Chemosphere 119 (2015) 498503

Contents lists available at ScienceDirect

Chemosphere
journal homepage: www.elsevier.com/locate/chemosphere

Effects of molecular weight on the diffusion coefcient of aquatic

dissolved organic matter and humic substances
J. Balch a, C. Guguen b,
a
b

Environmental and Life Sciences Graduate Program, Trent University, ON, Canada
Chemistry Department, Trent University, ON, Canada

h i g h l i g h t s

g r a p h i c a l a b s t r a c t

 Diffusive properties determined for

aquatic DOM and HS ranging from

500 to 1750 Da.
 Molecular weight of diffused material
is preserved.
 Power regression model of D vs.
molecular weight calculated for 11
aquatic DOM samples.

DOM

DOM??

DGT assembly
with binding gel
a r t i c l e

i n f o

Article history:
Received 26 March 2014
Received in revised form 27 June 2014
Accepted 4 July 2014

Handling Editor: Keith Maruya

Keywords:
Dissolved organic matter (DOM)
Diffusion coefcient
Diffusive gradients in thin lms (DGT)
UVvisible spectroscopy
Field ow fractionation

Diffusive Gel

Prefilter

a b s t r a c t
In situ measurements of labile metal species using diffusive gradients in thin lms (DGT) passive
samplers are based on the diffusion rates of individual species. Although most studies have dealt with
chemically isolated humic substances, the diffusion of dissolved organic matter (DOM) across the hydrogel is not well understood. In this study, the diffusion coefcient (D) and molecular weight (MW) of 11
aquatic DOM and 4 humic substances (HS) were determined. Natural, unaltered aquatic DOM was capable of diffusing across the diffusive gel membrane with D values ranging from 2.48  106 to
5.31  106 cm2 s1. Humic substances had diffusion coefcient values ranging from 3.48  106 to
6.05  106 cm2 s1, congruent with previous studies. Molecular weight of aquatic DOM and HS samples
(5001750 Da) measured using asymmetrical ow eld-ow fractionation (AF4) strongly inuenced D,
with larger molecular weight DOM having lower D values. No noticeable changes in DOM size properties
were observed during the diffusion process, suggesting that DOM remains intact following diffusion
across the diffusive gel. The inuence of molecular weight on DOM mobility will assist in further understanding and development of the DGT technique and the uptake and mobility of contaminants associated
with DOM in aquatic environments.
2014 Elsevier Ltd. All rights reserved.

1. Introduction

Corresponding author. Tel.: +1 (705) 748 1011; fax: +1 (705) 748 1625.
E-mail address: celinegueguen@trentu.ca (C. Guguen).
http://dx.doi.org/10.1016/j.chemosphere.2014.07.013
0045-6535/ 2014 Elsevier Ltd. All rights reserved.

Dissolved organic matter (DOM) is a major constituent of

aquatic environments, and plays a vital role in biogeochemical processes (Perdue and Gjessing, 1990; Kullberg et al., 1993) and the

