Structural characterisation of degradation products formed upon di-n-butyl phthalate radiolysis by high-performance liquid chromatography electrospray tandem mass spectrometry

Article abstract of DOI:10.1255/ejms.1096

Structural characterisation of 15 degradation products, formed upon di-n-butyl phthalate (DBP) radiolysis, has been achieved using a high-performance liquid chromatography-tandem mass spectrometry (HPLC-MS/MS) coupling. The dissociation behaviour of protonated DBP was first established to be further used to characterise structural deviation in the degradation products. Based on accurate mass measurements, compounds shown by HPLC-MS analysis were all found to be DBP oxidation products, amongst which various sets of isomers could be distinguished. Collision-induced dissociation experiments performed on each electrosprayed molecule first allowed unambiguous definition of the location of the additional oxygen atoms; that is, in the alkyl branch or on the aromatic ring. Although location of the oxygen atom in the alkyl branches could not always be precisely determined, relative abundances of some product ions allowed oxygenated functions to be identified.

Full text of DOI:10.1255/ejms.1096

A. Tintaru, V. Labed and L. Charles, Eur. J. Mass Spectrom. 16, 595–603 (2010)
595
received: 12 october 2009
n
revised: 30 March 2010 n Accepted: 25 May 2010 n Publication: 18 August 2010
EuroPEAN
JourNAL
of
MASS
SPEctroMEtry
Structural characterisation of degradation
products formed upon di-n-butyl phthalate
radiolysis by high-performance liquid
chromatography electrospray tandem mass
spectrometry
Aura Tintaru,a Véronique Labedb and Laurence Charlesa,*
auniversités Aix-Marseille I, II et III–cNrS, uMr 6264: Laboratoire chimie Provence, Spectrométries Appliquées à la chimie Structurale,
f-13397 Marseille, france. E-mail: laurence.charles@univ-provence.fr
bcEA/Marcoule DtcD/SPDE/L2ED, f-30207 Bagnols sur cèze, france
Structural characterisation of 15 degradation products, formed upon di-n-butyl phthalate (DBP) radiolysis, has been achieved using
a high-performance liquid chromatography–tandem mass spectrometry (HPLC-MS/MS) coupling. The dissociation behaviour of
protonated DBP was first established to be further used to characterise structural deviation in the degradation products. Based on accu-
rate mass measurements, compounds shown by HPLC-MS analysis were all found to be DBP oxidation products, amongst which vari-
ous sets of isomers could be distinguished. Collision-induced dissociation experiments performed on each electrosprayed molecule
first allowed unambiguous definition of the location of the additional oxygen atoms; that is, in the alkyl branch or on the aromatic ring.
Although location of the oxygen atom in the alkyl branches could not always be precisely determined, relative abundances of some
product ions allowed oxygenated functions to be identified
Keywords: di-n-butyl phthalate, mass spectrometry, HPLC-ESI-MS/MS, structural characterisation, radiolysis products
Introduction
Phthalic acid dialkyl esters, usually named phthalates, absorption of cosmetics or contamination of blood and
are industrially produced in large quantities and are used
in many applications as plasticisers, fixatives, deter-
gents, lubricating oils or solvents.1–5 these molecules are
fat soluble and have been found, albeit at low levels, in
cosmetics6,7 or foods,8 including milk and its derivatives.9
Nowadays, phthalates are also ubiquitous contaminants in
the biosphere and they have been found worldwide in the
saliva by medical devices.17–19
Duetotheirtoxicity, numerousresearchstudieshavefocused
on the development of novel analytical methods, mainly using
liquid (Lc) or gas (Gc) chromatography, to improve quantita-
tive determination of phthalates in various samples.1–3,5,20–22
Moreover, many studies have been dedicated to the degrada-
tion mechanisms of phthalates, biodegradation and photolysis
atmosphere,10 water,11 sediments12 and soil.13,14 Moreover, being the most extensively studied processes. the main
humans are exposed to phthalates in numerous ways, metabolites formed during biodegradation were reported
by migration of phthalates into foodstuffs,15,16 dermal
to be mono-alkyl phthalates, obtained after cleavage of one
ISSN: 1469-0667
doi: 10.1255/ejms.1096
© IM Publications LLP 2010
All rights reserved

596
Structural Characterization of Degradation Products Formed upon Di-n-Butyl Phthalate Radiolysis
alkyl branch, and phthalic acid, resulting from the release of
both branches.23–28 In addition to these molecules, oxidation
products were also identified during phthalate photolysis.29–32
was split down to 50mLmin−1 using a zero-dead-volume tee
connector. A linear gradient elution was carried out starting
from water/methanol (50 :50, %, v/v) containing 3 mmolL−1
Accurate structural characterisation of the degradation prod- ammonium acetate up to 90% methanol in 30min. the final
ucts was mostly achieved using mass spectrometric (MS)
methods. Gc-MS has been widely used to identify detected
ions by comparison with mass spectra libraries.23,27,28 A few
Lc-MS couplings have also been reported, using electrospray
composition was held constant for 10 min before re-equili-
brating the column.
