Note: Descriptions are shown in the official language in which they were submitted.
GC49
~ 2 S ~
The present in~ention relates to a process for the
preparation of diphthaloyl peroxide.
Hitherto, it has been proposed by A. Baeyer and
v. Villiger in l9Ol that diphthaloyl peroxide could be
produced by reaction between phthalic anhydride and hydrog~n
peroxide in dilute aqueous alkaline solution. Although
Baeyer and Villiger quoted no yields, we obtained yields
of only 8% of the theoretical maximum, based o~ phthalic
anhydride present initially, on repetition of their work. ;
Such yields are commercially unacceptable.
According to the present invention, there is
provided a process for the production of diphthaloyl
peroxide comprising the steps of forming a mobile slurry
or paste containing particulate phthalic anhydride
and aqueous hydrogen peroxide, maintaining the slurry or
paste mobile until at least some diphthaloyl peroxide
has been produced, and thereafter separat~ng the diphthaloyl
peroxide from the aqueous phase
'` Herein, the term "mobile", used in relation to
the terms "slurry" or "paste", indicates that the slurry
or paste is capable of being mixed under the prevailing
reaction conditions, and in the chosen apparatus.
~n general, we have found that by varying the ratio
of liquid to solid in the slurry or paste, its mobility
is varied, at any given set of reaction conditions in a
particular item of mixing apparatus. A convenient ratio
of solid to liquid, i.e. phthalic anhydride to aqueous
hydrogen peroxide, is the range of 0.5 to 2.0 gm per ml,
,
- 2 -
: . . ; : . . :
, ' ~
~. .. . . ~,
GC49
for the initial mixture. In particular, it is highly
preferable to use only enough, or only slightly more
than enough liquid than the minimum to form a mobile
slurry or paste. By 5electing appropriate agitators,
or example, æ-blade mixers, mixtur~ c~n~ainina
the minimum volume of liquid will generally be in the form
of thick paste. During the course of the reaction, as
the hydrogen peroxide is consumed, the mixture
becomes less mobile. Mobility can be easily restored by
the addition of further amounts of liquid, suitably
dilute mineral acid, water or agueous hydrogen peroxide.
Preferably, the amount of liquid added is no more than
the minimum amount required to restore mobility.
Alternatively, sufficient liquid may be present initially
to obviate the need to add further amounts of liquid.
The amount of liquid to be added, in general, depends not
only upon the extent to which the amount of liquid present
initially exceeded the minimum, but also upon the mixing
apparatus. The amount added often falls within the range
of 0 to 1 ml per ml of liquid present initially, so that
the ratio of solid to total amount of liquid is usually
in the range of 0.25 to 1.5 gm per ml, more often 0.5 to
1.0 gm per ml.
The reaction to produce diphthaloyl peroxide theoret-
ically requires two moles of phthalic anhydride for each
mole of hydrogen peroxide. Use of excess hydrogen peroxide
can be advantageous, in that it tends to reduce the amount
of phthalic anhydride remaining in the finished product.
Consequently, we prefer to use a mole ratio of phthalic
~i~ 3
:. ~ . . ~ . . . . .
~, . ~ , . . .
.~ . . . . ' ! '
' ' '
,
.
GC49
52
anhyd~ide tv hydrogen peroxide in the range of 2:1
to 2:20 especially in the range of 2:1 to 2:5,and
more particularly in the range of 2:1.5 to 2:5. A
mole ratio of more than 2 moles phthalic anhydride per
mole hydrogen peroxide can be employed, but this results
inevitably in the presence of phthalic anhydride in the
product.
Although we wish not to be bound by any theory, it is
our current belief that a substantial proportion of the
phthalic anhydride remains in particulate form, and reacts
with the hydrogen peroxide in the solid state. By using
particles having an average particle diameter of 10~ to
50,u or lower e.g. obtained by grinding commercially
available flake phthalic anhydride, the possibility of
residual phthalic anhydride in the product, and thus its
appearance during use of the product, can be minimised.
