Chronological development of element

0 downloads 0 Views 831KB Size Report
Jan 4, 2008 - and wine-grower's cellar, a total of 51 samples of the ... was controlled by cooling between 16 and 18. ∘ ..... must weight is given in the unit. ∘.
DOI 10.1515/ract-2013-2111 | Radiochim. Acta 2014; 102(4): 333–349

Markus Feige, Gabriele Hampel, Jens Volker Kratz*, Norbert Wiehl, Helmut König, and Andreas Wagner

Chronological development of element concentrations in grapes during growth and ripeness and during fermentation of must determined by instrumental neutron-activation analyses Abstract: The chronological development of element concentrations during growth and ripeness of grapes described in the literature has only been concerned with the macro elements Mg, K, and Ca. Concentrations of trace elements in must are only described as a snapshot at the end of the ripeness. Therefore, the motivation for the present work was to accompany the growth and the ripening process of grapes successively by systematically determining element concentrations in grapes of Riesling and Cabernet Sauvignon by neutron-activation analyses. While for a number of elements, the concentrations in the grapes increased as a function of grape development (e.g., Na, K, Rb, Al), other concentrations decreased (e.g., Mg, Ca, Mn). These decreases are not only to be attributed to a dilution by an increasing uptake of water during growth, but also by an active transport of the cations out of the berries. Furthermore, the interest focused on the influence of mineral substances on the process of fermentation and on the uptake of trace elements by the yeasts. Keywords: Element concentrations, Grapes and must, Chronological development, Fermentation, Wine, Instrumental neutron-activation analyses. || *Corresponding Author: Jens Volker Kratz, Institut für Kernchemie, Johannes Gutenberg-Universität, 55099 Mainz, Germany, E-mail: [email protected] Markus Feige, Gabriele Hampel, Norbert Wiehl: Institut für Kernchemie, Johannes Gutenberg-Universität, 55099 Mainz, Germany Helmut König: Institut für Mikrobiologie und Weinforschung, Johannes Gutenberg-Universität, 55099 Mainz, Germany Andreas Wagner: Weingut Wagner, 55270 Essenheim, Germany

1 Introduction The quality of wine is influenced by a number of factors. These include, besides the ecological conditions that influence the development of the vine and ripeness of the grapes, also the processing of the grapes by the vinedresser and the involved microorganisms such as yeast, lactobacilli, acetic acid bacteria and fungi on grapes and in the must [1]. The growth of the grapes is depending on the climate, on the location of the vineyard (terroir), the type of soil, and the supply of nutritive substances. The substances deposited in the grapes which guarantee the nourishment of the yeast during fermentation, are supplied by the vine. This requires an optimum supply of nutritive substances to the plant, e.g., by minerals and bioavailable nitrogen, that can vary from year to year as it is coupled to the supply of water. In dry years, the supply of minerals is less than in wet years [2]. Substances coming from the treatment of the vines and the grapes by the vine-grower include those contained in the fertilizers, in pesticides, and in contaminations stemming from the processing of the grapes. To understand and improve the wine-making, some elements are used as chemical indicators at each step of the process. During growth and ripeness of the grapes, the evolution of the concentrations of macro-elements like Mg, K, and Ca are often considered. Especially, Mg and Ca are involved in the development of the cells as they are bound to cell-wall components [3] and are present in higher concentrations in the growth period. In addition, Mg ions are required for the activity (cofactor) of various enzymes of the carbohydrate metabolism and the stability of many proteins such as ribosomes [4]. K is also a cofactor of enzymes and involved in transport processes and cell trugor. Therefore, the K concentration in the berry increases with increasing ripeness while the concentra-

Bereitgestellt von | Johannes Gutenberg Universitaet Mainz Angemeldet Heruntergeladen am | 28.01.16 12:01

334 | Markus Feige et al., Chronological development of element concentrations in grapes

tions of Mg and Ca decrease: The concentration of K increases from 1460 𝜇g/mL (23∘ Oechsle) to 1990 𝜇g/mL (69∘ Oechsle), while the concentration of Ca decreases from 125 𝜇g/mL (23∘ Oechsle) to 32 𝜇g/mL (69∘ Oechsle) and the concentration of Mg decreses from 76 𝜇g/mL (23∘ Oechsle) to 55 𝜇g/mL (69∘ Oechsle) [2]. In the literature, data on concentrations of trace elements are restricted to the ripened must as well as to the finished wine. For Riesling must, typical concentrations of Al range from 0.006 to 0.52 𝜇g/mL, for Cu from 0.092 to 0.99 𝜇g/mL, for Mn from 0.032 to 1.5 𝜇g/mL, for Fe from 0.098 to 2.1 𝜇g/mL. The development of the grape is described in the literature rather according to the development of the acid content, the content of aroma and colour substances. During fermentation of the must, the role of a number of elements is largely known: The macronutrients Mg, K, Ca, and Na are actively incorporated into the yeast cells whereby an alcohol content of 6% – 8% is increasing the uptake [5]. These elements play an important role, e.g., in the regulation of the cell metabolism as well as in the activation of enzymes [3, 6]. Thereby, in particular Mg, along with the trace-element Zn, are cofactors of fermentation enzymes such as the phosphoglycerate kinase (Mg2+ ) and the alcohol dehydrogenase (Zn2+ ), respectively [5]. K ions are important for the uptake of phosphate into the yeast cells and are thus essential for the energy supply of the cells. Elements like Cu, Fe, Zn, Mn, etc., are required only in trace concentrations and serve as cofactors (prosthetic group) of metallo enzymes [6]. The developments of concentrations of the macro-elements and of some trace-elements during fermentation are comprehensively described in the literature. These data are based on laboratory experiments with fermentations performed on a small scale only. It is described that during fermentation, the concentrations of K and Ca decrease, while the concentrations of Mg, Na, and Mn are little affected. Also, a decrease of the concentrations of Fe, Cu, and Zn due to an uptake by the cells or by the precipitation of rarely soluble salts, respectively, is reported [6]. In another reference, a “quantitative decrease” of toxic elements such as Pb, Cd, Cu, and Hg during fermentation has been reported [7, 8]. However, these data are based on laboratory experiments with fermentations performed on a small scale only. The aim of the present work was to investigate how Mg, K, and Ca are enriched in the grapes and how their concentrations vary with time during the growth and ripeness of the grapes. Because data on the successive development of trace-element concentrations during growth and ripening of the grapes are not available, we also determined – for the first time – the change of concentrations

of the trace elements Na, Al, Mn, Zn, Cu, and Rb. Furthermore, it was of interest to find out whether metal-ion concentrations of the same elements being either too low or too high can influence the progress of the fermentation. To this end, samples of grapes of Riesling and Cabernet Sauvignon from the vineyard Essenheimer Teufelspfad located on marly limestone soil in the German wine-growing region Rheinhessen were taken regularly. The analyses of a large number of samples were performed by instrumental neutron-activation analyses (INAA) because this technique allows for multi-element determinations in the sub-mg/mL range. With INAA, it is possible to detect up to 70 elements with a high degree of selectivity [9]. Moreover, no digestion as in the case of ICPMS is necessary thus avoiding contaminations. The approach in the present work goes beyond the above cited isolated laboratory experiments. Here, the laboratory experiments are replaced by systematic multielement analyses during the real successive ripening of the grapes in the vineyard as well as during the fermentation process in the wine cellar. In order to realise a connection from the laboratory to the vineyard and wine-grower’s cellar, a total of 51 samples of two types of grapes and the resulting must and wines were analysed with respect to their concentrations of selected elements, whereby each sample was measured three times. The large number of samples was necessary in order to detect short-time variations during wine grape ripening and wine-making, as here, in contrast to the laboratory experiments, the wine-grower alone exerts influence on the production of his wines, in particular by influencing/regulating the fermentation processes and the application of selected starter cultures.

2 Experimental 2.1 Sampling and pre-treatment In the present work, the laboratory experiments are replaced by systematic multi-element analyses during the real successive ripening of the grapes in the vineyard as well as during the fermentation process in the wine cellar for Riesling and Cabernet Sauvignon. In order to realise a connection from the laboratory to the vineyard and wine-grower’s cellar, a total of 51 samples of the two types of grapes and the resulting musts and wines were analysed with respect to their concentrations of selected elements. The large number of samples was necessary in order to detect short-time variations during wine grape ripening and wine-making, as here, in contrast to the laboratory experiments, the wine-grower alone exerts

Bereitgestellt von | Johannes Gutenberg Universitaet Mainz Angemeldet Heruntergeladen am | 28.01.16 12:01

Markus Feige et al., Chronological development of element concentrations in grapes

influence on the production of his wines, in particular by influencing/regulating the fermentation processes and the application of selected starter cultures. In addition, samples from two Riesling wines during fermentation were taken where one was spontaneously fermented while the other was inoculated with a yeast starter culture. The samples irradiated and measured in this work were taken in the period from 5 July 2007–11 February 2008. The samples were kept frozen at -21 ∘ C until irradiation. This was also true particularly for the samples of the fermenting must that were taken starting on 9 October 2007. The vessels used for working with and storage of the grapes, fresh must, fermenting must, and yeast were made from plastics or glass and were washed several times with deionized water (Millipore) before use to avoid minerals contaminations.

