Data
Data
KGaA, Weinheim
10.1002/14356007.a20 181.pub2
Phthalic Acid and Derivatives 1
Phthalic Acid and Derivatives
Peter M. Lorz, BASF AG, Ludwigshafen, Germany (Chap. 1, 2, 3 and 4)
Friedrich K. Towae, BASF AG, Ludwigshafen, Germany (Chap. 1, 2, 3 and 4)
Walter Enke, BASF AG, Ludwigshafen, Germany (Chap. 5)
Rudolf J ackh, BASF AG, Ludwigshafen, Germany (Chap. 6)
Naresh Bhargava, BASF Canada Inc., Cornwall, Ontario, Canada (Chap. 1, 2, 3, 4 and 5)
Wolfgang Hillesheim, GUBDH Consultancy, Schwetzingen, Germany (Chap. 6)
Related Articles Isophthalic acid is treated under Terephthalic Acid, Dimethyl Terephthalate, and Isophthalic
Acid
1. Phthalic Acid . . . . . . . . . . . . . 2
2. Phthalic Anhydride . . . . . . . . . 2
2.1. Physical Properties . . . . . . . . . . 2
2.2. Chemical Properties . . . . . . . . . 2
2.3. Resources and Raw Materials . . 3
2.4. Production . . . . . . . . . . . . . . . 3
2.4.1. Gas-Phase Oxidation . . . . . . . . . 3
2.4.1.1. Catalyst and Reaction Mechanism . 4
2.4.1.2. Apparatus and Important Process
Steps in the Gas-Phase Oxidation of
o-Xylene . . . . . . . . . . . . . . . . . 6
2.4.2. Fluidized-Bed Oxidation . . . . . . . 8
2.4.3. Liquid-Phase Oxidation of o-Xylene 9
2.5. Environmental Protection . . . . . 9
2.6. Quality Specications
and Analysis . . . . . . . . . . . . . . 9
2.7. Economic Aspects . . . . . . . . . . 10
2.8. Storage and Transportation . . . . 10
2.9. Uses . . . . . . . . . . . . . . . . . . . 10
3. Phthalimide . . . . . . . . . . . . . . 10
3.1. Properties . . . . . . . . . . . . . . . . 10
3.2. Production . . . . . . . . . . . . . . . 11
3.2.1. Production from Phthalic
Anhydride and Ammonia . . . . . . 11
3.2.2. Production from Phthalic
Anhydride and Urea . . . . . . . . . . 11
3.2.3. Production from o-Xylene . . . . . . 12
3.3. Uses . . . . . . . . . . . . . . . . . . . 12
4. Phthalonitrile . . . . . . . . . . . . . 12
4.1. Properties . . . . . . . . . . . . . . . . 12
4.2. Production . . . . . . . . . . . . . . . 12
4.2.1. Production from o-Xylene . . . . . . 12
4.2.2. Production from Phthalic
Acid Derivatives . . . . . . . . . . . . 13
4.3. Uses . . . . . . . . . . . . . . . . . . . 14
5. Phthalates . . . . . . . . . . . . . . . 14
5.1. Physical and Chemical Properties 14
5.2. Raw Materials . . . . . . . . . . . . . 14
5.3. Production . . . . . . . . . . . . . . . 14
5.4. Environmental Protection . . . . . 17
5.5. Quality Specications . . . . . . . . 17
5.6. Storage and Transportation . . . . 18
5.7. Uses . . . . . . . . . . . . . . . . . . . 18
5.8. Economic Aspects . . . . . . . . . . 18
6. Toxicology . . . . . . . . . . . . . . . 19
6.1. Use of and Exposure to Phthalic
Acid and Derivatives . . . . . . . . 19
6.2. Toxicological Proles . . . . . . . . 19
6.2.1. Phthalic Acid . . . . . . . . . . . . . . 19
6.2.2. Phthalic Anhydride . . . . . . . . . . 19
6.2.3. Phthalimide . . . . . . . . . . . . . . . 20
6.2.4. Phthalonitrile . . . . . . . . . . . . . . 20
6.2.5. Phthalate Esters . . . . . . . . . . . . 21
6.2.5.1. Metabolism and Toxicokinetics . . . 21
6.2.5.2. Acute Toxicity . . . . . . . . . . . . . 24
6.2.5.3. Irritation and Sensitizing Potential . 25
6.2.5.4. Repeated DEHP Dosing . . . . . . . 25
6.2.5.5. Genotoxicity and Mutagenicity . . . 26
6.2.5.6. Carcinogenicity . . . . . . . . . . . . 27
6.2.5.7. Reproductive Toxicity . . . . . . . . 28
6.2.5.8. Effects of Phthalate Esters
by Groups . . . . . . . . . . . . . . . . 33
6.3. Risk Assessment . . . . . . . . . . . 36
6.3.1. Biomonitoring and Human
Exposure to Phthalate Esters . . . . 37
6.3.2. Carcinogenicity . . . . . . . . . . . . 38
6.3.3. Toxicity to Reproduction . . . . . . . 39
6.4. Risk Management . . . . . . . . . . 39
7. References . . . . . . . . . . . . . . . 40
2 Phthalic Acid and Derivatives
1. Phthalic Acid
Phthalic acid [88-99-3], 1,2-benzenedicarboxyl-
ic acid, has remained unimportant industrially.
It is formed as a byproduct in the manufacture
of phthalic anhydride.
Physical Properties. Phthalic acid,
C
8
H
6
O
4
, M
r
166.14, forms colorless, mono-
clinic crystals which melt at 191
C (sealed
tube) and are converted into phthalic anhydride
with the elimination of water at 210211
C.
Some physical properties of phthalic acid are as
follows [1 3]:
Density (15
C) 1.593 g/cm
3
Heat of fusion 315.3 J/g
Specic heat of solid (099
C) 1.214 J g
1
K
1
Heat of combustion 19 657.03 J/g
Heat of formation 43 714.34 J/g
Heat of solution at 25
C 123.55 J/g
Flash point 168
C
2. Phthalic Anhydride
Phthalic anhydride [85-44-9], isobenzofuran-
1,3-dione, has been commercially produced
continuously since 1872 when BASF developed
the naphthalene oxidation process. It was the
rst anhydride of a dicarboxylic acid to be used
commercially and is comparable in its impor-
tance to acetic acid.
The most important derivatives of phthalic
anhydride are plasticizers and, to a lesser degree,
polyester resins and dyes.
About 60 years after its discovery in 1836 by
A. Laurent, a more effective commercial pro-
cess for its production was introduced, which
was based on mercury-catalyzed liquid-phase
oxidation of naphthalene.
The breakthrough that led to commercial pro-
duction of a quality product was the develop-
ment of the gas-phase oxidation of naphthalene
or o-xylene in an air streamwith vanadiumoxide
as catalyst [4].
2.1. Physical Properties
Phthalic anhydride, C
8
H
4
O
3
, M
r
148.12, forms
colorless needles or platelets, with a monoclin-
ic or rhombic crystalline form. Some important
physical properties of phthalic anhydride are as
follows [1, 5, 6]:
Density of solid (4
C) 1.527 g/cm
3
Specic vapor density (1013 mbar) 6.61 kg/m
3
Melting point 131.6
C
Boiling point (1013 mbar) 295.1
C
Heat of fusion 159.1 J/g
Heat of combustion 22 160.7 J/g
Heat of formation from naphthalene 12 058 J/g
Heat of formation from o-xylene 8625 J/g
Heat of sublimation 601 J/g
Heat of evaporation 441.7 J/g
Flash point 152
C
Ignition temperature 580
C
Upper limit of ammability (1013 mbar) 10.5 vol %
Lower limit of ammability (1013 mbar) 1.7 vol %
Lower dust explosion limit 25 g/m
3
The density of liquid phthalic anhydride in
the range 140240
C)
0.000905(t/
C)
2
(1)
Table 1 lists the solubility of phthalic anhy-
dride in various solvents.
Table 1. Solubility of phthalic anhydride
Solvent Temperature,
C Solubility, g/100 g
Water 20 1.64
Water 50 1.74
Water 100 19.0
Carbon disulde 20 0.7
Formic acid 20 4.7
Pyridine 20 80
Benzene soluble
Ethanol 20 soluble
Diethyl ether 20 slightly soluble
The reported explosion hazard data of ph-
thalic anhydride in air vary signicantly [4, 5].
Explosions can occur at concentrations below
100 g/m
3
, depending on the impurities present.
Recent incidents in production plants indicate
that phthalic anhydride concentrations exceed-
ing 35 g/m
3
in the reaction product gas are ca-
pable of ignition if heat-transfer salt enters the
reactor due to broken reactor tubes.
2.2. Chemical Properties
As a cyclic anhydride, phthalic anhydride is a
reactive compound, but in addition, the other-
wise very stable aromatic ring is capable of re-
action. Phthalic anhydride reactions which have
Phthalic Acid and Derivatives 3
achieved commercial importance are summa-
rized below. The most important is the reaction
withalcohols or diols togive esters or polyesters.
Unsaturated polyester resins are obtained by
polycondensation in the presence of maleic an-
hydride or fumaric acid.
One or both of the carboxy groups can react
with ammonia to give phthalic monoamide and
phthalimide [85-41-6] or phthalonitrile [91-15-
6].
Phthalein and rhodamine dyes, some of
whichhave beeninproductionfor over 100years
and have not yet lost their importance, are ob-
tained by reaction of phthalic anhydride with
phenols, aminophenols or quinaldine deriva-
tives. The FriedelCrafts reaction of phthalic an-
hydride with benzene derivatives followed by
ring closure to form anthraquinone derivatives
is of importance as a route to Indanthrene dyes.
Much attention has been devoted to the rear-
rangement of dipotassium phthalate to produce
terephthalic acid [100-21-0] [8], but due to tech-
nical problems the process is no longer used.
3,5-Dihydrophthalic acid can be produced by
electrochemical hydrogenation of phthalic an-
hydride [4]. Hydrogenation with a nickel cata-
lyst produces phthalide [529-20-4].
2.3. Resources and Raw Materials
In the 1960s a fundamental shift took place in the
raw material base for phthalic anhydride. From
1960 to 1975 the production was switched from
100 % coal-tar naphthalene to about 75 % o-
xylene. In 1991, 85 %of phthalic anhydride was
produced from o-xylene [9]. Naphthalene de-
rived frompetroleum, which is produced mainly
in the United States, has not gained importance
as a raw material for phthalic anhydride.
The changeover to o-xylene as the feedstock
was inevitable because the quantities of naph-
thalene derived from coal tar depend on the pro-
duction of coke and these were unable to keep
pace with the increasing demand for naphtha-
lene. On the other hand, naphthalene is an in-
evitable byproduct of coke production which
should be used commercially; therefore, the ph-
thalic anhydride process will continue to use
some naphthalene as feedstock in the future. o-
Xylene, which can readily be separated fromthe
mixture of xylenes containing roughly one third
o-xylene and two thirds p-xylene, is nowadays
available in adequate quantities from cracking
plants and reneries. However, in the past, vari-
ations in the demand for p-xylene have often
affected the availability and price of o-xylene,
as has the need to increase the amount of alkyl
aromatics in unleaded gasoline.
These facts and the need to be able to react
to varying raw material prices have resulted in
plants being planned which are capable of pro-
cessing o-xylene or naphthalene or mixtures of
the two with tailor-made catalysts.
2.4. Production
2.4.1. Gas-Phase Oxidation
Phthalic anhydride is predominantly produced
on an industrial scale by gas-phase oxidation of
o-xylene or naphthalene.
General Features. Preheated o-xylene is in-
troduced into a stream of hot air. The o-xylene
air mixture is passed through a tubular reactor
where the exothermic oxidation takes place on a
highly selective catalyst. The heat of reaction is
used to produce steam, only part of which is uti-
lized in the plant itself. Any excess steam can
be used elsewhere. The gases emerging from
the reactor are precooled. At high loadings of
phthalic anhydride in the product gases, some
liquid phthalic anhydride can be won in a liq-
uid condenser. The product gases are then fed
4 Phthalic Acid and Derivatives
to a switch condenser system where the phthalic
anhydride is condensed on the nned tubes as
a solid. The switch condensers are cooled by a
heat-transfer oil in an automated switching cy-
cle. During the heating cycle the deposited ph-
thalic anhydride is melted and collected in a stor-
age tank.
After the phthalic anhydride has been sepa-
rated, the exhaust gases still contain byproducts
and small quantities of phtalic anhydride and
must be cleaned by scrubbing with water, or cat-
alytically or thermally incinerated. If scrubbing
with water is employed, it is possible to concen-
trate the main byproduct, maleic acid, and the
scrubbing solution can then be processed fur-
ther to yield fumaric acid or maleic anhydride
[10, 11]. If the scrubbing of the exhaust gases is
combined with production of maleic anhydride,
the discharge of any polluted wastewater from
the plant can be avoided.
The crude phthalic anhydride is transferred
to a continuous thermal/chemical treatment sys-
tem, which converts the phthalic acid formed as
a byproduct into the anhydride. The crude prod-
uct is then puried in a continuous two-stage
distillation system [12].
The BASF process can be operated with a
wide range of o-xylene loadings up to 105
g/m
3
(STP). The reactor outlet gases are fur-
ther treated catalytically in a nishing reactor
to decrease the amount of byproducts and to
improve product quality. Flexible operation of
the reactor system enables the phthalic anhy-
dride yield to be optimized and decreases the
amount of residues and volatile organic com-
pounds. Simultaneously a smoother operation
of the catalyst in the tubular reactor prolongs
its life. Addition of SO
2
to activate the catalyst
is not required, and there is no need for a liquid
condenser before collecting the crude phthalic
anhydride in the switch condensers. Because of
the low byproduct content, chemical treatment
of the crude phthalic anhydride is unnecessary.
The plant design is optimized for low energy
consumption and a high net export of steam and
electrictity [13].
The Wacker process makes it possible to
use o-xylene or naphthalene or mixtures of the
two. With o-xylene, loadings of 90100 g/m
3
(STP) are possible [14, 15]. Modications of the
process are aimed at saving energy [16].
The Nippon Shokubai VGR Process. In
patent literature, the Nippon Shokubai VGR
(vent gas recycling) process is described. Its
characteristic feature is, that exhaust gas is re-
cycled and added to the mixture of o-xylene and
process air to reduce the oxygen concentration
to less than 10 vol %. This makes it possible to
work outside the limits of ammability despite
a high o-xylene loading (up to 85 g/m
3
STP).
Despite the low O
2
concentration, yields of up
to 116 g PA/100 g xylene (given as 116 %by PA
producers) have been reported. This is attributed
toa speciallydevelopedcatalyst system. It is also
possible to use naphthalene instead of o-xylene
in this process [17 21].
The AlusuisseFtalital LARProcess. Inthe
LAR (low air ratio) process, o-xylene loadings
in the process air of up to 134 g/m
3
(STP) are
suggested. This couldmake it possible toachieve
an appreciable reduction in energy and the size
of the equipment [22, 23]. In commercial appli-
cation o-xylene loadings up to 80 g/m
3
(STP)
are reached [107]. Catalysts in the form of rings
or half rings are used. The catalyst is capable
of processing both o-xylene and naphthalene or
mixtures of the two [24, 25].
The Rh one-Poulenc process employs o-
xylene as raw material. The crude product is
subjected to a chemical posttreatment before
being puried by two-stage distillation. The
waste gas is incinerated [26, 27].
The ELF Atochem/Nippon Shokubai pro-
cess uses a Nippon Shokubai catalyst for re-
acting o-xylene or naphthalene. With o-xylene,
feed loadings of up to 75 g/m
3
(STP) are used
industrially [107].
2.4.1.1. Catalyst and Reaction Mechanism
The oxidation of o-xylene and naphthalene is
nowadays carried out almost exclusivelyintubu-
lar reactors cooled by a molten salt. The spheri-
cal catalysts previously employed are only used
in isolated cases in old plants. In modern plants
ring-type catalysts have become established for
Phthalic Acid and Derivatives 5
Scheme 1.
energy saving reasons [28], but catalysts on half-
shell supports are also available [24, 25]. Com-
mercial catalysts are normally composed of an
inert ring-type support, made of silicate, silicon
carbide, porcelain, alumina, or quartz measur-
ing 510 mm in diameter, on which a thin ac-
tive layer of a mixture of nely divided titanium
dioxide and vanadium oxide is deposited. Com-
pounds of antimony, rubidium, cesium, niobium,
and phosphorus are added to improve the se-
lectivity [29 35]. The use of a two-zone cat-
alyst, i.e., a combination of catalysts with low
activity with catalysts of higher activity, has be-
come generally established [36]. For high load-
ings even three-zone catalysts are known. Some
catalysts require addition of SO
2
for activation
and longer service life [37, 38]. The reaction
is driven to an almost quantitative turnover of
o-xylene, to ensure low concentrations of dis-
turbing byproducts (e.g., phthalide). It is sum-
marized by the following equation:
The heat of formation is 1108.7 kJ/mol. To-
tal combustion of o-xylene yields 4380 kJ/mol,
i. e., almost four times the heat of formation of
phthalic anhydride. Oxidation in tubular reac-
tors produces a heat of reaction between 1300
and 1800 kJ/mol of o-xylene.
Scheme 1 shows the important byproducts
found in the reaction product. Although the in-
termediate alcohol has so far not been detected,
it is probable that the formation of phthalic an-
hydride passes through it. A standard interpre-
tation of the course of the reaction is not avail-
able in the literature. All the conclusions are very
dependent on the method of measurement used
and on the experimental setup. So far no com-
mercially used catalyst has been investigated in
depth. There are also no literature data relating
to aging phenomena. It is generally accepted that
a redox mechanism, in which the selective oxi-
dation proceeds at oxygen atoms in the lattice,
is involved. Attempts are made in [39 59] to
describe the course of the reaction.
According to literature data, it is now possi-
ble to achieve yields of 120 kg PA /100 kg o-
xylene [35], but it is generally accepted that re-
actor yields exceeding 112114 kg PA/100 kg o-
xylene (calc. 100 %) are improbable onanindus-
trial scale. This is equivalent to 8082 % of stoi-
chiometric yield. In o-xylene based processes,
after allowing for the losses during condensa-
tion, dehydration, and distillation, pure phthalic
anhydride yields of 110112 kg PA/100 kg o-
xylene may be expected.
The oxidation of naphthalene proceeds in ac-
cordance with the following equation:
6 Phthalic Acid and Derivatives
The heat of formation is 1788 kJ/mol; total
combustion of naphthalene yields 5050 kJ/mol.
Dependingonthe yieldachievedandthe byprod-
ucts formed, a heat of reaction of 21002500
kJ/mol is expected. Possible side reactions are
shown in Scheme 2.
Besides maleic anhydride, naphthoquinone is
also formed as a byproduct, which requires a
high performance of the purication stage. Data
on the course of the reaction can be found in
[60 63].
According to the literature, yields of up
to 102 kg/100kg naphthalene are achievable in
commercial plants, but the nal product yields
generally do not exceed 98 kg/100kg naphtha-
lene, i.e., 85 % of theory.
Scheme 2.
2.4.1.2. Apparatus and Important Process
Steps in the Gas-Phase Oxidation of
o-Xylene
Procedure Employing an Ignitable Xy-
leneAir Mixture. The current objective of ph-
thalic anhydride technology is essentially to
limit energy consumption as far as possible. This
is supplemented by efforts to construct as com-
plete a system as possible for utilizing the steam
produced in the plant, or alternatively by con-
sidering driving the air blower with a steam tur-
bine [64, 65]. It is possible to save energy costs
primarily by increasing the concentration of o-
xylene in the process air, thus reducing the quan-
tity of process air for the same capacity and in-
creasing the steam yield [66].