. _ _ .. . . .. .. .. , .. , .. _ _ . ., .. " .. , . .. _ _ . .. . .
It will be readily apparent that the mole ratio of
phthalic anhydride to hydrogen peroxide is related to the
ratio of solid to liquid employed initially, to the
concentration of hydrogen peroxide in the solution, and
to whether any additional amount of liquid added to
maintain bility of the mixture contains hydrogen peroxide
Thus, if the mole ratio of phthalic anhydride to hydrogen -
peroxide is low, e.g. 3:2, and the ratio of solid to ;
liquid used initially is low, e.g. 0.5:1, then the -
_ concentration of hydrogen peroxide is also relatively
.. .... . ~
low, in the region of 10% by weight. However, the
concentration is usually at least lOX, and frequently
at least 20X. Thus, using a mole ratio of phthalic
~: :
~. . , . . . . , .: . .
-: . . : : . - :
GC49
5~ ~
anhydride to hydrogen peroxide from 1:1 to 2:3, and
an initial solid to liquid ratio in the range of 1:1
to 1:1~5, the concentration of hydrogen peroxide falls
within the range of 20 to 45% by weight. Clearly,
higher mole ratios of hydrogen peroxide to phthalic
anhydride require higher hydrogen peroxide concentrations
or lower solids to liquid initial ratio, or addition of
hydrogen peroxide during the reaction.
The reaction between phthalic anhydride and hydrogen
peroxide can take place without adjustment of the pH of ~;
the system, i.e. at the pH obtained when commercially
available aqueous hydrogen peroxide solution i9 contacted
with the solid phthalic anhydride~ The addition of a
small amount of acld e.g.a mineral acid such as sulphuric
or phosphoric acid, can lead to a small increase in the
reaction yield, and the addition of a small amount of
alkali, e.g. an alkali metal hydroxide, such as sodium
hydroxide, can tend to make the mixture more mobile.
Desirably the mixture has a pH of from 0.5 to 3 as measured, and
preferably from 0.5 to 2.5. The mixture can also contain
a small amount of a non-ionic surfactant, such a9 an alkyl-
phenol ethoxylate e.g. trimethylnonylphenol ethoxylate,
suitably in an amount o~ from 0.05% to 0.5~ by weight,
based on the weight of the mixture.
In general, we have found that the mixture becomes
more mobile as the temperature is raised. Consequently,
: .` . -- 5
. ~
GC49
~08~ 5~
we prefer to employ a temperature of above ambient, more
preferably in the range of rom 25 to 50C, thereby
avoiding the increased decomposition of the product
which can occur at temperatures above 50C. The reaction
is normally terminated after about 5 hours or less,
since a longer reaction period does not tend to
significantly increase the yield of diphthaloyl pero~ide an~
can result in the formation of undesirable b~-products.
Significant yields can be obtained after reaction of at
least 1 hour, conveniently up to 4 hours, especially at
temperatures of about 35-40 C or higher. In one convenient
method of carrying out the process, particulate phthalic
anhydride and aqueous hydrogen peroxide are mixed at a
pre-selected temperature in the range 25 to 50C, the
temperature is maintained for a short period, such as
15 minutes, whilst sufficient liquid is added to maintain ~-
~he reaction mixture mobile, and the mixture is thereafter
permitted to cool to ambient temperature for the remainder
of the reaction period.
,, _ . ... . . .
Suitably, the diphthaloyl peroxide can be separated
from the aqueous phase by conventional techni~ues, such
as by filtering or centrifuging, particularly when the
volume of li~uid used in the reaction mix is significantly
greater than the minimum amount required to P ~ a
mobile slurry or paste. Further separation can then be
effected by drying the product in conventional apparatus,
such as spray driers or fluid bed driers. However,
~,,
-- 6
,.. ,.. ~-~ . .... . . . . . ,, .;,,. . .:
, ,, , . : ., ~ ~. ; , : ` -. . :.
,: . .: - - . ~ ~ . . . .
: .
