Atmospheric oxidation of poly(oxyethylene) alcohols. Identification of ethoxylated formates as oxidation products and study of their contact allergenic activity

Article abstract of DOI:10.1021/js980210y

Ethoxylated alcohols are widely used as surfactants. In the present study we have continued our investigations on the degradation with time upon air exposure of the ethoxylated alcohols at normal storage and handling. As a result, a new group of ethoxylated formates with the general formula C₁₂H₂₅(OCH₂CH₂)(n)OCHO (n = 0-4) was identified in C₁₂H₂₅(OCH₂CH₂)₅OH stored and handled at room temperature. To facilitate the identification work, reference compounds were synthesized. The formates showed no allergenic activity in the sensitization studies performed. In previous investigations on the same ethoxylated alcohol, we have identified formaldehyde and ethoxylated aldehydes among the oxidation products formed. Formaldehyde is a common contact allergen, and the ethoxylated aldehydes were shown to have a sensitizing capacity of the same magnitude as formaldehyde. The instability of the ethoxylated alcohols and formation of oxidation products may give an allergenic contribution to hand eczema caused by work with water and surfactants. To investigate the clinical significance in man an appropriate diagnostic patch testing in exposed humans is required.

Full text of DOI:10.1021/js980210y

Atmospheric Oxidation of Poly(oxyethylene) Alcohols. Identification of
Ethoxylated Formates as Oxidation Products and Study of Their
Contact Allergenic Activity
MARGARETA BERGH,* LI PING SHAO, KERSTIN MAGNUSSON, ELISABETH G A¨ FVERT, J. LARS G. NILSSON, AND
ANN-THERESE KARLBERG
Contribution from Department of Occupational Medicine, Dermatology, National Institute for Working Life, Solna, Sweden.
Received May 18, 1998.
Final revised manuscript received November 19, 1998.
Accepted for publication December 15, 1998.
is a free radical mechanism initiated by minor amounts of
Abstract 0 Ethoxylated alcohols are widely used as surfactants. In
the present study we have continued our investigations on the
degradation with time upon air exposure of the ethoxylated alcohols
at normal storage and handling. As a result, a new group of ethoxylated
formates with the general formula C12H25(OCH2CH2)nOCHO (n ) 0−4)
was identified in C12H25(OCH2CH2)5OH stored and handled at room
temperature. To facilitate the identification work, reference compounds
were synthesized. The formates showed no allergenic activity in the
sensitization studies performed. In previous investigations on the same
ethoxylated alcohol, we have identified formaldehyde and ethoxylated
aldehydes among the oxidation products formed. Formaldehyde is a
common contact allergen, and the ethoxylated aldehydes were shown
to have a sensitizing capacity of the same magnitude as formaldehyde.
The instability of the ethoxylated alcohols and formation of oxidation
products may give an allergenic contribution to hand eczema caused
by work with water and surfactants. To investigate the clinical
significance in man an appropriate diagnostic patch testing in exposed
humans is required.
free radicals present or catalyzed by metal salts, e.g. copper
sulfate.6 Peroxides and hydroperoxides are the primary
oxidation products followed by formation of carbonyl
compounds as secondary oxidation products.7
The prevalence of contact allergy in the general popula-
tion in Europe is about 10%. Of those sensitized, about
2-4% have ongoing allergic contact dermatitis, which is
8
the consequence of exposure to environmental chemicals.
About 90% of occupational contact dermatitis is located on
the hands, and half of all work-related hand eczemas are
9
caused by work with surfactants and water. Most diag-
noses of contact dermatitis from wet work are considered
to be irritant dermatitis.10 The diagnosis of allergy is
difficult to exclude from irritancy in cases of chronic
dermatitis. Surfactants are irritants, partially due to their
ability to solubilize lipid membranes, since they possess
1
0
both lipophilic and hydrophilic regions in their structures.
In our recent studies we found that the oxidation products
identified after air exposure of ethoxylated surfactants had
allergenic properties. Formaldehyde is a well-known con-
tact allergen, and the ethoxylated aldehydes (Figure 1)
were shown to be contact allergens in experimental sen-
3
sitization studies. In the literature some cases are reported
Introduction
of allergic contact dermatitis due to ethoxylated nonionic
surfactants and emulsifiers.1
1,12
Formaldehyde is described
The ethoxylated surfactants have wide applications and
are used in, for example, household cleaners, laundry
products, pharmaceuticals, and industrial and institutional
cleaners. In 1993 the total consumption of ethoxylated
alcohols was estimated to about 313000 tons in Western
to be a significant allergen in women with hand eczema
1
3
caused by occupational and domestic exposure. Thus,
allergenic oxidation products in ethoxylated surfactants
may cause hand eczema or aggravate an ongoing irritant
dermatitis in wet work.
1
Europe. The number of oxyethylene groups in nonionic
In our previous studies on the identification of oxidation
products formed from ethoxylated alcohols we have for the
first time shown the formation of ethoxylated aldehydes
ethoxylated surfactants determines application behavior,
e.g. detergency, emulsification, and wetting, at a given
temperature in all formulation work.
3
with allergenic properties. In the present study we have
In recent studies2
-4
we have shown that oxidation
further investigated the autoxidation of the pure ethoxy-
products were rapidly formed from ethoxylated nonionic
surfactants during storage and handling at room temper-
ature in daylight and also during storage in dark. We
lated dodecyl alcohol, C12
2 2 5
H25(OCH CH ) OH (below re-
ferred to as C12 ) by gas chromatography (GC) analysis
E
5
during one year. We have identified a new group of
oxidation products formed, the ethoxylated formates, and
studied their allergenic activity.
detected peroxides,2 formaldehyde,
,4
2,4
and a series of
ethoxylated aldehydes3 among the oxidation products in
2
our studies on Tween 80 (sorbitan monooleate) and
ethoxylated fatty alcohols.3 The ethoxylated surfactants
,4
have so far been considered to be stable products at normal
storage and handling. The products are usually stored at
Experimental Section
5
room temperature, since they become semisolid at lower
temperatures. However, ethoxylated surfactants are poly-
ethers and as such susceptible to oxidation at air exposure.
