Physicochemical Signatures of Natural Sea Films from Middle Adriatic

Article pubs.acs.org/JPCA

Physicochemical Signatures of Natural Sea Films from Middle Adriatic Stations Sanja Frka,*,† Stanislaw Pogorzelski,‡ Zlatica Kozarac,† and Božena Ć osović† †

Division for Marine and Environmental Research, Ruđer Bošković Institute, POB 180, HR-10002 Zagreb, Croatia Institute of Experimental Physics, University of Gdańsk, Wita Stwosza 57, 80-952 Gdańsk, Poland

‡

S Supporting Information *

ABSTRACT: Monolayer studies and a force−area quantification approach, in combination with electrochemical methods, are applied for physicochemical characterization of surface active substances (SAS) of the sea surface microlayers (MLs) from Middle Adriatic stations. Higher primary production during late spring−early autumn was reflected in the presence of MLs of higher surfactant activity containing on average molecules of lower molecular masses (Mw = 0.65 ± 0.27 kDa) and higher miscibility (y = 6.46 ± 1.33) and elasticity (Eisoth = 18.33 ± 2.02 mN m−1) modulus in comparison to structural parameters (average Mw = 2.15 ± 1.58 kDa; y = 3.51 ± 1.46; Eisoth = 6.41 ± 1.97 mN m−1) obtained for MLs from a period of lower production. A higher inhibition effect on the reduction process of Cd2+ was observed for SAS abundant MLs from a more productive period. This kind of distribution is explained as the consequence of competitive adsorption of hydrophobic lipid-like substances of lower Mw that act as end-members, highly influencing the surface structural properties of the natural air−water interface forming there segregated surface films during more productive period.

1. INTRODUCTION The sea surface microlayer (ML) is the widest environmental interface, where many important processes such as wind action, water transpiration, solar energy flux, and atmospheric inputs take place.1 A large fraction of the ML organic matter has a particularly strong interfacial affinity. Due to their molecular structures, such surface active substances (SAS) are adsorbed at the air−water interface; i.e. their concentration increases spontaneously and they form natural surface films. Electrochemical methods are widely used for selective measurement of the concentration of SAS in natural waters by following nonfaradic and/or faradic processes at the electrode/seawater interface.2 These methods have proved to be an efficient tool for physicochemical characterization of the natural surface microlayer.3−5 The biota of the underlying water column are the primary source of naturally derived organic material since the phytoplankton exudates and their degradation products are assumed to be one of the largest sources of SAS.6 Excretion products of many types of marine phytoplankton include complex polymeric compounds of high molecular weights and are frequently found in surface waters during phytoplankton blooms.7,8 Thus, it is reasonable to assume that the prevalence of oceanic films will closely track biological productivity, since this is a large source term. Lipids are also present in phytoplankton exudates, but they are generally less abundant than carbohydrates and proteins. However, their contribution to the surface physical properties of microlayer films may be disproportionately large considering their low concentration levels.9 Various biological, chemical, and physical processes lead to the alteration of the film chemical composition, surface © 2012 American Chemical Society

physical properties, surface concentration, and spatial distribution of film-forming components. There is a dynamic equilibrium which involves the supply and the removal processes of chemical substances. The supply processes include atmospheric inputs, bubble flotation, and hydrodynamic renewal from the subsurface water, the last being in turn affected by the underlying water column. Convergent circulations driven by wind, tidal forces, current shear, upwelling, and internal waves lead to localized concentrations of SAS on various spatial scales. The removal processes include injection of SAS into the atmosphere in an enriched form as part of the marine aerosol produced by bursting bubbles. This provides a mechanism for the selective transfer of materials to terrestrial environments and an important mechanism for the charge separation and electrification of the atmosphere. There is increased evidence for the importance of surface films in the transfer of masses, heat, and momentum across the air−sea interface. The viscoelastic behavior of the air−sea interface, a key parameter affecting air−sea exchange of mass, momentum, and heat, is strongly dependent on naturally occurring adsorbed surfactant materials. Due to the complex mixture of components, the composition and the molecular arrangement of microlayer films can vary in response to physical forcing. Such films are capable of undergoing relaxation processes on a Special Issue: Herman P. van Leeuwen Festschrift Received: December 23, 2011 Revised: April 28, 2012 Published: April 29, 2012 6552