It is possible to operate existing older plants,
which were originally designed for loadings of
40 g/m
3
(STP), with o-xylene loadings of up to
80 g/m
3
(STP) by employing modern catalysts
and modifying the reactor cooling system. The
attainable increase in loading depends primar-
ily on the feasibility of removing safely the in-
creased heat of reaction from the reactor.
The lower limit of ammability is exceeded
with a loading of ca. 44 g/m
3
. In order to ex-
ceed the limit of ammability, it is essential to
estimate the risks to the safety of the personnel
and of the plants. It is impossible to completely
eliminate the riskof ignitionof anexplosive mix-
ture, but the equipment may be designed to be
inherently safe, i.e., to withstand the highest pos-
sible pressure. Alternatively the equipment may
be safeguarded by pressure relief devices so that
pressure surges are relieved safely. Probability
of ignition inside the catalyst is minimal as the
autoignition temperature for phthalic anhydride
is 580
C, and the
retention times are 1024 h. In the case of ph-
thalic anhydride produced from o-xylene, the
treatment is intended to remove water and low-
boiling contaminants such as maleic anhydride
ando-tolualdehyde. The destructionof phthalide
is also achieved in pretreatments [87 90].
Distillation. The nal purication of the
crude phthalic anhydride is normally carried
out in a continuous distillation system. Fig-
ure 2 shows schematically a typical distillation
plant. The crude phthalic anhydride is rst pre-
treated in a predecomposer where byproduct ph-
thalic acid is converted to phthalic anhydride.
The product is then introduced into a column
in which the low-boiling constituents such as
maleic anhydride and benzoic acid are concen-
trated at the head and removed. The bottoms
are introduced into a distillation column, from
which pure phthalic anhydride is removed from
the head while the residue is discharged fromthe
bottom.
2.4.2. Fluidized-Bed Oxidation
Of the many uidized-bed processes for the oxi-
dation of naphthalene, the process developed by
Badger may still be in use today [91 93]. The
capacity of such plants has been increased by
injecting oxygen and air into the uidized bed,
thus enabling the naphthalene concentration to
be increased.
Liquid naphthalene is injected directly at the
bottom of the catalyst bed and evaporates im-
mediately, distributing itself over the entire bed.
Here it comes in contact with the catalyst and
reacts with the atmospheric oxygen fed in via a
distributor plate.
The vigorous agitation and mixing in the u-
idizedbedresults ina uniformtemperature being
maintained throughout the bed, with the temper-
ature being in the range 345385
C. The heat
is removed by cooling tubes installed in the bed
and is used to produce high-pressure steam. Any
entrained catalyst is separated by cyclones and,
after being retained in specially constructed ce-
ramic lters, is blown back into the reactor. Sev-
eral lter units are used, and one of them is al-
ways blown back with a stream of air in order
to remove catalyst from the lter surfaces. Up
to 60 % of the phthalic anhydride can be reco-
vered in a liquid condenser, the remainder being
desublimated in switch condensers. The crude
product is then puried in a distillation column
under vacuum. Aproposal for regenerating a de-
activated catalyst with SO
2
has been published
[94].
Kawasaki Steel Corporation is working on
the uidized-bed oxidation of o-xylene. Accord-
ing to patent literature a stoichiometric yield of
Phthalic Acid and Derivatives 9
up to 84 % at an o-xylene conversion of 94 % is
reached [95 99].
2.4.3. Liquid-Phase Oxidation of o-Xylene
In the liquid-phase oxidation of o-xylene, a mix-
ture of acetic acid, o-xylene, and catalyst, which
consists mainly of Co, Mn, and Br, is fed to the
rst vessel in a cascade. The reaction, which
is completed in the subsequent vessels, is ini-
tiated under pressure while the air is injected.
The water produced by the reaction is removed
in the rst vessel by azeotropic distillation with
o-xylene. The isomers of phthalic acid are re-
moved from the reaction mixture, and the ph-
thalic anhydride is obtained by crystallization.
The crude product is subjected to a pretreat-
ment which imposes special requirements be-
cause of the bromine content. Purication can
then be carried out by distillation. The o-xylene
must have an ortho isomer content > 99 % if
the process is to achieve high yields of 130 kg
of phthalic anhydride from 100 kg of o-xylene.
Companies such as H uls have investigated the
process [100] and, in addition, further work has
been done on it by Standard Oil of Indiana [101].
Sisas has described a two-step process: in the
rst liquid-phase oxidation step crudeo-toluic
acid is formed with a yield of 1.19 g per gram of
o-xylene at partial o-xylene turnover; unreacted
o-xylene is recycled. The second step is the gas-
phase oxidation of the o-toluic acid to phthalic
anhydride. The overall molar yield based on o-
xylene is up to 88 % [102].
2.5. Environmental Protection
Exhaust Gas. The exhaust gas from the
switch condensers still contains an apprecia-
ble quantity of organic substances and therefore
must be puried before it is discharged from the
plant. Scrubbing with water in a scrubbing tower
is suitable for extracting maleic anhydride, the
main byproduct in the exhaust gas. In a recov-
ery process, the organic acids fromthe scrubbed
gas are concentrated in several scrubbing towers
in which the scrubbing water is circulated until
the concentration is ca. 30 %. The scrubbing so-
lution is then worked up to recover maleic an-
hydride or fumaric acid. It is also possible to
produce sodium maleate by adding sodium hy-
droxide or sodium carbonate [103].
The efciency of the scrubbing depends on
the sophistication of the equipment and the num-
ber of scrubbing stages. The concentration of all
organic compounds can be reduced to < 150 mg
per m
3
exhaust air discharged from the plant.
The mainsubstance emittedfromphthalic an-
hydride plants is carbon monoxide, which can-
not be removed by exhaust-gas scrubbers. A
plant emits about 1100 t CO per 10 000 t of ph-
thalic anhydride produced, depending on the op-
erating conditions.
Apart from scrubbing towers, two processes
are at present available for purifyingexhaust gas:
1) Thermal combustion at 650850
C which
involves burning the organics and carbon
monoxide in the presence of an oil-fed or gas-
fed reverberatory ame.
2) Catalytic combustion at 250450
C, which
normally does not require additional fuel.
This process employs noble-metal or mixed-
oxide catalysts.
Such plants are supplied, for example, by
Haldr Topsoe in Denmark who employ their
Catox process or by Engelhard who market the
Torvex process.
A comparison of thermal and catalytic
exhaust-gas combustion for phthalic anhydride
plants is discussed in [104].
Catalyst. Depending on the operating proce-
dure, the catalyst normally has a service life of
28 years, after which it must be replaced be-
cause its activity is too low. Used catalysts may
be deposited, for example, at a secure landll
site; this depends on local regulations owing to
the toxicityof the additives. As a result of various
problems, washing and reusing the active mate-
rial has so far been uneconomical, even though
proposals are repeatedly being made in this con-
nection [105, 106].
2.6. Quality Specications and Analysis
The specications of technically pure phthalic
anhydride are listed in Table 2. A suitable
method of determining the acids by gas chro-
matography is to form derivatives, for example,
with diazomethane, but direct determination on
10 Phthalic Acid and Derivatives
Table 2. Specications of technical-grade phthalic anhydride
Property Average value Method Specication limit
Purity, % 99.9 GC 99.8 min
Solidication point,
C 131.0 ASTM 149367 130.8 min
Maleic anhydride, % 0.03 GC 0.05 max
Benzoic acid, % 0.03 GC 0.1 max
Color index of melt (APHA) 510 ASTM 336674 20 max
Heat stability (APHA) 1020 ASTM 336674 40 max
packedcolumns or capillarycolumns is alsopos-
sible.
2.7. Economic Aspects
In 1996 world production capacity for phthalic
anhydride was > 3.710
6
t/a and this gure is
expected to exceed 4.610
6
t/a in the year 2000.
A remarkable increase in capacity is occurring
in south-east Asia.
The world capacities for phthalic anhydride
(10
3
t) in 1996 can be broken down as follows:
Western Europe 869
Eastern Europe 641
North America 568
South America 256
Japan 320
South Korea 260
India 153
Taiwan 95
Remainder of Asia 155
Africa and Near East 124
2.8. Storage and Transportation
Pure phthalic anhydride is a very stable com-
pound and may be stored for a prolonged pe-
riod in the molten state without any change in its
properties. To prevent res, it is recommended
that storage containers and transportation are
blanketed with nitrogen.
Molten phthalic anhydride is subject to the
following safety and transport regulations:
EC directive no. 607009004
Symbols X
i
R phrases 36/37/38
UN no. 3256
Packaging group III
Transportation regulations
GGVE/RID 3.61 c
GGVS/ADR 3.61 c
ADN/ADNR 3.61 c
GGV See/IMDG code 3.3, UN-No. 3256-PG III
IATA Dangerous Goods
Regulation 3, UN-No. 3256
2.9. Uses
The most important application of phthalic an-
hydride is the production of plasticizers. The
main uses of phthalic anhydride are [13]:
Plasticizers 55 %
Unsaturated polyester resins 14 %
Alkyd resins 15 %
Other 16 %
3. Phthalimide
3.1. Properties
Phthalimide [85-41-6], 1,3-dioxoisoindoline,
C
8
H
5
O
2
N, M
r
147.14, crystallizes from solu-
tion as white needles or prisms, while sublima-
tion produces platelets. The compound is spar-
ingly soluble in water (0.3 g at 20
C, 0.9 g at
50
C, 2.2 g at 100
C
Heat of combustion 3560 kJ/mol
Heat of fusion 187.6 J/g
Specic heat at 100
C 1.21 J g
1
K
1
Vapor pressure at
120
C 0.10 mbar
150
C 0.95 mbar
180
C 5.93 mbar
220
C 30.7 mbar
254
C 187.6 mbar
Flash point 214
C
Ignition temperature 530
C
Phthalic Acid and Derivatives 11
With bases, phthalimide forms water-soluble
salts, which react with halogens to form the
corresponding N-chloro, N-bromo, or N-iodo
compounds. These N-halo compounds are also
obtained if alkali-metal phthalimides are re-
acted with hypochlorous or hypobromous acid.
When the N-halo compounds are heated, isatoic
anhydride [118-48-9] or anthranilic acid (o-
aminobenzoic acid) [118-92-3] is formed in a
Hofmann degradation [108, 109]. The reaction
of alkali-metal phthalimides with alkyl halides
to give N-alkyl phthalimides, and subsequent
hydrolysis or hydrazinolysis affords primary
amines (Gabriel synthesis) [110]:
3.2. Production
Phthalimide is produced almost exclusively
from phthalic anhydride and ammonia, but pro-
cesses based on phthalic anhydride and urea
or oxidative ammonolysis of o-xylene are also
known.
3.2.1. Production from Phthalic Anhydride
and Ammonia
The continuous processes for producing ph-
thalimide are particularly important industrially.
One of the continuous processes is carried out in
an externally heated, vertical reaction tube lled
with packing material [111]. A gas-tight con-
nection links the bottom of the reaction tube to
a sublimation chamber from which a discharge
armdischarges the phthalimide through a slot by
means of a screw conveyor. The exhaust gases
are dischargedfromthe sublimationchamber via
built-in bafes.
Molten phthalic anhydride and excess ammo-
nia are continuously fed into the reaction tube
at the top and are reacted at 250280
C. The
reaction gases are cooled to 170180
C in the
sublimation chamber, and solid phthalimide is
deposited and discharged. Water and excess am-
monia are removed via an exhaust-gas pipe. A
98 % yield of phthalimide is obtained with a pu-
rity of 99 %.
Another continuous process for producing
phthalimide from phthalic anhydride is a coun-
tercurrent process [112]. Molten phthalic anhy-
dride is continuously fed to the head of a reactor
while ammonia is continuously fed to the bot-
tom. The temperature increases from 150
C
at the top of the reactor to a maximumof 270
C
at the bottom. The molten phthalimide emerging
from the bottom of the reactor with a purity of
99 % is cooled and aked or dissolved in aque-
ous alkali, and supplied as a starting material
for further syntheses, e.g., to produce anthranilic
acid.
The exhaust gas from the head of the reactor,
which is composed of water vapor, sublimed ph-
thalimide and phthalic anhydride, and ammonia,
is fed to the bottomof a scrubbing column where
it is scrubbed in countercurrent with molten ph-
thalimide from the bottom of the reactor. The
melt taken fromthe bottomof the scrubbing col-
umn, which contains unreacted phthalic anhy-
dride, is returnedtothe reactor, while the exhaust
gas from the scrubbing column, which contains
predominantly water vapor and phthalimide, is
dissolved in aqueous alkali, and the solution is
used to produce anthranilic acid.
3.2.2. Production from Phthalic Anhydride
and Urea
Solvent-Free Process. Amixture of phthalic
anhydride and urea is introduced into a heatable,
tiltable reaction vessel. Mounted on the cover
of the vessel is an extraction tube for the re-
action gases, CO
2
and water vapor. The reac-
tor is heated to 130140
C
Heat of combustion 4013 kJ/mol
Heat of evaporation 67 kJ/mol
Specic heat at 30
C 1.30 J g
1
K
1
Vapor pressure at 20
C 0.05 mbar
Density 1.238 g/cm
3
Apparent density ca. 0.5 g/cm
3
Flash point 162
C
4.2. Production
Phthalonitrile is produced commercially fromo-
xylene, phthalic acid, phthalic anhydride, phtha-
lamide (1), or phthalimide (2).
4.2.1. Production from o-Xylene
Ammoxidation. In a single-stage continu-
ous process, o-xylene is converted to phthaloni-
trile by reaction with ammonia and oxygen in the
gas phase in a uidized-bed reactor. Generally,
metal oxide mixtures containing vanadium, an-
timony, chromium, and molybdenum, with fur-
ther active components such as iron, tungsten,
and alkali-metal oxides, on an alumina or silica
support are used as catalysts [116 123].
Phthalic Acid and Derivatives 13
The course of the reaction and the indus-
trial importance of ammoxidation have been de-
scribed in depth in [124 126]. The commer-
cial process of producing phthalonitrile from
o-xylene by oxidative ammonolysis is as fol-
lows [127, 128]: Agaseous mixture of o-xylene,
NH
3
, and O
2
is passed through a distributor plate
into a uidized-bed reactor. Either a vanadium
oxideantimony oxide catalyst or a vanadium
oxidechromium oxide catalyst on an alumina
support is used. The optimum temperature for
pressureless ammoxidation is 480
C. Temper-
atures below 480
C, ammonia
begins to burn. The cooling elements incorpo-
rated in the uidized bed make it possible to
keep the temperature constant despite the large
heat of reaction. The hot reaction gases from the
reactor are quenched in a product settler with an
aqueous suspension of phthalonitrile. After set-
tling and cooling, the phthalonitrile is separated
in decanters and dried.
The phthalonitrile produced in this way,
which contains < 0.1 % of acid and phthalim-
ide and < 0.1 %water, can be processed directly
without further purication to yield phthalocya-
nine pigments. The yield is 8085 %.
Ammonia is removed from the gas mixture
emerging fromthe product settler (the latter con-
tains NH
3
, CO
2
, CO, N
2
, and traces of HCN)
in an ammonia recovery plant, and the exhaust
gas is burnt. The ammonia is returned to the
process. Unreacted o-xylene and the interme-
diate o-toluonitrile can be worked up to iso-
late o-toluonitrile or can be fed back into the
uidized-bed reactor for complete conversion to
phthalonitrile.
Oxidative Ammonolysis. The continuous
production of phthalonitrile from o-xylene by
oxidative ammonolysis has been studied in a
pilot plant [115, 129]. In this process o-xylene
is reacted with ammonia in the presence of a
metal-oxide catalyst in the gas phase at 350
450
C in the presence of NH
3
on a boron
phosphatealumina catalyst.
Ammonia is recovered from the residual gas
mixture, which contains NH
3
, CO
2
, CO, N
2
, and
traces of HCN, and fed back into the process.
Noxious gases such as CO and HCN are burnt
in the regenerator and N
2
and CO
2
are elimi-
nated from the process.
4.2.2. Production from Phthalic Acid
Derivatives
Phthalonitrile can be produced from phthalic
acid, phthalic anhydride, phthalamide, or phthal-
imide by reaction with ammonia and elimination
of water at 300500
C in a xed-bed
reactor downstream of the heating apparatus us-
ing an Al
2
O
3
catalyst. The reaction gas emerg-
ing fromthe reactor is quenched with water, and
the phthalonitrile is separated from the aqueous
suspension in decanters and dried. The concen-
tration of NH
3
in the circulating gas is kept at
ca. 90 % by adding NH
3
.
Numerous other production methods have
been described [130, 131], but have no commer-
cial signicance. For example, phthalonitrile can
be prepared by eliminating water from phtha-
lamide in the presence of acid halides such as
phosgene or thionyl chloride. To prevent hydro-
lysis of the phthalamide by the acid formed, dilu-
ents such as benzene or chlorobenzene are added
14 Phthalic Acid and Derivatives
or the acids are bound by tertiary amines such
as N,N-diethyl-o-toluidine or pyridine. It is also
possible to use acylated secondary amines such
as N-methylformamide.
4.3. Uses
Phthalonitrile is used as a starting material
for producing phthalocyanine pigments (Ph-
thalocyanines), uorescent brightners, and pho-
tographic sensitizers.
5. Phthalates
A wide variety of dialkyl phthalates are pro-
duced and marketed (Table 3).
In the early 1900s, the availability of mono-
hydric alcohols was limited to those with a chain
length of up to four carbon atoms. This resulted
in the development and manufacture of four es-
ters; e.g., diethyl phthalate (DEP) was used as
a heat-transfer oil, and dibutyl phthalate (DBP)
was used to reduce the hygroscopicity of explo-
sives. Both DEP and DBP were also used as car-
riers in the perfume industry [132].
Also at this time camphor was being used as
a plasticizer for cellulose nitrate and cellulose
acetate. In the early 1920s it was found that ph-
thalates could replace the expensive plasticizer
camphor. This increased the demand for phtha-
lates [132]. In 1937, the Chemiker Taschen-
buch [133] listed seven phthalates for use as
plasticizers. The late 1920s saw the start of the
development of poly(vinyl chloride), PVC. PVC
is rigid and brittle at ambient temperature. How-
ever, when plasticized it was found to be suit-
able for replacing natural rubber [132]. In 1929
a U.S. patent for plasticized PVCwas granted to
Ostromislenski [134], and Kyrides applied for
U.S. patent for the use of di-2-ethylhexyl phtha-
late (DEHP) as plasticizer for PVC, which was
granted in 1933 [135].
The most important phthalates are DEHP, the
standard general purpose plasticizer for PVC,
followed by DINP and DIDP [136]. In trade,
DEHP is also referred to as DOP. Besides the
phthalates listed in Table 3, phthalate blends and
coesters are used for special applications [137].
5.1. Physical and Chemical Properties
All phthalates listed in Table 3 are clear, oily
liquids at room temperature. They are soluble in
common organic solvents and are miscible with
other PVC plasticizers. When added to plastics
and resins, they improve the workability during
fabrication, modify the properties, or give rise
to new improved properties not exhibited by the
original material [138].
Some of the important properties of plasticiz-
ers are listed below. These properties are inu-
encedbythe manufacturingprocess andcanvary
signicantly. The values given are typical ranges
in which commercial products are available. The
analytical methods are listed in Table 4.
APHA color 560
Acid number, mg KOH/g 0.020.1
Water content, wt % 0.050.1
Purity, area % or wt % 99.099.5
Odor odorless or low
5.2. Raw Materials
Phthalic anhydride (PA) and monohydric alco-
hols are generally the raw materials for phtha-
lates of commercial signicance. Phthalic anhy-
dride is mainly used in the molten form. The
alcohols are based on the C
2
C
4
olens, which
are produced in either the steam cracking pro-
cess or in petrochemical plants. These alco-
hols can be produced by (1) oligomerization
or Ziegler synthesis followed by hydroformy-
lation and hydrogenation, or (2) by aldol con-
densation with hydrogenation (e.g., the produc-
tion of 2-ethylhexanol frompropylene). The lin-
ear or slightly branched C
6
to C
11
alcohols are
based on ethylene. Isononanols are produced
from the C
4
fraction of the steam cracker out-
put or from C
8
C
9
olens from a poly gas unit
(olen oligomerization) [133].
