Lithium speciation and bioaccessibility in wheat, durum
and barley flour
Camilla Sundqvist
Independent work for bachelor exam in chemistry, KEO15G, VT19 Supervisor: Viktor Sjöberg
Examiner: Michaela Zeiner Date: 2019-06-03
Abstract
In this study, the total as well as the water-soluble and bioaccessible fractions of lithium in wheat, durum and barley flour were evaluated together with the isotope ratio. Lithium was determined by inductively coupled plasma – mass spectrometry (ICP-MS). The total lithium content was determined after an open-vessel digestion in HNO3. The water-soluble fraction was evaluated by extraction with deionized water. The bioaccessibility was determined in vitro under simulated stomach digestion conditions. The total lithium content in wheat ranged from 0.91 ng/g to 7.8 ng/g, in durum flour from 1.7 ng/g to 5.2 ng/g and for barley flour from 2.0 ng/g to 6.0 ng/g. For the water-soluble fraction, the ranges were for wheat flour 0.33 ng/g to 4.8 ng/g, for durum flour 0.83 ng/g to 2.5 ng/g and for barley flour 0.81 ng/g to 3.3 ng/g. The bioaccessible lithium content ranges in wheat flour from 0.38 ng/g to 5.3 ng/g, in durum flour from 0.98 ng/g to 1.9 ng/g and in barley flour from 1.3 ng/g to 3.1 ng/g. The water and bioaccessible fractions of the majority of the barley and durum flour samples had an isotope ratio differing statistically significantly from the natural ratio (p<0.05) which indicates enrichment of the heavier isotope.
Additionally, speciation of protein bound lithium in water and simulated stomach digestion extracts from respective flours were studied. Extracts from the respective flour types were chromatographically separated and fractionated with an ammonium nitrate buffer on a (size-exclusion-chromatography) SEC-column. Thirty fractions were collected, and their lithium contents were determined by ICP-MS. Lithium was mostly eluted in fraction 14, which indicates that most of the available lithium was protein-bound.
Table of content
Introduction ... 1
Objective and aim ... 3
Experimental ... 3
Materials and methods ... 3
Total extraction ... 4
Water extraction ... 4
Artificial gastric juice extraction ... 5
ICP-MS measurements ... 5
HPLC-ICP-MS analysis ... 6
Quality control ... 6
Statistics ... 7
Calculations ... 7
Results and discussion ... 8
Total extraction ... 8
Water extraction ... 11
Artificial gastric juice extraction ... 13
Comparison between flour types and extraction ... 15
Chromatographic speciation ... 16
Conclusion ... 20
References ... 22 Appendix ... A
Introduction
Lithium (Li) is the third element after hydrogen and helium that was formed during “Big Bang”. It is the lightest of the alkali metals with a natural occurrence of 0.0007% in the earth´s crust and is found widespread in aquatic and terrestrial environments (Krebs 2006). It has two stable isotopes, 6Li and 7Li with a naturally abundance of 7.6% and 92.4%,
respectively, resulting in an isotope ratio (IR 6Li/7Li) of ~0.082 (Krebs 2006).
Since the introduction of the lithium-ion battery technique, the production of lithium has increased during the last century (Franzaring et al. 2016). The environmental distribution is assumed to change in the future through the use and disposal of lithium-containing products. Lithium in nature is not found in its elemental state but commonly in minerals of different complexity such as spodumene (LiAlSi2O6)and lepidolite (K[Li,Al]3[Al,Si]4O10[F,OH]2) as well as lithium carbonate (Li2CO3), lithium chloride (LiCl) and lithium oxide (Li2O) (Aral and Vecchio-Sadus 2008). Lithium in geogenic soils is mostly unavailable for plant uptake while lithium ions added to soil are readily mobile and taken up by plants (Robinson et al. 2018).
Lithium is taken up by most plants, but the bioaccumulation differs between species and available content and some species tolerate and accumulate more lithium than others (Franzaring et al. 2016 and Jiang et al. 2014). The content in the plant is often found to be about 3 to 4 orders of magnitude lower than in the soil (Lemarchand et al. 2010).
Impact on humans occurs mainly due to dietary intake, since lithium is also found in foodstuff. Vegetables, grains, milk, eggs and water are some of the major dietary lithium sources although the amounts vary upon location and species and range from 0.001 µg/g to 33.68 µg/g for some high accumulating species (Ammari et al. 2011, Figueroa et al. 2013, González-Weller et al. 2013, Maria et al. 2015, Mir-Marqués 2012, Noël et al. 2012, Piergiovanni et al. 1997, Spiegel et al. 2009 and Vetter 2005).
Wheat, durum and barley are three commonly consumed crops. Wheat and barley are used in bakery products such as bread while durum is commonly used for the production of pasta. Lithium contents in wheat, durum and barley crops have been investigated before on a total basis range from 0.002 µg/g to 0.088 µg/g for durum and <0.002 µg/g to 0.136 µg/g for barley with values significantly differ between sites while no reported values for wheat was found (Spiegel et al. 2009). Lithium content in different wheat species were found to range from 1.1 µg/g to 13.6 µg/g and in durum from 5.6 µg/g to 6.7 µg/g (Piergiovanni et al. 2009). Although the total content gives valuable information about differences in lithium levels depending on location and food type it does not reveal the entire nutritional information since not all ingested lithium is available for absorption in the gastrointestinal tract (Bertin et al. 2016). Speciation and fractionation studies reveal further information about its chemical form and availability for uptake in the gastrointestinal tract. Chromatographic separation of simple water extracts may provide information about free metal ions readily available for absorption through the gut wall and those bound to water-soluble proteins (Erdemir and Gucer 2016).
In vitro methods can be used to estimate the fraction of an element that is available for absorption in the gastrointestinal tract, the bioaccessibility (Stelmach et al. 2014). The
methods typically aim at mimicking the oral, gastric and small intestinal phases by mimicking physiological conditions (temperature, pH, agitation, enzymes and their composition,
chemical composition and digestion time). Although worth mentioning, due to the simplicity it is not an accurate reproduction of the dynamic in vivo conditions of the biochemical processes taking place in the gastrointestinal tract such as the digestion secretion and brush border enzymes lining the small intestine tract helping with the final nutrient digestion. In order to mimic human digestion, reliable data on the conditions e.g. pH, enzymatic activities and total volumes are needed. Gastrointestinal juices from humans have been
characterised for (Ulleberg et al. 2011), showing gastric juice to have pepsin activities ranging from 7 U/mL to 70 U/mL and a pH from 1 to 4.
Data of bioaccessible content of lithium is limited, in Sarcocornia ambigua (sea asparagus), a plant belonging to the amaranth-family, the bioaccessible lithium content were determined to approximately 75% of the total lithium content. The pepsin activity used in the artificial gastric juice is unknown and the pH given as 1.2 (Bertin et al. 2016). Bioaccessible lithium content has also been studied from the leaves of black, Earl Grey and green teas.
Bioaccessible lithium content ranged from 24% to 59% of the total lithium content at the standard infusion time of 2 min with a pepsin activity of 670 U/mL at pH 1.9 in the artificial gastric juice(Erdemir and Gucer 2018).
A relatively large mass difference between two isotopes is one of the driving mechanisms for isotopic fractionation. Isotopic fractionation of metals in plants is controlled by mechanisms such as sorption to particles, complex formation and binding to organic compounds (Wei et al. 2018). Lighter isotopes diffuse faster than heavier ones into the roots as well as free ions compared to complexed ions (Rodushkin et al. 2004). Conversely, binding to
carrier-molecules such as proteins favours the heavier isotope since it involves covalently binding (Weiss et al. 2004). For lithium, it seems that soil conditions and its IR in the soil have higher impact on the lithium IR in the vegetation than the vegetation’s ability to enrich one of the isotopes (Lemarchand et al. 2010).
Speciation of lithium in food has not been widely investigated and not yet fully understood. It would therefore be of interest to study not only the total content of lithium in some selected crops but also the speciation and bioaccessibility of lithium to assess how the digestion contribute to the availability of lithium in the gastrointestinal tract. For this purpose, three different crops have been selected; wheat, durum and barley available as flour. Also,
knowledge of the IR may provide additional information whether the large difference in mass between the two lithium isotopes leads to any isotope fractionation, within and between the different flour types.
Due to the variation of lithium content that have already been observed one can except to find variable content of lithium in the different flours in the low µg/g-range at different IR
depending on origin and treatment of the sample. It is therefore necessary to start to find an optimized method that is able to meet the objectives and aims of this study.
Objective and aim
The objectives of this study are to
(1) collect data of the total lithium content
(2) characterize the speciation of lithium (water soluble, bioaccessible and total) (3) measure the isotope ratio in different pools of lithium
(4) qualitatively evaluate the correlation between protein-bound and free lithium in different types and batches of flour.
The aims are to
(1) quantify the total lithium content in wheat, durum and barley flour
(2) quantify the pools of water soluble and bioaccessible lithium in relation to the total pool of lithium
(3) calculate the isotope ratio for the different pools of lithium
(4) verify any correlation between protein-bound and free lithium in different types and batches of flour.