dx.doi.org/10.1021/jp212430a | J. Phys. Chem. A 2012, 116, 6552−6559

The Journal of Physical Chemistry A

Article

evaporated to dryness and stored under a nitrogen atmosphere until analysis. Prior to the analysis, dry extracts were redissolved to a concentration of 5 mg cm−3 and analyzed as DCM ex situ reconstructed microlayers. To obtain a final mass concentration, the dry weight of extracts was measured with an analytical electronic balance of high precision readings to five decimal places (Mettler Toledo, AX205). 2.2. Sample Analysis. Measurements of π−A isotherms were carried out at room temperature (21 ± 0.5 °C) in a round Teflon Langmuir trough (220 mL) of Fromherz type with movable barriers17 and a Wilhelmy plate system, all supplied by Mayer-Feintechnik (Göttingen, Germany). The π−A isotherms were recorded for original ML samples and for those formed by spreading ML extract solution on the water surface (DCM ex situ reconstructed films).3 The π−A isotherm measurements were determined at least two times with a new aliquot of the original sample and at least three times with a new spread of ex situ film to ensure consistent results. Stabilization time was 15 min prior to recording the π−A isotherms at a compression rate of 60 cm2 min−1. From the π−A isotherms different structural parameters such as the molecular weight (Mw) of the filmforming SAS mixture, miscibility (y), and elasticity (Eisoth) modulus have been calculated (see details in the Supporting Information) according to the isotherm scaling approach previously described elsewhere.13,14,18 An electrochemical method, phase-sensitive a.c. voltammetry, was applied in this work (see details in the Supporting Information). Concentration of surface active substances in original ML samples was obtained by a.c. voltammetry (out-ofphase signal) and has been expressed as an equivalent adsorption effect of a certain amount of model nonionic surfactant Triton-X-100.5,19,20 The relative standard deviation obtained for multiple analyses of the same solution containing SAS at a level of 0.1 mg dm−3 is less than 5%. Additional characterization of the adsorption behavior and the structure of original ML samples as well as those of ex situ films was carried out by a.c. voltammetry (in-phase signal) by investigating the influence of the adsorbed layer on the cathodic processes of cadmium ions used as electrochemical probes. Measurements toward negative potentials were applied after SAS accumulation by stirring for 120 s at a potential of −0.4 V (for original samples) or after film transfer (for ex situ microlayers). The inhibition effect (I) was determined as I = (i0 − i)/i0, where i0 is the peak current of cadmium(II) ions (c = 10−4 mol dm−3) in the absence and i is the peak current in the presence of the adsorbed layer at the electrode surface.

variety of time scales, which lead to a surface physical response that is not purely elastic.10,11 Instead, the film-influenced water surface will exhibit a complex stress−stain relation in response to the compression and dilatation.12 Surface films comprising many different molecular species, when subject to long-time compression, relax differently by desorption of more soluble species. The remaining film materials (the more hydrophobic ones) dominate the elastic response and generally exhibit enhanced damping capabilities.12 Improved knowledge of the sea surface microlayer distribution under different forcing conditions, its chemical composition, and rheological characteristics is of crucial importance for a better insight into the structural changes and the mechanisms of microlayer formation as well as for modeling the phenomena and processes at the air−sea interface. An understanding of the processes and physical mechanisms governing the exchange of gases between air and sea is essential when considering coupled models of atmosphere−ocean interactions and global climate, the interactions of which are of particular concern as they may affect or be affected by global changes. Instead of analyzing its chemical composition, it should be possible to scale the sea surface microlayer surface pressure−area (π−A) isotherms in terms of structural parameters, reflecting the natural film morphology and resulting from the generalized physical formalisms adopted to study multicomponent films of surface active substances.13,14 In particular, a set of structural film parameters appears to be a sensitive and quantitative measure of the film physicochemical composition, surface concentration, and miscibility of its film-forming components. The main expectation from this study is to offer a new perspective to the physicochemical characterization of the sea surface microlayer structure regarding the composition and the properties of dominant SAS material as well as to delineate the role of present surfactant films in modulating the seasonal change signals of its surface rheological parameters at the natural air− sea interface.

2. EXPERIMENTAL SECTION 2.1. Sampling and Sample Pretreatment. Sea surface microlayer samples were collected at two different locations in the central part of the Eastern Middle Adriatic coastal area. First is a small eutrophicated seawater Rogoznica Lake covering an area of 10276 m2 with a maximum depth of 15 m, and another is the Martinska station situated in the lower part of the highly stratified Krka River estuary in front of the Šibenik city. More details about the study area are given in the Supporting Information. The ML samples were collected in different seasons under different weather conditions in the period from March 2001 to August 2003 at the Martinska station and from March 2001 to August 2005 at Rogoznica Lake; see Table S1 in the Supporting Information for details. In the text, labels R and M indicate ML samples collected from Rogoznica Lake and the Martinska station, respectively. Sampling of the ML was performed from the calm sea surface using a 16-mesh stainless steel Garrett’s screen,15 and the average thickness of the recovered layer was calculated to be 260 ± 40 μm. The sea surface microlayer was studied as original samples without any pretreatment and as DCM ex situ reconstructed films after sample extraction by dichloromethane (p.a. grade, Kemika, Croatia) with a dielectric constant of 8.9.16 The aliquots (0.5 dm3) were extracted three times with 100 cm3 of dichloromethane. Afterward, combined extracts were

3. RESULTS AND DISCUSSION 3.1. Force−Area Natural Film Characteristics. The typical π−A isotherms of original and ex situ reconstructed marine films studied at the Middle Adriatic stations are presented in Figure 1a and 1b, respectively. Due to the complexity and the unknown composition of microlayer samples, isotherms are given using the area in square centimeters as the abscissa instead of the area per molecule as usual in monolayer studies. It is also possible to scale π−A isotherms using chemical attributes as proposed by Frew and Nelson where π−A isotherms of natural microlayer samples, even if not scaled on a specific area basis, are found useful to compare physical states of different microlayers if the experimental conditions are kept constant.21 The shapes of isotherms of the original ML samples generally indicated liquidexpanded or gaseous film types of natural surface films. Such a 6553



Article

SA material in the ML samples during the investigation period. The πmax values for the original MLs of both, Rogoznica Lake and the Martinska station, were in the range from 2.7 to 39.8 mN m−1 (average 15.3 ± 10.8 mN m−1). We observed a statistically significant linear relationship between πmax and SAS concentrations (r = 0.872, p < 0.0001, N = 17) obtained for all samples, indicating coupling of those two parameters at the 95% confidence level (Figure 2). Such a correlation implies that