Mass spectrometry
ionisation (ESI)31–33 or atmospheric pressure chemical ionisa- Accurate mass measurements were performed with a QStar
tion (APcI)26 operated in the negative ion mode, to charac- Elite mass spectrometer (Applied Biosystems ScIEX, concord,
terise degradation products based on their molecular mass. oN, canada) equipped with an electrospray source operated in
Methods involving tandem mass spectrometry (MS/MS) have
also been reported, to determine the structure of molecules
arising from the biotransformation of di-n-butyl phthalate26 or
to determine the migration level of di(2-ethylhexyl)phthalate
and two of its well-known breakdown products, namely
mono(2-ethylhexyl)phthalate and phthalic acid, formed upon
gamma-ray irradiation of polyvinyl chloride sheets.33
the positive ion mode. the capillary voltage was set at 5500V
and the declustering potential was 40V. In this hybrid instru-
ment, ions were measured using an orthogonal acceleration
time-of-flight mass analyser. A quadrupole was used for selec-
tion of precursor ions to be further submitted to cID in MS/MS
experiments. Air was the nebulising gas (10psi) while nitrogen
was used as the curtain gas (20psi) as well as the collision
gas. the collision energy was set according to the experi-
In this study, a HPLc-ESI-MS/MS coupling was developed to
characterise major degradation products formed upon di-n- ments. the Analyst software version, 2.1 (Applied Biosystems
butyl phthalate (DBP) radiolysis. the fragmentation pathways ScIEX) was used for instrument control, data acquisition and
observed during collision-induced dissociation (cID) of proto- data processing. Direct sample introduction was performed at
nated DBP were first established, to be further used as guide- a 5mLmin−1 flow rate using a syringe pump.
lines in the MS/MS study of degradation products, in conjunc-
tion with accurate mass measurements.
Results and discussion
HPLC-MS analysis
Experimental section
the HPLc-ESI-MS coupling has been developed from a previ-
ously reported method6 using a DBP standard sample. Due to
the potentially different polarity of the DBP degradation prod-
ucts, a gradient elution was used, starting with a 50:50 (%, v/v)
water–methanol mobile phase and increasing the methanol
content up to 90% in 30min. In these chromatographic condi-
tions, DBP was eluted at a 21.5min retention time (tr). Injection
of the irradiated DBP sample gave rise to the HPLc-MS
chromatogram presented in figure 1, which shows a main
peak assigned to DBP, on the basis of both retention time and
Chemicals
Di-n-butyl phthalate (DBP) and ammonium acetate were
purchased from Sigma Aldrich (St Louis, Mo, uSA). Methanol
(Sigma Aldrich) and water (SDS, Peypin, france) were HPLc
grade. All chemicals were used without further purification.
Irradiation of DBP
Gamma irradiation was performed with a gamma source (60co
source activity: 5.5×1014 Bq). Pure DBP was irradiated under
air and at ambient temperature. the sample was homogene- ion m/z value. Low abundance impurities were mostly eluted
ously irradiated by rotating it during treatment and the dose
rate was about 400 Gy h−1. Dosimetry was performed using
Perspex films. the dose integrated by the sample was around
0.25 MGy. Irradiated DBP was then solubilised in methanol
prior to DBP, presumably due to their higher polarity; however,
some very small peaks could also be observed at higher reten-
tion times. Extracted MS spectra revealed the presence of
different sets of compounds. In particular, the most abundant
and further diluted using a solution of water/methanol (50:50, impurities were five isobaric molecules detected at m/z 295.2;
% v/v) containing 3mmolL−1 ammonium acetate to reach a
that is, a 16Da mass increase compared to DBP. As indicated
by accurate mass data, all these m/z 295.2 ions would be
protonated molecules with the same elemental composition,
including an additional oxygen atom compared to DBP. two
peaks were observed in the extracted chromatogram corre-
final concentration of about 10mgmL−1.
High-performance liquid chromatography
HPLc experiments were performed on an Agilent 1100
series system (Agilent technologies, Santa clara, cA, uSA). sponding to m/z 293.2 and elemental composition found for
the chromatographic column (ZorBAX Eclipse XDB–c18, both compounds indicated the addition of one oxygen atom
4.6 × 250 mm, pore diameter: 80 Å, particle size: 5 mm), was
from Agilent technologies and was maintained at a constant
temperature of 35°c. for all experiments, the injection volume
was 10mL and the flow rate was set to 1mLmin−1. Prior to
entering the electrospray source, the chromatographic effluent
and one double bound compared to DBP. these results would
indicate DBP has been oxidised during the radiolysis process,
as previously reported during photolysis of different phtha-
lates.24,29,31 Additional impurities were detected with lower
abundance at m/z 223.1 (tr =3.3min), m/z 251.1 (tr =16.1min),

A. Tintaru, V. Labed and L. Charles, Eur. J. Mass Spectrom. 16, 595–603 (2010)
597
Figure 1. HPLC-ESI-MS chromatogram of irradiated DBP. Retention time (tR) of DBP was measured at 21.5min.
m/z 267.1 (tr =9.4min), m/z 307.1 (tr =19.0min, 20.6min and
fragmentation of the protonated phthalic anhydride at m/z
26.1min) and m/z 311.1 (tr =14.2min and 15.3min). cID experi- 149.0. Accurate mass measurements suggest a pathway
ments were then performed on each of these precursor ions
to characterise these DBP degradation products, using MS/
MS of protonated DBP as reference dissociation behaviour.