Experimental
Total lithium content in all digest and extract solutions as well as isotopic ratios will be analysed with inductively coupled plasma – mass spectrometry (ICP-MS). Water-soluble lithium species will be extracted using deionized water at a suitable liquid to solid ratio. Separation of free lithium and protein bound lithium will be performed using a high-performance-liquid-chromatography (HPLC) with a size-exclusion-chromatography (SEC) column and fluorescence/UV-Vis detection followed by qualitative detection in off-line mode with ICP-MS. The content of bioavailable lithium in the gastric phase will be studied using artificial digestion systems.
Materials and methods
Commonly consumed flours of durum (n = 5), barley (n = 3) and wheat (n = 5) were
purchased from local markets in Örebro, Kumla and Hallsberg. Different batches within the respective flours were selected. Each batch of flour was handled separately and thus a total number of 13 different samples was studied, see table 1.
Table 1 the sample identity and type of each flour sample.
Sample Type Identity Sample Type Identity
1 durum 1376679 8 barley 1365584 2 durum 1375338 9 wheat 1370796 3 durum 1371839 10 wheat 1375337 4 durum 1370030 11 wheat 1374193 5 durum 1378094 12 wheat 1377238 6 barley 1373507 13 wheat 1376680 7 barley 1375692
Total extraction Method optimization
Three setups were tested to find the proper mass of flour that allowed for an accurate
quantification of the total lithium content as well as the IR with ICP-MS. In setup A), 5 mL of deionised water (18.2 MΩ) (DI) and 5 mL HNO3 (sub-boiled distilled), was added to
approximately 1 g of each sample (n = 1) in 50 mL propylene test tubes (Sarstedt). Then the samples were heated to 65 ℃ in a water bath (Grant T100) and 1 mL of 30% H2O2 (analytical reagent grade, VWR) was added in 0.1 mL aliquots. After 2 h, five 1 mL aliquots of HNO3 were added and after 4.5 h the samples were allowed to cool down and the volume was adjusted to 50 mL with DI.
In setup B) and C) a total mass of 1 g and 3 g of each sample (n = 1) was digested in 1 mL DI and 10 mL HNO3 for approximately 4.5 h at 65 ℃ in a water bath. Samples containing 3 g of flour were pre-digested in 5 mL HNO3 and 1 mL DI overnight at room temperature before digestion at 40 ℃ for 2 h. Due to gas formation all samples were allowed to react before the temperature were raised to 65 ℃. Additionally, 5 mL of HNO3 were added in aliquots of 1 mL and the volume adjusted to 50 mL with DI. Samples containing 1 g were pre-digested in 5 mL HNO3 and 1 mL DI at room temperature for 2 h before putting into the water bath at 65 ℃. After 2 h, additionally 5 mL of HNO3 were added in aliquots of 1 mL and the solution was then filled up to 50 mL with DI.
After digestion and volume adjustment, the samples from setup A were filtered through 0.2 μm syringe filters (polypropylene, VWR) using 5 mL syringes (Braun Injekt®). Samples from setup B and C were not filtered.
The optimized method was further on used to prepare three more replicates of each sample. Water extraction
Method optimization
Two setups were performed to find an appropriate mass of flour that allowed for
quantification of the water-soluble lithium content as well as the IR with ICP-MS. Also to obtain detectable signals of lithium after the chromatographic separation using the ICP-MS in offline mode. In setup D) 4 g of flour and 40 mL (L/S 10) of DI were mixed in 50 mL test tubes, shaken well and put in an ultrasonic bath (45 kHz) at room temperature (VWR,
USC500TH) for 1 h. Then, all samples were put on an end-over-end shaker (Heidolph REAX 2). After 2.5 h, all samples were centrifuged (Eppendorf FEMETT 5804) at 8000 g for 10 min. The supernatants were collected and split in three for further preparation as follows: A) acidified, B) acidified and filtered, C) filtered and acidified in 10 mL test tubes
(polypropylene, Sarstedt). All samples were acidified with conc. HNO3 (10 µL/1 mL of sample). In addition to D) a second trial, having one replicate of the respective samples, E) using 20 mL DI and 4 g of flour was tested and treated in the same way as D). Although, for this setup, no splitting of the supernatant was done, the supernatant was instead filtered into 10 mL PP-tubes and acidified.
The optimized method was further on used to prepare three replicates of each sample. Artificial gastric juice extraction
Method
Two different artificial gastric juice solutions were prepared with 0.12 g of 0.7 FIP-U/mg pepsin (1 FIP-unit = 62.5 USN units) (Sigma Aldrich) and dissolved in 50 mL PP-tubes, with approximately 6.5 mL HCl-solution, Sigma Aldrich analytical reagent, (c = 1 mol/L) and 40 mL DI. Then, HCl-solution and DI were added for a final pH of 0.9 and 1 respectively, in a total volume of 50 mL. For a final concentration of 27 U/mL of pepsin (w/w) at a pH of 2, 14 mL of DI and 6 mL of the artificial gastric juice were added to 4 g of wheat samples (resultant pH = 1) and durum and barley samples (resultant pH = 0.9) (n = 3). For sample 6 to 8 the pH was adjusted to 2.0 with HCl-solution (c = 5 mol/L).
All samples were subsequently placed in a Shake´n´Stack, Thermo Hybaid at 37 ℃ for 2 h. After this all samples were centrifuged at 8000 g for 10 min and the supernatants collected. Then, wheat and durum samples were filtered into 10 mL pp-tubes. However, the supernatant of barley samples were again centrifuged at 8000 g for 10 min, before these supernatants were collected and filtered. Samples that were not immediately analysed were stored in the freezer at -20 ℃ until analysis.
Due to different buffer capacity of the respective flour, test experiments were made to determine the amount of acid required prior the extraction.
ICP-MS analysis
Analysis of samples were performed in a clean room using an Agilent 7500cx ICP-MS. For settings see table 2.
Table 2, ICP-MS operating conditions
ICP-MS operation conditions
Radio frequency power 1500 W
Carrier gas (argon) 0.90 L/min
Make up gas (argon)
Nebulizer 0.20 L/min MicroMist
Nebulizer pump
Spray chamber 0.10 rps Scott double pass Integration time
Replicates 0.10 s Two consecutive
Calibration
Calibration stock solution Isotopes Internal standard (10 µg/L) Collision cell External Multielement VI (Merck) 6Li, 7Li, 24Mg, 88Sr 103Rh Off
HPLC-ICP-MS analysis
One flour sample from each flour type with the highest concentration of lithium was selected for the HPLC-ICP-MS analysis (HP 1100 + Agilent 7500cx). For instrument settings see tables 2 and 3. This method is a chromatographic separation used for the detection of proteins using a 50 mmol/L phosphate buffer pH 6.8. This is not a proper buffer for the ICP-MS and a 50 mmol/L ammonium nitrate buffer, purchased from VWR analytical reagent, pH 6.8 was used instead. To ensure that the latter buffer gave a similar retention time and separation of the proteins two setups were performed. In setup F) Water and gastric juice extracts were prepared as described above and diluted (1:1) with a 50 mmol/L phosphate buffer at pH 6.8 (unknown origin and purity). After dilution the samples were filtered into glass vials. Chromatographic separation was performed according to table 3.In setup G) the same extracts, left standing cold overnight were treated as F) but with 50 mmol/L ammonium nitrate buffer at pH 6.8.
Table 3 HPLC operation conditions HPLC operation
conditions
Column Agilent Bio SEC-5 (5 µm, 150 Å 7.8 x 300 mm)
Mode Isocratic
Injection volume
Mobile phase 50 µL Ammonium nitrate buffer
Flow rate 1 mL/min (total of 30 min)
Replicates Two
Detection Fluorescence (FLD Ex = 221 nm and Em = 350 nm)
For samples in setup G) chromatographic separation was performed in the same way and thirty fractions collected. For a total volume of 2 mL/fraction each sample was run twice. Also, one wheat sample was spiked with 80 µL of a lithium chloride solution (analytical reagent, Merck; γ = 0.01 g/L) for a total concentration of 40 µg/L after the addition of DI. This was followed by the same water extraction procedure as already above. This was done to evaluate how added lithium interacts with flour and to see the distribution in the collected fractions.
Quality control
Plasticware was used as much as possible through the project, whenever glassware was used, it was washed three times with diluted HNO3 and DI. All experiments were carried out in triplicates. Blanks were run parallelly with all experiments. Internal standard (103Rh) was used for signal correction during elemental analysis.
Magnesium and strontium were used to evaluate any problem with the method in the open vessel digestion.
Retention time and peak separation using ammonium nitrate as mobile phase were evaluated against the phosphate buffer.
Statistics
A one-way ANOVA p = 0.05 was used for variation in absolute content ng/g in the respective flour types.
For significant difference, paired t-test was used for IR determination (p = 0.05) and Student´s t-test at p = 0.05 for difference between to samples.
For all statistical tests, the data were assumed having a normal distribution. All statistical tests were performed using Excel, Microsoft office 365 (version 1808).