2.1.1 Grapes For the determination of element concentrations during growth and ripening of the berries, grapes were cut from Riesling and Cabernet Sauvignon vines regularly starting on 5 July 2007 and were frozen at most two hours later for their preservation. The dates of sampling were 5, 19, 30 July, 22 August as well as 6, 21 and 29 September. On the last date, only Riesling grapes could be sampled as the Cabernet Sauvignon grapes were already harvested. In the early dates, several bunches of grapes were cut off from the vines in order to obtain an acceptable volume. In order to observe variations of the concentrations during growth and ripening period of the berries, they were thawn and pressed by hand in order to avoid contaminations. Before the pressing, the frozen berries were separated from the peduncles, and thawn embedded in polyethylene foil in order to avoid variations in the concentrations by condensing water. Saping was done in 50 mL centrifuge cones by squeezing the berries with a glass rod. Unsoluble parts were separated from the must by centrifugation. Muddy particles such as microorganisms were removed by filtering the must through a membrane filter of 0.2 𝜇m pore size. The liquid juice was frozen for preservation.

2.1.2 Fermenting musts Element concentrations of fermenting musts of Riesling and Cabernet Sauvignon were investigated. The larger part of the Riesling must was inoculated with a yeast starter culture (Anaferm 4, Oenolytic) a smaller part was sub-

| 335

ject to spontaneous fermentation. The Cabernet Sauvignon was inoculated with a yeast starter culture (SIHA 8, Begerow, Langenlonsheim) and was fermented within three days. During this time, sampling was not possible. Only after decantation and after it was filled in a wooden barrel for malo-lactic fermentation and maturing, a sample could be taken. In contrast to the rapid fermentation of the Cabernet Sauvignon, the fermentation of the Riesling was controlled by cooling between 16 and 18 ∘ C and took two weeks. Because the metabolic activity of the yeast is maximum at the beginning of the fermentation, periodic sampling was advisable only during the first two weeks. As the fermentation process started more slowly in case of the spontaneously fermenting Riesling, samples were taken here only every five days in the beginning and later at the same times as from the inoculated Riesling. After the end of fermentation, the young wine was separated from the yeast, muddy particles, and rarely soluble salts and was sulfurized for stabilization. For the separation of the yeast cells, the fermenting must was centrifuged 5 min at 5000 rpm. The supernatant was decanted and filtered through a membrane filter of 0.2 𝜇m pore size. The yeast was washed twice with Millipore water and centrifuged. Both supernatant and yeast were frozen until further processing.

2.1.3 Yeast For further processing, the yeast was freeze dried. Constant weight was subsequently achieved by drying the yeast samples three days at 85 ∘ C in a cabinet drier. As the original yeast cultures were freeze dried, they were only dried three days in a cabinet drier. Since the Cabernet Sauvignon fresh wine was subjected to malo-lactic fermentation, the yeast cell pellet obtained after decantation also contained cells of Oenococcus oeni. Therefore, part of the original cell material of Oenococcus oeni were treated in the same way as the original yeast cultures.

2.2 Sample preparation Because grape juice and the young wines contain potassium and tartaric acid, potassium hydrogentartrate (tartar) precipitates at low temperatures. In order to avoid erroneous determinations of concentrations by precipitates, the liquid samples were chemically homogenized by addition of ethanol (dissolving precipitated proteins) and nitric acid (dissolving tartar) in the ratio sample:ethanol:nitric acid = 40:6:1.

Bereitgestellt von | Johannes Gutenberg Universitaet Mainz Angemeldet Heruntergeladen am | 28.01.16 12:01

336 | Markus Feige et al., Chronological development of element concentrations in grapes

The ripeness of the grapes was determined refractometrically. To this end, the frozen grape juice was thawn and an aliquot of 2 mL was centrifuged for 15 min at 15 600 rpm to separate precipitates. The supernatants were measured at the wine-grower with his refractometer.

2.2.1 Additional experiments In addition to the measurement of element concentrations during growth and ripening of the grapes and during fermentation, the decontamination of Cu during fermentation of must by the yeast Anaferm 4 that was used to inoculate the Riesling, was investigated, the resulting element concentrations after treatment of the wine by bentonite¹ was determined, and it was checked which variations resulted from the application of a yeast food salt. For the former two experiments, grape juice from the pressing plant Heil, Laubus-Eschbach (Germany), was used. For the third experiment, 7 g of a yeast food salt (SIHA PROFERM plus, produced by Begerow) were solved in 1 L of Millipore water and the element concentrations were determined. To investigate the resorption of Cu by the yeast starter culture Anaferm 4, three tests with 100 mL of grape juice and varying amounts of Cu were run, inoculated with Anaferm 4, and fermented at 30 ∘ C under shaking. The fermentation started after 24 h and was finished after 6 d. The young wines were separated from the yeast cells by centrifugation at 9000 rpm. The dried yeasts were subjected to the same irradiation conditions as the Riesling yeast.

2.2.2 Irradiation protocols Short-term and long-term irradiations were performed in the TRIGA Mainz reactor at a thermal power of 100 kW. Short irradiations were conducted in a pneumatic tube system in which the samples could be introduced into the reactor core (𝜑th = 1.7 × 1012 n cm−2 s−1 ) and transported after the end of irradiation to the measurement laboratory within a few seconds. Liquid samples (3 mL), i.e. grape juice, fermenting must, grape juice from the bentonite experiment, grape juice from the Cu experiments, and grape juice from the experiment with the yeast food salt were ir-

radiated for 5 min. Solid samples (70 – 300 mg depending on the activity), i.e. cell material of the yeast starter cultures Anaferm 4 and SIHA 8 as well as of the lactic acid bacterium Oenococcus oeni were irradiated for 3 or 2 min, respectively. Long irradiations were performed for 6 h in the carrousel (𝜑th = 0.7 × 1012 n cm−2 s−1 ) that is rotating around the reactor core so that all samples see the same integral neutron flux. After the end of irradiation, the samples remained over night in the carrousel where short-lived activities were decaying.

2.3 Measurement protocols For the detection of short-lived activities, the irradiated samples were transferred in an inactive counting capsule immediately after the transport into the measurement laboratory. This caused a delay of 1 – 2 min after which a measurement of 10 min was started using a high purity Ge detector and a 4096-channel multi-channel analyzer. The recorded spectra were analyzed with the program Genie2000 [10]. In the 10-min measurement, the nuclides 27 Mg , 28 Al , 49 Ca , 52 V , and 66 Cu could be detected. Details of the measurement protocols are given in Table 1. After a decay time of 3 – 5 h, the same sample was measured again for 30 min to detect 56 Mn . Long-lived nuclides were produced in the 6-h irradiations. The samples were also transferred into inactive measurement capsules and counting of 1 h duration was started after a decay time of 24 h. This way, 24 Na and 42 K could be detected. Nuclides with half-lives of several days and longer could be detected in a measurement of 15 h duration after a decay time of ca. 10 d; these were 46 Sc , 51 Cr , 59 Fe , 60 Co , 65 Zn , and 86 Rb . Details are also contained in Table 1.

2.4 Standards In this work, the INAA was applied as a relative method, i.e. the activities of the nuclides contained in the irradiated samples were compared to activities of standards of known element concentrations and relative amounts 𝑅nuclide were calculated using eq. (1)

𝑅nuclide = 1 Bentonite is a layered silicate containing cations such as Mg2+ , Na+ , and K+ between the layers that can diffuse into the wine. By the resulting negative charges of the silicate layers, positively charged amino acids and proteins can be sorbed and precipitate with the silicate which makes necessary a second decantation.

𝐴 sample 𝑚standard 𝑚sample 𝐴 standard

(1)

where 𝐴 is the activity and 𝑚 is the mass. The concentrations of the elements contained in the samples were calculated by multiplying the relative amounts 𝑅nuclide with the element concentrations of the standard. As standards, liquid single-element ICP-MS standard solutions of 1000 ±

Bereitgestellt von | Johannes Gutenberg Universitaet Mainz Angemeldet Heruntergeladen am | 28.01.16 12:01

Markus Feige et al., Chronological development of element concentrations in grapes

| 337

Table 1: Irradiation measurement protocols, and nuclear properties of the indicator nuclides. Irradiation

Shortterm

Longterm

Decay time

Counting time

1–2 min

10 min

Indicator nuclide 27

Mg Al 49 Ca 52 V 66 Cu 56 Mn 24 Na 42 K 46 Sc 51 Cr 59 Fe 60 Co 65 Zn 86 Rb 28

3–5 h 24 h

30 min 1h

8–12 d

15 h

𝑇1/2

𝐸𝛾 [keV]

Abundance [%]

9.46 min 2.25 min 8.72 min 3.75 min 5.10 min 2.58 h 14.96 h 12.36 h 83.82 d 27.70 d 44.50 d 5.27 a 244.30 d 18.70 d

1014.44 1778.90 3084.40 1434.06 1039.20 846.75 1368.53 1524.67 1120.50 320.08 1099.22 1173.22 1115.52 1077.00

28.00 100.00 92.10 100.00 7.40 98.90 100.00 17.90 100.00 9.83 56.50 100.00 50.75 8.64

3 𝜇g/mL were used to produce multi-element standard solutions of known element concentrations. One multielement standard solution contained those elements that produce short-lived nuclides. The other contained those elements that produce long-lived nuclides. The multielement standard solutions were irradiated under identical conditions, irradiation position and duration, and their activities were determined 𝛾-spectrometrically under the same measurement conditions, sample geometry, counting position and duration. For the liquid samples, a geometry of 2 mL was chosen, for the lyophilic samples 0.5 mL.