J. Balch, C. Guguen / Chemosphere 119 (2015) 498503

overall aquatic ecosystem health (Aiken et al., 2011; Neale et al.,

2011). DOM comprises a variety of organic constituents ranging
in molecular weight (Beckett, 1987), structure, and optical properties (Coble et al., 1990; Yamashita and Tanoue, 2003). For example,
Cuss and Guguen (2012) found that the molecular weight distribution of DOM determined by asymmetrical ow eld-ow fractionation (AF4) ranged from approximately 300 to 4300 Da, with
smaller molecular weights attributed to polyphenol/tyrosine-like
constituents of DOM and fulvic/humic-like constituents attributed
to larger molecular weights. This result is in agreement with previous studies using size-exclusion chromatography, tangential ultraltration and AF4 (Her et al., 2003; Boehme and Wells, 2006; Maie
et al., 2007; Guguen and Cuss, 2011).
Diffusive gradients in thin lms (DGT; Davison and Zhang,
1994) is a widely used analytical technique for in situ metal speciation monitoring (Guguen et al., 2012; Ferreira et al., 2013;
Sherwood et al., 2009; Zhang and Davison, 1995; Zhang and
Davison, 2001; Sndergaard, 2007). The DGT technique is based
on the diffusion of chemical species across thin gels produced
with known characteristics that can be controlled (e.g., gel type
and thickness; Scally et al., 2003, 2006). Previously, it has been
theorized that the diffusion of DOM in agarose polyacrylamide
diffusive hydrogels may be restricted because of the amphiphilic
properties and pseudo-micellar behavior of DOM. Thus, it has
been theorized that complexed species including DOM-metal species are incapable of diffusing through the gel (van der Veeken
et al., 2010; von Wandruszka, 1998). However, the DGT diffusive
gel pore size has been estimated at 15 nm (Lead et al., 2000a;
Zhang and Davison, 2001), which could be large enough to allow
the diffusion of DOM and associated metals in natural waters.
Indeed, Uher et al. (2013) showed that large DOM-metal complexes such as humic complexes can contribute to the ux of
metal measured by DGTs.
Previous studies have shown the diffusion of several humic
substances through polyacrylamide gels to range from
1.81  106 to 3.23  106 cm2 s1 (Beckett, 1987; Lead et al.,
2000a,b, 2003; Wang et al., 2001; Furukawa and Takahashi,
2008), which is 4670% slower than free metals (approx.
6.00  106 cm2 s1) (Zhang and Davison, 2000). These studies
also revealed that the diffusion coefcients (D) of humic and fulvic substances depend on their molecular weight as well as the
solution pH and ionic strength. Bufe (1988) proposed a model
based on a cubic root dependency between D and the molecular
weight (MW) of various spherical organic macromolecules ranging from 342 Da for sucrose to 6  107 Da for a virus. The D values reported in most studies including Bufe (1988) have been
generated using commercially available humic and fulvic acid
standards, macromolecules and virus. However, it is possible, that
commercial standards have been chemically and structurally
altered by the extraction procedures (International Humic
Substan Society, 1981) and therefore may not truly reect the
structure and native characteristics of unaltered, aquatic DOM
(Kuwatsuka et al., 1992). Therefore, there is a need to further
characterize the diffusive properties of aquatic DOM in order to
provide correct interpretation of the measurements made using
DGT in freshwater environments.
In this study, we aimed at (1) determining the diffusive properties of aquatic DOM and humic substance samples over a wide
range of molecular weights (5001750 Da) using a diffusive cell
apparatus and (2) assessing the integrity of organic matter passing through the diffusive gel using the asymmetrical ow eldow fractionation (Guguen and Cuss, 2011). The use of both
standards and natural unfractionated DOM in this study allows
for a comparison between altered and natural DOM, and provides
further information regarding the diffusion of these organic
species.

499

2. Materials and methods

2.1. DOM samples
Seven aquatic sources of natural DOM were used: sand ltered
and raw Otonabee River (44.37N/78.28W; 273 12 lS cm1; R1
R2), Otonabee Wetland (44.40N/78.26W; 315 4 lS cm1; R3),
Trent University Nature Conservatory Wetland outlet (44.34N/
78.28W; 359 9 lS cm1; R4), Baxter Creek (44.15N/78.45W,
289 4 lS cm1; R5) and two stream samples (306 5 and
349 3 lS cm1; R67) near the Dorset Research Center, Dorset,
Ontario (45.24N/78.89W). One additional source of natural
DOM also tested included algal exudates generated from Scenedesmus acutus (Canadian Phycological Research Centre; University of
Waterloo, Canada) a freshwater, green alga (McIntyre and
Guguen, 2012) collected at 5, 6, 30 and 45 d of growth
(312 6 lS cm1; A1A4, respectively). These samples were chosen to provide a range of DOM types differing in source (e.g., riverine, terrestrial, algal), and molecular weight (5001750 Da; see
below for details). Each aquatic sample was ltered using a
0.7 lm pre-combusted glass ber lter (Whatman) and a
0.22 lm nitrocellulose lter (Millipore).
The humic substances (HS) samples purchased from the International Humic Substances Society (IHSS) included: Suwannee
River fulvic and humic (i.e. SRFA 2S101F and SRHA 2S101H) and
Nordic Aquatic fulvic and humic (i.e. NAFA 1R105F and NAHA
1R105H). Each HS solution was prepared at a concentration of
2.5 mg C L1 in 0.01 M HEPES adjusted to pH 7.1 using 0.02 M
NaOH.

