LA-UR-15-29536

Approved for public release; distribution is unlimited.

Title: JWL Equation of State

Author(s): Menikoff, Ralph

Intended for: tutorial for younger colleagues

Issued: 2017-07-11 (rev.1)

Disclaimer:
Los Alamos National Laboratory, an affirmative action/equal opportunity employer, is operated by the Los Alamos National Security, LLC for
the National Nuclear Security Administration of the U.S. Department of Energy under contract DE-AC52-06NA25396. By approving this
article, the publisher recognizes that the U.S. Government retains nonexclusive, royalty-free license to publish or reproduce the published
form of this contribution, or to allow others to do so, for U.S. Government purposes. Los Alamos National Laboratory requests that the
publisher identify this article as work performed under the auspices of the U.S. Department of Energy. Los Alamos National Laboratory
strongly supports academic freedom and a researcher's right to publish; as an institution, however, the Laboratory does not endorse the
viewpoint of a publication or guarantee its technical correctness.
 JWL Equation of State

Ralph Menikoff

October 30, 2016

Abstract

The JWL equation of state (EOS) is frequently used for the products
(and sometimes reactants) of a high explosive (HE). Here we review
and systematically derive important properties. The JWL EOS is of
the Mie-Grüneisen form with a constant Grüneisen coefficient and a
constants specific heat. It is thermodynamically consistent to specify
the temperature at a reference state. However, increasing the refer-
ence state temperature restricts the EOS domain in the (V, e)-plane
of phase space. The restrictions are due to the conditions that P ≥ 0,
T ≥ 0, and the isothermal bulk modulus is positive. Typically, this
limits the low temperature regime in expansion. The domain restric-
tions can result in the P -T equilibrium EOS of a partly burned HE
failing to have a solution in some cases. For application to HE, the
heat of detonation is discussed. Example JWL parameters for an HE,
both products and reactions, are used to illustrate the restrictions on
the domain of the EOS.

1
Functional form for JWL pressure
The pressure for the Jones-Wilkins-Lee (JWL) equation of state (EOS) is
defined by [Kury et al., 1965; Lee et al., 1968]
   
ωV0 ωV0
P (V, e) =A 1 − exp(−R1 V /V0 ) + B 1 − exp(−R2 V /V0 )
V R1 V R2
(1)
ω

+ e + ∆e ,
V
where ρ0 = 1/V0 is the initial density, ω is the Grüneisen coefficient V (∂e P )V ,
and A, B, R1 , R2 are parameters. The parameters A and B have dimensions
of pressure, while R1 , R2 and ω are dimensionless. They are subject to the
constraints that R1 > R2 > 0 and ω > 0. Moreover, for explosive products
A > B > 0. But when used for reactants A > −B > 0.
∆e is an offset in the origin of specific internal energy. For the reactants
of an explosive, ∆e can be chosen such that for the ambient state specified
by (V0 , P0 , T0 ), the specific energy e0 = 0. For the products, ∆e corresponds
to the heat of detonation when the initial state of the reactants is such that
e0 = 0 and P0 = 0. This will be shown in a later section.

Isentropes
For HE products, the isentrope through the CJ state is important for applica-
tions and simulations. Isentropes also play an important role in constructing
the thermal component for the JWL EOS. They are determined by the inte-
gral curves of the ODE
de/dV = −P (V, e) . (2)
It can easily be verified that the isentropes are given by

PC (V ) = A exp(−R1 v) + B exp(−R2 v) + C v −ω−1 , (3a)

h
eC (V ) = −∆e + V0 (A/R1 ) exp(−R1 v) + (B/R2 ) exp(−R2 v) (3b)
i
+ (C/ω)v −ω ,

2
where v = V /V0 and C is an integration constant with dimensions of pressure
that serves to label an isentrope. Later, when the thermal component of the
EOS is discussed, the label C will be related to the value of the entropy.
We note the following properties. For all isentropes, PC (V ) → 0 as
V → ∞. For explosive products, if C ≥ 0 then PC (V ) > 0 for all V > 0 and
PC (V ) → ∞ as V → 0. When C < 0, PC is negative for sufficiently large V .
The isentrope through a state (Vr , er ) corresponds to
h i
Cr = (er +∆e)/V0 −(A/R1 ) exp(−R1 vr )−(B/R2 ) exp(−R2 vr ) ω vr ω , (4)