This autoxidation is theoretically discussed in the surfac-
Ch em ica lssTetraethylene glycol (99%) was obtained from
Aldrich (Steinheim, Germany). Triethylene glycol (99%), formic
acid (98-100%), 1-dodecanol (98%), 1-bromododecane, p-toluene-
sulfonic acid (99%), and dimethyl sulfoxide (DMSO) were obtained
from Kebo Lab AB (Stockholm, Sweden). Sodium hydride (60%
dispersion in mineral oil, toluene-soluble bags) was obtained from
Acros Chimica N.V. (Geel, Belgium). Triethylene glycol mono
n-dodecyl ether C12H25(OCH2CH2)3OH (CAS Reg. no. 3055-94-5)
(referred to as C12E3) and pentaethylene glycol mono n-dodecyl
6
tant literature. The proposed mechanism for autoxidation
*
Corresponding author. Phone: + 46 8 730 9919. Fax: + 46 8 730
9
892. e-mail: mbergh@niwl.se.
©
1999, American Chemical Society and
10.1021/js980210y CCC: $18.00
Published on Web 02/17/1999
Journal of Pharmaceutical Sciences / 483
American Pharmaceutical Association
Vol. 88, No. 4, April 1999

C12H25(OCH2CH2)2OH (2). Yield: 59%. FT-IR (neat): 3467
O-H), 2850 and 2950 cm (C-H). 1H NMR (CDCl3): δ 3.78-
.65 (m, 8H, (CH2O) 4), 3.47 (tr, 2H, CH2CH2CH2O), 2.85 (s, 1H,
OH), 1.58 (m, 2H, CH2CH2CH2O), 1.24 (m, 18H, (CH2) 9), 0.87 (tr,
H, CH3). 13C NMR (CDCl3): δ 72.49, 71.56, 70.40, 70.08 (CH2O)
, 61.73 (CH2OH), 31.86 (CH3CH2CH2), 29.58 (2 C:s), 29.58 (3 C:s),
-1
(
3
3
4
2
9.42, 29.29 ((CH2)7), 25.99 (CH2CH2CH2O), 22.62 (CH3CH2), 14.05
+
+
+
(
(
(
CH3). GC-MS-PCI m/z (% rel inten): 275 [M + 1] (53), 274 M
+
+
3), 273 [M - 1] (19), 166 [C12H22] (23), 107 [HO(CH2CH2O)2H2]
100).
Figure 1s(a) Ethoxylated alcohol (pentaethylene glycol mono n-dodecyl ether)
. (b) The earlier identified ethoxylated aldehydes in air exposed C12
c) In the present study the identification and sensitizing capacity of the
ethoxylated formates are presented.
C12H25(OCH2CH2)4OH (3). Yield: 57%. FT-IR (neat): 3374
(O-H), 2847 and 2948 cm (C-H). 1H NMR (CDCl3): δ 3.63-
3.56 (m, 16H, (CH2O) 8), 3.40 (tr, 2H, CH2CH2CH2O), 2.74 (s, 1H,
OH) 1.55 (m, 2H, CH2CH2CH2O), 1.22 (m, 18H, (CH2)9), 0.80 (tr,
3
7
2
-1
C
(
12
E
5
5
E .
H, CH3). 13C NMR (CDCl3): δ 72.46, 71.45, 70.51, 70.48, 70.48,
0.46, 70.24, 69.93 (CH2O)8, 61.57 (CH2OH), 31.59 (CH3CH2CH2),
9.30 (5C:s), 29.15, 29.01 (CH2)7, 25.73 (CH2CH2CH2O), 22.30
ether C12H25(OCH2CH2)5OH (CAS Reg. no. 3055-95-6) (referred
to as C12E5) were purchased from Nikko Chemicals CO., Ltd.
(
1
CH3CH2), 13.69 (CH3). GC-MS-PCI m/z (% rel inten): 363 [M +
(Tokyo, J apan). The purity was stated to be 98% by the producer,
+
+
+
] (100), 362 M (9), 361 [M - 1] (61), 195 [HO(CH2CH2O)3-
which was confirmed with GC analysis. The other ethoxylated
alcohols were synthesized as described below. Standard chemicals
were of p.a. quality.
+
+
+
CH2CH2OH2] (54), 177 [(CH2CH2O)4H] (12), 166 [C12H22] (8),
33 [(CH2CH2O)3H] (16), 89 [(CH2CH2O)2H] (17), 45 [CH2CH2-
OH] (28).
+
+
1
+
In str u m en ta tion a n d Mod e of An a lysissFT-IR spectra were
recorded with a Perkin-Elmer 16 PC FT-IR instrument using a
sealed liquid cell with KBr windows. NMR spectroscopy was
performed on a J EOL EX 270 instrument in CDCl3 using tetram-
ethylsilane as internal standard. GC analyses were carried out
on a Hewlett-Packard HP 5890 gas chromatograph with a flame
ionization detector (FID). The gas chromatograph was equipped
with a fused silica capillary column (25 m × 0.22 mm i.d) coated
with 0.2 µm CP-sil 8 CB (Chrompack, Middelburg, The Nether-
lands). Nitrogen was used as carrier gas at a linear gas velocity
of 22 cm/s. An ELDS laboratory data system from Chromatography
Data System Inc. (Svartsj o¨ , Sweden) was used for registration and
processing of the detector signal.