2.2. Diffusion cell experiments

The diffusive gel was comprised of an acrylamide monomer
(15 vol%) cross-linked with an agarose derivative (0.3 vol%)
(Zhang and Davison, 1999). All experiments with aquatic DOM
and HS standards were conducted with the 0.8 mm hydrogel,
which is the thickness most commonly used in DGT studies (e.g.
Scally et al., 2003, 2006; Sndergaard, 2007; Furukawa and
Takahashi, 2008; Guguen et al., 2011; Han et al., 2013). The gel
thicknesses were veried using a digital caliper to an accuracy of
0.02 mm.
The diffusion cell apparatus (Zhang and Davison, 1999; Scally
et al., 2006) was comprised of two cube-shaped, acrylic plexi-glass
compartments, each with an interconnecting 1.2 cm diameter
opening. In this way diffusion of DOM from one diffusion cell into
the adjacent cell was moderated by the diffusive hydrogel placed
between the two acrylic plexi-glass cells (Wang et al., 2001;
Scally et al., 2006). The volumes of source and receiving compartments were 550 and 1200 mL, respectively. The volumes of each
cell were constructed to be larger than the amount of solution
added in order to leave enough room for the insertion of the stirrer
attached to the Nexxtech 12 V DC hobby motor, powered using 3 V.
A 2.5 cm diameter disc of diffusive gel was secured in-between the
interconnecting cell openings by placing the hydrated gel overtop
of one of the openings of one half of the assembly with the opening
facing the ceiling. The additional half of the assembly was then
carefully overlaid on top of the bottom assembly without disturbing the position of the gel and the whole assembly was then
clamped together.
It was important to maintain a similar pH and ionic strength of
the solutions on both sides of the diffusive gel to ensure that diffusion was not altered or inuenced by an imbalance in pH or ionic
strength between the source and receiving solutions. The source
and receiving solutions for HS samples consisted of 0.01 M HEPES
(200250 ls cm1) adjusted to approximately pH 7.1 (0.07; n = 8)

500

J. Balch, C. Guguen / Chemosphere 119 (2015) 498503

using 0.02 M NaOH. The HEPES buffer was chosen in order to maintain a constant pH throughout the duration of the experiment,
minimizing the inuence of pH variation on D (Lead et al., 2003).
In the aquatic DOM experiments, the receiving solution was
0.01 M NaNO3 adjusted to pH 7.1 and river conductivity using
0.02 M NaOH and Milli-Q water (Millipore; >18 O) to reect properties similar to those of natural waters. The aquatic DOM source
solutions had a pH of 7.06 0.34 (n = 12). No signicant differences
in pH and conductivities were observed between the solutions on
either side of the diffusion cells in all experiments (ANOVA;
p < 0.05). The same pH was chosen for both natural DOM and HS
to avoid any pH dependency (Lead et al., 2003; Scally et al., 2006).
Approximately 30 mL of solution were collected from both the
source and receiving compartments of the diffusive cell every
24 h for ve to six consecutive days (t = 0, 24, 48, 72, 96, 120 and
144 h). Longer periods of time (24 h) were used than previously
described in other studies (Scally et al., 2006; Furukawa and
Takahashi, 2008) in order to increase the concentration of diffused
DOM between time intervals and thus increase accuracy in calculating slopes. Samples were stored in 40 mL pre-combusted, amber
glass vials at 4 C and analyzed within 7 d. All experiments were
performed at room temperature and maintained at 23 C 0.4 C.
The diffusion coefcients were corrected to 25 C using the method
adapted from Zhang and Davison (1995) and Scally et al. (2006). All
diffusion coefcients experiments were conducted in duplicate
except for A14, R1R2, R6 and NAFA, due to tearing of the gel
membrane. With the amount of pressure placed on the gel membrane with it being secured between two planes of plexi-glass, it
was fragile and could rupture easily with increases in stirrer speed
and water turbulence. D values for A12 could not be measured
due to full rupturing of the gel membrane, however, initial MW
was measured for these samples and included in the MW results
for A14. R5 was used test the preservation of DOM MW as it
diffused across the gel membrane over time; no D values were
available for R5.

2.4. Molecular weight determination

The molecular weight (MW) of DOM and HS was estimated
using an asymmetrical-ow eld ow fractionator (AF4; Postnova
Analytical) coupled to a diode array detector (DAD) following the
procedure outlined by Guguen and Cuss (2011) and Cuss and
Guguen (2012). A calibration curve was generated twice a day
based on ve macromolecules of known molecular weight and
peak maximum absorbance (kmax): rhodamine B (479 Da;
kmax = 550 nm), bromophenol blue (692 Da; kmax = 591 nm), vitamin B12 (1.35 kDa; kmax = 355 nm), cytochrome C (12.4 kDa;
kmax = 405 nm) and lysozyme (14.3 kDa; kmax = 270 nm) (Guguen
and Cuss, 2011). The AF4 method allows for a determination of
MW of light absorbing organic macromolecules with a precision
of approximately 100 Da (Guguen and Cuss, 2011).
3. Results
3.1. Diffusion coefcients
The determinations of D for aquatic DOM and HS were based on
a linear increase in DOC concentration over time (r2 > 0.82) (Fig. 1)
with the slope directly proportional to D. The D value was temperature corrected (Zhang and Davison, 1995) and reported at 25 C.
The D values for both natural DOM and HS ranged from
2.48  106 to 6.05  106 cm2 s1 (Table 1; Fig. 2A). The D values
of fulvic and humic acids (4.53  106 0.35  106 and
4.80  106 1.40  106 cm2 s1) were on the higher end of the
range previously reported (1.95.1  106 cm2 s1 and 1.1
4.5  106 cm2 s1, for FA and HA respectively; Beckett, 1987;
Dixon and Larive, 1997; Morris et al., 1999; Lead et al., 1999;
Lead et al., 2000a,b; Wang et al., 2001). This is probably due to
the lower ionic strength used in this work as higher D values are
typically reported at lower ionic strength (Lead et al., 2000a).
3.2. Molecular weight

