural heritage.27−29 Water-insoluble carbonates were converted to soluble carbonates by following the procedure reported in the Materials and Methods section. The synthesized carbonates were characterized via TGA to ensure the quantitative conversion of NaHCO3 in Na2CO3. In addition, δ13C values of the initial insoluble carbonates, as well as those of the obtained soluble carbonates, were determined by IRMS, and the results excluded any isotopic fractionation. The obtained carbonates were then analyzed by IRMS, NDIRS, and13C NMR, and the results are reported in Table 4 as the mean values of three measurements for each sample. The results showed that NMR spectroscopy represents a valid tool to determine the carbon isotope composition of carbonate species in solution; more specifically, calcium carbonate showed a δ13C of −9.8 ± 1.0 ‰ obtained by NMR, which was very close to the data obtained by IRMS (δ13C = −9.0 ± 0.1 ‰). Positive results were also obtained for marble samples where δ13C = 1.0 ± 1.0 ‰ and 1.0 ± 0.1 ‰ by NMR and IRMS, respectively. Because of the great interest in determining the positionspecific carbon isotope ratios, at natural abundance, by NMR spectroscopy to identify the origin, authenticity, and traceability of several organic molecules in samples (some representative examples were glucose, glycerol, malic acid, and vanillin),30,31 we also tried to apply the developed methodology to small organic molecules, such as oxalic acid. The sample was prepared by dissolving 0.0045 g of CH313CO2Na, 0.120 g of oxalic acid, and 5.0 mg of Cr(acac)3 in 0.5 mL of D2O. The 13C spectrum of oxalic acid is reported in Figure 3. Taking into account that all molar carbon of sodium acetate is 13C on carbonyl carbon with an intensity of 1, the signal intensities of oxalic acid correspond to 0.5282. Based on this assumption, 13CR represents the molar carbon of the internal standard, while 13Coxalic acid is the molar 13C of oxalic acid obtained by multiplying the molar carbon of the internal

Figure 2. Linear calibration plot between the 13C/12C ratio obtained by 13C NMR and the δ13C (‰) obtained by IRMS.

calibration line obtained by plotting the average absorption intensity ratios against the reference δ13C values obtained by IRMS. As can be observed, the linear correlation was rather good (correlation coefficient of r2 = 0.9987), leading to a standard deviation of 1.0 ‰, in terms of δ13C units. Table 2. 13C NMR Chemical Shifts and Isotope Ratio Obtained by δ13C Analysisa sample

chemical shift (ppm)

T1 (13C) (s)

δ13C (‰)

NaHCO3 Cs2CO3 K2CO3 NH5CO3

161.20 168.15 168.20 160.94

16.050 42.855 47.944 23.845

−3.32 −5.20 −27.7 −43.25

C/12Ca (NMR)

13

0.01113 0.01103 0.01076 0.01058

a13

C NMR analyses were performed in the presence of 0.005 g of Cr(acac)3.

For major clarity, results are also reported in Table 2. The C/12C ratio obtained by the 13C NMR experiments was determined independently for each carbonate and bicarbonate as the mean value of four measurements for each sample. The values of δ13C (‰) of standards ranged from −3.32 ‰ (in NaHCO3) to −43.25 ‰ (in NH5CO3), which is consistent with data reported in the literature for these samples.27 3.2. Comparison among Techniques (13C NMR, IRMS, and NDIRS). Consistent with the objective of this study, four samples of carbonates and bicarbonates were analyzed to extend the choice of the developed methodology among the techniques applicable for the determination of carbon isotopic composition. Results are summarized in Table 3, and the 13

Table 3. Comparison among Techniques (IRMS, NMR, and NDIRS) for δ13C Analysis of Carbonates and Bicarbonates IRMS

NMR

NDIRS

samplea

δ13C

standard error of mean, SEM

δ13C

standard error of mean, SEM

δ13C

standard error of mean, SEM

Na2CO3 KHCO3 CsHCO3 (NH4)2CO3

−2.6 −28.9 −33.0 −45.7

0.1 0.1 0.1 0.1

−2.0 −30.0 −32.0 −46.0

1.0 1.0 1.0 1.0

−2.4 −29.7 −32.0 −45.5

0.2 0.2 0.3 0.1

a13

C NMR analyses were performed in the presence of the relaxation reagent (Cr(acac)3). 11416

