Content uploaded by Y. Jiang
Author content
All content in this area was uploaded by Y. Jiang on Feb 20, 2018
Content may be subject to copyright.
Evaluating the environmental impact of artificial
sweeteners: A study of their distributions,
photodegradation and toxicities
Ziye Sang
1
, Yanan Jiang
1
, Yeuk-Ki Tsoi, Kelvin Sze-Yin Leung*
Department of Chemistry, Hong Kong Baptist University, Kowloon Tong, Hong Kong Special Administrative Region
article info
Article history:
Received 29 May 2013
Received in revised form
30 October 2013
Accepted 2 November 2013
Available online 14 November 2013
Keywords:
Artificial sweetener
Emerging environmental contami-
nant
Photodegradation
Phototoxicity
UV/TiO
2
photocatalysis
Wastewater treatment
abstract
While having a long tradition as safe food additives, artificial sweeteners are a newly
recognized class of environmental contaminants due to their extreme persistence and
ubiquitous occurrence in various aquatic ecosystems. Resistant to wastewater treatment
processes, they are continuously introduced into the water environments. To date how-
ever, their environmental behavior, fate as well as long term ecotoxicological contributions
in our water resources still remain largely unknown. As a first step in the comprehensive
study of artificial sweeteners, this work elucidates the geographical/seasonal/hydrological
interactions of acesulfame, cyclamate, saccharin and sucralose in an open coast system at
an estuarine/marine junction. Higher occurrence of acesulfame (seasonal average:
0.22 mgL
1
) and sucralose (0.05 mgL
1
) was found in summer while saccharin (0.11 mgL
1
)
and cyclamate (0.10 mgL
1
) were predominantly detected in winter. Seasonal observations
of the four sweeteners suggest strong connections with the variable chemical resistance
among different sweeteners. Our photodegradation investigation further projected the
potential impact of persistent acesulfame and sucralose compounds under prolonged
exposure to intensive solar irradiation. Real-time observation by UPLCeESI/MS of the
degradation profile in both sweeteners illustrated that formation of new photo by-products
under prolonged UV irradiation is highly viable. Interestingly, two groups of kinetically
behaved photodegradates were identified for acesulfame, one of which was at least six
times more persistent than the parent compound. For the first time, acute toxicity for the
degradates of both sweeteners were arbitrarily measured, revealing photo-enhancement
factors of 575 and 17.1 for acesulfame and sucralose, respectively. Direct comparison of
photodegradation results suggests that the phototoxicity of acesulfame degradation
products may impact aquatic ecosystems. In an attempt to neutralize this prolonged
environmental threat, the feasibility of UV/TiO
2
as an effective mineralization process in
wastewater treatment was evaluated for both sweeteners. Under an environmental and
technical relevant condition, a >84% removal rate recorded within 30 min and complete
photomineralization was achieved within 2 h and delivering the best cost efficiency
comparing to existing removal methods. A compilation of distribution, degradation,
*Corresponding author. Tel.: þ852 34115297; fax: þ852 34117348.
E-mail address: s9362284@hkbu.edu.hk (K.S.-Y. Leung).
1
These authors equally contributed to this work.
Available online at www.sciencedirect.com
ScienceDirect
journal homepage: www.elsevier.com/locate/watres
water research 52 (2014) 260e274
0043-1354/$ esee front matter ª2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.watres.2013.11.002
toxicity and attenuation results presented in this paper will go through critical discussions
to explore some current issues and to pinpoint solutions for a better control in the emer-
gent contamination of artificial sweeteners.
ª2013 Elsevier Ltd. All rights reserved.
1. Introduction
Artificial sweeteners are being used as sugar substitutes in
considerable and increasing amounts in food and beverages,
especially for those who are diabetic and/or obese. They have
also been used in other personal care and pharmaceutical
products (Zygler et al., 2009) such as toothpastes. Although,
from the beginning of their use, there has been controversies
over their risk as potential carcinogens (Weihrauch and Diehl,
2004), these sweetener compounds are generally considered
to be safe for use in foodstuffs (Kroger et al., 2006; Ahmed and
Thomas, 1992; Cohen et al., 2008). Some of the low-calorie
sweeteners currently approved by different international au-
thorities as direct food additives include acesulfame, aspar-
tame, cyclamate, saccharin and sucralose (US FDA, 2006; EU,
2003). Other flavorings are continually being developed and
are increasingly commonly used in foodstuffs, especially
because they confer longer shelf-life. Just as these compounds
are metabolically inert in the human body, so scientists are
finding, they are also inert in the environment. Concern is
shifting from health concerns to ecosystem concerns. In
terms of environmental degradation, among the five most
commonly used artificial sweeteners named above, only
aspartame decomposes under normal usage conditions, and
safety clearance was given to the intake of even its breakdown
derivatives (US FDA, 1983). Outstanding chemical stability in
these sweeteners means they are passed out mainly un-
changed into the domestic wastewater treatment system,
with the intact compounds enter the aquatic environment
almost directly.
Of the variety of artificial sweeteners being used, only ace-
sulfame, cyclamate, saccharin and sucralose have been iden-
tified in wastewater effluents (Lange et al., 2012). Comparison
with the sweetener content of influent shows substantial but
variable resistance of these compounds to breakdown by
wastewater treatment. Common mechanical and secondary
microbial digestion can only partially mineralize and remove
sweetener pollutants. Acesulfame and sucralose have been
reported as the most persistent sweeteners with removal rates
as low as 40% and 20%, respectively (Scheurer et al., 2009).
Ironically, chemical and biological recalcitrance in compounds
has been valued as an ideal marker property for tracing the
influence of wastewater in the environment (Buerge et al.,
2009). Now, that attitude is shifting, as concern for the long-
term ecological effects are considered. Of the four most
widely allowed safe artificial sweeteners, acesulfame, cycla-
mate, saccharin and sucralose are currently not considered in
any existing effluent quality code, and no connection has been
established between their pervasive presence and any envi-
ronmental impact, until recently.
Widespread occurrence of acesulfame, cyclamate,
saccharin and sucralose have been recorded from nano- to
microgram levels in various rivers and lakes of European
countries (i.e. Switzerland, Germany, Austria, Sweden, Serbia,
Spain, UK, Belgium, Netherland, France, Italy and Norway)
and North America (Loos et al., 2009; Mead et al., 2009; Torres
et al., 2011). Sweeteners have also made their way into
groundwater networks through surface water infiltration and
percolation in soil aquifers, bringing levels to 34 mgL
1
for
acesulfame and 24 mgL
1
for sucralose (Van Stempvoort et al.,
2011). In contrast, lower concentrations of cyclamate and
saccharin were measured in these same waters. These dif-
ferences are probably due to subsurface attenuation processes
variably experienced among the sweetener species. To date,
the availability of occurrence data is confined to inland wa-
ters; very little is known about their fate in open coastal
environment, where complex distribution forces are driven by
the interplay of estuarine and oceanic hydrology and where a
much larger scale dispersion of pollutants becomes likely.
Especially, the behaviors of artificial sweeteners entering
macro-hydrological system climatically sensitive to temper-
ature, radiation, chemical and biological parameters, remain
limitedly explored.
While artificial sweeteners are marketed as metabolically
inert sugar substitutes, studies have revealed that they are not
entirely inert in the environment. One study has found a po-
tential xenobiotic interference in the normal biological func-
tions in ecosystem (e.g. photosynthesis and feeding behaviors
in zooplanktons) when and because these compounds invoke
organisms’ biological response to natural sugar (Kessler,
2009). For this newly emergent class of environmental con-
taminants, the long-term consequences of their ubiquitous
distribution and those potentiated by possible chemical
transformation over an extended pollution episode are
uncharacterized.
Artificial sweeteners degrade at varying rates under
different environmental conditions. Incubated in aerobic soils
for a period of 1e3 months, acesulfame and sucralose showed
signs of slow degradation, suggesting even the most persis-
tent sweeteners are not necessarily inert to microbial actions
(Buerge et al., 2011). In addition, positive observation of photo-
induced decomposition and initial by-product identification in
sucralose (Calza et al., 2013), indicating another probable
course of natural elimination after prolonged exposure to
sunlight. These findings significantly highlight another critical
issue: namely, the formation and accumulation of potentially
more deleterious by-products from natural degradation of
artificial sweeteners. Indeed, there have been well-
documented examples involving various other persistent
organic pollutants including polyaromatic hydrocarbons
(PAHs), pharmaceuticals, pesticides and personal-care prod-
ucts, in which the enhancement in degradation toxicity,
especially phototoxicity, has clearly implied unforeseen
environmental consequences over the long term (Petersen
water research 52 (2014) 260e274 261
and Dahllof, 2007; Klamerth et al., 2010). For instance, under
natural sunlight, sulfonamides increase as much as 16.5-fold
in toxicity. Because these drugs are so commonly used, they
pose a continuous threat (Jung et al., 2008). In contrast, little
attention has been paid to the degradation fate of artificial
sweeteners, until only recently a new study has initially
measured the photo-induced toxicity for sucralose subse-
quent to catalytic degradation. By comparing the activity of
marine bioluminescence bacteria in irradiated sucralose
samples, a real-time experiment revealed an observable
elevation in toxicity due to the production of photo-
transformation intermediates (Calza et al., 2013). The in-
crease in toxicity was not quantified in that study; however,
its relative results signal the importance of further investiga-
tion to verify, quantify and characterize these phototoxic ef-
fects. On a further note, current environmental safety
standards in total unawareness of their degradation fate and
phototoxic effect is likely to underestimate the actual impact
of artificial sweeteners, which the field with the present state
of knowledge is still unable to comprehend.