2.5 Treatment of uncertainties It was strictly ensured that the samples and the standards had the same geometry and were measured in the same position relative to the detector. Other than that, uncertainties such as volume uncertainties, mass uncertainties, and statistical uncertainties are unavoidable and were treated as follows. The uncertainties associated with the given concentrations were determined by error propagation. Volume uncertainties (Δ𝑉) and mass uncertainties (Δ𝑚) were estimated to be 1% each. The standard deviations 𝜎𝑠 obtained by the threefold measurement of each sample as well as the error resulting from counting statistics 𝜎𝑐 (each in percent) were also taken into account. Together with the uncertainty of the standard 𝜎std a total uncertainty of 2 𝜎total = ±√Δ𝑉2 + Δ𝑚2 + 𝜎𝑠2 + 𝜎𝑐2 + 𝜎std

(2)

results, where the uncertainty of the standard is also determined by error propagation as

𝜎std = ±√Δ𝑉2 + Δ𝑚2 + 𝜎𝑠2 + 𝜎𝑐2 .

(3)

3 Results In this paragraph, the results of the samples are given successively in a chronological sequence. The element concentrations in the grape juices are treated first, followed by those in the fermenting must. The element concentrations in the fermenting yeasts are compared with those in the starter cultures, and the results of the additional experiments on the Cu resorption by the starter culture Anaferm 4 and on the treatment of the wine by bentonite as well as the variations of element concentrations by adding of a yeast food salt are presented.

3.1 Element concentrations in the grape juice In the following, the concentrations of the macro-elements and trace elements in the grape juices during growth and ripening of the grapes are given in 𝜇g/mL. The macroelements (mineral substances) according to Eschnauer occur in a concentration range of 10 to 1000 𝜇g/mL while the trace elements occur in a concentration range of 1 to 10 𝜇g/mL (ppm range) [7]. The must weights that can be seen as a measure of the ripeness of the grapes were determined by refractometry and are plotted as a function of the date of sampling in Figure 1 giving the trend of the

Bereitgestellt von | Johannes Gutenberg Universitaet Mainz Angemeldet Heruntergeladen am | 28.01.16 12:01

338 | Markus Feige et al., Chronological development of element concentrations in grapes

Fig. 1: Must weight of Riesling and Cabernet Sauvignon determined by refractometry in Oechsle degree.

growth of the Riesling and Cabernet Sauvignon grapes. It is clear that the sampled grapes are always coming from single vinegrape plants and the measured values cannot represent the integral of all grapes. In order to get the integral it would have been necessary to sample grapes from all vines in the vineyard. With respect to the economical loss that this would have inflicted on the wine-grower, this was not considered. Thus, even though the data in Figure 1 (and all the element concentrations presented below) are only random tests, we can discuss some visible trends. The must weight is given in the unit ∘ Oechsle (Oechsle degree), i.e. the weight of 1 L of must in g minus the weight of the water of 1000 g/L. 1 L of must with a weight of 1079 g has a must weight of 79∘ Oechsle. The ∘ Oechsle data increased continuously for both grapes until a degree of 90∘ Oechsle was achieved which is typical for the German quality category “Spätlese”. There is a higher degree of ripeness visible for the Spätburgunder grapes compared to the Riesling grapes starting with the second sampling; in September, the data again resemble each other. In Tables 2 and 3 the concentrations for the macroelements Mg, K, and Ca are listed for the juice of both Riesling and Cabernet Sauvignon grapes. Both vines store particularly potassium, magnesium, and calcium in the grapes. The K concentration increases for both grapes parallel to the ripening. It is interesting that the concentration for Mg ions stays approximately constant at about 60 𝜇g/mL in both grapes. The concentrations of the Ca ions reach their maximum of more (Riesling) or less (Cabernet Sauvignon) than 200 𝜇g/mL at the beginning of the growth and decrease with increasing ripeness to about 80 𝜇g/mL for the Riesling and about 60 𝜇g/mL for the Cabernet Sauvignon. Whether this decrease is due to a dilution by an increasing uptake of water with increas-

Table 2: Macro-element concentrations in the juice of Riesling grapes. Date of sampling 2007-07-05 2007-07-19 2007-07-30 2007-08-22 2007-09-06 2007-09-21 2007-09-29

Mg

Concentration [𝜇g/mL] K

Ca

68.2 ± 6.2 61.9 ± 5.5 67.0 ± 5.6 61.8 ± 5.0 50.7 ± 4.6 54.0 ± 4.3 57.9 ± 4.8

596 ± 31 761 ± 36 1147 ± 56 988 ± 51 1308 ± 70 1515 ± 80 1564 ± 98

233 ± 21 260 ± 24 250 ± 24 132 ± 13 79.2 ± 7.9 81.9 ± 7.9 96.9 ± 9.3

Table 3: Macro-element concentrations in the juice of Cabernet Sauvignon grapes. Date of sampling 2007-07-05 2007-07-19 2007-07-30 2007-08-22 2007-09-06 2007-09-21

Mg

Concentration [𝜇g/mL] K

Ca

70.2 ± 5.5 58.2 ± 5.3 48.3 ± 4.5 55.6 ± 4.4 50.7 ± 4.2 56.9 ± 4.8

668 ± 29 778 ± 42 1397 ± 41 1258 ± 61 1807 ± 90 1623 ± 76

172 ± 16 147 ± 14 102 ± 11 88.9 ± 8.9 59.2 ± 6.1 63.1 ± 6.9

ing ripeness or to an active transport of the cations out of the berries, will be investigated in Sect. 4. In Tables 4 and 5, the concentrations of the trace elements in the juice of Riesling and Cabernet Sauvignon grapes are listed. While there is an enrichment of Na by a factor of two in both grapes, that is similar to the variation of the Ca concentrations, there is a decrease of the Mn concentrations by a factor of two to three. The concentrations of Al and Cu are in the sub-𝜇g/mL range with Al being stronger incorporated into the grapes than Cu. The con-

Bereitgestellt von | Johannes Gutenberg Universitaet Mainz Angemeldet Heruntergeladen am | 28.01.16 12:01

Markus Feige et al., Chronological development of element concentrations in grapes

| 339

Table 4: Trace-element concentrations in the juice of Riesling grapes. Date of sampling 2007-07-05 2007-07-19 2007-07-30 2007-08-22 2007-09-06 2007-09-21 2007-09-29

Concentration [𝜇g/mL] Mn Zn

Na

Al

2.46 ± 0.08 1.90 ± 0.05 3.35 ± 0.09 3.32 ± 0.10 3.47 ± 0.13 2.85 ± 0.11 4.76 ± 0.13

0.45 ± 0.03 0.50 ± 0.03 0.51 ± 0.03 0.45 ± 0.04 0.56 ± 0.04 0.62 ± 0.03 0.59 ± 0.03

2.84 ± 0.08 2.61 ± 0.08 3.14 ± 0.08 1.48 ± 0.05 1.08 ± 0.03 0.85 ± 0.02 0.92 ± 0.02

0.87 ± 0.04 0.65 ± 0.03 0.85 ± 0.04 0.45 ± 0.03 0.37 ± 0.04 0.32 ± 0.04 0.40 ± 0.03

Cu

Rb

0.31 ± 0.09 0.35 ± 0.11 0.39 ± 0.15 0.29 ± 0.08 0.33 ± 0.17 0.35 ± 0.09 0.40 ± 0.03

0.11 ± 0.01 0.13 ± 0.01 0.18 ± 0.01 0.28 ± 0.01 0.47 ± 0.03 0.25 ± 0.02 0.13 ± 0.01

Table 5: Trace-element concentrations in the juice of Cabernet Sauvignon grapes. Date of sampling 2007-07-05 2007-07-19 2007-07-30 2007-08-22 2007-09-06 2007-09-21

Concentration [𝜇g/mL] Mn Zn

Na

Al

1.76 ± 0.05 1.99 ± 0.07 2.77 ± 0.07 3.09 ± 0.11 3.20 ± 0.09 2.51 ± 0.08

0.46 ± 0.03 0.51 ± 0.03 0.48 ± 0.05 0.49 ± 0.03 0.66 ± 0.03 0.65 ± 0.04

2.63 ± 0.07 1.78 ± 0.05 1.45 ± 0.04 1.32 ± 0.04 0.97 ± 0.03 0.82 ± 0.02

0.78 ± 0.05 0.58 ± 0.03 0.53 ± 0.03 0.29 ± 0.02 0.40 ± 0.04 0.32 ± 0.03

centration of Al increases in both grapes from 0.45 𝜇g/mL to 0.60 𝜇g/mL, while the concentration of Cu is around 0.30 𝜇g/mL. This corresponds to the natural uptake of Cu through the roots as in the selected vineyards Cu sulphate heptahydrate was not applied as fungicide. Both, Riesling and Cabernet Sauvignon grapes in the beginning contained about 0.80 𝜇g/mL Zn which decreased with increasing ripeness to 0.30 𝜇g/mL. In the Cabernet Sauvignon grapes, the concentration of Rb increased with time from 0.21 𝜇g/mL to 0.41 𝜇g/mL; in the Riesling grapes, there was an increase from 0.11 𝜇g/mL to 0.47 𝜇g/mL followed by a decrease to 0.13 𝜇g/mL towards the end.