Figure 1. Surface-pressure isotherms of the original (A) and of DCM ex-situ reconstructed microlayers (B) of the Rogoznica Lake (R) and of the Martinska station (M). Figure 2. Correlation of maximum surface pressure (πmax) obtained during ML compression and surface active substance (SAS) concentration of the original microlayers from Rogoznica Lake (R) and the Martinska station (M).

behavior has been recorded previously for oceanic films by several investigators.9,22−24 This would imply that there is a great deal of uniformity of the surface active material affecting the shape of π−A isotherms that occur naturally at the surface of salt lakes, seas, and oceans worldwide, suggesting that surface properties are substantially averaged in complex mixtures. It is unlikely that the formation of condensed films is an important phenomenon under typical conditions of turbulence found at the sea surface, since the close-packed molecular configuration would not be maintained in the presence of wind and breaking waves under natural sea conditions. It should be noted that the π−A isotherms of some of the original and ex situ reconstructed microlayers studied in coastal waters of the Adriatic Sea exhibited the plateau regions and kinks representing the firstorder G−LE and LE−LC phase transitions or phase transitions of higher orders, respectively. Enthalpy H and entropy S of the 2D first-order phase transitions can be determined using the Clausius−Clapeyron equation applied to the isotherms registered at different temperatures.25 Furthermore, it is known that Brewster angle microscopy (BAM) provides information concerning monolayer morphology (homogeneity, domain existence, formation, and aggregation) and the existence of surface phases26 as well as the monolayer thickness from the relative intensity of light reflected from the interface.27 Also, BAM can significantly contribute to the characterization and visualization of the topography of sea surface microlayers showing the formation and development of the microlayer during a 24 h period.28 Application of fractal analysis to BAM images provides additional complementary means for characterization of structures and dynamical processes at the MLs providing information on phase transitions, interactions, and aggregation mechanisms at the interface and structural changes under applied pressure.28 Natural microlayers of the Middle Adriatic stations investigated under the same experimental conditions reached significantly different maximum surface pressures (πmax) during compression due to the substantial variability of the complex

reaching the maximum surface pressure could be the result of different SAS quantities as well as the consequence of different dominant classes of SAS material present in the original MLs in different seasons. Namely, variation in surface properties may be attributed not only to the concentration changes but also to the changed chemical nature of the surface active organic matter. Indeed, we proved previously that some of the investigated samples from both sampling locations had higher ratios of surface activity and of dissolved organic carbon concentration (SAS/DOC ratios), indicating the presence of more hydrophobic material in those microlayers.5 Particularly, this was noticeable for the ML samples R6, R10, R11, M2, and M6, collected in a warmer, more biologically productive period of the year. Considerably higher SAS content in those samples was not accompanied with an increase of DOC concentration, which is a characteristic of highly hydrophobic substances such as lipids. It is well-known that hydrophobic material represents a minor fraction of the ML organic matter, which is composed primarily of polysaccharides, proteins, and humic-type material.29,30 Although lipids are not the most abundant group of compounds in the ML, the study of these compounds is particularly important since, due to their reactivity and physical surface characteristics, they are contributing to the processes of the formation and stabilization of the microlayer.31 Thus, to gain insight into that minor hydrophobic fraction, lipid material was isolated, selected, and concentrated from the original MLs by extraction with dichloromethane and the DCM ex situ reconstructed films were investigated at the air−water phase boundary. The isotherms of dichloromethane films commonly characterized by the monotonous increase of the surface pressure during the compression were of a liquidexpanded type (Figure 1b) reaching maximum surface pressures from 10.0 to 56.0 mN m−1 (average 31.3 ± 14.0 mN m−1), which were typically higher than πmax of the 6554



Article

surface activities.5 Namely, microlayers within the cluster A are characterized by a higher SAS content, mostly higher than 1.3 mg dm−3 equiv Triton-X-100 in comparison to those from the cluster B. As previously concluded, the microlayers within the cluster A were collected during a warmer, more productive period while microlayers within the cluster B were collected during a less productive period, as it is known that during winter−early spring months the phytoplankton abundance is typically low since the phytoplankton populations in that period adjust to low light intensity.33 The production of marine surfactants derived from phytoplankton exudates and their degradation products appear to be seasonal and have been linked to biological productivity cycles.34−36 Wurl et al. found that the mean concentration of surfactants in the ML increases with the increase in primary production, being on average for oligotrophic, mesotrophic, and eutrophic waters 320 ± 66, 502 ± 107, and 663 ± 77 μg dm−3, respectively, expressed in equivalents of model surfactant Triton-X-100.37 In this study, higher organic matter (OM) production from late spring until early autumn is reflected in the increased concentrations of surface active material (>1.3 mg/L, expressed in equivalents of Triton-X-100) which forms films of higher elasticity (Eisoth↑; average 19.09 mN m−1), with the most complex film architecture (y↑; average 6.69) and the highest surface concentrations of low molecular mass film-forming material (Mw↓; 0.65 kDa) when compared to those films from a less biologically productive period. The same seasonal variability of structural parameters was observed for microlayers of the Baltic Sea regardless of the fact that the ML samples were collected by different sampling devices.38 The Mw values reported here for the microlayers from the more productive season agree well with the Mw values reported for the Atlantic Ocean microlayer samples collected during summer months using the same sampling technique.22 The observed Mw values characteristic for MLs from the more productive period (3.5, pointing to a decreased film homogeneity due to segregation of the surface film within the MLs in the form of patches or domains of film-forming components.18 Oleic alcohol as the film-forming material widely used in oceanographic slick studies reached a high y value (y = 10) spread on seawater indicating the formation of nearly separate layers at the interface (sandwich-like structure). High values of y = 8 were also found for the samples of natural film collected in the wakes of naval vessels.18 However, it is clear that the Mw values obtained for the microlayer samples during the less productive period (average Mw = 2.39 kDa) were considerably higher than those of model lipid substances, implying that the SA material in those microlayers must be an aggregated mixture of the material