5.3. Production
The formation of phthalates is as follows:
Phthalic Acid and Derivatives 15
Table 3. Commercially important phthalates
Chemical denomination Code 1* CAS M
r
Viscosity Density Refractive
DIN 7723 registry no. at 20
C, at 20
C, index at
mPa s g/m
3
20
C
Dimethyl phthalate DMP [131-11-3] 194.19 18 1.191 1.516
Diethyl phthalate DEP [84-66-2] 222.24 13 1.117 1.502
Dibutyl phthalate DBP [84-74-2] 278.35 21 1.047 1.493
Diisobutyl phthalate DIBP [84-69-5] 278.35 41 1.039 1.490
Butyl benzyl phthalate BBP [85-68-7] 312.36 59 1.121 1.538
Diisopentyl phthalate DIPP [84777-06-0] 306.40 30 1.024 1.490
Diheptyl phthalate DHP [68515-50-4] 362.51 49 0.993 1.487
Di-2-ethylhexyl phthalate DOP(DEHP) [117-81-7] 390.56 80 0.983 1.487
Diisooctyl phthalate DIOP [27554-26-3] 390.56 7584 0.984 1.488
Di-n-octyl phthalate DNOP [117-84-0] 390.56 39 0.980 1.485
Di(hexyl-octyl-decyl) phthalate HXODP(610P) [68515-51-5] 44 0.974 1.484
Di(octyl-nonyl-decyl) phthalate ONDP(810P) [71662-46-9] 49 0.969 1.483
Di(heptyl-nonyl) phthalate HNP(79P) [68515-41-3] 44 0.986 1.486
Di(heptyl-nonyl-undecyl) phthalate HNUP(711P) [68515-42-4] 50 0.972 1.484
Di(nonyl-decyl-undecyl) phthalate NDUP(911P) [68515-43-5] 73 0.962 1.484
Diisononyl phthalate DINP
**
[28553-12-0] 418.62 80 0.975 1.486
Diisononyl phthalate DINP
**
[68515-48-0] 418.62 105 0.974 1.486
Di(3,5,5-trimethylhexyl) phthalate DINP
**
[14103-61-8] 418.62 110 0.968 1.486
Diisononyl phthalate DINP
**
[28553-12-0] 418.62 165 0.978 1.488
Diisodecyl phthalate DIDP [26761-40-0] 446.67 130 0.966 1.486
Diundecyl phthalate DUP [3648-20-2] 474.72 70 0.955 1.482
Diisoundecyl phthalate DIUP [85507-79-5] 474.72 200 0.964 1.487
Diisotridecyl phthalate DTDP [68515-47-9] 530.83 310 0.948 1.482
Di(methoxyethyl) phthalate (DMEP) [117-82-8] 282.29 55 1.170 1.502
Di(butoxyethyl) phthalate (DBEP) [117-83-9] 366.45 42 1.061 1.486
* Codes in parentheses are used in English.
** DINP types are based on different branched isononyl alcohols.
The rst step, alcoholysis of PA to give the
monoester, is rapid and goes to completion. The
reactiongenerallystarts at elevatedtemperatures
and proceeds exothermically.
The second step is the conversion of the mo-
noester to a diester with the formation of water.
This is a reversible reaction and proceeds more
slowly than the rst, thus determining the overall
rate of reaction. To shift the equilibriumtowards
the diester, the water of reaction is removed by
distillation. The rate of reaction can be inu-
enced by the choice of catalyst and the reaction
temperature. For fast conversionrates, highreac-
tion temperatures are generally used. However,
these are inuenced by the boiling point of the
alcohol and/or the type of catalyst.
Sulfuric acid [132], methanesulfonic acid
[139, 140] and p-toluenesulfonic acid [132] are
effective esterication catalysts [140]. Sodium
hydrogensulfate [132] or the monoester itself
[132, 141] canalsobe used. Acidcatalysts canbe
used for esterication reactions at temperatures
up to 140165
C, but the
conversion to diester is not complete, and the re-
covery of unreacted monoester is necessary.
Currently, nearly all major phthalate produc-
ers use amphoteric catalysts for the esterica-
tion of high boiling alcohols. The rst claim
for the use of titanates as esterication catalysts
was made in 1955 in the United States [142].
Since then titanates and tin(II) compounds have
16 Phthalic Acid and Derivatives
Figure 3. Simplied ow sheet for the production of phthalates. Terms in parentheses are optional.
been widely used as esterication catalysts [140,
143 145].
The reaction temperatures for the amphoteric
catalysts are about 200
C
4
alcohols, the overhead vapors of the reac-
tor are fed into a rectication column where the
evaporated alcohol is dried before it is recycled
to the reactor. If higher alcohols of very low wa-
ter solubility are used, the vapors from the re-
actor can be fed directly into a condenser which
drains into a alcoholwater separator. The water
is separated here by settling and the alcohol is
recycled to the reactor.
Usually a 2025 % excess of alcohol is used
[140]. The catalyst can be added with the alco-
hol. Amphoteric catalysts are preferably added
after monoester formation has taken place at
160
C
[140], and lime or soda ash or sodiumhydroxide
are added with water. The mixture is stirred and
Phthalic Acid and Derivatives 17
the water is then removed by vacuum distilla-
tion.
After neutralization, residual alcohol is re-
moved by steam distillation [140]. The temper-
ature, pressure, and steam for injection are ad-
justed so that the product after this treatment is
nearly alcohol free and dry. The nal ltration
is performed with a rotary vacuum lter or plate
and frame lters. If the hydrolyzed catalyst was
not removed with the water phase in the neutral-
ization step, the catalyst can be removed in the
ltration step by adding a lter aid.
The continuous BASF process described in
[132, 141] can be carried out autocatalytically
or preferably by using an amphoteric catalyst.
The esterication is carried out in a multistage
cascade of agitated vessels. The reaction mix-
ture leaving the last reactor is pumped to a vac-
uum ashing chamber to recover the excess al-
cohol [132]. To remove the catalyst, the crude
ester is treated with a mild aqueous alkali solu-
tion in a mixer/settler system. To remove resid-
ual volatiles such as water, alcohol, and other
low-boiling compounds a steam distillation is
carried out under reduced pressure. For the -
nal treatment a lter aid is added. The product
is subjected rst to coarse and then to nal l-
tration.
5.4. Environmental Protection
The wastewater from phthalate processes is
contaminated with phthalates, alcohols, and
other organic compounds. Therefore, the pro-
cess wastewater must be treated before discharg-
ing to a municipal system or a water body. Lo-
cal regulations dictate the treatment and disposal
of this water. Generally, biological treatment
is required. Studies on phthalates have shown
that they can be biodegraded by many species
of bacteria [146 148]. In Germany only three
phtha-lates (diethyl phthalate, diallyl phthalate,
benzyl butyl phthalate) are listed in the catalog
for water-endangering materials [149]. They are
classied as materials with moderate hazard po-
tential (WGK 2).
In the United States and Canada there are no
current regulations limiting the concentration of
phthalates in wastewater.
In the TA Luft [150], only DOP is mentioned
as material of class II (intermediate danger po-
tential). The other phthalates have been classi-
ed according to the hazard potential for the en-
vironment by the rules given in [150]. Exxon
[146] lists the C
4
C
11
phthalates, with the ex-
ception of DOP, in class III. Because of their
low vapor pressure, the phthalates easily meet
the concentration limits for phthalates specied
by TA Luft.
Owing to the high working temperatures,
air emissions from compounding and extru-
sion processes for exible PVC generally have
high phthalate concentrations. Therefore emis-
sion controls may be required for these dis-
charges. Exxon [146] gives the following data
for the environmetal levels of DBP and DEHP:
Fresh and estuarine/sea waters: 10 and 0.7
ppb, respectively
Sediments: a few ppb up to 100 ppm
Air: 3 ng/m
3
over the sea and up to 130
ng/m
3
in cities.
Detailed information on DBP and DEHP dis-
tribution in the environment are given in [147,
148].
In North America, waste containing DEHP
and DBP is regarded as hazardous waste and
must be disposed off in a secured landll site.
5.5. Quality Specications
For manufacturing exible plastics the follow-
ing properties are important for the selection of
a plasticizer [134]: (1) compatibility with plas-
tics, (2) solvating temperature (a measure of the
solvating capability on plastics), (3) efciency
for plasticizing, (4) volatility in plastics, (5) ex-
tractability and diffusion losses in plastics, and
(6) electrical properties. The properties required
in the nal products determine the type and the
quality specication of the plasticizer.
For manufacturing exible PVC for blood
bags or other goods in medical application the
plasticizer used (e.g., DEHP) should have a pu-
rity of min. 99.5 %, a low alcohol content, a
low acid number (max. 0.02 mg KOH/g), and
no odor [151].
Flexible vinyl sheets used for automotive
interior trim should be plasticized with a ph-
thalate that ensures low fogging of the inte-
rior windshield. Also low-temperature exibil-
ity, light and thermal stability, and processability
18 Phthalic Acid and Derivatives
are required. BASF [152] recommends a low-
branched DINP or a HNUP for this application.
Phthalates usedfor plasticizingPVCfor wire,
cable, and conductor sheathing need high elec-
trical volume resistivity. Products with volume
resistivities > 1.010
11
cm at 20
C are re-
quired. Typical resistivity of DOP is 1.010
11
,
and of DIDP 1.010
12
cm.
Quality tests are carried out both on the ph-
thalate itself and on the plasticized polymers.
Table 4 lists standards for testing phthalates as
PVC plasticizers.
Nowadays, almost all demands of plasticizers
for use in exible plastics can be met by phtha-
lates.
Table 4. Standards for testing of phthalates
Property Methods
DIN ASTM
Phthalates
Viscosity at 20
C 51 561 D 445
Density at 20
C 51 757 D 1045
Refractive index at 20
C 51 423 D 121861
APHA color ISO 6271 D 1209
Acid value 53 402 D 1045
Pour point ISO 3016 D 9757
Water content 51 777 D 1364
Purity gas chromatography
Plasticized PVC
Solvating temperature 53 408
Tensile strength 53 455 D 412
Elongation at break 53 455 D 412
Brittleness temperature 53 372 D 746
Clash & Berg torsion
stiffness at 310 N/mm
2
53 447 D 1043
Shore B hardness 53 505 D 2240
Volume resistivity at 25
C 53 482 D 1169
5.6. Storage and Transportation
Because phthalates are neutral liquids they can
be stored in carbon steel tanks. The more expen-
sive stainless steel or aluminum alloy tanks are
also used. All phthalates have a ash point above
100
C.
87 % of the phthalates produced are used for
formulating exible PVC, which is consumed
for manufacturing the following goods [156]:
Wire and cable 25 %
Film and sheeting 23 %
Flooring 15 %
Proles and tubing 10 %
Plastisol spread coatings 11 %
Other plastisols 8 %
Miscellaneous (shoe soles, blood
bags, gloves)
8 %
5.8. Economic Aspects
The major phthalate producers in Europe, the
United States, and Japan are listed in Table 5.
The capacities of these companies range from
70 000 to 350 000 t/a.
In 1988 world consumption of plasticizers,
including nonphthalates, was 3.8510
6
t [157].
Based on these gures and uses in other applica-
tions, world consumption for phthalates is esti-
mated at 3.2510
6
t, of which DEHP accounts
for ca. 2.110
6
t. The estimated total consump-
tion of all phthalates and consumption of DEHP
(in 10
3
t) by geographic region is:
Phthalic Acid and Derivatives 19
Total DEHP
Western Europe 900 465
North America 730 155
Eastern Asia 530 490
Japan 320 245
Others 720 765
World total 3250 2120
Prognoses for the world market show an an-
nual increasing demand of ca. 2.5 %.
Table 5. Major phthalate producers and their trade names
Company Location Trade name
Aristech Chemical Corp. United States PX
ATOCHEM Europe Garbeex
BASF Europe, Palatinol
United States
BP Chemicals Europe Bisoex
Chemische Werke H uls Europe, Vestinol,
United States Nuoplaz
Eastman Chemical United States Kodaex
Enichimica Europe Sicol
Exxon Chemical Europe, Jayex
United States
Hoechst Europe Genomoll
ICI Europe Hexaplas
Kyowa Hakko Japan Kyowacizer
Mitsubishi Kasei Vinyl Japan
Neste Chemicals Europe
New Japan Chem. Japan
6. Toxicology
6.1. Use of and Exposure to Phthalic
Acid and Derivatives
Phthalate esters, phthalic acid, phthalic anhy-
dride, phthalimide, and phthalonitrile are inter-
mediates that are produced and further used in
industry under well-controlled conditions. Oc-
cupational exposure may occur but is expected
to be low. Exposure of the general public is low
and negligible. However, the case of phthalate
esters is different because these are generally
used as plasticizers, not only for a wide range of
products for industrial use such as resins, paints,
electrical cable, and equipment lining, but also
for numerous consumer products like PVCoor
tiles, shoes, rubber boots, gloves, and house-
hold plastic materials including food contain-
ers. Plasticizers are also contained in toys and in
medical equipment such as blood bags, infusion
lines, and devices for hemodialysis and extracor-
poreal membrane oxygenation (ECMO). Thus,
both occupational exposure of workers during
production and the exposure of the general pub-
lic may occur. The highest levels of exposure
are expected for children and patients undergo-
ing intensive medical treatment. This promoted
intense investigation of the exposure to and the
mode of action of phthalate esters. The most
prominent diethylhexyl phthalate (DEHP)
may be regarded as a model phthalate ester and
is therefore described below in more detail.
6.2. Toxicological Proles
6.2.1. Phthalic Acid
Only few investigations exist; however, the tox-
icity appears to be low. Acute toxicity (LD
50
,
mouse, i.p.): 550 mg/kg [158].
Metabolism. Most of the material is excreted
in the urine, either directly (as in dogs) or par-
tially conjugated (as in rats and rabbits). A mi-
nor part may be decarboxylated and excreted as
benzoic acid [159].
6.2.2. Phthalic Anhydride
Phthalic anhydride is an inhalation allergen
and may cause occupational asthma in exposed
persons [160 164]. Hexahydrophthalic acid
showed also respiratory sensitizing properties
[163].
Acute toxicity data indicate minor toxicity.
Acute toxicity: LD
50
, rat, oral: 15004000
mg/kg [165]; LC
50
, rat inhalation: > 210 mg
m
3
h
1
(dust) [166]; LD
50
, rabbit, dermal: >
10 000 mg/kg [166]. Dermal irritation: nonirri-
tating or slightly irritating in rabbit skin [165].
Eye irritation (rabbit): irritating [165, 166]. Der-
mal sensitization: sensitizing in the mouse ear
swelling test [162].
Subacute Toxicity. Feeding experiments
with 250, 1000, and 3800 ppm in the diet (17,
67, and 253 mg kg
1
d
1
) were tolerated for 28
d without signicant effects [166]. Gavage ex-
periments in female rats, beginning with 20 mg
kg
1
d
1
and subsequently increasing doses up
to 4800 mg kg
1
d
1
over a period of 8 weeks,
caused toxic effects on the kidneys and the gas-
tric mucosa at doses exceeding 1200 mg kg
1
20 Phthalic Acid and Derivatives
d
1
[167]. Inhalation experiments with sub-
limed material in rats, mice, rabbits, and cats
showed some irritation and impaired respiration
at 10 000 mg/m
3
for 4 h per day for two weeks
[165]. In female guinea pigs, vapor concentra-
tions of 615 mg/m
3
for 30 min per day over 4
and 8 d caused irritative effects on the eyes and
respiratory tract [167]. Repeated inhalation of
pure phthalic anhydride dust in concentrations
of 8.5 mg/m
3
for 3 h per day for several periods
of 4 consecutive days, alternating with 10-d re-
covery periods for a total experimental period
of 8 months, caused irritation of the respiratory
tract and frequent pneumonias [167].
Genotoxicity. Nomutagenic effects were de-
tected in the Ames test. No sister chromatid ex-
changes or chromosome aberrations were found
in CHO cells. All experiments were conducted
with and without metabolic activation [168,
169].
Chronic Toxicity and Carcinogenicity.
Long-termbioassays with 7500 and 15 000 ppm
in the diet of rats (500 and 1000 mg kg
1
d
1
)
did not lead to an increased tumor incidence
compared to control animals. Similarly, mice
dosed with 3500 and 7100 mg kg
1
d
1
for 32
weeks and with 900 and 1800 (females) or 1800
and 3600 (males) mg kg
1
d
1
for a further 72
weeks did not respond with an increase of tu-
mors [170, 171].
Reproductive Toxicity. Only screening ex-
periments with intraperitoneal injection in CD-
1 mice have been reported. 80 mg/kg injections
were carried out from gestation days 8 to 10
or 11 to 13. Malformations (cleft palates, mal-
formations of ribs and vertebrae) were detected
at doses 55.5 mg kg
1
d
1
[172 174]. The
experiments are inconclusive for the quantitative
assessment of a teratogenic risk, since the expo-
sure route was articial and mice are prone to
respond to experimental stress with these types
of malformations.
6.2.3. Phthalimide
Few data exist; however, a low toxicity may be
assumed. The material is hydrolyzed to phthalic
acid and ammonia [175].
Acute toxicity. LD
50
, (rat, oral): > 5000
mg/kg (Bayer AG (1978), cited in [176]); LD
50
,
(mouse, oral): 5000 mg/kg [177]. The material
was not irritating to the rabbit skin and eye [176].
No data on sensitizing properties are known to
exist.
Repeated-Dose Toxicity. There were no
substance-related effects in male and female rats
after inhalation of the material at 0.051, 0.154,
and 0.523 mg/L (4 weeks, 6 h/d, 5 d/week) re-
garding mortality, body weight gain, clinical
chemistry including hematology and urinalysis,
or gross organ changes (IBT (1979); cited in
[176]). No long-term studies are known to exist.
Genotoxicity. Phthalimide was not geno-
toxic in an Ames test performed according to
the OECD test guideline using S. typhimurium
strains TA98, TA100, TA1535, and TA1537 at
concentrations of 85000 g per plate, with and
without metabolic activation (Bayer AG(1993);
cited in [176]).
Reproductive Toxicity. No reliable studies
are reported in [176].
6.2.4. Phthalonitrile
Acute Toxicity. LD
50
; (rat, oral): 30125
mg/kg [178, 179]. No irritation of rabbit skin
(BASF AG (1969); cited in [178, 179] or eye
(BASF AG (1993); cited in [178]) was noted.
Medical records of workers involved in the pro-
duction and use of phthalonitrile do not indicate
sensitizing properties (BASF AG (1995); cited
in [262]).
Repeated-Dose Toxicity. Neurotoxicity
was noted in rats and mice following repeated
dosing in the form of convulsions and increased
excitation [178]. Increased activity and body
weight decreases were seen in a 13-week rat
neurotoxicity study at 10 and 25 mg kg
1
d
1
.
There was no histopathological correlation to
the behavioral changes. The NOAEL was 3 mg
kg
1
d
1
in this study (Bio Research (1994),
cited in [178]). Epileptiformic convulsions last-
ing minutes were observed in workers after oc-
cupational exposure with a latency period of sev-
eral hours to days [180]. Clinical examinations,
Phthalic Acid and Derivatives 21
clinical chemistry, hematological parameters,
neurological studies, and EEG studies detected
no abnormal ndings in 81 workers who had
been exposed for an average of 8.5 years, 11
of whom had suffered from acute intoxications
(Kleinsorge et al. (1979); cited in [178]). Also
dermal exposure may be relevant [179]. Since
the material does not appear to be metabolized
to cyanide, the typical measures taken against
cyanide intoxications are not promising.