Given these situations and needs, a comprehensive envi-
ronmental study should be undertaken to evaluate and char-
acterize the environmental fate of the four most commonly
used artificial sweeteners acesulfame, cyclamate, saccharin
and sucralose. This work is a first step in that study. An original
occurrence profile in an open maritime system was firstly
investigated incorporating dual contribution from seasonal
and neighboring estuarine movements. To compensate for
blind spots in existing environmental impact reviews on
sweeteners, we studied the photo-degradation of persistent
acesulfame and sucralose to determine their transformation
profile and intermediate products, as well as to systematically
evaluate any short-term marine toxicity induced by UV irra-
diation. To our knowledge, this is the first time that the
phototoxicity of acesulfame and sucralose were systematically
quantified and compared. In long term to alleviate the threat
from and sustain a safer use of artificial sweeteners, the
application of a more efficient retention/removal barrier in
wastewater treatment becomes a necessary task. Nowadays,
photo-disinfection by means of UV irradiation has already
received recognition in wastewater and drinking water purifi-
cation works. It is also important to remark that with the use of
oxidants, photocatalytic treatment is an increasingly prom-
ising way to cope with refractory organic pollutants (Chong
et al., 2010). For this reason, the present study explored the
feasibility of TiO
2
-assisted photocatalysis to achieve complete
mineralization. Results from this aspect of our study offer a
practical solution for the problem of developing effective
pollution control of artificial sweeteners.
2. Experimental
2.1. Chemicals and reagents
Chemical standards for the artificial sweeteners sucralose,
acesulfame potassium, saccharin and sodium cyclamate were
purchased from SigmaeAldrich (MO., USA). Individual stock
solutions of 1000 mg L
1
of each of the four artificial sweet-
eners were prepared with Milli-Q water and stored in the dark
at 4 C. For quantitative analysis in the occurrence study, the
stock solutions were diluted to obtain 1000 mgL
1
aqueous
mixtures as intermediate standards. All solutions were stored
at 4 C.
For chromatographic separation and analyses, ammonium
acetate buffer salt (Scharlau Chemie S. A., Barcelona, Spain)
and HPLC grade methanol (Tedia, OH, USA) were used to
prepare the mobile phases. For photo-catalytic degradation
and removal studies, titanium dioxide (99.5%) with Aero-
xide
P25 specifications (i.e. a c.a. 21 nm particle size and a
35e65 m
2
g
1
surface area) was used, as supplied by Sigma-
Aldrich. Milli-Q water (Millipore, Blilerica, MA, USA) of
18.2 UM cm was used to prepare all aqueous solutions.
Nitrogen gas (99.9% purity) used for drying SPE eluate was
obtained from Linde HKO Limited (Hong Kong).
Reagents for Microtox bioassays, including lyophilized bac-
teria, reconstitution solution, diluent (2% NaCl), and osmotic
adjusting solution (22% NaCl), were purchased from SDIX
(Newark, DE, USA). 100 mLL
1
phenol used as the method con-
trol was obtained from Thermo Scientific (Waltham, MA, USA).
2.2. Sample locations and collection
Seawater samples were collected from thirteen locations
along the coast of Hong Kong (Fig. 1) in order to produce a
comprehensive distribution profile. The city of Hong Kong is
located on the Southeastern coast of China towards the
Northwest part of the South China Sea. Because Hong Kong is
situated in the Pearl River Delta, the western coasts of Hong
Kong Island and Kowloon (i.e. River Trade Terminal, Tuen
Mun, Tsing Lung Tau) receive a substantial discharge from the
estuary while coastal waters on the East side (i.e. Kiu Tsui, Po
Doi O, Tit Cham Chau) remain relatively pristine. Sampling
locations for the current study were established on both sides
of and in the center of Victoria Harbor. Special attention was
paid to areas with high levels of urbanization (e.g. North Point
and Tsuen Wan) and particularly those associated with sub-
marine wastewater treatment outfall (e.g. Yau Tong, Tsim Sha
Tsui, Harbor Tunnel, and Stonecutters Island).
Samples were collected from all sites on 2011 July 22nd
(peak of the hot, wet summer) and 2012 January 17th (peak of
the cold, dry winter) in order to capture seasonal differences.
Hong Kong Observatory data showed an average temperature
in 2011 July of 31.4 C and a mean humidity of 81% with var-
iable rainfall totaling 226.8 mm. In 2012 January, the Obser-
vatory recorded a monthly average of 17 C with 82% humidity
and 42.1 mm total rainfall.
Surface seawater was sampled at 2-m depth using a 3-L
Van Dorn sampler and stored in opaque high-density poly-
ethylene bottles. Collected samples were filtered through
0.45 mm membrane and subject to immediate analysis. All
samples were frozen and stored at 20 C until use.
2.3. Laboratory experiments
2.3.1. Quantitative analysis using solid-phase extraction
An individual standard addition calibration curve was estab-
lished for each seawater sample by spiking the standard
mixture into 100 mL. Sample clean-up and pre-concentration
used Strata-X 33 mm Polymeric Reversed Phase SPE cartridges
water research 52 (2014) 260e274262
(200 mg/3 mL, Phenomenex, CA, USA). SPE procedures adopted
the modified extraction method previously used for an LC-MS/
MS-based study of seven artificial sweeteners (Scheurer et al.,
2009), which included the sweeteners targeted in the present
study. After conditioning the cartridge with 3 3 mL water
followed by 3 3 mL methanol, 100 mL seawater spiked ac-
cording to standard addition calibration was loaded onto the
column at a flow rate of 1 mL min
1
. Subsequent to 10-min
vacuum drying of the SPE sorbent, the retained analytes were
wet for another 10 min,and then eluted with 3 aliquots of 3 mL
methanol. The eluted enriched sample was then subject to ro-
tary evaporation (BU
¨CHI Rotavapor RII, Flawil, Switzerland) at
ca. 40 C, reduced to a volume of ca. 1 mL and brought to com-
plete dryness by nitrogen. Finally, samples were reconstituted
with precisely1 mL water then syringe-filteredthrough 0.22-mm
nylon (Grace, IL, USA) for UPLCeMS/MS analysis.
The performance of the modified extraction method was
evaluated in order to establish valid analytical data for the
seawater study. A five-point standard addition calibration
curve was prepared by extracting a series of analyte-free
samples spiked with different levels of sweetener mixtures
at 0, 0.05, 0.10, 0.50, 1.00 mgL
1
. The method obtained good
linearity with correlation coefficients greater than 0.9950.
Extraction efficiency was also evaluated by first fortifying
blank seawater samples with variable levels from 0.06 to
1mgL
1
for standard addition procedures. This gave satisfac-
tory analyte recoveries in a range between 73.5 and 128.4% for
all four target compounds. The standard addition method
exhibited high repeatability with the relative standard de-
viations (R.S.D.) <9.2% for low (0.01 mgL
1
), mid (0.10 mgL
1
)
and high (0.50 mgL
1
) levels when five replicates were tested.
Limit of detection (LOD) and limit of quantification (LOQ),
defined as three and ten times of the signal-to-noise ratio of
the SIM analyte peak were 0.006 and 0.01 mgL
1,
respectively,
for this work. These results indicated similar sensitivity
among four sweeteners.
2.3.2. Liquid chromatography tandem mass spectrometry
An Acquity Ultra Performance LC system (Waters, MA., USA)
was employed for all photo-degradation and removal studies
of artificial sweeteners. It consisted of an auto-injector with a
10-mL sample loop and a temperature controlled column
compartment. An injection volume of 5 mL was used for all
samples. Chromatographic analysis for four sweeteners
extracted from seawater was performed with an Acquity BEH
C8 column (2.1 i.d. 100 mm, 1.7 mm, Waters) using an in-
jection volume of 5 mL. Mobile phases were: (A) 20 mM
ammonium acetate in water and (B) methanol. The gradient
profile started with 98% mobile phase A which decreased in
14 min to 25%; after holding for 1 min, mobile phase compo-
sition returned to its original 98% in 1 min to re-equilibrate for
6 min. A flow rate of 0.2 mL min
1
was maintained throughout
the 22 min programme.