corresponding original microlayers (Figure S1 in the Supporting Information). The shape of the isotherms indicates the presence of phospholipid-type lipid material in DCM films. This supports the findings of a previous electrochemical investigation of the adsorption behavior for different ex situ reconstructed films and model lipid compounds.5 Thus, the expanded-type isotherms of dichloromethane films were expected since phospholipid compounds are known to form more expanded films due to the presence of polar head groups and disable the formation of condensed monolayers by electrostatic and/or steric factors.32 We should also take into account that the reconstructed material might be a complex lipid mixture forming thus less condensed, i.e. more expanded, films than monolayers of individual compounds. 3.2. Structural Signatures of Natural Sea Surface Microlayers. In this study structural parameters including molecular weight (Mw), miscibility (y), and elasticity modulus (Eisoth) have been obtained from π−A isotherms of the original MLs from the selected Middle Adriatic stations (see Figure S2 in the Supporting Information). The parameters for the original MLs of Rogoznica Lake possessed wide ranges of values: Mw = 0.29−4.94 kDa (average 1.23 ± 1.29 kDa), y = 1.82−8.35 (average 5.20 ± 2.31), Eisoth = 3.64−19.52 mN m−1 (average 11.78 ± 6.77 mN m−1). This was also the same case for the Martinska station: Mw = 0.80−4.97 kDa (average 2.00 ± 1.65 kDa), y = 1.87−6.42 (average 3.92 ± 1.83), Eisoth = 5.48−20.76 mN m−1 (average 11.08 ± 7.03 mN m−1). Cluster analysis enabled grouping of MLs according to the variability of structural parameters into two distinct clusters (Figure 3). Cluster A includes microlayers from both

Figure 3. Bray−Curtis Similarity61 of the structural parameters (Mw, Eisoth, and y) obtained for the original sea surface microlayer samples from the Rogoznica Lake (R) and from the Martinska station (M). The clustering data showed that the samples were clustered into two main groups; Bray−Curtis similarity 55.83%, group average.

investigated locations sampled during late spring−summer− early autumn months (samples R2, R6, R9, R10, R11, M2, and M6) while cluster B involves MLs from winter−early spring months as well as the samples from June 2003 (samples R7 and M4) collected from a calm sea just after a strong wind event. Thus, cluster A is characterized by the presence of microlayers containing molecules of lower molecular masses (average Mw = 0.65 ± 0.27 kDa) and of higher miscibility (average y = 6.46 ± 1.33) and elasticity (average 18.33 ± 2.02 mN m−1) modulus compared to Mw (average Mw = 2.15 ± 1.58 kDa), y (average y = 3.51 ± 1.46), and Eisoth values (Eisoth = 6.41 ± 1.97 mN m−1) obtained for MLs from cluster B. A similar classification was previously noticed for the same microlayers according to their 6555



Article

3.3. Interaction of Surface Active Constituents with Electrochemical Probe. Additional information on the adsorption behavior of the mixture and the structure of the adsorbed SAS can be obtained by investigating the influence of the adsorbed layer on the electrode processes of other ions and molecules such as cadmium(II) ions which are used as a probe. This influence was followed through the change in inhibition effect (I), which shows a decrease in the permeability of the adsorbed SAS layer for the reduction of cadmium(II) ions at the mercury electrode. Inhibition effects on the cadmium reduction process have been studied with a number of adsorbable organic substances, representing either naturally occurring substances in the aquatic systems or different model substances.46 It was found previously that the effect of SAS upon the electrode reaction of cadmium depends on the concentrations and the types of SAS present in different environmental samples. It is important to mention that, when compared to copper, which is predominantly forms complexes with organic matter in the bulk solution,47,48 the interaction of cadmium occurs in the adsorbed layer at the surface. Table 1