Mass spectrometric (MS) analyses were performed on a Finni-
gan Incos 50 quadrapole instrument equipped with a Varian 3400
gas chromatograph with an on-column injector. The MS analyses
were carried out in electron impact (EI) and positive ion chemical
ionization (PCI) modes. Introduction of the sample into the ion
source was made via GC using on-column technique. The gas
chromatograph was equipped with a fused silica capillary column
Preparation of Ethoxylated Formates 4-8sSamples of the
appropriate alcohol 1-3 and C12E3(4 mmol) were heated at 85 °C
in formic acid (10 mL) with p-toluenesulfonic acid as catalyst for
4 h.1
4-16
The reaction mixture was neutralized with a saturated
(10 M) sodium hydroxide solution and washed with water, dried
over MgSO4, and concentrated in a vacuum. The crude product
was chromatographed on a silica gel column eluted with an
increasing content of ethyl acetate 30-70% in dichloromethane
to give the pure ethoxylated formates 4-8 as clear oils in 80-
100% yield. Identification was performed with FT-IR, NMR, and
MS.
C12H25OCHO (4). Yield 100%. FT-IR (neat): 2924 and 2854
-1
1
(
(
2
(
C-H aliphatic), 1732 (CdO), 1180 cm
(C-O).
H NMR
CDCl3): δ 8.03 (s, 1H, OCHO), 4.14 (tr, 2H, CH2OCHO), 1.63 (m,
H, CH2CH2O), 1.29 (m, 18H, (CH2)9), 0.85 (tr, 3H, CH3). C NMR
13
CDCl3): δ 161.35 (OCHO), 64.26 (CH2O), 32.04 (CH2CH2CH3),
2
9.67 (5C:s), 29.47 29.31 ((CH2)7), 25.93 (CH2CH2O), 22.80 (CH2-
+
CH3), 14.24 (CH3). GC-MS-PCI m/z (% rel inten): 215 [M + 1]
+
+
(
2.62), 213 [M - 1] (5.25), 169 [C12H25] (69.2).
(
25 m × 0.25 mm i.d.) coated with 0.2 µm CP-sil 8 CB (Chrompack,
C12H25OCH2CH2OCHO (5). Yield 95%. FT-IR (neat): 2924 and
Middelburg, The Netherlands), and helium was used as carrier
gas. The temperature programming of the gas chromatograph oven
was as follows: 35 °C for 1.0 min followed by a temperature
increase of 10 °C/min up to 295 °C. The GC-MS transferline was
held at 310 °C. The ion source was held at a temperature of 150
-1
1
2
8
3
(
7
2
852 (C-H), 1732 (CdO), 1180 cm (C-O). H NMR (CDCl3): δ
.10 (s, 1H, OCHO), 4.31 (tr, 2H, CH2OCHO), 3.67 (tr, 2H, CH2O),
.47 (tr, 2H, CH2CH2O), 1.58 (m, 2H, CH2CH2O), 1.26 (m, 18H,
13
CH2)9), 0.87 (tr, 3H, CH3). C NMR (CDCl3): δ 161.01 (OCHO),
1.55, 68.21, 63.13 (CH2O)3, 31.90 (CH2CH2CH3), 29.58 (5C:s),
9.43, 29.32 ((CH3)7), 26.02 (CH2CH2O), 22.68 (CH2CH3), 14.24
°C and the electron energy was 70 eV in the EI mode. In PCI mode
the ion source was held at 80 °C, the electron energy was 110 eV,
and the ion source pressure was about 1 Torr. At chemical
ionization, methane of >99.995% purity was used as reagent gas,
and the instrument was tuned by optimizing the reactant ions
+
+
(
(
CH3). GC-MS-PCI m/z (% rel inten): 259 [M + 1] (30.5), 258 M
+
+
0.39), 257 [M - 1] (1.77), 169 [C12H25] (6.32), 73 [CH2CH2-
+
OCHOH] (100).
+
+
+
C12H25(OCH2CH2)2OCHO (6). Yield 92%. FT-IR (neat): 2900
(
CH5 , C2H5 , and C3H5 ) to an approximate ratio of 5:4:1. The
-
1
1
and 2850 (C-H), 1732 (CdO), 1180 cm (C-O). H NMR (CDCl3):
MS scan range in all analyses was m/z 50-600, and the scan cycle
time was 0.6 s.
δ 8.09 (s, 1H, OCHO), 4.33 (tr, 2H, CH2OCHO), 3.75-3.60 (m,
6
1
1
(
2
[
2
7
H, (CH2O)3), 3.52 (tr, 2H, CH2CH2O), 1.57 (m, 2H, CH2CH2O),
.25 (m, 18H, (CH2)9), 0.87 (tr, 3H, CH3). 13C NMR (CDCl3): δ
61.07 (OCHO), 71.73, 70.80, 70.15, 68.98, 63.18 (CH2O)5, 32.04
Syn th esissPreparation of Dodecylethoxylated Alcohols 1-3s
Sodium hydride (0.80 g, 5.5 mmol) was stirred in DMSO (dry, 8
mL) at room temperature for 30 min. The appropriate glycol
H(OCH2CH2)nOH (n ) 1, 2, or 4) (77 mmol) was added slowly,
and the mixture was stirred at room temperature under nitrogen
for 2 h. 1-Bromododecane (5.0 g, 20 mmol) was added dropwise,
and the mixture was heated at 90 °C overnight under nitrogen.