Genotoxicity. The material does not cause
mutagenic effects in the Ames test [181] and
the CHO/ HGPRT gene mutation assay [182].
The substance was negative in the mouse mi-
cronucleus test in vivo (LMP (1987); cited in
[178]). Chromosomal aberrations were not in-
creased in 20 workers, seven of whom had ex-
perienced acute intoxications (Fleig and Thiess
(1979); cited in [178]).
Carcinogenicity. After oral and subcuta-
neous administration to rats and mice, and after
percutaneous administration to mice, induction
of leukemia was reported. The study description
is, however, vague and does not allow a conclu-
sive assessment [183].
6.2.5. Phthalate Esters
Diethylhexyl phthalate has been most exten-
sively investigated, due to the large production
volume and widespread use. The following sec-
tions on the general toxicological prole of ph-
thalate esters therefore focus on DEHP as the
lead substance. However, signicant differences
between groups of phthalate esters of varying
chain lengths need consideration. Generally, ac-
cording to the length of the linear alkyl backbone
of the alcohol moiety, groups of low(C
1
C
3
), in-
termediate (C
4
C
6
), and high molecular weight
(C
7
C
13
) phthalate esters are distinguished.
6.2.5.1. Metabolism and Toxicokinetics
Phthalate esters in general are rapidly hydro-
lyzed to the corresponding monoesters and the
respective alcohol in the intestine [184]. These
are readily absorbed from the gut [185], with
some species difference in absorption and ex-
cretion. Short-chain phthalates such as dimethyl
phthalate are either excreted unchanged in the
urine or completely hydrolyzed to phthalic acid
and excreted [186].
The pattern of DEHP metabolites in the urine
has been elucidated [187, 188]. MEHP appears
in native or depending on the species in
glucuronidated form (at the carboxylate moi-
ety) in the urine along with degradation products
from -and -1-oxidation in the alcohol side
chain. -Oxidation leads to a carboxylic moi-
ety at the C-terminal site of the alcohol chain.
-1-Oxidation leads to the 5-hydroxy and 5-
keto derivatives of MEHP [188]. The latter was
shown to be the metabolite ultimately responsi-
ble for peroxisome proliferation [189]. The -
1-oxidative pathway is more pronounced in the
rat than in the cynomolgus monkey [190].
The metabolism of DEHP is shown in Fig-
ure 4.
The liberated primary alcohols may be oxi-
dized to the respective aldehyde and carboxyl-
ic acid via enzymatic reactions catalyzed by al-
cohol and aldehyde dehydrogenases. The fatty
acids are further degraded in the intermediary
metabolism, i.e., the fatty acid and citrate cy-
cles. This pathway holds true for linear pri-
mary alcohols and for primary alcohols which
are methyl-branched at even-numbered posi-
tions, whereas branched primary alcohols with a
methyl substituent at an odd-numbered position
and 2-ethyl-branched alcohols do not undergo
-oxidation. Similarly, secondary and tertiary
alcohols are not metabolized to the fatty acid
and aldehyde, respectively. When -oxidation
does not allow rapid metabolism, alternative re-
actions (chain hydroxylation and conjugation)
occur prior to urinary excretion [191].
Di-n-octyl phthalate was similarly metabo-
lized, i.e., the monoester was formed and side-
chain oxidations occurred. Phthalic acid was
also detected in minor quantities [192].
Toxicokinetic studies using radiolabeled
DEHP by gavage in pregnant and nonpregnant
female Wistar rats and CD-1 mice appeared
in unpublished reports sponsored by the Eu-
ropean Council for Plasticizers and Intermedi-
ates [193 197]. Determinations were made af-
ter single oral doses of 200 or 1000 mg/kg b.w.
and after ve daily doses at these levels. Results
indicated rapid absorption, linearity of peak ra-
22 Phthalic Acid and Derivatives
Figure 4. DEHP metabolism. The key metabolites are circled. (From Koch et al. [188], used with kind permission of Springer
Science and Business Media and Prof. Dr. J. Angerer.)
diolabel plasma levels, and the area under the
curve (AUC) with dose in both species irrespec-
tive of the pregnancy status. A rapid increase
and biphasic radiolabel blood curve was noted
in all mice but not in rats. This probably mir-
rors enterohepatic cycling of conjugates, e.g.,
glucuronides, which are formed in mice but not
in rats. The plasma half-life times ranged bet-
ween 7.1 and 10.3 h at the lowdose and between
5.5 and 13.5 h at the high dose. The ndings
were conrmed when radiolabeled DEHP was
given after ve preceding doses. MEHP domi-
natedDEHPinbloodbothat the lowandthe high
dose level, as was seen from the AUC in rats (by
a factor of 1.52) and mice (factor of ca. 2.5
3.5). In both species peak blood concentrations
of DEHP and MEHP were comparable at the
low dose (3040 nmol/g); at the high dose peak
MEHP concentration was approximately twice
as high as that of DEHP. In rats, MEHP metabo-
lites were ca. 10 %compared to the MEHP peak
concentration and AUC. The MEHPmetabolites
were not detectable in mouse blood.
Species differences in metabolism between
rats and mice were also noted by the urinary and
fecal excretion pattern in a study using repeated
Phthalic Acid and Derivatives 23
doses (10 doses, each 200 and 1000 mg/kg). In
rats, the -1-oxidation metabolites, 5-OH- and
5-oxo-MEHP, accounted for 65 % compared to
35 %-oxidation metabolites, both in nonpreg-
nant and pregnant rats, and at both dose levels.
Excretion of metabolites occurred almost exclu-
sively in urine. Unchanged DEHP and MEHP
were excreted in comparable amounts: ca. 10 %
of the administered dose each, and almost ex-
clusively in the feces.
In mice, DEHP was also excreted exclusively
in feces (610 % of the dose), whereas MEHP
(1525 % of the dose) was excreted 2/3 in feces
and 1/3 in urine. Thus, much larger amounts of
DEHP and MEHP were excreted in mice than
in rats. The majority of the MEHP metabolites
was excreted in urine (86 %; feces 14 %). -1-
oxidation metabolites tended to decrease dur-
ing the treatment (from 63 to 55 %) parallel to
an increase in -oxidation (from 37 to 45 %)
[193 197].
From these metabolism studies [193 197]
with DEHP in pregnant and nonpregnant rats
and mice it can be concluded that:
DEHP and its metabolites are rapidly ab-
sorbed in both species as DEHP, MEHP, and
MEHP-derived metabolites.
The systemic exposure to MEHP is higher in
pregnant mice thaninpregnant rats, but DEHP
exposure is equivalent.
Metabolites from -1- (metabolites VI and
IX, i.e. 5-OH- and 5-oxo-MEHP) and -
oxidation pathways are excreted at higher lev-
els in rats than in mice, but the subsequent -
oxidationof metabolite V(6-carboxy-MEHP)
to form metabolite I (2-ethyl-4-carboxybutyl
phthalate) is more active in mice than in rats.
Repeated DEHP administration induced a
modication of the metabolic pathway:
a decrease of DEHP excretion in nonpreg-
nant and pregnant rats and in pregnant mice
and an increase in nonpregnant mice
an increase of MEHP excretion in pregnant
mice
an increase of -1-oxidation in rats and a
decrease in mice
an increase of -oxidation in pregnant mice
at the high dose level.
An unpublished 65-week oral-dose toxicity
study of DEHP in marmosets [198, 199] de-
scribed pharmacokinetics and organ distribution
following single oral doses of 100 and 2500
mg/kg DEHP to animals aged 3 and 18 months;
a third group was treated at 18 months after a 65-
week pretreatment with unlabeled DEHP. Ring-
labeled
14
C-DEHP was used to determine ra-
diolabel levels in blood (at 1, 2, 4, 8, 12, 24,
48, 72, 120, and 168 h after dosing) and in exc-
reta. Organ distribution including male and fe-
male reproductive organs was examined 2 h af-
ter a second dosing, at least two weeks after the
rst treatment. Peak blood levels were reached
within 1 or 2 h after dosing. Plasma half-life
times of the radiolabel ranged between 2.7 and
8 h. The organ/plasma ratio was generally 0.2 to
0.4. The authors called particular attention to the
small amounts of label distributed to the testes
compared to rodents, in which large amounts of
MEHP are distributed to the testes. Pretreatment
for 65 weeks did not alter the distribution [198,
199].
No difference was seen when comparing
blood concentration time courses of pregnant
and nonpregnant rats and marmosets, respec-
tively, receiving single or repeated doses of 30,
500, and 1000 mg DEHP per kg body weight. In
both groups of rats MEHPAUCs were about two
orders of magnitude higher than those of DEHP.
Maximum MEHP blood levels were reached at
0.5, 2, and 4 h in animals receiving 30, 500,
and 1000 mg/kg, respectively. The authors con-
cluded that kinetics were linear for DEHP but
saturated for MEHP, based on increased time to
reach maximum blood levels [200].
The blood concentration time courses of
pregnant and nonpregnant marmosets were sim-
ilar except that the values of MEHP were lower
in the high-dose pregnant group than in the non-
pregnant group. Generally, the AUC of MEHP
was ca. 10- to 20-fold higher than that of DEHP.
Comparison of species revealed that the
AUCs of DEHP were similar in both species,
while those of MEHP were 10-fold higher in
rats than in marmosets [200].
The above conrms earlier studies in which
rats dosed with 2000 mg kg
1
d
1
of DEHP for
14 d by gavage eliminated (on day 13) 56.2 %
of the administered daily dose in the urine, and
58.7 % in the feces. The peak blood level on day
14 was 368 g equivalents of DEHP per gram
blood. In a parallel marmoset study the same
dose regimen led to only 13 g equivalents of
DEHP per gram blood. Tissue levels were also
24 Phthalic Acid and Derivatives
Table 6. Urinary metabolite excretion in a human 24 h after ingestion of DEHP, % of dose
Metabolite Estimated elimination t
1/2
,
h
DEHP dose, g/kg
4.7 28.7 650
MEHP 5 6.26 4.3 7.3
5-OH-MEHP 10 23.1 22.7 24.1
5-oxo-MEHP 10 17.3 13.0 14.6
Mono(2-ethyl-5-
carboyxypentyl)
phthalate
1215 15.5 19.4 20.7
Mono[2-(carboxy-
methyl)hexyl]
phthalate
24 3.7 5.2 3.8
Total percentage of DEHP
dose
65.8 64.6 70.5
much lower in marmosets, and no peroxisome
proliferation or testicular effects could be de-
tected [201].
In humans, urinary excretion of metabolites
is also considered to reect exposure and ab-
sorption. Two volunteers, each ingesting 30 mg
of DEHP, excreted 11 and 15 % of the dose in
the urine within 24 h. Four daily doses of 10 mg
lead to a urinary recovery rate of 1525 %[202].
One volunteer ingesting deuterium-labeled
DEHP (48 mg, 0.64 mg/kg) excreted 47 % of
the dose within 2 d as MEHP (7.34 %), 5-OH-
MEHP (24.7 %), and 5-oxo-MEHP (14.9 %).
Peak level of MEHP in urine occurred 2 h after
dosing; the maximum urinary concentrations of
the MEHP oxidation products occurred after 4
h. Serumlevels of MEHP were higher than those
of its oxidation products at all times, consistent
with the more rapid elimination of polar metabo-
lites. The estimated serum elimination half-life
of all three metabolites was < 2 h [203].
The proportions of -1- and -oxidation
were investigated in the 24-h urine of one vol-
unteer ingesting three different doses (4.7, 28.7,
and 650 g/kg) of deuterium-labeled DEHP
(Table 6). The -oxidation metabolites were
mono(2-ethyl-5-carboyxypentyl) phthalate and
mono[2-(carboxymethyl)hexyl] phthalate. The
results indicate that excreted MEHP alone does
not properly reect the exposure because -1-
and -oxidation contribute signicant and vary-
ing proportions of the entire metabolism. The
authors argued that serum MEHP level is not
a useful marker for DEHP exposure due to its
short half-life. However, they stated that serum
levels were of the same orders of magnitude de-
spite the fact that the human dose was 501000
times lower than in animal studies [378].
There are species differences in Phase-II
metabolism between rats and mice (see above)
and primates. After intravenous administration
of DEHP to the African green monkey, 80 %
of the urinary metabolites were excreted as glu-
curonide conjugates, in contrast to rats, which
did not excrete conjugates in the urine [204].
In mice receiving 200 mg/kg, high proportions
of the urinary MEHP (58 %) and MEHP-derived
metabolites (49 %) were excreted as glucuronide
conjugates [196]. This indicates that species dif-
ferences in the rate of glucuronidation occur and
may cause considerable variations in biological
half-lives and tissue levels. As a consequence,
the higher glucuronidation (detoxication) rate
of the metabolites in primates may contribute
to the observed lower MEHP peak concentra-
tions and AUCs in marmosets compared to rats
[198 200] and the higher resistance to DEHP-
induced toxicity.
6.2.5.2. Acute Toxicity
Having entered the organism, one of the two es-
ter bonds of phthalic diesters is cleaved and the
alcohol is released [184, 185, 187, 202, 205].
Toxicological effects may result from the ph-
thalic acid monoester and its sequel products
and/or from the alcohol. In some cases (e.g.,
dimethoxyethyl phthalate) the impact of the al-
cohol is of greater importance [206 208]. The
acute toxicity of diallyl phthalate is relatively
high (LD
50
, rat, oral: 8001700 mg/kg) and it
is irritating [215, 216], whereas most phthalates
show very low acute toxicity. Acute oral toxic-
ity in the rat and in the rabbit is summarized in
Table 7.
Phthalic Acid and Derivatives 25
Table 7. Acute toxicity of phthalate esters [209]
Phthalate ester LD
50
(rat, oral),
mg/kg
LD
50
(rabbit,
dermal), mg/kg
Diallyl phthalate 1 500
Dimethyl phthalate >5 000 >12 000
Diethyl phthalate >9 000 >20 000
Dibutyl phthalate >8 000 >20 000
Diisobutyl phthalate >15 000 >10 000
Diethylhexyl
phthalate
>30 000 >24 500
Diisononyl phthalate >10 000 >3 100
Diisodecyl phthalate >20 000 >3 600
Diundecyl phthalate >15 800 >7 900
6.2.5.3. Irritation and Sensitizing Potential
As a rule, most phthalates do not cause signi-
cant irritation [210 212] or sensitization of the
skin [213, 214]. Diallyl phthalate [215, 216],
however, may exert a weak irritating potential.
Earlier reports of respiratory sensitizations ap-
pear to be related to contamination of the esters
with phthalic anhydride and maleic anhydride.
6.2.5.4. Repeated DEHP Dosing
Effects in rodents on liver, kidneys, and testes
were seen, as well as increased incidences of tu-
mors in liver and testes (Section 6.2.5.6) and tox-
icity to reproduction (Section 6.2.5.7). Subacute
and chronic effects as well as reproductive ef-
fects depend to some extent on chain length and
structure of the alcohol moiety. Thus, a differen-
tiation between short-, medium-, and long-chain
esters appears to be justied (Section 6.2.5.8).
Generally, the effects of phthalates are due to the
monoester and metabolites derived therefrom,
rather than the liberated alcohol and its metabo-
lites. Exceptions tothis rule are the more toxic al-
lyl alcohol and the teratogenic methoxyethanol.
2-Ethylhexanol leads to 2-ethanoic acid, which
is known to have a potential for teratogenicity.
However, this is insufcient to explain the tera-
togenic effects seen with DEHP.
DEHP and several other phthalates act on the
liver lipid metabolism in a qualitatively simi-
lar manner to a group of structurally unrelated
compounds which are used or investigated as hy-
polipidemic drugs to reduce serum cholesterol
and triglycerides. Increased enzyme activities
involved in mitochondrial and peroxisomal fatty
acid - and -oxidation appear to be early ef-
fects of DEHP [217 219]. Initially, short tran-
sient inhibition of -oxidation and lipid accu-
mulation may occur, but subsequently can be
overcome [220 222]. In rodents, but not in
primates, the numbers and volumes of peroxi-
somes increase, and the liver undergoes a short
period of DNA replication and cellular prolif-
eration with a maximal mitogenic response af-
ter 24 h [223 226]. This results in a dose-
related increase in liver weight which may dou-
ble [224 231]. Plasma cholesterol values de-
crease in the same way as is observed with hy-
polipidemic drugs.
A study in SpragueDawley rats reports a
continuous increase of peroxisomal enzyme ac-
tivities upon prolonged treatment over two years
[232]. In Fischer rats, increased peroxisomal en-
zyme activity was still present after two years,
but not hepatomegaly or cellular proliferation
[233]. 25 mg/kg of DEHP appears to be the
no-effect level for the increase of peroxisome-
related enzyme activity [228]. The no-effect lev-
els for DINP and DIDP are somewhat higher;
DBP is much less active [234].
Peroxisome proliferation and liver weight
increase occurred in mice receiving DEHP in
the diet (>1000 mg/kg, 24 weeks), but not in
mice which were decient in PPAR (PPAR-
knock-out mice) [235]. Peroxisome proliferator-
activated receptors (PPAR) are physiologically
thought to play a role in lipid homeostasis in that
they induce the expression of enzymes involved
in lipid metabolism (-oxidation, cytochromes
P450). The three different types that are known
to exist [PPAR, - (or -), and -] respond
to endogenous (fatty acids, prostaglandins, leu-
cotrienes) and exogenous ligands (peroxisome
proliferators: brates, phthalate esters, phenoxy
herbicides). The sequence of action is the ini-
tial formation of a proliferatorreceptor com-
plex which becomes a transcription factor after
association with the 9-cis-retinoic acid receptor
[236, 237]. The transcription factor stimulates
the expression of genes having specic promoter
regions (peroxisome proliferator responsive ele-
ments, PPRE). The promoter regions are identi-
cal for all types of PPARs; differences in the en-
zyme induction pattern and intensity result from
differences in the vicinity of the promoter re-
gion, cofactors, and the ligand itself, which, like
26 Phthalic Acid and Derivatives
DEHP, may have overlapping afnity to all three
types of PPAR [238, 239].
In rodent liver, PPAR mediates an in-
crease of cell proliferation, inhibition of apop-
tosis, and increased expression of peroxiso-
mal, mitochondrial, and microsomal enzymes
[231, 240 242]. Under physiological condi-
tions PPAR is equally involved in the growth
regulation of fat cells and epithelial cells of di-
verse organs [239], but its role in processes stim-
ulating and inhibiting tumor growth are not yet
fully understood [243]. PPAR is also involved
in cell cycle regulation and conversion of colon
mucosa cells into carcinoma cells [243, 244].
The concentration of PPARs is high in ro-
dents but low in primates including humans. In
marmosets receiving 100, 500, and 2500 mg
kg
1
d
1
DEHP for 13 weeks (oral, 91 con-
secutive days) peroxisome proliferation was not
detected, neither by morphology nor by bio-
chemistry, except for minor changes indicative
of induction of some enzyme activities (increase
in P450 and microsomal protein content) in all
males and in females at the intermediate and
highdose levels. The number of peroxisomes per
cell was not affected. Only the mean volume of
peroxisomes showed a slight increase (1.31.4-
fold) at 500 and 2500 mg/kg b.w. Gross pathol-
ogy, organ weights, histopathology, and electron
microscopy did not show any substance-related
ndings in any of the treated groups. Especially
the liver, pancreas, andtestes or testicular param-
eters (Sertoli cells, Leydig cells, spermatogonia,
spermatids, spermatozoa) were not affected. The
NOAEL for systemic toxicity was 500 mg kg
1
d
1
in marmosets [245].