3.2 Element concentrations in the fermenting musts In this section, the element concentrations of the young Riesling wines that were fermented under varying conditions are given in Tables 6 to 11 and in Figure 2 as well as those of the Cabernet Sauvignon wine in the stage of malolactic fermentation. The first Riesling sample is a must sample taken on the 2007-10-10 during the pressing of the Riesling grapes. It contained the starting concentrations for both the spontaneously fermented Riesling as well as for the Riesling inoculated with a starter yeast culture. The first Cabernet Sauvignon sample was taken on the 2007-1009.

Cu

Rb

0.25 ± 0.07 0.30 ± 0.07 0.40 ± 0.14 0.31 ± 0.08 0.30 ± 0.12 0.38 ± 0.10

0.21 ± 0.02 0.24 ± 0.02 0.30 ± 0.02 0.20 ± 0.02 0.33 ± 0.02 0.41 ± 0.02

3.2.1 Riesling inoculated with a yeast starter culture In Table 6, the concentrations of the macro-elements Mg, K, and Ca during fermentation are listed, in Table 7, the concentrations of the trace elements. The last sample was taken on the 2008-02-11 after the treatment of the wine with bentonite before filling the wine into bottles. In Table 7, it is obvious that during the fermentation, Cu could only

Table 6: Macro-element concentrations of the Riesling inoculated with a yeast starter culture. Date of sampling 2007-10-10 2007-10-12 2007-10-15 2007-10-17 2007-10-19 2007-10-22 2007-10-24 2007-10-26 2007-10-29 2007-10-31 2007-11-07 2007-11-16 2008-01-04 2008-02-11

Mg

Concentration [𝜇g/mL] K

Ca

60.5 ± 4.9 58.2 ± 6.0 55.0 ± 4.8 62.3 ± 5.2 59.9 ± 4.9 63.4 ± 5.4 63.9 ± 5.2 67.5 ± 5.8 65.9 ± 5.5 67.8 ± 5.4 66.6 ± 6.0 62.6 ± 5.4 63.8 ± 5.1 67.2 ± 5.6

1195 ± 60 1524 ± 102 1504 ± 94 1392 ± 84 1209 ± 84 1097 ± 68 985 ± 64 1007 ± 64 904 ± 58 894 ± 57 875 ± 53 797 ± 49 634 ± 41 573 ± 36

98.7 ± 9.9 89.1 ± 8.6 94.4 ± 9.9 105 ± 10 102 ± 10 110 ± 11 113 ± 12 110 ± 10 103 ± 10 104 ± 10 98.3 ± 9.4 88.2 ± 8.5 80.4 ± 8.4 95.1 ± 9.27

Bereitgestellt von | Johannes Gutenberg Universitaet Mainz Angemeldet Heruntergeladen am | 28.01.16 12:01

340 | Markus Feige et al., Chronological development of element concentrations in grapes

Fig. 2: Concentration of the macro element K during fermentation of the Riesling inoculated with the yeast starter culture Anaferm 4. The last data point refers to the concentration after the treatment of the wine with bentonite. Table 7: Trace-element concentrations of the Riesling inoculated with a yeast starter culture. Date of sampling 2007-10-10 2007-10-12 2007-10-15 2007-10-17 2007-10-19 2007-10-22 2007-10-24 2007-10-26 2007-10-29 2007-10-31 2007-11-07 2007-11-16 2008-01-04 2008-02-11

Concentration [𝜇g/mL] Mn Zn

Na

Al

4.67 ± 0.13 5.13 ± 0.16 5.16 ± 0.17 5.49 ± 0.14 4.99 ± 0.22 5.50 ± 0.16 5.04 ± 0.16 5.47 ± 0.19 5.24 ± 0.18 4.99 ± 0.15 5.37 ± 0.14 5.28 ± 0.15 5.20 ± 0.17 14.49 ± 0.44

0.63 ± 0.05 0.75 ± 0.05 0.67 ± 0.04 0.81 ± 0.08 0.80 ± 0.04 0.79 ± 0.04 0.84 ± 0.06 0.81 ± 0.04 0.89 ± 0.05 0.86 ± 0.06 0.84 ± 0.04 0.81 ± 0.05 0.90 ± 0.05 1.65 ± 0.08

1.45 ± 0.04 1.40 ± 0.04 1.42 ± 0.04 1.57 ± 0.04 1.53 ± 0.04 1.56 ± 0.04 1.60 ± 0.04 1.64 ± 0.04 1.62 ± 0.04 1.64 ± 0.04 1.61 ± 0.05 1.50 ± 0.04 1.56 ± 0.04 1.61 ± 0.05

1.20 ± 0.05 1.80 ± 0.10 0.73 ± 0.06 1.26 ± 0.10 1.21 ± 0.05 1.37 ± 0.10 1.47 ± 0.07 1.60 ± 0.09 1.68 ± 0.13 1.89 ± 0.06 2.03 ± 0.11 2.00 ± 0.07 2.00 ± 0.07 1.94 ± 0.08

be detected until the 2007-10-22; in the following samples, the Cu concentration was below the detection limit. In Figure 2 the variation of the K concentration during fermentation and ripeness of the wine is depicted. It is characterized by a strong decrease during the first twelve days by 520 𝜇g/mL. In the following weeks, the decrease continues more slowly, i.e. by another 400 𝜇g/mL in the time between the 2007-10-24 and the 2008-01-04. After the separation of the wine from the yeast by decantation, the treatment of the wine by bentonite has little influence on the final K concentration. The amount of Ca increases slightly in the early phase of fermentation from 100 𝜇g/mL to 113 𝜇g/mL and decreases later to 80 𝜇g/mL. The bentonite treatment increases the concentration again to the amount present in the beginning. The amount of Mg changes not signif-

Cu

Rb

0.47 ± 0.11 0.41 ± 0.07 0.37 ± 0.14 0.39 ± 0.08 0.38 ± 0.12 0.32 ± 0.10

0.64 ± 0.03 0.72 ± 0.04 0.78 ± 0.04 0.74 ± 0.04 0.69 ± 0.04 0.65 ± 0.05 0.62 ± 0.03 0.65 ± 0.03 0.60 ± 0.05 0.63 ± 0.03 0.61 ± 0.04 0.56 ± 0.03 0.51 ± 0.03 0.46 ± 0.03

icantly. The amount of Na stays constant during fermentation while it increases by a factor of three through the treatment with bentonite. Besides the K concentration, the Zn-ion concentration varies strongly in the beginning of the fermentation, reaches a minimum on the 2007-10-15 and increases within three weeks to a constant value of 2.0 𝜇g/mL which is not significantly changed by the bentonite treatment.

3.2.2 Spontaneously fermented Riesling The concentrations of the macro-elements and trace elements are listed in Tables 8 and 9. The fermentation was stopped already by mid of December and the wine was sulfurized for stabilization. Similarly to the Riesling inoc-

Bereitgestellt von | Johannes Gutenberg Universitaet Mainz Angemeldet Heruntergeladen am | 28.01.16 12:01

Markus Feige et al., Chronological development of element concentrations in grapes |

Table 8: Macro-element concentrations of the spontaneously fermented Riesling. Date of sampling 2007-10-10 2007-10-15 2007-10-19 2007-10-22 2007-10-24 2007-10-26 2007-10-29 2007-10-31 2007-11-07 2007-11-16 2008-01-04

Mg

Concentration [𝜇g/mL] K

Ca

60.5 ± 4.9 57.7 ± 5.4 56.1 ± 4.5 55.7 ± 4.8 55.6 ± 5.6 55.5 ± 4.6 57.1 ± 4.7 58.7 ± 5.0 58.0 ± 6.0 58.5 ± 4.7 58.7 ± 4.6

1195 ± 60 1362 ± 63 1257 ± 84 1214 ± 61 1111 ± 53 1149 ± 66 1024 ± 46 975 ± 43 932 ± 46 809 ± 41 721 ± 36

98.7 ± 9.9 100 ± 10 91.6 ± 9.1 98.3 ± 9.9 97.4 ± 9.7 96.6 ± 9.5 102 ± 10 98.0 ± 9.5 100 ± 10 97.1 ± 9.7 97.5 ± 10.0

ulated by the yeast culture Anaferm 4, there is a marked decrease of the K concentration, here from 1195 𝜇g/mL to 721 𝜇g/mL. The other macro-element concentrations are roughly constant at 100 𝜇g/mL for Ca and at 60 𝜇g/mL for Mg. While the amount of Rb stays approximately constant, there is again a significant increase of the Zn concentration from 1.20 𝜇g/mL to 2.86 𝜇g/mL.