Calculations
All calculations were calculated using Excel, Microsoft office 365 (version 1808). Content in ng/g
For calculating the content in ng/g, all sample concentrations obtained from 7Li ICP-MS were blank corrected and multiplied by the respective final volume. Then the amount in µg was divided by its mass of flour and multiplied by 1000 for a total content of ng/g (wet mass). Further the standard deviation was calculated using the formula STDAV.S of the respective replicates in each sample. A mean value from the replicates in each sample was calculated and the RSD calculated by taking the standard deviation divided by the mean value of each sample.
Procedure blanks
Procedure blanks were calculated in ng/g content as described above. For blank values below the instrumental LOD, the LOD-value was used in the calculations.
IR
Due to concentrations in the low µg/L-range, i.e. between 0.1 µg/L and 1 µg/L, the
background noise impact on the calculation of the IR was significant. Consequently, variation in concentration and different ratio in 6Li/7Li both led to an observed but probably imaginary isotopic fractionation. To overcome this problem and evaluate any difference in IR, standard solutions were assumed having a natural IR. Linear regressions using standards of 0.1 µg/L and 1 µg/L vs their IR were used. This was done for each ICP-MS run. The equation obtained were then used to calculate the expected natural IR from the instrument based on sample concentrations in µg/L and to compare this with that obtained with the raw IR from the sample. For all samples the raw IR were calculated as: cps signal for 6Li divided by the cps signal for 7Li. To evaluate any significant difference between the calculated IR and the raw data IR, paired t-test was used (p = 0.05) to confirm any potential isotope fractionation. Extraction yield
The extraction yield was calculated as the mean value of total lithium content (water or artificial gastric juice extracts) divided by the mean value of the total lithium content (total extract) of each sample multiplied by 100.
Instrumental LOD
The instrumental LOD is calculated as: (three*the standard deviation of the blank –intercept) / slope of the calibration curve.
Fractionation
For mass-balance calculations, the mean value of the two first fractions were assumed the background and subtracted from all analysed fractions. Values below the instrumental LOD were rejected; <0.02 µg/L for wheat samples of the spiked water and artificial gastric juice extract and <0.01 µg/L for the remaining. The total concentration in µg/L was then calculated by multiplying the concentration by the volume (2 mL). Then the concentration in the
respective injection volume was calculated by dividing the mass by the injection volume (100 µL) and correcting for the dilution with a factor of 2.
Normalized lithium concentration
For the normalized lithium concentration, the lowest value (after subtracting of the assumed background) was added to all values given a lowest value of zero.
Results and discussion
For raw data from method optimizing, see appendix section A, for total, water and artificial gastric juice extract concentrations section B, IR raw data, section C and for mass balance section D.
Total extraction Method optimization
All three setups resulted in a jelly-like solution. In addition, a white flaky precipitate was visible in the samples from setup A) and in barley flour samples setup B). This was suggested to be the formation of silicic acid that precipitate at low pH (Karlsson 2019). A sample mass of 1 g was not enough to distinguish the signal for 6Li from the background noise but worked well for the content determination based on 7Li.Setup C)using 3 g of each sample digested in HNO3 was hence the preferable extraction method.
Optimized method
Procedure blanks had a lithium content of 0.23 ng/g. Although four replicates of each flour sample were prepared, there was some problem with the lab work leading to that only two replicates were used in the result. First, due to the relatively high mass of 3 g there was a lot of gas formation when HNO3 was added. Consequently, one replicate boiled over and could not be used. Another replicate of each sample was successfully prepared but something happened between extraction and analysis to the respective sample leading to most samples below the instrumental detection limit <0.01 µg/L.
IR
P-values from table 4 clearly show that any significant difference is seen at very low confidence interval (C.I.) and this high uncertainty indicate that the difference between the natural IR and that found in the respective samples is low. Some samples were too close to the background of the instrument for a proper IR calculation and were therefor not included. Table 4, the raw IR and the expected natural calculated IR from the respective samples and flour types (n=2) except for sample 8 and 9 n=3.
sample 6Li/7Li raw 6Li/7Li natural calc. IR p-value
1 0.13 0.12 0.31 2 0.10 0.12 0.14 3 0.13 0.13 0.54 4 0.13 0.12 0.80 5 0.14 0.13 0.31 6 0.13 0.12 0.62 7 0.13 0.12 0.44 8 0.12 0.12 0.97 9 0.11 0.13 0.75
The result from the total extraction is illustrated in figure 1 including total extraction with 1 g of flour (n = 1) and 3 g of flour (n = 2). The lithium content in durum flour samples ranges from 1.7 ng/g to 5.2 ng/g, in barley flour samples from 2.0 ng/g to 6.0 ng/g and in wheat flour samples from 0.91 ng/g to 7.8 ng/g. There is no significant variation within the durum flour samples. However, a significant variation is seen within the barley flour samples and within the wheat flour samples respectively. The elemental variation is not surprising and has been reported before (Spiegel et al. 2009 and Ammari et al. 2011) since lithium concentrations in different plant species are related to location where they grow (Ammari et al. 2011).
Figure 1, total content in durum flour (sample 1-5), barley flour (sample 6-8) and wheat flour (sample 9-13) expressed in ng/g (n = 3) 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 co nt . [ ng/g] samples
Total extraction
Visible is also that there is substantial variation within the samples with a relative standard deviation (RSD) that range from 8.8% to 60%. Most instrumental RSD are <15% and randomly distributed indicating low impact from the analysis. To evaluate if the variation within the samples was related to low content or the method, two other elements, magnesium and strontium, were determined with respect to their RSD (figure 2). The mean content from each sample of flour is plotted vs. its RSD (n = 3). No correlation between content and RSD is seen for either lithium, magnesium or strontium. This indicates that not the method is
affected. The obtained high RSD thus come from the very low concentrations found and the inhomogeneity in each sample.
Figure 2, contents for lithium (ng/g), magnesium (mg/g) and strontium (µg/g) vs. RSD n=3.
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 0 2 4 6 8 RS D (% ) cont.
Total extraction
lithium magnesium strontiumWater extraction Method optimization
After leaching at setup D) L/S 10 with DI and preparing the liquid samples according to A), B) or C) all setups resulted in similar concentrations. For setup B) the acidification caused something in the samples to precipitate making it very hard to filter all samples. For setup E) all samples contained more or less particles and made it necessary to filtrate before the analysis to avoid any risk of clogging.
Water leaching of sample 1 and 6 at L/S 10 resulted in concentrations of 0.21 µg/L and 0.15 µg/L, respectively, while leaching at L/S 5 gave slightly higher concentrations 0.27 µg/L and 0.21 µg/L, respectively. Sample 9 gave higher concentrations at L/S 5 (0.83 µg/L) than L/S 10 (0.55 µg/L). Hence the procedure using an L/S 5 followed by filtration before acidification was used throughout the study.
Optimized method
Blanks were close to or just above the detection limit of the instrument (LOD<0.01) and the procedure blank were determined to 0.082 ng/g.
IR
As can be seen from table 3, samples can be put into two group based on their p-values were a significant difference between the natural and sample IR is found at a high C.I. 95% and at low C.I.<79% with a higher degree of uncertainty. Thus, there is a significant difference indicating an enrichment of the heavier isotope in the water leaching of all durum flour and barley flour samples and wheat sample 10 while an IR closer to the natural is found in the other wheat flour samples.
Table 5 the raw IR and the expected natural calculated IR from the respective samples and flour types (n = 4). Sample 6Li/7Li raw 6Li/7Li natural calc. IR p-value
1 0.088 0.11 0.0028 2 0.085 0.11 0.0026 3 0.094 0.12 0.0011 4 0.11 0.12 0.029 5 0.11 0.12 0.0072 6 0.091 0.12 0.0026 7 0.093 0.12 0.015 8 0.073 0.097 0.020 9 0.074 0.078 0.21 10 0.11 0.13 0.015 11 0.12 0.13 0.24 12 0.12 0.13 0.33 13 0.12 0.13 0.30
Lithium content
At an L/S 5 the water-soluble lithium content in durum flour samples ranged from 0.83 ng/g to 2.5 ng/g, in barley flour samples from 0.85 ng/g to 3.3 ng/g and in wheat flour samples from 0.33 ng/g to 4.7 ng/g (compare figure 3). There is no significant variation within the durum flour samples. However, the elemental content is variable within the barley flour samples and within the wheat flour samples.
Figure 3, water extraction of durum flour (sample 1-5), barley flour (sample 6-8) and wheat flour (sample 9-13) expressed in ng/g, n=4.
The average extraction yield of lithium ranges from 30% to 62%, see table 6. Table 6, the extraction yield of each sample in %.
Sample % Durum flour (1) 42 Durum flour (2) 43 Durum flour (3) 62 Durum flour (4) 40 Durum flour (5) 37 Barley flour (6) 38 Barley flour (7) 37 Barley flour (8) 58 Wheat flour (9) 59 Wheat flour (10) 30 Wheat flour (11) 45 Wheat flour (12) 42 Wheat flour (13) 51 0 1 2 3 4 5 6 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 co nt . [ ng/ g] samples
Water extraction
Artificial gastric juice extraction Method
Barley flour samples were after the first centrifugation very viscous, with small suspended particles, and very hard to filter. The recovered liquid phase was therefore centrifuged. Upon analysis the variation in the IS signal was large, probably due to the sample matrix, see appendix section E. Even though, there tend to be small variations in concentration in the respective sample except for one were the concentration was 1 order of magnitude higher than the other and therefore excluded as outlier. Another replicate of the same sample was also excluded do to handling problem of the operator.