(a)
Collision-induced dissociation of [DBP+H]+
the fragmentation pattern of protonated DBP was first char-
acterised by direct infusion of a DBP methanolic solution
containing ammonium acetate (3mmol L−1). under positive
mode ESI, protonated DBP was readily detected at m/z 279.2.
the MS/MS spectrum of [DBP+H]+ shows a very intense product
ion at m/z 149.0 [figure 2(a)], commonly reported during frag-
mentation of any phthalates, regardless of the dissociation
mode and the ionisation mode polarity.12,26 this product ion
was assigned to a protonated phthalic anhydride, obtained
after elimination of both alkyl branches from the phthalate
precursor ion.21 the proposed two-step mechanism would
first involve a proton transfer to allow the release of a butanol
molecule from the precursor ion. In the so-formed m/z 205.1
ion, a proton transfer, from the remaining alkyl chain to the
carboxyl oxygen atom, would then induce the opening of the
six-member ring and the acidic proton would be transferred
to the ethylenic carbon via a 1,6-rearrangement, allowing
two species to be released and observed in their protonated
form at m/z 149.0 and m/z 57.1, respectively [Scheme 1(a)].
Based on accurate mass data, the very low abundance product
ion detected at m/z 223.1 would indicate the elimination of
a 1-butene neutral from the m/z 279.2 precursor ion. the
proposed 1,5-proton transfer depicted in Scheme 1(b) to
account for this 56.1 Da neutral loss would also allow the
production of m/z 57.1, depending on the location of the proton
at the time of dissociation. MS/MS experiments performed at
higher collision energies allow three additional product ions to
be observed at m/z 121.0, m/z 93.0 and m/z 65.0 (figure 2(b)).
formation of these ions could be accounted for by considering
(b)
Figure 2. ESI-MS/MS spectra of [DBP+H]+ at m/z 279.2 using
a collision energy of (a) 10eV and (b) 45eV (laboratory frame).
Inset: Structure of DBP.

598
Structural Characterization of Degradation Products Formed upon Di-n-Butyl Phthalate Radiolysis
consisting of the successive release of three co molecules.
the first co loss would readily occur by the opening of the
phthalic anhydride, induced by the protonation of the oxygen
atom, giving rise to m/z 121.0.7,21,34 In the following step, the
remaining carbonyl group would allow the formation of a four-
member ring, involving two carbon atoms of the aromatic
cycle. A rearrangement would further yield a more stable ionic
form, in which a 1,5-proton transfer from the hydroxyl to the
carboxyl oxygen atom would allow the formation of m/z 93.0
via a straightforward elimination of a co neutral. to account
for the formation of m/z 65.0, opening of the aromatic ring in
m/z 93.0 should be envisaged via a 1,4-proton migration. In
the so-formed structure, a 1,3-transfer of the hydroxyl proton
followed by the heterolytic cleavage of a c–c bond would lead
to the formation of m/z 65.0, the stability of which should be
enhanced due to charge delocalisation.
(a)
(b)
Since any structural deviations from DBP are expected to
give rise to significant modifications of the previously depicted
cID behaviour, MS/MS spectra of the protonated impurities
have thus been scrutinised for these differences. this meth-
odology will be described in details hereafter with the example
of the five 294Da isomeric compounds while the main data
and structural conclusions will be presented for all other
identified impurities.
(c)
(d)
(e)
Protonated molecules at m/z 295.2
MS/MS spectra obtained for the five m/z 295.2 protonated
compounds are presented figure 3. from these data, two
groups could clearly be distinguished based on product ions
formed upon cID. the first group includes compounds A, B
and C, whose protonated forms show a similar fragmenta-
tion pattern compared to DBP [figures 3(a)–(c)]. In particular,
the same m/z 205.1 and m/z 149.0 products ions, reported
to be respectively formed from [DBP+H]+ after elimination
of a butanol neutral from one alkyl branch and a 1-butene
molecule from the second branch, are also formed during
the dissociation of [A+H]+, [B+H]+ and [C+H]+. Based on accu-
rate mass measurements, a c4H10o2 composition was found
for the 90.1Da neutral released from the protonated A, B or C
precursor ion to produce m/z 205.1, indicating that the addi-
tional oxygen in these DBP degradation products is located
in one of the two alkyl chains. In addition to the m/z 205.1
and m/z 149.0, four other products ions could be observed in
the MS/MS spectra of [A+H]+, [B+H]+ and [C+H]+. A m/z 277.1
ion, arising from dehydration of the m/z 295.2 precursor ion,
is formed with a high abundance upon cID of [B+H]+ and
[C+H]+ while it is hardly detected from [A+H]+. this favoured
Figure 3. MS/MS spectra of the m/z 295.2 precursor ion
obtained after electrospray ionisation of (a) A (tR =12.5min), (b)
B (tR =13.4min), (c) c (tR =14.1min), (d) D (tR =18.7min) and (e)
E (tR =19.8min). All spectra were obtained using a 10eV colli-
sion energy (laboratory frame).