The UPLC system was hyphenated to a Quattro Ultima triple
quadrupole mass spectrometer (Waters, MA., USA) with an
electrospray interface operated in negative ionization mode at
2.5 kV capillary voltage. The source and desolvation tem-
peratures were set at 150 C and 300 C, respectively. Nitrogen
was used with a cone gas flow of 150 L hr
1
for nebulization and
of 600 L hr
1
for desolvation. The quantitative analysis was
performed in time-scheduled Selected Ion Monitoring (SIM).
The corresponding [M H]
molecular ions for acesulfame,
Fig. 1 eMap of Hong Kong showing sampling locations.
water research 52 (2014) 260e274 263
cyclamate, saccharine and sucralose were monitored at
respective mass-to-charge (m/z) ratios of 162, 178, 182, and 395,
respectively. Instrument control, data acquisition and pro-
cessing were performed using MassLynx software.
Two sets of instrumental conditions using different col-
umns and elution conditions were specifically applied to
retrieve photocatalytic degradation profiles of acesulfame and
sucralose. To analyze irradiated samples of sucralose, an
Acquity BEH C18 (2.1 i.d. 50 mm 1.7 mm, Waters) column was
used with a C18 guard column. Deionized water and methanol
were applied as mobile phases A and B, respectively. No buffer
salt was added to minimize adduct formation that interferes
detection and elucidation of the photodegradation products.
Separation was afforded under constant 0.4 mL min
1
flow
rate with a 18-min gradient programme: 98% mobile phase A
was initially run for 1 min before linearly decreasing to 10%
over 10 min, after which it was held constant for another min.
Then the composition was rapidly ramped back up to 98% and
equilibrated for 5 min.
To separate acesulfame and its irradiation by-products, an
Acquity BEH C8 columnwas used packed with (2.1 i.d. 100 mm
1.7 mm, Waters)with an inlet filter. At a flow rate of 0.3 mL min
1
,
the same mobile phases made up of deionized water (A) and
methanol (B) were used for a 13-min gradient: mobile phase A
initially ran for 1 min, was then decreased to 80% over 0.5 min,
then further brought down to 5% over 5.5 min; after 1-min at
that level, the composition was stepped back up to 98% mobile
phase A over another min was and then finally equilibrated.
2.3.3. Photodegradation profiling
To evaluate the degradation toxicity of sucralose and acesul-
fame, these two sweeteners were firstly photo-degraded for
toxicity screening.
Photocatalytic reaction was carried out in a lab-made
irradiation chamber of dimensions 32 (W) 32 (D) 36 (H)
cm
3
and equipped with a concentric series of sixteen UV
germicidal lamps (Sankyo G8T5, Japan) emitting at 253.7 nm
wavelength and 128 W total power. A calibrated time switch
was also accessorized for precise time control of the irradia-
tion experiment.
To prepare photocatalytic reaction, accurately weighed
TiO
2
was suspended in a 25-mL sweetener solution with a
sweetener:catalyst ratio ¼1:20 (m/m) in a 50 mL Pyrex conical
flask. Two initial concentrations of 400 mg L
1
(with 200 mg
TiO
2
) and 20 mg L
1
(with 10 mg TiO
2
) each sweetener were
observed, looking for any dependence of pathway on the ki-
netic parameter. At specific time intervals after initiation of
UV irradiation, 1-mL aliquots of the suspension were with-
drawn, filtered through 0.22 mm nylon and analyzed.
2.3.4. Bioluminescent Microtox test
The phototoxicity of sucralose and acesulfame was evaluated
by conducting acute toxicity screening in irradiated samples.
According to the obtained degradation profiles, degraded
sweeteners were collected separately at the precise irradiation
time of peak breakdown. Collected samples were concen-
trated and lyophilized with a freeze-dryer (Ilshin, Model FD
5512, Netherlands) to a dry powder. The combined mass of
degradates mixture was accurately weighed and reconstituted
with Milli-Q water for subsequent toxicity testing.
The acute toxicity bioassay was conducted using a Micro-
tox Model 500 analyzer (Modern Water, Guildford, UK) with
the marine bacterium Vibrio fischeri as a bioluminescent indi-
cator. The toxicant-induced inhibition of luminescence of V.
fischeri was measured by recording the decay after 5 and
15 min of exposure to a serial dilution of the sweetener’s
degradates and comparing to a blank control, which con-
tained the same 2% sodium chloride reagent as other samples
only without the addition of degradates. The effective con-
centration (EC
50
) at which the irradiated sweetener caused
50% inhibition was calculated using MicrotoxOmni software.
Parallel tests were performed in phenol as reference toxicant
for positive control with an EC
50
criterion set in a typical
13e26 mg L
1
range for quality assurance.
2.3.5. Removal treatment
The experimental arrangements (including irradiation pa-
rameters, catalyst composition and hardware set-up) of the
removal treatment for sucralose and acesulfame were set up
to mirror the photocatalytic conditions described in Section
2.3.3. Initial levels at 1 and 5 mg L
1
were respectively taken to
evaluate removal efficiency of sucralose and acesulfame.
Removal progress was monitored at time intervals by with-
drawing 1-mL aliquot to UPLCeMS/MS analysis. A total irra-
diation time of 300 min was adopted as that time period had
proven sufficient for complete degradation of both
sweeteners.
3. Results and discussions
3.1. Environmental occurrence
3.1.1. Regional overview
The present work in Hong Kong revealed a median level of
0.06 mgL
1
for sucralose. This falls in the lower end of the
general European range. In comparison, acesulfame with an
average of 0.22 mgL
1
was the significant contributor to the
local artificial sweetener contamination in Hong Kong waters.
These results match the occurrence worldwide: In the surface
and underground water environments in Spain, acesulfame
was the most prevalent sweetener detected (53.7 mgL
1
,
Ordonez et al., 2012), Canada (33.6 mgL
1
,Van Stempvoort
et al., 2011), Switzerland (7.35 mgL
1
,Berset and Ochsenbein,
2012), China (4.65 mgL
1
,Gan et al., 2013), and Germany
(2.00 mgL
1
,Scheurer et al., 2009). After acesulfame, the
highest contaminations were also found in Spain where
profiling studies uncovered the highest figures between 5.3
and 19.7 mgL
1
in surface waters, following the descending
order of saccharin >cyclamate >sucralose. The same trend
was found in Switzerland and presently in Hong Kong (Berset
and Ochsenbein, 2012; Buerge et al., 2009). Aside from this
similarity, there were great discrepancies in contaminant
levels among these countries. For examples, total sweetener
contents in a Spanish river were 24e141 times higher than
that in the open harbor of Hong Kong. This is possibly, if not
likely due to direct effluent discharge from wastewater treat-
ment plant and untreated sewage overflow to the receiving
inland watercourse in the Spanish river (Berset and
Ochsenbein, 2012). On the other hand, Swiss lakes showed a
water research 52 (2014) 260e274264
good match with our oceanic occurrence at sub-ppb concen-
trations. They represent a relatively larger water body and
environmental capacity, plus perhaps greater control of
sewage effluent as well as less use of these sweeteners in the
region. Therefore, both the natural environmental features as
well as man-made factors could have contributed to the dis-
tribution, accumulation and fate of the contaminants within
an individual aquatic compartment. The occurrence and
behavior of sweeteners in the marine environment are influ-
enced by complex oceanographic mechanisms (e.g. tidal ac-
tion, oceanic current) that remain uncharacterized and barely
explored. The first marine study of sucralose was carried out
in North American waters, along the southeastern coast of the
U.S., where a wide range of the sweetener was observed (Mead
et al., 2009). The wide concentration range of sucralose
spanning from ppt to sub-ppb level along the southeastern
coast of US was different from a later observation in China.
Another small scale marine study has been lately reported in
Bohai Bay, a polluted gulf in northern China (Gan et al., 2013).
Profiles of sweeteners in the seawaters of Bohai Bay and Hong
Kong share some similarities in terms of ppb levels and the
relative distribution of (highest to lowest)
acesulfame >saccharin >cyclamate >sucralose. While most
artificial sweeteners were found at the same levels in both
waters, acesulfame levels in the northeastern bay area was
found to be at least 10 times more than in Hong Kong. Unique
characteristics of different maritime environments suggest
even more subtle influences as a combination of usage pref-
erence, seasonal effects and local neighboring geographic
features are in force.
3.1.2. Occurrence in Hong Kong surface seawater
Four artificial sweeteners were analyzed in Hong Kong sub-
tropical waters. Conforming to the natural coastline, seawater
samples were obtained in thirteen designated locations to
retrieve a continuous profile from the Eastern side (Sai Kung)
to the West (Tuen Mun) through Victoria Harbor channel
(Fig. 1). The seasonal variations on sweetener distribution
were evaluated in summer wet (September) and winter dry
(January) periods to compare relative significance of various
seasonal disturbances (e.g. influence from vicinal estuarine
discharge, monsoon, typhoon, oceanic current, etc.). A sum-
mary of the occurrence of the four sweeteners studied, with
seasonal and geographic distribution, is illustrated in Fig. 2.