In general, contained all blank samples lithium close to or below the instrumental LOD (<0.01 µg/L) and the procedure blank were determined to 0.067 ng/g. However, one blank contained 0.136 µg/L lithium. This blank was prepared parallel to one replicate of wheat flour samples. All samples except for the first filtered were close to twice the concentration found in the other two replicates. A potential source of this contamination is a glassware used for cleaning of the syringe between samples in the filtration step. To evaluate this, a blank of DI from the glassware was analysed having a concentration of 0.06 µg/L lithium. Further, this replicate was prepared in another lab and since there had been no problem with contaminated blanks previously it might be connected to the lab environment. The procedure blank for these samples were determined to 0.68 ng/g.
IR
Almost all samples from the respective flour types have a significant difference (p<0.05), see table 7, all suggesting an enrichment of the heavier isotope. At a slightly higher level of uncertainty, at a C.I. of 92%, this is also true for durum flour sample 1 and wheat flour sample 10. For wheat flour sample 9 and barley flour sample 7, the p-value indicate a high degree of uncertainty thus an IR closer to the natural is expected. Sample 8 could not be evaluated. Table 7 the raw IR and the expected natural calculated IR from the respective samples (n = 3).
Sample 6Li/7Li raw IR 6Li/7Li natural calc. IR p-value
1 0.082 0.097 0.071 2 0.076 0.10 0.0022 3 0.084 0.10 0.042 4 0.082 0.11 0.0042 5 0.079 0.10 0.032 6 0.090 0.11 0.013 7 0.11 0.11 0.56 9 0.075 0.070 0.12 10 0.093 0.11 0.059 11 0.092 0.11 0.032 12 0.10 0.12 0.0030 13 0.088 0.11 0.018
Lithium content
Artificial gastric juice extracted lithium contents in durum flour samples range from 0.98 ng/g to 1.9 ng/g, in barley flour samples from 1.3 ng/g to 3.1 ng/g and in in wheat flour samples from 0.38 ng/g to 5.3 ng/g, see figure 4. The elemental variation within durum flour samples and within wheat flour samples respectively are significant. Only one replicate of barley flour samples, sample 8, was successfully analysed and a Student´s t-test was used to evaluate any difference between lithium content of sample 6 and 7 (barley flour) indicating no difference (p<0.05).
Figure 4, artificial gastric juice extraction of durum flour (sample 1-5), barley flour (sample 6-8) and wheat flour (sample 9-13) expressed in ng/g, n=3 except for one barley flour sample n=1.
The average extraction yield of lithium ranges from 24% to 67%, table 8 and did not
significantly differ from the water extraction except for sample 6 and 7 (barley flour) that has a significant higher lithium content.
Table 8, the extraction yield of each sample in %.
Sample % Durum flour (1) 46 Durum flour (2) 37 Durum flour (3) 55 Durum flour (4) 43 Durum flour (5) 37 Barley flour (6) 54 Barley flour (7) 54 Barley flour (8) 58 Wheat flour (9) 67 Wheat flour (10) 24 Wheat flour (11) 47 Wheat flour (12) 34 Wheat flour (13) 54 0 1 2 3 4 5 6 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 co nt . [ ng/ g] samples
Comparison between flour types and extraction
The readily available pool of lithium in wheat, durum and barley flours based on water leaching, artificial gastric juice and the total content is represented in figure 5. There is a significant difference in lithium content between the water and artificial gastric juice extracts of barley flour samples 6 and 7 indicating a higher content of lithium in the extract treated with artificial gastric juice. At a slightly higher degree of uncertainty (p<0.07) this difference is also seen for wheat flour sample 9. This suggest that lithium is released from non-water-soluble molecules after the digestion by artificial gastric juice.
No statistically significant difference was seen between the other samples, indicating that only a minor part of lithium is bound to molecules prone to be digested by artificial gastric juice.
Figure 5 mean ± standard deviation (ng/g) over lithium levels released from samples using different extraction methods. Total extraction (n=3), water extraction (n=4) and artificial gastric juice extraction (n=3). Sample 1-5 (durum flour), sample 6-8 (barley flour) and sample 9-13 (wheat flour).
Even if the artificial gastric juice extraction gives some information of the bioaccessible lithium content it is important to remember that the uptake occurs in the intestinal tract and that other enzymes operate there that may release lithium from other non-water-soluble compounds. Also, that the less acidic environment in the small intestine may lead to reformation of lithium-containing compounds and changes in uptake routes (Erdemir and Gucer 2016).
The water and bioaccessible fractions from the artificial gastric juice range from 30% to 62% and from 24% to 67% in the single samples. There seems to be no trend in a higher water and bioaccessible content when comparing the different flour types. However, this content seems instead to vary within the different flour types suggesting that growing and processing conditions have a higher impact than the flour type itself.
Two samples of barley flour, 6 and 7 have almost identical lithium content in all three extractions. However, sample 6 has in the artificial gastric juice extract an IR that is
0 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 10 11 12 13 co nt en t [ ng/ g] samples
Comparison of extraction methods
significant different from the natural while sample 7 does not. This suggest that lithium is bound to different compounds in the respective flours and depending on location, season and weather have different IR in the soil. Even if it cannot be statistically evaluated there is a tendency seen for variation in the IR between the different extraction methods but also between samples in and between respective flour types.
Chromatographic speciation
The quality of the chromatographic separations using either phosphate buffer or ammonium nitrate were in good agreement. Proteins in the ammonium nitrate buffer were eluted
approximately one minute earlier with lower peaks, probably due to some protein degradation occurring overnight see figure 6. No more proteins could be identified after 20 min whereas the last 10 min are cut off. For further illustrations, see appendix section F.
Figure 6, chromatographic separation of proteins in water extract from wheat flour using a phosphate buffer and ammonium nitrate buffer. Time in min on the x-axis and intensity of luminescence unit (LU) on the y-axis
Sample 1 (durum flour), sample 8 (barley flour) and sample 9 (wheat flour) was used for the chromatographic separation. Proteins were detected after chromatographic separation by excitation at 221 nm and measurement of the emission at 350 nm using a fluorescence
detector. Due to the behaviour of the size exclusion column, larger molecules are eluted faster than smaller. Due to handling problem of the operator, water extract from the wheat sample is excluded from the chromatographic speciation. A main issue with fractionation is the dilution of the analyte by the mobile phase resulting in analyte concentrations closer to the
instrumental LOD. Therefore, a spiked wheat flour sample was used to give further information about potential interaction between lithium and proteins. The calculated mass-balance for lithium for the fractions are presented in table 9. As can be seen from the
numerical values, some flour samples had a higher content of lithium after fractionation than before. For the water extraction, the spiked wheat flour sample was most affected while the
barley flour sample did not show any difference. For the durum flour sample, on the other hand was the mass balance after fractionation not possible to calculate due to being below LOD. For the artificial gastric juice samples wheat and barley flour showed a large
discrepancy with a higher content in the fractionation. However, in the gastric juice extraction good correlation in mass balance was obtained for the durum flour sample . Blank values of ammonium nitrate buffer and gastric juice solution was found to be 0.02 µg/L and 0.13 µg/L indicate some problem with contamination.
Table 9 mass balance calculated for the respective flour types and extractions n = 1. All concentrations in µg/L.
water extraction gastric juice extraction
Flour type Sum of all fractions total
sample Sum of all fractions sample total
Wheat - - 2.86 0.94
Wheat spiked 44.39 39.15 - -
Durum <LOD 0.37 0.43 0.35
Barley 0.76 0.72 2.53 0.61
After subtraction of the blank a lithium as well as protein profile for each flour type and extraction is seen in figure 7-9. The last 10 min of the chromatographic separation have been cut off since no lithium or proteins could be detected after 20 min. Lithium in the spiked wheat flour water sample is eluted in two fractions corresponding to an elution time between 13-14 min and 14-15 min figure 7. This elution time correlating to several low intensity protein peaks. The lithium elution profile in artificial gastric juice extract of the wheat flour sample looks slightly different. Lithium was mostly found in two fractions with an elution time between 3-4 min that do not correlate with any protein elution and 13 and 14 min where several low intensity protein peaks are found. Lithium found in the other fractions seen in the figure are below the instrumental LOD.
Lithium in the artificial gastric juice extract of the durum flour sample is mostly found in fraction 14, with an elution time between 13-14 min corresponding to an elution time of several low intensity proteins, figure 8. Lithium found in the other fractions, also the first dot in the figure where found to be below the instrumental LOD. As already mention, lithium was below the instrumental LOD in all fractions of water extract.
Lithium is eluted in fraction 14 of the water extract of the barley flour sample corresponding to an elution time between 13-14 min correlating to where several low intensity proteins are eluting, figure 9. The elution profile from the artificial gastric juice extract looks slightly different, with lithium eluting between 5-6 minutes and 13-14 minutes. Which for the latter elution time, correspond to the elution of several low intensity proteins.
Due to possible contamination and low concentrations found in samples it is not possible to determine an accurate concentration. However, a major part of lithium seems to be bound to organic compounds identified as protein.