m/z 223.1 is also shown to proceed at a different rate from
[A+H]+ compared to [B+H]+ and [C+H]+, further confirming
water elimination in the case of protonated B and C mole- that the oxygen atom is involved in a different function in
cules suggests the presence of a hydroxyl function in one
alkyl branch of these two isomers while A would contain a
different type of oxygenated function. Although the different
relative intensities of m/z 205.1 and m/z 277.1 measured in
figures 3(b) and (c) is probably due to a different location of
the oH moiety in B and C, this information was not sufficient
for these two isomers to be distinguished. Elimination of a
c4H8o neutral from the m/z 295.2 precursor ion to produce
the case of A. two additional weak abundance product ions,
detected at m/z 221.1 and m/z 73.1 in the MS/MS spectra of
[A+H]+ and [C+H]+, could be accounted for by considering the
same reaction mechanisms as described in Scheme 1(a).
the release of a butanol molecule (74.1Da) to form m/z 221.1
from both precursor ions would confirm one alkyl branch
is intact. further dissociation of this m/z 221.1 product ion
would release phthalic anhydride and a c4H8o molecule,

A. Tintaru, V. Labed and L. Charles, Eur. J. Mass Spectrom. 16, 595–603 (2010)
599
(a)
(b)
Scheme 1. Proposed dissociation pathways of [DBP+H]+ at m/z 279.2 to account for (a) the formation of m/z 149.0 and m/z 57.1, via the
intermediate m/z 205.1 product ion and (b) the production of the complementary m/z 223.1 and m/z 57.1 product ions.
respectively observed in their protonated forms at m/z 149.0
and m/z 73.1. No realistic assumption could be made to
account for the absence of these m/z 221.1 and m/z 73.1 in the
MS/MS spectrum of [B+H]+. Moreover, MS/MS experiments
performed at higher collision energies (data not shown) did
not allow additional informative product ions to be produced
from [A+H]+, [B+H]+ and [C+H]+ since the same m/z 121.0, m/z
93.0, m/z 65.0 and m/z 57.1 ions previously observed during
cID of [DBP+H]+ were detected.
In contrast to A, B and C compounds, D and E molecules were
classified in a second group due to the absence of any signal
at m/z 205.1 and m/z 149.0 in the MS/MS spectra of [D+H]+
and [E+H]+. these two product ions could actually be observed
with a 16Da shift, at m/z 221.1 and m/z 165.0, respectively. the
MS/MS fragmentation pathway observed for the protonated
form of D and E is actually the same as for [DBP+H]+; that is, a
sequential loss of 74.1Da and 56.1Da neutrals. these results
allowed us to conclude that the additional oxygen atom in D
were observed in the MS/MS spectrum of [E+H]+. the pres-
ence of m/z 147.0 reveals a dehydration reaction has occurred
from the m/z 165.0 product ion. Since aromatic oH functions
do not usually give rise to dehydration, the observed water
elimination could only be envisaged by considering the oH
group is in ortho-position towards one carbonyl group in the
phthalic anhydride moiety. Accordingly, the hydroxyl H atom
could be transferred to this carbonyl o atom in a concerted
electron pair migration (Scheme 2). further protonation of the
so-formed hydroxyl function would then allow dehydration to
proceed. the absence of any signal at m/z 147.0 in the MS/MS
spectrum of [D+H]+ would thus indicate the hydroxyl group on
the aromatic ring is in the meta-position with respect to the
other substituents on the aromatic ring of D. further dissocia-
tion of this m/z 147.0 product ion would account for the addi-
tional four peaks observed in figure 4(b). A first pathway would
consist of two successive co neutral losses to yield m/z 119.0
and m/z 91.0, respectively. An alternative mechanism for the
and E is located on the aromatic moiety of the phthalate mole- dissociation of m/z 147.0 would consist of a decarboxylation
cules. Data obtained while increasing the collision energy in
cID experiments allowed D to be distinguished from E (figure
4). the formation of all product ions arising from the dissocia- 2.