DOI: 10.1021/acs.analchem.7b02473 Anal. Chem. 2017, 89, 11413−11418

Article

Analytical Chemistry Table 4. Comparison among Techniques (IRMS, NMR, and NDIRS) for δ13C Analysis of Carbonates IRMS

NMR

NDIRS

samplea

δ13C

standard error of mean, SEM

δ13C

standard error of mean, SEM

δ13C

standard error of mean, SEM

CaCO3 marble

−9.0 1.0

0.1 0.1

−9.8 1.0

1.0 1.0

−8.9 −1.0

0.2 0.3

a13

C NMR analyses were performed in the presence of the relaxation reagent (Cr(acac)3).

■

AUTHOR INFORMATION

Corresponding Author

*Tel.: +39 89969366. E-mail: [email protected]. ORCID

Raffaele Cucciniello: 0000-0002-3291-7273 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was financially supported by Fondi di Ateneo per la Ricerca di Base (No. FARB 2016), University of Salerno (Grant No. ORSA 167988), Cle.Pr.In. srl, and SNECS Databenc project (No. CUP E68C14000050005). We are grateful to Dr. Patrizia Oliva for technical assistance.

Figure 3. 13C NMR spectrum of oxalic acid (denoted by a bold cross, +) IAEA-C8 (international reference material for stable isotopes) in D2O with CH313CO2Na (denoted by a solid circle, ●) and relaxation agent Cr(acac)3.

■

(1) Proto, A.; Cucciniello, R.; Rossi, F.; Motta, O. Environ. Sci. Pollut. Res. 2014, 21, 3182−3186. (2) Goldstein, A. H.; Shaw, S. L. Chem. Rev. 2003, 103, 5025−5048. (3) Pironti, C.; Proto, A.; Camin, F.; Cucciniello, R.; Zarrella, I.; Motta, O. Talanta 2016, 160, 276−281. (4) Caytan, E.; Botosoa, E. P.; Silvestre, V.; Robins, R. J.; Akoka, S.; Remaud, G. S. Anal. Chem. 2007, 79, 8266−8269. (5) Motta, O.; De Caro, F.; Quarto, F.; Proto, A. J. Infect. 2009, 59, 90−94. (6) Fry, B. Carbon Isotope Techniques; Academic Press: San Diego, CA, 1991. (7) Danezis, G. P.; Tsagkaris, A. S.; Camin, F.; Brusic, V.; Georgiou, C. A. TrAC, Trends Anal. Chem. 2016, 85, 123−132. (8) Van Geldern, R.; Nowak, M. E.; Zimmer, M.; Szizybalski, A.; Myrttinen, A.; Barth, J. A. C.; Jost, H. Anal. Chem. 2014, 86, 12191− 12198. (9) Hildebrand, P.; Beglinger, C. Clin. Infect. Dis. 1997, 25, 1003− 1005. (10) Zanasi, R.; Alfano, D.; Scarabino, C.; Motta, O.; Viglione, R. G.; Proto, A. Anal. Chem. 2006, 78, 3080−3083. (11) Esler, M. B.; Griffith, D. W. T.; Wilson, S. R.; Steele, L. P. Anal. Chem. 2000, 72, 216−221. (12) Robins, R. J.; Pétavy, F.; Nemmaoui, Y.; Ayadi, F.; Silvestre, V.; Zhang, B.-L. J. Biol. Chem. 2008, 283, 9704−9712. (13) Augusti, A.; Betson, T. R.; Schleucher, J. Chem. Geol. 2008, 252, 1−8. (14) McKelvie, J. R.; Elsner, M.; Simpson, A. J.; Sherwood Lollar, B.; Simpson, M. J. Environ. Sci. Technol. 2010, 44, 1062−1068. (15) Reid, M. L.; O’Donnell, C. P.; Downey, G. Trends Food Sci. Technol. 2006, 17, 344−353. (16) Dordevic, N.; Camin, F.; Marianella, R. M.; Postma, G. J.; Buydens, L. M. C.; Wehrens, R. Aust. J. Grape Wine Res. 2013, 19, 324−330. (17) Bayle, K.; Grand, M.; Chaintreau, A.; Robins, R. J.; Fieber, W.; Sommer, H.; Akoka, S.; Remaud, G. S. Anal. Chem. 2015, 87, 7550− 7554. (18) Hu, M. Y.; Deng, X.; Thanthiriwatte, K. S.; Jackson, V. E.; Wan, C.; Qafoku, O.; Dixon, D. A.; Felmy, A. R.; Rosso, K. M.; Hu, J. Z. Environ. Sci. Technol. 2016, 50, 12373−12384. (19) Diefenbacher, J.; Piwowarczyk, J.; Marzke, R. F. Rev. Sci. Instrum. 2011, 82, 076107.

