Acid-induced decomposition of Di-tert-butyl peroxide in n-heptane solution up to high temperatures and pressures

Article abstract of DOI:10.1524/zpch.1999.210.Part_1.061

The decomposition of di-tert-butyl peroxide (DTBP) in n-heptane with varying amounts of pivalic acid (PA) being added was studied via online FT-IR spectroscopy at elevated temperatures, 180 to 200°C, and pressures up to 2000 bar. A pronounced acid-induced DTBP decomposition is found without any PA consumption. Decomposition rates are adequately described by first-order kinetics. The pressure and the temperature dependence of acid-induced decomposition are investigated. The data suggest that the rate enhancement is due to the formation of a DTBP-PA complex. The acid-induced DTBP decomposition yields the same products as does the non-induced process (without acid being present). It thus appears justified to use the rate data observed for the model system DTBP-PA-n-heptane for the simulation of initiation rates in technical polymerizations where DTBP acts as the initiator and where acid is present, e.g. as a monomeric species. by Oldenbourg Wissenschaftsverlag, Muenchen.

Full text of DOI:10.1524/zpch.1999.210.Part_1.061

Bd.
S. 61-81
(1999)
Zeitschrift für
©
Chemie,
210,
Physikalische
München
Oldenbourg Wissenschaftsverlag,
by
Acid-Induced
Decomposition
Peroxide
Solution
of
in
Di-terf-butyl
-Heptane
to
and Pressures
High
up
Temperatures
*
Lars
Michael Buback and
Wittkowski
By
für
Institut
Universitat
Göttingen, Germany
Physikalische Chemie,
Göttingen,
TammannstraBe
37077
6,
16, 1998;
(Received
January
accepted
5, 1998)
May
Peroxides
Methacrylic
IDecomposition kinetics I High
acid
I
pressure
in
polymerization
The
of
with
spectroscopy
amounts
at elevated
acid-induced
rates are
di-feri-butyl peroxide (DTBP)
added was studied via online
«-heptane
varying
decomposition
acid
of
FT-IR
(PA) being
pivalic
180 to
and
2000
to
bar.
A
200°C,
temperatures,
DTBP
pressures up
pronounced
found
is
without PA
Decomposition
decomposition
described
any
consumption.
first-order kinetics. The
by
are
and the
adequately
temperature dependence
rate
pressure
The data
of acid-induced
that the
enhance-
decomposition
investigated.
DTBP-PA
suggest
ment is due to the formation of
a
acid-induced
The DTBP
complex.
decompo-
sition
the same
as does the non-induced
acid
yields
products
(without
process
being
It
to
thus
use the
data observed
rate
for the model
present).
justified
system
appears
for the simulation of initiation
the initiator and where acid is
in
rates
technical
DTBP-PA-n-heptane
polymerizations
as a
monomeric
as
where DTBP acts
present,
species.
e.g.
Die Kinetik des Zerfalls von
unterschiedlicher
in
wird
bei
zwi-
Ditertiarbutylperoxid
an Pivalinsâure
(DTBP)
(PA)
n-Heptan-Lösung
bei erhöhten
Zugabe
Mengen
Temperaturen,
Infrarotspektroskopie
sáureinduzierter
schen 180
und 200
und Drücken bis 2000 bar durch
studiert. Es wird ein
wobei sich die Saurekonzentration
°C,
quantitative
unter
DTBP
Zerfall
des
ausgeprágter
MeBbedingungen
im Verlauf einer Zerfallsreaktion
beobachtet,
nicht verándert. Die Reaktion laBt sich
durch ein
1. Ord-
sàureinduzierten
Zerfalls
jeweils
Geschwindigkeitsgesetz
beschreiben. Die Druck- und die
DTBP-Zerfalls werden untersucht. Die wesentliche Ursache des
des
nung
Temperaturabhângigkeit
beschleunigten
wird in der
eines
Der saureinduzierte
DTBP-
Bildung
DTBP-PA-Komplexes gesehen.
Zerfall führt zu denselben Produkten wie der
Sàure
ohne
nicht-induzierte,
erfolgende
Zerfall.
Daten
heit von Saure zu
erscheint somit
die am
Es
zur Simulation technischer
erhaltenen kinetischen
gerechtfertigt,
Polymerisationen
bei
Modellsystem
mit DTBP als Initiator
bei Anwesen-
von Sáuremonomeren.
etwa
nutzen,
Polymerisationen
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 9/22/15 1:29 PM

Buback and
Lars
62
Wittkowski
Michael
1. Introduction
in free-radical
used
initiators
as
are
polymer-
peroxides
extensively
Organic
izations
ficients
coef-
such
rate
To model
As
[1].
decomposition
precise
processes,
measured
is
that
not
are
easily
peroxide decomposition
required.
have
on-
rates
the
of
course
a
reaction,
decomposition
polymerization
during
solvents
of
determined in
been
line Fourier-transform
a
[2].
Using
non-polymerizable
variety
and
several
IR
[3]
diacyl peroxides
spectroscopy,
extended
in
studied in
over an
have been
to
[4]
-heptane
tempera-
peroxyesters
and
used
kbar
are
as
of few
a
ture
high-pressure
pressures
To mimic
range
ethene
up
found in
situations
the
that are
monomers or
high-pres-
polymerization.
in
of
a
free-radical
sure
copolymerizations
polymerization
polar
has
monomer
where
least
sition kinetics
at
one
at
polarity, peroxide decompo-
significant
has also
studied in solutions of
toluene,
been
high
pressure
acetonitrile
and
chloroform,
dichloromethane,
chlorobenzene,
propionitrile,
first-order
almost
co-
showed that the
increase
rate
These
[5].
decomposition
order of
investigations
one
in
efficients of
from
magnitude
peroxides
by
diacyl
acetonitrile.
solvent
to
a
going
It
hydrocarbon
acid influences
added
whether
worthwhile to
in
carboxylic
study
appears
the
an
same or
to
solvent
to
an
perhaps
unpolar
peroxide decomposition
materials.
been
above-mentioned
the
than
extent
even
into
Investigations
polar
larger
carried
and
have
out on
acid-induced
and
already
decomposition
peroxide
ambient
at
observed.
to
[6—9]
significant
peroxides
diacyl
pressure
detailed
peroxyesters
For a
such
of
been
enhancements have
rate
effects
of
understanding
the
acid-induced
reasonable
it
appears
decomposition
investigate
as
basic
looked
the
which
be
(DTBP)
upon
may
di-ierr-butyl peroxide
relevant
peroxide
technically
component.
in several
arises
acid-induced
as
The interest in
decomposition
peroxide
with
ethene
The
of
free-radical
COOH
acid,
(meth)acrylic
copolymerizations,
e.g.
are
acid co-
groups
ethene-(meth)acrylic
necessarily
present.
are
and technical interest.
considerable scientific
of
They
polymerizations
and
to
conditions
close
this
carried
out at
are
high
temperatures,
For
fairly
pressures
in-
the
used in ethene
reason,
homopolymerization.
high-pressure
need
also
acid
with
into
being
peroxide decomposition
present
vestigations
decided
and
at
We
to
be
to
study
high
temperatures.
(DTBP)
performed
pressures
with
in
of
the
-heptane
acid)
di-teri-butyl peroxide
decomposition
acid
different
added.
of
amounts
being
(2,2-dimethylpropionic
pivalic
a
classical
bond scission
of
is
DTBP the
homolytic
showing simple
example
peroxide
de-
has
of literature exists
DTBP
and
The
on
ambient
at
a
body
large
single
of DTBP
of
[10—18].
decomposition
pressure
cyclohexane,
composition
such
wide
studied in
as
been
a
solvents,
variety
already
al-
i-amyl
tetrahydrofurane,
dimethylaniline, cyclohexene,
triethylamine,
and acetic acid
alcohol, acetonitrile,
nitrobenzene,
cohol,
by Huyser
/-butyl
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 9/22/15 1:29 PM