:
: : , .~.. ... .. , , - : , . ~ .: .
s~ ~ .
1 especially when substan~ially the minimum amount of liquid had
been used in the reaction st~g~, it can be convenient to omit
a filtration or centrL~uge stage, and pass directly to a spray
drier or fluid bed drier. ITowever, diphthaloyl peroxide in the
pure dry state is hazardous, e.g. impact sensitive, so that in
practice it is h:ighly desirable to contact the damp diphthaloyl
peroxide intimately with a diluent, as described in our copending
Canadian patent application no. 252,390 in which a notice of
allowance was issued on April 10, 1979, before drying. Such
diluents can be mixed in after the reaction is completed, e.g.
magnesium sulphate. However other diluents which are substantially
inert towards diphthaloyl peroxide can be present during the
reaction itself. Such other diluents include fatty acids, e.g.
lauric acid, and aluminosilicates e.g. zeolites or bentonite.
Preferably, the amount of diluent or diluents used is sufficient
to ~ully desensitize the product.
One of the by-products in the reaction between
phthalic anhydride and hydrogen peroxide is monoperoxyphthalic
acid which decomposes "in situ" more rapidly than diphthaloyl
peroxide. Consequently, it is preferable to remove monoperoxy-:
phthalic acid. One way of e~fecting this, is to wash the
diphthaloyl peroxide with a small amount of water and/or a non-
acidic organic solvent. The organic solvent can be hydrophilic,
e.g. acetone or a low molecular weight aliphatic alcohol e.g.
isopropanol, or it can be hydrophobic, e.g. a chlorinated hydro-
carbon chloroform, or a liquid hydrocarbon. Washing with
, :
:,
'
- 7 -
,~ .
.j .
GC4g
3L~i 3~L'~5~
organic solvents,Pre~erablY low molecular weight alcohols,can
also remove phthalic anhydride. Con~eniently, the
washlng can be effected either prior to or after
separation of the product from the reaction liquor,
if such separation stage is employed. One other way of
removing monoperoxyphthalic acid is to employ a
reducing agent e.g. sodium sulphite. This can also
remove excess hydrogen peroxide, and advantageously
forms, "in situ", sodium sulphate, a highly satisfactory
10 diluent. Suprisingly, sodium sulphite appears not to
react marXedly with diphthaloyl peroxide, and thus the
yield of the reaction remains substantially unaltered.
Preferably, diphthaloyl peroxide is treated with
sufficient sodium sulphite to remove residual mono-
15 peroxyphthalic acid and hydrogen peroxide, and also
washed with an organic solvent as described above, in
order to remove residual phthalic anhydride.
Certain embodiments according to the present
invention will now be described by way of Example only.
In each Example, phthalic anhydride was added to
aqueous hydrogen peroxide containing,
where indicated, additives Al to A5,in a beaker held in
a water bath maintained at the temperature shown, and
a stirrable slurry resulted. The slurry was stirred
25 continuously with a paddle stirrer, power to the stirrer
being increased after about 15 minutes when the slurry
began to ~hicken rapidly. Additional amounts of
aqueous hydrogen peroxide (same concentration) and/or water
. '. .
.::, ., :: : : : - . :. .......... : .:: : :. .: :
, , - . , . : :: : . . . .:
GC49
were then slowly added to the slurry, over a period of
about 30 minutes, restoring the slurry to approximately
its original mobiliky. The slurry was then cooled to
ambient for the remainder of the reaction period. In
Example 10, a diluent, lauric acid, together with
demineralised water was added after 1 hour of the reaction
period. In Examples 11 to 16, the lauric acid together
with an additional amount of water was stirred in at the
end of the reaction period and mixed for 10 minutes to
give a homogeneous mixture.
.. .. .. .