The reaction mixture was dissolved in ethyl acetate and washed
with water. The organic phase was dried over MgSO4 and
evaporated. The crude product was chromatographed on a silica
gel column eluted with an increasing content of ethyl acetate 4-6%
in hexane, followed by 20% methanol in ethyl acetate. The products
CH2CH2CH3), 29.74 (5C:s), 29.59, 29.47 (CH2)7, 26.20 (CH2CH2O),
2.80 (CH2CH3), 14.24 (CH3). GC-MS-PCI m/z (% rel inten): 303
+
+
+
M + 1] (38.0), 302 M (8.45), 301 [M - 1] (17.6), 275 [M + 1 -
+
+
+
8] (5.63), 166 [C12H22] (19.7), 135 [H2(OCH2CH2)2OCHO] (71),
3 [H2OCH2CH2OCHO] (95.8).
+
C12H25(OCH2CH2)3OCHO 7. Yield 90%. FT-IR (neat): 2945
-
1
1
and 2835 (C-H), 1724 (CdO), 1180 cm (C-O). H NMR (CDCl3):
δ 8.09 (s, 1H, OCHO), 4.32 (tr, 2H, OCH2OCHO), 3.75-3.56 (m,
1
1
1
0H, (CH2O)5), 3.44 (tr, 2H, CH2CH2O), 1.55 (m, 2H, CH2CH2O),
.25 (m, 18H, (CH2)9), 0.88 (tr, 3H, CH3). 13C NMR (CDCl3): δ
60.93 (OCHO), 71.51, 70.66, 70.62, 70.55, 70.01, 68.82, 63.00
1
-3 were obtained as clear oils in 57-66% yield and identified
with FT-IR, NMR, and MS.
(
CH2O)7, 31.88 (CH2CH2CH3), 29.58 (5C:s), 29.45, 29.31 ((CH2)7),
C12H25OCH2CH2OH (1). Yield: 66%. FT-IR (neat): 3370 (O-
-
1
1
26.06 (CH CH O), 22.64 (CH CH ), 14.09 (CH ). GC-MS-PCI m/z
H), 2850 and 2950 cm (C-H). H NMR (CDCl3): δ 3.72 (tr, 2H,
CH2O), 3.53 (tr, 2H, CH2O), 3.42 (tr, 2H, CH2CH2CH2O), 2.11 (s,
2
2
2
3
3
+
+
+
(
(
% rel inten): 347 [M + 1] (29.2), 346 M (2.89), 345 [M - 1]
+
+
13.6), 317 [M + 1 - 28] (4.82), 179 [(CH2CH2O)3H] (49.5), 166
1
H, OH), 1.58 (m, 2H, CH2CH2CH2O), 1.26 (m, 18H, (CH2)9) 0.85
+
+
1
3
22 2 2
[C12H ] (21.91), 73 [CH CH OCHO] (100).
(tr, 3H, CH3). C NMR (CDCl3): δ 71.68, 71.43 ((CH2O)2), 61.85
(CH2OH), 31.90 (CH3CH2CH2), 29.63 (5 C:s), 29.47, 29.33 ((CH2)7),
C12H25(OCH2CH2)4OCHO (8). Yield 80% FT-IR (neat): 2942
-1
1
2
6.09 (CH2CH2O), 22.66 (CH3CH2), 14.05 (CH3). GC-MS-PCI m/z
and 2845 (C-H), 1732 (CdO), 1178 cm (C-O). H NMR (CDCl3):
δ 8.10 (s, 1H, OCHO), 4.33 (tr, 2H, CH2OCHO), 3.76-3.60 (m, 14
H, (CH2O)7), 3.45 (tr, 2H, CH2CH2O), 1.58 (m, 2H, CH2CH2O), 1.26
+
+
+
(% rel inten): 231 [M + 1] (17), 230 M (2), 229 [M - 1] (17),
+
+
1
69 [C12H25] (16), 63 [HOCH2CH2OH2] (100).
4
84 / Journal of Pharmaceutical Sciences
Vol. 88, No. 4, April 1999

(
(
(
m, 18H, (CH2)9), 0.89 (tr, 3H, CH3). 13C NMR (CDCl3): δ 160.94
OCHO), 71.52, 70.58 (5C, s), 70.10, 68.81, 63.00 (CH2O)9, 31.88
CH2CH2CH3), 29.55 (5C:s), 29.45, 29.31 (CH2)7, 26.04 (CH2CH2O),
Table 1sSensitizing Potential of Compound 5 and Cross-Reactivity
Studies with C12 and an Ethoxylated Aldehyde in Guinea Pigs Using
the Modified CCET Method without Adjuvant
E
5
2
2.64 (CH2CH3), 14.07 (CH3). GC-MS-PCI m/z (% rel inten): 391
+
+
+
no. of animals with positive reaction after exposurea
(% w/w in water)
[
M + 1] (8.12), 390 M (1.63), 389 [M - 1] (11.8), 363 [M + 1 -
+
+
+
2
8] (90.2), 223 [H2(OCH2CH2)4OCHO] (36.3), 166 [C12H22]
5
24.0), 133 [(CH2CH2O)2H] (17.8), 73 [CH2CH2OCHO]+ (100).
+
C12E5b aldehydec
(
guinea pigs
10
5
1
5
1
water
St or a ge a n d H a n d lin g of E t h oxyla t ed Alcoh olssTwo
samples of undiluted C12E5 (98%) were used in the experiment.
Sample 1 (5 g) was stored in a closed 10 mL vessel in darkness at
room temperature (20-22 °C) for 12 months. Sample 2 (5 g) was
stirred gently in an open 10 mL Erlenmeyer flask in daylight at
room temperature (20-22 °C) for 1 h, 4 times a day, during 12
months, mimicking what we considered normal handling in
laboratories and industries. The top of the flask was covered with
aluminum foil to prevent contamination and to diminish the
evaporation.