3.2.3 Cabernet Sauvignon The Cabernet Sauvignon must was subject to fermentation for three days, was drawn off, and was then transferred for malo-lactic fermentation in a wooden barrel. During the transfer, a first sample of the fermented must could be taken. Lactic acid bacteria play a decisive role in the removal of malic acid which can unfavourably influence the acidity and the taste of the wine. The ratio of malic acid

to tartaric acid is about 1:1 in the growing berries and decreases with increasing ripeness [11]. In general, after fermentation, red wines are subjected to malo-lactic fermentation – to reduce the total acid content. On the contrary, in white wines, a comparably higher acid content is desirable. For malo-lactic fermentation, the young wine was inoculated with the lactic acid bacteria Oenococcus oeni after ethanolic fermentation. In this case, the lactic acid bacteria do not compete with the wine yeast for sugars. They use only partly the remaining sugars which leads to a biological stabilization of the wine. An additional nutrition supply such as vitamins and amino acids for the oenococci is provided during lysis of yeast cells. Lactic acid bacteria such as Oenococcus oeni are available as freeze-dried starter cultures. Malo-lactic fermentation has several advantages. Firstly, the acid content of the wine is markedly reduced. Secondly, the amount of sulphurous acid can be reduced which is required to inhibit growth of microorganisms and to provide reducing conditions as Oenococcus oeni can eliminate acetaldehyde and pyruvate as well as uses the remaining sugars. In addition, in red wines, the colour stability can be increased. The mechanism of malo-lactic fermentation consists of the chemical transposition of malic acid to lactic acid whereby one of the two carboxylic acid groups of the malic acid is removed by the malolactic enzyme involving Mn2+ as cofactor in the active centre: L-malate→L-lactate + CO2 . In Tables 10 and 11, the concentrations of the macroelements and the trace element composition of fermented must during malo-lactic acid fermentation are listed, respectively. One recognises that the K content during malolactic acid fermentation is reduced by less than 15%. The content of Ca ions decreases by about 50% while the Mg content stays constant within the uncertainty limits. The decrease of the Na concentration yields 75% of the concen-

Table 9: Trace-element concentrations of the spontaneously fermented Riesling. Date of sampling 2007-10-10 2007-10-15 2007-10-19 2007-10-22 2007-10-24 2007-10-26 2007-10-29 2007-10-31 2007-11-07 2007-11-16 2008-01-04

Concentration [𝜇g/mL] Mn Zn

Na

Al

4.67 ± 0.13 4.52 ± 0.15 3.65 ± 0.14 3.73 ± 0.12 3.85 ± 0.14 3.71 ± 0.18 4.14 ± 0.12 3.87 ± 0.11 4.00 ± 0.12 3.91 ± 0.14 4.16 ± 0.13

0.63 ± 0.05 0.59 ± 0.04 0.51 ± 0.04 0.51 ± 0.03 0.51 ± 0.03 0.52 ± 0.05 0.68 ± 0.04 0.55 ± 0.03 0.54 ± 0.03 0.65 ± 0.03 0.63 ± 0.04

1.45 ± 0.04 1.42 ± 0.04 1.39 ± 0.04 1.40 ± 0.04 1.45 ± 0.04 1.40 ± 0.04 1.47 ± 0.04 1.45 ± 0.04 1.46 ± 0.04 1.42 ± 0.04 1.43 ± 0.04

1.20 ± 0.05 1.46 ± 0.06 1.23 ± 0.06 1.32 ± 0.06 1.39 ± 0.09 1.50 ± 0.10 1.72 ± 0.08 1.61 ± 0.11 1.74 ± 0.06 2.76 ± 0.10 2.86 ± 0.11

341

Cu

Rb

0.47 ± 0.11 0.86 ± 0.09 0.71 ± 0.13 0.72 ± 0.11 0.63 ± 0.09 0.63 ± 0.09 0.65 ± 0.09 0.77 ± 0.09 0.57 ± 0.09 0.50 ± 0.09

0.64 ± 0.03 0.68 ± 0.04 0.66 ± 0.04 0.63 ± 0.03 0.64 ± 0.04 0.66 ± 0.04 0.60 ± 0.03 0.63 ± 0.03 0.58 ± 0.02 0.58 ± 0.04 0.52 ± 0.03

Bereitgestellt von | Johannes Gutenberg Universitaet Mainz Angemeldet Heruntergeladen am | 28.01.16 12:01

342 | Markus Feige et al., Chronological development of element concentrations in grapes

Table 10: Macro-element concentrations in Cabernet Sauvignon during malolactic fermentation. Date of sampling 2007-10-09 2007-10-26 2007-10-31 2007-1107 2007-11-16 2008-01-04

Mg

Concentration [𝜇g/mL] K

Ca

75.5 ± 6.0 87.2 ± 6.6 83.4 ± 7.7 80.7 ± 6.3 82.1 ± 6.4 82.7 ± 6.8

1393 ± 92 1646 ± 105 1503 ± 65 1495 ± 74 1378 ± 62 1396 ± 66

111 ± 11 76.8 ± 7.5 71.8 ± 7.2 65.7 ± 7.5 62.4 ± 6.9 54.1 ± 5.6

tration on the 2007-10-26. The concentration of the Mn ions and Rb ions during malo-lactic acid fermentation stays constant within the error limits. This is similar for Al. The concentration of Zn ions in the red wine during malo-lactic acid fermentation increases until 2007-10-31 and decreases slighty thereafter. Cu ions could not be detected during malo-lactic acid fermentation as their concentration was already decreased during fermentation below the detection limit.

3.3 Element concentrations in yeast In this section, the element concentrations are presented that are accumulated in the yeast cells of the inoculated Riesling and of the Cabernet Sauvignon after rackening of the wine. They are compared to the concentrations in the starter yeast cultures. In case of the Riesling, the starter culture Anaferm 4, for the Cabernet Sauvignon the yeast culture SIHA 8, and for the malo-lactic fermentation the Oenococcus oeni culture Uvaferm BETA was used.

3.3.1 Riesling Table 12 contains the element concentrations that could be detected in the fermenting yeast as well as in the starter

culture. It is evident that the concentrations can both strongly decrease and increase. For Na, the content in the fermenting yeast is only a small fraction of the original content in the starter culture. The content of Al is reduced by a factor of two. In contrast to this, the concentration of Cu in the fermenting yeast increases by a factor of nine to ten reaching a concentration of slightly less than 90 𝜇g/g. The content of K in the yeast increases by a factor of three during fermentation. The content of Zn and Mg decreases from 104 to 4 𝜇g/mL and from 1146 to 147 𝜇g/mL in the course of the fermentation, respectively. For Zn, this is evidently reflected in the marked increase in the wine during fermentation, see Table 7. Ca is enriched in the yeast by a factor of five. The variations of the concentrations of Mn and Rb during fermentation are small. Sc and Co concentrations before and after the fermentation are in the ppb or ppm range, respectively. Generally, because the starter culture contains high concentrations of elements such as K, Na, or Zn, it can release a large fraction of these to the medium. In addition to the release of such elements to the fermenting must, their concentrations in the yeast are decreased because of the cell division. Vice versa, the yeast cells during fermentation take up, besides sugar and other nutritive substances, certain minerals that they remove from the medium, e.g., K and Ca.

3.3.2 Cabernet Sauvignon As the Cabernet Sauvignon was inoculated with a starter yeast culture for ethanolic fermentation as well as with Oenococcus oeni during malo-lactic fermentation, it is necessary to compare the precipitated fermenting cells with both starter cultures, which is done in Table 13. Also here, the observations described in Sect. 3.3.1 concerning increase or decrease of concentrations of the various cations in the wine hold, except that now, there is a double influence by the inoculation of the wine by two starter cultures.