tion of [D+H]+ could be explained using the same mechanisms
as for [DBP+H]+ [Schemes 1(a) and (b)], giving rise to peaks
detected with a 16 Da shift compared to data from figure
2(b) (except for m/z 57.1). In contrast, five additional peaks
reaction to produce m/z 103.0, which would further eliminate
a co molecule to give rise to m/z 75.0, as illustrated in Scheme
unfortunately, structural investigation by nuclear magnetic
resonance (NMr) performed on HPLc preparative fractions of
the targeted compounds did not allow more precise informa-
tion about the oxygen atom position in the alkyl branch of A,

600
Structural Characterization of Degradation Products Formed upon Di-n-Butyl Phthalate Radiolysis
HPLc-ESI-MS and observed at tr =12.3min and tr =13.9min,
respectively. upon cID, both protonated molecules at m/z
293.1 eliminate a butanol molecule to yield m/z 219.1,
indicating an intact butyl branch and an 88.1Da neutral from
the modified second branch to produce m/z 205.1. the main
product ion detected at m/z 149.0 further confirmed that these
compounds were phthalate derivatives and that the structural
modification affect one alkyl branch in both cases. As a result,
these degradation products were assumed to arise from the
oxidation of one methylene group in a DBP butyl chain into a
carbonyl moiety. these two isomers could be distinguished
based on two additional peaks only observed, with similar
intensities, in the MS/MS spectrum of [G+H]+ at m/z 223.1 and
m/z 71.0. Elimination of a c4H6o neutral to produce m/z 223.1
and the formation of m/z 71.0 (c4H7o+: m/z 71.0491, −9.3ppm)
suggested complementary ions for which a reaction mecha-
nism could be proposed. It would consist of a six-member
rearrangement in the modified alkyl branch, such as the reac-
tion depicted to occur from [DBP+H]+ in Scheme 1(b), which
could only proceed if the additional co group is not in the b
position relative to the ester moiety. As a result, although the
structural alteration could not be accurately localised in F,
the structure of the modified branch G would be –cH2–(co)–
cH2–cH3 (Scheme 3).
(a)
(b)
Protonated molecules at m/z 307.2
Figure 4. MS/MS spectra of the m/z 295.2 precursor ion
obtained after electrospray ionisation of (a) D and (b) E, using a
50eV collision energy (laboratory frame).
Accordingtoaccuratemassdata, thepresenceofthreeisomers
with an elemental composition such as c17H22o5 was revealed
in the extracted chromatogram at m/z 307.2. the later eluting
compound J was readily identify in MS/MS as a DBP degrada-
tion product containing a cH=o group as a substituent on the
aromatic ring. Indeed, the [J+H]+ precursor ion was shown to
sequentially eliminate a 74Da and a 56Da neutral, to yield a
main product ion at m/z 177.0. However, localisation of the
substituting group on the aromatic ring could not be further
determined (Scheme 3) since no additional information could
be obtained while increasing the collision energy in the cID
experiment. MS/MS data obtained for the two other H and I
B and C, due to the very low concentration of the degradation
products. Proposed structures for compounds A–E are
summarised in Scheme 3.
Protonated molecules at m/z 293.1
two compounds, further named F and G, with the same
elemental composition (c16H20o5) could be separated in
Scheme 2. Proposed dissociation pathways of the m/z 165.0 product ion formed during CID of [E+H]+ to account for the formation of m/z
147.0, m/z 119.0, m/z 103.0, m/z 91.0 and m/z 75.0.

A. Tintaru, V. Labed and L. Charles, Eur. J. Mass Spectrom. 16, 595–603 (2010)
601
Scheme 3. Structures proposed for DBP radiolysis by-products detected in HPLC-MS.
isomers indicated that both molecules still contain an intact 106Da neutral, the protonated phthalic anhydride at m/z 149.0
butyl branch as well as an aromatic ring with no additional
substituent. Due to the lack of any dehydration reaction, it
could also be concluded that the two additional oxygen atoms
and a m/z 89.1 product ion. Based on this information, it could
be concluded that structural alterations in K and L consist of
two oH groups located in the same alkyl branch. release of this
within the second branch were involved in a carbonyl moiety. modified branch from m/z 311.1 precursor ions, according to
the only difference between these isomers was the production the reaction depicted in Scheme 1(a), would lead to the forma-
rate of two complementary ions at m/z 223.1 and m/z 85.0, tion of m/z 205.1 while a rearrangement such as that illustrated
+
which was found to be higher in the case of H (tr =19.0min)
compared to I (tr = 20.6 min). the mechanistic assumption
previously proposed in Scheme 1(b) could also account for
the formation of m/z 223.1 and m/z 85.0 product ions from
m/z 307.2, considering the formation of a c4H4o2 molecule as
a neutral or in its protonated form, respectively. Since these
complementary ions are detected from the cID of both H and
I precursor ions, this would imply a H atom is available in
the group in the b position towards the ester moiety for the
proposed reaction to proceed; that is, a second branch such
in Scheme 1(b) would produce m/z 223.1 or a c4H9o2 ion at
m/z 89.1, depending on the location of the charge at the time
of dissociation. the two MS/MS spectra could only be distin-
guished by a slight difference in abundance of m/z 237.1, m/z
223.1 and m/z 89.1; however, this information was not sufficient
for us to assign a more precise structure than –c3H11(oH)2 to
the modified branch in K and L (Scheme 3).