Group 1d
exposed (n ) 15)
4
8 h
2 h
6 h
0
0
0
0
0
0
0
0
0
1a
2
1
1
2
1
0
0
0
7
9
Group 2
controls (n ) 15)
48 h
0
0
0
0
1
0
0
0
0
0
1
1
1
1
0
1
1
1
7
2 h
Detection of Oxidation P r odu cts in Eth oxylated Alcoh olss
Samples 1 and 2 were analyzed with GC-MS analysis using the
synthesized references.
96 h
a
The figures are the number of animals with confluent erythema 48, 72,
Samples 1 and 2 were also analyzed with GC every fourth week
after start of the exposure. The content of the ethoxylated formates
in samples 1 and 2 was quantified using the synthesized reference
compounds. Aliquots of 2 × 10 mg were taken out from each
sample. Two sample preparations (1.0 mg/mL) from each sample
were prepared and dissolved in dichloromethane, methyl stearate
was added as internal standard, and a duplicate analysis on each
sample was performed. On-column injections (1 µL) were made
at an injector temperature of 35 °C. The column oven was kept at
b
12 5
and 96 h after application of the test material. Ethoxylated alcohol, C E ,
5% (w/w in water). c Ethoxylated aldehyde with four ethylenoxide groups,
d
C H (OCH CH ) OCH OCHO, 1% (w/w in water). Induction: 10% (w/w in
12
25
2
2 4
2
water) of 5 in water.
Results
Sp ectr a l Ch a r a cter isticssIn FT-IR the specific hy-
droxyl resonance of the synthesized alcohols 1-3 was
3
5 °C for 2 min whereafter the temperature of the column was
-1
observed at 3370-3467 cm . The FT-IR spectra of the
raised with a rate of 10 °C/min to 210 °C. The column temperature
was then raised with a rate of 5 °C/min to a final value of 240 °C
which was kept for 10 min.
synthesized ethoxylated formates 4-8 had a specific car-
-
1
bonyl resonance at 1724-1732 cm and the C-O reso-
-
1
15
nance at 1180 cm
from the ester group. The NMR
Stu d ies on th e Sen sitizin g Ca p a city of 5sThe sensitization
experiment was performed using female Dunkin-Hartley guinea
pigs (weight 250-300 g) from AB Sahlins F o¨ rs o¨ ksdjursfarm,
Malm o¨ , Sweden. The animals were kept on a standard diet from
Beekey, North Humberside, England, and water ad libitum. The
animals were randomly assigned to one exposed, group 1 (n ) 15),
and one control group, group 2 (n ) 15).
signals were compared with literature data for poly-
oxyethylene) alcohols2
0,21
and formates and accorded with
16
(
structures 1-3 and 4-8. MS analyses in the GC-EI mode
yielded no molecular ions of the ethoxylated alcohols and
formates. Aliphatic ethers normally exhibit weak molecular
ion peaks.22 In the GC-MS-PCI analyses of the synthesized
+
The sensitization study was performed according to the Cumu-
alcohol ethoxylates, 1-3, the molecular ion M and [M +
1
7
+
+
lative Contact Enhancement Test (CCET) method in a modified
1] and [M - 1] ions were observed. Cleavage of the
1
8,19
form with closed epidermal challenge testing.
At induction the
ethoxylated chain led to fragments of the general formula
+
animals received an occlusive epidermal application on the shaved
upper back on days 0, 2, 7, and 9. About 200 mg of the test material
was applied on pieces of filter paper (4 × 2 cm) at each of the four
inductions. The FCA injections at the third induction were omitted
according to our earlier experience of sensitization studies on
[(CH CH O) H] . These data, together with the FT-IR and
2
2
n
NMR data are consistent with the structures 1-3. In the
GC-MS-PCI analyses of the synthesized formates 4-8 the
molecular ion M and [M + 1] and [M - 1]+ ions were
+
+
observed together with specific fragments with the general
3
surfactants. Challenge testing was performed on day 21 on the
+
+
2 2 n 2 2 n
formula [(CH CH O) H] and [CH CH O) OCHO] . The [M
shaved left flank using Finn Chambers (aluminum chambers, 8
mm i.d from Epitest, Helsinki, Finland) with approximately 15
mg of the test material applied in each chamber.
+
-
1] ion fragment corresponded to R-cleavage. No adducts
+
+
with methane, [(M + C
2
H
5
3 5
) ] and [(M + C H ) ] were seen.
-
4
These data, together with the FT-IR and NMR data are
The exposed group was induced with 5 10% w/w (2.6 × 10
consistent with the structures 4-8, C12
2 2 n
H25(OCH CH ) -
mol/g) in water, while the animals in the control group received
water alone. Both groups were challenged with 10, 5, and 1% w/w
OCHO, n ) 1-4.
-
4
-4
-5
Det ect ion of Oxid a t ion P r od u ct s in E t h oxyla t ed
Alcoh olssThe ethoxylated formates, C12H25(OCH CH ) -
2 2 n
(
2.6 × 10 , 1.3 × 10 , and 2.6 × 10 mol/g) of 5 in water, with
-4
C12E5 5% (1.3 × 10 mol/g) in water, and C12E4OCH2CHO 1%
-
5
w/w (2.5 × 10 mol/g) in water (Table 1). Water was applied as a
vehicle control. The chambers were removed after 24 h, and the
reactions were assessed at 48, 72, and 96 h after start of the
exposure. The minimum criterion for a positive reaction was a
confluent erythema.
OCHO, n ) 1-4, were all detected in samples 1 and 2 of
C E with GC-MS-PCI analyses using 4-8 as reference
1
2
5
compounds. The alcohols, C12
were identified in samples 1 and 2 of C12
PCI analyses using C12
as reference compounds.