2.3. Diffusion coefcient determination

D slope  Dg=A  C  3600

where A (cm2) is the exposed area of gel, Dg (cm) is the thickness of

the gel between the two compartments. The slope (mg h1), refers
to the mass of DOC based on the concentration calculated from
absorbance data (Tipping et al., 2009; Carter et al., 2012) measured
over time in 24 h intervals. C is the concentration of DOC (mg cm3)
in the source solutions at time 0 h, measured using absorbance and
converted to concentration (Carter et al., 2012). The 3600 converts
the time from hour intervals into seconds (Scally et al., 2006). The
diffusion coefcients are reported here at the standardized temperature of 25 C (Zhang and Davison, 1995).

The molecular weights of DOM and humic substances were

determined using AF4 using the macromolecule-based calibration
(Guguen and Cuss, 2011) (Fig. 3). For example, the algae DOM
at peak maximum has a shorter retention time than of the river
DOM, suggesting a lower MW for algae relative to river DOM.
The MW reported in this study ranged from approximately
507 18 to 1748 27 Da (Table 1; Fig. 2B) and t within the range

0.9

0.8

mass of DOC [mg]

The absorbance spectra of both source and receiving compartments were measured using a Shimadzu UV 2550 spectrophotometer equipped with a 10-cm quartz cell. Absorbance values
between 250 and 700 nm were blank corrected using Milli-Q water
(Millipore; >18 O). Light scattering was also subtracted from each
sample using the average absorbance between 650 and 700 nm
from the blank. The dissolved organic carbon (DOC) concentrations
were estimated based on absorbance measured at two wavelengths 254 and 270 nm using a two-component model (Tipping
et al., 2009; Carter et al., 2012). The presence of a non-absorbing
component was also included in the DOC estimate (Carter et al.,
2012).
The diffusion coefcients D were calculated for each type of
DOM using the equation adapted from Scally et al. (2006):

0.7

0.6

0.5

0.4
0

20

40

60

80

100

120

140

160

Time [hr]
Fig. 1. Examples of the linear increase of diffused dissolved organic carbon vs.
elapsed time (s R1, d R6; NSRHA; r2 = 0.99, 0.96, and 0.99 respectively). Slope of
line created using SigmaPlot (11.0) linear line of best t.

501

J. Balch, C. Guguen / Chemosphere 119 (2015) 498503

Table 1
Molecular weight (Da) and diffusion coefcient (cm2 s1) values of DOM samples with standard deviation.
Sample ID

Sample name

MW (Da)

D (cm2 s1)

A1
A2
A3
A4
R1
R2
R3
R3
R4
R4
R5
R6
R7
R7
SRFA
SRHA
SRHA
NAFA
NAHA
NAHA

Algal exudate Day 6

Algal exudate Day 7
Algal exudate Day 30
Algal exudate Day 45
Otonabee River sand ltered
Otonabee River in situ
Otonabee Wetland
Otonabee Wetland
Nature conservation
Nature conservation
Baxter Creek
Dorset Stream1
Dorset Stream2
Dorset Stream2
Suwannee River fulvic acid
Suwannee River humic acid
Suwannee River humic acid
Nordic aquatic fulvic acid
Nordic aquatic humic acid
Nordic aquatic humic acid

507 18
544 37
721 52
973 21
1198 19
920 140
989 54
1065 54
865 18
891 18
1110 55
1482 130
1710 27
1748 27
1089 105
1490 0
1490 0
961 50
1405 8
1394 8

N/A N/A
N/A N/A
5.04E-06 1.65E-07
3.13E-06 3.05E-07
3.83E-06 7.27E-08
4.72E-06 4.66E-08
4.14E-06 7.70E-08
4.16E-06 6.05E-07
5.30E-06 4.93E-07
5.31E-06 2.13E-07
N/A N/A
2.94E-06 1.48E-07
2.94E-06 1.32E-07
2.48E-06 9.02E-08
4.78E-06 4.65E-07
5.96E-06 1.29E-07
6.05E-06 3.86E-08
4.28E-06 3.05E-06
3.71E-06 1.30E-07
3.48E-06 3.66E-08

Fig. 2. Range of (A) D and (B) MW values for aquatic DOM (Algae n = 2 and 4, respectively; River/wetland outlets n = 9 and 9, respectively) and humic substances (FA n = 2 and
4, respectively and HA n = 4 and 4, respectively) found in this study.