On the other hand, it was pointed out that
DEHP can affect multiple signalling pathways
by transcriptional activation of PPAR-regulated
genes (suppression of apoptosis, protooncogene
expression) and can induce biological effects
(morphological cell transformation, decreased
levels of gap junction intercellular communica-
tion) in several tissues independent of peroxi-
some proliferation and receptor binding [246].
Effects in kidneys, including changes in
weight (both increase and decrease), function,
and histopathology, were seen in several rodent
studies (rat, oral, 2.1 mg kg
1
, 3 times per week,
for 312 months; cystic changes and decreased
creatinine clearance [247]; Fischer 344 rat, oral,
147 mg kg
1
d
1
, 104 weeks; increased
weight, chronic nephropathy [248]; mouse,
99 mg kg
1
d
1
, 104 weeks; decreased weight,
chronic nephropathy [249, 250] but not in pri-
mates [245].
6.2.5.5. Genotoxicity and Mutagenicity
DEHP and its metabolites have shown little
if any genotoxic activity in a wide variety of
short-term tests [251 257]. A few in vitro ex-
periments on DEHP and MEHP were weakly
positive and probably related to the cytotoxic-
ity of MEHP [258 261]. The metabolites 5-
OH- and 5-oxo-MEHP and mono(5-carboxyl-2-
ethylpentyl) phthalate were also negative in the
Ames test [262]. Positive results were seen with
MEHP (39 g/mL) in the Comet assay for DNA
single-strand breaks using human leucocytes in
the absence of metabolic activation [263]. Chro-
mosomal aberrationwas seenwithMEHP(3.9
g/mL) in Syrian hamster embryo cells [264].
Two dominant lethal studies [265, 266] using
very high doses and exposure routes of little rel-
evance (subcutaneous, intraperitoneal) showed
positive results in vivo. However, the effects
were of borderline signicance and appear to
reect testicular toxicity rather than true geno-
toxicity. After oral administration of DEHP and
MEHP, nodominant lethal effects were observed
[267, 268].
DEHP has been shown not to react covalently
with rat liver DNA after gavage administration
following a 3-week prefeeding period [269].
When high doses of DEHP, known to cause per-
oxisomal proliferation, were repeatedly admin-
istered, an increase of 8-hydroxyguanosine in
the DNA was found in the rodent liver but not in
the kidney. This could reect an oxidative DNA
alteration [270], which, e.g., is also observed in
rats fed on a choline-free diet [271], and other
conditions of oxidative stress [272], but a second
study on the same end point using more animals
of the same strain, route, and dose (Fischer 344
rats, 12 000 mg DEHP per kg feed) but shorter
duration (11 and 22 weeks, 6 rats per group) was
negative [273]. Determination of 8-hydroxyde-
oxyguanosine in the DNAis very difcult to per-
form. In conclusion, there is no evidence that
DEHP is genotoxic; the few positive results are
contradictory or could result from cytotoxicity
[274].
Phthalic Acid and Derivatives 27
Other phthalates so far investigated showed
similarly negative genotoxicity test results in
most of the experiments [213, 227, 260].
6.2.5.6. Carcinogenicity
DEHP is again the phthalate ester most thor-
oughly investigated. Earlier chronic toxicity
studies of DEHP showed no tumors in Wis-
tar or Sherman rats over two-year periods at
dietary levels of 4000 and 5000 ppm [275,
276]. Sprague-Dawley rats did not respond with
liver tumors or preneoplastic lesions when fed
with doses of 200, 2000, and 20000 ppm for
2 years; a continuous and sustained dose-and
time-dependent peroxisomal proliferation was
observed [232]. Due to the lowanimal numbers,
all these studies are of limited value in assessing
the carcinogenicity risk.
In more recent long-termrat and mouse feed-
ing studies, increased incidences of liver tumors
were seen. For example, increased incidences
of hepatocellular carcinoma, adenoma, and pre-
neoplastic foci were seeninFischer 344rats with
50 animals per dose group and sex, exposed to
6000 and 12000 ppm DEHP in the diet, and in
B6C3F1 mice with 3000 and 6000 ppm [232].
The actual uptake from the feed was estimated
to be 320770 mg kg
1
d
1
in rats, depending
on dose, sex, and food consumption, and 600
1800 mg kg
1
d
1
in mice [277]. A clear dose-
dependent increase in the incidence of hepato-
cellular adenomas was noted in male/female rats
fed 6000 and 12 000 ppm(322/394 and 674/774
mg kg
1
d
1
, respectively), and in male/female
mice fed 3000 and 6000 ppm (672/799 and
1325/1821 mg kg
1
d
1
, respectively) for two
years [278 280]. With a dose of 3000 mg kg
1
d
1
, 78.5 % of the treated rats developed liver
tumors and 29 % pancreatic islet cell adenomas
[281].
The relationshipbetweenhepatic peroxisome
proliferation, cell proliferation, and carcino-
genicity has been evaluated in chronic studies
of DEHP in rats and mice [248 250]. Animals
were fed a diet containing DEHP for 104 weeks;
additional groups treated for only 78 weeks were
subsequently placed on a DEHP-free diet for 26
weeks to examine reversibility of the effects. In
rats, relative liver weight was signicantly in-
creased at 147 mg kg
1
d
1
, and this correlated
with increased activity of palmitoyl CoA oxi-
dase, a peroxisomal enzyme marker. Treatment
with DEHP resulted in hepatocellular adenomas
and carcinomas at weeks 78 and 104; in general,
the incidence of tumors at week 78 was low ex-
cept for the high-dose group. Cessation of treat-
ment resulted in a one-third to one-half decrease
in the incidence of total neoplasia relative to the
groups treated for 104 weeks and in a reduc-
tion in liver weight and palmitoyl CoA oxidase
activity. The increased incidence of hepatic tu-
mors was signicant at the 147 mg kg
1
d
1
dose level and higher in rats [248, 249].
Similar ndings were reported in B6C3F1
mice. Cessation of treatment in mice resulted
in a 50 % decrease in the incidence of total neo-
plasia in males relative to groups treated for 104
weeks, but the incidence in females decreased
only slightly. The increased incidence in hepato-
cellular tumors achieved statistical signicance
at dose levels of 292 mg kg
1
d
1
DEHP [249,
250].
The role of peroxisome proliferation was fur-
ther elucidated in an experiment with PPAR-
null and wild-type [PPAR(+/+)] mice that were
fed 0.1 % Wy-14,643, a model peroxisome pro-
liferator. After 11 months of feeding 100 % of
the wild-type animals had multiple hepatocel-
lular neoplasms, including carcinomas and ade-
nomas, while the PPAR(/) mice were un-
affected [282].
Leydig cell tumors occurred in high inci-
dences of 9095 %in untreated F-344 rats and in
animals at intermediate DEHP dose levels. The
incidence was less than one-third in the high-
dose groups [248, 249, 278, 280]. High inci-
dences of this tumor are specic for this strain
[283].
Similar results at comparable dose ranges
(1.22.0 % in the diet) were obtained in con-
secutive studies with DEHP in smaller collec-
tives of Fischer rats [230, 233, 284, 285] and in
bioassays with two different forms of diisononyl
phthalate and with a phthalate of a mixture of
heptanols, nonanols, and undecanols (C
7
C
11
fraction) [210, 286]. Other investigators [287]
employing lower dose levels (up to 6000 ppm)
did not nd a statistically signicant increase
of liver tumors with a diisononyl phthalate. Un-
der these conditions (6000 ppm over two years)
the material did not exhibit peroxisome prolif-
eration at the end of the 2-year observation pe-
28 Phthalic Acid and Derivatives
riod. Different forms of diisononyl phthalates
and related materials may exhibit slightly dif-
ferent proles in relation to their structure and
the type of branching and molecular shape.
Butyl benzyl phthalate, which is not a sig-
nicant peroxisome proliferator, caused some
mononuclear cell leukemia in Fischer rats at
1.2 % in the diet, but no liver tumors [210, 288].
Diallyl phthalate, similarly, did not cause
liver tumors, but necrotic and brotic lesions of
the liver and bile duct hyperplasia in rats, and
gastric inammation and hyperplasia in mice
were observed. These effects were probably me-
diated via release of allyl alcohol. The doses in
this study were 100 and 300 mg/kg in rats, and
50 and 150 mg/kg in mice [171].
In hamsters, inhalation of atmospheres satu-
rated with DEHP vapor (ca. 15 mg/m
3
; ca. 710
mg/kg) for two years or intraperitoneal injection
up to a total dose of 54 g/kg, administered by
weekly injections, did not increase the numbers
of tumors [251]. Similarly, feeding of compara-
bly lowlevels (300 and 1000 ppm) of DEHP did
not lead to liver tumors [284].
Initiation/PromotionExperiments. No ini-
tiating activity was found with DEHP after
single oral administration of 10 g/kg or af-
ter 12 weeks feeding of 1.2 %, followed by
0.05 % feeding of phenobarbital as promoter
[289]. Generally, peroxisome proliferators have
been recognized as a group of epigenetic ro-
dent liver carcinogens which widely differ in
their carcinogenic activity [290, 291]. Tumor-
promoting activity of DEHP was investigated
with several experimental procedures. The re-
sults were either negative or weak [292 296];
in some regimens antipromoting activities were
also recorded [295, 297, 298]. There is evidence
that peroxisome proliferators exert signicant
promoting effects only late in the rodents life
[299]. Di-n-octyl phthalate, which is not a per-
oxisome proliferator, was shown to be a liver
tumor promoter [300], probably due to lipid ac-
cumulation and hepatomegaly [201].
There are several hypotheses on the mecha-
nism of action. The most important and most
complex seem to be the receptor-mediated
mechanisms (Section 6.2.5.4), which are also as-
sociated with liver weight increase and replica-
tive DNA synthesis [225]. Some authors fo-
cused on reduced expression of growth factor
receptors [292, 301, 303]. Another hypothesis
is that oxidative stress plays a role. Increased
intracellular H
2
O
2
and reactive O
2
species re-
leased from the peroxisomes [290, 304 306]
may cause continuous accumulation of 8-hy-
droxydeoxyguanosine in the DNA [270]. Leak-
age of H
2
O
2
fromperoxisomes is reectedbyin-
creases in oxidized glutathione in the bile [291].
Conjugated dienes and accumulation of lipofus-
cin pigments in lysosomes were also observed
and indicate oxidative stress to the cell [224,
277, 291, 307]. It is, however, not clear whether
oxidative stress is cause or consequence in the
course of phthalate ester tumorigenesis.
No direct genetic activity is observed with
phthalate esters (see also Section 6.2.5.5). This
also applies to hypolipidemic drugs, even if they
stimulate peroxisomes at very lowdoses and are
strongly carcinogenic in rats and mice [291].
These rodent liver tumors are therefore regarded
as epigenetic anda sequel of the liver effects. The
present knowledge allows the assumption that
threshold levels for these early-appearing liver
effects also reect a threshold for the appearance
of liver tumors in rodents. In addition, phthalate
esters do not cause peroxisome proliferation in
nonrodents including primates.
6.2.5.7. Reproductive Toxicity
Testicular Toxicity. Phthalate esters of
medium and longer chain length (C
3
C
6
, DEHP) cause testicular damage [290,
308 322]. The lesion is characterized by an
initial effect on Sertoli cells [323, 324] leading
to the exfoliation of spermatocytes and sper-
matids with atrophic degeneration of seminifer-
ous tubules containing cytoplasmic remnants of
Sertoli cells and spermatogonia [323, 325].
Minimal vacuolization of Sertoli cells was re-
ported to occur in rats receiving 38 mg kg
1
d
1
DEHP for 13 weeks; at a 10-fold higher
dose level, 375 mg kg
1
d
1
, mild vacuolization
and testicular atrophy were noted. The authors
derived a no-adverse-effect level (NOAEL) of
3.7 mg kg
1
d
1
. However, because the same
effects were seen in control animals in a sec-
ond experiment of the same study[333], the true
NOAEL should be considered to be 37.6 mg
kg
1
d
1
. Germ cell detachment may also be
observed in vitro with MEHP [323, 334] but
Phthalic Acid and Derivatives 29
Table 8. Effects of DEHP on male fertility
Species, strain Dose, mg kg
1
d
1
(duration) Effect (dose, mg kg
1
d
1
) Ref.
Rat, SD 0, 143, 737 (17 weeks) testes weight (737) [326]
Rat, F-344 0, 322, 674 (103 weeks) atrophy of seminiferous tubules
(674)
[280]
Rat, F-344 0, 330, 1000, 3000 (13 d) slight atrophy of testes (1000) [327]
Rat, F-344 0, 320, 1250, 5000, 20000 (60 d) testes weight , testosterone, LH
, FSH (284)
[328]
Rat, SD 0, 10, 100, 1000, 2000 (5 d) palmitoyl CoA oxidase (10), liver
weight (100); testes weight , no.
Sertoli cells , spermatocytes ,
spermatids , testicular zinc
(1000)
[324, 329]
Rat, Wistar 0, 50, 100, 250, 500 (30 d) slight testes weight , change of
testicular enzymes (50); signicant
testes weight (100);
histopathological changes, testes
(250)
[330]
Rat, F-344 0, 6, 29, 147, 789 (104 wk) bilateral aspermatogenesis in aged
rats (29); testes weight , bilateral
aspermatogenesis (789)
[248]
Rat, Wistar 0, 113, 340, 1088 advanced atrophy of seminiferous
tubules, aspermia in epididymes
(1088)
[331]
Rat, SD 0, 0.12, 0.78, 2.4, 7.9, 23, 77, 592,
775 (unspecied weeks)
sperm density ; male accessory
organ weights ; atrophy of
seminiferous tubules; sloughed
epithelial cells in epididymis (592)
[332]
Marmosets 0, 100, 500, 2500 (91 d) testes and testicular parameters
(Sertoli cells, Leydig cells,
spermatogonia, spermatids,
spermatozoa) not affected (2500)
[199]
not with2-ethylhexanol [335]. Further hormonal
sequels of the testicular atrophy are decreases
of serum testosterone and increases of serum
FSH (follicle stimulating hormones) and LH
(luteinizing hormones) [328]. Some of the early
andlate effects andthe respective dose levels fol-
lowing administration of DEHP are summarized
in Table 8. Extensive studies have been con-
ducted on the effects on Sertoli and Leydig cells,
as well as testosterone and sex hormone regula-
tion in male rodents. The absence of any effect in
marmosets [199] indicates that marked physio-
logical and endocrine differences exist between
rodents and primates regarding Sertoli cell and
Leydig cell physiology and susceptibility, and
sex hormone regulation. Rodent Leydig cells are
regarded as more sensitive than human Leydig
cells [283]. Atrophy of the seminiferous tubules
is generally seen in rodents at doses >250 mg
kg
1
d
1
. The exact mechanismis still unknown
[209].
Early functional effects, e.g., reduced sperm
counts, may be seen at lower dose levels,
whereas in acute studies relatively high (> 250
mg/kg) oral doses are required to induce visible
testicular damage. In experiments of longer du-
rationthe effective levels are considerablylower.
Serum testosterone [328] and male fertility in
continuous breeding protocols appear to be sen-
sitive parameters. The no-effect levels (NOAEL)
are in the range 2550 mg kg
1
d
1
. A dose of
1250 ppm (70 mg kg
1
d
1
) in the diet of rats
decreased serum testosterone within 60 d [328].
This is a dose range whichcausedmalformations
in a mouse teratology study [304, 336].
Reversal of testicular effects is slow and in-
complete. After administration of 2000 mg kg
1
d
1
DEHP for two weeks to rats, only slight re-
versal of the testicular lesions occurred after 45
d [318]. Incomplete reversal was also observed
after 60-d feeding of 20 000 ppm of DEHP in
the diet of rats and a 70-d postobservation pe-
riod [328]. The effects are age- and, to some ex-
tent, species-related [321]. Young rats (4-weeks-
old) appear to be more susceptible (lower doses,
shorter duration required) than adult rats. This
age-dependent susceptibility has been success-
fully modeled with MEHP in primary testicular
cultures from rats of varying age [334].
30 Phthalic Acid and Derivatives
As to species differences, DEHP can also
induce testicular atrophy in mice [315, 337],
guinea pigs [325], ferrets [338], and hamsters
[229]. Marmosets appear to be more resistant;
oral doses of DEHP (2000 mg/kg for 2 weeks
[201]; 2500 mg kg
1
d
1
for 91 consecutive
days [199]) causednotesticular toxicity. Alower
intestinal resorption and higher rate of excretion
may contribute to these ndings [218].
The mechanism of testicular toxicity is still
unclear, but zinc is thought to play a role. Testic-
ular zinc content is decreased while urinary zinc
increases [308, 314, 319]. Detailed studies with
di-n-pentyl phthalate [339] have not been able
to indicate whether the changes in testicular zinc
content are solely a secondary phenomenon. A
zinc-supplemented diet partially protected rats
fromDEHP-inducedtesticular damage, whereas
rats on a zinc-decient diet were more sensi-
tive [327]. Hypolipidemia and hepatomegaly ap-
peared to occur independently of zinc supple-
mentation. On the other hand, there is evidence
that reduced testicular zinc content was probably
a result of testicular toxicity after DEHP treat-
ment rather than a mediator of DEHP toxicity
[340].
Diethyl and di-n-octyl phthalate showed no
adverse effects on fertility and the testes [337,
345]. Comparative studies using several mono-
and di-esters were conducted in male Sprague-
Dawley rats, orally treated for four weeks with
phthalate diesters at 500 mg kg
1
d
1
or ph-
thalate monoesters at 250 mg kg
1
d
1
[313].
Animals were evaluated for body weight, testic-
ular and epididymal weights, epididymal sperm
count, andspermmotionparameters. There were
no signicant effects of treatment on body or re-
productive organ weight. The order of effects
on sperm motility for the diesters was DEHP >
DnOP > DEP > DUP > DIDP > BBP, and for
the monoesters MBuP >MEP >MEHP (MBuP
= monobutyl phthalate; MEP = monoethyl ph-
thalate) [313].
The relevance of rodent data to humans is still
under debate. The absence of testicular effects
in monkeys even at 2500 mg kg
1
d
1
[199],
and the failure to conrm vacuolization of Ser-
toli cells in rats [331, 332] are major factors in
this discussion because the NOAEL of 3.7 mg
kg
1
d
1
for vacuolization was used by several
institutions during their evaluations and risk as-
sessments [274, 342, 343].
Effects on Fertility and Development.
There are two high-quality rat studies available
which deserve a more detailed description.
In a two-generation reproductive study in
Wistar rats DEHP was administered in feed at
0, 1000, 3000, and 9000 ppm, resulting in esti-
mated intakes of 0, 113, 340, and 1088 mg kg
1
d
1
[331]. Cauda epididymal sperm were as-
sessed for motility, morphology, and head count
in F0 and F1 males at necropsy. Testicular sperm
head count was also performed. Male pups were
evaluatedfor the presence of nipples andareolas.
Day and weight at vaginal opening and preputial
separation were assessed in pups. F2 pups were
also evaluated by functional observation battery
and motor activity testing, and in a water-maze
test of learning and memory.
The high-dose level (9000 ppm) was associ-
ated with a decrease in feed consumption and
weight gain at several intervals during the study.
F2 pups in the high-dose group were smaller and
gained less weight from birth through the as-
sessment of functional observation battery and
water-maze testing. Grip strength was reduced
in males and hind-limb splay was reduced in
both sexes in the high-dose group. There were
no other treatment-related ndings in the func-
tional observation battery or in the water maze.
Differential ovarian follicle counts of F0 and F1
adults showed a decit in growing follicles and
corpora lutea in the high-dose group.