00
44
88
1122
1166
2200
EElluuttiioonn ttiim
mee ((m
miinn))
00
4400
8800
112200
L Luu
m m
i inn
e ess
c cee
n ncc
e e
U U
n nii
t tss
00
00..0044
11
11..22
R R
e ell
a att
i ivv
e e
l liitt
h hii
u umm
c c
o onn
c cee
n ntt
r raa
t tiioo
n n
LLiitthhiiuumm ((AAGGJJ))LLiitthhiiuumm ((WW)) WWhheeaatt,, aarrttiiffiicciiaall ggaassttrriicc jjuuiiccee ((AAGGJJ))WWhheeaatt,, wwaatteerr ((WW))
W
Whheeaatt fflloouurr
Figure 7, chromatographic separation of proteins in water and gastric juice extracts of wheat flour samples and the corresponding elution time for lithium, notice the break in the y-axis of normalized lithium concentration.
No rm aliz ed lit hiu m c on cent rat io n
Figure 8, chromatographic separation of protein in water extract of the durum flour sample, and the corresponding elution time for lithium.
No rm aliz ed lit hiu m c on cent rat io n
Figure 9, chromatographic separation of protein in water and gastric juice extracts of barley flour samples, excitation 221 nm and emission 350 nm and the corresponding elution time for lithium.
Conclusion
Lithium is found in the low ng/g range in all types of flour and flour extracts analysed. The total content of lithium in wheat flour is found between 0.91 ng/g and 7.8 ng/g, in durum flour between 1.7 ng/g and 5.2 ng/g and in barley flour between 2.0 ng/g and 6.0 ng/g. The water-soluble content of lithium in wheat flour was found to vary between 0.33 ng/g and 4.8 ng/g, of durum flour between 0.83 ng/g and 2.5 ng/g and of barley flour between 0.81 ng/g and
3.3 ng/g. Bioaccessible lithium content of wheat flour is determined to lay between 0.38 ng/g and 5.3 ng/g, of durum flour between 0.98 ng/g and 1.9 ng/g and barley flour between 1.3 ng/g and 3.1 ng/g. A variable content of wheat and barley flour is seen between their respective samples for all extractions. Variation of lithium content within durum flour was seen in the artificial gastric juice content. The extraction yield for both extractions did not differ for the majority of the samples.
Due to the low abundance of lithium in the respective flour types, no measurement of the IR can be made without using mathematical treatment of the data and no statistical treatment between extraction methods was possible. IR could not be determined for some samples in the
No rm aliz ed lit hiu m c on cent rat io n
total extraction and no significant difference from the natural IR were seen for the other samples. Water as well as artificial gastric juice extraction tend to enrich the heavier isotope. A thorough statistical evaluation would require more samples and go beyond the scope of this essay.
Protein-bound lithium was found in most water and artificial gastric juice extract. Wheat, durum and barley flour had different protein signatures in their respective water extracts and this signature changed for the respective flour types with the artificial gastric juice treatment.
References
Ammari TG, Al-Zu'bi Y, Abu-Baker S, Dababneh B, Gnemat W, Tahboub A. (2011). The occurrence of lithium in the environment of the Jordan Valley and its transfer into the food chain. Environ Geochem Health. 2011 Volume 33(5), p. 427-437.
https://doi.org/10.1007/s10653-010-9343-5
Aral H, Vecchio-Sadus, A. (2008). Toxicity of lithium to humans and the environment-A literature review. Ecotoxicol Environ Saf. 2008 Volume 70, p. 349-356.
https://doi.org/10.1016/j.ecoenv.2008.02.026
Bertin Labronici R, Maltez H. F, Santos de Gois J, Borges D. L.G., da Silva Campelo Borges G, Gonzaga L. V, Fett R. (2016). Mineral composition and bioaccessibility in Sarcocornia ambigua using ICP-MS. J Food Compost Anal. 2016 Volume 47 p. 45-51.
https://doi.org/10.1016/j.jfca.2015.12.009
Długaszek M, Kurpiewska Ż, Mierczyk J. (2015). Lithium content in the tea and herbal infusions. Eur Food Res Technol. 2015 Volume 241(2), p.289-293.
https://doi.org/10.1007/s00217-015-2456-4
Erdemir U.S, Gucer S. (2018). Correlation of lithium bioaccessibility from tea (Camellia sinensis L.) with tea type and consumption habits. Food Chem. 2018 Volume 244, p. 364-370.
https://doi.org/10.1016/j.foodchem.2017.10.053
Figueroa L.T, Razmillic B, Zumeata O, Aranda G.A, Barton S.A, Schull W.J, Young A.H., Kamiya Y.M, Hoskins J.A, Ilgre E.B. (2013). Environmental Lithium Exposure in the North of Chile—II. Natural Food Sources. Biol Trace Elem Res. 2013Volume 151, p.122–131. https://doi.org/10.1007/s12011-012-9543-1
Franzaring J, Schlosser S, Damsohn W, Fangmeier A. (2016). Regional differences in plant levels and investigations on the phytotoxicity of lithium. Environ Pollut. 2016 Volume 216, p. 858-865. https://doi.org/10.1016/j.envpol.2016.06.059
Gonzälez-Weller D, Rubio Gutiérrez A.J, González G.L, Caballero Mesa J.M, Gironés C.R, Ojeda A.B, Hardisson, A. (2013). Dietary intake of barium, bismuth, chromium, lithium, and strontium in a Spanish population (Canary Islands, Spain). Food Chem Toxicol. 2013Volume 62, p. 856-868. https://doi.org/10.1016/j.fct.2013.10.026
Jiang L, Wang L, Mu S-Y, Tian C-Y. (2014). Apocynum venetum: A newly found lithium accumulator. Flora. 2014 Volume 209, p. 285-289.
https://doi.org/10.1016/j.flora.2014.03.007
Karlsson, Stefan “Personal comment.” Örebro University, Örebro, Sweden.
Krebs R. E. (2006). The history and use of our earth´s chemical elements: a reference guide. 2nd edit, London: Greenwood press.
Lemarchand E, Chabaux F, Vigier N, Millot R, Pierret M-C. (2010). Lithium isotope systematics in a forested granitic catchment (Strengbach, Vosges Mountains, France). Geochim Cosmochim Acta. 2010 Volume 74, p. 4612-4628.
https://doi.org/10.1016/j.flora.2014.03.007
Mir-Marqués, A, Cervera M.L, Guardia M. (2012). A preliminary approach to mineral intake in the Spanish diet established from analysis of the composition of university canteen menus. J Food Compost Anal. 2012 Volume 27(2), p.160–168.
https://doi.org/10.1016/j.jfca.2012.06.006
Noël L, Chekri R, Millour S, Vastel C, Kadar A, Sirot V, Leblanc J.C, Guérin T. (2012). Li, Cr, Mn, Co, Ni, Cu, Zn, Se and Mo levels in foodstuffs from the Second French TDS. Food Chem. 2012 Volume 132, p. 1502-1513. https://doi.org/10.1016/j.foodchem.2011.12.009 Piergiovanni A. R, Rizzi R, Pannacciulli E, Gatta D. (1997). Mineral composition in hulled wheat grains: a comparison between emmer (Triticum dicoccon Schrank) and spelt (T. spelta L.) accessions. Int J Food Sci Nutr. 1997 Volume 48(6), p. 381-386.
https://doi.org/10.3109/09637489709028586
Robinson B. H, Yalamanchali R, Reiser R, Dickinson N. M. (2018). Lithium as an emerging environmental contaminant: Mobility in the soil-plant system. Chemosphere. 2018 Volume 197, p. 1-6. https://doi.org/10.1016/j.chemosphere.2018.01.012
Rodushkin I, Stenberg A, Andrén H, Malinovsky D, Baxter DC. (2004). Isotopic fractionation during diffusion of transition metal ions in solution. Anal Chem. 2004 Volume 76(7):2148-51. https://doi.org/10.1021/ac035296g
Spiegel H, Sager M, Oberforster M, Mechtler K, Stüger H.P, Baumgarten, A. (2009). Nutritionally relevant elements in staple foods: influence of arable site versus choice of variety. Environ Geochem Health. 2009 Volume 31, p. 549-560.
https://doi.org/10.1007/s10653-009-9254-5
Stelmach E, Pohl P, Szymczycha-Madeja A. (2014). Evaluation of the bioaccessibility of Ca, Fe, Mg and Mn in ground coffee infusions by in vitro gastrointestinal digestion. J Braz Chem Soc. 2014 Volume 25(11), p. 1993-1999. http://dx.doi.org/10.5935/0103-5053.20140183 Ulleberg K. E, Comi I, Holm H, Herud B. E, Jacobsen M, Vegarud E. G. (2011). Human Gastrointestinal Juices Intended for Use in In Vitro Digestion Models. Food Dig.