Protonated molecule at m/z 267.1
A c13H14o6 molecule, further named M, was found to be a
as –(cH2)2–c2Ho2 or a –(co)–cH2–c2H3o (Scheme 3). However, phthalate derivative with no additional substituent on the
no realistic assumption could be proposed for a more precise
assignment.
aromatic ring (main fragment ion at m/z 149.0) and a peak
observed at m/z 193.0 in the MS/MS spectrum of this m/z 267.1
precursor ion indicated one intact butyl branch. two additional
product ions formed upon cID of [M+H]+ were detected at m/z
249.1 and m/z 205.1 and would arise from the elimination of a
water molecule and a carbonic acid [according to the reaction
mechanism presented in Scheme 1(a)], respectively. Based on
these findings, it could be concluded that M was a DBP degra-
dation product with a carboxylic acid group instead of a butyl
chain in one branch (Scheme 3).
Protonated molecules at m/z 311.1
Very similar MS/MS spectra were obtained for the two c16H22o6
isomers eluted at tr =14.2min (K) and tr =15.3min (L). cID of
both precursor ions at m/z 311.1 were shown to produce a low
abundance m/z 293.1 after dehydration, m/z 237.1 after the
release of butanol (indicating the presence of an intact butyl
branch), m/z 223.1 (loss of 88Da), m/z 205.1 by eliminating a

602
Structural Characterization of Degradation Products Formed upon Di-n-Butyl Phthalate Radiolysis
Protonated molecule at m/z 251.1
2. K. Kato, M.J. Silva, L.L. Needham and A.M. calafat,
“Determination of 16 phthalate metabolites in urine using
automated sample preparation and on-line preconcen-
tration/high-performance liquid chromatography/tan-
dem mass spectrometry”, Anal. Chem. 77, 2985 (2005).
doi: 10.1021/ac0481248
3. P. Serodio and J.M.f. Nogueira, “considerations
on ultra-trace analysis of phthalates in drinking
water”, Water Res. 40, 2572 (2006). doi: 10.1016/j.
watres.2006.05.002
4. H.M. Koch, J. Muller and J. Angerer, “Determination of
secondary, oxidised di-iso-nonylphthalate (DINP) metab-
olites in human urine representative for the exposure to
commercial DINP plasticizers”, J. Chromatogr. B 847, 114
(2007). doi: 10.1016/j.jchromb.2006.09.044
5. c.P. feas, M.c.B. Alonso, E. Pena-Vazquez, P.H. Hermelo
and P. Bermejo-Barrera, “Phthalates determination in
physiological saline solutions by HPLc-ES-MS”, Talanta
75, 1184 (2008). doi: 10.1016/j.talanta.2008.01.019
6. D. De orsi, L. Gagliardi, r. Porra, S. Berri, P. chimenti,
A. Granese, I. carpani and D. tonelli, “An environmen-
tally friendly reversed-phase liquid chromatography
method for phthalates determination in nail cosmet-
ics”, Anal. Chim. Acta 555, 238 (2006). doi: 10.1016/j.
aca.2005.09.029
7. H.y. Shen, H.L. Jiang, H.L. Mao, G. Pan, L. Zhou and y.f.
cao, “Simultaneous determination of seven phthalates
and four parabens in cosmetic products using HPLc-
DAD and Gc-MS methods”, J. Sep. Sci. 30, 48 (2007). doi:
10.1002/jssc.200600215
8. S. Jara, c. Lysebo, t. Greibrokk and E. Lundanes,
“Determination of phthalates in water samples
using polystyrene solid-phase extraction and liquid
chromatography quantification”, Anal. Chim. Acta
407, 165 (2000). doi: http://dx.doi.org/10.1016/S0003-
2670(99)00829-6
Structural elucidation of the N molecule, observed at a
16.1min retention time in the extracted trace at m/z 251.1, was
quite straightforward. the presence of an intact butyl branch
was indicated by an m/z 177.0 product ion, which would further
eliminate a 28 Da neutral to produce m/z 149.0. the 46 Da
+
molecule released upon cID of this c13H15o5 precursor ion
to generate the low m/z 205.1 product could be assigned to
formaldehyde. As a result, N was found to be phthalate deriva-
tive with one butyl chain as the first branch and an aldehyde
moiety as the second branch (Scheme 3).