Season appears to have a significant influence in redistribut-
ing artificial sweeteners in Victoria Harbor. Nevertheless,
comparison of the concentration levels of the four sweeteners
in all show a consistent relative occurrence year round, in this
order (highest >lowest): acesulfame (0.34 mgL
1
)>saccharin
(0.25 mgL
1
)>cyclamate (0.23 mgL
1
)>sucralose (0.2 mgL
1
).
Unlike most reports from outside Hong Kong, both the
detection rate and occurrence value of sucralose were the
lowest among all sweeteners studied. This observation could
be an effect due to local usage preference. Because Hong Kong
relies heavily on importation for its food supply, sweetener
consumption in Hong Kong has close links with the interna-
tional safety standards enforced in the production countries
(e.g. European Union, Joint FAO/WHO Expert Committee on
Food Additives, etc.) as well as Chinese legislation. According
to GB2760-2001 Standards, sucralose is assigned the lowest
maximum usage in various food types such as processed
fruits, cold drinks, jelly, sauces, and candies, while up to 5
times more are allowed for three other sweeteners (MOH,
2011). Similarly, the lowest permitted level of usage was also
established for sucralose by the Scientific Committee on Food
of the European Union (Mortensen, 2006). Nonetheless, the
relative environmental occurrence of acesulfame, saccharin
and cyclamate poorly matched with the usage pattern as
allowed by these standards. This suggests the presence of
more complex mechanisms governing the distribution of
these compounds.
Based on evaluation of seasonal averages, a series of win-
ter:summer (W:S) concentration ratios was obtained for
cyclamate (W:S ¼2.5), saccharin (W:S ¼1), acesulfame
(W:S ¼0.72), and sucralose (W:S ¼0.2). Even closer compari-
son in each sampling location indicates that cyclamate and
saccharin exhibit higher contamination rates during winter,
mostly on the western side of the Harbor (Figs. 2b and c). In
contrast, consistently lower acesulfame and sucralose levels
were found in January as shown in Fig. 2a and d. Various
biochemical factors combined with the hydrodynamic in-
fluences of monsoonal winds as well as nearby Pearl River
outflow were likely to be the cause for these observed
changes.
Located at the southeast of the Pearl River (PR) Delta, Hong
Kong is positioned to be strongly influenced by Southeastern
China oceanic waters and the freshwater input from PR es-
tuary, as well as local sewage outfall. The interactions of these
three factors vary seasonally and spatially. During summer,
the rainy season delivers a maximal PR estuarine freshwater
discharge, accounting for 80% of its annual outflow, to the
West of Hong Kong (Zhao, 1990). This estuarine plume is
further driven eastward by prevailing southern/southwestern
monsoons to the inner coastal area and even the southern
waters. The large volume of neighboring freshwater input
constitutes a substantial effect of dilution to local concentra-
tion levels of sweeteners, especially for cyclamate and
saccharin (Atoui et al., 2013).
It is also worth noting that parallel contribution from bio-
logical degradation may also lead to lower distribution of
these two sweeteners (Fig. 2b and c). High nutrient injection
events due to PR heavy pollution load (Ho et al., 2008) together
with elevated water temperature consequently result in
favorable growth of bacteria and microbes. Saccharin and
cyclamate, being established as readily biodegradable in
highly eutrophic environments (Schuerer et al., 2010), has
likely experienced accelerated degradation during the sum-
mer months ea natural attenuation process which was
commonly followed by other emerging contaminants,
including endocrine disrupting compounds, pharmaceuticals
and personal care products (Yu et al., 2013).
Despite sharing the same dietary application, some artifi-
cial sweeteners exhibit entirely different profiles from other
sweeteners. There was a prevalence of acesulfame and
sucralose sweeteners in summer water samples that showed
no sign of being affected by the natural season-mediated
removal processes. Both compounds have been proven
resistant to breakdown and able to sustain much longer life-
times under microbiologically active conditions (Robertson
et al., 2013). As a result of their environmental recalcitrant
water research 52 (2014) 260e274 265
nature, seasonal monsoons driving a regional hydrodynamic
became a predominant force in the characteristic distribution
of acesulfame and sucralose. High summer occurrence in the
areas sitting west of Stonecutters Island reflects a certain level
of influence by the massive PR pollutant discharge (Cai et al.,
2004). In addition, under south/southwest monsoonal ef-
fects, deep continental shelf water transporting landward
induces in the coastal area an upwelling water current that
spreads submarine-discharged wastewater to the surface
(Zhou et al., 2012). Evidently this influence is at work in Vic-
toria Harbor because higher acesulfame levels were recorded,
along the Victoria Harbor channel (from Yau Tong to Stone-
cutters Island) where submarine sewage outfalls are inten-
sively located, whereas waters in the Eastern bays (Kiu Tsui
and Po Doi O) remained relatively pristine (Figs. 2a and d). The
hydrological trajectory was reversed in winter when north-
easterly winds promote a downwelling current that protects
the submarine effluent preserves in the Harbor channel. As a
Fig. 2 eOccurrence levels of four artificial sweeteners, (a) sucralose, (b) saccharin, (c) cyclamate and (d) acesulfame, in 13
locations of the Hong Kong marine environment during wet summer (July) and dry winter (January) seasons.
water research 52 (2014) 260e274266
result of seasonal hydrodynamics, acesulfame and sucralose
sustain low W:S ratios, in contrast with saccharin and cycla-
mate sweeteners, which are more susceptible to microbial
degradation.
The present findings effectively demonstrate the
complexity of factors that govern the behaviors and fate of
artificial sweeteners in the environment. In particular, distri-
bution results of acesulfame, the highest occurring sweetener
of all, indicate a severe lack of efficiency in current domestic
wastewater treatment for removing artificial sweeteners from
wastewater. If this is true for other emerging contaminants,
then the environmental impact of these chemicals is poten-
tially and alarmingly broad. The same environmental specu-
lation is true for sucralose, which is frequently found in some
European inland waters. In order to define more precisely
what happens to these pollutants in the environment, a series
of degradation and ecotoxicological studies were performed
focusing on the most persistent of the artificial sweeteners,
acesulfame and sucralose.
3.2. Degradation toxicity
Chemically designed to give no nutritional value and minimal
respiratory metabolic output, artificial sweeteners are highly
stable and have been shown resistant to breakdown during
sewage treatment. In fact, extensive environmental occur-
rence data worldwide has implied that the influence of their
immutability on their distribution and fate in the environ-
ment is monumental. Of the four artificial sweeteners, ace-
sulfame and sucralose have been consistently detected in
untreated wastewaters and treated effluent plume (Schuerer
et al., 2010; Robertson et al., 2013), signifying a strong recal-
citrance of the sweeteners and a severe lack of removal effi-
ciency through conventional wastewater treatment
processes. As they are discharged, these high stability
sweeteners continue to persist in the environment with half-
life of up to several years (Lubick, 2008), such that acesulfame
was still detected in a wastewater plume after a residence
time of about 1.5 years (Scheurer et al., 2009). For this reason,
ironically, both acesulfame and sucralose have been valued as
potential chemical markers for tracing domestic wastewater
in the aquatic environment (Buerge et al., 2009; Van
Stempvoort et al., 2011; Oppenheimer et al., 2011). With this
level of persistence together with their tendency to disperse,
these compounds are likely to affect a much larger area within
a body of water as they accumulate over time. Yet, there is
little scientific data revealing the fate and chronic effect of
artificial sweeteners on the environment. Photo-
enhancement of toxicity due to exposure to natural sunlight
has been widely documented in PAHs, antibiotics and other
pharmaceutical products (Petersen and Dahllof, 2007; Jung
et al., 2008; Dantas et al., 2010; Klamerth et al., 2010). This
evidence implies that the natural irradiation is an important
factor in determining the actual consequences of unregulated
persistent organic pollutants. In this part of study, photo-
catalytic simulation experiments followed by initial toxicity
tests were performed to assess the long term hazard of photo-
induced degradation of the persistent sweeteners acesulfame
and sucralose.
The present study was designed to simulate the photo-
induced degradation that occurs when sweeteners undergo
prolonged exposure to UV irradiation. To monitor the long
reaction process in much shorter timeframe, irradiation ex-
periments were assisted by heterogenous TiO
2
catalyst which
has been commonly applied in the phototoxicity study of
other organic compounds (Sousa et al., 2012; Selli et al., 2008).