2011Volume 2, p. 52-61. https://doi.org/10.1007/s13228-011-0015-4
Vetter J. (2005). Lithium content on some common edible wild-growing mushrooms. Food Chem. 2005 Volume 90(1-2), p. 31-37. https://doi.org/10.1016/j.foodchem.2004.03.019
Wei R, Guo Q, Wen H, Liu C, Yang J, Peters M, Hu J, Zhu G, Zhang H, Tian L, Han X, Ma J, Zhu C, Wan Y. (2016). Fractionation of Stable Cadmium Isotopes in the Cadmium Tolerant Ricinus communis and Hyperaccumulator Solanum nigrum. Sci Rep. 2016 Volume 6(24309).
Wiederhold J. G. (2015). Metal stable isotope signatures as tracers in environmental geochemistry. Environ Sci Technol. 2015 Volume 49, p. 2606–2624.
https://doi.org/10.1021/es504683e
Weiss DJ, Mason TF, Zhao FJ, Kirk GJ, Coles BJ, Horstwood MS. (2005). Isotopic discrimination of zinc in higher plants. New Phytol. 2005 Volume 165(3), p. 703-10.
Appendix
Section ARaw data from the first evaluation to find a proper mass of flour for the objective and aim of this study.
Table A1, raw data from the method optimization.
7 Li [1]
d Name SETUP lakning Massa L/S ofilt/filt Conc. [µg/L] Conc. RSD
D1V1 190411 D)A vatten 4.04 10 ofilt 0.18 8.85
D1V1 190415 E) vatten 4 5 filt+syra 0.26 4.64
D1V2 190412 D)B vatten 4.04 10 syra+filt 0.23 9.23
D1V3 190412 D)C vatten 4.04 10 filt+syra 0.21 0.63
D3 x1 A) tot 0.99 50 filt 0.03 1.03
D32 190418 C) tot 3.02 16 ofilt 0.14 7.58
K1V1 190411 D)A vatten 3.99 10 ofilt 0.14 10.93
K1V1 190415 E) vatten 4.02 5 filt+syra 0.19 4.56 K1V2 190412 D)B vatten 3.99 10 syra+filt 0.16 9.24 K1V3 190412 D)C vatten 3.99 10 filt+syra 0.15 2.77 K3 190418 B) tot 0.99 50 ofilt 0.12 43.09 K3 x1 A) tot 0.97 50 filt 0.07 9.36 K32 190418 C) tot 3 16 ofilt 0.32 7.29
V1V1 190411 D)A vatten 4.04 10 ofilt 0.51 9.72
V1V1 190415 E) vatten 4.01 5 filt+syra 0.83 6.73 V1V3 190412 D)C vatten 4.04 10 filt+syra 0.55 4.89 V3 190418 B) tot 0.98 50 ofilt 0.03 29.11 V3 x1 A) tot 1.14 50 filt 0.03 19.48 V32 190418 C) tot 3.04 16 ofilt 0.10 0.28 D3 190418 B) tot 1.02 50 ofilt 0.04739334 25.1183395 .
Table A2, raw data from the optimization of the method.
6 Li [ 1 ] 7 Li [ 1 ] SETUP CPS CPS RSD CPS CPS RSD D1V1 190411 D)A 104 9.967486 1078.405 8.9627205 D1V1 190415 E) 143.005 6.2648459 1755.15 4.9698711 D1V2 190412 D)B 135.665 0.3492143 1258.425 6.5556667 D1V3 190412 D)C 106.665 11.488455 1156.745 2.852891 D3 190418 B) 40 9.4398755 195.005 8.4596811 D3 x1 190410 A) 57.665 8.9882818 355.015 3.3202738 D32 190418 C) 48.665 3.8795338 333.345 5.6573633 K1V1 190411 D)A 88.335 2.6656089 836.715 10.14203 K1V1 190415 E) 130.335 10.487109 1535.12 5.9889796 K1V2 190412 D)B 92 3.0743773 938.395 5.2754486 K1V3 190412 D)C 83 10.785508 868.385 4.6145133 K3 190418 B) 47.665 28.675913 295.01 24.76947 K3 x1 190410 A) 54 13.958812 531.69 9.3094613 K32 190418 C) 66.67 4.2424286 575.03 2.8700889 V1V1 190411 D)A 212.34 7.3261511 2716.99 6.2471305 V1V1 190415 E) 344.67 3.2824756 4809.22 8.4813645 V1V3 190412 D)C 207.335 14.552413 2638.65 2.0554105 V3 190418 B) 36.335 6.4804337 171.675 6.8661541 V3 x1 190410 A) 40.335 5.8377726 320.015 10.312227 V32 190418 C) 49.33 5.7336856 278.345 0.8459522 V3V2 190412 D)B 89.665 19.454998 750.045 6.2853308 Table A3, raw data from the optimization of the method
6 Li cps 7 Li cps
CALBLK.D 190410 48.335 24.387028 131.67 19.687503 CALBLK.D 190411/12 48.665 15.503605 191.67 11.067566 CALBLK.D 190418 39.665 13.067169 145.005 1.6238513 CALBLK.D 190415 66 7.1353502 340.015 16.639152
Section B
Raw data from total, water and artificial gastric juice extraction.
Total extraction: D1-D5 stands for durum sample 1-5. K1-K3 stands for barley sample 6-8 and V1-V5 for wheat sample 9-13 in the result and discussion. Difference in second number only indicate replicate.
Water extraction: D1V-D5V stands for durum sample 1-5. K1V-K3V stands for barley sample 6-8 and V1V-V5V for wheat sample 9-13 in the result and discussion. Difference in second number only indicate replicate.
Artificial gastric juice extraction: D1G-D5G stands for durum sample 1-5. K1G-K3G stands for barley sample 6-8 and V1G-V5G for wheat sample 9-13 in the result and discussion. Difference in second number only indicate replicate.
Table A4, raw data from extractions.
Sample Mass 7 Li 24 Mg 88 Sr Sample mass 7 Li 24 Mg 88 Sr
Total extraction g [µg/L] conc [µg/L] conc [µg/L] conc Artificial gastric juice extraction g [µg/L] conc [µg/L] conc [µg/L] conc
D1 190418 0.99 0.10 12193.42 8.58 D1G1 190430 3.99 0.3472897 127293.75 106.58404 D12 190418 3 0.26 40818.71 27.89 D1G2 190430 4.01 0.3755211 131742.6 108.85696 D14 190510 3 0.1343832 40087.136 30.242174 D1G3 190430 4 0.3531712 130561.82 110.84522 D2 190418 0.96 0.07 11859.83 8.35 D2G1 190430 3.99 0.262576 117912.44 99.102695 D22 190418 3 0.26 40885.14 27.92 D2G2 190430 4 0.2745587 122365.22 100.73267 D24 190510 3.06 0.1903261 39225.215 28.304111 D2G3 190430 4.02 0.2824714 105652.18 87.726845 D3 190418 1.02 0.05 10568.51 8.32 D3G1 190430 3.99 0.226436 115750.47 104.35457 D32 190418 3.02 0.14 36303.74 26.91 D3G2 190430 4 0.2706072 110913.36 100.92669 D34 190510 3.06 0.1112553 37794.791 28.115974 D3G3 190430 3.99 0.2009 106104.81 96.613737 D4 190418 1.05 0.05 11258.08 8.83 D4G1 190430 4.02 0.2177303 103005.14 98.091126 D42 190418 2.98 0.17 35051.92 26.84 D4G2 190430 4.03 0.2061186 109139.2 101.32868 D44 190510 2.97 0.1201681 31048.786 25.840501 D4G3 190430 4 0.207781 104966.69 100.84988 D5 190418 0.95 0.09 8814.66 6.97 D5G1 190430 4 0.229135 108826.97 100.23024 D52 190418 3.04 0.16 35903.24 26.87 D5G2 190430 3.99 0.2269838 108833.46 100.07539 D54 190510 3.06 0.1016492 32793.349 26.379607 D5G3 190430 4.03 0.1966048 99510.094 93.347625 K1 190418 1 0.07 17614.19 28.30 K1G1 190510 4.03 0.3051319 166280.26 368.96688 K12 190418 2.95 0.15 69369.75 105.40 K1G2 190508 4.02 0.28 185269.69 411.19 K14 190510 3.02 0.1221432 71090.188 115.69417 K1G3 190508 4.04 0.31 170692.67 374.08 K2 190418 0.99 0.06 21035.55 34.22 K2G1 190510 4.02 0.2745414 154185.58 351.20166 K22 190418 2.99 0.19 72053.79 114.26 K2G2 190508 4.01 0.3461329 152490.18 339.94369 K24 190510 3 0.1266474 68530.424 116.86772 K2G3 190508 4.03 0.27 144527.25 338.70 K3 190418 0.99 0.12 20618.60 22.16 K3G2 190508 4.03 0.62 140126.81 221.57 K32 190418 3 0.32 66581.21 69.61 V1G1 190502 4 1.0577541 61732.114 87.981962 K34 190510 3.03 0.2751402 64026.485 72.963732 V1G2 190502 4.03 1.0184119 59070.186 84.363954 V1 190418 0.99 0.16 4169.99 5.50 V1G3 7/5-19 4.01 0.9383346 58064.688 84.550509 V12 190418 3.02 0.47 20273.29 24.87 V2G1 190502 4.02 0.1117841 47683.184 141.26552 V14 190510 3.02 0.4043411 21391.895 26.906401 V2G2 190502 4.03 0.1218677 46836.176 137.73322 V2 190418 0.97 0.08 6096.75 16.57 V2G3 7/5-19 4 0.1375776 46912.214 152.97229 V22 190418 3 0.12 18950.50 48.15 V3G1 190502 4.01 0.1335182 46284.202 209.2494 V24 190510 2.97 0.0717692 18683.757 47.863814 V3G2 190502 4.02 0.1352412 49845.774 218.70071 V3 190418 0.98 0.03 6259.66 24.56 V3G3 7/5-19 4.02 0.1238308 46170.203 227.7657 V32 190418 3.04 0.10 19600.02 74.33 V4G1 190502 4 0.0953746 42386.913 145.18518 V34 190510 3.05 0.0639079 18791.13 75.649796 V4G2 190502 3.99 0.0993874 44352.497 146.33306 V4 190418 0.98 0.03 4250.54 13.24 V4G3 7/5-19 4.02 0.07708 38302.426 156.24985 V42 190418 3.01 0.08 16049.82 48.00 V5G1 190502 4.03 0.1435404 44600.35 154.2294 V44 190510 3.04 0.0559002 16489.349 49.317001 V5G2 190502 4 0.1530718 52958.369 184.78284 V5 190418 0.94 0.03 4725.52 14.75 V5G3 7/5-19 4.02 0.116107 44424.007 179.29235 V52 190418 2.98 0.09 15690.88 47.28 V54 190510 3 0.0641573 16324.142 50.395244