Protonated molecule at m/z 223.1
the very early eluting O compound (tr =3.3 min) could also be
identified based on its cID behaviour. MS/MS of the m/z 223.1
precursor ion mainly generated protonated phthalic anhydride
at m/z 149.0 after the elimination of a butanol molecule. A low
abundance peak at m/z 205.1 indicated a dehydration reac-
tion, which would be followed by the release of a co neutral
to yield m/z 177.1. these results strongly suggest that O was
monobutylphthalate (Scheme 3), previously reported to be a
DBP biodegradation by-product.28
Conclusion
fifteen compounds arising from di-n-butyl phthalate radiolysis
were distinguished using HPLc-ESI-MS/MS. Although functional
moieties could not always be precisely localised in all by-prod-
ucts, a combination of accurate mass data and MS/MS informa-
tion allowed structural modification to be identified. this coupling
has several advantages over other MS-based methods previously
reported in phthalate degradation studies. first, extraction of
the degradation products from the irradiated medium is not
requested, as in Gc-MS, prior to analysis. In addition, cID of the
electrosprayed ions allowed a more detailed analysis compared
to Gc-MS. In particular, location of the oH group on the aromatic
ring of two isomers could not be revealed in Gc-MS.
9. L.K. Sorensen, “Determination of phthalates in milk and
milk products by liquid chromatography/tandem mass
spectrometry”, Rapid Commun. Mass Spectrom. 20, 1135
(2006). doi: 10.1002/rcm.2425
10. X.K. Wang, W.L. Guo, P.r. Meng and J.A. Gan, “Analysis
of phthalate esters in air, soil and plants in plastic film
greenhouse”, Chin. Chem. Lett. 13, 557 (2002).
11. H. Katsumata, A. Begum, S. Kaneco, t. Suzuki and K.
ohta, “Preconcentration of phthalic acid esters in water
samples by Saccharomyces cerevisiae immobilized on
silica gel”, Anal. Chim. Acta 502, 167 (2004). doi: 10.1016/j.
aca.2003.10.002
Acknowledgement
L. charles acknowledges support from Spectropole, the
Analytical facility of Aix-Marseille university, by allowing
special access to the instruments purchased with European
funding (fEDEr oBJ2142-3341).
12. Z.P. Lin, M.G. Ikonomou, H.W. Jing, c. Mackintosh and
f. Gobas, “Determination of phthalate ester congeners
and mixtures by Lc/ESI-MS in sediments and biota of an
urbanized marine inlet”, Environ. Sci. Technol. 37, 2100
(2003). doi: 10.1021/es026361r
13. Q.H. Zhou, Z.B. Wu, S.P. cheng, f. He and G.P. fu,
“Enzymatic activities in constructed wetlands and
d i- n-butyl phthalate (DBP) biodegradation”, Soil
References
1. K. Kato, S. Shoda, M. takahashi, N. Doi, y. yoshimura
and H. Nakazawa, “Determination of three phthalate
metabolites in human urine using on-line solid-phase
extraction–liquid chromatography–tandem mass
spectrometry”, J. Chromatogr. B 788, 407 (2003). doi:
10.1016/S1570-0232(03)00041-2

A. Tintaru, V. Labed and L. Charles, Eur. J. Mass Spectrom. 16, 595–603 (2010)
603
Biol. Biochem. 37, 1454 (2005). doi: 10.1016/j.soil-
bio.2005.01.003
pseudomonas-pseudoalcaligenes under anaerobic con-
ditions, with nitrate as the only electron-acceptor”, Appl.
Environ. Microbiol. 44, 576 (1982).
14. S. Shailaja, M. ramakrishna, S.V. Mohan and P.N.
Sarma, “Biodegradation of di-n-butyl phthalate
(DnBP) in bioaugmented bioslurry phase reactor”,
Bioresource Technol. 98, 1561 (2007). doi: 10.1016/j.
biortech.2006.06.009
15. H. Kataoka, M. Ise and S. Narimatsu, “Automated on-
line in-tube solid-phase microextraction coupled with
high performance liquid chromatography for the analy-
sis of bisphenol A, alkylphenols, and phthalate esters
in foods contacted with plastics”, J. Sep. Sci. 25, 77
(2002). doi: 10.1002/1615-9314(20020101)25:1/2<77::AID-
JSSc77>3.0.co;2-3
16. J.f. Jen and t.c. Liu, “Determination of phthalate esters
from food-contacted materials by on-line microdialysis
and liquid chromatography”, J. Chromatogr. A 1130, 28
(2006). doi: 10.1016/j.chroma.2006.06.029
17. t. Dine, f. Khalfi, B. Gressier, M. Luyckx, c. Brunet, L.
Ballester, f. Goudaliez, J. Kablan, M. cazin and J.c.
cazin, “Stability study of fotemustine in PVc infusion
bags and sets under various conditions using a stability-
indicating high-performance liquid chromatographic
assay”, J. Pharm. Biomed. Anal. 18, 373 (1998). doi:
10.1016/S0731-7085(98)00096-X
24. J.L. Wang, P. Liu and y. Qian, “Microbial-degradation of
di-n-butyl phthalate”, Chemosphere 31, 4051 (1995). doi:
10.1016/0045-6535(95)80006-7
25. c.A. Staples, D.r. Peterson, t.f. Parkerton and W.J.