Especially, for such applications as the removal and degra-
dation of estrone, carbofuran and methyl orange micro-
pollutants, the Aeroxide P25 with well-defined specifications
(i.e. a 21 nm particle size, an extensive 50 m
2
g
1
surface area
and a rutile:anatase phase ratio about 3:2) is earning
increasing respect as a prototype catalyst to deliver excep-
tional photosensitized oxidative efficiency over other larger
sized rutile and anatase composites (Fenoll et al., 2013; Han
et al., 2012; Da Dalt et al., 2013). Formation of superoxide
and hydroxyl radicals at the wateresemiconductor interface
promotes the degradation reaction of sweeteners as a conse-
quence of light absorption by TiO
2
(Hoffman et al., 1995). The
transformation profiles of the parent compound as well as the
product ions were followed real-time by UPLCeESIeMS anal-
ysis. In order to further verify any behavioral variation specific
to the starting concentration of the original compound, we
have also performed parallel experiments applying initial
levels of 20 and 400 mg L
1
to observe possible change in re-
action pathways. Irradiated samples were subsequently
collected at a defined timestamp to quantify photo-enhanced
toxicity of both sweeteners with special focus on their
ecological impact to marine environment. To this end, marine
bacteria V. fischeri was adopted in this work based on the
Microtox acute toxicity test standardized by ISO 11348-3 (ISO,
2007). By measuring the respiratory inhibition of the biolu-
minescent bacteria induced by the original/irradiated sam-
ples, absolute EC
50
(half maximal effective concentration)
values were obtained for toxicity comparison. To the best of
our knowledge, a comprehensive comparison between the
photo-degradation toxicity of sucralose and acesulfame has
yet been investigated.
3.2.1. Sucralose
Sucralose decomposition assisted by photocatalysis and
thermal treatment has been previously studied. Subjecting
sucralose to dry heat reportedly led to glycosidic bond cleav-
age by a different pathway and at a different dissociation
position than its non-chlorinated analog sucrose (Rahn and
Yaylayan, 2010). In the presence of two electronegative chlo-
rine atoms, degradation of sucralose releases a fructose moi-
ety whereas degradation of sucrose releases a fructofuranosyl
cation. Calza et al. have explained these degradation results
by an alternative route, namely by the photo-Fenton reaction
of sucralose. However, unlike in thermal decomposition, the
position of glycosidic bond breakage mediated by homoge-
neously catalyzed reaction remained inconclusive. These re-
sults strongly suggest that molecular transformation depends
on the mechanism and reaction conditions (e.g. presence of
water and reactive radical species) inducing it.
In the current study, we have recreated a realistic simula-
tion of sucralose’s transformation simply by applying physical
stimulants, namely photo irradiation and a chemically inert
photocatalyst. MS analysis using negative ESI mode detected
water research 52 (2014) 260e274 267
four UV/TiO
2
photocatalytic product ions with m/zratios of 216,
411, 447, and 457 as shown in Fig. 3. A major degradation in-
termediate with a molecular ion at m/z216 could be tentatively
identified as a fructose moiety, namely 1,6-dichloro-1,6-
dideoxyfructose of an empirical formula C
6
H
10
O
4
Cl
2
. This
observation was closely consistent with that in thermal
degradation, which is that bond cleavage occurs adjacent to the
six-membered glycoside. If true, this confirms an important
detachment detail previously not reported in the photo-Fenton
study. Furthermore, the inspection of the base peak at m/z411
revealed the possibility of monohydroxylation to the parent
molecule in our photo-oxidative reaction, indicating a new
transformation pathway which would be absent in dry pyrol-
ysis treatment. Our results also provide evidence showing the
presence of an even more complex reaction that may involve
oxidation and recombination, ultimately leading to the previ-
ously unreported formation of ions with higher molecular
masses.
Real-time monitoring during our photocatalytic experi-
ment revealed that all four intermediates emerged upon
initiation of UV irradiation. Due to the lack of reference
chemicals, it was not possible to perform absolute quantifi-
cation; we could only compare relative abundance among the
parent and product ions. The similarity of the abundance
trends for the four product peaks as shown in Fig. 3b suggests
that they were all primary degradation ions for which all
transformation reactions took place at comparable rates. The
product ions were rapidly formed while degradation of
sucralose was observed roughly following a first-order decay
(Fig. 3a). Accumulation of these ions peaked at approximately
one-third of the time it took for complete sucralose decom-
position. After that point, all products decayed rapidly,
reaching flat-line shortly after sucralose. Comparison of the
degradation profiles between 20 and 400 mg L
1
revealed the
same transformation trend: the intermediate ions were short-
lived and were degraded twice as fast as the parent com-
pounds. An identical set of degradation ions was also present
in similar proportions regardless of initial sweetener con-
centrations of irradiation experiment; therefore the higher
concentration (i.e. 400 mg L
1
) was selected for downstream
phototoxicity evaluation of sucralose.
In order to measure the environmental impact consequent
to photo-degradation, Microtox experiments were carried out
on irradiated samples with the maximum concentration of
the intermediate ions that would best reflect the phototoxicity
of sucralose. Irradiated samples were collected, lyophilized,
accurately weighed to obtain a set of absolute EC
50
values for
valid comparison. Based on the evaluation of bioluminescent
inhibition in V. fischeri, acute toxicity of sucralose has been
significantly enhanced from 2670 mg L
1
to 156.2 mg L
1
due
to the production of by-products. Toxic contribution by TiO
2
catalyst was considered, and the test therefore included TiO
2
in a sweetener-free negative control sample which showed no
inhibition effect. This leads us to believe that the formation of
photo-degradates were solely responsible for the observed
toxicity enhancement. Although it could be argued that both
sucralose and by-products with experimental EC
50
values
>100 mg L
1
belong to the same category of “not harmful” as
defined by the European Union with regard to marine organ-
isms (European Commission, 1996), such magnification power
in photo reaction pushes irradiated sucralose close to the
“harmful” (of a criterion of 10e100 mg L
1
) category.
3.2.2. Acesulfame
Like sucralose, the extent of understanding the degradation of
acesulfame is currently limited to food chemistry. In acidic
environments, acesulfame undergoes hydrolytic decomposi-
tion to form acetoacetamide and acetoacetamide-N-sulfonic
Fig. 3 ePhotocatalytic degradation profiles of (a) sucralose at 20 mg L
L1
and (b) its four product ions monitored in real time.
water research 52 (2014) 260e274268
acid (WHO Food Additive Series 16). Transformation can also be
achieved by ozone-based chemical treatment in wastewater.
In the presence of ozone, ring cleavage at the carbon double
bond in the six-member acesulfame ring has been reported;
the reaction involves an aldehyde hydrate as a major ozonation
intermediate and produces acidic by-products (e.g. acetic acid,
formic acid and oxalic acid) (Scheurer et al., 2012). Yet, data was
insufficient to allow reasonable depiction for its natural
degradation fate of acesulfame in environmental waters.
Upon catalytic UV irradiation of acesulfame in our study,
we were able to identify a total of twelve base ion peaks based
on UPLCeESIeMS MRM analysis (Fig. 4). According to their
kinetic behavior during the experiments these ions could be
categorized into two groups with distinctive transformation
profiles. Group I, which comprised four ions with base peaks
at m/z154, 180, 194 and 212 (Fig. 4b), was characterized by
rapid transformation: highest abundance was recorded at
around the half-life of acesulfame parent ion. Among these
four product ions, the presence of acetoacetamide-N-sulfonic
acid (m/z180) due to ring breakage could be tentatively iden-
tified. Group II comprised another group of molecular ions
with m/z82, 96, 136, 138, 215, 231 and 328 (Fig. 4c) with much
slower reaction response. The appearance of m/z96 was
consistent with aminosulfonic acid, an unreported hydrolytic
fragment of the broken ring structure. During the real-time
irradiation study, Group II ions reacted so slowly that we
were unable to observe an exhaustion point within a span of
time. Although the depletion rate of Group I ions was at least
twice that of Group II ions, there was not sufficient evidence to
implicate a sequential relationship between the groups. In
contrast to the sweetener sucralose, photo-degradation of
acesulfame was shown to involve more complex trans-
formation processes which ultimately prolonged its persis-
tence in form of its degradation products. Evaluation of the
risk led by these degradation residues becomes meaningful in
defining its long-term environmental safety.
The original acesulfame compound as a legally allowed
food additive was proven minimally toxic in a variety of
aquatic biological models, including zebra fish, golden orfe,
Daphnia magna, etc. The same conclusion could be drawn from
our Microtox screening with an EC
50
value of 72,190 mg L
1
.
Primary findings of current screening test also showed
Fig. 4 ePhotocatalytic degradation profiles of (a) acesulfame at 20 mg L
L1
, of which twelve products ion were monitored in
real time in (b) Group I and (c) Group II.
water research 52 (2014) 260e274 269
significantly elevated toxicity in irradiated samples. Due to
photo treatment, the acute inhibition effect in V. fischeri was
significantly amplified to EC
50
¼125.5 mg L
1
, at a measure-
able magnification factor of 575. Irradiated samples contain-
ing the maximum amount of degradation by-products were
collected and tested after 19 h of UV treatment. At this point
where >99.5% degradation of acesulfame was accounted for,
the measured toxicity enhancement has to be attributed to
the accumulation of its degradates. This number also repre-
sents an unusual finding when compared to other persistent
organic pollutants in which the enhancement factor is rarely
over 20 times. At such level of elevation, several pharmaceu-
tical products and pesticides have already received much
attention as pollutants in water environments (Jung et al.,
2008; Dantas et al., 2010). Data herein reveals even high
toxicity elevation in artificial sweeteners that signifies po-
tential for ecological impact which has been overlooked. The
combination of degradation and toxicity data clearly demon-
strates that during prolonged irradiation, stable compounds
much more toxic than the parent acesulfame were formed,
adding a new dimension to the issue concerning the fate of
artificial sweeteners. The traditional perspective which fo-
cuses on their basic occurrence becomes inadequate to pro-
vide formal assessment in environmental risk while scientific
data strongly mandates more considerations throughout
transformation stages of the molecule.