Table A5, raw data from extractions
Water extraction
Water extraction
Sample mass 7 Li 24 Mg 88 Sr Sample mass 7 Li 24 Mg 88 Sr
g conc [µg/L] conc [µg/L] conc [µg/L] g conc [µg/L] conc [µg/L] conc [µg/L]
D1V1 190415 4 0.26 84361.62 46.90 K2V2 190418 3.99 0.18 154533.56 232.34 D1V1 190417 3.96 0.40 97484.55 46.87 K2V3 190418 4.01 0.22 146580.01 212.87 D1V2 190418 3.98 0.39 102129.44 49.94 K3V1 190415 3.98 0.60 142451.60 131.44 D1V3 190418 4.03 0.27 87211.17 41.62 K3V1 190417 4.02 0.67 147599.67 126.71 D2V1 190415 4.02 0.19 81212.59 44.13 K3V2 190418 4.02 0.58 145144.55 127.52 D2V1 190417 4.02 0.27 84842.25 40.43 K3V3 190418 4.02 0.61 139604.98 120.64 D2V2 190418 3.99 0.32 100502.51 48.64 V1V1 190415 4.01 0.83 51468.57 55.50 D2V3 190418 4.03 0.51 84660.71 39.50 V1V1 190417 3.99 0.92 48895.14 48.10 D3V1 190415 3.98 0.17 76634.28 48.10 V1V2 190418 4.03 0.96 54320.30 55.48 D3V1 190417 4.02 0.43 82002.51 44.36 V1V3 190418 4.03 0.84 50361.98 50.15 D3V2 190418 4.01 0.25 96618.61 53.61 V2V1 190415 3.99 0.17 53710.47 123.32 D3V3 190418 4.01 0.21 80828.87 42.68 V2V1 190417 4 0.14 43933.97 92.54 D4V1 190415 3.99 0.18 81378.85 51.81 V2V2 190418 4.05 0.16 44812.48 96.72 D4V1 190417 4 0.17 65042.53 37.62 V2V3 190418 4.03 0.13 43766.29 93.32 D4V2 190418 4.01 0.25 89954.98 51.87 V3V1 190415 3.98 0.09 42846.99 152.55 D4V3 190418 4.04 0.20 75409.52 42.58 V3V1 190417 4 0.14 42450.57 141.88 D5V1 190415 4 0.21 85159.98 54.82 V3V2 190418 4.04 0.14 45141.63 154.35 D5V1 190417 4.05 0.24 78988.25 43.83 V3V3 190418 4.03 0.12 41253.21 139.34 D5V2 190418 3.99 0.25 92211.22 52.01 V4V1 190415 3.98 0.07 35424.22 91.68 D5V3 190418 4.01 0.18 74345.04 40.45 V4V1 190417 3.98 0.14 39492.63 96.18 K1V1 190415 4.02 0.19 142273.44 204.38 V4V2 190418 4.03 0.13 37046.93 91.41 K1V1 190417 4.01 0.24 183439.05 245.76 V4V3 190418 4.01 0.11 38751.03 95.94 K1V2 190418 4 0.21 164969.43 223.55 V5V1 190415 3.98 0.12 41526.13 108.15 K1V3 190418 4.03 0.20 162002.46 214.56 V5V1 190417 4.02 0.16 38382.88 94.38 K2V1 190415 3.99 0.16 128968.33 208.18 V5V2 190418 4.02 0.14 42008.57 105.10 K2V1 190417 4.02 0.24 143517.56 213.71 V5V3 190418 4.01 0.09 37630.96 95.53
Section C
Raw data for isotope ratio calculations.
Table A6, raw data from isotope ratio calculations
Total extraction Artifical gastric juice extraction
Sample µg/L 6 Li 7 Li sample µg/L 6 Li 7 Li D12 190418 0.2560 64 496.69 D1G1 190430 0.3473 120.67 1583.465 D14 190510 0.1344 51 423.355 D1G2 190430 0.3755 129.335 1636.8 D22 190418 0.2593 73.665 566.695 D1G3 190430 0.3532 141.335 1558.465 D24 190510 0.1903 55 488.355 D2G1 190430 0.2626 103 1315.1 D32 190418 0.1366 73 761.71 D2G2 190430 0.2746 97.33 1316.76 D34 190510 0.1113 48.665 333.345 D2G2 190430 0.2825 116 1510.13 D42 190418 0.1653 57.335 485.02 D3G1 190430 0.2264 104.33 1125.075 D44 190510 0.1202 53.33 365.02 D3G2 190430 0.2706 105.335 1348.43 D52 190418 0.1623 58.335 525.025 D3G3 190430 0.2009 87.665 1071.735 D54 190510 0.1016 51 363.35 D4G1 190430 0.2177 91.335 1118.41 K12 190418 0.1515 68 460.025 D4G2 190430 0.2061 87.335 1030.07 K14 190510 0.1221 53.665 343.35 D4G3 190430 0.2078 84.665 1063.405 K22 190418 0.1910 59.665 523.36 D5G1 190430 0.2291 78.335 1120.075 K24 190510 0.1266 55.665 381.68 D5G2 190430 0.2270 88.33 1115.075 K3 190418 0.1181 64.335 536.69 D5G3 190430 0.1966 94.33 1076.74 K32 190418 0.3177 47.665 295.01 K1G1 190510 0.3051 174.67 1870.165 K34 190510 0.2751 66.67 575.03 K1G2 190508 0.2791 137 1513.455 V1 190418 0.1552 97.665 1065.07 K1G3 190508 0.3059 170.67 2008.525 V12 190418 0.4727 48 341.68 K2G1 190510 0.2745 104.665 921.73 V14 190510 0.4043 91 808.38 K2G2 190508 0.3461 106.335 961.735 V24 190510 0.0718 71.665 830.045 K2G3 190508 0.2687 149.335 1588.465 V32 190418 0.1020 127.67 1495.115 K3G2 190508 0.6179 178.665 1730.185 V42 190418 0.0795 57.665 356.685 V1G1 190502 1.0578 333.34 4859.245 49.33 278.345 V1G2 190502 1.0184 331.34 4819.2 48.665 246.675 V1G3 7/5-19 0.9383 452.34 5064.305 V2G1 190502 0.1118 68.335 793.38 V2G2 190502 0.1219 85.335 863.385 V2G3 7/5-19 0.2376 140.665 1480.125 V3G1 190502 0.1335 78 913.39 V3G2 190502 0.1352 69.335 840.045 V3G3 7/5-19 0.2238 144.335 1318.43 V4G1 190502 0.0954 68.335 698.37 V4G2 190502 0.0994 63.665 666.7 V4G3 7/5-19 0.1771 120 1060.075 V5G1 190502 0.1435 69 891.72 V5G2 190502 0.1531 73.665 855.05 V5G3 7/5-19 0.2161 129.335 1255.09
Table A7 raw data from the isotope ratio calculations
Water extraction Water extraction
Sample µg/L 6 Li 7 Li sample µg/L 6 Li 7 Li D1V1 190415 0.2564 143.005 1755.15 K2V2 190418 0.1831 62.665 618.365 D1V1 190417 0.4049 118 1311.77 K2V3 190418 0.2201 65.335 708.37 D1V2 190418 0.3867 89.665 1116.74 K3V1 190415 0.6035 278.005 3585.535 D1V3 190418 0.2703 88.665 871.715 K3V1 190417 0.6721 167.005 2210.23 D2V1 190415 0.1869 123.335 1383.435 K3V2 190418 0.5815 123.335 1790.155 D2V1 190417 0.2673 88 983.4 K3V3 190418 0.6123 131 1863.5 D2V2 190418 0.3169 86 960.055 V1V1 190415 0.8332 344.67 4809.22 D2V3 190418 0.5098 120 1656.805 V1V1 190417 0.9249 220.67 3098.735 D3V1 190415 0.1693 117.335 1321.76 V1V2 190418 0.9611 219.005 2785.335 D3V1 190417 0.4302 134.335 1466.78 V1V3 190418 0.8402 188.005 2566.95 D3V2 190418 0.2522 75.665 766.71 V2V1 190415 0.1674 117.67 1140.085 D3V3 190418 0.2095 68.665 698.37 V2V1 190417 0.1429 67.33 586.7 D4V1 190415 0.1755 131 1238.425 V2V2 190418 0.1625 64 598.365 D4V1 190417 0.1663 60 571.695 V2V3 190418 0.1283 60.335 533.36 D4V2 190418 0.2532 80.67 766.71 V3V1 190415 0.0933 92 863.39 D4V3 190418 0.1980 74.67 693.37 V3V1 190417 0.1404 76 561.695 D5V1 190415 0.2053 122.335 1428.44 V3V2 190418 0.1380 65 515.025 D5V1 190417 0.2438 103.335 901.72 V3V3 190418 0.1227 65.335 488.355 D5V2 190418 0.2452 81.67 738.375 V4V1 190415 0.0656 82.665 715.04 D5V3 190418 0.1844 74.665 653.37 V4V1 190417 0.1375 67.67 530.025 K1V1 190415 0.1918 130.335 1535.12 V4V2 190418 0.1316 59.335 518.36 K1V1 190417 0.2437 60.665 678.37 V4V3 190418 0.1127 62 446.69 K1V2 190418 0.2148 70.67 696.705 V5V1 190415 0.1217 90 938.395 K1V3 190418 0.2045 63 695.04 V5V1 190417 0.1604 73 636.7 K2V1 190415 0.1618 125.335 1301.76 V5V2 190418 0.1439 51.67 501.695 K2V1 190417 0.2409 70.335 861.72 V5V3 190418 0.0915 61.665 403.35
Calibration curves of standard solutions that were assumed having a natural IR using standards of 0.1 and 1 µg/L vs their IR.