capable of forming a more complex interfacial molecular structure. This conclusion is in line with the findings of Pogorzelski for microlayers of the Baltic Sea.14 The highest values of Mw and Eisoth obtained for the MLs taken just after a strong wind event when intensive water mixing took place (samples R7 and M3) pointed to the fact that the microlayer exposed to a more dynamic disruption by wind resulted in structural film behavior more similar to that of subsurface water. Higher Mw values were also previously recorded for the Atlantic Ocean and the Baltic Sea water samples taken with a bottle if referred to the corresponding microlayer samples.14,22 There certainly exists continuous interchange of the material between the microlayer and bulk, under the action of turbulent mixing processes induced by wind, waves, currents, bubbles, and tides. Thus, in the case of surface material removal, the ML is reformed by the material coming to the surface from the bulk.28,40 The time scales of the renewal process are dependent on the degree of mixing in the water, which may take a few hours for stagnant conditions but only a few minutes or seconds when the water is bubbled.41,42 The obtained values for the elasticity modulus (Eisoth) for the natural microlayers from the Middle Adriatic (3.64−20.76 mN m−1) were comparable with the values reported for the microlayers from different natural marine systems sampled with the same type of sampler. Thus, the Eisoth values for the microlayer from the Northern Adriatic were in the range of 13 to 22 mN m−1.4,43 The mean Eisoth values of 16.7 ± 2.2 and 18.2 ± 1.0 mN m−1 were detected for the natural microlayers of the Chesapeake Bay and open Atlantic Ocean, respectively.22 The microlayers sampled in the presence of visible slicks in the area of the La Jolla Bay exhibited Eisoth values up to 30 mN m−1 while, without slicks, the MLs reached elasticity values of 2−4 mN m−1.23 The Eisoth values obtained in this study for the MLs from the Middle Adriatic as well as those previously reported for other marine systems were considerably lower than the Eisoth values reported for single compounds. For comparison, the elasticity modulus of the monolayers of some pure compounds are as follows: cholesterol and stearic acid 833.3 and 526.3 mN m−1, respectively; some proteins between 33.33 and 66.67 mN m−1; sedimentary humic acids of different molecular weights range between 27.78 and 41.67 mN m−1;44 chlorophyll a 47.62 mN m−1, vitamin A 52.63 mN m−1, and sodium dodecylbenzene sulfonate (NaDBS) 58.82 mN m−1.22 Thus natural films as a complex mixture of different organic substances, with different surface properties, are less condensed, i.e. more expanded, than the monolayers of single compounds. However, the inherent elasticity of an air−water interface influenced by different film types varies significantly. Since the natural films from the more productive period were more elastic than those from the rest of the year, this could indicate that lipid material was the dominant material of SAS of natural microlayers during productive months at particular Middle Adriatic stations as previously noticed by Frka et al.5 Furthermore, the potential role of nitrogen-containing surfactants in controlling film elasticity Eisoth is also of particular interest.12 Thus, the enhanced contributions of relatively soluble biopolymeric materials such as proteins in the MLs would be reflected in lower C/N values. It should be noted that the relation Eisoth versus C/N was previously obtained with a high correlation coefficient value. 45 Generally, higher C/N ratios were accompanied by a higher maximum elasticity modulus, indicating that incorporation of nitrogen-rich materials or other biopolymeric materials tends to lower film elasticity.

Table 1. Concentrations of Surface Active Substances (SAS) and Inhibition Effects on the Cadmium(II) Ion Reduction Process at the Electrode Surface for Different Environmental Samplesa Sample; reference ML samples, Middle Adriatic; this work ML samples, Tromsø region, Norway; ref 52 Seawater, Rovinj Harbor, Croatia; ref 46 Seawater, Rovinj Fish Cannery, Croatia; ref 46 Seawater, Split Plastics Factory, Croatia; ref 46 Sava River water, Croatia; ref 46 Rainwater, Zagreb, Croatia; ref 51

SAS eq. Triton-X-100 (mg dm−3)

Inhibition effect

0.21−1.36

0.08−0.68

0.12−0.22

0.37−0.55

0.84

0.31

2.00

0.60

0.60

0.78

0.23−1.65 0.03−0.15

no effect −1.98−0.40

a Inhibition effect (I) is determined as a relative decrease in cadmium reduction wave I = (i0 − i)/i0.

presents a literature survey on the influence of SAS from different environmental samples on the electrode reaction of cadmium ions at the electrode surface. A high content of SAS in natural waters does not necessarily result in a high inhibition effect. The content of SAS in seawater samples taken near pollution sources (Rovinj, Split) is considerably higher than the usual values for unpolluted seawater samples in the Adriatic sea, which results in an inhibitory effect, especially in the samples in which synthetic compounds predominated.50 Synthetic compounds, such as commercial detergents, significantly slow down the kinetics of the electrode processes even at very low concentrations of surfactants. Furthermore, the high content of SAS in the samples of the Sava river water taken at different stations and in different seasons showed no inhibition effect on cadmium reduction at natural pH (pH ∼8). Such behavior can be expected because adsorption studies with model substances clearly show that the majority of naturally occurring surface active substances, such as biopolymers (proteins) and geopolymers (humic substances), have a negligible effect on the mass and charge transfer processes at the electrode surface; e.g. they create adsorbed films which are porous for cadmium ions.46 Most of the investigated rainwater samples showed an 6556