Table 11: Trace-element concentrations in Cabernet Sauvignon during malolactic fermentation. Date of sampling 2007-10-09 2007-10-26 2007-10-31 2007-1107 2007-11-16 2008-01-04

Na

Al

Concentration [𝜇g/mL] Mn

Zn

Rb

4.48 ± 0.16 4.88 ± 0.15 3.68 ± 0.13 3.53 ± 0.13 3.15 ± 0.13 3.27 ± 0.11

0.83 ± 0.04 1.09 ± 0.05 1.05 ± 0.06 1.02 ± 0.05 1.04 ± 0.07 1.04 ± 0.05

1.63 ± 0.05 1.69 ± 0.05 1.66 ± 0.05 1.60 ± 0.05 1.63 ± 0.04 1.61 ± 0.04

2.21 ± 0.11 2.74 ± 0.11 2.84 ± 0.15 2.65 ± 0.09 2.60 ± 0.09 2.52 ± 0.12

0.76 ± 0.05 0.84 ± 0.04 0.84 ± 0.04 0.80 ± 0.04 0.80 ± 0.03 0.79 ± 0.03

Bereitgestellt von | Johannes Gutenberg Universitaet Mainz Angemeldet Heruntergeladen am | 28.01.16 12:01

Markus Feige et al., Chronological development of element concentrations in grapes

Table 12: Comparison of the element concentrations in the fermenting yeast of the Riesling must and in the freeze-dried starter culture Anaferm 4. Element

Na Al Cu K Zn Mg Ca Mn Rb Sc Co

Concentration [𝜇g/mL] Fermenting yeast Starter culture 4.18 ± 0.49 52.9 ± 5.4 86.9 ± 3.5 16310 ± 2140 3.55 ± 0.38 147 ± 95 5594 ± 479 9.54 ± 0.61 12.2 ± 1.4 0.0016 ± 0.0002 0.014 ± 0.004

1185 ± 47 118 ± 8 9.80 ± 2.69 5425 ± 437 104 ± 11 1082 ± 55 1146 ± 84 6.17 ± 0.35 15.4 ± 1.2 0.0151 ± 0.0006 1.70 ± 0.05

Table 13: Comparison of the concentrations in the fermenting yeast of the Cabernet Sauvignon must and the Saccharomyces starter culture SIHA 8 and the Oenococcus starter culture Uvaferm BETA. Element

Concentration [𝜇g/mL] Fermenting Starter cultures yeast SIHA 8 Uvaferm BETA

Mn Rb Zn Sc Co Al K Na

27.1 ± 2.4 17.7 ± 1.5 15.8 ± 1.7 0.203 ± 0.006 0.062 ± 0.013 3759 ± 348 13606 ± 1044 1747 ± 61

6.34 ± 0.42 19.8 ± 1.6 114 ± 12

242 ± 22 4.09 ± 0.41 5.40 ± 0.84

0.342 ± 0.009 58.3 ± 4.1 6883 ± 591 611 ± 18

0.060 ± 0.004 6195 ± 703 1191 ± 104 7288 ± 652

| 343

were added. One was mixed with 0.50 mL (solution B), the other with 1.0 mL (solution C). This resulted in Cu concentrations of 4.66 𝜇g/mL in solution B and 9.33 𝜇g/mL in solution C. After fermentation, these concentrations decreased to 1.66 𝜇g/mL and 2.55 𝜇g/mL, respectively, which corresponds to a reduction of the Cu concentration to 35% in solution B and to 27% in solution C. Compared to the starter culture with a Cu concentration of 9.80 𝜇g/g dry mass (ppm), the Cu in the dried yeast of solution B was enriched almost 30 fold (293 𝜇g/g) and in the yeast of solution C 60 fold (603 𝜇g/g). This is in qualitative agreement with the “quantitative decrease” reported in [6–8]. It is also consistent with the observation concerning the Cu concentration in the yeast after fermentation reported in Sect. 3.3.1.

3.4.2 Addition of a yeast food salt In order to investigate which metal ions end up in the wine by adding a yeast food salt to the must during fermentation, 7 g of a food salt provided by the wine-grower were suspended in one litre of Millipore water. Three mL each of this suspension were subject of a short-term and a long-term irradiation resulting in concentrations that were highest for Mg (52.1 𝜇g/mL), K (44.8 𝜇g/mL) and Zn (12.5 𝜇g/mL) and lowest for Mn (0.021 𝜇g/mL) and Co (0.0033 𝜇g/mL). Cu and Ca could not be detected so that these concentrations must have been below the limits of detection, i.e. 0.19 𝜇g/mL and 5.11 𝜇g/mL, respectively.

3.4 Additional experiments

3.4.3 Bentonite treatment

The biosorption of Cu by wine-associated lactobacilli has been demonstrated [8]. In this paragraph, it is reported how much Cu can be removed from the must fermented with the yeast starter culture Anaferm 4. Subsequently, the consequences of a treatment of the fermenting wine with a yeast food salt and by a treatment with betonite are presented.

In this experiment, 413.1 mg of bentonite were suspended in 10 mL of Millipore water and were allowed to swell for one hour. The supernatant was decanted. The swollen bentonite was suspended in 150 mL of sterilized grape juice (assuming that grape juice is chemically somewhat related to wine). After two days, the bentonite was separated from the juice by centrifugation at 9000 rpm for 15 min. Compared to the original grape juice, the Na concentration increased from 23.3 𝜇g/mL to 30.7 𝜇g/mL, i.e. by less than one third. Also Co, Al, and Ca are enriched in the treated juice. The amount of Co increases from 0.006 𝜇g/mL to 0.010 𝜇g/mL, the amount of Al from 3.15 𝜇g/mL to 5.61 𝜇g/mL. Mg is enriched starting from 63.8 𝜇g/mL by one eights, Ca starting from 129 𝜇g/mL by one quarter. The amount of Mn does not change significantly. Only the concentrations of Rb and Zn are slightly reduced by roughly 10%. All in all, the treatment by ben-

3.4.1 Removal of Cu by Anaferm 4 In this experiment, three samples of grape juice were mixed with various amounts of a Cu solution and fermented with the starter culture Anaferm 4. To one sample, no Cu solution was added (solution A) and served as background sample. To the other two samples different volumes of a Cu solution of concentration 1000 ± 3 𝜇g/mL

Bereitgestellt von | Johannes Gutenberg Universitaet Mainz Angemeldet Heruntergeladen am | 28.01.16 12:01

344 | Markus Feige et al., Chronological development of element concentrations in grapes

tonite does not change any of the cation concentrations in a dramatic way.

4 Discussion 4.1 Grapes The vine roots take up mineral substances from the soil that the vine needs for its metabolism and anabolism. The minerals are unequally distributed in different parts of the plant. The mineral substances fulfil various tasks. Firstly, they can be involved directly in the metabolism as cofactors in metalloenzymes [6]. Secondly, they are involved in maintaining the osmotic pressure and the electro neutrality [4]. In addition, they can influence cell compartments by their electrical charge and play a role in energy generation. In the beginning of the growth of the grapes, the content of mineral substances that are involved in cell growth and structure increases and stays constant in the ripeness phase [2]. In the latter, mainly water and sugars are accumulated in the grapes, organic acids like malic acid are decomposed. Through this, the must weight increases (see Figure 1). In order to investigate whether the decreases of concentrations of several cations in the ripeness phase is caused by a dilution through the increased uptake of water or whether the total mass of the respective ions within the berries decreases, the following way was persued. Based on the masses of the grapes and on the masses of the resulting musts the percentage water content was determined. Using the known densities that were already used to calculate the element concentrations, the must volume in relation to 100 g mass of the grapes was deduced and correlated with the respective element

concentration. This way, the element masses contained in 100 g mass of the grapes could be determined which allow to determine whether the total contents decreased or whether a decrease of the concentrations by dilution was the case. In the following, examples for the masses of elements found in Riesling and Cabernet Sauvignon normalized to 100 g mass of the grapes are presented. Within this exercise, uncertainties of 20% were assumed as in the manual pressing of the grapes, variable forces may have been applied. Figure 3 shows the mass of K ions normalized to 100 g mass of the berries during ripening of Riesling and Cabernet Sauvignon grapes. Figure 4 depicts the mass of Ca ions normalized to 100 g mass of the berries during ripening of Riesling and Cabernet Sauvignon grapes. The increase of the K concentration during growth of the grape as documented in Tables 2 and 3, is related to the fact that K is a decisive factor associated with the uptake of sugar by the berries. By the stronger uptake of K, the transparency of the cell walls and membranes is increased and the incorporation of substances is assisted [2]. By a comparison of the K concentrations (Tables 2 and 3) with the total mass of K in the must of 100 g of grapes, Figure 3, an interesting detail might be indicated. While the K concentration at the end of the ripening phase does not increase that much any more, Figure 2 indicates a further increase in the mass of the K ions. Based on this observation, a dilution effect might be postulated based on the increased uptake of water into the berries in the ripeness phase. In Tables 2 and 3, a strong decrease of the Ca concentrations as a function of time is documented. The concentrations of Ca decrease with increasing concentrations of K (“Kalk-Kali-Gesetz” [12]). By the parallel decrease of the Ca concentrations and the Ca masses, Figure 4, a dilution

Fig. 3: Absolute mass of K ions during ripeness of the Riesling and Cabernet Sauvignon grapes normalized to 100 g mass of the berries.

Bereitgestellt von | Johannes Gutenberg Universitaet Mainz Angemeldet Heruntergeladen am | 28.01.16 12:01

Markus Feige et al., Chronological development of element concentrations in grapes

| 345

Fig. 4: Absolute mass of Ca ions during ripeness of the Riesling and Cabernet Sauvignon grapes normalized to 100 g mass of the berries.

Fig. 5: Decrease of the must weight during fermentation of the Riesling inoculated with the yeast starter culture Anaferm 4 and of the spontaneously fermented Riesling.

effect can be excluded. There must be an active transport of Ca ions out of the berries. At the beginning of the growth of the grapes, all grapes are green and are performing photosynthesis. With increasing ripeness of the grapes, the chlorophyll is reduced and colour substances are built into the skin of the grapes [13]. For the photosynthesis, Mg ions are needed that are the central ions in chlorophyll. However, Mg has additional functions. It serves as phosphate carrier and is important for enzymatic metabolisms [12]. The slight decrease in the Mg concentrations documented in Tables 2 and 3 must be confronted with a constant mass of Mg normalized to 100 g mass of the grapes, indicating clearly a dilution effect. The concentration of Na ions increases in both grapes as is documented in Tables 4 and 5. In the representation of the total mass of Na, this trend is also visible.