In our studies, sucralose has decayed relatively slowly
compared to acesulfame, meaning a stronger resistance in
sucralose to degradative stimuli. Worldwide occurrence
studies which documented higher detection frequency lead us
to agree with the notion that sucralose potentially poses
greater hazard than other sweeteners. Now we have numer-
ical toxicity data and kinetic profiles that enable direct com-
parison between sucralose and acesulfame. New evidence
clearly shows that acesulfame, which degrades into more
persistent and toxic by-products in exposure to sunlight,
poses a longer-term ecological impact than sucralose. How
should wastewater treatment plants handle these potential
threats? One possible measure is to intercept artificial
sweeteners before they enter the environment. Ideally, an
attenuation mechanism that can effectively mineralize arti-
ficial sweeteners in domestic wastewater becomes urgently
needed.
3.3. Removal treatment
Worldwide environmental researchers consistently agree that
the stability of sucralose and acesulfame sweeteners and their
resistance to chemical and biological degradation can offer a
tracking tool for wastewater contamination in surface waters.
Sucralose and acesulfame have been reported to be inert to
partly degradable in conventional wastewater treatment
processes. Aerobic bioreactor showed no effects in degrading
both persistent sweeteners (Schuerer et al., 2010), while only
less than 2.5% consumption of sucralose was recorded in
anaerobic experiments (Torres et al., 2011). Similarly, chemi-
cal treatment with chlorine was proven not to work effectively
with the two targets. Under the most intensive chlorination at
100 mM for 96 h, sucralose removal up to 79% could be ach-
ieved, yet this level is difficult to realize in typical treatment
practice. Filtration by granular activated carbon (GAC) has
been more useful to decontamination other hydrophobic
organic pollutants, and to a limited extent, sucralose (Lange
et al., 2012). However, massive sorbent waste consequent to
the use of GAC would create even more complicated disposal
and regeneration issues.
Several advanced methods employing more powerful
mechanisms have been evaluated. Ozonation is able to ach-
ieve partial removal for acesulfame at a rate of 18e60% and
sucralose at 8e15% (Schuerer et al, 2010). Performance was
susceptive to slight changes in ozone contact times and
dosage. Further, its corrosive nature, high capital cost and
energy consumption have made ozonation a less ideal choice
for treatment. UV irradiation that is commonly used for water
disinfection, on the other hand, presents a much milder and
more economical option. UV irradiation, without any use of
catalytic reagent, has showed limited effect on both stable
sweeteners, with the best elimination rate recorded for ace-
sulfame as 35% (Torres et al., 2011; Soh et al., 2011). Based on
our observation in the current TiO
2
-assisted degradation
study, complete degradation under oxidative irradiation in-
dicates that mineralization for both compounds is highly
feasible in the presence of a photocatalyst. In fact, solar
application has been working in conjunction with different
oxidizers, including hydrogen peroxide, hypochlorite, chlo-
rine dioxide, as well as photo-sensitized TiO
2
and fenton re-
agents, in advanced oxidation processes which are
increasingly viewed as effective barriers against a wide range
of micropollutants (Sichel et al., 2011; Metz et al., 2011; Kurbus
et al., 2003; Bauer et al., 1999). In particular, heterogeneous
catalyst TiO
2
, which can be applied in immobilized or sus-
pension forms, offers unsurpassed advantages over other
solution-based reagents in terms of easier post-treatment
separation and regeneration (Bauer et al., 1999). Preliminar-
ily, the removal efficiency of sucralose in TiO
2
photocataysis
has been evaluated in a tentative study (Calza et al., 2013).
Experimental work carried out in long-wavelength (360 nm
UVA range) condition with a low-power lamp (40 W) has
reportedly achieved mineralization after a 4-hr irradiation
period. Results without considering parallel germicidal
application in wastewater treatment could only provide a
realistic depiction to a limited extent.
In this work, the efficiency and application feasibility of UV/
TiO
2
photocatalytic treatment targeting sucralose and acesul-
fame were reviewed at environmentally and technically rele-
vant levels. Real-time removal tests were carried out under the
same physical condition as the photodegradation study:
sweetener samples were irradiated by 128 W germicidal UVC
lamps at 254 nm in a TiO
2
suspension (sweetener:catalyst
ratio ¼1:20 m/m). As shown in Fig. 5a, a solution of acesulfame
of catalytic radiation was able to achieve at least 99.9% removal
within 45 min, and up to 100% with an extra 15 min. For
sucralose, due to its stronger persistence, a longer time was
required to reach complete degradation: it took 120 min to
remove 99.8% (Fig. 5b). With the assistance of TiO
2
, this pho-
tocatalytic approach presents a more powerful barrier to
sweetener micropollutants than irradiation treatment that is
primarily intended for disinfection purposes. In fact, the use of
a catalyst and its applied ratio have shown a kinetic influence
on the production of reactive hydroxyl radical species, and
water research 52 (2014) 260e274270
thus the rate of degradation reaction. To initially compare the
dose effect of catalyst, a preliminary experiment was per-
formed using a lower a ratio at 1:10 which resulted in a signif-
icant delay of total acesulfame depletion beyond 120 min.
Knowing that acesulfame is a less stable compound than
sucralose, a much longer time was expected in sucralose to
completely mineralize at low catalytic dose. To better model a
real-life treatment application, future development of photo-
sensitized reaction is recommended to employ a catalyst dose
equivalent to or more than 1:20, at which a 20e30-min contact
time is sufficient to fulfill a minimal quality criterion at 84e97%
removal for both sweeteners. In comparison with other
oxidative techniques such as ozone that only give <80%
removal at similar time allowance, this level of performance of
UV/TiO
2
is superior in delivering a good efficiency in a
reasonable period of time.
Although TiO
2
-based photodegradation batch experiments
has demonstrated good application feasibility for enhancing
removal of sweeteners, a vast majority of studies focusing on
more general organic pollutants have indicated major con-
straints in adapting bench-top systems to the municipal
treatment scale. One is the challenge in providing a cost-
effective solideliquid separation for slurry-type photo-
catalytic reactors (Chong et al., 2010). When considering the
practicality of photocatalytic water treatment technology, an
effective process preferably takes place in a continuous-flow
configuration with the catalyst immobilized or functional-
ized onto a fixed inert carrier; some recent examples are
aluminum foil, glass and quartz fibers, ceramic tiles, glass
slides, luffa sponge, etc (Botia et al., 2012; Carbonaro et al.,
2013; Devipriya et al., 2012). In lab-scale, immobilization of
TiO
2
on transparent quartz fiber reportedly has permitted a
better transmittance of UV through the total reactor volume
when compared to a suspension system (Bauer et al., 1999).
For the current photocatalytic removal technique to be scaled
up to a field-scale treatment unit targeting persistent sweet-
eners, a flow-through simulation reactor must be built to work
out the critical parameters controlling the photo-
mineralization kinetics specifically for sweetener pollutants,
such as catalyst loading, surface morphology, light photon
distribution, mass transfer on photosensitized surface, etc. In
addition, considering that the active reaction unit in real
operation must continuously accommodate a fresh stream of
influent water, a continuous-flow system instead of a circu-
lated batch reactor should be evaluated for deterioration, in-
hibition and deactivation of photocatalyst (Carbonaro et al.,
2013). Note that lab-simulation data based on pure standard
require further assessment and verification in real waters (i.e.
raw water, wastewater). Naturally present organic matters
and other water constituents can be potential interferences to
photodegradation (Yang and Ying, 2013). In order to potentiate
an effective treatment, the impact of water matrices influ-
encing positively or negatively on the removal efficiency of
artificial sweeteners must be characterized. For a field appli-
cation to sustain heavy-duty continuous treatment that meets
the emerging standards in micropollutant control, future
research will be directed towards polishing regenerative
strategies, extending catalyst longevity, and promoting a
better solar collecting technology in alternative solar-driven
process for a more cost-effective operation.