It was concluded that reproductive perfor-
mance and fertility were affected at the 9000
ppm dose level (1088 mg kg
1
d
1
) with a
NOAEL of 3000 ppm (340 mg kg
1
d
1
), and
that developmental toxicity was noted at 3000
and 9000 ppm, with an increase in stillbirth, an
increase in postnatal day (PND) 04 pup mor-
tality, retardation of F2 pup body weight, al-
tered male anogenital distance, and retained nip-
ples/areolas. A delay in sexual maturation was
also noted in F1 offspring at 9000 ppm. The
NOAEL for developmental toxicity was consid-
ered by the study authors to be 1000 ppm, as was
the NOAEL for systemic toxicity (113 mg kg
1
d
1
) [331].
In a multigeneration continuous-breeding
study Sprague-Dawley rats were fed diets con-
taining 1.5 (control group exposed to back-
ground DEHP levels), 10, 30, 100, 300, 1000,
or 7500 ppm DEHP from the rst day of the
Phthalic Acid and Derivatives 31
study until the day of necropsy [302, 332]. Due
to a lack of reproductive effects in the rst lit-
ter produced, the study was repeated with two
additional doses, 1.5 (control) and 10 000 ppm.
Ranges of DEHP intake in the F0, F1, and F2 an-
imals were estimated at 0.090.12, 0.470.78,
1.42.4, 4.87.9, 1423, 4677, 392592, and
543775 mg kg
1
d
1
. At about ve weeks
of age, F0 rats were fed the DEHP-containing
diets for six weeks prior to mating and were
then cohabitated for nine weeks. The rst two
litters delivered during the cohabitation period
(F1a and F1b) were counted, weighed, assessed
for anogenital distance, and then discarded. The
third litter (F1c) was raised by the dam. Fol-
lowing weaning of pups, vaginal cytology was
monitored in F0 females for 14 d. After comple-
tion of crossover studies described below, some
of the F0 rats were necropsied. Sperm analy-
ses were conducted, and organs were collected
for histopathological evaluation. F1 pups were
counted, weighed, and examined for anogenital
distance and nipple retention during the lacta-
tion period. On PND 16, one female per litter
was evaluated for vaginal opening, and a sec-
ond was selected for F1 mating. One male per
litter was selected for mating, and 4 or 5 males
per litter were evaluated for testicular descent
and preputial separation. At weaning on PND
21, pups were given diets containing the same
DEHP concentrations as their parents. On PND
81, the F1 rats chosen for mating (17 per sex
per group) were randomly assigned to breeding
pairs (preferably nonsibling) and cohabited for
nine weeks. The study conducted in F0 parents
and F1 offspring was repeated in F1 parents and
F2 offspring, except that no F3 offspring were
mated. Selected F3c males were necropsied on
PND 6364 and selected females on PND 60
74.
During numerous time periods of the study
and especially at necropsy, body weight gains
were decreased in rats fromthe 7500 and 10 000
ppm groups. Dam body weights during delivery
and lactation were decreased by 820 % in the
F0 10 000 ppm group. Increases and decreases
in feed intake were observed at most dose levels.
In the F0 7500 and 10 000 ppm groups, feed in-
take was decreased during lactation. The liver
was identied as a target of toxicity with in-
creases in liver weight and hepatocellular hy-
pertrophy observed at dose levels 1000 ppm.
Changes in organ weights and lesions were also
observed in the kidney at 7500 ppm and the
adrenal gland at 10 000 ppm. The only reproduc-
tive effects observed in the F0 parents occurred
at 10 000 ppm and included decreases in sperm
counts and velocity, reductions in testis and epi-
didymis weights, and increased numbers of rats
with small testes. Histopathological ndings as-
sociated with small testes included minimal to
marked atrophy of seminiferous tubules charac-
terized by loss of germ cells.
The lowest dose level producing dose-related
effects in F1 offspring was 7500 ppm, and those
effects included decreases in number of live
pups per litter, reducedmale anogenital distance,
and delays in vaginal opening, preputial sepa-
ration, and age at testicular descent. Additional
effects noted in the F1 offspring fromthe 10 000
ppm group included decreased live pup weight
at birth and during the lactation period and in-
creased ratio of female anogenital distance to
body weight. Fertility was compromised in the
F1 rats from the 10 000 ppm group, which did
not produce any viable litter. Other effects ob-
served in F1 parents were similar to those ob-
served in F0 parents but occurred at lower dose
levels. Histopathological ndings observedinall
animals of the 7500and10 000ppmgroups were
consistent with those observed in the F0 genera-
tion and included minimal-to-marked seminifer-
ous tubule atrophy and occasional sperm-release
failure. Additional reproductive effects observed
in F1 rats were reduced sperm counts at 7500
ppm and higher and increased uterus and ovary
weights at 10 000 ppm.
In the F2 pups, delays in preputial separa-
tion and testicular descent occurred at each dose
level. Other effects occurring in F2 pups of the
7500 ppmgroup included delayed vaginal open-
ing and reductions in live pup weight at birth and
during the lactation period, male anogenital dis-
tance, and survival during the lactation period.
In the F3 offspring, seminiferous tubule atro-
phy was observed in 10/10 males from the 7500
ppm group. All other reproductive effects in F2
adults occurred at 7500 ppm and included de-
creases in pregnancy index, the number of litters
per pair, male reproductive organweights, sperm
counts, and sperm motility. The F3 generation
was the only generation in which there was an
increase in males with nipples. Other F3 effects
included reduced sperm counts and weights of
32 Phthalic Acid and Derivatives
dorsolateral prostate, testis, and epididymis at
7500 ppm.
The study authors discussed the relevancy of
small male reproductive organ sizes observed in
both F1 and F2 rats of the 300 ppmgroups. They
noted that although incidences were low, the
effects were consistent with phthalate-induced
toxicity. The incidence of small testes and epi-
didymides exceeded historical control data from
the laboratory. Therefore, the effects were con-
sidered as potentially treatment-related. How-
ever, the study authors concluded that the over-
all signicance of the effects cannot be deter-
mined due to lack of histopathological data and
lack of adverse reproductive effects at 300 and
1000ppm. Usingcrossover studies, the male was
identied as the sex in which adverse fertility ef-
fects occurred.
In conclusion, DEHP was clearly a repro-
ductive and developmental toxicant at 7500 and
10 000 ppm based on changes in fertility and
pregnancy indices, litter data, sperm parame-
ters, sexual development, and/or histopatholog-
ical changes in testes. Intake at 7500 ppm was
estimated at 392592 mg kg
1
d
1
, and that at
10 000 ppm at 543775 mg kg
1
d
1
. Based
on the incidence of testicular abnormalities, the
CERHR Expert Panel considered 300 ppm to be
an effect level, giving a NOAEL of 100 ppm,
about 35 mg kg
1
d
1
[344]. Sertoli cell vac-
uolation, which was the end point driving the
LOAEL in the original CERHR Expert Panel
evaluation of DEHP [342, 344], was not in-
creased by DEHP treatment in this study [302,
332].
In CD-1 mice exposed to 0.3 % DEHP in the
feedina continuous breedingprotocol, complete
suppressionof fertilityat the highest dose (430
mg kg
1
d
1
) was observed. At 0.1 % a signif-
icant reduction in fertility was induced. No ef-
fects were seen at 0.01 % ( 20 mg kg
1
d
1
).
A cross-over mating employing treated males
and females fromthe 0.3 %groups indicated sig-
nicant effects on both the male and female re-
productive function [304, 345].
DBP exposure in mice resulted in a reduction
in the number of litters at 1.0 % in the diet, but
not at 0.3 %. Females were more affected than
males [345]. Di-n-hexyl phthalate affected both
sexes of mice in a dose-related fashion [345] at
0.3, 0.6, and 1.2 %. Di-n-pentyl phthalate com-
pletely inhibited fertility at 1.25 % and 2.5 %.
Di-n-propyl phthalate caused complete inhibi-
tion at 5.0 % and reduced fertility at 0.5 % and
2.5 %. Di-n-pentyl phthalate was more toxic to
males, and di-n-propyl phthalate more toxic to
females [337]. Diethyl and di-n-octyl phthalate
showed no adverse effects on fertility and the
testes [337, 345].
An abnormal sexual development was seen
in male rats with in utero and lactational ex-
posure to DEHP (oral gavage, dose levels 0,
375, 759, or 1500 mg kg
1
d
1
, dosing dur-
ing gestational day 3 through postnatal day
21) [346]. Dose-related effects on male off-
spring included reduced anogenital distance,
areola and nipple retention, undescended testes,
and incomplete preputial separation. Testis, epi-
didymis, glans penis, prostrate and seminal vesi-
cle weights were reduced at postnatal days
21, 63, and/or 105112. Further increased inci-
dences of prostrate and seminal vesicle agenesis,
reduced sperm counts, and testicular, epididy-
mal and penile malformations were noted. The
observationthat manyof the exposedmales were
sexually inactive was suggested to reect an in-
hibition of the sexually dimorphic central ner-
vous systemdevelopment. DEHP was attributed
an anti-androgenic action the anogenital dis-
tance reduction was only seen in male rats, to-
gether with the retention of areolas and nipples,
whereas DEHP had no signicant effect on vagi-
nal opening or the rst estrus. The exact mode
of action is still unclear, but anti-androgenicity
must not be confusedwithestrogenicity[346]. In
an uterotrophic assay in ovariectomized female
rats, DEHP (1000 mg kg
1
d
1
for 5 d) had no
estrogen activity as it failed to change sex hor-
mone levels or uterus estrogen receptor levels,
or to increase uterus weight, and nally there
was no difference compared to controls during
histopathology in uterus [347, 348].
Prenatal Toxicity; Embryotoxicity; Ter-
atogenicity. Embryotoxic and teratogenic ef-
fects of DEHPandDBPdependonspecies, dose,
and route of administration. Mice are more sen-
sitive than rats, and feeding appears to be more
effective than gavage or inhalation.
In mice, DEHP and DBP caused terato-
genicity after oral administration [337, 345,
349 353]: In CD-1 mice 0.1 and 0.15 % of
DEHPin the diet produced both maternal and fe-
tal toxicity [304, 352]. Malformations (eye and
Phthalic Acid and Derivatives 33
tail defects, exencephaly, aortic and pulmonary
arch defects, and skeletal defects) were observed
at 0.05 %in the absence of maternal or fetal tox-
icity. No effects were reported at the lowest di-
etary level in this study (0.025 % DEHP, equiv-
alent to ca. 35 mg kg
1
d
1
) [354].
DBP caused reduction of the numbers of lit-
ters and of live pups per litter at the 1.0 % level,
but not at 0.3 % in a continuous breeding study
[345]. In another feeding study in mice, with
1 % DBP in the diet, a slight increase in mal-
formations (exencephaly and spina bida) was
observed [349, 350].
Butyl benzyl phthalate at 2 % in the diet of
rats caused complete resorption of all litters (em-
bryonic or fetal death), whereas 1 % and 0.5 %
were without effect [190]. In another rat feed-
ing study [355] 2 % of butyl benzyl phthalate
in the diet did not lead to complete resorption
of the litters; the number of malformations was
greatly increased. In mice, malformations oc-
curred, with a dose-dependent increase at 0.5
and 1.25 %. The maternal and developmental
no-adverse-effect level was 0.1 % [355].
In rats, only embryotoxic effects were ob-
served, with little indication of teratogenicity,
even at dose levels that led to some maternal tox-
icity [331, 336]. Postimplantation loss was sig-
nicantly increased in F0 and F1 dams at 1088
mg kg
1
d
1
, and viability of the F1 and F2
pups was reduced in the groups at 340 mg kg
1
d
1
. Genital organ malformations (hypospadia,
cleft prepuce, small penis) were noted in only
two high-dose F2 pups [331].
The number of fetuses with external, soft tis-
sue, and skeletal malformations was increased
in a Wistar rat study [0, 40, 200, 1000 mg kg
1
d
1
, exposure during gestational day (GD) 6
15]. Malformations were seen at the maternally
toxic high-dose level, and were mainly conned
to the tail, brain, urinary tract, gonads, vertebral
column, and sternum [356].
Following exposure to DEHP in utero (GD
14 to PND 3, 750 mg kg
1
d
1
) malforma-
tions of reproductive organs were observed in
82 % of DEHP-treated males at necropsy. The
types of malformations included permanent nip-
ples, clefting of phallus and hypospadias, vagi-
nal pouches, agenesis of prostate, seminal vesi-
cles, or coagulating glands. Sperm production
and sperm numbers were reported to be not af-
fected by DEHP treatment, as was sexual be-
havior in adult rats, except that males with
malformed penises were unable to achieve in-
tromission. Testicular defects included hemor-
rhage, granuloma, brosis, reduced size or atro-
phy, and nondescent associated with abnormal
gubernacula or ligaments. Dihydrotestosterone-
dependent end points, i.e., reduced AGD and
areolas, were affected [357]. Inhalation of
DEHP aerosols up to 300 mg/m
3
did not cause
embryotoxic or teratogenic effects [300].
As to the mechanism, zinc depletion and in-
terference with sex hormone regulation have
been discussed. Since DEHP in the diet might
interfere with zinc resorption and metabolism
the prenatal toxicity may partly be due to zinc
depletion, which is known to cause teratogen-
ic effects [358]. Dihydrotestosterone-dependent
end points, i.e., reduced AGD and areolas, were
affected by DEHP, BBP, and DINP, but not by
DMP, DEP, and DOTP [357].
No teratogenicity or embryotoxicity was
found with dimethyl [359] and diethyl phtha-
late [360] in rat feeding studies (0.255.0 % in
the diet) or with dimethyl phthalate in rats af-
ter occlusive epicutaneous administration (0.5,
1.0 and 2.0 mL kg
1
d
1
; 2 h/d; gestation days
615) [361].
Di(methoxyethyl) phthalate is teratogenic
due to the release of highly teratogenic meth-
oxyethanol [206 208].
6.2.5.8. Effects of Phthalate Esters by
Groups
Relations between structure and effects of ph-
thalate esters seem to exist, and it appears to be
justied to group themaccording to the structure
of the alcohol moiety, i.e., linear chainlengthand
structure (linear or branched).
Short-Chain Phthalate Esters (C
1
C
3
).
Dermal exposure to dimethyl phthalate in hu-
mans occurs widely due to its use in insect re-
pellants; so far, no signicant toxicological side
effects were observed [213]. Dimethyl and di-
ethyl phthalate are used in cosmetics at concen-
trations of < 10 % and are dermally resorbed.
They are considered to be nonirritants, nonsen-
sitizers, and nonphototoxic agents [213]. Feed-
ing studies in rats and mice with diethyl phtha-
late caused liver weight increases at high doses
34 Phthalic Acid and Derivatives
( 0.5 % in the diet), but no testicular toxic-
ity [308 310, 345]. Peroxisome proliferation
is only marginal and requires high doses [223].
Developmental toxicity is regarded as insigni-
cant [359 361].
Dimethoxyethyl phthalate has a teratogenic
potential [206 208] due to the released meth-
oxyethanol.
Diallyl phthalate is acutely relatively toxic
(LD
50
, rat, oral: 8001700 mg/kg) and irritating
[215, 216]. In a long-term bioassay with gav-
age administration there was no clear carcino-
genic activity, but allyl alcohol-related toxicity
was observed (see Section 6.2.5.6).
Medium-Chain Phthalate Esters (C
4
C
7
).
Di(2-ethylhexyl)phthalate (DEHP) is the sub-
stance examined most extensively, and countless
publications on toxicological prole, biochem-
istry, and metabolism exist. Far fewer data have
been generated on any other phthalate ester.
In rodents, but not in other species, hepatic
alteration (peroxisome proliferation and hep-
atomegaly[210, 227, 228, 317]) is observed. The
peroxisome proliferation is somewhat more pro-
nounced in the group of longer chain phthalates,
especially of the branched-chain type.
Branched-chain phthalates commonly share
a potential to induce peroxisomes and hep-
atomegaly in livers of rats and mice at high
doses. This was demonstrated with diethylhexyl
phthalate (DEHP), diisononyl phthalate and di-
isodecyl phthalate. These materials may cause
peroxisome proliferation and hepatomegaly in
rodents (DEHP at doses exceeding 50 mg kg
1
d
1
) [201, 210, 223 230, 232, 234, 278]. In
long-term feeding experiments, DEHP [362]
and related phthalates [210] also led to an in-
crease of liver tumors in rodents, mediated by
receptor PPAR. This mechanism is specic to
rodents and not relevant to humans. The role
of peroxisome proliferator activated receptors
(PPARs) is outlined in Section 6.2.5.4. Nonro-
dents and primates have no or very low levels of
PPARs and do not show these liver effects [189,
201, 290, 363, 364]. Hence no carcinogenic-
ity is expected for humans, though some uncer-
tainty remains because there might be receptor-
independent effects (see Section 6.3.2).
Apart fromthe liver, effects were also seen in
kidneys and the thyroid gland. Chronic admin-
istration of 0.5 % of DEHP in the diet also led
to kidney toxicity in rats and mice with organ
weight increase, cystic changes, dysfunctions
[365 368], and alterations in thyroids [366,
369], with a fall in plasma thyroxine levels as-
sociated with hyperactivity and changes of the
thyroidal colloid spaces. Effects on the thyroid
gland were also observed with other peroxisome
proliferators [369].
The straight-chain analogues di-n-hexyl and
di-n-octyl phthalate also showthyroid effects al-
though they differ in terms of hepatic toxicity
[226, 366]. Di-n-octyl phthalate causes a differ-
ent type of renal toxicity than DEHP [370].
Medium-chain phthalate esters exhibit de-
generative testicular effects upon repeated oral
administration in mice, rats, and hamsters
[308 320]. The testicular effects are also ob-
tained with DEHP (C
8
, branched) to a simi-
lar degree but are not observed with the linear
longer chain di-n-octyl phthalate or with short-
chain phthalates. Testicular effects were noted
with DEHP after oral intake in rats and mice,
[234, 309, 313, 317, 318, 321] but were lacking
in monkeys [199]. Oral uptake of >50 mg kg
1
d
1
of DEHP also caused embryotoxic and ter-
atogenic effects in mice [337, 345, 349 353].
In rats, some embryotoxicity was observed in
feeding studies at maternally toxic levels [352],
but not upon inhalation [300].
Dibutyl phthalate (DBP) also caused embryo
toxicity and teratogenicity at a concentration of
1 %in the diet of ICR/ICL mice [315, 349, 350].
After a 6-month inhalation exposure of rats to
50 mg DBP/m
3
for 6 h per day, an accumula-
tion of DBP in the brain and increase of brain
weight were found; at 5 mg/m
3
these effects
were marginal [371].
Butyl benzyl phthalate at 2.5 and 5.0 % in
the feed caused testicular effects, enlargement of
liver and kidneys and decrease of bone marrow
cellularity and thyroid function in rats [372].
Longer Chain Phthalate Esters (C
8
C
12
).
Di-n-octyl phthalate did not exert testicular tox-
icity in vivo [337, 341, 345] nor signicant per-
oxisome proliferation in the liver [362]. It was
shown to be a liver tumor promoter in a rat feed-
ing experiment [373], probably as a sequel of fat
accumulation and hepatomegaly [226].
The straight-chain DEHP analogues di-n-
hexyl and di-n-octyl phthalate also showthyroid
effects although they differ in terms of hepatic
Phthalic Acid and Derivatives 35
toxicity [366]. Di-n-octyl phthalate causes a dif-
ferent type of renal toxicity than DEHP [370].
The effects of a group of high molecular
weight phthalate esters (HMWPEs) compris-
ing C
7
C
12
alkyl backbone esters (di-2-pro-
pylheptyl ester; bis(C
7
C
9
) branched and linear
alkyl esters; di-C
11
branched and linear alkyl es-
ters; di-C
11
alkyl ester; bis(C
11
C
14
) branched
and linear alkyl esters, C
13
-rich; di-C
13
alkyl
ester) have recently been assessed as a category
of related chemicals in the course of the global
OECD-ICCA-HPV program [374]
HMWPEs have a low order of acute toxi-
city by the inhalation, dermal, intraperitoneal,
and oral routes of exposure. Chemicals of the
HMWPE category are not irritating to the skin
or eyes (only slight conjunctival irritation for di-
C
13
alkyl ester), nor are they skin sensitizers
(maximization test or comparable, or Buehler
method). Although some data for these end
points are older, the weight of evidence is con-
sistent.