H
25(OCH
2
CH
2
)
n
OH (n ) 1-4),
with GC-MS-
E
5
E and the synthesized alcohols 1-3
3
The experiment was performed with the equimolar concentra-
tions for induction and challenge as used in the experiments with
the ethoxylated aldehyde, C12H25(OCH2CH2)4OCH2CHO.3 The
challenge concentrations of 5 were in pretests shown to be
nonirritating in three untreated guinea pigs. Patch testing with
concentrations 20-1% of 5 in water gave no skin reactions after
The amount of the identified formates increased continu-
ously with time. The formates seem to be rapidly formed
since they were detected in small amounts in the GC
analysis already in a newly opened bottle of the pure
ethoxylated alcohol. The limits of detection for the sub-
stances 1-5 in the GC-analysis were estimated to be in
the range 0.001-0.05 ng/µL using a signal-to-noise ratio
:1 (S/N ) 3). The total content of formates in C12
.3% in the sample stored in a closed vessel in darkness
(sample 1) and 4.0% in the sample handled in daylight
(sample 2) after 12 months. The content of the individual
4
8 and 72 h. The experiment was approved by the local ethics
committee.
Sta tistica l An a lysessThe result from the animal experiment
was analyzed with Fisher’s exact test. The number of reactions to
each applied test substance in the exposed animals was compared
with the number of reactions in the control group. A p value <0.05
was statistically significant.
3
3
5
E was
Journal of Pharmaceutical Sciences / 485
Vol. 88, No. 4, April 1999

Figure 3sGC chromatogram showing the separation of different degradation
products formed at air exposure of C12 . The peak from the original
ethoxylated alcohol is assigned C12 , peaks A−E are the homologue series
of ethoxylated formates, C12 25(OCH CH OCHO n ) 0−4, eluting in the
order of boiling point; peaks F−I are the homologue series of ethoxylated
alcohols, C12 25(OCH CH OH n ) 1−4, eluting in the order of boiling point;
peaks J−N are the homologue series of ethoxylated aldehydes, C12 25(OCH
CH OCH CHO (n ) 0−4), eluting in the order of boiling point. The peak
assigned * is the internal standard, methyl stearate, with a retention time of
8.63 min.
E
5
E
5
H
2
2 n
)
H
2
2 n
)
H
2
-
2
)
n
2
1
a number of degradation products as illustrated by the gas
chromatogram in Figure 3. Among the oxidation products
identified, some have significant allergenic properties. So
far, nonionic ethoxylated surfactants have been regarded
as stable products and are normally stored at room
temperature, since they become semisolid at lower tem-
5
peratures. The practical consequence will be that a product
containing ethoxylated alcohols can have quite a different
chemical composition after storage and handling compared
to the original product. In addition to an increase of their
3
harmful effects on skin with time, there might also be a
change in their surface active properties.25 To avoid
significant decomposition, the pure ethoxylated alcohols
must be stored in the refrigerator (8 °C), which was
demonstrated in a previous study.4
Figure 2sThe content (%) of the individual ethoxylated formates 4−8 at
different time points in C12 : (a) sample 1 stored in a closed vessel in
darkness at room temperature; (b) sample 2 stored and handled in daylight
at room temperature.
E
5
To the best of our knowledge, ethoxylated formates are
described here for the first time. The formates, C12
CH OCHO, n ) 0-4, were identified and their structures
elucidated in the oxidation mixture of C12 . Prior to the
identification work, the dodecyl poly(oxyethylene) formates
that theoretically might be formed were synthesized and
used as reference compounds in the analyses. The major
2
H25(OCH -
formates in samples 1 and 2 is shown in Figure 2. Formate
2 n
)
4
, C12H25OCHO, was formed in the highest concentration.
E
5
All calculations were performed relative to the internal
standard. The coefficient of variation (CV) was below 10%
at repeated measurements (n ) 10). Duplicate analyses
were performed of each sample. Approximately 5% of the
total sample volume (5 g) was used for analysis, which was
not regarded to contribute to the degradation. Figure 3
shows the GC separation of oxidation products in sample
formate, C12H25OCHO (4), is apparently formed due to
cleavage between the poly(oxyethylene) groups and the
3
alkyl chain. The ethoxylated aldehydes earlier identified
are formed by cleavage of the poly(oxyethylene) chain
resulting in loss of oxyethylene units and also by oxidation
of the terminal hydroxyl group yielding the dominant
ethoxylated aldehyde, C12H25(OCH CH ) OCH CHO. This
2 2 4 2
indicates that the autoxidation proceeds with parallel and
different mechanisms.
2
air exposed for 12 months. The oxidation products that
we have identified are assigned in the chromatogram with
retention times corresponding to homologue retention
characteristics accordning to boiling point.
Sen sitizin g Ca p a city of 5sNo sensitizing response
was observed to 5 in the animal experiment (Table 1). No
cross-reactivity was observed to the corresponding alcohol,
The amount of ethoxylated formates increased with time
of air exposure as determined with GC analysis. Different
compounds are predominant in the degradation mixture
depending on the time of air exposure. In our previous
studies2 the formation of peroxides, formaldehyde and
ethoxylated aldehydes was shown. The amount of peroxides
was determined with the unspecific iodometric titration
method23 whereas the aldehydes were quantified using GC-
and LC methods. Peroxides were initially formed followed
by formation of formaldehyde and the ethoxylated alde-
C
12
E
5
, or to the ethoxylated aldehyde, C12
2 2 4
H25(OCH CH ) -
OCH
2
CHO. Some irritation was seen when the animals
were tested with the alcohol and the aldehyde.