The decrease of the concentrations of Mn ions in Tables 4 and 5 cannot be explained by a dilution effect alone, because the total mass of Mn normalized to 100 g mass of the grapes also decreases. This must be attributed to an active transport of the Mn ions out of the grapes. Cu is, like Mn, an essential trace element that is taken up by the plant and serves the growth. The total masses in the Riesling and Cabernet Sauvignon grapes stay approximately constant. A slight dilution effect is visible for Al. Zn ions are indispensable growth factors and a shortage causes disorder in the development of the grapes [12]. The development of the concentrations is in parallel to that of Mn. Its decrease cannot be attributed to a dilution as the total mass normalized to 100 g mass of the grapes also decreases. The concentrations of Rb and the total masses of Rb preclude a dilution effect.

Bereitgestellt von | Johannes Gutenberg Universitaet Mainz Angemeldet Heruntergeladen am | 28.01.16 12:01

346 | Markus Feige et al., Chronological development of element concentrations in grapes

Fig. 6: Concentration of the trace element Zn during fermentation of the Riesling inoculated with the yeast starter culture Anaferm 4. a) addition of the yeast culture, b) cell division, c) addition of a yeast food salt, d) decreasing fermentation, e) concentration after the treatment with bentonite.

4.2 Fermenting Musts As we have seen above, the fermentation has a decisive influence on the composition of the must and consequently on the corresponding wines. The yeast cells take up nutritive substances from the must that are essential for their metabolism and growth. On the other hand, they excrete their metabolic end products. In respect to wine making, the most important product is ethyl alcohol produced during fermentation of the sugar contained in the must. Besides, byproducts such as biogenetic amines, esters, aldehydes etc. are formed [5]. By autolysis of the yeast cells at the end of the fermentation, further cell contents are liberated that influence the growth of malo-lactic bacteria and also the sensoric properties of the wine. In Figure 5, the decrease of the must mass of the two Riesling wines during fermentation is depicted as it was documented by the wine-grower. There is a significant difference visible in the fermentation behaviour between a spontaneous fermentation and an induced fermentation with a selected yeast starter culture. Metal ions play a significant role in the progress of the fermentation. They are contained in the must in varying concentrations and serve as cofactors groups in enzymes such as Zn, Cu, or Mn or serve to regulate the metabolism [6]. In Figure 2, the variation of the concentrations of K in the inoculated Riesling is shown which represent the largest fraction of the metal ions contained in the must. The large decrease of the K concentration in the course of the fermentation can be partly explained by the precipitation of tartar. The increase of the K concentration at the beginning of the fermentation can be explained by

the fact that the must sample was stored in the refrigerator until its processing which already reduced the solubility of the K hydrogen tartrate. As there is a higher increase of the K concentration right after the beginning of the fermentation in the inoculated Riesling compared to the spontaneously fermented Riesling, the influence of the added yeast culture cannot be excluded as it contains 5425 ppm K. The concentration of K in the Cabernet Sauvignon seems to increase slightly similar to the Riesling case which was also caused by the storage of the fermented solution in the refrigerator before processing. Even though the K concentration does not significantly change during the malolactic fermentation, K is enriched in the precipitate. In a comparison of the Ca concentrations in the two differently fermented Rieslings, it is obvious that the Ca concentration in the spontaneously fermented Riesling does not change within the uncertainties, while in the Riesling inoculated with a starter culture, the Ca concentration increases significantly. As the Ca concentration in the starter yeast culture is 1146 ppm, it is supposed that Ca ions are released into the must. The following slight decrease of the Ca concentration is associated with the formation of Ca tartrate that accumulates in the precipitate. Even though Mg is needed by the yeast cells, the Mg concentration is by far sufficient so that no significant decrease of Mg is observed neither in the fermenting Riesling nor in the Cabernet Sauvignon. By the treatment of the Riesling wines with bentonite, the concentrations of Na, Ca, and Al is increased in the finished wines. A significant variation of the concentrations of the other elements is not visible.

Bereitgestellt von | Johannes Gutenberg Universitaet Mainz Angemeldet Heruntergeladen am | 28.01.16 12:01

Markus Feige et al., Chronological development of element concentrations in grapes

Mn is the prosthetic group in various enzymes and is essential in the decarboxylation of malic acid into lactic acid by the malo-lactic enzyme [14]. The necessary concentration is by far present so that no significant decrease is observed. While the addition of a yeast starter culture increases the Al concentration, Cu concentrations decrease in the fermenting wines whereby the decrease is stronger with increasing Cu concentration. Even though Brandolini reports about Cu-resistant yeast cultures, that are fermenting even in high Cu concentrations, whereby under the reducing conditions during fermentation Cu ions are reduced and are incorporated into the cells [15]. At the low Cu concentration in these wines, this cannot be the only reason for the decreasing Cu concentrations. The removal of Cu from the wines under fermenting condition may be associated with the formation of sulfite and sulfide from sulfate ions by reduction and these catch Cu ions and other heavy metals by the formation of rarely soluble sulfides [5]. The element whose concentration varies most besides K and Cu, is Zn. Zn is a growth factor and plays an important role in the fermentation. The Zn concentration increases rapidly upon addition of the yeast starter culture, see Figure 6; by the subsequent cell division and the increasing alcohol content, it decreases rapidly. The following increase of the Zn concentration is related to the addition of yeast food salt that contains apart from nitrogen compounds unsaturated fatty acids that are essential for building up cells but also Zn. This is enriched in the cells and is subsequently released into the wine by the autolysis of the cells at the end of fermentation. The result of the addition of the yeast food salt to the fermenting wine is largest for Zn as Zn is contained in the must to 1.2 𝜇g/mL only and the food salt contains 12.5 𝜇g/g Zn. For the inoculated Riesling and for the Cabernet Sauvignon, some element concentrations are increased by the addition of the starter cultures. These are in particular for the Cabernet Sauvignon K, Na, and Al. Generally, the existence of heavy-metal ions in the must or wine does not reduce its quality. On the contrary, by catching unwanted fermentation side products, they can even increase the quality of the wine. Heavy metals such as Pb, Fe, Cu, Cd, and Hg are almost completely removed by the precipitation of rarely soluble salts or resorption by the yeast during fermentation so that they can no more be detected [5, 7]. A biosorption by microbial cells may also be possible [8].

| 347

5 Summary The successive development of element concentrations during growth and ripening of wine grapes has been described in the literature only for the macro elements Mg, K, and Ca. Concentrations of trace elements have been given only as a snapshot for ripe grapes at the end of the ripening period [2], where the focus has been on the potential toxicity of heavy metals [16–20]. Other than that, such measurements were performed in order to determine the influence of winemaking practices on metal-ion concentrations [21, 22], the elemental nutrient concentrations on the fermentation kinetics [6, 23], and the geographical origin of wines [24–31]. The motivation for the present work was firstly to continuously accompany the growth and ripening process of the grapes by systematic element analyses which reflect the development of the element concentrations as a function of developmental stage. Secondly, it was of interest to investigate possible influences of mineral substances on the fermentation process, in particular that of trace elements. This disposition was caused by frequent sluggish and stuck fermentations that may have various reasons and result in a low wine quality. Therefore, in the literature, some attempts have been described aiming at the investigation of the development of element concentrations during fermentation [6]. As these were based on ideal laboratory conditions, the goal of the present work was, apart from the systematic assessment of element concentrations during growth and ripening of the grapes, to investigate the fermentation process under the real conditions in the wine-grower’s cellar in order to be able to judge whether laboratory experiments apply also for real wine-making. Furthermore, additional elements that were not considered in the above mentioned laboratory experiments could be detected by neutron activation analysis. With the INAA, it was possible to determine in the grapes of Riesling and Cabernet Sauvignon and the resulting musts the macro elements Mg, K, and Ca as well as the trace elements Na, Al, Mn, Cu, Zn, and Rb quantitatively and to observe the developments of their concentrations during growth and ripening of the grapes, during fermentation of the must, and during malo-lactic fermentation resulting in a biological acid reduction that was conducted with the Cabernet Sauvignon. In addition to the grapes and musts, the fermenting yeasts and their element concentrations were determined and compared with the element concentrations of the respective starter cultures. Consequences of Cu ions of various concentrations on the fermenting process, the application of a yeast food salt, and the treatment of the wine by bentonite were also investigated.