Adams, “the environmental fate of phthalate esters:
A literature review”, Chemosphere 35, 667 (1997). doi:
10.1016/S0045-6535(97)00195-1
26. N.G. coldham, M. Dave and M.J. Sauer, “Analysis of
di-n-butylphthalate biotransformation in cattle by
liquid chromatogaphy ion trap mass spectrometry
mass spectrometry”, J. Mass Spectrom. 33, 803 (1998).
doi: 10.1002/(SIcI)1096-9888(199809)33:9<803::AID-
JMS689>3.0.co;2-0
27. S.M. Lee, B.W. Koo, S.S. Lee, M.K. Kim, D.H. choi, E.J.
Hong, E.B. Jeung and I.G. choi, “Biodegradation of
dibutylphthalate by white rot fungi and evaluation on
its estrogenic activity”, Enzyme Microb. Technol. 35, 417
(2004). doi: 10.1016/j.enzmictec.2004.06.001
28. S.M. Lee, J.W. Lee, B.W. Koo, M.K. Kim, D.H. choi and I.G.
choi, “Dibutyl phthalate biodegradation by the white rot
fungus, Polyporus brumalis”, Biotechnol. Bioeng. 97, 1516
(2007). doi: 10.1002/bit.21333
18. M.A. faouzi, f. Khalfi, t. Dine, M. Luyckx, c. Brunet, B.
Gressier, f. Goudaliez, M. cazin, J. Kablan, A. Belabed
and J.c. cazin, “Stability, compatibility and plasticizer
extraction of quinine injection added to infusion solu-
tions and stored in polyvinyl chloride (PVc) contain-
ers”, J. Pharm. Biomed. Anal. 21, 923 (1999). doi: 10.1016/
S0731-7085(99)00204-6
19. t. Niino, t. Ishibashi, t. Itoh, S. Sakai, H. Ishiwata, t.
yamada and S. onodera, “comparison of diisononyl
phthalate migration from polyvinyl chloride products
into human saliva in vivo and into saliva simulant in vitro”,
J. Health Sci. 48, 277 (2002). doi: 10.1248/jhs.48.277
20. K. Kato, M.J. Silva, L.L. Needham and A.M. calafat,
“Determination of total phthalates in urine by iso-
tope-dilution liquid chromatography–tandem mass
spectrometry”, J. Chromatogr. B 814, 355 (2005). doi:
10.1016/j.jchromb.2004.10.056
29. A.I. Balabanovich, S. Denizligil and W. Schnabel,
“Photolysis of dialkyl phthalates and poly(vinyl chloride)
plasticized with dialkyl phthalates at wavelengths
exceeding 290 nm”, J. Vinyl Addit. Technol. 3, 42 (1997).
doi: 10.1002/vnl.10164
30. A. Balabanovich and W. Schnabel, “on the photolysis of
phthalic acid dimethyl and diethyl ester: a product analy-
sis study”, J. Photochem. Photobiol. A 113, 145 (1998). doi:
10.1016/S1010-6030(97)00334-1
31. o. Bajt, G. Mailhot and M. Bolte, “Degradation of dibutyl
phthalate by homogeneous photocatalysis with fe(III) in
aqueous solution,” Applied Catal. B 33, 239 (2001). doi:
10.1016/S0926-3373(01)00179-5
32. t.K. Lau, W. chu and N. Graham, “the degradation of
endocrine disruptor di-n-butyl phthalate by uV irradia-
tion: A photolysis and product study”, Chemosphere 60,
1045 (2005). doi: 10.1016/j.chemosphere.2005.01.022
33. r. Ito, N. Miura, M. ushiro, M. Kawaguchi, H. Nakamura,
H. Iguchi, J. ogino, M. oishi, N. Wakui, y. Iwasaki, K.
Saito and H. Nakazawa, “Effect of gamma-ray irradia-
tion on degradation of di(2-ethylhexyl)phthalate in poly-
vinyl chloride sheet”, Int. J. Pharm. 376, 213 (2009). doi:
10.1016/j.ijpharm.2009.04.019
21. M. castillo and D. Barcelo, “characterisation of organic
pollutants in textile wastewaters and landfill leachate
by using toxicity-based fractionation methods followed
by liquid and gas chromatography coupled to mass
spectrometric detection”, Anal. Chim. Acta 426, 253
(2001). doi: 10.1016/S0003-2670(00)00828-X
22. K. Luks-Betlej, P. Popp, B. Janoszka and H. Paschke,
“Solid-phase microextraction of phthalates from water”,
J. Chromatogr. A 938, 93 (2001). doi: 10.1013/S0021-
9673(01)01363-2
23. G. Benckiser and J.c.G. ottow, “Metabolism
of the plasticizer di-normal-butylphthalate by
34. X.H. Guo, A.P. Bruins and t.r. covey, “characterization
of typical chemical background interferences in atmos-
pheric pressure ionization liquid chromatography–mass
spectrometry”, Rapid Commun. Mass Spectrom. 20, 3145
(2006). doi: 10.1002/rcm.2715