-4
Discussion
Air exposure during storage and handling of the pure
ethoxylated alcohol C12
5
E at room temperature results in
4
86 / Journal of Pharmaceutical Sciences
Vol. 88, No. 4, April 1999

hydes as secondary oxidation products. There was a
decrease in the content of peroxides after a certain time
during the oxidation process, while the content of aldehydes
content of the identified oxidation products constitutes less
than 10% of the total content in the product after 6 months.
This indicates that a major part of the oxidation and
2
-4
steadily increased. The ethoxylated alcohols, C12
H
25(OCH
2
-
5
degradation products from C12E still remains to be identi-
CH OH, (n ) 1-4) that we identified in the oxidation
2
)
n
fied in the complex oxidation mixture. Our data indicates
that the composition of ethoxylated alcohols may change
rather rapidly upon storage. It has not been a topic of this
study to investigate this in detail, but our data support
that a stability study is requested.
mixture with GC-MS may be formed by decomposition of
the corresponding peroxides or hydroperoxides, since alkyl
hydroperoxides are reported to decompose to alcohols at
temperatures above 90 °C.24 The boiling point of C12
5
E is
in the range of 202-216 °C. Therefore, the commonly used
GC technique involves injection via a split/splitless injector
at a temperature of 280 °C. Despite the use of on-column
injection technique in this study to avoid degradation of
thermally unstable compounds at injection, the hydroper-
oxides were not detected with GC-MS. The hydroperoxides
could have decomposed in the column when the column
temperature increased during the analysis. To identify
ethoxylated hydroperoxides alternative methods have to
be employed.
It is important to further investigate the skin effects of
the widely used ethoxylated surfactants, since a majority
of the cases of occupational dermatits is caused by work
with water and surfactants. Various types of ethoxylated
surfactants are also used as emulsifiers in creams and
lotions used on the skin. The clinical significance to man
will require an appropriate diagnostic patch testing in
exposed humans. The sensitizing capacity of other oxida-
tion/degradation products will be studied and also the
influence of oxidation on the skin-irritating properties.
In the predictive sensitization studies, Freund’s complete
adjuvant (FCA) is often used. FCA consists of Arlacel
(mannide mono-oleate), a sorbitan emulsifier to promote a
References and Notes
stable water-in-oil emulsion, and dried heat-killed Myco-
bacterium tuberculosis organisms to enhance the nonspe-
1
2
. Modler, F.; Willhalm, R.; Yoshida, Y. Surfactants, household
detergents and their raw material. In CEH (Chemical
Economics Handbook); Stanford Research Institute: Palo
Alto, CA, 1994; pp 583.8001A-583.0003J .
. Bergh, M.; Magnusson, M.; Nilsson, J . L. G.; Karlberg, A.-T.
Contact allergenic activity of Tween 80 before and after air
exposure. Contact Dermatitis 1997, 37, 9-18.
cific immune response and raise the sensitivity of the
method.26 However, also an increased irritability of the skin
may be seen. Determination of the sensitizing potential of
chemicals, such as surfactants, with irritating properties
is difficult. The challenge testing should be performed with
the chemical in nonirritating concentrations which might
then be too low to detect an allergenic effect. In our
investigation of the sensitizing capacity of the ethoxylated
aldehydes3 two parallel CCET protocols were used, one
with intradermal FCA injections and one without. No
increased allergenic activity was observed when FCA was
administered, but an increased irritation was initially seen
in the FCA-treated controls compared to the non-FCA-
treated. A dose-response relationship was seen in both
experiments. Thus, in our subsequent experiments with
3. Bergh, M.; Shao, L. P.; G a¨ fvert, E.; Nilsson, J . L. G.; Karlberg,
A.-T. Contact allergens from surfactants. Atmospheric
oxidation of polyoxyethylene alcohols, formation of ethoxy-
lated aldehydes and their allergenic activity. J . Pharm. Sci.
1
998, 87, 276-282.
4. Bergh, M.; Magnussson, K.; Nilsson J . L. G.; Karlberg, A.-T.
Formation of formaldehyde and peroxides by air oxidation
of high purity polyoxyethylene surfactants. Contact Derma-
titis 1998, 39, 14-20.
5
. Evetts, S.; Kovalski, C.; Levin, M.; Stafford, M. High-
temperature stability of alcohol ethoxylates. J OACS 1995,
72, 811-816.
6
. Donbrow, M. Stability of the polyoxyethylene chain. In
Surfactant Science Series volume 23, Nonionic surfactants:
Physical chemistry; Schick, M. J ., Ed.; Marcel Dekker, Inc.:
New York, 1987; pp 1011-1067.
4
ethoxylated alcohols and formates no FCA treatment was
performed. In these experiments the compounds showed
no significant allergenic activity (Table 1). Some irritation
was seen showing the difficulties when testing with ir-
ritants. However, irritation in some of the control animals
can also be observed in sensitization studies of other
substances.17
Since alcohols and esters are rare sensitizers, the
absence of a sensitizing capacity for the ethoxylated
formate was not surprising.27 To cause sensitization, a
compound (hapten) has to penetrate the skin and react
with macromolecules in the skin to form a complete antigen
that is recognized as foreign. Molecules which can function
as haptens in allergic contact dermatitis are either elec-
trophilic or have structures that can easily form radicals.27
Such molecules are able to react with the nucleophilic skin
proteins forming stable covalent bonds. In our studies,
7. Decker, C.; Manchal, J. Polyoxy e´ thyl e` ne: produits d’oxidation
et sch e´ ma cin e´ tique. Makromol. Chem. 1973, 66, 155-178.
8
. Hypersensitivity induced by chemicals: Report on a WHO
consultation. In Allergic hypersensitivities induced by chemi-
cals: recommendations for prevention. Vos, J . G., Younes,
M., Smith, E., Eds.; CRC Press: Boca Raton, FL, 1996; pp
3
-18.