4. Conclusion
Refractory organic micropollutants such as artificial sweet-
eners have been regarded as viable tracers of
anthropogenically-sourced wastewater in inland surface wa-
ters. Now the persistence and ubiquitous presence of these
chemicals is raising increasing concern because little is
known about their behavior and fate. In the present study, a
comprehensive seasonal profile in an open marine environ-
ment was established for acesulfame, cyclamate, saccharin
and sucralose. Connection between their intrinsic molecular
recalcitrance and more complex periodical influence on
estuarine/oceanic hydrology was established. Of these four,
further work focused on acesulfame and sucralose because
occurrence data implicated much stronger persistence in
these compounds than others. Showing to sustain microbio-
logical activity during the summer, acesulfame and sucralose
were found to persist in areas linked to polluted estuarine
discharge and local wastewater driven by deep continental
shelf upwelling current. In our observation, both of these
chemicals are likely to impact the ecosystem of surrounding
larger water compartments and have long term conse-
quences. Our photodegradation study which tracked the
behavior of artificial sweetener over extended pollution pe-
riods clearly illustrated that both acesulfame and sucralose
were substantially degraded under simulated natural UV
conditions. Real-time observation revealed that the photo-
induced transformation of acesulfame leads to a collection
of more persistent by-products that are >500 times more toxic
Fig. 5 eRemoval efficiencies of the sweeteners (a) 5 mg L
L1
acesulfame and (b) 1 mg L
L1
sucralose under
photocatalytic conditions.
water research 52 (2014) 260e274 271
than the parent compound. Results implied that acesulfame is
significantly more phototoxic than other organic pollutants
reported thus far. For the first time, degradation toxicity of
acesulfame and sucralose is directly compared, and this
comparison refutes the general assumption that sucralose is a
more environmentally endangering sweetener species
because of higher persistence of the parent compound.
With the aim of finding a way to reduce the environmental
impact of these sweeteners and their associated degradation
products, our investigation was furthered on the mineraliza-
tion treatment in order to intercept their environmental
entrance. Photo-treatment integrates with advanced oxidation
by TiO
2
catalyst is here suggested as a feasible cost-effective
mechanism that is able to promote total mineralization
within 30 min. This level of performance is seen to exceed other
developing advanced oxidation treatments such as with ozone
and photo-Fenton reagent. While it has been typically applied
as a disinfection tool, UV irradiation working along with TiO
2
photocatalyst presents a powerful solution for the control of
persistent contaminants in effluents. This highlights the need
for an upgrade in the current irradiation processes to accom-
plish decontamination in addition to disinfection.
In this aspect, a technological development in artificial
sweetener attenuation should be focused on engineering the
suspension-type batch reactor into a fixed-bed field operation.
Continuous progress along this line will need to emphasize
not only on the elimination of original sweetener compounds
but also on the safety of photo-induced by-products. As new
evidence was disclosed in the present work about the poten-
tial phototoxic impacts of these compounds, more extensive
work is needed to accurately assess the biochemical and
ecotoxicological implications of photocatalytic treatment. In
this connection, priority research in the field of artificial
sweetener photo-degradation should be directed to the char-
acterization of toxicological end-points and identification of
the causative by-products by effect-directed analysis (EDA).
Meanwhile, it is equally important to make proactive efforts to
control the formation and accumulation of the degradation
by-products, by establishing appropriate treatment end-point
for persistent artificial sweeteners. On top of the removal of
precursor sweetener compounds, data of photo-induced
toxicity raise new concerns in the fate of degradation prod-
ucts in wastewater treatment and indicate the need to pursue
better removal strategies. More information on specific
biochemical properties such as bioaccumulation, biodegrad-
ability and retention behavior in adsorption treatment should
also be taken into account in fine-tuning a precise engineering
design. For better risk management, issues concerning the
inclusion of artificial sweeteners and ecotoxicologically rele-
vant photo-degradates products in quality indicators of
effluent require further discussion.
Acknowledgments
This work has been supported by a General Research Fund
(GRF) grant (Project No. HKBU 201210) from the Research
Grants Council of Hong Kong Special Administrative Region,
PR China. Z.Y.Sang and Y.N.Jiang gratefully acknowledge their
receipts of postgraduate studentship from the University
Grants Committee. We acknowledge Dr Patrick Y K Yue for his
technical assistance in the toxicity test.
references
Ahmed, F.E., Thomas, D.B., 1992. Assessment of the
carcinogenicity of the nonnutritive sweetener cyclamate. Crit.
Rev. Toxicol. 229 (2), 81e118.
Atoui, A., Hafez, H., Slim, K., 2013. Occurrence of toxic
cyanobacterial blooms for the first time in Laker Karaoun,
Lebanon. Water Environ. J. 27 (2), 42e49.
Bauer, R., Waldner, G., Fallmann, H., Hager, S., Klare, M.,
Krutzler, T., Malato, S., Maletzky, P., 1999. The photo-fenton
reaction and the TiO
2
/UV process for waste water treatment e
novel developments. Catal. Today 53 (1), 131e144.
Berset, J.-D., Ochsenbein, N., 2012. Stability considerations of
aspartame in the direct analysis of artificial sweeteners in
water samples using high-performance liquid
chromatography-tandem mass spectrometry (HPLC-MS/MS).
Chemosphere 88 (5), 563e569.
Botia, D.C., Rodriguez, M.S., Sarria, V.M., 2012. Evaluation of UV/
TiO
2
and UV/ZnO photocatalytic systems coupled to a
biological process for the treatment of bleaching pulp mill
effluent. Chemosphere 89 (6), 732e736.
Buerge, I.J., Buser, H.-R., Kahle, M., Muller, M.D., Poiger, T., 2009.
Ubiquitous occurrence of the artificial sweetener acesulfame
in the aquatic environment: an ideal chemical marker of
domestic wastewater in groundwater. Environ. Sci. Technol.
43 (12), 4381e4385.
Buerge, I.J., Keller, M., Buser, H.-R., Muller, M.D., Poiger, T., 2011.
Saccharin and other artificial sweeteners in soils: estimated
inputs from agriculture and households, degradation, and
leaching to groundwater. Environ. Sci. Technol. 45 (2),
615e621.
Cai, W., Dai, M., Wang, Y., Zhai, W., Huang, T., Chen, S., Zhang, F.,
Chen, Z., Wang, Z., 2004. The biogeochemistry of inorganic
carbon and nutrients in the Pearl River estuary and the
adjacent Northern South China Sea. Continent. Shelf Res. 24
(12), 1301e1319.
Calza, P., Sakkas, V.A., Medanna, C., Vlachou, A.D., Dal Bello, F.,
Albanis, T.A., 2013. Chemometric assessment and investigation
of mechanism involved in photo-Fenton and TiO
2
photocatalytic degradation of the artificial sweetener sucralose
in aqueous media. Appl. Catal. B: Environ. 129, 71e79.
Carbonaro, S., Sugihara, M.N., Strathmann, T.J., 2013.
Continuous-flow photocatalytic treatment of pharmaceutical
micropollutants: activity, inhibition and deactivation of TiO
2
photocatalysts in wastewater effluent. Appl. Catal. B Environ.
129, 1e12.
Chong, M.N., Jin, B., Chow, C.W.K., Saint, C., 2010. Recent
developments in photocatalytic water treatment technology: a
review. Water Res. 44 (10), 2997e3027.
Cohen, S.M., Arnold, L.L., Emerson, J.L., 2008. Safety of saccharin.
Agro Food Industry Hi-tech 19 (6), 24.
Da Dalt, S., Alves, A.K., Bergmann, C.P., 2013. Photocatalytic
degradation of methyl orange dye in water solutions in the
presence of MWCNT/TiO
2
composites. Mater. Res. Bull. 48 (5),
1845e1850.
Dantas, R.F., Rossiter, O., Ribeiro Teixeira, A.K., Simoes, A.S.M., da
Silva, V.L., 2010. Direct UV photolysis of propranolol and
metronidazole in aqueous solution. Chem. Eng. J. 158 (2),
143e147.
Devipriya, S.P., Yesodharan, S., Yesodharan, E.P., 2012. Solar
photocatalytic removal of chemical and bacterial pollutants
water research 52 (2014) 260e274272
from water using Pt/TiO
2
-coated ceramic tiles. Int. J.
Photoenergy, 1e7. Article number:970474.
EC, 1996. Technical Guidance Documents in Support of Directive
96-67-EEC on Risk Assessment of New Notified Substances
and Regulation (EC) No. 1488/94 on Risk Assessment of
Existing Substances (Parts I, II, III and IV). EC catalogue
numbers CR-48-96-001, 002, 003, 004-EN-C. Office for Official
Publications of the European Community, 2 rue Mercier, L-
2965 Luxemburg.
EU, 2003. European Parliament and Council Directive 2003/115/EC
of 22 December 2003 Amending Directive 94/35/EC on
Sweeteners for Use in Foodstuffs.
Fenoll, H., Hellin, P., Flores, P., Martinez, C.M., Navorro, S., 2013.
Degradation intermediates and reaction pathway of
carbofuran in leaching water using TiO
2
and ZnO as
photocatalyst under natural sunlight. J. Photochem. Photobiol.
A: Chem. 251, 33e40.
Gan, Z., Sun, H., Wang, R., Feng, B., 2013. A novel solid-phase
extraction for the concentration of sweeteners in water and
analysis by ion-pair liquid chromatography-triple quadrupole
mass spectrometry. J. Chromatogr. A 1274, 87e96.