1600

6000

1400

4000

1200
Void
Peak

MW [Da]

Absorbance [mAU]

5000

3000
2000

1000
800
600
400

1000

200
0

0
6

10

12

14

16

18

20

Time [min]
Fig. 3. Fractograms of A3 (black) and R5 (gray) in the source compartment. Note:
the void peak as indicated by the arrows consists of a small, unfocused material and
appears immediately following elution of the material from the AF4 membrane.

A3

R5

Fig. 4. Time series of MW in the receiving cells at 28 h (black), 49 h (dark gray) and
149 h (light gray) for A3 and R5. The average MW in the source compartment
(white) is provided as comparison. Error bars represent standard deviation
(>100 Da) (n = 2). Note: no MW at 28 h for A3 is available due to very weak signal
(i.e. small amount of material diffused across the gel).

J. Balch, C. Guguen / Chemosphere 119 (2015) 498503

of DOM and HS previously measured using AF4 (Guguen and Cuss,

2011; Cuss and Guguen, 2012 and references therein). The algal
DOM (i.e. A1A4) exhibited the lowest MW values (507973 Da)
whereas the stream samples (R67) showed the largest MW values
(i.e. 14821748 Da). The aquatic DOM and HS were in the intermediate size range (i.e. 8651198 and 9611490 Da, respectively).
SRFA had a molecular weight smaller than SRHA (1025 91 vs.
1445 52 Da, respectively; p < 0.05) and compared well with previous results (Guguen and Cuss, 2011; Remucal et al., 2012).
To assess the impact of diffusion on the molecular weight of
natural DOM, the MW of samples A3 and R5 was monitored at
28 h, 49 h and 149 h. in the source and receiving compartments
(Fig. 4). The molecular weight of the diffused DOM was relatively
stable (100 Da) and comparable to the source solutions. It can
be noted that the absorbance intensity of A3 in the diffused compartment at time 28 h was too low to be detected by AF4 and thus
no MW was available for A3 at 28 h. For example, the A3 MW in
the receiving cell ranged from 710 to 781 Da, which is consistent
with the molecular weight of the source DOM (721 Da). This result
suggests that the average MW is preserved through diffusion
process.
4. Discussion
4.1. DOM vs. HS
Despite a larger variability in D values of aquatic DOM, no signicant difference was found between aquatic DOM and HS samples (3.98  106 1.04  106 vs. 4.71  106 1.10  106
cm2 s1; p > 0.05) (Table 1; Fig. 2A). The larger range in DOM D values reects the natural heterogeneity of the aquatic DOM mixture
composed of humic acids, fulvic acids, low molecular weight (MW)
organic acids, carbohydrates, proteins, and other compound classes
resulting in a variable size distribution. For example, the >1
10 kDa DOM accounted for 5% to 7585% of DOM in aquatic environments (Guo and Santschi, 2007). No signicant difference in D
values was found between HA and FA (p > 0.05). This contrasts
with Lead et al. (2000b) who found a small difference in D values
between FA and HA using uorescence correlation spectroscopy.
In conclusion, this study has demonstrated that natural, aquatic
DOM diffuses similarly to isolated HS, conrming the use of FA
and HA as DOM proxies in the calculation of DGT-labile species
(Guguen et al., 2011; Han et al., 2013).
Despite the fact that both natural DOM and HS were capable of
diffusing through the DGT hydrogels, their D values (Fig. 2A) were
within 1 order of magnitude of previously reported D of free metal
species (Zhang and Davison, 1999; Scally et al., 2006). The time
required to travel across a gel thickness Dg can be estimated as follows (Scally et al., 2006):

Dg 2
2D

For example, copper with a D of 6.23  106 cm2 s1 would take
approximately 17 min to travel across a gel thickness of 0.08 cm
while DOM would take between 21 and 36 min for the slowest
and the fastest diffusion rates reported here (A2 and R7, respectively; Fig. 2A) to travel across the gel. Although DOM diffuses
more slowly than free metals, it is likely that it will signicantly
diffuse across the gel with a typical DGT deployment time of few
days.
4.2. Relationship between diffusion coefcient and molecular weight
The dependence of the D values on molecular weight of natural
DOM is depicted in Fig. 5. The natural DOM (circles) and HS sam-