Fig A1 Calibration curve std. solution from: 190418 (tot extract.) Fig A2 Calibration curve std. solution from 190510 (tot extract).
Fig A3 Calibration curve for std. solution from: 190417/18 (water extract) Fig A4 Calibration curve for std. solutions from 190415 (water extract).
y = -0,0704x + 0,1432 0 0,05 0,1 0,15 0 0,5 1 1,5 6L i/7L i µg/L
Calibration curve 190418
y = -0,0446x + 0,1239 0 0,05 0,1 0,15 0 0,2 0,4 0,6 0,8 1 1,2 6L i/7L i µg/LCalibration curve 190510
y = -0,0532x + 0,1201 0 0,05 0,1 0,15 0 0,2 0,4 0,6 0,8 1 1,2 6L i/7L I µg/LCalibration curve 190415
y = -0,0704x + 0,1432 0 0,05 0,1 0,15 0 0,2 0,4 0,6 0,8 1 1,2 6L i/7L i µg/LCalibration curve 190417/18
Fig A5 Calibration curve for std. solutions from 190430/190502 (gastric extract) Fig A6 Calibration curve for std. solutions from 190508/10 (gastric extract)
Fig A7 Calibration curve for std. solutions from 190507 (gastric extract)
y = -0,0479x + 0,1147 0 0,02 0,04 0,06 0,08 0,1 0,12 0 0,2 0,4 0,6 0,8 1 1,2 6L i/7L i µg/L
Calibration curve 190430/190502
y = -0,0446x + 0,1239 0 0,05 0,1 0,15 0 0,2 0,4 0,6 0,8 1 1,2 6L i/7L i µg/LCalibration curve 190508/10
y = -0,061x + 0,1363 0 0,02 0,04 0,06 0,08 0,1 0,12 0,14 0 0,2 0,4 0,6 0,8 1 1,2 6L i/7L i µg/LCalibration curve 190507
Section D
Raw data from the fractionation:
VIV1= water extraction of wheat flour (LOD <0.01). VIV2=water extraction of wheat flour spiked (LOD <0.02). DIV1= water extraction of durum flour (LOD <0.01). DIG= artificial gastric juice extraction of durum flour (LOD <0.01). K3V1= water extraction of barley flour (LOD <0.01). K3G= artificial gastric juice extraction of barley flour (LOD <0.01). VIG= artificial gastric juice extraction of wheat flour (LOD <0.02). Table A8 raw data from fractionation
7 Li [ 1 ] 7 Li [ 1 ] 7 Li [ 1 ] 7 Li [ 1 ]
Sample Name Conc. µg/L Conc. RSD Sample Name Conc. µg/L Conc. RSD Sample Name Conc.µg/L Conc. RSD Sample Name Conc. µg/L Conc. RSD V1V1 F01 0.017247 49.942591 V1V2 F01 0.0217312 48.039931 D1V1 F01 0.0133074 10.806055 DIG F01 0.0312148 4.4256734 V1V1 F02 0.0166131 14.182297 V1V2 F02 0.0190436 28.286029 D1V1 F02 0.01 26.027895 DIG F02 0.0257594 8.5115815 V1V1 F03 0.0067776 48.795758 V1V2 F03 0.0164146 61.082934 D1V1 F03 0.0066212 10.639783 DIG F03 0.0222915 42.634918 V1V1 F04 0.0079915 40.781742 V1V2 F04 0.0157586 38.937453 D1V1 F04 0.0113902 43.962672 DIG F04 0.0178393 11.943162 V1V1 F05 0.0532942 10.967928 V1V2 F05 0.0161307 7.0652307 D1V1 F05 0.01 32.064395 DIG F05 0.0206188 34.261078 V1V1 F06 0.0179633 30.37459 V1V2 F06 0.02 8.0553111 D1V1 F06 0.01 186.90915 DIG F06 0.0271576 35.433005 V1V1 F07 0.0250893 0.5161407 V1V2 F07 0.02 11.674592 D1V1 F07 0.01 66.497241 DIG F07 0.0222729 13.428228 V1V1 F08 0.0113805 27.306145 V1V2 F08 0.0201349 16.962983 D1V1 F08 0.01 89.743972 DIG F08 0.0197265 12.895467 V1V1 F09 0.0097842 28.051514 V1V2 F09 0.0205532 55.004239 D1V1 F09 0.0064989 163.5583 DIG F09 0.0174976 11.971079 V1V1 F10 0.0155434 46.503529 V1V2 F10 0.0186993 25.750072 D1V1 F10 0.01 453.63642 DIG F10 0.01 4.545292 V1V1 F11 0.0143272 13.387488 V1V2 F11 0.02 19.270573 D1V1 F11 0.01 N/A DIG F11 0.0160203 8.7257902 V1V1 F12 0.0100544 48.076679 V1V2 F12 0.0173446 26.626659 D1V1 F12 0.01 N/A DIG F12 0.0174846 30.084957 V1V1 F13 0.0146291 9.775953 V1V2 F13 0.0276216 4.8388136 D1V1 F13 0.01 31.926741 DIG F13 0.0229291 5.5993476 V1V1 F14 0.0215649 27.073789 V1V2 F14 1.087662 0.4408845 D1V1 F14 0.0124848 29.869099 DIG F14 0.0392334 17.024553 V1V1 F15 0.0088404 20.053407 V1V2 F15 0.0628498 6.5092923 D1V1 F15 0.0169099 31.026037 DIG F15 0.0179466 2.5722311 V1V1 F16 0.0205694 41.147193 V1V2 F16 0.0278201 13.125086 D1V1 F16 0.01 6.1035665 DIG F16 0.0200506 27.264468 V1V1 F17 0.0085632 21.968862 V1V2 F17 0.0275603 31.790149 D1V1 F17 0.0079279 73.502407 DIG F17 0.0205523 30.409413 V1V1 F18 0.01 95.895904 V1V2 F18 0.0173475 12.709108 D1V1 F18 0.0063148 57.72076 DIG F18 0.0211067 64.993842 V1V1 F19 0.0080408 75.717214 V1V2 F19 0.0167452 4.4337494 D1V1 F19 0.0068158 3.5832447 DIG F19 0.01 24.617643 V1V1 F20 0.0114801 88.811654 V1V2 F20 0.0187308 79.798126 D1V1 F20 0.0080526 7.3892212 DIG F20 0.0186593 4.3330522 V1V1 F21 0.0156173 0.6945434 V1V2 F21 0.02 29.362903 D1V1 F21 0.01 13.204917 DIG F21 0.015481 27.876373 V1V1 F22 0.0120011 5.7211011 V1V2 F22 0.0178485 50.155996 D1V1 F22 0.01 74.551042 DIG F23 0.0172436 58.218657 V1V1 F23 0.008954 53.413034 V1V2 F23 0.0230637 7.8522305 D1V1 F23 0.01 1754.2335 missing V1V1 F24 0.0064295 124.22703 V1V2 F24 0.0219255 28.110461 D1V1 F24 0.01 71.591291 DIG F24 0.0170814 15.623627 V1V1 F25 0.01 154.38999 V1V2 F25 0.01857 8.9592134 D1V1 F25 0.01 74.984796 DIG F25 0.0157932 40.775335