Acid-Induced
and Van
authors at
135°C these numbers
63
of
Peroxide in
Solution
Di-tert-butyl
Decomposition
The
-Heptane
rate
first-order
between 115 and 135°C
measured
these
rather similar. At
solution within factor
coefficients,
[10].
kobs,
Scoy
by
are
temperatures
a
with
values
for
fcobs
-heptane
agree
kobs
temperatures
variation of
of
of
in
measured the various
two.
includ-
solvents,
Extrapolation
to
around
num-
170°C
almost identical
more
ing -heptane,
yields
bers. No
with solvent
than
kobs
significant
polarity,
by
say
even
is
50
for the 180
200 °C
range
to
cent,
expected
per
major
(pure)
upon
in molecular
to
in
solvent from
environment,
changes
changing
e.g.
acetonitrile and
to
acetic acid. This
is
con-
cyclohexane
(pure)
finding
sistent with
studies
rate
DTBP
in acetonitrile
on
performed
decomposition
variation of
at
and
for
[11], where,
temperature
by (only)
significant
pressure
e.g.
500 bar
and
enhancement of
an
40
cent
180°C,
kobs
compared
per
measured in
The kinetics of
to
has been found.
at
kobs
[12]
n-heptane
DTBP
2
kbar
and
up
to
decomposition
pressures
up
200°C has
to
been determined in
the influence
measured via
the
In
[11].
already
(pure) n-heptane
study
of
acid
DTBP
on
FT-IR
in
here,
presented
-heptane
pivalic
quantitative
decomposition
2
is also
to kbar
spectroscopy
up
and 200°C.
2.
Experimental
The
kinetics
measured in
were
stainless
made from
a
steel
decomposition
batch-type
autoclave
The cell for
with
windows
to
two
(W.St.-Nr.
crystalline
1.4122)
equipped
poly-
silicon.
°C
and 2000
200
bar has
operation
up
been described
the
At the relevant reaction
above
already
[12].
temperatures
fast
allow
too
autoclave
to
for intro-
conditions and
avoid
150°C,
as
peroxide
sample
decomposition
proceeds
the
the
solution into the
ambient
at
ducing
within several
the desired
to
To
cell,
of
minutes,
heating
temperature.
the
the
reached,
before
the desired
problem
having
peroxide mostly
decomposed
reaction
the
maximum
mental
is
the evacuated
cell is heated
to
temperature
only
optical
reaction
The
solution
is
to
a
sample
temperature.
pre-compressed,
up
of
3
in the
of the
kbar,
pressure
pressure-generating
part
experi-
room
maintained
to
at
where
is
rate
set-up
temperature,
decomposition
even
low
reaction
within
extended
storage
times.
cell,
any
sufficiently
prevent
After
the reaction
in the
a
reaching
optical high-pressure
temperature
valve between the cell and the
takes about
is
It
pressure-generating
opened.
system
5
10 seconds until the desired
to
reaction conditions
includes fine
of and
of the
reached within the cell. This time
are
period
tuning
standard
desired value.
generator,
to
a
pressure
The
valves and the
are
capillaries
pressure
gauges,
Pressure
measured
DMS transducer
was
a
(HBM-MeBtech-
equipment.
by
Class
nik,
maximum
3
recorded with
was
2,
a
kbar).
Temperature
pressure
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 9/22/15 1:29 PM

Buback and Lars
Wittkowski
64
Michael
and
CIA
S
CGA
con-
and
sheathed
trolied
(Ni/Cr-Ni,
250,
Alsthom)
in
thermocouple
PID
a
The
instrument
(Eurotherm 820).
pressure
accuracy
by
and
bar
±
is
6
was
determination
±0.25°C,
respectively.
temperature
con-
cell.
after
Collection of
data
started
conditions in
immediately
reaching
spectral
the
and
were
stant
optical high-pressure
temperature
pressure
IFS 88 FT-IR/NIR
recorded
on
Bruker
a
The
spectrometer
spectra
a
a
and
a
CaF, beam
source,
For the
liquid-nitrogen-cooled
splitter,
halogen
using
acid
a
with
InSb detector.
methacrylic
being present
experiments
was
data collec-
the de-
used. To
recorded
Fourier transformation
MCT detector
speed
spectral
nitrogen-cooled
up
the
tion,
were
(and stored)
during
interferograms
only
absorbance
of
calculation
of
and
reaction.
these data
measured
composition
reaction. The
served
the
after
from
was
completion
performed
spectra
IR
reaction
the evacuated cell at
on
temperature
spectrum
reference
the
as
spectrum.
in
of
ml. Prior
about 100
to
Peroxide solutions were
quantities
prepared
the
solution
was
thaw
four freeze and
the
The
cycles.
experi-
degassed by
close 0.15
experiment,
all
in
concentration
to
DTBP
was
mol/kg heptane
measured with an
ments
each
of
the
with
amount
being
accuracy
component
was
PA
the solutions
series of
solution. The
con-
±
added
to
0.001
Pivalic acid
to
one
of
(PA)
yield
g.
up
In
meth-
used
0.178
added
centrations
to
was
experiments
mol/kg heptane.
materials
the
to
DTBP
Merck),
acid
DTBP/heptane
acrylic
acid
reactions
in the
i-butanol
tane
were:
(99.5%, Fluka),
(>98%, Merck),
pivalic
and
acid
(99.5%, Fluka),
(>99.5%,
n-hep-
methacrylic
Merck).
(>99%,
3. Results
200°C
series
measured
at
and
acid
In
1
absorbance
decomposition
is shown
an
spectral
DTBP
Fig.
500 bar
in
mol
with
of
the
yield
pivalic
-heptane
during
a
The initial
absorbance
concentration of 0.122
added
to
/kg heptane.
being
The
mol
0.15
concentration was
3600 and 3650 cm
This
DTBP/kg heptane.
peroxide
is due
which increases
to
between
reaction,
during
reactions of the
acetone
-abstraction
is formed in
radicals. The other
from the
í-butanol
intermediate
and
icals
of the
sorbance
(TB).
product
are
(AC)
rad-
H-abstraction
major products
r-butoxy
reaction of
methane.
which
r-butoxy
originate
They
/^-scission
to
formation
AC and leads
CH4
by
directly yields
'
is
the ab-
due to
3550
radicals. The band at
cm
produced methyl
is found
this acid absorbance
No
in
of the PA monomer.
reaction.
change
the
decomposition
during
solutions
on
containing varying
experiments
spectroscopic
Independent
used within
concentrations
the
similar
of PA
showed that
as
of TB and
amounts
kinetic
(at
for
the PA
Beer-Lambert's law holds
experiments)
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 9/22/15 1:29 PM

65
Acid-Induced
of
Peroxide in
Solution
Decomposition
Di-ierr-butyl
n-Heptane
3700
3650
3600
3550
3500
the
/
cm"'
1.
of
infrared
the
absorbance
at 200°C
collected
and 500 bar in the
Fig.
Hydroxyl region
spectra
during
decompo-
DTBP
sition of
acid
of
(0.15
heptane)
mol/kg
presence
pivalic
The f-butanol
band
around 3550 cm"1
around 3625 cm"1 increases
(0.122
heptane).
(TB)
during
mol/kg
whereas the acid
remains constant.
decomposition
absorption
1
monomer
TB
in
band
solution
3550 cm"1
at
cm
at
and for the
the and
band around 3630
conditions of this
work
This infor-
absorbance of
[19].
n-heptane
mation
with the observation
band in the kinetic
of an almost invariant
together
3550 cm"'
the
concentration
indicates
that the PA
1)
experiment
(Fig.
constant
the
of the
out at
course
carried
reac-
during
experiments,
stays
decomposition
tion.
Also within the other
of
180,
temperatures
190 and 200°C
and at
between 100
and 1900
indication
no
bar,
pressures
of
3550
in PA concentration is
the absorbance
around
change
provided
by
any
reasonable
cm"1.
Thus it appears
the
solution
present:
to
first-order
both the
use
a
law
rate
(1)]
[Eq.
DTBP
concentration
for
in
fitting
during
PA and for
decomposition
some
the
without
(1)
-heptane
containing
decomposition
acid
being
d C[)TBP
^obs
CdtBP
;
dr
are
butanol
and
identified
the
acetone
(TB)
(AC)
Tertiary
products
only
containing
of this
oxygen
that
from IR and GC
in the
are
analysis
samples
TB
rate
and AC
be
first-order
to
reactions,
study. Assuming
produced by
&pACa,rallel
characterized
coefficients
and
allows
to
kTB
by
respectively,
derive
fraction
of
that the
DTBP that
reacts
to
(2).
TB,
Eq.
By assuming
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 9/22/15 1:29 PM

Michael Buback and Lars Wittkowski
66
5.0
4.5
4.0
3.5
3.0
2.5
h
2.0
0.05
ciò
o
o
O
o
o*
rP
r, °Q
°
°
CP
o
%
o
O
O
0.00
O
O
O
O
-0.05
100
50
150
200
t/s
OH
mode of TB
of DTBP in
absorbance of the
2.
stretching
(see text)
«-heptane
Fig.
Upper part:
Integrated
of reaction time
t
the
measured as
200 °C and 500
concentration was 0.15
a
function
during
decomposition
The initial DTBP
at
at
concentration of 0.122
Lower
bar
an acid
mol/kg heptane.
residuals with
the curve
to
part:
respect
mol/kg heptane.
fitted
(2).
by Eq.
a
DTBP
ratio, kobs
concentration,
+
constant
kTB/(kTB
kAC),
stays
decomposition
is
during
particular
a
number for this
measured
without
even
reaction,
directly
knowing
TB
via
accessible from the
cTB,
spectroscopically
(2).
Eq.
ctb
In
2
(2)
Ctb
Cdxbp
and
{~~kobs t}
exp
TB
refer
concentration after
to
(2),
c^B
CdTBP
complete decompo-
The overall de-
Eq.
the initial DTBP
sition and
to
concentration,
respectively.
is found
non-linear
rate
coefficient, /cubs,
by
regression (using
composition
of the
measured TB
the
algorithm)
spectroscopically
Levenberg-Marquard
concen-
Within this
the TB
time data.
concentration
vs.
fitting procedure
between 3620 and
the absorbance at
the TB absorbance
is
tration,
cTB,
replaced by
integrated
'
baseline defined
horizontal
3645
a
cm
by
against
1
non-linear
of the
cnr
shown in
An
3720
1).
(not
advantage
Fig.
important
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 9/22/15 1:29 PM

67
at
in
Acid-Induced
of
Peroxide
Solution
Di-iert-butyl
Decomposition
-Heptane
of
Table 1. Overall rate coefficient
for the
DTBP in
The initial
&„,,s
decomposition
and
-heptane
different
concentration
acid concentrations,
pivalic
peroxide
temperatures,
pressures.
'"
0.15
coefficients
was
rate
derived from
a
fit of the
mol/kg heptane;
temperatures
fit of the literature data for 190°C at
(2)
literature data for 500 bar and
derived from
from 140 to
rate coefficients
from 100 to 2000 bar.
200°C;
a
pressures
to
Rate coefficients deduced from Ref.
or l2).
(*)
[12], according
procedures
10-Vs-'
at 500 bar
£„bB
cPA
mol/kg heptane
180°C
190°C
200°C
0(D
0
4.22*
4.12
4.67
5.16
7.09
9.42
9.92
9.98
10.0*
9.63
9.76
11.2
14.7
15.4
18.2
16.5
23.0*
22.0
0.029
0.033
0.078
0.122
0.167
0.178
23.7
28.8
28.9
33.8
31.4
107s-'
at 190°C
500 bar
k,hs-
cPA
mol/kg heptane
100 bar
1000 bar
1500 bar
1900 bar
*
*
*
*
*
11.1
0,2>
10.3
9.2
8.2
9.55
13.7
16.4
7.6
9.11
14.1
0.033
0.119
0.178
11.8
14.9
15.1
11.2
14.6
16.5
10.1
14.0
16.2
15.4
a
fact that
relates
the
absorbance
constant
to
regression
due
background,
[19],
e.g.
solvent absorbance is considered in the
to
and
kobs. fitting
procedure
therefore does
interfere the
for
the
not
analysis
this
Fig. yields
kind of
reaction
to
analysis
Applying
decomposition
depicted
1
the TB
in
absorbance
time
shown in the
vs.
plot
part
upper
an
of
2. Each data
is derived from
Fig.
point
independent
spectroscopic
The number of data
2
num-
in
measurement.
exceeds the
points
is
Fig.
largely
a
ber of
measured
shown in
1. This because
fraction
of the
Fig.
only
spectra
actually
In the lower
the
has been
of
2
difference
plotted.
fitted
Fig.
spectra
part
measured and
absorbance is
between
of residuals demonstrates the
This
(2)]
[by
Eq.
presented. plot
of kinetic
data
the
adequate representation
by
first-order
law
and
spectroscopic
rate
(2)].
(1)
[Eqs.
data
of the
measured
the
of
DTBP
course
Analysis
during
the other
and
rate
at
PA concentrations
decomposition
temperatures, pressures,
behaviour.
first-order
shows the
of
via
The
same
coefficients,
kobs,
type
in
to
derived from
The
Included in Table
Table 1.
be better
are
analysis
(2),
given
Eq.
of
1
values is estimated
than ±10 per
study
cent.
of DTBP de-
kobs
accuracy
the
values from the earlier
are
kobs
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 9/22/15 1:29 PM

and
Wittkowski
Lars
68
Michael Buback
/
c PA
(mol/kg heptane)
coefficient
corre-
concentration of the observed rate
3.
on
acid
in
of ±10%.
(cPA)
Fig.
Dependence
DTBP
an
pivalic
180°C and 500 bar. The error bars
at
for
(kobs)
to
spond
decompositions
rc-heptane
uncertainty
These data
marked
(at
points
in
are
[12].
«-heptane
0)
(at
cPA
composition
(pure)
work
500 bar and
three such data
180, 190,
literature
the
with
asterisk. Within the
have been measured:
an
0)
present
only
and 200°C. The
data.
at
cPA
with
is
in
be
numbers
As
are
very
pleasing
agreement
toward
from Table
seen
enhanced
can
kobs
1,
significantly
low
At
The
influence of
is less
pronounced.
PA
higher
pressure
whereas
temperature.
lowers
seen.
concentrations
higher
DTBP
PA concentration for
the
at
levels of cPA,
kobs
pressure
can
no
be
is
clear trends
the
of
and
Plotted in
3
on
kobs
dependence
Fig.
at
an
180°C
the
500 bar. Within
of
carried
out
uncertainty
decompositions
the
few
values
be fitted
concen-
±10
cent,
per
error
bars,
kobi
represented by
may
PA
at
line.
is indicated
As
a
a
by
experiments
by straight
very
of
0.178
does
continue
lin-
trations
not
to
concentrations.
A closer look
kobs
exceeding
mol/kg
-heptane,
with
3
such
PA
increase
at
at
cPA
higher
earily
indicates that for
data in
and in Table
1
>0.1 mol
cPA
the
Fig.
PA/kg
in
is less
within
and,
the increase
kabs
pronounced
experimental
heptane
of
addition of
is
further PA
clear enhancement
the
no
kobs
uncertainty,
upon
fit the
data
at concen-
Nevertheless it
to
vs.
seen.
kobs
cPA
justified
appears
a
linear
trations
0.178 mol
relation. In
for
at constant
to
so,
/kg heptane by
decomposition
doing
up
each of the
studies carried
out
is achieved.
3
experimental
and
correlation of the
shown in
as
a
,
quality
Fig.
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 9/22/15 1:29 PM

Acid-Induced
of
in
69
Peroxide
Solution
Decomposition
Di-ierr-butyl
-Heptane
4. Discussion
into
of both mechanistic and
use
are
Investigations
peroxide
decomposition
the latter
initiators in
influence the
due
the
of these
As the
to
mate-
mono-
interest,
practical
being
widespread
free-radical homo- and
as
rials
mers
copolymerizations.
and
of
may
reactivity
selectivity
decomposition
decomposition
peroxide
their
direct
in
the
or even
a
through
polarity
by
participation
detailed studies into the
variation of molecular
kinetics
at con-
peroxide decomposition
process,
siderable
solvent
environment,
temperature
rate are
by changing
type,
solvent
and
are
concentration,
required. Particularly
pressure,
on
effects
are
when COOH
to occur
strong
decomposition
expected
The
interest in
such
is due
to
The
practical
groups
present.
acrylic
systems
and
acid
(co)monomers.
in
solution of
methacrylic
being important
present study
DTBP
on
focuses
with different
decomposition
-heptane
of
acid
amounts
added
mimic situations
to
as occur
in
(PA)
pivalic
being
acid
As these
technical reactions
studies
ethene-(meth)acrylic
copolymerizations.
are
carried
out
and
at
an
also
our
are
high
pressures
temperatures,
per-
formed within
The
been measured
extended and
range.
DTBP
kinetics
few such
(see
of
in
has
decomposition
(pure) -heptane
already
in
A
have
also been carried
out
[12].
experiments
It
the
an
Table
±
is
within
is found
with the entire
to note
that,
agreement
1).
per
present
investigation
gratifying
of 10
excellent
cent,
experimental uncertainty
with the numbers from
the earlier
(where
study
literature
has been
comparison
of
made).
body
existing
As
be
from
3 and from
can
seen
the entries in Table
enhanced
the observed
the addition
1,
Fig.
coefficient, kobs,
in
first-order
is
rate
significantly
upon
0.1
cPA
of
concentrations
at
to
acid,
pivalic
particular
mol/kg
up
decomposition
hep-
tane. The
Acetone
measured
but it is
is i-butanol.
spectroscopically
product
PA
is also formed
in the
absence of
or at
at
(AC)
only
very
low levels of PA
identified via
the
mode around
easily
carbonyl stretching
cm-1
1720
Acetone is
reaction of intermediate
a
[4].
/?-scission
produced by
radicals.
Thus methane
^-scission)
-abstraction of the
In
i-butoxy
produced by
be
will
methyl
radicals
this
another
solution
of
(from
product.
these small amounts of methane
be detected
not
however,
may,
rc-heptane
IR
via online
spectroscopy.
1
The kinetic observations
be
Scheme which
may
represented
by
applies
both
to
:
in the absence and in the
of PA
Scheme
decomposition
presence
1
should
It
be noted that the individual reactions in
not
are
balanced for
is
are:
from the
The obser-
solvent).
the
hydrogen (which
picked
up
(i)
in the
vations
into Scheme
that
1
TB and AC
are
going
only
oxygen-
Gas-chromatographic
are seen
containing products
spectra.
indicates
that
other
formed in
no
are
analysis
oxygen-containing
products
above 0.1
relative
to
cent
TB and
The concentration of
quantities
decomposed
peroxide; (ii)
per
with and without PA
AC
occur
being
(iii)
present;
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 9/22/15 1:29 PM

Michael Buback and
Wittkowski
Lars
70
*'
)
)
DTBP
DTBP
+
2
TB
k:
+
2
AC
TB
2
CH4
PA
+
)
DTBP
PA
PA
2
+
+
>
DTBP
2
AC
2
PA
+
CH4
Scheme
Reaction channels
1. Scheme for the
of
respect
in
DTBP
with PA
hydrogen (see
-heptane
being present.
decomposition
balanced with
are not
to the amount of
text).
a
acid does
the
of
reaction
up
not
course
PA
pivalic
to
during
decomposition
vary
conversion. Thus
in Scheme
DTBP
is added
both sides of
third
the
on
complete
and fourth
1.
equation
The relation between these individual coefficients and the
experi-
+
+
+
is:
measured
the individual reaction
As
ka2
kobs
k,
k2
&al
kobs.
cPA
cPA
cannot
quantity
mentally
and AC
the
dis-
be
TB
yielding
pathways
within
tution is made:
of
substi-
the
the
tinguished
following
experiments
study,
present
~ ~
k\
kai
a
k2
kHE
>
(3)
(4)
"
kd2
kPA
coefficient
As
the observed overall rate
be
re-
kobs
consequence,
presented by
may
:
"f"
&obs
&HE
^PA
(5)
CpA
where
refers
the first-order
to
coefficient of
rate
DTBP
to
kHB
decomposition
(pivalic)acid-
in
and
rate
the second-order
coefficient of
because
kPA
-heptane
It should be noted
induced
constant
of
that,
cPA
decomposition.
remaining
each
the
in
term
coefficient.
reaction,
(5),
during
decomposition
product
Eq.
first-order
be looked
as
a
rate
£pA
may
CpA,
From
pseudo
upon
been shown
kPA,
according
cm
the fitted linear
of kobs
which has
CpA,
dependence
and
are
the
for
to
the
180°C in
coefficients
500 bar
and
ordinate
rate
3,
kHE
Fig.
the
available from the intersection with
and from
values derived
at constant
(5),
Eq.
of the
The
and
line,
respectively.
kHE
kPA
slope
this
straight
are
in Table
2
for
three
by
procedure
(500 bar)
given
temperatures
constant
and for five
increase with
at
(190°C).
pressures
temperature
pressure
values
The
and decrease with
is
kHE
kPA
temperature
temperature,
pressure.
enhanced toward
but
increases
also
with
too,
higher
pressure,
least
1500 bar.
to
at
up
Shown in
4
is the
of
and of acid-in-
kHE
Fig.
dependence
the
temperature
duced
first-order
rate
coefficient
decomposition, represented by
pseudo
for the
chosen concentration of
0.1
kPA
cPA
EA(kHE), largelycPA
arbitrarily
mol/kg hep-
The activation
of
plotting
tane.
exceeds
The
fac-
are
kHE,
EA(kPA).
energy
difference is visualized in
tor, A,
activation
vs.
pre-exponential
energy
in
5. This
takes into
that both
account
Fig.
diagram
quantities
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 9/22/15 1:29 PM

Solution
71
Acid-Induced
of
Peroxide in
Decomposition
Di-rert-butyl
-Heptane
Table 2.
sition
coefficients of non-induced
Rate
and induced
decomposition (¿he)
calculated via linear
decompo-
to
DTBP
Data
of
in
are
(kPA)
-heptane.
regression according
The
in
and
km
is
determined
the
of
root
the variance
Eq. (5).
uncertainty
kPA
by
square
of
the
and of the
from linear
intercept
slope
regression.
1
77°C
500
107s"1
s
mol™'
molm'
kHK
(at
bar)
kPA 107kg
180
190
200
±
±
±
±
±
±
4.1
9.6
22
0.3
0.6
1
36
46
60
3
6
9
107s"1
s~'
plh&r
190°C)
(at
kHk
kPA
107kg
±
±0.1
±
±
±
±
±
100
500
1000
1500
1900
11.3
10.2
9.0
8.1
7.8
0.5
24
36
41
47
46
5
3
2
1
±
±
±
0.2
0.1
0.6
5
correlated via
relation
confidence intervals
the Arrhenius
A
(k
EA/(R
highly
exp{
T)}).
Given in
fidence level of 95
The JCI's in
5
the
are
to
a
con-
Fig.
joint
(JCI's) referring
cent
are
[20].
per
5
rather extended
confidence level of 95
of
as a
Fig.
consequence
cent
applying
three
a
the F-distribution
at
data
to at
only
per
Calculation of
rate
with
the
JCI's
coefficients from Table
and with the
^-distribution
temperatures.
2
variance
of
2
in
results smaller
The
ellipses.
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 9/22/15 1:29 PM

Wittkowski
72
Michael Buback and Lars
180
160
140
120
100
so
c
jj_ _lJ_lJ_lJ_ij_
IO13
IO9
10"
IO15
/ß"1
acid-induced
Arrhenius
of DTBP in
for
of the
decomposition
confidence intervals
5. 95% Joint
parameters
«-heptane
Fig.
(in*
0.1
500 bar,
at
non-induced
and
cPA
(fcHi
)
mcoPl/Ak)g heptane.
for the de-
determined
are
factor
A
and
EA
3. Overall activation
Table
frequency
corresponding
energy
uncertainties of
and
A
at 500 bar. Estimated
of DTBP
EA
composition
the
values
limits of JCI's and
difference between the outer
from the
Reaction
(see text).
optimum
1
mol
A/s
EAfk]
System
"'
at
between
measured
[12].
concentration of 0.1
PA
500 bar and
Calculated from rate coefficients
Ref.
temperatures
and 200°C from
140
(2)
for
Calculated
a
mol/kg heptane.
JCI's and of
the
some
structure
error
detail elsewhere
of
the
taking
properly
calculating
procedure
is described in
into
of the data sets
account,
quite
[20].
activation
energy
and
for the
factor
3
means
the
are
Given in Table
pre-exponential
of
de-
non-linear
of
and
determined
decomposition
regression
by
DTBP
for the
in
(pivalic)acid-induced
corresponding
-heptane
activation
the
and
from Table
As
be
can
seen
3,
in
composition.
exponential
energy
from
pre-
in
data
data of Ref.
factor from
this
[12]
rc-heptane,
(pure)
of Table
2
in excel-
and from the
data
are
of
kHE
work,
-heptane
(pure)
lent
identical.
of data
The JCI's of these three
sets
are
nearly
agreement.
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 9/22/15 1:29 PM

Peroxide in
73
Solution
Acid-Induced
of
Decomposition
Di-rert-butyl
-Heptane
-4.0
I
bar
6. Pressure
of the rate coefficients
for acid-induced
non-
and
0.1
dependence
(kPA
cPA)
Fig.
induced
of
in
DTBP
solution
at
(¿HE)
-heptane
190°C,
cPA
mol/kg
decomposition
heptane.
of the
extended JCI
values
in
function
of
as
a
leads
the
of
to
kPA
cPA
Analysis
temperature
from
which the activation
5
±
depicted
Fig.
very
kPA
energy
this value it
is
deduced
be:
46 33 kJ mol-1.
to
In
EA(kPA)
deriving
is assumed that the
of
is
due
to
energy
for the non-induced
kPA
cPA
dependence
entirely
temperature
the
in
is
this activation
kPA.
EA(kPA)
Although
uncertainty
fairly large,
is far below
the activation
EA(kHE),
undoubtedly
energy
process.
Before
the
the
to
activation
two
and of
kHE
are so
turning
question why
energies
the
of
Fig.
0.1
190°C,
mol/
different,
kPA
(at
pressure
cPA
cPA
dependence
will
addressed.
vs.
be
In
6
the
coefficients
rate
for
kg heptane)
from
Table
are
2,
plotted
pressure.
V*
The activation volume
for the non-induced
,
—RT(d\nk/dp)T,
8.2 ±0.7
is found
cm3
derived
be:
(k^)
in
to
excel-
mol"1,
decomposition
V"*(/che)
cm3
lent
the
with V*
procedure
8.5 ±0.7
mol"1
agreement
by applying
same
the data from Ref.
is
to
[12].
fitting
activation
The
volume for
in absolute
curved which
not
kPA
different,
vs.
strikingly
only
·
value but also in
the
is
Moreover,
ln(/cPA
cPA)
sign.
plot
that
the
derived activation
volume
the
on
formally
depends
says
From
pressure.
fit
-25
the
500 bar
activation
activation
V*
are
to
volumes
mol"1
The
to
data,
parabolic
(kPA)/cm3
and
of: -37
1
-14 1000 bar
calculated.
at
at
at
bar,
volumes
low
at
close
are
negative
relatively large
pressure
values found in associative
whereas the
or
bond-forming
positive
processes
characteristic for bond-scission
for
activation
observed
as
are
kHE,
volumes,
reactions and thus
cess.
the
for
meet
a
expectations
peroxide
decomposition
pro-
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 9/22/15 1:29 PM

Wittkowski
Michael Buback and Lars
74
in
for the
difference
What
rate
be the
reason(s)
rc-heptane
may
DTBP
decompo-
of PA?
significant
and
the
in
sition
of
in
in the absence
in
presence
via
There is
debate that
no
decomposition
-heptane proceeds
(pure)
radicals within the
factor,
homolytic 0-0 bond scission
primary
yielding i-butoxy
of acti-
The measured values of
kinetic
vation
With acid
[13].
pre-exponential
step
energy,
with this view.
consistent
of activation volume
and
being
are
fully
a
follow
reaction
alternative reaction
does
not
the
simple
obviously
present,
or
contribute
via
The
route.
pathways
may
following
may
DTBP
ionic
dominant in the
of PA :
an
become
(a)
influence
indirect
decomposes
presence
an
with the
rate
(c)
;
in that acid reacts
There is
mechanism; (b)
DTBP
affect
metal ions that
metal walls
An induced
to
decomposition
produce
with acid molecules
free-radical reaction
occurs
participating
medium is
asso-
of the solvent
A
The
an
to
in
transfer
the acid
(d)
polarity
hydrogen
processes;
rise
in
increase
enhanced
kobs (e)
;
specific
giving
by
DTBP and PA
associated
free-radical
allows for
ciation between
a
pathway
a
much smaller activation
with
energy.
(a)
will
be discussed.
now
of
The relevance
(e)
arguments
mechanism
Ionic
(a)
ionic mechanism.
an
indication of an
some
of
The low value
On the other
EA(kPA)
provides
is
no
acid-
yield
medium for
suitable
hand,
overly
n-heptane
reaction would
reaction
a
to
Moreover,
protonation
proceed.
catalyzed
an
be
and
an ester.
intermediate carbocation which
with
to
must
yield
Thus
and
expected
rearrange
acid anion
form
to
react
a
or
pivalic
2-methoxypropene
in
mechanism should be
of
ionic
a
an
major change
accompanied by
type
whether such
oc-
To
see
concentration
two
changes
decomposition products.
in
molar solution
DTBP
to
one
of
DTBP
of a 0.1
cur,
/i-heptane,
decompositions
PA
a
molar
1:1
PA
to
without
am-
ratio of
and one
with
acid,
a
yield
adding
have
in sealed
under helium
been carried out
°C. GC
glass
atmosphere
than
the
more
which contains
of
the
175
at
liquid phase
analysis
poule
distribution listed in Table 4. This
%
of the
95
yields
product
the
at
out
products
of Akzo
expected
Deventer
contained the
has been
Research Centre
carried
which
GC
Nobel.
analysis
has
also been
The
analyzed
phase
gas
and i-butanol.
methane, ethane,
acetone,
products:
±
within
of
balance
10
entries in Table
4
demonstrate
The
an
that,
per
accuracy
of
and for the DTBP
for the individual
mass
amounts
cent
(sum
of
after
DTBP
and i-butanol
acetone
and the
amount
times the
the
for
amounts
0.5
concentration of
is
acid.
differ-
should be noted
the reaction of
and the
not
products
distinctly
reaction),
type
without
with and
It
ent
pivalic
decompositions
concentration of
which
that the
C7-dimer,
-heptane, unambiguously
these
significant
proves
shows that radicals
an
radicals with the solvent
free
are
ionic
it
be ruled
that
out
reactive
For
the
reasons
can
species.
contributes
the acid-induced DTBP
mechanism
to
decompo-
dominantly
reaction.
sition
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 9/22/15 1:29 PM

75
in
Solution
of
Acid-Induced
Table 4.
Peroxide
Decomposition
Di-tórí-butyl
-Heptane
distributions for
atmosphere
of DTBP in
the
decomposition
acid
n-heptane
Liquid phase product
Pivalic acid
being
present.
°C
with and without
at 175
under He
pivalic
M.
DTBP
balance refers to the
were 0.1
The
and initial DTBP concentrations
oxygen-
containing species.
Product
amount
Product
DTBP)
(mol/mol
without PA
with PA
0.619
1.272
0.004
0.572
0.612
1.184
0.003
0.520
Acetone
?-Butanol
DTBP
CVdimer
DTBP-balance
95.0%
90.1%
Metal ion
(b)
catalysis
metal
As
enhance
is
ions
reason
well-known,
For
may
significantly
peroxide decompo-
a
sition
this
the autoclave has been manufactured from
are
rates.
stainless steel which is
in
active metals. There
two
reac-
catalytically
poor
metal ion
is
The
other
tion
not
(1)
why
catalysis
overly likely:
arguments
follows first-order kinetics which makes it hard
believe that
to
closely
the
that
occurs
a
of active
in
affords for
(metal ion)
way
production
species
the observed
first-order
and
no
has been mentioned
rate
as
kinetics;
(2)
simple
concentrations
further
toward
PA
be
enhancement is
before,
found.
surface is
higher
It
that the
should
of the internal
material.
reactor
noted,
silicon,
finally
major
part
a
window
the
Thus,
polycrystalline
although
a
contribution from metal ion
be
such mechanism
cannot
excluded,
catalysis
be of minor
at
best,
can,
importance.
free-radical
where
Induced
(c)
process
initiator is
Induced
reaction
free-radical chain
a
decomposed
by
processes
are
known
contribute
to
to
[13].
strongly
in
atom
hydroperoxide reactivity
the 0—0
With
to
DTBP is
no
bond, however,
-position
hydrogen
candidate for
contribution of induced
obvious
sition.
transfer
induced
a
no
significant
decompo-
no
acid has
free-radical chain-
Moreover,
pivalic
particularly
strong
If chain transfer would
the
contribute
acid-
to
activity.
significantly
observation
sec-
(see
experimental
introductory
decomposition,
DTBP
almost the
in
in
of
acetic
at
same rate
and
decomposing
cyclohexane
unlikely
tion)
pure
acid could
be understood. It thus
are
that
acid-induced
hardly
appears
for the observed
reactions
free-radical induced
responsible
of
DTBP.
decomposition
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 9/22/15 1:29 PM

Michael Buback and
Wittkowski
Lars
76
Solvent
(d)
polarity
studies
As has been shown
anoyOperoxide,
on
bis(3,5,5-trimethylhex-
by high-pressure
solvent
rate,
may
vary
polarity
going
significantly
decomposition
7
factor of in
from
acetonitrile. The solvent
a
to
e.g.
by
-heptane
DTBP
on
is
influence
be much less
indicated
seems to
as
decomposition
by
the ambient
ies into solvent
data of
and
DTBP
stud-
[10].
High-pressure
VanScoy
decomposition
who
Huyser
pressure
are
of
available with
decomposition
not
DTBP
dependence
measured
500 bar he found
the
of
a
Drefahl,
exception
study by
and
10"3
180°C
40
5.8
in
s
acetonitrile
For
kobs
[11].
(pure)
same
above
the
at
and
which is
about
cent
lcHE
(only) by
per
value of the
for
per
where the ratio
conditions. The
0.178
cPA
cent
corresponding
investigation
present
10~3 s_1
is
which is
about 140
by
9.98·
mol/kg heptane
in
acid
is
above
As the
value measured
molecules
the
solution,
clearly
kHE.
kobs
PA
above
in the
of
to
1:600,
equals
kobs
«-heptane
it
be concluded that bulk solvent
is
acetonitrile,
may
polarity
highly polar
for the acid-induced enhancement observed for DTBP. At
not
such low
responsible
PA
trans-
in
to
concentrations of
relative
also
n-heptane
changes
be of
to
reason
relevance for acid-induced
coefficients
are
unlikely
port
very
It
any
DTBP
or
is for this
that diffusion
effects
decomposition.
viscosity
will
be
considered
the
that
for the
account
not
among
sig-
arguments
may
nificant
in
kobs.
changes
association
Specific
(e)
volumes
activation
A
the
that
an
for
V*
The
associative
sition.
kPA,
(kPA),
suggest
large negative
to
contributes
ambient
measured for Diels-Alder
acid-induced
37
reactions,
process
significantly
decompo-
·
for
mol
cm'
is close
')
The AW*
value
(kPA)
(
pressure
volumes
which
activation
are as-
to
via
transition
That
acids
sumed
structures
to occur
[21].
cyclic
carboxylic
is
form such
well-known from the existence of their
The
structures
cyclic
dimer
may
PA
dimerization
of
in di-
pressure
[22].
equilibrium
hydrogen-bonded
species.
solution of
temperature
has
been studied in extended
lute
and
rc-heptane
already
2000 bar and
on
to
175°C,
respectively
DTBP-PA
of the DTBP
up
ranges,
direct evidence
measured
the existence of
There is
no
complexes,
from
the
the
course
during
decompo-
spectra
complexes
e.g.
sition.
and
are
If such
will coexist with PA
identified from the IR
observations made for the PA-
in
not
monomer
formed,
they
dimer
and thus will not be
easily
species
spectra.
various
are at
existence of such DTBP-PA
be noted that
It
should, however,
DTBP
of
least
with
the
induced
disagreement
decomposition
post-
argu-
the
addition
In
to
complexes.
ulating
A
also
least
at
be
about
ment
V*(kPA),
EA(kPA)
qualitatively,
explained
may,
that such
formed
DTBP-PA
In
case
are
rapidly
by
complexation.
complexes
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 9/22/15 1:29 PM

77
of
Peroxide in
Solution
rate
Acid-Induced
Decomposition
Di-íerí-butyl
-Heptane
as
coefficient
the -induced
their
kd
to
to
kd,
rate,
decomposition
compared
DTBP-PA
and the
com-
to
for
constant
is related
kPA
Kc,
(6):
equilibrium
formation,
plex
by Eq.
(6)
kPA
kd-
Kc.
for
and
leads
to:
kd
Kc
Introducing
dependences
temperature
+
EA(kPA)
EA(kd)
AHC
(7)
This
is
for
known the
where
Hc
is the
value
not
complexation enthalpy.
DTBP-PA
The dimerization
of PA in
species.
enthalpy
n-heptane,
pHos0tulated
mol"',
reference value with the under-
50 kJ
in
illustrates that
serve as some
may
will
lower than zf
absolute
be
below
into
values,
that,
HD.
(7)
Hc
standing
Eq.
be
that
should
for the
EA(kd).
EA(kPA)
energy
(significantly)
Assuming
activation
radicals of acid-
difference
the
decomposition
i-butoxy
is smaller than
DTBP
EA(kPA)
leads to
an
EA(kHE),
complexed
appreciable
between
and
This is what is
from the
seen
^ ).
experimental
data in
4.
Fig.
in
on
the
of
Another
favour
PA
from
dimer
comes
complexation
hypothesis
argument
PA
of
concentration.
Toward
the
the
kobs
cPA
higher
dependence
is
relative
to the
concentration enhanced
are
concentration. As the PA
monomer
formation, kobs
complex
dimer
to
available for
will
continue
not
not
species
increase with
solution.
is observed for PA concentrations above
as
cPA
linearily
0.1
mol/kg
the
on
of acid-induced enhancement of
with
com-
the
The
kobs
origin
arguments
A
DTBP-PA
reaction which
via
seems to
be summarized:
a
cPA
proceeds
may
that
radicals
plex
decomposes by producing f-butoxy
provide
best
for all
kHE
The
explanation
experimental findings.
dependence
pressure
of
of
and
and
0.1
be
190°C)
kPA
(at
(at
cPA
cPA
may
mol/kg heptane)
which
have been deduced from
(8)
(9),
respectively,
represented by
Eqs.
the
:
data in Table
in(kHE/&-1)
2
2.12
104
-4.48
190°C)
(at
(8)
(p/bar)
·
10~7
+
104
-6.06 9.5
3.0
ln(¿PA
The
Cpa/s"1)
(p/bar)2
mol/kg)
(p/bar)
(at
0.1
190°C,
(9)
cPA=
500
the Arrhenius
of
and
&HE
kPA
(at
(at
cPA
cPA
(10)
bar)
dependence
temperature
0.1
is
and
expressions,
mol/kg heptane)
given by
Eqs.
(11),
respectively:
1
·
1014
500
(at
7.5
ytHE/s
18041/(77K)}
exp {-
bar)
(10)
(11)
·
102
6.3
kPA CpA/s->
{-5533/(77K)}
exp
500
0.1
bar,
(at
cPA
mol/kg).
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 9/22/15 1:29 PM

Michael
Buback and Lars Wittkowski
78
60
0.05 mol
/kg
/kg
/kg heptane
heptane
heptane
mol
0.10
0.15 mol
0
1X0
190
200
210
re
220
230
240
DTBP
overall
to
500 bar. The data are calculated from
of
in
7.
of acid-induced
Fig.
decomposition
decomposition
Percentage
as
a
function of
at
(10)
Eqs.
temperature
«-heptane
for
concentrations of monomeric
and
three
PA,
0.10, and 0.15
0.05,
(11)
cPA
mol/kg
heptane.
the
PA
mol
overall de-
concentration of
refers
to 0.1
(where
(8) —(11)
Using
Eqs.
per
the fraction of acid-induced
to
kg -heptane)
decomposition
In
be calculated.
this fraction
7
kPA cPA/k„hs,
(in
per
composition,
is
Fig.
may
function of reaction
for -induced
DTBP
of
as
a
cent)
plotted
temperature
bar and three PA
The
at
concentrations.
500
decompositions
percentage
decreases
acid-induced
increases with and
It should be noted
monomeric
less than 10
This observation is of interest with
cPA
decomposition
significantly
with
cussion,
the
dis-
to
that,
according
preceding
temperature.
PA. Above
cent to
to
refers
-induced
240°C,
CpA
decompo-
sition contributes
overall DTBP
by
decomposition.
per
respect
to ensure
acid
co-
to
ethene-methacrylic
order
the
in
at
of
which,
performed
polymerizations
homogeneity
polymerizing
such
of this
an
are
high
temperatures.
system,
is
the
the
also
to
DTBP
measure
acid
It
beyond
decompo-
scope
paper
sition
course
of
ethene-methacrylic
during
copolymerization
which is the
relevant
studied here. It is
behind the
acid
technically
n-heptane/pivalic
system
model
of
one
(meth)-
polymerizing
only
system
aspect
acid
initiator
with DTBP
the
that will be
addressed in this
for acid-induced
as
acrylic
briefly
section
discussion
final
The
:
preceding
provides
arguments
due
monomeric acid. If the acid is
the
to
a
monomer com-
decomposition being
with
is the
has
to
be asked
in
as
case
acid,
(meth)acrylic
question
be
may
ponent,
whether the
acid
which
polymerized
largely engaged
segments,
intramolecular
composition.
also
rise
acid-induced
to
de-
association,
give
peroxide
the model
has been
acid
from
To
an
DTBP/n-
polymeriza-
answer,
give
changed
The
system
to
(MAA).
heptane/PA
DTBP/n-heptane/methacrylic
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 9/22/15 1:29 PM

79
Acid-Induced
Peroxide in
Solution
of
Decomposition
Di-ierf-butyl
rc-Heptane
TB
DTBP
8.
Time evolution of
and
concentration
in
MAA
their characteristic
200 °C
at
and 500 bar. The
Fig.
IR
(represented by
a
absorbance) during
decomposition
-heptane
initial
MAA and DTBP concentrations were 0.160
the TB concentrations
and 0.150
mol/kg
mol/kg heptane
hep-
tane,
absorbances)
(integrated
respectively. Only
represented by
were
first-order fit for
to
a
via this
k„h,
&„b5.
open symbols
subjected
Back-extrapolation
t
value indicates that in the initial reaction
below
40
s,
where MAA monomer
period,
faster rate the
residuals in the lower
is
and
DTBP
text).
at
a
of
8
(see
Fig.
present,
see
decomposes
part
tion of this
and acid concentrations of 0.150 mol
has been studied
500 bar
initial
and
at
200°C,
system
peroxide
mol
and 0.160
DTBP/kg heptane
The
of the C=C
reaction is
mode
MAA/kg
polymerization
heptane, respectively.
spectroscopi-
monitored via the
of
cally
decay
stretching
absorption
1625 and 1650
MAA
between
and the
DTBP
cm"1,
monomer,
decompo-
sition is
concentrations
mined
There
via the TB band. The
deter-
as
measured,
before,
are
spectroscopically
in
8.
(in arbitrary
plotted
units)
absorption
of MAA. After about 40 seconds
Fig.
is
of acid
which demonstrates
a
monomer
rapid
decay
the
rate
MAA
(mono-
time
high polymerization
of
after
is consumed. The kinetic
the TB concentration
vs.
analysis
mer)
data has been restricted
the
to
MAA
period
complete
consumption,
s
is for
circles in
that
i>40
The first-order
rate coefficient
Fig. 8).
(open
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 9/22/15 1:29 PM

Michael Buback and
Wittkowski
Lars
80
is in full
2.1
can
found
IO"2
be
with
kHE
from this
0
s"\
seen
data, kobs
agreement
residuals in the lower
from the
in
1.
Table
As
at
kobs
300
interval,
within the time interval 40
the first-order
law
s
to
s
rate
provides
part,
2.1
coefficient for this
excellent fit of the data. The
rate
an
•
feobs
TB
into
concentration data back
10"2
be used
the
to
s~\
extrapolate
may
reaction
is
where MAA
below 40
monomer
the
s,
period,
present.
early
measured
concentrations from the
differences.
TB
ones
Subtraction of
back-extrapolated
8)
The residuals
are
circles in
zero.
clearly
significant
yields
(full
Fig.
in the
this time
deviation is such that
off
The
TB
The
interval,
to
during
increased
at an
presence
measured
at
rate as
of
i>40 s.
DTBP
kobs
MAA,
compared
decays
Monomeric MAA increases
8 is:
from
Fig.
important
message
does
MAA
within the
whereas
as
PA,
rate,
segments
decomposition
with
acid-induced
to
be ineffective
seem to
decomposition.
regard
polymer
in the ab-
remains
of acid
the total
amount
kahs
constant,
DTBP
Although
groups
same as
sence
for
in
monomeric MAA is the
of
decomposing
(pure)
-heptane.
ethene-MAA
copolymeriza-
Concerning high-pressure high-temperature
addition acid-
from
rather
8
is
in
to
the
tions initiated
induced DTBP
the influence of
utes to
detail
that,
DTBP,
Fig.
by
message
ineffective
at
high
decomposition being
temperature,
contrib-
monomeric MAA
will be
acid is further reduced
as
only
in
These
enhancement.
more
this
rate
investigated
aspects
Another inter-
ethene-MAA
from the
[19].
copolymerizations
directly during
of
relates
study
present
the
to
possibility
point originating
esting
tuning
free-radical
in
Studies
initiator
amounts
rate
acid.
adding
polymerization by
decomposition
these lines
of
are
small
underway
and tuned
along
carboxylic
initiated
DTBP
ethene
acid
for the
polymerization
the realm
within
by
high-pressure
acetic
[19].
by
Acknowledgements
This work
Mechanismen
AKZO
"Molekulare
with
of the SFB 357
carried
out
was
The
in
unimolekularer Reaktionen".
cooperation enjoyed
Dr. B.
NOBEL
with
Center
Mr. R.
(Research
Deventer),
particular
and
Gerritsen is
results and the
experimental
for the interaction
also
as
Dr. J.
discussion of
Fischer,
acknowledged,
gratefully
Meijer
in GC
is the
that
analysis by
(Ludwigs-
support
with BASF AG
for ethene
firm.
We
are
grateful
Dr. R.
with
and
in
Klimesch,
hafen),
being provided
particular
W.
des
indebted the
AG.
is
to
BASF
Verbandes
L.
Stipendien-Fonds
"Stiftung
by
for financial
the BMBF
Industrie V." and
der Chemischen
to
e.
support.
References
Academic
Or-
Press,
and
Science
The Elements
(1982).
A.
lando
1.
Rudin,
Polymer
Engineering,
of
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 9/22/15 1:29 PM

in
Solution
of
81
Peroxide
Acid-Induced
Di-teri-butyl
Decomposition
in
-Heptane
2. J.
J.
C.
Handbook, 3rd edition, eds.
and E. H.
Masson,
Wiley
Immergut,
Polymer
Bandrup
61.
John
and
York,
New
Sons,
(1989).
Chem.
.
193
M. Buback and C. Hintón,
3.
4.
5.
6.
Physik.
(1996)
D.
Ph.
S.
Klingbeil,
Thesis,
(1995).
Göttingen
M.
Buback and C.
Z.
Chem. 199
229.
(1997)
Hintón,
Physik.
P. D.
Am.
Leffler, J.
Chem. Soc. 72
3030.
Bartlett and J. E.
(1950)
3912.
T.
and
J.
Chem. Soc. 73
7. A.
A. F.
Am.
Ferris,
(1951)
Blomquist
84
3697.
8. H. Hart and R. A.
J. Am. Chem. Soc.
(1962)
Cipriani,
79
J. E. Leffler and C. C.
J. Am. Chem. Soc.
3068.
9.
10.
11.
Petropoulos,
(1957)
E. S.
and
Diploma
J.
Chem. 77
3524.
R. M. Van
(1968)
Scoy,
Huyser
Org.
A.
Thesis,
Drefahl,
(1984).
Göttingen
Z. Naturforsch. 34
Vol.
M. Buback and
a
1482.
John
12.
H. Lendle,
(1979)
in
ed. J.
K.
and New York
Sons,
13. T.
Free Radicals,
1,
Kochi,
Vol.
Koenig,
Wiley
113.
(1973)
ed. J. K.
John
Kochi,
and
14. T.
and H. Fischer, in Free
1,
Radicals,
Angew.
Koenig
Wiley
157.
and H.
New York
Sons,
(1973)
G.
C.
Makromol. Chem. 73
95.
(1978)
15.
16.
P.
Luft,
Walling
Seidl,
Phys.
Mehrling
2361.
J.
P. Waits,
Chem. 71
and
.
(1967)
H. Kiefer and
G.
Jr.,
J. Am. Chem. Soc. 89
6667.
(1970)
17.
T.
(1967)
Traylor,
92
R. N.
and
J.
R.
J. Am. Chem. Soc.
2440.
18.
Neuman,
Bussey,
19. L.
20. A.
21. M.
Ph. D. Thesis,
J. Chem. Ed.
Wittkowski,
(1998).
Göttingen
van
72
138.
(1995)
Herk,
and
Chem. Ber. 127
Naturforsch. 42
2241.
K.
C.
Ott,
Buback,
L. E
Tietze,
(1994)
Buback,
Gerke,
M.
a
187.
22. E. M. Borschel and
.
(1987)
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 9/22/15 1:29 PM