7e-6

6e-6

D [cm2.s-1]

502

5e-6

4e-6

3e-6

2e-6
600

800

1000

1200

1400

1600

1800

MW [Da]
Fig. 5. Correlation between D and MW (plain proposed model; dotted Bufes
model (1988)) using natural DOM (circles) HS (squares) were not used in tting the
exponential decay model. The model developed in this study used the exponential
decay function tting line in SigmaPlot (11.0) with r2 = 0.75, ANOVA; p < 0.05.

ples (squares) showed a decrease in diffusion coefcients with

increasing molecular weight of DOM, which is consistent with previous studies (Bufe, 1988). The natural DOM and HS (with the
exception of NAHA) demonstrated a signicant exponential relationship between D and MW. The reason for the fast diffusion of
NAHA remains unclear but Zhang and Shimaoka (2013) suggested
that the manufacturing process may result in a wider range of diffusion values (Zhang and Shimaoka, 2013). However, the observed
D values of low MW DOM (<950 Da) were 30% higher than that
based on Bufes model (dotted line; Fig. 5). As a result, a new
power regression model (D = 1.1  103 * (MW)0.81; r2 = 0.78)
based on 11 aquatic DOM is proposed. The root mean square deviation (RMSD) of the proposed model was 0.89  106 which is 30%
lower than that of obtained using Bufes model. The results of the
model developed demonstrate a lower standard deviation between
the model and observed data as well as a higher trend in D values
of lower MW than predicted using Bufes model. The Bufes
model was dened using globular organic macromolecules with
MW proportional to volume of a hydrated spherical molecule.
However, DOM and HS are much more heterogeneous and polydisperse than globular macromolecules as revealed by microscopic
techniques (Wilkinson et al., 1999). Overall, the new proposed
model gave a higher accuracy in predicting D using MW for both
laboratory/standard DOM and in-situ DOM pertaining to DGT
experiments conducted in the eld.
5. Conclusions
It was determined that both natural DOM and humic substances
can diffuse through the most commonly used diffusive hydrogel
used in DGTs. DOM has the capability to diffuse through gel membranes with decreasing rates of diffusion directly related to
increasing molecular weight. The proposed power regression
model allows to accurately determining the D values of natural
DOM of size ranging from 500 to 1750 Da. The goodness of t
in the low DOM MW range (i.e. 500950 Da) was improved by
30% relative to the Bufes model. No noticeable change in MW
was found in aquatic DOM diffusing across the diffusion gel, suggesting that DOM quality is preserved. The results of this study
demonstrate the importance of studying the diffusive characteristics of DOM in environmental monitoring, as inaccurate estimations of metal bioavailability and toxicity may occur with the use
of DGT samplers without careful consideration of the diffusion of
DOM associated species.