Bereitgestellt von | Johannes Gutenberg Universitaet Mainz Angemeldet Heruntergeladen am | 28.01.16 12:01

348 | Markus Feige et al., Chronological development of element concentrations in grapes

It was found that the element concentrations as a function of the developmental stage in the two different grapes paralleled largely. K was strongly enriched in the grapes of both Riesling and Cabernet Sauvignon. On the other hand side, the Ca concentration decreased during maturation of the grapes. While also the Mn concentrations decreased as a function of time, the Na concentration increased in the Riesling grapes by a factor of two and somewhat less but also significantly in the Cabernet Sauvignon grapes. Also the Al concentrations increased by one third. A significant increase was also observed for Rb in both wines, which subsequently decreased again to less than one half. The Cu concentration stayed constant as a function of maturation, and the Zn concentrations decreased to values between 0.3 𝜇g/mL and 0.4 𝜇g/mL. By an alternative presentation of the element contents in the grapes for some elements, it was possible to demonstrate that their concentrations decrease by dilution, while for others it was shown that their contents decreased most likely by an active transport of the cations out of the berries. To achieve this result, the water content of the grape was determined in the pressing of the grapes and the must volume normalized to 100 g mass of the grapes was calculated. The total element masses contained in 100 g grapes was plotted as a function of time and compared to the concentrations of the elements. The Riesling must was investigated during the process of fermentation. After pressing, the must was divided in two parts where the larger part was inoculated with a yeast starter culture and a smaller part was fermented spontaneously. For both wines, the sugar concentrations decreased, less pronounced for the spontaneously fermented Riesling than for the inoculated Riesling. While the Ca concentration stayed constant in the spontaneously fermented Riesling, it increased first for the inoculated Riesling and decreased subsequently to less than the starting concentration. Both Rieslings showed a decreasing concentration of Mg ions at the beginning of the fermenta-

tion which increased again during the course of the fermentation. Na and Al increased slightly at the beginning and stayed constant subsequently. The strongest concentration fluctuations were observed for K, Cu, and Zn. Cu decreased during the fermentation induced by the starter culture and it could no longer be detected after two weeks. Most interesting was the time dependence of the Zn concentrations that was both influenced by the addition of the starter culture and by the later addition of the yeast food salt. The comparison of the element concentrations in the fermenting yeast with those in the starter cultures showed that some elements were enriched in the cell pellet while others are transferred from the starter culture into the fermenting must and consequently into the young wine. The investigation of influence of the yeast food salt yielded a high content of Zn ions. It could be shown that Zn is an essential trace element for the development of the fermentation process. The treatment of the wine with bentonite resulted in the observation that Na, Al, and Ca left partly the bentonite and were dissolved in the wine. With different concentrations of Cu up to 10 𝜇g/mL, it was shown that these did not perturbe the fermentation. There was a significant decrease of the Cu concentrations during fermentation of the must that was even stronger for higher Cu concentrations. In summary, the present results demonstrate that laboratory experiments on the influence of different mineral substances on the fermentation process can be transferred to the conditions of real wine-making in the wine cellar.

Acknowledgement: Thanks are due to Martina Schlauder (TA) for assistance with microbiological work.

Received February 18, 2013; accepted August 22, 2013.

References 1. König, H., Unden, G., Fröhlich, J. (eds.): Biology of Microorganisms on Grapes, in Must and in Wine. Springer Verlag, Heidelberg (2009). 2. Würdig. G., Woller, R.: Chemie des Weines, Verlag Eugen Ulmer (1989). 3. Richter, G.: Stoffwechselphysiologie der Pflanzen, 6𝑡ℎ Edition, Thieme Verlag, ISBN 978-3-134-42006-7 (1997).

4. Ebermann, R., Elmadfa, I.: Lehrbuch Lebensmittelchemie und Ernährung, Springer-Verlag Wien, ISBN 978-3-211-48649-8 (2008). 5. Dittrich, H., Groẞmann, M.: Mikrobiologie des Weines, 3𝑟𝑑 Edition, Verlag Eugen Ulmer, ISBN 3-8001-4470-0 (2005). 6. Azpilicueta, C. A.: Influence of decantation of viura must on the cation content. Evolution during wine fermentation and stabilization. Food Res. Int. 32, 683 (1999).

Bereitgestellt von | Johannes Gutenberg Universitaet Mainz Angemeldet Heruntergeladen am | 28.01.16 12:01

Markus Feige et al., Chronological development of element concentrations in grapes

7. Eschnauer, H.: Spurenelemente und Ultra-Spurenelemente in Wein, Naturwissenschaften 73, 281 (1986). 8. Schut, S., Zauner, S., Hampel, G., König, H., Claus, H.: Biosorption of copper by wine-relevant lactobacilli, Int. J. Food Microbiol. 145, 126 (2011). 9. Analytical Chemistry: A Modern Approach to Analytical Science, 2𝑛𝑑 Edition, WILEY-VCH Weinheim, ISBN 3-527-30590 (2004). 10. Genie 2000, Version 3.2, Copyright 2009, Canberra Industries Inc. (2009). 11. Steidl, R., Leindl, G.: Biologischer Säureabbau, Verlag Eugen Ulmer, ISBN 3-8001-3913-8 (2002). 12. Vogt/Götz: Weinbau, 5. neub. Aufl., Verlag Eugen Ulmer, ISBN 3-8001-1205-1 (1977). 13. Hulme, A. C.: Biochemistry of fruits and their products, Vol. 2, 172, Academic Press, London and New York, SBN 12-361202-0 (1971). 14. Dittrich, H.: Chemie und Mikrobiologie. Der deutsche Wein, 100, Gräfe und Unzer GmbH, München, ISBN 3-7742-1010-1 (1978). 15. Brandolini, V., Tedeschi, P., Capece, A. et al.: Saccharomyces cerevisiae wine strains differing in copper resistance exhibit different capability to reduce copper content in wine, World, J. Microbiol. Biotechnol. 18, 499 (2002). 16. Lara, R., Cerutti, S., Salonia, J. A. et al.:Trace element determination of Argentine wines using ETAAS and USN-ICP-OES, Food Chem. Toxicol.43, 293 (2005). 17. Wagner, E., Augustin, T., Brunn, H.: Metallgehalt in Wein – Bestimmung in deutschen und ausländischen Weinen, DLR7, 553 (2010). 18. Catanio, S., Curvelo-Garcia, A. S., de Sousa, R. B.: Contaminant elements in wines: A review, Ciencia e tecnica vitivinicola 23, 3 (2008). 19. Naughton, D. P., Petróczi, A.: Heavy metal ions in wines: metaanalysis of target hazard quotients reveal health risks, Chemistry Central Journal 2, 22 (2008). 20. Tariba, B.: Metals in wine – Impact on wine quality and health outcomes, Biol. Trace Elem. Res. 144, 143 (2011). 21. Rossano, E. C., Szilágyi, Z., Malorni, A., Pocsfalvi, G.: Influence of winemaking practices on the concentration of rare earth elements in white wines studied by inductively coupled plasma mass spectrometry, J. Agric. Food Chem. 55, 311 (2007).

| 349

22. Castineira Gómez, M. D. M., Brandt, R., Jakubowski, N., Andersson, J. I.: Changes of the metal composition in german white wines through the winemaking process. A study of 63 elements by inductively coupled plasma-mass spectrometry, J. Agric. Food Chem. 52, 2953 (2004). 23. Schmidt, S. A., Dillon, S., Kolouchova, R., Henschke, P. A., Chambers, P. J.: Impacts of variations in elemental nutrient concentrations of Chardonnay musts on Saccharomyces cerevisiae fermentation kinetics and wine composition, Appl. Microbiol. Biotechnol. 91, 365 (2011). 24. Coetzee, P. P., Steffens, F. E., Eiselen, R. J. et al.: Multi-element analysis of South African Wines by ICP-MS and their classification according to geographical origin, J. Agric. Food Chem. 53, 5060 (2005). 25. Greenough, J. D., Mallory-Greenough, L. M., Fryer, B. J.: Biogeochemical controls on the trace element composition of wine, Goldschmidt Conference Abstracts A215 (2006). 26. Galgano, F., Favati, F., Camso, M., Scarpa, T., Palma, A.: Analysis of trace elements in southern Italian wines and their classification according to provenance, LWT Food Science and Technology 41, 1808 (2008). 27. Serugal, M., Nemet, I., Laslavic, B.: Heavy metals content of some Croatian wines, Deutsche Lebensmittel-Rundschau 104, 46 (2008). 28. Razic, S., Onjia, A.: Trace element analysis and pattern recognition techniques in classification of wine from central Balkan countries, American, J. of Enology and Viticulture 61, 506 (2010). 29. Van der Linde, G., Fischer, J. L., Coetzee, P. P.: Multi-element analysis of South African wines and their provenance soils by ICP-MS and their classification according to geographical origin using multivariate statistics, South African, J. of Enology and Viticulture 31, 143 (2010). 30. Laurie, V. F., Villagra, E., Tapia, J., Sarkis, J. E. S., Hortellani, M. A.: Analysis of major metallic elements in Chilean wines by atomic absorption spectroscopy, Ciencia e Investigacion Agraria 37, 77 (2010). 31. Bentlin, F. R. S., Pulgati, F. H., Dressler, V. L., Pozebon, D.: Elemental analysis of wines from South America and their classification according to country, J. Braz. Chem. Soc. 22, 327 (2011).

Bereitgestellt von | Johannes Gutenberg Universitaet Mainz Angemeldet Heruntergeladen am | 28.01.16 12:01