Diphthaloyl peroxide (DPP) was thereafter recovered
by one of techniques A to E, together with an impurity
monoperoxyphthalic acid (MPPA). In technique A, the
slurry was filtered and water washed, both under suction,
and then dried. In technique ~, the slurry was stirred
with water for a minute, centrifuged and then dried. In
technique C, technique A was followed, with the addition
of washing with isopropanol under suction the water
washed filter cake. In technique D, technigue C was
followed, substituting a 20% methylated spirit in water
solution for the isopropanol. In technique E, finely
, ground sodium sulphite heptahydrate (90 g) together
with water (40 ml) were stirred into the slurry, with
cooling using an ice bath and the white cream dried
1 25 under vacuum.
In Æxamples 2 to 4 and 6 to 10, the phthalic
~:
', .
, . ,
i'~ `''' - 9
S~
1 ~nhydride was a commercially available flake material, and in
the remaining Examples, the flake material had been ground to
an average particle size of 50~. Additive Al is suf~icient
aqueous sodium hydroxide to raise the pH to approximatley pH 3,
A2 is 0.5 ml of 2N sulphuric acid, A3 is 3.0 ml of lN sodium
hydroxide. A4 is 0.1 ml of a non-ionic surfactant, trimethyl-
nonylphenol ethoxylate available commercially under the name
TERGITOL * TMN, and A5 was a combination of the non-ionic sur-
factant and 0.5 ml of 2N sulphuric acid.
The reaction conditions and amounts of reagents and :~
water employed are summarised in the Table, in which the mole
ratio shown is that of the total amount of hydrogen peroxide to
phthalic anhydride (PAn) the yield is molar, based on the amount
of phthalic anhydride added, and the content by weight based on :~
the resultant dried product.
~ ~ .
~ , :
.
'.
* Trade Mark
-- 10 --
,. ,i q.~,
~4Y
S2
THE TABLE
Examp1e ¦Initial Amounts ~Concn. Add- Dlluent
No. Amounts Added(mls) ¦ of it- Additi . - .
P~n ~tml2 ~22 ~ml) ¦ H/w2 ives ~ nt (2m0l)
_ t-- . _ __ ~ . ~ ,
1 30 30 0 30 35
2 30 30 0 30 35 ,.
3 30 30 0 12 35
4 30 17 0 14 35
32 30 10 30 35 ;
6 32 30 10 30 35
7 32 30 10 17 35
8 32 30 10 0 35
9 32 30 10 2 35 Al
1280 1300 400 0 35 1000 1000
11 32 3010 5 35 10 20
12 32 3010 5 35A2 10 20
13 32 3010 5 35A3 10 20
14 32 3010 5 35A4 lo 20
32 3010 5 35A5 10 20
16 32 30 10 5 35
17 32 30 10 51~ 10 20
18 20 15 7 1212.5 fi 12
19 ~ 30 llO_ ~ 35 10 20
',:1
.'' `',
''` ,
'`,' ' ' ' . .. ~ ' . ' ~
GC49
S~ ~
The Table ( cont . )
_ . __
Ex- Mole Reac- Reac~io Re- ~ield % Content %
am- Ratio tion Time cover DPP MPPA DPP MPPA
ple temp .(hot rs) Tech-
No . (C ) nique
_ . . _ _ _ _ l
1 1.8 25 20 B150 10 7814
2 1.8 25 5 B¦ 5711 7216
3 1.8 35-40 4 A159 7 7o10
4 1.1535-4Q 4 A¦ _ _ 6617
2.2525-30 4 C132 3 837
6 2.2525-30 4.5 B154 12 6819
7 2.25 35 4 B158 20 6826
8 2.25 40 4 A75 7 938
9 2.2525-30 4 ~44 13 6521
2.2535-40 1.67 C _ _ 421
11 2.2535-40 4 ~65 4 694
12. 2.2535-40 4 D67 3 673
13 2.2535-40 4 D30 2 513
14 2.2535 - 40 4 D74 3 743
2.2535-40 4 D71 2 702
16 2.2535-40 4 E74 2 321
17 1.2 35-40 4 D78
18 0.6 35-4Q 4 D53 4 373
, 19 2.25L 35 -40 2 D ~ 6 ~ S
., .
; _12_