J. Balch, C. Guguen / Chemosphere 119 (2015) 498503

Acknowledgements
This study would not have been possible without the assistance
of Chad Cuss and Antoine Perroud and their sample analysis,
expertise and knowledge of instrumentation. Acknowledgments
are also made towards two anonymous reviewers and editor of
Chemosphere for their comments as well as NSERC and Canada
Research Chair program for their nancial support to this project.
References
Aiken, G., Gilmour, C., Krabbenhoft, D., Orem, W., 2011. Dissolved organic matter in
the orida everglades: implications for ecosystem restoration. Crit. Rev.
Environ. Sci. Technol. 41, 217248.
Beckett, R., 1987. Determination of molecular weight distributions of fulvic and
humic acids using ow eld-ow fractionation. Environ. Sci. Technol. 21, 289
295.
Boehme, J., Wells, M., 2006. Fluorescence variability of marine and terrestrial
colloids: examining size fractions of chromophoric dissolved organic matter in
the damariscotta river estuary. Mar. Chem. 101, 95103.
Bufe, J., 1988. 1988 Complexation Reactions in Aquatic Systems: An Analytical
Approach. Ellis Horwood Ltd., Chichester, UK.
Carter, H., Tipping, E., Koprivnjak, J., Miller, M., Cookson, B., Hamilton-Taylor, J.,
2012. Freshwater DOM quantity and quality from a two-component model of
UV absorbance. Water Res. 46, 45324542.
Coble, P.G., Green, S.A., Blough, N.V., Gagosian, R.B., 1990. Characterization of
dissolved organic matter in the Black Sea by uorescence spectroscopy. Nature
348, 432435.
Cuss, C.W., Guguen, C., 2012. Determination of relative molecular weights of
uorescent components in dissolved organic matter using asymmetrical ow
eld-ow fractionation and parallel factor analysis. Anal. Chim. Acta 733, 98
102.
Dixon, A.M., Larive, C.K., 1997. Modied pulsed-eld gradient NMR experiments for
improved selectivity in the measurement of diffusion coefcients in complex
mixtures: application to the analysis of the suwannee river fulvic acid. Anal.
Chem. 69, 21222128.
Davison, W., Zhang, H., 1994. In situ speciation measurements of trace components
in natural waters using thin-lm gels. Nature 367, 546548.
Ferreira, D., Ciffroy, P., Tusseau-Vuillemin, M., Bourgeault, A., Garnier, J., 2013. DGT
as surrogate of biomonitors for predicting the bioavailability of copper in
freshwaters: an ex situ validation study. Chemosphere 91, 241247.
Furukawa, K., Takahashi, Y., 2008. Effect of complexation with humic substances on
diffusion of metal ions in water. Chemosphere 73, 12721278.
Guguen, C., Cuss, C.W., 2011. Characterization of aquatic dissolved organic matter
by asymmetrical ow eld-ow fractionation coupled to UVvisible diode array
and excitation emission matrix uorescence. J. Chromotogr. A 1218, 4188
4198.
Guguen, C., Clarisse, O., Perroud, A., McDonald, A., 2011. Chemical speciation and
partitioning of trace metals (Cd Co, Cu, Ni, Pb) in the lower Athabasca River and
its tributaries (Alberta, Canada). J. Monit. Environ. http://dx.doi.org/10.1039/
c1em10563a.
Guguen, C., Burns, D.C., McDonald, A., Ring, B., 2012. Structural and optical
characterization of dissolved organic matter from the lower Athabasca River,
Canada. Chemosphere 87, 932937.
Guo, L., Santschi, P.H., 2007. Ultraltration and its applications to sampling and
characterization of aquatic colloids. In: Wilkinson, K.J., Lead, J.R. (Eds.),
Environmental Colloids and Particles: Behaviour, Separation and
Characterization. John Wiley & Sons Ltd., Chichester, UK. http://dx.doi.org/
10.1002/9780470024539.ch4.
Han, S., Naito, W., Hanai, Y., Masunaga, S., 2013. Evaluation of trace metals
bioavailability in Japanese river waters using DGT and a chemical equilibrium
model. Wat. Res. 47, 48804892.
Her, N., Amy, G., McKnight, D., Sohn, J., Yoon, Y., 2003. Characterization of DOM as a
function of MW by uorescence EEM and HPLC-SEC using UVA, DOC, and
uorescence detection. Water Res. 37, 42954303.
International Humic Substances Society, 1981. Outline of Extraction Procedures,
Distributed to the Members of International Humic Substances Society.
Kullberg, A., Bishop, K.H., Hargeby, A., Jansson, M., Petersen, R.C., 1993. The
ecological signicance of dissolved organic carbon in acidied waters. Ambio
22, 331337.
Kuwatsuka, S., Watanabe, A., Itoh, K., Arai, S., 1992. Comparison of two methods of
preparation of humic and fulvic acids, IHSS method and NAGOYA method. Soil
Sci. Plant Nutr. 38, 2330.