Article

increase in the voltammetric reduction peak of Cd2+ ions suggesting an accumulation of metal ions in the adsorbed layers of organic molecules. These results are in accordance with earlier investigations on model substances as representatives for precipitation samples, like humic acids and unsaturated fatty acids.49−51 In this study, the SAS of natural microlayers have shown a consistent inhibition effect on cadmium ion reduction at the electrode surface regardless of the sampling location (see Figure S3 in the Supporting Information). For the original microlayers, inhibition was in the range from 0.08 to 0.68 (average 0.42 ± 0.23). The highest inhibition effect was noticed for those original microlayers characterized by lower Mw and higher y and Eisoth values, e.g. MLs previously grouped within cluster A. A significant inhibition effect was also noticed previously for original MLs from Tromsø, Norway (see Table 1), characterized by a considerably lower SAS content than those detected in the Middle Adriatic samples.52 An analysis using Pearson’s correlation coefficient shows a statistically significant linear relationship (r = −0.538, N = 15) between the inhibition effect and SAS concentrations, indicating the coupling of those two parameters at the 95% confidence level for the Middle Adriatic samples (graph not presented). This implies that the inhibitory effect could be not only the result of different SAS quantities but also the consequence of specific dominant classes of SAS material present in the original ML during different seasons. When the inhibition effects for the original and for the corresponding ex situ films are compared, a stronger inhibition with DCM films can be observed than with the original ML samples. Generally, the highest inhibition was observed for dichloromethane films, which occasionally (samples R10 and M2) induced a total inhibitory effect on cadmium(II) ion reduction at the HMDE surface. Thus, ex situ films which were extracted from the original microlayers previously grouped within cluster A (R6, R9, M2, and M6) were characterized by lower values of Mw (average 0.70 ± 0.34 kDa) and by higher values of y (average 8.11 ± 4.66) and Eisoth (average 26.59 ± 5.06 mN m−1), showing consequently a strong inhibition effect on Cd2+ reduction. For comparison, the films extracted from microlayers within cluster B (R5, R7, and R8) were characterized by higher Mw (average 1.17 ± 0.25 kDa) and by lower y (average 4.57 ± 2.55) and Eisoth (average 9.16 ± 5.03 mN m−1) showing a lower inhibition effect with the exception of R5 (ML from February 2002). Data presenting structural parameters Mw, y, and Eisoth obtained for particular DCM ex situ films are shown in Table S2 in the Supporting Information. Since the highest inhibition was observed for all corresponding dichloromethane films we could attribute such behavior to the predominance of phospholipid-like lipid compounds as film forming material in those natural films. This is not surprising since the electrode surface completely covered by phospholipid egg lecithin has shown an almost complete blocking effect on cadmium reduction.53 The electrochemistry of Cd as well as that of other metal probes such as Cu, Eu, Pb, Zn, and V have been investigated at a model lipid dioleoyl lecithin (DOPC) coated mercury electrode, which significantly inhibited all electrode processes in the capacitance minimum potential region.54 For those investigated metals, the rate limiting step for reduction appears to be associated with the transport of the metal ion across the lipid monolayer. Furthermore, adsorbed layers of different mixtures of protein and lecithin have been permeable for the transport of cadmium ions while the degree

of permeability depended on the ratio between the lipid and the protein components in the mixed layers.53 The investigated mixtures were composed of a constant concentration of one component, either lecithin or albumin, and an increasing concentration of the other component of the mixture. The inhibition effect caused by the mixture, in which the concentration of lecithin is constant and the concentration of albumin increases, decreases very slowly with further addition of albumin. The inhibition effect of the mixture in which albumin is the constant component and lecithin is the increasing one continuously changes with the addition of lecithin until the effect of the mixture coincides with the effect of lecithin at a given concentration. If lecithin is in excess, it displaces the albumin molecules from the electrode, while when albumin is present in excess it combines with the adsorption layer of lecithin. Evidently, the inhibition effect characteristic of microlayer SA compounds on cadmium reduction is the result of the presence of the overall inhibitory material present in each sample, but the degree of permeability depends on the ratio between the components of the mixture which possess different degrees of inhibition for the transport of cadmium ions. It is reasonable to believe that, in the complex natural surface films, competitive adsorption processes occur resulting in the different dominant composition of SA material. Comparison of the inhibitory effects of the original ML and the corresponding dichloromethane ex situ films led to the conclusion that polar lipid material represents an important fraction of the surface active material in the investigated microlayers collected during the warmer season at the Middle Adriatic stations.5 3.4. Dynamic Concept of the Sea Surface Microlayer. The sea surface microlayer is a site of a number of unique and dynamic, nonequilibrium processes such as wind stress, water transpiration, solar energy flux, and atmospheric inputs. Vertical microprofiles of the distribution of the chemical species near the interface are presently largely unknown. However, physicochemical properties of the ML could be qualitatively and quantitatively characterized by general parameters concerning a larger group of compounds such as surface active compounds. Considerable effort has been focused on the measurements of surface pressure−area isotherms of natural MLs placed in a Langmuir trough. It should be pointed out that by such an approach a natural microlayer composed of a wide variety of chemical species with different degrees of surface activity becomes a subject of spontaneous segregation of its structural components. The thickness of the microlayer depends on the method of sampling, and evidently, thinner layers show a chemical composition closer to the actual composition of the air−sea boundary layer, resulting in greater microlayer enrichment regarding the subsurface water. Investigation of structural properties of natural films by the monolayer technique as well as the applied scaling approach enables the study of surface films composed of accumulated SA material in the monolayer, explicitly localized at the air−water interface. Such a thin layer is more representative of the actual range of ML thickness regardless of the dilutions made by different sampling methods used for its collection. Historically, a model of the ML implicates subdivisions of the surface film into strata. The organic substances have thereby been considered either as dry or as wet surfactants.31 Dry surfactants, such as lipids, are considered as hydrophobic amphiphilic molecules which form the top layer above a protein− polysaccharide layer of the “so called” wet surfactants.55−57 6557



Article

ones from less productive season. This complex approach, using both electrochemical and monolayer techniques, accompanied with the novel scaling approach led to the conclusions about the role of lipid material as structural components in the surface active fraction of the original MLs, thus providing additional insight into the characterization of organic matter in the natural sea surface microlayers of the Middle Adriatic stations.

Yet, there also exists the accepted concept where the ML is a nonstratified, complex macromolecular/polymeric matrix composed primarily of oxygenated molecules of higher molecular weight when compared to the hydrophobic material such as hydrocarbons and free fatty acids.9,58 Thus, Hunter and Liss suggested that the organic films on the surface consist mostly of complex macromolecular material with the molecules supported at the interface by inter- and intramolecular interactions.29 In a recent report Wurl and Holmes (2008) provided data on transparent exopolymer particle (TEP) enrichment in the ML and confirmed the hypothesis of the sea surface microlayer as a hydrated gelatinous matrix formed by a complex structure of carbohydrates, proteins, and lipids.59 The degree of surfactant molecule aggregation within the natural film and the characteristic time scale of the transition processes can be derived from the measurements of surfactant adsorption kinetics.14,60 In the present study we obtained that significant variability in chemical composition of the microlayer surface active fraction caused a distinctive discrepancy in their structural parameters. Particular importance is assigned to a certain end-member material of hydrophobic SAS or the lipid-like material which strongly influenced the surface properties of the air−sea interface. That hydrophobic lipid material is present in OM to a very low degree, but due to its extremely high surface affinity, its presence at the air−water interface is the result of the competitive adsorption and segregation from other macromolecular constituents. The amount and the composition of organic surface active substances are closely related to the periods of primary production. Thus, during a period of higher OM production we could expect as well a higher production of lipid material. This material has the capability to accumulate at the air−sea interface, occupying the available space rather than large, more water-soluble surfactants such as polysaccharides, proteins, and humus-like material, forming thus structured domains above the complex macromolecular matrix. On the other hand, during a less productive period as well as in the case of mixing and dilution of SAS, the situation becomes favorable for the accumulation of less surface active material with larger Mw and higher Eisoth values due to the depletion of lipid material in the microlayer.

■

ASSOCIATED CONTENT

S Supporting Information *

Material includes detailed information about the study area, principles of isotherm scaling concept and background of electrochemical determination of surface active substances in natural waters. Table 1 gives details about the general conditions of sampling the sea surface microlayers. Table 2 presents static structural parameters (Mw, Eisoth, and y) obtained for DCM ex situ films. Figure 1 presents the maximum surface pressures obtained during compression of the original MLs and of corresponding DCM ex situ films. Figure 2 shows the seasonal change of the structural parameters (Mw, Eisoth, and y) obtained for the original MLs. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was funded by the grant from the Croatian Ministry of Science, Education and Sports, project: Nature of organic matter, interaction with traces and surfaces in environment (No. 098-0982934-2717).

■

REFERENCES

(1) Liss, P. S.; Duce, R. A. The Sea Surface and Global Change; Cambridge University Press, 1997. (2) Van den Berg, C. M. G. Electroanalytical Chemistry of Sea-Water. In Chemical Oceanography; Riley, Y. P., Skirrow, G., Eds.; Academic Press: New York, 1988; Vol. 9; p 198. (3) Kozarac, Z.; Ć osović, B.; Frka, S.; Möbius, D.; Hacke, S. Colloid Surf., A 2003, 219, 173. (4) Gašparović, B.; Kozarac, Z.; Saliot, A.; Ć osović, B.; Möbius, D. J. Colloid Interface Sci. 1998, 208, 191. (5) Frka, S.; Kozarac, Z.; Ć osović, B. Estuarine, Coastal Shelf Sci. 2009, 85, 555. (6) Ž utić, V.; Ć osović, B.; Marčenko, E.; Bihari, N.; Kršinić, F. Mar. Chem. 1981, 10, 505. (7) Sakugawa, H.; Handa, N. Geochim. Cosmochim. Acta 1985, 49, 1185. (8) Sakugawa, H.; Handa, N. Oceanol. Acta 1985, 8, 185. (9) Van Vleet, E. S.; Williams, P. M. Limnol. Oceanogr. 1983, 28, 401. (10) Van den Tempel, M.; Lucassen-Reynders, E. H. Adv. Colloid Interf. Sci. 1983, 18, 281. (11) Lucassen-Reynders, E. H. Dynamic properties of film-covered surface; ONRL-workshop proceedings-role of surfactant films on the interfacial properties of the sea surface, 1986. (12) Bock, E. J.; Frew, N. M. J. Geophys. Res. 1993, 98, 14599. (13) Pogorzelski, S. J. Colloid Surf., A 2001, 189, 163. (14) Pogorzelski, S. J.; Kogut, A. D. Oceanologia 2001, 43, 223. (15) Garrett, W. D. Limnol. Oceanogr. 1965, 10, 602. (16) Smith, M. B. Organic Synthesis; McGraw-Hill, Inc.: New York, 1994; p 125.

4. CONCLUSION This study offers a new aspect to the sea surface microlayer physicochemical structure regarding the composition and the properties of dominant SAS material with the focus on the hydrophobic low-molecular lipids that act as end-members, highly influencing the surface structural properties of natural films at the air−water interface. Lipid material is present in OM in a very low degree but due to the extremely high surface affinity their presence at the air−water interface is the result of their competitive adsorption and segregation from other macromolecular constituents, amount and the composition of which are closely related to periods of primary production. The lipid films strongly influence the seasonal change signals of SAS concentration and surface rheological parameters of natural films at the air−water interface. Hence, higher OM production from late spring until early autumn is reflected in the increased concentration of SA material which forms films of higher elasticity with the most complex film architecture and the highest surface concentrations of the lowest molecular mass film-forming material. Such organic films showed a higher inhibitory effect on cadmium reduction in comparison to the 6558