9
. Rycroft, R. J . G. Occupational contact dermatitis. In Textbook
of Contact Dermatitis, 2nd ed.; Rycroft, R. J . G., Menn e´ , T.,
Frosch, P. J ., Eds.; Springer-Verlag: Berlin, 1995; pp 343-
357.
1
1
1
0. Effendy, I.; Maibach, H. I. Surfactants and experimental
irritant contact dermatitis. Contact Dermatitis 1995, 33,
2
17-225.
1. Dooms-Goossens, A.; Blockeel, I. Allergic contact dermatitis
and photoallergic contact dermatitis due to soaps and
detergents. Clin. Dermatol. 1996, 14, 67-76.
2. Pasche-Koo, F.; Piletta, P.-A.; Hunziker, N.; Hauser, C. High
sensitization rate to emulsifiers in patients with chronic leg
ulcers. Contact Dermatitis 1994, 31, 226-228.
three ethoxylated compounds have so far been tested for
4
their contact allergenic properties, i.e., C12
E
5
, C12
H
25(OCH
2
-
13. Cronin; E. Formaldehyde is a significant allergen in women
3
with hand eczema. Contact Dermatitis 1991, 25, 276-282.
2 4 2 2 2 4
CH ) OCH CHO and C12H25(OCH CH ) OCHO. The alco-
1
1
4. Dymicky, M. A general method for the preparation of
formates. Org. Prep. Proc. Int. 1982, 14, 177-181.
5. Martinez, J .; Laur, J . Active esters of formic acid as useful
formylating agents: Improvements in the synthesis of formyl-
amino acid esters, N-R-Formyl-Met-Leu-Phe-OH, and Formyl-
Met- Lys-Pro-Arg, a phagocytosis stimulating peptide. Syn-
thesis 1982, 11, 979-981.
hol has no electrophilic properties, and contact allergenic
4
activity could not be observed. The formate has a very low
reactivity as electrophile, whereas the aldehyde is an
apparent electrophilic group. In our tests, only the alde-
hyde3 showed contact allergenic activity which is in ac-
cordance with its electrophilic properties.
1
1
6. Bhart, I.; William, R.; Gelbaum L. T. Methanol-formic acid
esterification equilibrium in sulfuric acid solution: Influence
of Sodium Salts. Ind. Eng. Chem. Res. 1993, 32, 981-985.
7. Tsuchyia, S.; Kondo, M.; Okamoto, K.; Takase, Y. The
cumulative contact enhancement test. In Contact Allergy Test
in Guinea Pigs; Andersen, K. E., Maibach, H. I., Eds.;
Karger: Basel, 1985; pp 208-219.
In a sample of C12
5
E stored at room temperature only
7
0% of the original product remained after 6 months.
Similar analysis after 12 months storage was not per-
formed. The results in this study together with earlier
investigations of the autoxidation of C12E show that the
5
Journal of Pharmaceutical Sciences / 487
Vol. 88, No. 4, April 1999

1
8. Boman, A.; Karlberg A-T.; Wahlberg, J . E. Experiences with
Freund’s complete adjuvant test (FCAT) when screening for
contact allergens in colophony. Contact Dermatitis 1988, 18,
25. Blute, I.; Svensson, M.; Holmberg, K.; Bergh, M.; Karlberg,
A.-T. Solution behavior of a surfactant aldehyde - the
oxidation product of an alcohol ethoxylate. Colloids Surf., A,
in press.
2
5-29.
1
2
9. Karlberg, A.-T.; G a¨ fvert, E.; Hagelthorn, G.; Nilsson, J . L.
G. Maleopimaric acid - a potent sensitizer in modified rosin.
Contact Dermatitis 1990, 22, 193-201.
2
6. Herbert, W. J . Handbook of Experimental Immunology. In
Application of Immunological Methods; Weir, D. M., Ed.;
Blackwell Scientific Publications: Cambridge, MA, 1979; Vol.
0. Ribeiro, A. A.; Dennis, E. A. Motion in nonionic micelles and
mixed micelles with phospholipids- C-13 spin-lattice relax-
ation study on p- tert-octylphenylpolyoxyethylene ethers. J .
Phys. Chem. 1976, 80, 1746-1753.
3
, pp A 3.1-A 3.12.
2
7. Roberts, D. W.; Lepoittevin, J .-P. Hapten-protein interac-
tions. In Allergic contact dermatitis. The molecular basis;
Lepoittevin, J .-P., Basketter, D. A., Goossens, A., Karlberg
A-T., Eds.; Springer-Verlag: Berlin, 1998; pp 81-111.
2
2
1. Ribeiro, A. A.; Dennis, E. A. A carbon -13 and proton nuclear
magnetic resonance study on the structure and mobility of
nonionic alkyl polyoxyethylene ether micelles. J . Phys. Chem.
1
977, 81, 957-963.
2. McLafferty, F. W.; Turecek, F. Detailed mechanisms of ion
fragmentation. Interpretation of mass spectra; University
Science Books: Mill Valley: CA, 1993; pp 261-264.
Acknowledgments
2
2
3. European Pharmacopoeia III; Council of Europe: 67075
Strasbourg Cedex, ISBN 9287129916; p 2.5.5.
We would like to thank Gunnel Hagelthorn for her skillful
assistance with the animal testing. Financial support from the
Swedish Council for Work Life Research is gratefully acknowledged.
4. Foglia, A.; Sibert, L. S.; Vail, P. D. Gas-liquid and high-
performance liguid chromatographic analysis of aliphatic
hydroperoxides and dialkyl hydroperoxides. J . Chromatogr.
1
993, 637, 157-165.
J S980210Y
4
88 / Journal of Pharmaceutical Sciences
Vol. 88, No. 4, April 1999