The primary ndings in the repeated-dose
rat studies were in the liver and kidney and to
a lesser degree in the thyroid. Effects on the
liver are indicative of peroxisomal proliferation,
including increased PCoA, liver weights, and
liver hypertrophy, and are not relevant to hu-
mans. It has been shown that these effects are
mediated through peroxisome proliferation ac-
tivated receptor alpha (PPAR) and that lev-
els of PPAR are much higher in rodents than
humans. Thus, one would expect humans to
be substantially less responsive than rodents to
peroxisome-proliferating agents. Empirical evi-
dence for this hypothesis has been provided by
studies in primates, in which repeated adminis-
tration of DINP had no effects on liver, kidney,
or testicular parameters, including peroxisome
proliferation. The kidney effects were a result
of a dose-dependent -2-globulin nephropa-
thy. Such effects are sex- and species-specic to
male rats and also are not relevant to humans.
Thyroid effects are likely to be a compensatory
effect associated with the peroxisomal prolifer-
ation in the liver. The results were consistent for
all members of the category, withNOAELs rang-
ing between 10 and 282 mg
1
kg d
1
. The 10
mg
1
kg d
1
value was delivered in an OECD
422 study in which rats were dosed for 45 d, and
the effect observed at 50 mg
1
kg d
1
was in-
creased liver weight. This is likely to be related
to peroxisome proliferation. All the NOAELs
are driven by liver and/or kidney effects.
HMWPE category members are nongeno-
toxic. All of them have been tested in the Ames
reverse mutation assay using Salmonella ty-
phimurium and all were nonmutagenic with or
without metabolic activation. Similarly, a range
of substances covering the majority of the car-
bon numbers in this category was found to be in-
active in the mouse lymphoma tests. Additional
testing of di-C
13
phthalate ester showed that the
test substance did not induce either structural
chromosomal aberrations or polyploidy in Chi-
nese hamster liver cells up to the limit concentra-
tion of 4.75 mg/mL, in the absence or presence
of an exogenous metabolic activation system.
Although HMWPEs have not been tested
for carcinogenic properties (i.e., chronic toxi-
city or bioassay studies), previous experience
with a wide range of phthalates including di-
n-octylphthalate and DINP suggests that high
doses might produce liver changes in rodents
via interaction with PPAR. However, these
are considered to be not relevant to humans
and not indicative of a potential human risk.
Three chronic toxicity/carcinogenicity studies
on DINP have been conducted: two in rats and
one in the mouse. In the rat studies, the major
ndings were liver and kidney changes princi-
pally related to the induction of peroxisome pro-
liferation. There was an increase in liver tumors
in both male and female rats and also a small
increase in kidney tumors in the male rats. Both
of these tumors are considered to be rat-specic
and without relevance to humans. In the mouse
study, there were liver tumors as well, also the
consequence of peroxisomal proliferation, but
no tumors of other types. It should be noted that
di-n-octyl PE was a liver tumor promoter in a rat
feeding study [373], probably as a sequel of fat
accumulation and hepatomegaly [226].
Although not all members of the category
have been tested for reproductive toxicity (di-
2-propylheptyl, PE, di-C
11
PE, or di-C
13
PE),
there are data for the lower [bis(C
7
C
9
) PE],
intermediate [bis(C
9
C
11
) PE], and higher (di-
C
13
PE) molecular weight representatives indi-
cate no signicant reproductive toxicity at doses
up to 500 mg
1
kg d
1
(or 250 mg
1
kg d
1
di-
C
13
PE). Effects included transiently decreased
body weights or slightly decreased ovary and
epididymidal weights. These effects are minor
36 Phthalic Acid and Derivatives
and are not directly related to reproductive toxic-
ity. Furthermore, the category members bis(C
7
C
9
) PE and bis(C
9
C
11
) PE have been recently
shownnot tobe associatedwithdetectable repro-
ductive effects and do not affect fertility, simi-
larly to DINP and DIDP.
Data from the developmental toxicity tests
for the HMWPE conducted in rats on di-2-pro-
pylheptyl, bis(C
7
C
9
), bis(C
9
C
11
), and di-C
13
PEs, have shown minimal maternal toxicity at
doses up to 1000 mg
1
kg d
1
(limit dose) or
250 mg
1
kg d
1
(di-C
13
PE). Either no ef-
fects were produced or the effects were associ-
ated with decreased food consumption and body
weight loss in dams. Only the di-2-propylheptyl
PE showed maternal toxicity at the limit dose
and associated effects of resorption, decreased
litter size, or fetal survival associated with the
above two symptoms of maternal toxicity. In
the two-generation study on DIDP, a decrease in
offspring survival, more marked in F2, was ob-
served. In the di-C
13
PE one-generation study
(F1 generation) a decrease in survival indices
was observed, leading to a NOAEL of 50 mg
1
kg d
1
for offspring rats, whereas NOAELs for
parental rats were 250 mg
1
kg d
1
. These
may be considered related to developmental ef-
fects. No such changes were seen in either gen-
eration of the separate studies on bis(C
7
C
9
)
and bis(C
9
C
11
) PE; these results are consis-
tent with those from DIDP studies. However,
none of the HMWPEs tested produced develop-
mental effects. Increased frequencies of devel-
opmental variants including dilated renal pelvis
and supernumerary lumbar ribs were produced
in the studies on bis(C
7
C
9
) PE and bis(C
9
C
11
) PE, but are common ndings in rats. Al-
though not all members of the HMWPE Cat-
egory have been tested for developmental tox-
icity, there are data for the lower [di-2-pro-
pylheptyl and bis(C
9
C
11
) PE], intermediate
[bis(C
9
C
11
) PE], and higher (di-C
13
) molecu-
lar weight representatives. Like DINPandDIDP,
they have shown no signicant developmen-
tal toxicity. It is reasonable to conclude that
other members of the category would behave
similarly, as shown by the weight of evidence.
Thus, it can be concluded that this category of
HMWPE induces no biologically signicant de-
velopmental effects in rodents.
In conclusion, the weight of evidence shows
that the HMWPEs have a low acute and sub-
chronic toxicity. They are not irritating to the
skin or eyes. They are not skin sensitizers. They
are not mutagenic. No or only minimal develop-
mental toxicity and no adverse effects on repro-
ductive capability have been observed in rodent
studies. Thus, there is minimal concern about
these PEs resulting in reproductive toxicity in
humans. Although not tested for carcinogenic-
ity, the members of this category do not showthe
potential for producing genetic effects. Also, the
same mechanism of action through peroxisome
proliferation can be anticipated for induction of
liver tumors in rodents, and this is presumed not
to be relevant to humans [374].
6.3. Risk Assessment
Extensive toxicological data on the effects of ph-
thalic acid intermediates and on most of the ph-
thalate esters have been collected and are avail-
able to the public [374, 375]. Final EU Risk As-
sessment Reports are also available for dibutyl-,
diisononyl-, and diisodecyl phthalate [375]. In
contrast, only draft versions exist for butyl ben-
zyl phthalate and DEHP [375] which are still un-
der discussion because of uncertainties for some
key elements, which included the human expo-
sure situation, carcinogenicity, toxicity to repro-
duction, and implications of species differences.
Carcinogenicity. IARC evaluated the short-
and long-term studies regarding genotoxicity
and carcinogenicity [376]. It was concluded that
there is sufcient evidence in experimental an-
imals for the carcinogenicity of DEHP, but that
there is inadequate evidence in humans for the
carcinogenicity of DEHP. DEHP was therefore
not classiable as to its carcinogenicity to hu-
mans (Group 3).
In making its overall evaluation of the car-
cinogenicity to humans of DEHP, it was taken
into consideration that:
DEHP produces liver tumors in rats and mice
by a non-DNA-reactive mechanisminvolving
peroxisome proliferation.
Peroxisome proliferation and hepatocellular
proliferation have been demonstrated under
the conditions of the carcinogenicity studies
of DEHP in rats and mice.
Peroxisome proliferation has not been docu-
mented in human hepatocyte cultures exposed
Phthalic Acid and Derivatives 37
to DEHP or in the liver of exposed nonhuman
primates.
Therefore, the mechanism by which DEHP
increases the incidence of hepatocellular tumors
in rats and mice is not relevant to humans [376].
Similar results were obtainedfor the other phtha-
late esters. Consequently, none of these is clas-
sied as being carcinogenic to humans.
The overall evaluation is widely accepted
[274, 343]. It was, however, pointed out that
there are some effects of DEHP that are not
receptor-mediated, that peroxisome prolifera-
tion is not established as an obligatory step in the
carcinogenesis of DEHP, and that therefore the
species differences are not viewed as a valid hy-
pothesis for the noncarcinogenicity to humans.
It was argued that, independent from peroxi-
some proliferationandPPAR, DEHPcanaffect
multiple signaling pathways by transcriptional
activation of PPAR-regulated genes (suppres-
sion of apoptosis, protooncogene experession)
and can induce biological effects (morphologi-
cal cell transformation, decreased levels of gap
junction intercellular communication). All these
effects could contribute to carcinogenicity, and
the precise mechanism of carcinogenic action
remains to be determined [246, 274].
Toxicity to Reproduction. Several of the
phthalate esters are classied as reproduction
toxicants according to European legislation
(di-n-butyl, di-n-pentyl, n-pentyl isopentyl, di-
isopentyl, butyl benzyl, di-2-ethylhexyl, and di-
2-methoxyethyl phthalate); however, none is
listed as being mutagenic or carcinogenic [377,
401].
6.3.1. Biomonitoring and Human Exposure
to Phthalate Esters
Urinary excretion of phthalate ester metabolites
(such as monoesters in conjugated or unconju-
gated form) may be taken as an indirect measure
for intestinal absorption. After in vitro hydro-
lysis of all metabolites phthalic acid may be
taken as sum parameter. Urinary MEHP was
extensively used in animal and human stud-
ies for DEHP exposure estimates. However,
there were uncertainties, as during the complex
DEHP metabolism MEHP is further metabo-
lized. This suggests that MEHP in blood or urine
underestimates the exposure to DEHP, and also
that the sum of MEHP and secondary urinary
metabolites, namely, 5-OH- and 5-oxo-MEHP,
provides a better exposure estimate [188, 203,
378 380]. This is widely acknowledged, but
debate is ongoing as uncertainties remain re-
garding the value of earlier studies where the
oxidized MEHPmetabolites were not measured.
This issue was treated among others by the
NTP-CERHR Expert Panel during the evalua-
tion of DEHP-related risks to human reproduc-
tion in October 2005 [344]. Since the rst NTP-
CERHR report on DEHP identied data gaps
[342] substantial progress was made regarding
biomonitoring of phthalate esters in the general
population [382].
In 85 urban Germans aged 734 years urinary
MEHP predicted a median DEHP intake of 13.8
g kg
1
d
1
(95th percentile: 38.3); inclusion
of the secondary MEHP metabolites gave a me-
dian DEHP intake of 10.3 g kg
1
d
1
(95th
percentile: 52.1) [379, 380].
Urinary MEHP and, for the rst time, also
5-OH- and 5-oxo-MEHP, were measured in
an U.S. National Health and Nutritional Ex-
amination Survey (NHANES) for the period
2001/2002 that included 2782 individuals. It was
noted that MEHPlevels were comparable to pre-
vious reports for U.S. residents [383]. The lev-
els of the secondary metabolites were roughly
510-fold higher than MEHP [384]. Thus, they
were similar to or up to twofold higher than in
German adults and children [379].
Exposure estimates are always higher in chil-
dren than in adults and highest in medically
treated infants, due to the high dose and their
low body weights. More recent estimates indi-
cate that in adults and children the exposure to
different phthalates is comparable, with the ex-
ception of the use of medical devices in chil-
dren reaching enormous doses. Though doses
are higher in children than in adults, the more
recent estimates are far lower than those used
previously by the NTP-CERHR Expert Panel in
exception of the use of DEHP in medical de-
vices 2000. Based on the recent estimates and
rened NOAELs the margins of safety (MOS)
for all scenarios in Table 9 were reported to be
> 10 000, with the exception of use DEHP in
medical devices (MOS 5) [382].
High exposure may result from DEHP leak-
ing out of medical devices during medical treat-
38 Phthalic Acid and Derivatives
Table 9. Phthalate ester exposure estimates [342, 382]
Phthalate Exposed group Exposure estimate [342] in g
kg
1
d
1
More recent exposure estimates
[382] in g kg
1
d
1
Dibutyl phthalate adult 210 0.86 (3.86)
b
young child not provided 2.65
Butyl benzyl phthalate adult 2 0.43 (2.08)
b
young child 6 1.64
DEHP adult 330 0.61 (3.51)
b
young child (healthy) 1020 2.57
DEHP, medical-device use critically ill infant 18003300 12 000
Diisononyl phthalate adult <330 <LOD (0.73)
ab
children using toys mean <320 2.91 (10.71)
Diisodecyl phthalate adult <330 best analog: diisononyl phthalate
children >330
a
LOD = Level of detection.
b
Values are in the format: mean value (95th percentile).
ment of patients and parenteral nutrition of
neonates [382, 385, 386]. Estimates are gener-
ally far below 1 mg
1
kg d
1
in adults, but may
reach values of 18 mg
1
kg d
1
during extra-
corporeal membrane oxygenation (ECMO) and
large-volume blood transfusions. On the other
hand, the plasticizer most commonly employed
for medical use, DEHP, and its corresponding
monoester MEHP appear to stabilize the mem-
branes of stored erythrocytes [387]. In neonates,
dose estimates increase in the order: parenteral
nutrition without lipids (0.03 mg kg
1
d
1
) and
with lipids (2.5 mg kg
1
d
1
) <ECMO (14 mg
kg
1
d
1
) < exchange transfusions (22 mg
1
kg d
1
) [388]. These exposures are far beyond
the TDI value of 50 g kg
1
d
1
[389], but
only recently, a promising substitute has become
available, that lacks toxicity to reproduction and
testes [386]: diisononyl cyclohexane-1,2-dicar-
boxylate (DINCH). DINCH has been success-
fully introduced into the market and a name plate
production capacity of 100 000 tons is currently
under construction (BASF AG, personal com-
munication).
In conclusion, the occupational exposure lev-
els resulting from handling and workplace ex-
posure are generally low and well controlled.
Exposure estimates for subgroups of the general
public covers a wide range for all of the phthalate
esters, but more recent data indicate that the ex-
posure levels are much lower than was assumed
during the last evaluation of toxicity to reproduc-
tion in the year 2000. Only under special circum-
stances such as parenteral exposure (e.g., eluates
from medical devices) may the systemic intake
be considerable and of potential toxicological
relevance [205, 342, 382, 390, 391]. Dermal re-
sorption has been measured in vitro [341] and
in vivo [392] for several phthalate esters and ap-
pears to decrease with increasing chain length.
The contribution of this route to the total expo-
sure to phthalate esters is considered to be minor
compared to the oral and parenteral routes. The
signicantly enhanced knowledge of exposure
situations will improve the assessment of the
risks of reproduction toxicity, and it is reason-
ably expected that the MOS are satisfactory in
most scenarios, with the exception of phthalate
ester uptake from medical devices. An alterna-
tive to DEHP lacking toxicity, DINCH, has only
recently been developed and is suited to solve
the exposure problems associated with intense
medical treatment of patients and neonates.
6.3.2. Carcinogenicity
Hepatomegaly, peroxisome proliferation and re-
lated liver effects in rodents are described in Sec-
tion 6.2.5.4. DEHP exerted a much lower ex-
tent of peroxisome proliferation [229] in ham-
sters. Hepatocytes of guinea pigs, ferrets, mar-
mosets, and humans did not respond with per-
oxisome proliferation [189, 363]. Marmosets
did not show peroxisome proliferation after oral
administration of DEHP (2000 mg kg
1
d
1
for 14 d), whereas in a parallel rat experiment
marked peroxisome proliferation was observed
[201]. Peroxisome proliferation was also not
observed in cynomolgus monkeys which were
dosed with 500 mg kg
1
d
1
of DEHP for 3
weeks [393].
Several reasons for this species difference
are assumed: The primary metabolite, MEHP
Phthalic Acid and Derivatives 39
(see Section 6.2.5.1), undergoes much less glu-
curonidationinrats. Therefore, renal excretionis
much slower in rats, and tissue levels are higher.
This has been demonstrated in experiments with
rats and monkeys [201, 393]. Furthermore, it
was shown that in rats, due to saturation of a -
nal -oxidation step in the -oxidation process,
a secondary metabolic pathway (-1 oxidation)
is induced to a far greater extent, leading to the
5-keto derivate of MEHP, which was shown to
be the ultimate peroxisome proliferator [393].
In addition to these kinetic differences, there
may also be intrinsic toxicodynamic differences
among the species. The overall evidence indi-
cates that human hepatocytes either do not re-
spond to peroxisome proliferators or do so only
to a much lesser extent [291]. The conclusion
is that, in the case of phthalates, rodent experi-
ments overestimate the risks of chronic toxicity
to humans.
6.3.3. Toxicity to Reproduction
DEHP is the model substance to investigate ad-
verse effects and end points in rodents, in pri-
mates, and in vitro. Evidently DEHP is a repro-
duction developmental and developmental tox-
icant in rodents, but not in primates, that ex-
erts adverse effects on the male reproductive
organs and reduces sperm parameters, reduces
fertility (increased postimplantation loss) at ele-
vated dose, exerts developmental toxicity to the
progeny as substantiated by decreased viability
and adverse effects on reproductive parameters
of male progeny (adverse effects on reproduc-
tive organs, sperm parameters). The underly-
ing mechanisms need further elucidation; tox-
icity of metabolites to Sertoli cells and interac-
tion with sex hormone regulation are clearly in-
volved, potentially via zinc depletion or receptor
interaction. Species differences exist: primates
are much more resistant than rodents.
However, mechanisms, potential threshold
levels, and species differences are under intense
investigation. An update of the evaluation by the
CERHR Expert Panel in 2000 [342] was dis-
cussed in October 2005 and the nal version
of the update Expert Panel Report is available
since November 2005 [344]. The database is
much weaker for other phthalates, but for man-
aging risks the precautionary principle was ap-
plied, i.e. many of the short-, intermediate-, and
long-chain phthalate esters are considered to act
through similar mechanisms as does DEHP, and
lead to the same adverse effects.
6.4. Risk Management
Phthalates have been under regulatory scrutiny
for more than 20 years because of questions
about carcinogenicity and reproductive toxicity,
and this has led to some regulatory measures for
worker and consumer protection.
In Germany permissible exposure levels
(PEL) existed for phthalic anhydride (1 mg/m
3
)
and 11 phthalate esters. The values stemmed
from several European countries and ranged
from3 mg/m
3
(e.g., diethyl, butyl benzyl, diben-
zyl, diisodecyl phthalate) and 5 mg/m
3
(e.g.,
diallyl, dicyclohexyl, diheptyl, dioctyl, dinonyl
phthalate) to 10 mg/m
3
for DEHP [394]. How-
ever, the scientic basis of the PEL values was
regarded to be insufcient, and consequently
the PEL values were abolished with the excep-
tion of DEHP, which is currently the only ph-
thalate for which a scientically derived AGW
a
value of 10 mg/m
3
(short-termvalue: 8-fold) ex-
ists. Work is ongoing to establish AGW values
for a number of phthalates for which the sci-
entic basis is regarded to be sufcient. These
are dicyclohexyl, diheptyl, dioctyl (excluding
di-n-octyl and DEHP), and dinonyl phthalate, as
well as isophthalic and terephthalic acids [395].