Article

(17) Fromherz, P. Rev. Sci. Instrum. 1975, 46, 1380. (18) Pogorzelski, S. J. Colloid Surf., A 1996, 114, 297. (19) Ć osović, B.; Vojvodić, V. Limnol. Oceanogr. 1982, 27, 361. (20) Ć osović, B.; Vojvodić, V. Mar. Chem. 1987, 22, 363. (21) Frew, N. M.; Nelson, R. K. J. Geophys. Res. 1992, 97, 5281. (22) Barger, W. R.; Means, J. C. Clues to the structure of marine organic material from the study of physical properties of surface films. In Marine and Estuarine Geochemistry; Sigleo, A. C., Hattori, A., Eds.; Lewis: Chelsea, 1985; p 47. (23) Frew, N. M.; Nelson, R. K. J. Geophys. Res. 1992, 97, 5291. (24) Jarvis, N. L. Limnol. Oceanogr. 1967, 12, 213. (25) Adamson, A. W. Physical Chemistry of Surface; John Wiley & Sons, Inc.: New York, USA, 1982. (26) Möbius, D. Curr. Opin. Colloid Interface Sci. 1998, 2, 137. (27) Patino, J. M. R.; Niño, P. R.; Sánchez, C. C. J. Agr. Food Chem. 2003, 51, 112. (28) Kozarac, Z.; Risović, D.; Frka, S.; Möbius, D. Mar Chem. 2005, 96, 99. (29) Hunter, B.; Liss, P. S. Organic sea surface films. In Marine Organic Chemistry; Duursma, E. K., Dawson, R., Eds.; Elsevier Scientific Publishing Company: Amsterdam-Oxford-New York, 1981; Vol. 31; p 259. (30) Gašparović, B.; Plavšić, M.; Ć osović, B.; Saliot, A. Mar. Chem. 2007, 105, 1. (31) Gladyshev, M. I. Biophysics of the Surface Microlayer of Aquatic Ecosystems; IWA Publishing: London, 2002. (32) Gaines, G. L. J. Insoluble Monolayers at Liquid-Gas Interfaces; Interscience: New York, USA, 1966. (33) Mangoni, O.; Modigh, M.; Mozetic, P.; Bergamasco, A.; Rivaro, P.; Saggiomo, V. Estuarine, Coastal Shelf Sci. 2008, 77, 633. (34) Ć osović, B.; Vojvodić, V.; Pleše, T. Water Res. 1985, 19, 175. (35) Gašparović, B.; Ć osović, B. Mar. Chem. 2001, 75, 301. (36) Gašparović, B.; Ć osović, B. Estuarine, Coastal Shelf Sci. 2003, 58, 555. (37) Wurl, O.; Wurl, E.; Miller, L.; Johnson, K.; Vagle, S. Biogeosciences 2011, 8, 121. (38) Mazurek, Z. A.; Pogorzelski, J. S.; Boniewicz-Szmyt, K. J. Mar. Syst. 2008, 74, S52. (39) Liss, P. S. Chemistry of the sea surface microlayer. In Chemical Oceanography; Riley, J. P., Skirrow, G., Eds.; Academic Press: New York, 1975; Vol. 2, p 193. (40) Dubreuil, F.; Daillant, J.; Guenoun, P. Langmuir 2003, 19, 8409. (41) GESAMP (IMO/FAO/Unesco-ICO/WMO/WHO/IAEA/ UN/UNEP Joint Group of Experts on the Scientific Aspects of Marine Environmental Protection) The Sea-Surface Microlayer and its Role in Global Change; 1995. (42) Dragčević, D.; Pravdić, V. Limnol. Oceanogr. 1981, 26, 492. (43) Kozarac, Z.; Möbius, D.; Romero, M. T. M. Water Res. 2000, 34, 1463. (44) Hayase, K.; Tsubota, H. J. Colloid Interface Sci. 1986, 114, 220 and references therein. (45) Pogorzelski, S. J.; Kogut, A. D.; Mazurek, A. Z. Hydrobiologia 2006, 554, 67. (46) Kozarac, Z.; Ć osović, B.; Vojvodić, V. Water Res. 1986, 20, 295. (47) Buffle, J. Complexation Reactions in Aquatic Systems. An Analytical Approach; Ellis Horwood Pub.: Chichester, 1988. (48) Plavšić, M. Anal. Lett. 2003, 36, 143. (49) Krznarić, D.; Ć osović, B. J. Colloid Interface Sci. 1983, 96, 425. (50) Plavšić, M.; Ć osović, B. Mar. Chem. 1991, 36, 39. (51) Orlović-Leko, P.; Kozarac, Z.; Ć osović, B.; Strmečki, S.; Plavšić, M. J. Atmos. Chem. 2010, 66, 11. (52) Frka, S. Physico-chemical characteristics of the sea surface microlayer, PhD Thesis, University of Zagreb, 2008. (53) Kozarac, Z.; Ć osović, B. Bioelectrochem. Bioenerg. 1984, 12, 353. (54) Nelson, A.; Van Leeuwen, H. P. J. Electroanal. Chem. 1989, 273, 183. (55) Norkrans, B. Surface microlayers in aquatic environments. In Advances in Microbial Ecology; Alexander, M., Ed.; Plenum: New York, 1980; Vol. 4.

(56) Hardy, J. T. Prog. Oceanogr. 1982, 11, 307. (57) Hermansson, M. The dynamics of dissolved and particulate organic material in surface microlayers. In The Biology of Particles in Aquatic Systems; Wooton, R. S., Ed.; CRC Press: Boca Raton, FL, 1990. (58) D’Arrigo, J. S. J. Colloid Interface Sci. 1984, 100, 106. (59) Wurl, O.; Holmes, M. Mar. Chem. 2008, 110, 89. (60) Birdi, K. S. Handbook of surface and colloid chemistry; CRC Press: New York, 1997. (61) Bray, J. R.; Curtis, J. T. Ecol. Monogr. 1957, 27, 326.

6559


Physicochemical Signatures of Natural Sea Films from Middle Adriatic

Recommend Documents