503

Lead, J.R., Balnois, E., Hosse, M., Menghetti, R., Wilkinson, K., 1999. Characterization
of Norwegian natural organic matter. Size, diffusion coefcients and
electrophoretic mobilities. J. Environ. Intern. 25, 245258.
Lead, J., Wilkinson, K.J., Cutak, B.J., Larive, C.K., Assemi, S., Beckett, R., 2000a.
Diffusion coefcients and polydispersities of the Suwannee River fulvic acid:
comparison of uorescence correlation spectroscopy, pulsed-eld gradient
nuclear magnetic resonance and ow eld-ow fractionation. Environ. Sci.
Technol. 34, 35083513.
Lead, J., Wilkinson, K.J., Starchev, K., Canonica, S., Bufe, J., 2000b. Determination of
diffusion coefcients of humic substances by uorescence correlation
spectroscopy: role of solution conditions. Environ. Sci. Technol. 34, 13651369.
Lead, J., Starchev, K., Wilkinson, K.J., 2003. Diffusion coefcients of humic
substances in agarose gel and in water. Environ. Sci. Technol. 37, 482487.
Maie, N., Scully, N.M., Pisani, O., Jaff, R., 2007. Composition of a protein-like
uorophore of dissolved organic matter in coastal wetland and estuarine
ecosystems. Water Res. 41, 563570.
McIntyre, A.M., Guguen, C., 2012. Binding interactions of algal-derived dissolved
organic matter with metal ions. Chemosphere 90, 620626.
Morris, K.F., Cutak, B.J., Dixon, A.M., Larive, C.K., 1999. Analysis of diffusion
coefcient distributions in humic and fulvic acids by means of diffusion ordered
nmr spectroscopy. Anal. Chem. 71, 53155321.
Neale, P.A., Antony, A., Gernjak, W., Leslie, G., Escher, B.I., 2011. Natural versus
wastewater derived dissolved organic carbon: implications for the
environmental fate of organic micropollutants. Water Res. 45, 42274237.
Perdue, E.M., Gjessing, E.T., 1990. Organic Acids in Aquatic Ecosystems. Wiley, New
York.
Remucal, C.K., Cory, R.M., Sander, M., McNeil, K., 2012. Low molecular weight
components in an aquatic humic substance as characterized by membrane
dialysis and orbitrap mass spectrometry. Environ. Sci. Technol. 46, 93509359.
Scally, S., Davison, W., Zhang, H., 2003. In situ measurements of dissociation kinetics
and labilities of metal complexes in solution using DGT. Environ. Sci. Technol.
37, 13791384.
Scally, S., Davison, W., Zhang, H., 2006. Diffusion coefcients of metals and metal
complexes in hydrogels used in diffusive gradients in thin lms Anal. Chim.
Acta 558, 222229.
Sherwood, J.E., Barnett, D., Barnett, N.W., Dover, K., Howitta, J., Ii, H., Kewc, P.,
Mondona, J., 2009. Deployment of DGT units in marine waters to assess the
environmental risk from a deep sea tailings outfall. Anal. Chim. Acta 652, 215
223.
Sndergaard, J., 2007. In situ measurements of labile Al and Mn in acid mine
drainage using diffusive gradients in thin lms. Anal. Chem. 79, 64196423.
Tipping, E., Corbishley, H.T., Koprivnjak, J.F., Lapworth, D.J., Miller, M.P., Vincent,
C.D., Hamilton-Taylor, J., 2009. Quantication of natural DOM from UV
absorption at two wavelengths. Environ. Chem. 6, 472476.
Uher, E., Tusseau-Vuilleminc, M., Gourlay-Franced, C., 2013. DGT measurement in
low ow conditions: diffusive boundary layer and lability considerations.
Environ. Sci.: Processes Impacts 15, 13511358.
van der Veeken, P.L.R., Chakraborty, P., van Leeuwen, H.P., 2010. Accumulation of
humic acid in DET/DGT Gels. Environ. Sci. Technol. 44, 42534257.
Von Wandruszka, R., 1998. The micellar model of humic acid: evidence from pyrene
uorescence measurements. Soil Sci. 163, 921930.
Wang, Y., Combe, C., Clark, M.M., 2001. The effects of pH and calcium on the
diffusion coefcient of humic acid. J. Membrane Sci. 183, 4960.
Wilkinson, K.J., Balnois, E., Leppard, G.G., Bufe, J., 1999. Characteristic feature of the
major components of freshwater colloidal organic matter revealed by
transmission electron and atomic force microscopy. Colloids Surf. A 155, 287
310.
Yamashita, Y., Tanoue, E., 2003. Chemical characterization of protein-like
uorophores in DOM in relation to aromatic amino acids. Mar. Chem. 82,
255271.
Zhang, H., Davison, W., 1995. Performance characteristics of diffusion gradients in
thin lms for the in situ measurement of trace metals in aqueous solution. Anal.
Chem. 67, 33913400.
Zhang, H., Davison, W., 1999. Diffusional characteristics of hydrogels used in DGT
and DET techniques. Anal. Chim. Acta 398, 329340.
Zhang, H., Davison, W., 2000. Direct in situ measurements of labile inorganic and
organically bound metal species in synthetic solutions and natural waters using
diffusive gradients in thin lms. Anal. Chem. 72, 44474457.
Zhang, H., Davison, W., 2001. In situ speciation measurements. Using diffusive
gradients in thin lms (DGT) to determine inorganically and organically
complexed metals. Pure Appl. Chem. 73, 915.
Zhang, H., Shimaoka, T., 2013. Formation of humic substances in weathered MSWI
bottom ash. Sci. World J. 2013 (384806), 15. http://dx.doi.org/10.1155/2013/
384806.