standard with the signal intensities of the acid, as the following equations suggest: 13

CR =

0.0045 g = 5.485519329 × 10−5 mol 82.0343 g/mol

Coxalic acid = 13CR × IOA = 2.897446556 × 10−5 mol

13

13

Coxalic acid 12

C

13

=

Coxalic acid

Ctot − 13Coxalic acid

REFERENCES

= 0.01091

The δ13C of oxalic acid was calculated based on the linear calibration depicted in Figure 2, and the value obtained (δ13C = −19.0 ± 1.0 ‰) was consistent with that measured via IRMS analysis (δ13C = −18.3 ± 0.5 ‰).

4. CONCLUSIONS The application of 13C NMR spectroscopy to determine the bulk 13C/12C carbon isotope ratio, at natural abundance, in inorganic carbonates and bicarbonates with high accuracy (1.0 ‰) is unprecedented, to the best of our knowledge. Measurement parameters, such as relaxation time (T1), the presence of a relaxation agent, the pulse intervals (D1), and the use of an internal standard were optimized. Thanks to the optimization of the experimental parameters, analysis time was reduced in the range of 1−2 h, based on the chemical difference of the compounds investigated. The δ13C values obtained by 13C NMR were consistent with those obtained by other techniques (IRMS and NDIRS), both for carbonates/bicarbonates and oxalic acid. In summary, this work shows that 13C NMR methodology can be used as a valuable alternative to IRMS and NDIRS for δ13C analysis of carbonate and bicarbonate matrices, extending the choice of techniques applicable for the determination of carbon isotopic composition. 11417

DOI: 10.1021/acs.analchem.7b02473 Anal. Chem. 2017, 89, 11413−11418

Article

Analytical Chemistry (20) Surface, J. A.; Skemer, P.; Hayes, S. E.; Conradi, M. S. Environ. Sci. Technol. 2013, 47, 119−125. (21) Surface, J. A.; Wang, F.; Zhu, Y.; Hayes, S. E.; Giammar, D. E.; Conradi, M. S. Environ. Sci. Technol. 2015, 49, 1631−1638. (22) Moore, J. K.; Surface, J. A.; Brenner, A.; Skemer, P.; Conradi, M. S.; Hayes, S. E. Environ. Sci. Technol. 2015, 49, 657−664. (23) Measurement uncertainty revisited: Alternative approaches to uncertainty evaluation, Technical Report No. 1/2007, EUROLAB, 2007 (www.eurolab.org). (24) Guide to the evaluation of measurement uncertainty for quantitative tests results, Technical Report No. 1/2006, EUROLAB, 2006 (www. eurolab.org). (25) Caer, V.; Trierweiler, M.; Martin, G. J.; Martin, M. L. Anal. Chem. 1991, 63, 2306−2313. (26) Caytan, E.; Botosoa, E. P.; Silvestre, V.; Robins, R. J.; Akoka, S.; Remaud, G. S. Anal. Chem. 2007, 79, 8266−8269. (27) Craig, H. Geochim. Cosmochim. Acta 1953, 3, 53−92. (28) Cucciniello, R.; Proto, A.; Rossi, F.; Motta, O. Atmos. Environ. 2013, 79, 666−671. (29) Guerranti, C.; Benetti, F.; Cucciniello, R.; Damiani, D.; Perra, G.; Proto, A.; Rossi, F.; Marchettini, N. Atmos. Pollut. Res. 2016, 7, 754−761. (30) Höhener, P.; Silvestre, V.; Lefrançois, A.; Loquet, D.; Botosoa, E. P.; Robins, R. J.; Remaud, G. S. Chemosphere 2012, 87, 445−452. (31) Gilbert, A.; Silvestre, V.; Robins, R. J.; Tcherkez, G.; Remaud, G. S. New Phytol. 2011, 191, 579−588.

11418

DOI: 10.1021/acs.analchem.7b02473 Anal. Chem. 2017, 89, 11413−11418