Han, J., Liu, Y., Singhal, N., Wang, L., Gao, W., 2012. Comparative
photocatalytic degradation of estrone in water by ZnO and
TiO
2
under artificial UVA and solar irradiation. Chem. Eng. J.
213, 150e162.
Ho, A.Y.T., Xu, J., Yin, K., Yuan, X., He, L., Jiang, Y., Lee, D.H.W.,
Anderson, D.M., Harrison, P.J., 2008. Seasonal and spatial
dynamics of nutrients and phytoplankton biomass in Victoria
Harbour and its vicinity before and after sewage abatement.
Mar. Pollut. Bull. 57 (6e12), 313e324.
Hoffman, M.R., Martin, S.T., Choi, W.Y., Bahnemann, D.W., 1995.
Environmental applications of semiconductor photocatalysis.
Chem. Rev. 95 (1), 69e96.
ISO, 2007. Water Quality eDetermination of Inhibitory Effect of
Water Samples on the Light Emission of Vibrio Fischeri
(Luminescent Bacteria Test) ePart 3: Method Using Freeze-
dried Bacteria. ISO11348e3:2007, Geneva, Switzerland.
Jung, J., Kim, Y., Kim, J., Jeong, D.-H., Choi, K., 2008.
Environmental level of ultraviolet light potentiate the toxicity
of sulfonamide antibiotics in Daphnia magna. Ecotoxicology 17
(1), 37e45.
Kessler, R., 2009. Sweeteners persist in waterways. Environ.
Health Perspect. 117 (10), A438.
Klamerth, N., Rizzo, L., Malato, S., Maldonado, M.I., Aguera, A.,
Fernandez-Alba, A.R., 2010. Degradation of fifteen emerging
contaminants at mg
-1
initial concentrations by mild solar
photo-Fenton in MWTP effluents. Water Res. 44 (2), 545e554.
Kroger, M., Meister, K., Kava, R., 2006. Low calorie sweeteners and
other sugar substitutes: a review of the safety issues. Compr.
Rev. Food Sci. Food Saf. 5 (2), 35e47.
Kurbus, T., Slokar, Y.M., Le Marechal, A.M., Voncina, D.B., 2003.
The use of experimental design for the evaluation of the
influence of variables on the H
2
O
2
/UV treatment model textile
waste water. Dyes Pigm. 58 (2), 171e178.
Lange, F.T., Scheurer, M., Brauch, H.-J., 2012. Artificial
sweetenersda recently recognized class of emerging
environmental contaminants: a review. Anal. Bioanal. Chem.
403 (9), 2503e2518.
Loos, R., Gawlik, B.M., Boettcher, K., Locoro, G., Contini, S.,
Bidoglio, G., 2009. Sucralose screening in European surface
waters using a solid-phase extraction-liquid
chromatographyetriple quadrupole mass spectrometry
method. J. Chromatogr. A 1216 (7), 1126e1131.
Lubick, N., 2008. Artificial sweetener persists in the environment.
Environ. Sci. Technol. 42 (9), 3125.
Mead, R.N., Morgan, J.B., Avery, J., Keiber, R.J., Krik, A.M.,
Skrabal, S.A., Willey, J.D., 2009. Occurrence of the artificial
sweetener sucralose in coastal and marine waters of the
United States. Mar. Chem. 116 (1e4), 13e17.
Metz, D.H., Meyer, M., Dotson, A., Beerendonk, E., Dionysiou, D.D.,
2011. The effect of UV/H
2
O
2
treatment on disinfection by-
product formation potential under simulated distribution
system conditions. Water Res. 45 (13), 3969e3980.
Mortensen, A., 2006. Sweeteners permitted in European Union:
safety aspects. Scand. J. Food Nutr. 50 (3), 104e116.
Ministry of Health of the People’s Republic of China (MOH), 2011.
Food Safety National Standards - Standards for Use of Food
Additives GB 2760-2011 (Beijing, China).
Oppenheimer, J., Eaton, A., Badruzzaman, M., Haghani, A.W.,
Jacangelo, J.G., 2011. Occurrence and suitability of sucralose as
an indicator compound of wastewater loading to surface
waters in urbanized regions. Water Res. 45 (13), 4019e4027.
Ordonez, E.Y., Quintana, B.J., Rodil, R., Cela, R., 2012.
Determination of artificial sweeteners in water samples by
solid-phase extraction and liquid chromatography-tandem
mass spectrometry. J. Chromatogr. A 1256, 197e205.
Petersen, D.G., Dahllof, I., 2007. Combine effects of pyrene and
UV-light on algae and bacteria in an arctic sediment.
Ecotoxicology 16 (4), 371e377.
Rahn, A., Yaylayan, V.a, 2010. Thermal degradation of sucralose
and its potential in generating chloropropanols in the
presence of glycerol. Food Chem. 118 (1), 56e61.
Robertson, W.D., Van Stempvoort, D.R., Solomon, D.K.,
Homewood, J., Brown, S.J., Spoelstra, J., Schiff, S.L., 2013.
Persistence of artificial sweeteners in a 15-year-old septic
system plume. J. Hydrol. 477, 43e54.
Scheurer, M., Brauch, H.-J., Lange, F.T., 2009. Analysis and
occurrence of seven artificial sweeteners in German waste
water and surface water and in soil aquifer treatment (SAT).
Anal. Bioanal. Chem. 394 (6), 1585e1594.
Scheurer, M., Godejohann, M., Wick, A., Happel, O., Ternes, T.A.,
Brauch, H.-J., Ruck, W.K.L., Lange, F.T., 2012. Structural
elucidation of main ozonation products of the artificial
sweeteners cyclamate and acesulfame. Environ. Sci. Pollut.
Res. 19 (4), 1107e1118.
Schuerer, M., Storck, F.R., Brauch, H.-J., Lang, R.T., 2010.
Performance of conventional multi-barrier drinking water
treatment plants for the removal of four artificial sweeteners.
Water Res. 44 (12), 3573e3584.
Selli, E., Bianchi, C.L., Pirola, C., Cappelletti, G., Ragaini, V., 2008.
Efficiency of 1,4-dichlorobenzene degradation in water under
photolysis, photocatalysis on TiO
2
and sonolysis. J. Hazard.
Mater. 153 (3), 1136e1141.
Sichel, C., Garcia, C., Andre, K., 2011. Feasibility studies: UV/
chlorine advanced oxidation treatment for the removal of
emerging contaminants. Water Res. 45 (19), 6371e6380.
Soh, L., Connors, K.A., Brooks, B.W., Zimmerman, J., 2011. Fate of
sucralose through environmental and water treatment
processes and impact on plant indicator species. Environ. Sci.
Technol. 45 (4), 1363e1369.
Sousa, M.A., Goncalves, C., Vilar, V.J.P., Boaventura, R.A.R.,
Alpendurada, M.F., 2012. Suspended TiO
2
-assisted
photocatalytic degradation of emerging contaminants in a
municipal WWTP effluent using a solar pilot plant with CPCs.
Chem. Eng. J. 198, 301e309.
Torres, C.I., Ramakrishna, S., Chiu, C.A., Nelson, K.G.,
Westerhoff, P., Krajmalnik-Brown, R., 2011. Fate of sucralose
during wastewater treatment. Environ. Eng. Sci. 28 (5),
325e331.
US FDA, 1983. Food additives permitted for direct addition to food
for human consumption; aspartame. Final rule. Fed. Regist. 48,
31376e31382.
US FDA, 2006. Artificial sweeteners: no calories.sweet! FDA
Consumer Mag. 40 (4), 27e28.
water research 52 (2014) 260e274 273
Van Stempvoort, D.R., Roy, J.W., Brown, S.J., Bickerton, G., 2011.
Artificial sweeteners as potential tracers in groundwater in
urban environments. J. Hydrol. 401 (1e2), 126e133.
Weihrauch, M.R., Diehl, V., 2004. Artificial sweeteners-do they
bear a carcinogenic risk? Ann. Oncol. 15 (10), 1460e1465.
WHO Food Additives Series 16. Acesulfame Potassium.
Monographs and Evaluations. Joint FAO/WHO Expert
Committee on Food Additives.
Yang, B., Ying, G.-G., 2013. Oxidation of benzophenone-3 during
water treatment with ferrate(VI). Water Res. 47 (7), 2458e2466.
Yu, Y., Wu, L., Chang, A.C., 2013. Seasonal variation of endocrine
disrupting compounds, pharmaceuticals and personal care
products in wastewater treatment plants. Sci. Total Environ.
442 (1), 310e316.
Zhao, H., 1990. Ocean Press, Beijing (in Chinese).
Zhou, W., Yin, K., Harrison, P.J., Lee, J.H.W., 2012. The influence of
late summer typhoons and high river discharge on water
quality in Hong Kong waters. Estuarine, Coastal Shelf Sci. 111,
35e47.
Zygler, A., Wasik, A., Namiesnik, J., 2009. Analytical
methodologies for determination of artificial sweeteners in
foodstuffs. Trac Trends Anal. Chem. 28 (9), 1082e1102.
water research 52 (2014) 260e274274