The MAK list gives additionally MAK values
for isophthalic and terephthalic acids besides
DEHP, but these values are not binding. The data
basis for phthalic acid, phthalic anhydride, and
diallyl phthalate was regarded as being insuf-
cient for deriving an MAK value [396]. Inhala-
tion of vapors, mists, and aerosols is regulated
in the USA by TLVs (TWA 5 mg/m
3
, STEL
10 mg/m
3
for DEHP, dibutyl phthalate, and di-
methyl phthalate).
Regulatory scrutiny led to the voluntary re-
moval of DEHP from childrens teethers and
toys in 1986, and subsequent pressure to re-
move others. Since December 1999, the EU has
prohibited toys containing more than 0.1 % of
six phthalates (di-n-butyl, butyl benzyl, di-2-
ethylhexyl, dioctyl, diisononyl, and diisodecyl
phthalate) that are intendend to be placed in
the mouth of children under three years of age
40 Phthalic Acid and Derivatives
[397]. This temporary ban was strengthened
in July 2005 in that di-n-butyl, butyl benzyl,
and di-2-ethylhexyl phthalate are prohibited in
all toys and childcare articles; for the sake of
the precautionary principle the regulation re-
mained unchanged for dioctyl, diisononyl, and
diisodecyl phthalate [398]. The more recent Di-
rective 2005/84/EC prohibits the use of plasti-
cized materials containing more than 0.1 % of
DEHP, DBP, or BBP in toys and childcare ar-
ticles; or more than 0.1 % of DINP, DIDP, or
DNOP in toys or childcare articles which can be
placed in the mouth by children. This new reg-
ulation was adopted by the EC Member States
during 2006, and is effective since January 2007
[399].
There is a general call for substitution of haz-
ardous materials by less hazardous or nonhaz-
ardous alternatives. Several applications have
been identied or are under investigation in
which DEHP could be substituted by less toxic
plasticizers, e.g., diisononyl, diisodecyl, or di-2-
propylheptyl phthalate, adipates, citrates, alky-
lated sulfonic esters, trimellitates (in which
a third ester group is attached to the ben-
zene ring of phthalates), terephthalates, and
cyclohexane-1,2-dicarboxylic esters (DINCH)
[400], but there may be limitations that are of
technical (e.g., stability), economical (e.g., in-
creased costs), or toxicological and ecological
nature (e.g., unknown effect proles of new
substances). Nonetheless, research efforts al-
ready led to very promising new solutions and
products, notably the substitution of DEHP by
DINCH in medical-care products [386].
The former MAK and TRK-values are all
abolished and will be stepwise replaced by
science-focused AENs (= Arbeitsplatzgrenzw-
erte)
7. References
1. R. C. Weast: Handbook of Chemistry and
Physics, 60th ed., CRC Press, Boca Raton,
Fla., 1979/1980 , C 436.
2. Handbuch des Chemikers, VEB-Verlag
Technik, Berlin 1956.
3. Kirk-Othmer, 2nd ed., 15, 445.
4. H. Suter: Phthals aureanhydrid und seine
Verwendung, Dr. Dietrich Steinkopf-Verlag,
Darmstadt 1972.
5. K. Nabert, G. Sch on: Sicherheitstechnische
Kennzahlen brennbarer Gase und D ampfe,
2nd ed., Deutscher Eichverlag, 1970.
6. R. H. Perry: Chemical Engineers Handbook,
6th ed., McGraw-Hill, New York 1973.
7. BASF, unpublished results, Ludwigshafen
1989.
8. B. Raeke, Angew. Chem. 70 (1958) 1.
9. K. Weissermel, H.-J. Arpe: Industrielle
organische Chemie, 4th ed., VCH
Verlagsgesellschaft, Weinheim, Germany
1994, p. 415.
10. F. Wirth, Hydrocarbon Process. 54 (1975) 107.
11. BASF, DE 2 356 049, 1973 (F. Wirth et al.).
12. BASF, DE 1 266 748, 1966 (G. Schaefer et
al.).
13. BASF company brochure: Phthalic
Anhydride Advanced Catalytic
PA-Technology, 1995.
14. Hydrocarbon Process. 76 (1997) no. 3, 146.
15. Europa Chemie (1997) no. 4, 12.
16. O. Wiedemann et al., Chem. Eng. (N.Y.) 86
(1979) 62.
17. Nippon Shokubai Kagaku Kogyo, DE 2 948
163, 1978 (T. Sato et al.).
18. Nippon Shokubai Kagaku Kogyo, US 4 284
571, 1978 (T. Sato et al.).
19. T. Sato et al., Hydrocarbon Process. 62 (1983)
no. 10, 107.
20. Nippon Shokubai Kagaku Kogyo, US 4 046
780, 1975 (Y. Nakanishi et al.).
21. Nippon Shokubai Kagaku Kogyo, US 4 356
112, 1976 (Y. Nakanishi et al.).
22. L. Verde, A. Neri, Hydrocarbon Process. 63
(1984) no. 11, 83.
23. Hydrocarbon Process. 64 (1985) no. 11, 156.
24. Alusuisse Italia S.P.A., EP 37 492, 1980 (A.
Neri et al.).
25. Alusuisse Italia S.P.A., US 4 489 204, 1980
(A. Neri et al.).
26. Hydrocarbon Process. 59 (1979) no. 11, 208.
27. Rh one Progil, DE 2 416 457, 1973 (R.
Dumont et al.).
28. BASF, DE 2 510 994, 1975 (K. Blechschmitt
et al.).
29. BASF, DE 2 547 624, 1975 (K. Blechschmitt
et al.).
30. BASF, DE 2 914 683, 1979 (P. Reuter et al.).
31. Nippon Shokubai Kagaku Kogyo, DE 3 045
624, 1979 (Y. Nakanishi et al.).
32. Nippon Shokubai Kagaku Kogyo, JP 56 73
543, 1981.
33. Nippon Shokubai Kagaku Kogyo, JP 57 105
241, 1982.
Phthalic Acid and Derivatives 41
34. Nippon Shokubai Kagaku Kogyo, JP 57 180
430, 1982.
35. Mitsui Toatsu Chemicals Inc., JP 54 5 891,
1979 (K. Oshima).
36. BASF, DE 2 546 268, 1975 (K. Blechschmitt
et al.).
37. N. G. Glukhovskii et al., SU 978 910, 1981;
Chem. Abstr. 98 (21): 178 962 b.
38. D. W. B. Westerman et al., Appl. Catal. 3
(1982) 151.
39. M. S. Wainwright et al., Can. J. Chem. Eng. 55
(1977) 557.
40. W. Fiebig et al., Chem. Tech. (Leipzig) 28
(1976) 673.
41. D. Vankove et al., J. Catal. 36 (1975) 6.
42. J. Herten, G. F. Froment, Ind. Eng. Chem.
Process Des. Dev. 7 (1968) 516.
43. P. H. Calderbank, Chem. Eng. Sci. 32 (1977)
1435.
44. M. S. Wainwright et al., Catal. Rev. Sci. Eng.
19 (1979) 211.
45. D. D. McLean et al., Can. J. Chem. Eng. 58
(1980) 608.
46. A. G. Lyubarksii et al., Kinet. Katal. 14 (1973)
956.
47. K. Hertwig et al., Chem. Tech. (Leipzig) 23
(1971) 584.
48. J. Skrzypek et al., Chem. Eng. Sci. 40 (1985)
611.
49. M. Blanchard, D. Vanhove, Bull. Soc. Chim.
Fr. 1971, 3291.
50. R. R omer et al., Chem. Ing. Tech. 46 (1974)
653.
51. G. Emig et al., Chem. Ing. Tech. 47 (1975)
717, 888, 1012.
52. I. E. Wachs et al., Stnd. Surf. Sci. Catal. 19
(1984) 275.
53. I. E. Wachs et al., Appl. Catal. 15 (1985) 339.
54. R. Saleh et al., Appl. Catal. 31 (1987) 87.
55. V. Nikolov et al., Stnd. Surf. Sci. Catal. 34
(1987) 173.
56. I. E. Wachs et al., J. Catal. 91 (1985) 366.
57. G. F. Froment et al., Appl. Catal. A 120 (1994)
17.
58. G. F. Froment et al., Chem. Eng. Scie. 51
(1996) 2091.
59. G. C. Bond et al., J. Catal. 164 (1996) 276.
60. N. L. Franklin et al., Trans. Inst. Chem. Eng.
34 (1956) 280.
61. F. de Maria et al., Ind. Eng. Chem. 53 (1961)
259.
62. F. Bernardini et al., Chim. Ind. (Milan) 48
(1966) 9.
63. A. Farkas, Hydrocarbon Process. 49 (1970)
121.
64. Davy Powergas, DE 2 602 895, 1976 (L.
Hellmer et al.).
65. R. M. Dow et al., Hydrocarbon Process. 56
(1977) 167.
66. Chem. Eng. 84 (1977) 65.
67. BASF, DE 1 601 162, 1967 (F. Lorenz et al.).
68. Deggendorfer Werft und Eisenbau, DE 2 207
166, 1972 (O. Wanka et al.).
69. Deggendorfer Werft und Eisenbau, DE 2 543
758, 1975 (R. Vogl).
70. Davy McKee AG, DE 2 855 629, 1978 (G.
Keunecke et al.).
71. Davy McKee AG, DE 2 855 630, 1978 (G.
Keunecke et al.).
72. BASF, DE 3 207 208, 1982 (E. Danz et al.).
73. BASF, DE 3 007 627, 1980 (W. Muff).
74. Davy McKee AG, DE 3 339 392, 1983 (H.
Saffran).
75. Metallgesellschaft AG, DE 3 411 732, 1984
(V. Franz et al.).
76. GEA Luftk uhler GmbH, DE 3 512 903, 1985
(W. Rudowski).
77. Metallgesellschaft AG, DE 1 227 443, 1964
(W. Scheiber).
78. Pittsburgh Coke and Chemical Company, US 2
850 440, 1955 (M. O. Shrader et al.).
79. Mitsui Toatsu Chem., DE 1 948 374, 1968 (K.
Hayakawa et al.).
80. Japan Gas Chem., JP 45/10 333, 1970.
81. Rh one-Progil, DE 2 452 171, 1976.
82. Nippon Shokubai Kagaku Kogyo, JP-Kokai
58 8074, 1981.
83. Nippon Steel Chem., JP-Kokai 58 45 434,
1977.
84. Nippon Shokubai Kagaku Kogyo, DE 3 225
079, 1981 (Y. Kita et al.).
85. Chemische Werke H uls AG, EP 120 199, 1983
(F. Gude et al.).
86. Chemische Werke H uls AG, DE 3 309 310,
1983 (F. Gude et al.).
87. BASF, US 3 507 886, 1966 (H. Suter et al.).
88. Chemiebau Dr. A. Zieren GmbH, US 3 655
521, 1968 (H. Gehrken et al.).
89. Nippon Steel Chem. Co, JP-Kokai 57 49 549,
1977.
90. Nippon Steel Chem. Co, JP-Kokai 58 45 435,
1977.
91. Hydrocarbon Process. 46 (1967) 215.
92. The Badger Company Inc., US 4 435 580,
1982 (C. D. Miserlis).
93. The Badger Company Inc., US 4 435 581,
1982 (C. D. Miserlis).
94. Kawasaki Steel Corp., JP-Kokai 63 141 645,
1986.
42 Phthalic Acid and Derivatives
95. Kawasaki Steel Corp., US 5 252 752, 1991 (T.
Aona et al.).
96. Kawasaki Steel Corp., JP 61 90 281, 1994 (Y.
Asami et al.).
97. Kawasaki Steel Corp., EP 451 614, 1991 (T.
Aona et al.).
98. Shokubai Kasei Kogyo, JP 80 99 040, 1994.
99. Kawasaki Steel Corp., JP 57 32 096, 1994.
100. Chemische Werke H uls AG, DE 1 643 827,
1967 (F. List et al.).
101. Standard Oil Company, US 4 215 051, 1979
(H. Schroeder et al.); US 4 215 052, 1979 (H.
Schroeder et al.); US 4 215 053, 1979 (D. A.
Palmer et al.); US 4 215 054, 1979 (H.
Schroeder et al.).; US 4 215 055, 1979 (D. A.
Palmer et al.); US 4 215 056, 1979 (H.
Schroeder et al.).
102. Sisas, EP 256 352, 1987 (F. Celeste et al.).
103. Mitsubishi Gas Chemical Co, JP 53/40 711,
1978.
104. V. N. Orlik, Khim. Tekhnol. 6 (1981) 51.
105. S. Slavov, Khim. Ind. (Soa) 56 (1984) 252.
106. S. Slavov, Geterog. Katal. 1 (1983) 81.
107. Chem. Sys. Inc., Phthalic Anydride, 933, 05.
1995.
108. BASF, DE 2 902 978, 1979 (G. Kilpper et al.).
109. BASF, EP 4635, 1978 (G. Kilpper et al.).
110. Beilstein E II 21 (1953) 348.
111. LENTIA GmbH, DE 2 056 891, 1970 (J.
Schweighofer).
112. BASF, DE 2 334 379, 1973 (E. Hetzel et al.);
DE 2 334 916, 1973 (E. Hetzel et al.); DE 2
911 245, 1979 (G. Kilpper et al.).
113. O. Joklik, AT 286 271, 1969.
114. Dawes Lab Inc., US 3 819 648, 1971 (W. R.
Boehme et al.).
115. M. C. Sze et al., Hydrocarbon Process. 55
(1976) no. 2, 103.
116. Sun Res. + Develop. Co, DE 2 256 596, 1972
(R. D. Bushick et al.).
117. Degussa, DE 1 948 714, 1969 (T. L ussling et
al.).
118. Allied Chem. Corp., US 2 833 807, 1956 (A.
Farkas et al.).
119. Mitsubishi Gas. Chem. Co, DE 2 427 191,
1979 (M. Saito et al.).
120. Institut Nefttecnimitscheskich USSR, DE 2
627 068, 1979 (S. D. Mechtjev et al.).
121. BASF, EP 699 476, 1996 (L. Karrer et al.).
122. Takeda Yakuhin Kogyo, JP 55 17 360, 1980 (I.
Onishi et al.).
123. BASF, EP 766 985, 1997 (L. Karrer et al.).
124. G. Stefani, Chim. Ind. 54 (1972) 984.
125. K. K. Moll et al., Chem. Tech. (Leipzig) 20
(1968) 600.
126. T. Kudo, CEER Chem. Econ. Eng. Rev. 2
(1970) 43.
127. BASF, DE 1 172 253, 1962. (H. Kr oper et al.).
128. BASF, DE 1 643 630, 1967 (M. Decker et al.).
129. Lummus Corp., US 3 776 937, 1972 (A. P.
Gelbein);
DE 2 008 648, 1970 (A. P. Gelbein).
130. Ullmann, 3rd ed., 13, 733735.
131. Beilstein, E III 9, 41994200.
132. H. Suter: Phthals aureanhydrid und seine
Verwendung, D. Steinkopff Verlag, Darmstadt
1972.
133. H. Rein, Chemiker Taschenbuch, 58th ed., Part
2, Springer-Verlag, Berlin 1937.
134. L. Meier: Weichmacher f ur PVC, in G. W.
Becker, D. Braun (eds.): Kunststoff-Handbuch,
vol. 2/1: Polyvinylchlorid, Hanser-Verlag,
M unchen 1986.
135. Monsanto US 1 923 938, 1933 (L. P. Kyrides).
136. G. L utzel, J. Holzmann: Ein neuer
Weichmacher f ur PVC auf Basis DINP,
Kunststoffe 74 (1984) no. 4, 10881091.
137. Modern Plastics Encyclopedia, McGraw-Hill,
New York (1988).
138. US. International Trade Commission,
Synthetic Organic Chemicals, US. Production
and Sales,1987. US. Government Printing
Ofce, Washington, 1987.
139. Bulletin S-107 B, Methane Sulfonic Acid,
Pennwalt Corporation, Pa 19 102, USA.
140. TIL, Tioxide UK Ltd., Booklet TIL 12, Tilcom
Catalysts For Ester Manufacture, Billingham
1982.
141. Ullmann, 4th ed. 18, 534538.
142. F. X. Werber, B. F. Goodrich, US 3 056 818,
1962.
143. Du Pont Company, Chemical & Pigment
Dept., Booklet E-69 9961 4/87 (2 M), Du
Pont Performance Products, Wilmington, DE,
1987.
144. Dynamit Nobel, Technical Information, ZN
666/2.80/500/2439, Titanates as Catalysts for
the manufacture of plasticizers, Troisdorf,
1980.
145. M & T, Chemicals Inc., How to make DOP
and other Phthalate Plasticizers easier with
FASCAT 2001, and M & T Technical Data
Sheet No. 341, Rev. 9/78, and Material Safety
Data Sheet 3/82 for Stannous Oxide, Rahway,
NJ 1978.
146. Exxon Chemical International. A practical
guide to protecting man and the environment,
Kraainem, Belgium 1988.
Phthalic Acid and Derivatives 43
147. BUA Report 22 on Di-butyl phthalate,
VCH-Verlagsgesellschaft, Weinheim, Dec.
1987.
148. BUA Report 4 on Di-2-ethylhexyl phthalate,
VCH-Verlagsgesellschaft, Weinheim, Jan.
1986.
149. Gemeinsames Ministerialblatt (GMBl), Bonn,
Germany 1990.
150. Gemeinsames Ministerialblatt (GMBl), G
3191 A, Bonn, W. Germany, 28. Feb. 1986.
151. BASF, Technical Leaet, Palatinol AH-L
(med), Ludwigshafen, 1989.
152. B. L. Wadey, J. Holzmann, SAE Technical
Paper Series, 870 318, Plasticizer for
Automotive Interior Trim, SAE, 400
Commonwealth Drive Warrendale, PA 15 096,
USA 1987.
153. Verordnung uber Anlagen zur Lagerung,
Abf ullung und Bef orderung brennbarer
Fl ussigkeitenVbF, Bonn, v. 27. Feb. 1980.
154. Regulations for water
(Wasserhaushaltsgesetz-WHG) Bonn, last
edition Sept. 23., 1986 (BGBl. I, S. 1529).
155. BASF, Hinweise zur Lagerung von
Weichmachern f ur PVC, TI-CIW/ES 003 d,
Ludwigshafen, Dec. 1987.
156. Verband Kunststofferzeugende Industrie,
Argumente 3, Zum Thema Weichmacher,
Frankfurt, 1988.
157. BASF, Marketing CIW, Ludwigshafen, 1988.
158. Registry of Toxic Effects of Chemical
Substances (NIOSH; 1990), Proc. Soc. Exp.
Biol. Med. 49 (1942) 471.
159. E. Sandmeyer, C. Kirwin, jr., in G. D. Clayton,
F. E. Clayton (eds.): Pattys Industrial Hygiene
and Toxicology, vol. II A, Wiley-Interscience,
New York (1982) p. 2350.
160. M. Wernfors, J. Nielsen, A. Sch utz, S.
Skerfving: Phthalic Anhydride-Induced
Occupational Asthma, Int. Arch. Allergy
Appl. Immunol. 79 (1986) 7782.
161. J. Nielsen, H. Welinder, A. Sch utz, S.
Skerfving: Specic Serum Antibodies
Against Phthalic Anhydride in Occupationally
Exposed Subjects, J. Allergy Clin. Immunol.
82 (1988) 126133.
Alternative Dermal Sensitization Test: The
Mouse Ear Swelling Test (MEST), Toxicol.
Appl. Pharmacol. 84 (1986) 93114.
162. S. C. Gad et al.: Development and Validation
of an Alternative Dermal Sensitization Test:
The Mouse Ear Swelling Test (MEST),
Toxicol. Appl. Pharmacol. 84 (1986) 93114.
163. V. van Kampen, R. Merget, X. Baur: