NC-based energetic materials - stability, decomposition,

and ageing

Dr. Manfred A. Bohn

Fraunhofer-Institut für Chemische Technologie (ICT)
Postfach 1240, D-76318 Pfinztal-Berghausen, Germany

Presentation on the Meeting

Nitrocellulose - Supply, Ageing and Characterization
April 24 to 25, 2007
AWE, Aldermaston, England
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 1
 Content
Abstract
Stability ’versus‘ ageing
Stability tests for NC and NC-based materials
Bergmann-Junk, Abel, Vacuum Stability, Mass Loss, Dutch Mass Loss Test
Aspects of NC decomposition
Thermolysis, hydrolysis, chain splitting, bond dissociation enthalpies
Accelerated ageing - objectives
Measurements on NC alone – Comparison of results of four stability tests
Modelling - benefits
Modelling of primary (added at production) stabilizer consumption
models ‘S: exponential + linear’, ‘S: nth order’, ‘S: extended’
Applications
Prediction of a real ageing over 16 years of a singlebase propellant used in automotive seat belt
restraint systems
Combined modelling of stabilizer consumption and the course of consecutive products
Forming reactivities of consecutive products of DPA
Modelling of molar mass decrease of NC
model 1: chain splitting by decomposition of chain element (= monomer unit)
model 2: chain splitting by bond breaking, number of chain elements is constant
Application, examples and meaning
molar mass decrease and shift of ductile-brittle transition temperature
molar mass decrease and decrease of mechanical strength
Ageing of New Type gun propellants
Literature
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 2
 Abstract

Nitrocellulose (NC) has itself re-established as an essential ingredient in recent gun

propellant formulations containing RDX and newer type of plasticizers. All the efforts to
replace NC by other polymeric binders have been successful in a rather small part only.
Surely NC decomposition is present also in the modern formulations and the ageing of
such propellants has to be considered carefully. To master the ageing of NC based
energetic materials knowledge of the basic decomposition behaviour of NC is necessary.
Data of bond dissociation energies and activation energies of NC decomposition will be
compiled and discussed. Experimental results and quantum mechanically obtained data are
used for this. The way or the severity of NC decomposition will be discussed and
formulated in a scheme. Kinetic models to describe NC decomposition are presented for
mass loss and molar mass degradation.

Stabilization is a main issue with NC-based materials. Stabilizer reactions and their
modelling are presented. With examples of ageing of gun propellants and rocket
propellants including new type of gun propellants containing RDX, the way of prediction
of the use time (life time) is shown. Several experimental methods as mass loss, stabilizer
consumption, molar mass degradation, heat generation can be used and their suitability
and the special demands for prediction purposes are demonstrated.

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 3
 Stability ‘versus’ Ageing

Stability tests use reduced time-temperature loads compared to ageing investigations,

typically only one temperature and one time period.
Stability tests determine the momentary state of the material.

In contrast the determination of ageing behaviour gives a prediction possibility based on

information obtained from an extended time-temperature behaviour of the material.

For both aspects molecular decomposition is the base, especially with chemical stability.
With ageing also physical phenomena are included as migration of surface treatment
agents (burning phlegmatizer for gun propellant grains)

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 4
 Stability tests
The NC-decomposition causes
● formation of permanent gases
● formation of NOx
● generation of reaction heat
● polymer chain degradation

The older tests are based mainly on the generation of NOx caused by a time – temperature load
Using the effects of NO2 – namely oxidizing and forming of HNO3 , many tests have been developed
and most of these are connected with famous names:
Abel, Bergmann-Junk, Hanson-Metz, Vieille, Methyl-Violet-Test
time to visible NOx (the well-known 65.5°C (=150 °F) surveillance test, introduced by the
US Navy, about 1905)
Self ignition temperature, Silvered Vessel-Test (Dewar-Test)

Then somewhat later mass loss and gas generation (vacuum stability test) were used

Newer methods use the same basic effects but the detection is much improved
● NOx-Chemiluminescence
● GPC (SEC)
● microcalorimeter (heat generation rate), ARC (adiabatic self heating)
● gas analysis by FTIR-EGA, GC, GC-MS
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 5
Bergmann-Junk-Test (Bergmann-Junk-Siebert-Test)
Developed from E. Bergmann and A. Junk in Germany, improved by Siebert

Criterion:
Substance Test Test
consumption of 0.01n NaOH
substance amount temp. time
to neutralize HNO3 formed by
[g] [°C] [h]
absorbed and oxidized NO2
Nitrocellulose 2 132 2 ≤ 12.5 ml/g
Sb gun propellant (propellant not
5 132 5 ≤ 10 to 12 ml/g *)
containing blasting oils)

Db gun propellant with blasting oil

5 115 16 ≤ 10 to 12 ml/g *)
content up to 10 mass-%
Db gun propellant with blasting oil
5 115 8 ≤ 10 to 12 ml/g *)
content greater than 10 mass-%

Db rocket propellant 5 115 16 ≤ 10 to 12 ml/g *)

*) limit value is often fixed for the individual propellant in the range 10 to 12ml per g of propellant
This test is not used with triple base propellants
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn 6
Bergmann-Junk-Test (Bergmann-Junk-Siebert-Test)

Dimensions in mm

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn 7
Abel-Test - German conditions and typical results

test Sample test

Substance temp. amount limit
[°C] [g] [min]
Nitroglycerine,
BTTN, TMETN, 80 2 ≥ 15
DEGN
Nitrocellulose 80 1.3 ≥ 10
TNT 80 1 ≥ 30

KJ starch paper, wetted by water -

glycerine (1:1) to half length. At the
interface colour change occurs caused by
NO2 in oxidizing iodide to iodine and this
forms a coloured product with starch.

Time taken at this point

of colour change

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn 8
 Vacuum stability test – German test conditions and criteria

Substance Test temp. Test Criterion for ‘stable’

substance
amount [g] [°C] time [h] ml/g at STP
Sb gun propellant (propellant
2.5 100 40 ≤ 1.2 to 2
not containing blasting oils)
High explosives as PETN, HMX,
RDX, explosive charges, HTPB- 2.5 100 40 ≤ 1.2 to 2
composite propellants
Db gun propellant with blasting
2.5 90 40 ≤ 1.2 to 2
oil
Db rocket propellant 2.5 90 40 ≤ 1.2 to 2
Recent tendency to unify conditions
for NC-based propellants 2.5 100 40 ≤2
Non-NC high explosives 2.5 100 40 ≤ 1.2
2 ml/g or 5ml/2.5g is the upper allowed limit. Between 1.2 and 2 ml/g the stability is considered as
borderline. Preferred are values ≤ 1.2ml/g or ≤ 3ml/2.5g

STP: standard temperature (=0°C) and standard pressure (=1 atm)

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn 9
 Mass Loss Tests
Mass loss (ML) at 90°C for GP and RP
In most cases one ages 2g (up to 5g) of propellant in tubes with loosely inserted ground
stoppers over 18 days
Sample can be used as delivered (if not to wet !, water content not higher than 1.5 mass-%)

Criterion: ML ≤ 3 % and no red-brown fumes (NOX) up to 18 days

initial mass loss can be subtracted, ML after one day as baseline

Applicable for all types of propellants (single, double, triple, semi-nitramine, new type,…)
Other test times - test temperatures (always
with 3 % limit)
Necessary equipment
For 50 years at For 15 years at
25°C 25°C • well temperature controlled oven
• glass tubes with ground stoppers
100°C 6 days 1.8 days
• analytical balance, well positioned
90°C 18 days 5.4 days
80°C 54 days 16.2 days
75°C 93.5 days 28.0 days
70°C 162 days 48.6 days
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn 10
 Dutch Mass Loss Test

Mass loss after 72 h, reduced by the mass loss after the first 8 h in open tube

4g of sample in tubes closed with ground stoppers

at 105°C: double and triple base GP and RP
at 110°C: single base GP
Criterion: ML (72h-8h) ≤ 2%, all GP and RP

In principle also usable with high explosives as RDX, HMX, PETN

Also used for semi-nitramine and new-type propellants
Pure substances can be tested at 110°C

For other formulations the test temperature depends on ingredients;

with energetic plasticizers 105°C is suitable
Other conditions and test criterion remain the same

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn 11
Assessment of ageing and stability with mass loss
Three types of GP doublebase GP L5460 shows after20d strong ML increase with visible NOx
triplebase GP Q 5560 shows smooth increase of ML without showing visible NOx
new type GP (see later) shows smooth increase of ML without showing NOx
9
ML [%]
8 mass loss at 90°C
GP, L5460, 120mm
7

6 GP Q 5560, 27mm MRCA

5
limit: ≤ 3%, 18 d
4

2
new type GP (contains RDX)
1
test time [d]
0
0 10 20 30 40 50 60 70 80 90
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn/ 12
 Comparison of mass loss: ball powder K6210 – 120mm GP L5460
Difference in stability can be revealed clearly with ML (18 days at 90°C corresponds to 50 years at 25°C isothermal)
db GP K6210 is stabilized with DPA (diphenylamine). In spite of not fulfilling the strong criterion it can be used
safely for certain time. Doblebase GP L5460 is stabilized with AK II (akardite II).
6
ML [%] mass loss test at 90°C

K6210 at 18 days ML <= 3%

4
L5460

db ball powder K6210

2

db GP L5460 (120mm)
1

time [d]
0
0 2 4 6 8 10 12 14 16 18 20 22 24

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn/ 13
 Comparison of heat generation rate: ball powder K6210 – 120mm GP L5460
Difference in stability can be revealed also clearly with heat generation rate (dQ/dt) and heat generation (Q)

90 70
dQ/dt [µW/g] db ball powder K 6210, Qex = 4059 J/g Q [J/g]
80 db 120mm GP L5460, Qex = 4720 J/g 63
80°C
70 56

K6210 49
60
dQ-K6210 42
50
dQ-L5460
35
40 Q-K6210
28
Q-L5460
30
21
L5460
20 14
10 7
time [d]
0 0
0 1 2 3 4 5 6 7 8 9 10 11 12
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn/ 14
 Aspects of NC decomposition

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 15
 What determines the NC decomposition

NC and NC-based products decompose chemically.

This is because of the relatively small bond dissociation enthalpy in the nitrate ester group.

average bond dissociation

chemical bond name
enthalpy at 25°C [kJ/mol]
CO − NO2 nitrate ester 165 to 170
C−C carbon - carbon 342
C−H carbon - hydrogen 416
C−O carbon - oxygen 343
N − O (in NO2) nitro group 305
C-NO2 (aromatic) carbon – nitro group 295

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn 16
Decomposition of NC

Assumption: by splitting-
off of one NO2 group
from any ester group the
corresponding glucose
ring will be destroyed.
This means the mean
molar masses of NC
decrease all the time
caused by the slow
intrinsic decomposition of
NC, which cannot be
stopped by stabilizers.

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 17
Weighing of reaction rates by the investigation temperature via activation energy

The weighing has an important effect: at higher temperatures the process with the higher activation
energy dominates the decomposition, at lower temperatures it is the one with smaller activation energy.

-14 ln(kA [1/s]) 90°C Arrhenius plot of the three

main decomposition types of NC
-18 temperature range 10°C to 120°C
60°C
-22 40°C
-CO-NO2 -COH + HNO2
-26 ester thermolysis, 163 kJ/mol
ester hydrolysis, 100 kJ/mol
-30
chain hydrolysis, 56 kJ/mol

-34 ET+EH
-CO-NO2 -CO• + •NO2
ET+EH+CH 1/T [1/K]
-38
0.0024 0.0028 0.0032 0.0036

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯−−−−⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn 18
 Stabilizer consumption: activation energy as function of temp.

decrease of stabilizer
DPA in single base GP
A5020 (20 mm machine
gun, older lot) /1/
change in Ea at about
65°C
higher temp. range has
more thermolytical
ester group
decomposition
lower temp. range has
more hydrolytical ester
group decomposition

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
Stabcon 2005 Bohn / 19
 Molar mass decrease of NC: activation energy as function of temp.

decrease of molar mass of

NC in single base GP A5020
(20mm machine gun, older
lot) /1/
change in Ea at about 65°C
higher temp. range has
more thermolytical ester
group decomposition with
consecutive chain scission
lower temp. range has
more hydrolytical NC chain
splitting

Remark: the lot of A5020

investigated was already an
older one, means products
with effect on hydrolysis
present

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
Stabcon 2005 Bohn / 20
 Temperature range dependence of Ea for PVN (polyvinylnitrate)
Also with other nitric acid ester compounds a temperature dependence of Ea is observable /2/

EatyQ [kJ/mol] PVN-M (polyvinylnitrate)

125
EatyQ as function of conversion (reached Q)
70°C to 110°C and temp. range sections

120 90-110°C

115

70-110°C
110

70-110
70-90
70-90°C
105 90-110
Obtained from heat generation measurements
Q [J/g]
100
0 25 50 75 100 125 150 175 200 225 250 275 300 325 350 375 400
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 21
 Nitrate esters NE decompose intrinsically, means by themselves, Decomposition scheme
because of the small bond energy in the nitrate ester group, of nitric acid esters NE
about 170 kJ/mol for thermolytical split-off of NO2; Further more
one has formation of water and acids, and additional products as
gases and organics.
The decomposition
products NO2 and acids
k act as catalysts P in
NE ⎯⎯NE
⎯→ P + R-NE + ∆HRID intrinsic decomposition
decomposing NE; a further
NE + P
k
⎯⎯auto
⎯⎯→ 2 P + R-NE + ∆HRAD autocatalytic decomp. decomposition reaction
channel is opened – the
k so named autocatalytic
S+P ⎯⎯SP
⎯→ P-S + ∆HRS stabilizer reaction
decomposition.

The stabilizer S binds the NO2 and nitric acid and makes
the autocatalysts P in-effective. No autocatalytic
decomposition appears during active phase of stabilization.
– But the intrinsic decomposition goes on and causes mass
loss, energy loss, chain splitting of NC, gas and heat
generation.
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn 22
 Experimental bond dissociation enthalpies
kcal/mol kJ/mol
20 83.7
Nitrate RO - NO2 kJ/mol Nitrite RO - NO kJ/mol
22 92.0
iso-Propyl nitrate 172.0 ± 4.2 iso-Propyl nitrite 171.5 ± 7.5
24 100.4
n-Propyl nitrate 177.0 ± 4.2 n-Propyl nitrite 167.8 ± 7.5 26 108.8
Methyl nitrate 172.4 ± 4.2 Methyl nitrite 174.9 ± 3.8 28 117.2
Alkyl nitrate 170.3 ± 2.1 Alkyl nitrite 170.7 ± 4.2 29 121.3
30 125.5
From CRC Handbooks 32 133.9
34 142.3

Please be aware of the difference in the terms 36 150.6

Relation kcal to kJ 38 159.0
Bond dissociation enthalpy
refers to a specified bond which is split 1 kcal = 4.184 kJ 40 167.4
Bond enthalpy 42 175.7
refers to the energy obtained when the
molecule is split into the elements 44 184.1
see for example with hydrazine in /3/ 46 192.5
50 209.2
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 23
NC-simulate and NC-
radicals to calculate
the bond dissociation
enthalpies of the
split-off of NO2
radical from the three
glucose ring positions
2, 3 and 6 NC-simulate Pos. 2

Pos. 6

NO2 radical
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
Pos. 3
2007 Bohn / 24
 PVN-simulate and PVN-radical to calculate the bond dissociation enthalpy of the
split-off of NO2 radical

PVN (polyvinylnitrate) simulate Split-off of NO2 from middle position

PVN was seen as a replacement for NC in former time.

Today it is a speciality energetic polymer, used as additive in TNT HE charges.

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 25
Bond dissociation enthalpies at 298.15 K in different molecular environment
Substance [kJ/mol]
Calculation method:
ME-NC-2NO2, (BDE in the nitrate group at position 2) 1.2 160.8
DFT, BLYP, GGA all ME-NC-3NO2, (BDE in the nitrate group at position 3) 1.0 133.0
electron treatment
ME-NC-6NO2, (BDE in the nitrate group at position 6) 1.3 174.4
Calculated values correc-
ted by normalizing to PVN simulate (polyvinylnitrate) 170.8
experimental average
value of alkyl nitrate of Nitroglycerine, split-off of NO2 from middle position 171.2
170.3 kJ/mol; error in calc. Nitroglycerine, split-off of NO2 from side position 174.5
values ± 8 kJ/mol
PETN 170.0

DFT: (electron) density Methyl nitrate 185.0

functional theory
Iso-Propyl nitrate 170.3
B: Becke gradient correc-
ted exchange potential n-Propyl nitrate 179.9
LYP: Lee, Yang, Parr
gradient corrected elec- Iso-propyl peroxonitrate: i-PropOO-NO2; split-off of NO2 radical 118.6
tron correlation potential Iso-propyl peroxonitrate: i-PropO-ONO2; split-off of NO3 radical 92.5
GGA: generalized gradient
approximation Iso-propyl peroxonitrite: i-PropO-ONO; split-off of NO2 radical 64.5
Further conditions see /4/ Iso-propyl peroxonitrite: i-PropOO-NO; split-off of NO radical 87.2
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 26
Relation between nitrate, peroxo nitrite and peroxo nitrate
CH3

170 kJ/mol CH2 O NO2

. NO 2
CH3

CH3 65 kJ/mol CH3 87 kJ/mol CH3

.NO2 .NO
CH2 O. CH2 O O NO CH2 O O.
N O
CH3 + .O CH3 CH3
. NO2
.NO3
CH3 119 kJ/mol
92 kJ/mol
CH2 O O NO2
CH3

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 27
 Accelerated ageing

ageing means generally: property change with time and temperature

Objectives
● To reach equivalent ageing in much shorter times than at target temperature,
for example at 25°C
● To find out the ageing process(es)
● To determine activation energies for the ageing process(es) or other parameters
to quantify the temperature dependence of the ageing
● To use the temperature dependence describing parameters to predict the ageing
at any time-temperature load
● To simulate real ageing using time-temperature profiles

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn 28
Arrhenius Plot according to STANAG 4582 (so-named heat flow (HFC) STANAG)
13
ln(kQ [µW/g]) Arrhenius plot of reaction rate constant kQ(T)
11 according to STANAG 4582
9 T S=60°C
7 120 kJ/mol 80 kJ/mol
135°C
5 the composed Arrhenius line is used

3
kQ2 with EaQ2 = 80 kJ/mol
1 ZQ2 = 3.4099 E+13 µW/g

-1
ln(kq1 [µW/g]) 15°C
-3 ln(kq2 [µW/g])
ln(kq [µW/g]) kQ1 with EQ1 = 120 kJ/mol
-5 ZQ1 = 6.3719 E+19 µW/g
Ea divide
1/T [1/K]
-7
0.0022 0.0024 0.0026 0.0028 0.003 0.0032 0.0034 0.0036
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn 29
HFC-STANAG 4582 (Heat generation rate (HGR)-STANAG)

Main points Acceleration

(dQ/dt)L
TT [°C] tT [d] with regard to
[µW/g]
Fixing of test times tT and upper limits for 10 years at 25°C
HGR (heat flow) (dQ/dt)L at a series of test 60 123 9.8 30
temperatures TT
65 64.9 18.5 56
To fulfil the HGR limits during test time 70 34.8 34.5 105
at test temperature gives a safe time 71 30.8 39.0 119
interval against autoignition of 10 years
at isothermally 25°C 75 19.0 63.1 192
80 10.6 114 345
The test times and HGR limits are valid 89 3.83 314 954
only for a decomposition kinetics as it
90 3.43 350 1065
is shown by NC-based propellants

HGR limits have been determined with a finite element

programme for the solution of the heat balance
equation for the autoignition of a big caliber (230 mm in
diameter) cartridge
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn 30
 Arrhenius relation between ‘test (tt,Tt)‘ and ‘in-service (tE,TE)‘

With one activation energy valid in the considered temperature range TE to Tt

⎛ Ea ⎞
ki = Z ⋅ exp⎜⎜ − ⎟⎟
⎝ RTi⎠

k2 ⎛ Ea ⎛ 1 1 ⎞ ⎞ 1
Reaction rate constant
= exp⎜⎜ − ⋅ ⎜⎜ − ⎟⎟ ⎟⎟ t∝ reciprocal proportional
k1 ⎝ R ⎝ T2 T1 ⎠ ⎠ k to time to reach a state
or conversion

t1 ⎛ Ea ⎛ 1 1 ⎞ ⎞
= exp⎜⎜ − ⋅ ⎜⎜ − ⎟⎟ ⎟⎟ By forming the ratio the time-
t2 ⎝ R ⎝ T2 T1 ⎠ ⎠ temperature scaling is achieved

⎛ Ea ⎛ 1 1 ⎞ ⎞
tE = t t ⋅ exp⎜⎜ + ⋅ ⎜⎜ − ⎟⎟ ⎟⎟ Caution
⎝ R ⎝ TE Tt ⎠ ⎠ this procedure may be
simplifying to strong

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 tBohn 31
 Extrapolation of test conditions to in-service conditions with factor F (works in
both directions: ac-celeration and de-celeration)
Factor 3 for 10°C temperature change corresponds to activation energy range of 80 to 120 kJ/mol
in the temperature range 20°C to 95°C
10
F [-] factor F= k(T) / k(T-10K) Ea [kJ/mol]
9 for dT = 10°C 160
140
8 120
100
7 80

2
T [°C]
1
0 10 20 30 40 50 60 70 80 90 100 110 120
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn 32
Relation between test time - test temperature and in-service time - temperature

For chemically caused ageing processes the following formula has proven by experience as
suitable (Ea about 80 to 120 kJ/mol, temperature range 20°C to 95°C)

tE [a] = t T [d] ⋅ F( TT − TE ) / ∆TF / 365 .25d

tE time in years at temperature TE

tT test time in days at test temperature TT
F reaction rate change factor per 10°C temperature change
De-celeration with temp. decrease / ac-celeration with temp. Increase
TT test temperature in °C
TE in-service temperature in °C
∆TF temperature interval for actual value of F, here ∆TF is always 10°C

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn 33
 Comparison of the STANAG extrapolation with the factor 3 extrapolation
100000
time tt [d] time-temperature relations
normalized to tE = 10 years at TE = 25°C
10000 TS=60°C

80 kJ/mol 120 kJ/mol

curves with Ea1 and Ea2,
1000 10 years at 25°C
intersecting at 60°C

100
with Ea2 = 80 kJ/mol
factor 3
10 t [d] Ea1
extrapolation
t [d] Ea2
t [d] Ea1+Ea2
1
tt [d] F-3
25°C reference with Ea1 = 120 kJ/mol
Ea divide shifted up to intersection at 60°C
0.1

T t [°C]
0.01
0 10 20 30 40 50 60 70 80 90 100 110 120
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn 34
 Overview of some used test times – test temperatures extrapolated to 25°C

100000
time tt [d] 80 a
TS = 60°C time-temperature relations
40 a
20 a tE =10 years at TE = 25°C
10000 80 kJ/mol 120 kJ/mol
extensions to other in-service times
Ea1 = 120 kJ/mol, Ea2 = 80 kJ/mol
10 a
1000 5a STANAGs
Example how to read: This point gives 5 years at 25°C
TE = 25°C
100 ML, 90°C, 18d

1 x te [d] = 10 a 0.5 x te [d] DMLT, 110°C, 105°C

10 2 x te [d] 4 x te [d]
8 x te [d] DMLT-110°C, 72h
DMLT-105°C, 72h VST-100°C, 48h Dutch HGR-Test Bergmann-Junk
1 VST 100°C, 40h VST 90°C, 40h
ML 90°C, 18d DHGT 85°C, 7d
VST, 100°C, 90°C
Abel, 80°C, 1 h BJ, NC, 132°C, 2h
0.1 BJS, GP, 115°C, 8h STANAG 4541, 60°C, 60d
STANAG 4117, 65°C, 120d 25°C reference Abel, 80°C
Ea divide Tt [°C]
0.01
0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn 35
 Implicit limit in the HFC STANAG
Limit for the allowed energy loss of a gun (GP) or rocket (RP) propellant: ≤ 3%
For 10 years at 25°C

TT (dQ/dt)L
tT [d] QL [J/g] at 80°C, after 10.6 d Limit is
[°C] [µW/g] QEX(gw)
3% of
60 123 9.8 104.1 HGR Q % of [ QEX(lw) ] QEX(gw)
[µW/g] [J/g] QEX(gw) [J/g]
65 64.9 18.5 103.7 [J/g]
L5460 4238
70 34.8 34.5 103.7 12.8 16.1 0.38 127
db, 120mm [ 4610 ]
71 30.8 39.0 103.8 K6210 3730
68.2 57.1 1.53 112
75 19.0 63.1 103.6 db ball, 9mm [ 4040 ]
80 10.6 114 104.4
89 3.83 314 103.9 QEX(gw) value with reaction water gaseous
90 3.43 350 103.7 QEX(lw) value with reaction water liquid

To extend the predicted time from

10 to 20 years implies to use the Energy loss limit is for the actual use of the propellant
double test times but also only half more important than the danger of its autoignition
the values for (dQ/dt)L
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn 36
 Measurements on NC alone – Comparison of stability tests

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 37
 NC lot 1: HGR and HG at 90°C, measured under air and under argon
Measured at 90°C with TAMTM in closed steel ampoules with glass inserts, sample amounts about 0.2g
100000 4000
dQ/dt [µW/g]
NC-1
90000 Q [J/g] 3600
90°C
80000 dQ-NC53-Ar 3200
dQ-NC53-Ar-2
70000 dQ-NC53-Lu 2800
zero in argon
60000 2400
Q -NC53-Ar
50000 Q -NC53-Ar-2 2000
Q -NC53-Lu
40000 in air 1600

30000 1200
N=12.5 mass-%
QEX-wliqu = 3931 J/g
20000 800
QEX-wgas = 3619 J/g
10000 400

0 0
time [d]
-10000 -400
-2 1 4 7 10 13 16 19 22 25

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 38
 NC lot 1: HGR and HG at 90°C, measured under air and under argon
Measured at 90°C with TAMTM in closed steel ampoules with glass inserts, sample amounts about 0.2g

3000 100
dQ/dt [µW/g] NC-1 Q [J/g]
2700 90°C 90
comparison between
2400 air - argon 80

2100 70
dQ-NC53-Ar
1800 dQ-NC53-Lu 60
dQ-NC53-Ar-2
1500 zero 50
Q -NC53-Ar
1200 40
Q -NC53-Lu
900 Q -NC53-Ar-2 30

600 20

300 10

0 0
time [d]
-300 -10
-1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 39
 NC lot 2: HGR and HG at 90°C, measured under air and under argon
Measured at 90°C with TAMTM in closed steel ampoules with glass inserts, sample amounts about 0.2g
1500 15
dQ/dt [µW/g] NC-2 Q [J/g]
1350 90°C 13.5

1200 under argon 12

132°C, 2 h correspond to
1050 90°C, 3.75 d with F = 2.5 10.5
90°C, 8.07 d with F = 3.0 under air
900 9

750 7.5

600 6

450 4.5

300 3

150 1.5

0 0
time [d]
-150 -1.5
-0.1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2 1.3
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 40
 Heat generation rates of dry and unstabilized NC samples – region 0 to 2 days
Measured at 90°C with TAMTM in closed steel ampoules with glass inserts in air, sample amounts about 0.2g

120000
dQ/dt [µW/g] NC, 90°C
110000
comparison of NC-lots in air
100000

90000 dQ-NC-1
dQ-NC-2
80000
dQ-NC-3
70000 dQ-NC-4
dQ-NC-5
60000
dQ-NC-6
50000 dQ-NC-7
zero
40000

30000

20000

10000

0
time [d]
-10000
-0.2 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 41
 Heat generation rates of seven dry and unstabilized NC samples
Measured at 90°C with TAMTM in closed steel ampoules with glass inserts in air, sample amounts about 0.2g
120000
dQ/dt [µW/g]
110000 NC, 90°C
comparison of NC-lots in air
100000

90000 dQ-NC-1

80000 dQ-NC-2
dQ-NC-3
70000
dQ-NC-4
60000 dQ-NC-5

50000 dQ-NC-6
dQ-NC-7
40000
zero
30000

20000

10000

0
time [d]
-10000
-1 0 1 2 3 4 5 6 7 8 9 10 11 12
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 42
 Mass loss of dry and unstabilized NC lot 1 in the ‘linear range’ of pre-
autocatalysis (means a reaction of zero order)
1.6
ML [%] NC-1
1.4 mass loss in stoppered vials
90°C, air
P1 [%] region of 'pre-autocatalysis'
1.2 P2 [%]
M [%]
1

visible NOx appeared;

0.8 this is named as start
of 'autocatalysis'
0.6
ML [%] = 0.165 [%/d] * t [d] + 0.5724 [%]
2
R = 0.904
0.4

0.2 The ‘linear range’ of pre-autocatalysis means a reaction of zero order in ML

time [d]
0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 43
 Mass loss of dry and unstabilized NC lot 1 at 90°C up to autocatalysis
50
ML [%] NC-1
45 mass loss in stoppered vials
90°C, air
40 region up to 'autocatalysis'
P1 [%]
35
P2 [%]
M [%]
30

25

20 s is , )
y
al ning
visible NOx a t
15 appeared t oc egin
u
o f a at b
10 n ge ially
Ra spec
Pre-autocatalysis range (e
5
time [d]
0
0 1 2 3 4 5 6 7 8 9 10 11

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 44
 Comparison of stability data of seven NC lots obtained with four methods

10.0 0.9
time to NOx [d] correlation of NC stability data kml [%/d]
time to max HGR [d]
9.0 0.8
BJ-value [ml/g 0.01 n BJ value [ml/g 0.01n NaOH]
NaOH] t(90°C) to NOx [d]
8.0 t(90°C) to max HGR [d] 0.7
kml [%/d]
7.0
0.6
6.0
Bergmann-Junk 0.5
5.0
ML reaction rate constant 0.4
4.0
0.3
3.0
time to visible NOx
0.2
2.0
time to max. HGR
1.0 0.1
NC NC
sample
lot
0.0 0
1 2 5 6 3 7 4
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 45
 Photographs of NC residues after ageing at 90°C in mass loss vials

Brown to black
residues after the
‘cool‘ burning-out
of the NC samples.
Sample amount
about 0.6g to 0.8g
No ignition
occurred in any
case !

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 46
 Photographs of NC residues in glass inserts after measurements in TAM at 90°C

Brown to black
residues after the
‘cool‘ burning-out
of the NC samples.
Sample amount
about 0.2g
No ignition
occurred in any
case !

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 47
 Scheme of NC decomposition considering temperature-time exposure see /5/

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 48
 Mitigation of ageing

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 49
 Chemical formulas of the most used stabilizers in NC-based propellants

H CH3
N HN H CH3
N
N O=C
C2H5
DPA O=C N
C2H5
N
NO2
H NO2
N
EC, Centralite I Ak II, Akardite II MNA, pNMA

2NDPA, C 2H 6
2-NO2-DPA HN
N OH
CH3 O=C
NO O=C
CH3 N
N N
OH

N-NO-DPA MC, Centralite II Ak III, Akardite III Resorcinol

N-nitroso-DPA
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn 50
STANAGs for assessment of chemical stability / ageing using stabilizer consumption

STANAG 4117 STANAG 4541 STANAG 4542 STANAG 4527

SB with DPA
DB with DPA only DB with 2-NO2-DPA all NC-based propellants
SB with DPA and EC
(≤ 15% NG) (≥ 15% NG) multi-temperature test
DB with EC

60 days at 65.5°C
60 days at 60°C 35 days at 60°C 40 – 90°C
or 120 days at 65.5°C

120 d: DPAeff ≥ 0.3 mass-% • Kinetic modelling with

abs. decr. ≤ 0.5 mass% DPAeff ≥ 0.5 mass-%; 2-NO2-DPA ≥ 0.2 mass-% 0th or 1st order
• Arrhenius parameter
120 d: EC ≥ 0.3 mass-% rel. decrease ≤ 50% abs. decr. ≤ 0.5 mass-% • definition of evaluation
rel. decrease ≤ 50% is incomplete

• Existing standards use different test conditions, the reason is their historical development
it seems today not justified to define different test criteria as it was done in former times
each NC-based propellant should be assessed in the same way.
But: there exist different ranges of stability with different type of propellants !!
• Not all propellants are covered by these STANAGs named above
• Kinetic analysis and calculation of chemically determined safe time period is defined not precise enough
• Therefore: a unification was developed with AOP 48 Edition 2 which will replace all these STANAGs
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn 51
Standards to assess chemically determined use time with stabilizer consumption

STANAG 4117 STANAG 4541 STANAG 4542 STANAG 4527

SB with DPA DB with DPA DB with 2-NO2-DPA all NC-based propell.
SB with DPA and EC multi-temperature test
(≤ 15% NG) (≥ 15% NG)
DB with EC

STANAG 4620 AOP 48 Edition 1

Ö "Empty document" • Compilation of the 4 procedures from the 4 earlier STANAGs
Ö Refers to AOP 48 • No changes of test procedures
Ö In ratification • In ratification

AOP 48 encourages the use AOP 48 Edition 2

of kinetic description of • Same test procedure for all typical stabilizers and GP types
stabilizer consumption • Test conditions harmonized to other STANAGs (4582-HFC,
with a model of nth order, 4147-Ed.3-Compatibility)
see later. dS( t ) • In preparation, meanwhile near to ratification (May, 2007 )
= −k | ⋅ S( t )n
dt
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn 52
 Modelling of stabilizer consumption

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 53
 Benefits of modelling the ageing
See /6/

• minimize the risk of a mis-evaluation

• increase the reliability of the prediction of the ageing effects by a better description of
the measured data and by an inherently correct extrapolation along the time axis
• recognize a faulty measuring arrangement if the measured data are not conform to the
model
• identify a faulty production of a GP or RP
• support the find out of ageing reactions or ageing processes
• support the find out of measures against the effects of ageing processes
• support the planning and the realization of the acquisition of experimental data and of the
procurement of substance data to assess ageing and safety
• support the decision about the employability of energetic components and of formulations
• full use of the service time capacity of a GP or RP

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn 54
 Kinetic description of stabilizer consumption - principles
Facts on stabilizer consumption, presented in way of reaction scheme /6,14/

Reaction scheme
k
NE ⎯⎯NE
⎯→ P + R-NE intrinsic decomposition
k
NE + P ⎯⎯auto
⎯⎯→ 2 P + R-NE autocatalytic decomposition
k
S+P ⎯⎯SP
⎯→ P-S stabilizing reaction

a good stabilizer suppresses the autocatalytic decomposition of nitrate ester NE

by catching product P

A correct description of stabilizer depletion needs at least two reaction rate

constants:
kNE for NE decomposition
kSP for the stabilizer reaction: P reacts with S

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 55
Kinetic description of consumption of the primary stabilizer S – reaction scheme I
/7,11, 14/
Primary stabilizer is the
stabilizer added at production
reaction scheme RS I
k
NE ⎯⎯NE
⎯→ P + R-NE intrinsic decomp.
k
NE + P ⎯⎯auto
⎯⎯→ 2 P + R-NE autocat. decomp.
k
S+P ⎯⎯SP
⎯→ P-S stabilizer reaction

dNE( t )
= −kNE ⋅ NE( t ) − kauto ⋅ NE( t ) ⋅ P( t )
dt
dP( t ) System of DEq for the above
= +kNE ⋅ NE( t ) + kauto ⋅ NE( t ) ⋅ P( t ) − kSP ⋅ P( t ) ⋅ S( t ) reaction scheme
dt
dS( t )
= −kSP ⋅ P( t ) ⋅ S( t )
dt

This DEq system is very probably not solvable analytically in closed form.
It is integrated and iterated numerically to determine the values of the three rate
constants.
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn 56
 Kinetic description of consumption of the primary stabilizer S – RS II
Approximation:
stabilizer S suppresses the autocatalytic decomposition and the corresponding reaction can
be deployed

reaction scheme RS II Primary stabilizer is the

stabilizer added at production
k
NE ⎯⎯NE
⎯→ P + R-NE intrinsic decomp.
k
S+P ⎯⎯SP
⎯→ P-S stabilizer reaction

Corresponding differential equation system

dNE( t )
(1) = −kNE ⋅ NE( t )
dt
dP( t )
(2) = kNE ⋅ NE( t ) − kSP ⋅ S( t ) ⋅ P( t )
dt
dS( t )
(3) = −kSP ⋅ S( t ) ⋅ P( t )
dt
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 57
 Kinetic description of consumption of the primary stabilizer S – RS II
Simplified solution of DEq system of reaction scheme RS
dNE( t )
(1) = −kNE ⋅ NE( t )
dt
dP( t )
(2) = kNE ⋅ NE( t ) − kSP ⋅ S( t ) ⋅ P( t )
dt
dS( t )
(3) = −kSP ⋅ S( t ) ⋅ P( t )
dt

Assumption: P(t) is constant, P(t) = Pc ; this holds at the beginning of stabilizer consumption

dS( t )
= −k′SP ⋅ S( t ) with k′SP = kSP ⋅ Pc
dt

Stabilizer is consumed according to reaction of first order

S( t ) = S(0) ⋅ exp( −k′SP ⋅ t ) means exponential decrease of stabilizer consumption

Further simplification:
for short times the first order reaction can be approximated by zero order reaction
S( t ) = S(0) ⋅ (1 − k′SP ⋅ t ) Linear decrease of stabilizer concentration, but only at the very beginning

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 58
 Kinetic description of consumption of the primary stabilizer S – RS II

Way of solution of not simplifying the DEq system, means with P(t) not assumed as
constant – it results in just one differential equation for the stabilizer decrease /7/
eq.(1) has the solution :
dNE( t ) NE( t ) = NE(0) ⋅ exp( −kNE ⋅ t )
(1) = −kNE ⋅ NE( t )
dt
dP( t ) eq.(2) and eq.(3) give :
(2) = kNE ⋅ NE( t ) − kSP ⋅ S( t ) ⋅ P( t )
dt dP( t ) dS( t ) −k ⋅t
dS( t ) − = kNE ⋅ NE( t ) = kNE ⋅ NE(0) ⋅ e NE
(3) = −kSP ⋅ S( t ) ⋅ P( t ) dt dt
dt
with the solution :

− kNE ⋅ t
P( t ) − S( t ) = C − NE(0) ⋅ e
the expression for P(t) from the right is
inserted in eq.(3) of DEq-system. This and C is the integration constant :
results in an expression containing C = NE(0) + P(0) − S(0)
only S(t)

= −kSP ⋅ S( t ) ⋅ ⎛⎜ S( t ) + C − NE(0 ) ⋅ e− kNE ⋅ t ⎞⎟

dS( t ) NE(0): initial conc. of nitrate ester
⎝ ⎠ P(0): initial conc. of product P
dt
S(0): initial conc. of stabilizer S
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 59
 Kinetic description of consumption of the primary stabilizer S – RS II

Way of solution of the single differential equation for stabilizer decrease

Introducing reasonable approximations

• C ≅ NE(0 )
C is dominated by NE(0)
• NE(0) ⋅ exp( −kNE ⋅ t ) ≅ NE(0 ) ⋅ (1 − kNE ⋅ t ) the exponential can be well approximated by
Taylor series up to second term

dS( t )
≈ −kSP ⋅ S( t ) ⋅ (S( t ) + NE(0 ) ⋅ kNE ⋅ t ) Resulting rate equation for S
dt

S( t ) =
(
exp − NE(0 ) ⋅ kNE ⋅ kSP ⋅ t2 )
1 π kSP ⎛ NE(0 ) ⋅ kNE ⋅ kSP ⎞ Integrated rate equation
+ ⋅ ⋅ erf ⎜⎜ ⋅ t ⎟⎟
S(0 ) 2 NE(0 ) ⋅ kNE ⎝ 2 ⎠
This is a kinetically correct description of stabilizer depletion, because two reaction
steps are involved - decomposition of NC and stabilizer reaction - so two reaction rate
constants must appear in the description.

2 ⎛⎜ z3 z5 z7 z9 ⎞
erf(z ) = ⋅ ⎜z − + − + − .. + .. ⎟⎟ for z < ∞ ; erf(0) = 0
π ⎝ 1!⋅3 2!⋅5 3!⋅7 4!⋅9 ⎠
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn 60
 Modelling with RS II, model ’S: extended‘

Good description of
the data of DPA
consumption in a
single base
propellant with
model ‘S: extended‘

————————————————————————————————————————————————
2007 Bohn / 61
 Model ‘S: exponential + linear’ part 1 of 2 see /8/

Reactions of first order or zero order alone cannot give satisfying descriptions of stabilizer
decrease. To achieve at a better description the combination was tried.

Combination of the reactions of first and zero order

⎛ dS( t , T ) ⎞
⎜ ⎟ = − k1 ( T ) ⋅ S( t , T ) − k0 ( T )
⎝ dt ⎠ T

(
k1 ( T ) = Z1 ⋅ exp − Ea1 / RT)
k0 ( T ) = Z0 ⋅ exp(− Ea0 / RT )
Arrhenius behaviour

Integrated rate equation

⎛
S( t , T )=⎜ S(0) +
⎜
k0 ( T ) ⎞⎟
k1 ( T ) ⎟⎠
( ) k (T)
⋅ exp − k1 ( T ) ⋅ t − 0
k1 ( T )
⎝

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn/ 62
 Model ‘S: exponential + linear’ part 2 of 2
Calculation of service times tyS(T)
Definition of degree of stabilizer consumption
S( t , T )
yS = therewith the times tyS(T) to reach yS can be calculated
S(0)

⎛ ⎞
⎜ k0 ( T ) ⎟
⎜ 1+ 1 ⎟
1 ⎜ S(0) ⋅ k1( T ) ⎟
ty S ( T ) = ⋅ ln times to reach yS these are the looked for usetimes
k1 ( T ) ⎜ k0 ( T ) ⎟
⎜ yS + ⎟
⎜ S( 0 ) ⋅ k ( T ) ⎟
⎝ 1 ⎠

⎛ S(0) ⋅ k ( T ) ⎞
⋅ ln⎜ + 1⎟
1
t0 ( T ) = 1 times to reach yS = 0
⎜
k1 ( T ) ⎝ k0 ( T ) ⎟
⎠

the model has a good extrapolation ability

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn/ 63
 Stabilizer consumption: decrease of 2-nitro-DPA in a RP

description with
model 'S: expon. +
linear' in comparison
with description
according to first
order reaction

Data points marked

with arrows must have
been omitted with first
order reaction (model
‘S: exponential’) to
achieve acceptable
description at the
beginning, but not
with the model
'S: exponential + linear'
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 64
 Model ‘S: exponential + linear’ for stabilizer consumption - extrapolation ability

Extrapolation ability of
model ’S: exponential +
linear‘
Black line: all points
have been used
Red line: points up to
red arrow have been
used
Blue line: points up to
blue arrow have been
used
Green line: points up to
green arrow have been
used
The resulting scattering
is within narrow range

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 65
Models to describe stabilizer consumption

name of model
Eq. kinetic rate equation kinetic type
suitability, how to handle
dS( t ) first + zero order, ‘S: exponential + linear’ /8/
1 = −k 1 ⋅ S( t ) − k 0
dt two reactions very suitable, easy to handle
dS( t ) first order, ‘S: exponential’
2 = −k 1 ⋅ S( t )
dt one reaction not suitable, easy to handle
dS( t ) zero order, ‘S: linear’
3 = −k 0
dt one reaction conditioned suitable, easy to handle
dS( t ) nth order, ‘S: nth order’ (AOP 48 model)
4 = −k | ⋅ S( t )n
dt one reaction well suitable, easy to handle

dS( t ) two reactions, ‘S: extended’ /7, 14/

5 = −k SP ⋅ S( t ) ⋅ (S( t ) + NE(0) ⋅ kNE ⋅ t ) NE decomp. and suitable, to handle with advanced fit
dt
consump. of S codes (OriginTM, Origin Lab Corp.)

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 66
 Models to describe stabilizer consumption – integrated expressions

name of model
Eq. equation time t0 to yS = S(t)/S(0) = 0
suitability, how to handle
⎛ k (T) ⎞ k (T)
S( t , T ) = ⎜⎜ S(0) + 0 ⎟⎟ ⋅ exp(− k 1( T ) ⋅ t ) − 0 t 0 (T) =
1 ⎛ S(0) ⋅ k 1( T ) ⎞
⋅ ln⎜⎜ + 1⎟⎟ ‘S: exponential + linear’ /8/
1 ⎝ k 1( T ) ⎠ k 1( T ) k 1( T ) ⎝ k 0 ( T ) ⎠ suitable, easy to handle

2 S( t , T ) = S(0) ⋅ exp(− k1( T ) ⋅ t ) ‘S: exponential’

t0(T) → ∞ not suitable, easy to handle

S(0 ) ‘S: linear’

3 S( t , T ) = S(0) − k 0 ( T ) ⋅ t t 0 (T) = conditioned suitable, easy
k 0 (T)
to handle
⎛ 1 ⎞
⎜ ⎟
S( t , T ) = S(0 ) ⋅ [1 − (1 − n) ⋅ k( T ) ⋅ t ] ⎝ 1−n ⎠ 1 ⎛ 1 ⎞
t 0 (T) = ⎜ ⎟ ‘S: nth order’
4 k| (T)
k( T ) ⎝ 1 − n ⎠
with k( T ) = suitable, easy to handle
S(0 )1−n only for n < 1

S( t ) =
(
exp − NE(0) ⋅ k NE ⋅ k SP ⋅ t 2 ) ‘S: extended’ /7, 14/
1 π k SP ⎛ NE(0) ⋅ k NE ⋅ k SP ⎞ t0(T) → ∞
5 + ⋅ ⋅ erf⎜ ⋅ t⎟ suitable, fairly to handle
S(0) 2 NE(0) ⋅ k NE ⎜ 2 ⎟
⎝ ⎠ with advanced fit codes

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 67
 Description of stabilizer consumption with model ‘S: extended’
comparison of single temp. fit (sT) and all temp. fit (aT)

Model ‘S: extended‘

can describe
pronounced tailing
sometimes found

Data of stabilizer
consumption of
propellant P2
determined in
1987/1988 after
accelerated ageing
at 70°C, 80°C and
90°C.

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 68
 Description of stabilizer consumption with model ‘S: nth order’ (AOP 48 Ed. 2
model) - only all temperature (aT) fit is shown
0.9
DPA [mass-%]
Because the 0.8 P2 NC-propellant
description with model 'S: nth order' 90°C experiment
reaction order n is
0.7 all temp. fit 80°C experiment
<1, the times t0(T)
70°C experiment
to reach S(t,T) = 0
0.6 zero
can be calculated
90°C n th order
0.5 80°C n th order
70°C
70°C n th order
0.4
80°C
0.3

0.2 90°C
128.6 d
33.0 d
0.1

0
time [d]
9.1 d
-0.1
0 10 20 30 40 50 60 70 80 90 100 110 120 130

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 69
 Description of stabilizer consumption with model ‘S: exponential + linear’
comparison of ‘single temperature fit’ (sT) and ‘all temperature fit’ (aT)
0.9
time t0(T) to DPA [mass-%]
P2 NC-propellant
reach S(t,T) = 0 0.8 90°C experiment
stabilizer consumption
can be 80°C experiment
determined at begin of natural ageing (1987)
calculated 0.7 70°C experiment
description with model
always. 'S: expon. + linear' zero
0.6 90°C expon.+lin.-aT to=7.05 d
The t0(T) times
are shorter 80°C expon.+lin.-aT to=25.86 d
0.5 70°C expon.+lin.-aT to=102.36 d
than the ones 70°C
90°C expon.+lin.-sT to=6.63 d
calculated with 0.4
80°C expon.+lin.-sT to=25.83 d
model ‘S: nth
80°C 70°C expon.+lin.-sT to=107.02 d
order’ 0.3

0.2 90°C

0.1 107.02 d
7.05 d 25.86 d
0
6.63 d 25.83 d 102.36 d time [d]
-0.1
0 10 20 30 40 50 60 70 80 90 100 110 120 130

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 70
 Model ‘S: nth order’ (AOP 48 model) application with two stabilizers of different
reactivity

Model ‘S: nth

order‘ can cause
artificial shapes
of stabilizer
decrease curves
A bend upwards
with n<0 is not
reasonable for
stabilizer
consumption, as
shown with the
blue curve in
figure on the
right.

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 71
 Model ‘S: exponential + linear’ application with two stabilizers of different
reactivity

All models must

be applied in
two steps, each
stabilizer is
described
separately
DPA is much
more reactive
than Ak II.
As long as DPA
is active Ak II is
not changed in
concentration

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 72
 Comparison of models: because of conservative character in extrapolation
model ‘S: exponential + linear’ is taken for evaluation
0.9
The times t0(T)
DPA [mass-%] P2 NC propellant
are shortest with 0.8
comparison of three models
model
all temperature fits
’S: exponential + 0.7 data for 70°C extracted
linear‘
The tailing of 0.6
70°C experiment
model ‘S: nth
0.5 zero
order’ is already
70°C expon.+lin.-aT to=102.36 d
too strong to 70°C nth order-aT te=128.6 d
0.4
include the 70°C extended-aT te=160 d
stabilizer
0.3
decrease up to
'S: extended'
zero content in 0.2 'S: nth order'
the calculations
with profiles 0.1 'S: extended'
'S: exponential + linear'

Note: AOP 48, Ed 2 0

recommends not to time [d]
use model ‘S: nth -0.1
order‘ at S(t) below
20% of S(0) 0 20 40 60 80 100 120 140 160 180

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn 73
 Prediction of a real ageing over 16 years of a singlebase
propellant used in automotive seat belt restraint systems
See also /9,10/

What is needed for this task:

• The representative time-temperature profile (usually compiled for one year) at ageing
site (best would be continuous measurements of the temperature course)
• Accelerated ageing of propellant sample to parameterize its stabilizer decrease
• A model description of the stabilizer consumption which follows it up to zero content
of primary (means added at the production) stabilizer
• The Arrhenius parameters for the stabilizer decrease obtained with the model
description
• Application of the model description with the time-temperature profile data
• Measurements on stabilizer consumption during the course of real ageing of the
propellant

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 74
 Arrhenius parameters obtained with model for stabilizer consumption
‘all temperature fit’ – propellant P2, model ‘S: exponential + linear’

rate constant k1 rate constant k0

S(0)
Ea1 Ea0
model [mass- lg(Z1 [1/d]) lg(Z0 [m.-%/d])
[kJ/mol] [kJ/mol]
%]
all temp. fit 0.78 138.1 19.001 138.9 18.796
R2
0.9935
(from fit directly)

Data of stabilizer consumption of propellant P2 determined in 1987/1988 after

accelerated ageing at 70°C, 80°C and 90°C.
Now after 19 years they are used to make the prediction for the real ageing
result of propellant P2, which was aged in Florida over 16 years in set belt
cartridges /10/
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 75
 Experimental data on evolution of stabilizer concentrations
Concentrations of DPA and DPA consecutive products of single base P2
propellant in mass-% as function of ageing time at the test center in Miami
/9,10/.

ageing time at test center

Stabilizer base
1a 4a 9a 16 a
line
DPA 0.780 0.690 0.585 0.030 0.000
N-NO-DPA 0.135 0.370 0.110
2-Nitro-DPA 0.010 0.100 0.160 0.230
4-Nitro-DPA 0.020 0.170 0.300
2,4’-Dinitro-DPA 0.000
2,2’-Dinitro-DPA 0.030
2,4,4’-Trinitro-DPA 0.017

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯−−−−−⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 76
 Introduction

Company TRW Alfdorf, Germany has arranged a

long term natural ageing program with belt
restraint units equipped with ‘life’ gas generator
units of two types taken from the regular
production /9,10/
The ageing started in October 1987 in a test center near Miami, Florida, USA.
The restraint units have been mounted in such a way to simulate the situation in a car

In regular intervals samples were taken out and investigated

• Pressure-time curves made by TRW
• Chemical ageing and stability tests made by ICT

The ageing ended in October 2003. An ageing time of 16 years was realized at
ambient conditions.
But the propellants experienced elevated temperatures because of the housing of the
belt restraint units
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2006 Bohn / 77
 Time-temperature profiles for Miami, Florida, USA
40
time interval [d] time-temperature profiles
35
Miami BMP
Miami AMB
30 Miami AMB: Tm = 24.9°C

25

20
Miami BMP: Tm = 34.3°C

15

10 55°C

33°C
5
T [°C]
0
10 15 20 25 30 35 40 45 50 55 60

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯−−−−⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 78
 Model ‘S: exponential+ linear’ for use with time-temperature profile data or
continuous time-temperature measurements

⎛ k (T ) ⎞ k (T )
S( ∆ti , Ti ) = ⎜⎜ S( ∆ti−1, Ti−1) + 0 i ⎟⎟ ⋅ exp(− k1( Ti ) ⋅ ∆ti ) − 0 i
⎝ k1( Ti ) ⎠ k1( Ti )

i = 1, 2, 3 ...n number of profile section i

S(∆ti-1,Ti-1) for j=1 and i=1 S(∆ti-1,Ti-1) = S(0)
for j>1 and i=1 the value of S after profile passing j
is the start value for profile passing j+1
∆ti time interval of profile section i
Ti temperature of profile section i
j number of one full profile passing (profile cycle)

n
t (of one profile cycle ) = ∑ ∆t i
(with profile BMP it is one year)
i=1
n
ty S = j ⋅ ∑ ∆t i
i=1

With continuous temperature data (these have also time ∆ti at Ti until next Ti+1) this same
procedure is taken
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 79
 Prediction of stabilizer consumption in GG-P2 according to BMP profile load
and 4 loads with temperature-shifted BMP, model ‘S: exponential + linear’
0.9
DPA [mass-%] propellant loaded by different profiles based on BMP profile
0.8 BMP shifted by -4°C
BMP shifted by -2°C
0.7 BMP
BMP shifted by +2°C
0.6 BMP shifted by +4°C
BMP zero
BMP -2°C
0.5 0.2* S(0)
experim. data
BMP -4°C
0.4

0.3
BMP +2°C S(0)*0.2
15 29
0.2 8

0.1
BMP +4°C
11 21
0
time [a]
-0.1
0 4 8 12 16 20 24 28 32 36
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2006 Bohn / 80
 Modelling of stabilizer consumption
and the course of consecutive products

See also
/7,11,12,13, 14/

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 81
 How to handle the total stabilizing capability of a typical stabilizer

Determination of
DPA consumption
and the evolution
of consecutive
products of DPA
by Dr. Fred Volk
(about begin of the
1980s)

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 82
Modelling of stabilizer and its consecutive products – RS III
Reaction scheme for the course of DPA and its consecutive products up to the first nitration step /
nitrosation step /7/

reaction scheme III (RS III)

dxNC
= −kNC xNC − kautoxNC xP
NC ⎯ ⎯
k
⎯→ P + R
NC dt
dxP
= kNC xNC + kautoxNC xP + kDPNNOxNNO
k
NC + P ⎯ ⎯⎯⎯→ 2 P + R
auto
dt
DPA + P ⎯ ⎯⎯
NNO →
k
N-NO-DPA − (kNNO + k4N + k2N )xPxDPA
DPA + P ⎯ ⎯
k
⎯→ 4-nitro-DPA
4N − (kNNO− xNNO + k4N− x 4N + k2N− x2N )xP
k dxDPA
DPA + P ⎯ ⎯
⎯→ 2-nitro-DPA
2N
= kDPNNOxNNO − (kNNO + k4N + k2N )xP xDPA
dt
N-NO-DPA ⎯ ⎯⎯⎯
DPNNO k
→ DPA + P dxNNO
= kNNOxP xDPA − kNNO− xP xNNO − kDPNNOxNNO
dt
k
N-NO-DPA + P ⎯ ⎯⎯ ⎯→ DPANNO-
NNO− dx 4N
= k4NxP xDPA − k4N− xP x 4N
k dt
4-nitro-DPA + P ⎯ ⎯⎯
4N−
→ DPA4N-
dx2N
k = k2NxP xDPA − k2N− xP x2N
2-nitro-DPA + P ⎯ ⎯⎯
2N−
→ DPA2N- dt
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 83
Modelling of stabilizer and its consecutive products - P(t) is not constant.
RS III (continued) - with data newly measured from aged singlebase GP see /7/
Full lines: model description with the above reaction scheme
Broken lines: connect the measured data for each of the four substances
0.009
reactant singlebase GP, 90°C
0.008 [mol/mol ONO2] RS III
xP(0)=0.00075
0.007
KK = 0.99
0.006 DPA calc.
DPA exp.
0.005
N-Nitroso-DPA calc.
0.004 N-Nitroso-DPA exp.
4-Nitro-DPA calc.
0.003 4-Nitro-DPA exp.
2-Nitro-DPA calc.
0.002 2-Nitro-DPA exp.

0.001
time [d]
0
0 5 10 15 20 25 30
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 84
Stabilizer and its consecutive products, RS III (continued) – values of the rate
constants obtained by the modelling
kauto = 10 kNC and kDPNNO ≠ 0 (N-NO-DPA to DPA) all reaction rate constants in 1/d
temp. = 90°C, different initial concentrations xP(0) for product P

constant xP(0) [mol/(mol ONO2)] xP(0) as fit paramet.

0.0 0.0005 0.00075 0.00128
kNC 0.00173 0.00159 0.00155 0.00157
kNNO 340.8 300.1 258.6 156.0
k4N 104.3 98.3 88.9 57.9
k2N 44.9 35.7 32.5 20.9
kDPNNO 0.089 0.0718 0.0513 0.0135
kNNO- 20.8 25.75 27.31 25.43
k4N- 22.1 26.1 25.0 18.2
k2N- 9.66 10.43 9.79 6.66
R2
st.dev. 2.07 E-4 1.67 E-4 1.56 E-4 1.50 E-4
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 85
 Modelling of high nitrated DPA products, P(t) is not constant – RS IV – part 1
Just for presentation, the result is shown in two figures. The first part looks like the result of RS III,
but it is not. The modelling has included all products up to tri-nitrated species at once! /7/
0.009
reactant
0.008 [mol/mol ONO2] singlebase GP
RS IV
0.007 xP(0)=0.0005
DPA calc.
part 1
DPA exp.
0.006 N-nitroso-DPA calc.
N-nitroso-DPA exp.
0.005 4-nitro-DPA calc.
4-nitro-DPA exp.
0.004 2-nitro-DPA calc.
2-nitro-DPA exp.
0.003

0.002

0.001
time [d]
0
0 5 10 15 20 25 30
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 86
 Modelling of high nitrated DPA products, P(t) is not constant – RS IV – part 2

0.003
reactant
singlebase GP
[mol/mol ONO2] RS IV
0.0025 4,4'-dinitro-DPA calc. xP(0)=0.0005
4,4'-dinitro-DPA exp. part 2
0.002 2,4'-dinitro-DPA calc.
2,4'-dinitro-DPA exp.
2,2'-dinitro-DPA calc.
0.0015 2,2'-dinitro-DPA exp.
2,4,4'-trinitro-DPA calc.
0.001 2,4,4'-trinitro-DPA exp.

0.0005

time [d]
0
0 5 10 15 20 25 30
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 87
Relative reactivities of consecutive products with regard to DPA

From the modelling of the consecutive products of DPA their relative reactivity with
respect to DPA can be formulated /7/.
The advantages
• determined in a gun propellant matrix (single base GP) at 90°C
• based on reaction kinetics, reaction with the autocatalytic product

relative reactivity ERX obtained by RS IV

DPA 1.0
N-NO-DPA has a lower
N-NO-DPA 0.041 reactivity than the mono-
4-N-DPA 0.074 nitro-DPAs
2-N-DPA 0.055 And it has a much lower
4.4’-DN-DPA 0.008 reactivity than DPA itself

2.2’-DN-DPA 0.004
2.4’-DN-DPA 0.016
2.4.4’-TN-DPA 8 E-5

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 88
 -5.5
Mechanistic modelling of ln(k NC [1/d])
stabilizer reaction with RS I see /13/ -6.0 DPA decrease in a single base GP
k NC determ . according to RS I
90°C
-6.5
Activation energy Ea of NC decompos. Ea = 145.8 ± 4.8 kJ/m ol
-7.0 85°C lg(Z[1/d]) = 18.097 ± 0.72
is with 146 kJ/mol composed of two KK = 0.9978
reactions: thermolysis and hydrolysis -7.5
80°C
-8.0
7.5
ln(k D PA [1/d]) -8.5 75°C
D P A decrease in a sin gle base G P
7.0
k D PA determ . according to R S I
-9.0
90°C 70°C
6.5 E a = 132.4 ± 28 kJ/m ol
-9.5
lg(Z[1/d]) = 21.915 ± 4.1
6.0 K K = 0.9226 65°C
80°C -10.0

5.5 -10.5
75°C 1/T [1/K]
85°C
5.0 -11.0
0.00270 0.00275 0.00280 0.00285 0.00290 0.00295 0.0030
4.5

4.0 70°C

3.5 65°C Ea of stabilization reaction is with

3.0 132 kJ/mol in accordance with the
1/T [1/K ] loss of aromaticity of benzene ring
2.5
0.00270 0.00275 0.00280 0.00285 0.00290 0.00295 0.00 3
(resonance energy)
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 89
 Modelling of molar mass decrease of NC
See also /14,15/

Generally there are two main models to describe the statistical chain splitting of NC or any polymer

Model 1 chain scission by decomposition of a chain element (= monomer unit)

with time the mass of the polymer sample decreases
the number of chain elements is decreasing
model 1 was developed at ICT

Model 2 chain scission by opening of the bond (bond splitting) between two monomer units
with time the mass of the polymer sample stays constant
the number of chain elements (monomer units) stays constant
model 2 is several times given in the literature, sometimes it is referred as Ekenstam model

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 90
 Development of the kinetic model for polymer degradation – model 1 - part 1 /15/

Model 1: chain scission by chain element (=monomer unit)

number of polymer chain in
decomposition
the sample, from 1 to nC
1 2 3 4 i
number of E- E- E- E - ... E- E- E 1
chain
elements E, i+1
from 1 to N
E- E- E -.... E- E 2
.
N-1 N .
E- E- E- E- E nC

total number of chain elements: N(0) N( t , T ) = N(0 ) ⋅ exp (− kM ( T ) ⋅ t )

total number of chains: nC(0) Mn(0)
N(0) = nC (0) ⋅ Pn(0) = nC (0) ⋅
Pn=Mn/m: degree of polymerization m
Mn( t , T )
m: molar mass of chain element E N( t , T ) = nC ( t , T ) ⋅ Pn( t , T ) = nC ( t , T ) ⋅
m
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 91
 Development of the kinetic model for polymer degradation – model 1 - part 2

K( t , T ) = N(0 ) − N( t , T ) Formation of one new chain by each decomposition of one chain element
nC ( t , T ) = nC (0) + N(0) − N( t , T )

Mn(0) m 1
1+
Mn(0) nC (0) ⋅ 1+ 1+
N(0)
= exp(+ kM ( T ) ⋅ t )
m Mn( t , T ) Pn( t , T )
nC ( t , T ) = nC (0) ⋅ m = = =
Mn( t , T ) m 1 N( t , T )
1+
Mn( t , T ) nC ( t , T ) ⋅ 1+ 1+
m Mn(0) Pn(0)
m

⎛ m ⎞ ⎛ m ⎞
⎜1 + ⎟ = ⎜1 + ⎟ ⋅ exp (+ kM ( T ) ⋅ t )
⎝ Mn( t , T ) ⎠ ⎝ Mn(0 ) ⎠

Approximated formulas
⎛ m ⎞ ⎛ m ⎞
ln⎜1 + ⎟ = ln⎜1+ ⎟ + kM( T ) ⋅ t
⎝ Mn( t , T ) ⎠ ⎝ Mn(0) ⎠ m m
= + kM( T ) ⋅ t
Mn( t , T ) Mn(0)
corresponding rate expression
⎛ ⎛ Mn( t , T ) ⎞ ⎞ ⎛ ⎛ Mn( t , T ) ⎞ ⎞
⎜ d⎜ ⎟⎟ ⎜ d⎜ ⎟⎟ 2
⎛ 2⎞
⎜ ⎝ m ⎠ ⎛
⎟ = −k ( T ) ⋅ ⎜ Mn( t , T ) ⎞
⎜ ⎝ m ⎠ ⎟ = −k ( T ) ⋅ ⎜ ⎛⎜ Mn( t , T ) ⎞⎟ + ⎛⎜ Mn( t , T ) ⎞⎟ ⎟ M ⎟
⎜ dt ⎟ M ⎜⎝ m ⎠ ⎝ m ⎠ ⎟ ⎜ dt ⎟ ⎝ m ⎠
⎜ ⎟ ⎝ ⎠ ⎜ ⎟
⎝ ⎠T ⎝ ⎠T

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 92
 Development of the kinetic model for polymer degradation – model 1 - part 3

time tyMn is the looked for life time;

⎛ m ⋅ D( t , T ) ⎞ ⎛ m ⋅ D(0) ⎞
this is the time to reach the degree of ln⎜ 1 + ⎟ = ln⎜ 1+ ⎟ + kM ( T ) ⋅ t
degradation yMn=Mn(t)/Mn(0) ⎝ An ⋅ Mw ( t , T ) ⎠ ⎝ An ⋅ Mw (0 ) ⎠

⎛ 1 An ⋅ Mw (0 ) ⎞
⎜ + ⎟ m ⋅ D( t , T ) m ⋅ D(0 ) D=
Mw
1 ⎜ yMn m ⋅ D(0 ) ⎟ = + An ⋅ kM ( T ) ⋅ t
tyMn ( T ) = ⋅ ln Mw ( t , T ) Mw (0 ) Mn
kM ( T ) ⎜ 1 + An ⋅ Mw (0 ) ⎟
⎜ m ⋅ D(0 ) ⎟
⎝ ⎠
An: conversion factor from absolute to
relative Mn values: Mnabs = An Mnrel
Model conditions
• statistical chain splitting.
• the chain splitting occurs by decomposition of chain elements; with NC the triggering
is the thermolytic and /or hydrolytic degradation of the ≡CO-NO2 groups
• the total number of chain elements is much greater than the number of chains
• the total number of chain elements decreases with first order, rate constant kM(T)
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 93
 Kinetic model for polymer degradation based on model 2 – part 1
Model 2: chain scission by bond breaking between chain elements (monomer units)

total number of bonds at time zero: B(0)

B(0) = ∑ B j (0) = ∑ (N j (0) − 1) = N(0) − nC (0)
total number of chain elements: N(0)
total number of chains: nC(0)
number of bonds in chain j: Bj(0)
number of chain elements in chain j: Nj(0)
B( t , T ) = N(0) − nC ( t , T ) = N( t , T ) − nC ( t , T ) with N(0) = N(t, T)

⎛ d(B( t , T ) ⎞
⎜ ⎟ = −kB ( T ) ⋅ B( t , T ) B( t , T ) = B(0) ⋅ exp( − k B ( T ) ⋅ t )
⎝ dt ⎠T

nC ( t , T ) m
n ( t, T ) =
1− C N(0) Mn( t , T ) 1−
m
1−
1
B( t , T ) N(0) − nC ( t , T ) N(0)
= =
= exp(− k B ( T ) ⋅ t )
Mn( t , T ) Pn( t , T )
B(0) N(0) − nC (0) n (0 ) =
1− C nC ( 0 ) m m 1
N(0) = 1− 1−
N(0) Mn(0) Mn(0) Pn(0)

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn/ 94
 Kinetic model for polymer degradation based on model 2 – part 2

m 1
1− 1−
⎛ m ⎞ ⎛ m ⎞
= exp(− kB ( T ) ⋅ t )
Mn( t, T ) Pn( t , T ) ln⎜1 −
= ⎟ = ln⎜1 − ⎟ − kB (T) ⋅ t
m 1 ⎝ Mn( t , T )⎠ ⎝ Mn(0)⎠
1− 1−
Mn(0) Pn(0)
Signs in formula are different to the model
based on chain element decomposition
⎛ ⎛ Mn( t , T )⎞ ⎞
⎜ d⎜ ⎟⎟ ⎛ ⎛ Mn( t , T )⎞ ⎛ Mn( t , T )⎞ 2 ⎞
⎜ ⎝ m ⎠⎟
= −k B ( T ) ⋅ ⎜ − ⎜ ⎟ +⎜ ⎟ ⎟ Approximated formulas
⎜ dt ⎟ ⎝ ⎝ m ⎠ ⎝ m ⎠ ⎠
⎜ ⎟
⎝ ⎠ m m
T = + kB ( T ) ⋅ t
Mn( t , T ) Mn(0)
⎛ An ⋅ Mw(0) 1 ⎞
⎜ − ⎟
1 m ⋅ D (0) y ⎛ ⎛ Mn( t , T ) ⎞ ⎞
ty Mn ( T ) = − ⋅ ln⎜ Mn ⎟
⎜ d⎜ ⎟⎟
⎜ An ⋅ Mw(0 ) ⎟ 2
⎜ ⎝ m ⎠ ⎟ = −k ( T ) ⋅ ⎛⎜ Mn( t , T ) ⎞⎟
kB (T)
⎜ −1 ⎟
⎝ m ⋅ D(0) ⎠ ⎜ dt ⎟ B
⎝ m ⎠
⎜ ⎟
Signs in formula are different to the model
⎝ ⎠T
based on chain element decomposition In approximated form model 2
and model 1 are equal
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn/ 95
Definition of terms – mean molar masses

The polymer fraction i is characterized by some quantities; it is thought to be molecular uniform,

means it has only molecules of one molar mass Mi and the number of mols is ni
The polymer sample is characterized by a distribution function hxi, which can be of different kind

∑ hn (M) ⋅ M ∑ hm (M) u ,i i u ,i
Mn: mean molar mass averaged according to the mol
Mn = i
= i

∑ hn (M ) hm (M)
∑ M i
u ,i u ,i
number in the polymer fraction i (mol number weighing)
i i

∑ hn (M) ⋅ M ⋅ M ∑ hm (M) ⋅ M
u ,i i i u ,i i Mw: mean molar mass averaged according to the mass
Mw = i
= i

∑ hn (M ) ⋅
i
M ∑ hm (M) u ,i i
i
u ,i
of the polymer fraction i (mass weighing)

∑ hn (M) ⋅ M ⋅ M ⋅ M ∑ hm (M) ⋅ M ⋅ M ∑ hz (M) ⋅ M

u ,i i i i u ,i i i u ,i i
Mz = i
= = i i

∑ hn (M ) ⋅ M ⋅
i
M ∑ hm (M ) ⋅ M
u ,i ∑ hz (M) i i
i
u ,i i
i
u ,i

Mz: mean molar mass averaged according to the z-mass of the polymer fraction i (z weighing)
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn/ 96
 Definition of terms - distribution functions

hmu,i (M)
hnu,i (M) Mi hni(M) polymer fraction mol number related
hni (M) = =
∑ hmu,i (M) normalized molar mass distribution
∑
hnu ,i (M)
i
ni is mol number in polymer fraction i
i Mi

hzu,i (M)
hmu,i (M) Mi hmi(M) polymer fraction mass related molar
hmi (M) = =
∑ hmu,i(M) ∑ hzu,i (M) normalized molar mass distribution
mi = Mi*ni
i Mi
i
mi is the mass of polymer fraction i

hzu,i (M)
hzi (M) = hzi(M) polymer fraction z-weight related
∑ hzu,i(M) normalized molar mass distribution
i zi = Mi*mi = Mi*Mi*ni
zi is the z mass of polymer fraction i
hnu,i and hmu,i and hzu,i are un-normalized distribution functions

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn/ 97
 Examples of molar mass degradation

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 98
 Modelling of the molar mass decrease of NC – model 1

Modelling of NC
degradation in a
double base rocket
propellant
on the ordinate the
model expression of
model 1 is shown

The full lines are the model

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 99
 Modelling of the molar mass decrease of NC – model 1

Modelling of NC
degradation in a
double base rocket
propellant
on the ordinate
mean molar mass
Mw is shown, back
calculated from the
model expression of
model 1
the description of
the data is very good
The full lines are the model

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 100
 Model for degradation of molar mass of NC - extrapolation ability – model 1

This graph shows the

good extrapolation
ability of the model 1.
The figure has two lines:
red and black.
The red one is
determined by using the
model 1 with data up to
30 days.
The black one is
determined with data
up to 60 days.
Both curves coincide.

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 101
Molar mass distribution functions of aged PVN – determined by GPC /2/
PVN, polyvinylnitrate, an energetic speciality polymer
0.0045
hm(M) PVN-M The systematic in the
0.004 MMD based on polymer fraction mass ageing is expressed in
80°C a common crossing
0.0035 un-aged ‘point’ of the curves
80°C 1d at about M = 1.3E+05
0.003 80°C 2d g/mol
80°C 3d
0.0025 80°C 7d
80°C 10d
0.002 80°C 14d
80°C 18d
0.0015 zero

0.001

0.0005

0
M [g/mol]
-0.0005
1.E+03 1.E+04 1.E+05 1.E+06 1.E+07
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 102
 Molar mass degradation of PVN - description with model 1

220000
Mn [g/mol] PVN-M
description of Mn with kinetic model
200000 based and statistical chain splitting 80°C Mn-calc
high molecular part 75°C Mn-calc
70°C Mn-calc
180000

160000

140000 70°C

75°C
120000
80°C

100000

time [d]
80000
0 10 20 30 40 50 60

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 103
 Molar mass degradation of PVN – description with model 1

For the chain ln(kMn [1/d]) PVN-M

scission in PVN -11 MMD evaluation with model
an activation high molecular part, Mn(0) = 189600 g/mol
energy of 146
kJ/mol is 80°C
determined /2/.
-11.5
Such Ea values are
found also NC EaMn = 146.1 ± 13 kJ/mol
based propellants lg(ZMn [1/d]) = 16.748 ± 2
2
R =0.9918
-12
75°C

-12.5

70°C
1/T [1/K]

-13
0.00282 0.00284 0.00286 0.00288 0.0029 0.00292 0.00294
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 104
 Application and meaning of molar mass decrease

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 105
 Molar mass decrease - brittle ductile transition

shift of the brittle-ductile

transition temperature to
higher temperature values
with decreasing molar
mass = decreasing chain
length of NC
ageing causes increase in
brittle tendency
danger during firing
rounds with large grain GP

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 106
 Example for molar mass decrease - brittle ductile transition

.
GP KN6540 (M30
type), triple base
GP, 110mm tank
gun
stress - strain
curves of non-
aged and aged
samples as func-
tion of temper-
ature determined
at high strain rate,
about 5000%/sec
the brittleness non-aged
increases with aged, 80°C, 29d
ageing

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 107
 Molar mass decrease - brittle ductile transition

GP KN6540 (M30
type), triple base GP,
brittle stress
110mm tank gun

shift of the brittle

ductile, yield stress
ductile transition
temperature with
ageing to higher
values
.
non-aged
aged, 80°C, 29d

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 108
 Correlation of molar mass decrease and decrease of mechanical strength

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 109
Ageing of db rocket propellant – correlation between molar mass decrease of NC
and decrease mechanical properties

Correlation between molar mass

decrease, stabilizer decrease and
decrease of strain at break and
tensile stress at break for the
doublebase RP DBE 44/36 /16/

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn 110
 Ageing of new type gun propellants
See also /17/

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 111
Types of gun propellants - overview
Main types of GP (numbers give applied ranges in mass-%)
Single base NC( N content 13 to 13.4 mass-%) 90-95, stabilizer 1-2, plasticizer 2-5,
graphite
Double base NC (N content 12.3 to 12.6 mass-%), 30 to 60, blasting oils 25 to 50
(NG, DEGN, special types also with TMETN, BTTN), stabilizer 1-2,
graphite
NH2
Triple base NC (N-content 12.3 to 12.6 mass-%) 30 to 40, blasting oils 20 to 30 HN C
nitroguanidine 25 to 40, stabilizer 1 to 2, plasticizer 5 to 10, graphite NH
NO2
Semi nitramine NC (N-content 12.3 to 12.6 mass-%) 30 to 40, blasting oils 20 to 30,
RDX 25 to 40, stabilizer 1 to 2, plasticizer 5 to 10, graphite
New-type of ICT NC (N-content 12.3 to 12.6 mass-%) 30 to 40, DNDA (dinitro-diaza)-type
plasticizer 20 to 30, RDX 25 to 40, stabilizer 1 to 2, graphite n n+1 n+2
CH3 N CH2 N CH3

NO2 NO2
All show a standard decomposition behaviour with respect to NC and blasting oil decomposition.
But in triple base GP the NO2 is in part captured by nitroguanidine (NQ), because of reaction
between NO2 and the NH groups of NQ. This and the dilution effect for NC by NQ leads to the fact
that with triple base propellants no visible NO2 can be observed and no strong autocatalysis.

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 112
 Type of gun propellants - shapes

Main types of GP according to shape and treatment

cylindrical grain with several holes, mostly 1, 7, 19
long tube
short tube
flake
stripe
ball (sphere, ellipsoids)
cube
ring
horseshoe
nudle (short rods)
Surface treatment:
phlegmatization to adapt burning behaviour
energetic enforcement by blasting oil treatment (Wimmis EI type)

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Pecul 113
 n n+1 n+2
Energetic plasticizer DNDA CH3 N CH2 N CH3

NO2 NO2
recent plasticizer based on the n,(n+2)-Di Nitro - n,(n+2)-Di Aza group

DNDA 57: liquid dinitro-diaza plasticizer

Mixture of three components: DNDA5 : DNDA6 : DNDA7, about 43 : 45 : 12 mass-%

CH3 N CH2 N CH3

2,4-dinitro-2,4-diaza-pentane, DNDA 5 (54°C)
NO2 NO2

CH3 N CH2 N CH2 CH3

2,4-dinitro-2,4-diaza-hexane, DNDA 6 (32°C)
NO2 NO2

CH3 CH2 N CH2 N CH2 CH3

3,5-dinitro-3,5-diaza-heptane, DNDA 7 (75°C)
NO2 NO2

DNDA 57 (Tm = -9°C) gels with NC

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 114
 Defining the LTC effect for gun propellants

LTC: low temperature coefficient (coefficient with small value)

means
A gun propellant which shows small to very small dependence of
maximum chamber pressure pmax on charge temperature TC
temperature range of interest is from -40°C (-54°C) to +63°C (+71°C)

These New Type GP of ICT show this interesting effect of low temperature coefficient
of maximum chamber gas pressure
This effect can be used the reduce weight of the gun system (small and light vehicles)
or to increase the charge mass to enhance the muzzle velocity of the projectile

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯−−−−−⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 115
 Comparison of typical gas pressure curves of conventional and LTC propellants
maximum gas pressure

LTC Propellant
Pm

Conventional Propellant

4 0 0 0

-50 -40 -30 -20 -10 0 10 20 30 40 50 60

propellant temperature (°C)

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 116
 Some characteristic data of GP formulations

LTC autoignition adiab. flame force QEX VEX at 25°C

GP web
effect temp. [°C] temp. Tad [K] [J/g] [J/g] [ml/g]

TLP 1N 7 yes 185 2905 1170 3768 961

TLP 2N 7 yes 220 2906 1178 4198 939
TLP 3N 7 yes 193 2910 1180 4201 939
TLP 4G 7 yes 198 2908 1185 4071 938
TLP 5W 19 yes 189 2510 1060 4124 931
TLP 6 19 yes 199 2540 1080 4177 929
For comparison the data of two typical conventional GP
JA2 ( L 5460, db) 7 no 168 3390 1139 4610 753
MRCA (Q 5560, tb) 19 no 172 3078 1040 3758 857

Adiabatic flame temperature Tad, force, heat of explosion QEX and gas volume VEX have been
calculated by ICT Thermodynamic Code using the data of the ICT Thermochemical Data Base
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 117
 Adiabatic self heating determined by an ARCTM of typical conventional and of
new type GP, including the XM39 GP with a binder based on CAB
100
h [°C/min] Comparison of adiabatic self heat rate

TLP 1N
TLP 4G TLP 2N
TLP 6
10
JA2 TLP 5W
M 30
single base CAB-LOVA
1

0.1 CAB-LOVA

T [°C]
0.01
125 135 145 155 165 175 185 195 205
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 118
Assessment of ageing and stability with mass loss
9
ML [%] mass loss at 90°C
8 1N

7 JA2 5W
2N
6 MRCA
4G
5
3N
4 limit: ≤ 3%, 18 d

2
6
1
test time [d]

0
0 10 20 30 40 50 60 70 80 90 100

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 119
 Modelling of ML data with model ‘ML: minor 1st ord.; main 1st ord. + autocat.’
See also /6/
Reaction scheme
A ⎯⎯→ k1
B+C+S inherent decomp. of main comp. A: C = gases, S= solids

A + B ⎯⎯→
k2
2 B + C + S autocatalytic decomp. of main comp. A; B acts autocatalytically

⎯⎯→
⎯
k
V V V_ decomp. of minor comp. V; describes initial ML

Reaction kinetic formulation gives the following equation for total mass loss ML
⎧m ⎛ M (0) + MS(0) + MN ⎞ ⎫
ML(t, T) = 100%⋅ ⎨ C ⋅ ⎜ (1− aV ) − B ⎟ ⋅ (1− MAr(t, T)) + aV ⋅ (1− exp(−k V (T) ⋅ t))⎬
⎩mA ⎝ M(0) ⎠ ⎭
MN: mass of a
non-reacting Decomposition of main component A Decomp. of minor comp. V
ingredient N in A

k1(T) + (F + 1) ⋅ k2(T) Decomposition of main

MAr(t, T) =
( ) (( ) )
k2(T) + k1(T) + F ⋅ k2(T) ⋅ exp k1(T) + (F + 1) ⋅ k2(T) ⋅ t component according to ‘1st
order + autocatalytic‘
Parameters to be determined from the measurement curve:
k1, k2, kV, (aV)
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 120
 Assessment of ageing and stability with mass loss

NT-GP 6

Modelling of ML data with kinetic model

‘ML: minor: 1st order; main: 1st order+ autocat.‘

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 121
 Times tyML to reach 3% mass loss in years calculated with Arrhenius parameters
obtained from mass loss measurements at 70°C, 80°C and 90°C

See /17/ for more explanations

1N 2N 3N 4G 5W 6 MRCA JA2 JA2, HT

web 7 7 7 7 19 19 19 7 7
EaML [kJ/mol] 116.5 104.3 101.8 105.6 105.7 102.2 147.9 102.0 149.0
15.23 13.28 12.86 13.45
log(ZML [1/d]) 5 6 5 7 13.56 12.888 19.837 13.282 20.102
time tyML to reach 3% mass loss ML in years at preset temperatures
65°C [a] 1.6 1.8 2.0 1.9 1.6 2.2 2.8 0.8 2.2
50°C [a] 10.6 10.3 10.6 11.1 9.2 11.9 31.9 4.5 26.3
35°C [a] 87.8 68.1 67.1 75.1 62.3 75.6 465.1 28.3 391.2

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯−−−−−⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 122
Assessment of ageing and stability by heat generation rate (dQ/dt) and heat
generation Q at 90°C - Part 1: NT-GP 1N, 2N, 3N and 4G
50
200 dQ/dt [µW/g] Q [J/g]
heat generation rate (left ordinate) and
heat generation (right ordinate, bold lines) 45
180 at 90°C
40
160
1N
Q are the curves directed to the right upwards 35
140

2N 30
120 3N

25
100
4G
20
80

60 1N 15
2N
40 10
3N
20 5
4G time [d]
0 0
0 1 2 3 4 5 6 7 8
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 123
Assessment of ageing and stability by heat generation rate (dQ/dt) and heat
generation (Q) at 90°C - Part 2: NT-GP 5W, 6 and JA2, MRCA,.
160 35
dQ/dt [µW/g] heat generation rate (left ordinate) and Q [J/g]
heat generation (right ordinate, bold lines) MRCA
140 at 90°C 30
JA2
120
25

100 5W
20

80 6

15
60 MRCA

JA2
10
40
5W
5
20
6
time [d]
0 0
0 1 2 3 4 5 6 7
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 124
 Times tyEL to reach 3% energy loss EL
Times tyEL in years calculated with Arrhenius parameters obtained from heat
generation rate measurements at 60°C, 70°C, 80°C and 90°C.
Reference quantity Qref is the individual heat of explosion QEX
See /17/ for more explanations

1N 2N 3N 4G 5W 6 MRCA JA2
QEX [J/g] 3768 4198 4201 4071 4124 4177 3758 4610
web 7 7 7 7 19 19 19 7
EaQ [kJ/mol] 112.3 115.6 121.6 118.2 111.5 113.9 130.0 128.2
log(ZQ
17.971 18.217 19.059 18.436 17.633 17.847 20.426 20.056
[µW/g])
time tyEL to reach 3% energy loss EL in years at preset temperatures
65°C [a] 0.85 1.74 2.12 2.57 1.53 2.22 1.62 2.45
50°C [a] 5.4 11.7 15.8 18.1 9.6 14.6 13.8 20.3
35°C [a] 42 95 142 154 73 115 146 207
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 125
 Literature – part 1 of 3
/1/ M.A. Bohn, F. Volk.
Aging Behavior of Propellants Investigated by Heat Generation, Stabilizer Consumption, and Molar Mass Degradation.
Propellants, Explosives, Pyrotechnics 17 (1992) 171-178.

/2/ M.A. Bohn, J. Aniol, M. Dörich, K. Hartlieb, H. Pontius

Stability, Ageing and Usetime Prediction of Polyvinylnitrate (PVN).
Paper 73 on the 37th International Annual Conference of ICT, June 27 to 30, 2006, Karlsruhe,Germany.
Proceedings 2006, pages 73-1 to 73-18. Fraunhofer-Institut für Chemische Technologie (ICT), D-76318 Pfinztal.

/3/ M.A. Bohn, T.M. Klapötke

DFT and G2MP2 Calculations of the N-N Bond Dissociation Enthalpies and Enthalpies of Formation of Hydrazine,
Monomethylhydrazine and Symmetrical and Unsymmetrical Dimethylhydrazine.
Z. Naturforschung 59b, 148-152 (2004)

/4/ M.A. Bohn and M.E. Grillo

Quantum mechanical calculations used to reveal decomposition ways of Ammonium Dinitramide (ADN).
Paper 74 on the 37th International Annual Conference of ICT, June 27 to 30, 2006, Karlsruhe, Germany.
Proceedings 2006, pages 74-1 to 74-17. Fraunhofer-Institut für Chemische Technologie (ICT), D-76318 Pfinztal.

/5/ M.A. Bohn, F. Volk, O. Brauer

Determining the Safe Lifetime of a Plasticized Nitrocellulose.
Proceed of the 8th International Syposium on ‘Loss Prevention and Safety Promotion in the Process Industries’ Vol. II
Ed. J.J. Mewis, H.J. Pasman, E.E. De Rademaeker
Elsevier Science B.V., Amsterdam, New York, Oxford Tokyo, 1995.

/6/ M.A. Bohn

Kinetic Description of Mass Loss Data for the Assessment of Stability. Compatibility and Aging of
Energetic Components and Formulations Exemplified with ε-CL20.
Propellants Explosives Pyrotechnics 27. 125-135 (2002).
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 126
 Literature – part 2 of 3
/7/ M.A. Bohn, N. Eisenreich
Kinetic Modelling of the Stabilizer Consumption and of the Consecutive Products of the Stabilizer in a Gun Propellant.
Propellants, Explosives, Pyrotechnics 22, 125-136 (1997).

/8/ M.A. Bohn

Prediction of Life Times of Propellants - Improved Kinetic Description of the Stabilizer Consumption.
Propellants, Explosives, Pyrotechnics 19, 266-269 (1994).

/9/ M.A. Bohn

Ammunition monitoring in field situations by stabilizer consumption and molar mass decrease as predictive tools.
Paper 25 on the 37th International Annual Conference of ICT, June 27 to 30, 2006, Karlsruhe, Germany.
Proceedings 2006, pages 25-1 to 25-19. Fraunhofer-Institut für Chemische Technologie (ICT), D-76318 Pfinztal.

/10/ M.A. Bohn, F. Volk, S. Lütze, S. Strempel, K.-H. Köppel

Monitoring of the Natural Ageing over 16 years of two Single Base Propellants Used in Car Safety Systems.
Paper 72 on the 37th International Annual Conference of ICT, June 27 to 30, 2006, Karlsruhe, Germany.
Proceedings 2006, pages 72-1 to 72-18. Fraunhofer-Institut für Chemische Technologie (ICT), D-76318 Pfinztal.

/11/ M.A. Bohn

Kinetic modelling of the concentrations of the stabilizer DPA and some of its consecutive products
as function of time and temperature.
J. Thermal Analysis and Calorimetry 65, 103-120 (2001).

/12/ M.A. Bohn

Modelling of Stabilizer Reactions in Gun and Rocket Propellants.
in ‘Ageing Studies and Lifetime Extension of Materials’ pages 449 - 466. Editor: L.G. Mallinson.
Kluwer Academic / Plenum Publishers, New York, Boston, Dordrecht, London, Moscow, 2001.

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 127
 Literature – part 3 of 3

/13/ M.A. Bohn, N. Eisenreich

Temperaturabhängigkeit der Zersetzung der Nitrocellulose bestimmt mit Modellierung des TGA-Massenverlusts und
Stabilisatorverbrauchs (Temperature dependence of the decomposition of nitrocellulose determined by TGA mass loss
and stabilizer consumption).
Paper 62 in Proceed. of the 31st Internat. Annual Conference of ICT 2000. pages 62-1 to 62-10. June 27-June 30, 2000,
Karlsruhe, Germany. Fraunhofer-Institut für Chemische Technologie (ICT). D-76318 Pfinztal-Berghausen. Germany.

/14/ M.A. Bohn

The Use of Kinetic Equations to Evaluate the Ageing Behaviour of Energetic Materials – Possible Problems.
Proceed. 11th Symp. on ‘Chemical Problems Connected with the Stability of Explosives’ (held in Båstad, May 24 to 28,
1998), pages 89-151. Editor: Jan Hansson, Ola Listh. 1999, Sundbyberg, Sweden.

/15/ A. Pfeil, H.H. Krause, N. Eisenreich.

The Consequences of Beginning Slow Thermal Decomposition on the Molecular Weight of Nitrated Cellulose.
Thermochimica Acta 85 (1985) 395-398.

/16/ F. Volk, M.A. Bohn, G. Wunsch.

Determination of Chemical and Mechanical Properties of Double Base Propellants during Aging.
Propellants, Explosives, Pyrotechnics 12 (1987) 81-87.

/17/ M.A. Bohn, D. Müller

Insensitivity aspects of NC bonded and DNDA plasticizer containing gun propellants.
Paper 47 on the 37th International Annual Conference of ICT, June 27 to 30, 2006, Karlsruhe, Germany.
Proceedings 2006, pages 47-1 to 47-11. Fraunhofer-Institut für Chemische Technologie (ICT), D-76318 Pfinztal.

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 128
 .

Thank you for your attention

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
2007 Bohn / 129