where vr = Vr /V0 . The isentrope can then be used to obtain the sound speed
c2 (Vr , er ) = −Vr2 (dPCr /dV ) (Vr ) = −V0 vr2 (dPCr /dv) (vr )
n h i Cr o (5)
= V0 vr2 A R1 exp(−R1 vr ) + B R2 exp(−R2 vr ) + (ω + 1) ω .
vr
If A, B, Cr > 0, then the isentropic bulk modulus, K = ρ c2 , is positive for
all V > 0. Moreover, c → 0 as V → ∞. Later issues that arise when Cr < 0
will be discussed.
For any isentrope, the JWL equation of state has the Mie-Grüneisen form,
ωh i
P (V, e) = PC (V ) + e − eC (V ) . (6)
V
Equation (1) is identical to Eq. (6) with C = 0.
Typically, for explosive products, the JWL parameters are fit to data for
the release isentrope from the detonation state. When used for explosive
reactants, the parameters are fit to data for the shock locus through the
initial state (V0 , e0 ).

Formulation of temperature
To obtain a complete EOS from the JWL pressure, Eq. (1), a thermodynamic
consistent temperature needs to be specified. Utilizing the thermodynamic
relation
(∂V T )S = −(T /V ) ω , (7)
and assuming a constant specific heat, CV , the JWL temperature can be
defined as follows:

3
 1. Select the temperature at a reference point; Tr at (Vr , er ).
For products EOS, it is natural to use the CJ state as the reference
point. For reactants or inert, it is natural to chose the ambient state
(atmospheric pressure and room temperature) as the reference point.

2. The reference isentrope is given by Eqs. (3a) and (3b) with the param-
eter Cr given by Eq. (4).

3. Along the reference isentrope, the temperature is determined by Eq. (7).

Since the Grüneisen coefficient, ω, is constant,
 ω
Vr
TCr (V ) = Tr . (8a)
V

Equation (7) implies that the temperature on all isotherms have this
form. Hence, the T = 0 isotherm is also an isentrope.

4. For a constant specific heat, the temperature off the reference isentrope
is given by
e − eCr (V )
T (V, e) = TCr (V ) + , (8b)
CV
where eCr (V ) is given by Eqs. (3b) and (4).

5. The entropy can then be defined by

h i
S(V, e) = CV ln(T (V, e)/Tr ) + ω ln(V /Vr ) + Sr , (9)

where Sr is the entropy at the reference state (Vr , er ).

Equations (8) and (9) can be combined to give

  
S−Sr

e(V, S) = eCr (V ) + CV TCr (V ) exp CV − 1 . (10)

This satisfies the fundamental thermodynamic relation

de = −P dV + T dS . (11)

Since e(V, S) is a thermodynamic potential, Eq. (10) determines a thermo-

dynamically consistent complete EOS. Equations (1), (8) and (9) define the
complete JWL EOS with (V, e) as independent thermodynamic variables.

4
 The isentrope label C in Eq. (3) can be mapped into the entropy by
substituting Eq. (3b) into Eq. (9);

S(C) = S(V, eC (V )) . (12)

The right hand side of Eq. (12) is independent of V since Eq. (2) is consistent
with Eq. (11). Moreover, S(C) is a monotonic function; i.e., 1-to-1 map
for C ≥ C0 , where C0 corresponds to the T = 0 isentrope. The function
S(C) depends on the choice of the reference temperature Tr . The reference
temperature also affects the domain of the EOS. The domain will be discussed
in a later section.
We note that S → −∞ as T → 0. Physically, S should go to 0 at T = 0.
This requires CV → 0 as T → 0, which is contrary to the assumption of a
constant specific heat. The JWL EOS is not meant for low temperature.
For a more general derivation of a complete Mie-Grüneisen EOS with a non-
constant specific heat see [Menikoff; 2012].

Simplified form
Combining Eqs. (8b), (4), (3a) and (3b), the temperature can be expressed
as
 
A B Eer
CV T (V, e) = e + ∆e − V0 exp(−R1 v) + exp(−R2 v) + ω , (13a)
R1 R2 v
 hA i
ω B
Er = vr CV Tr − er − ∆e + V0
e exp(−R1 vr ) + exp(−R2 vr ) . (13b)
R1 R2
Then using Eq. (1) to eliminate e + ∆e, the pressure can be expressed as
er /V0 
CV T E
P (V, T ) = A exp(−R1 v) + B exp(−R2 v) + ω − ω+1 . (14)
V v

Both T (V, e) and P (V, T ) depend on the choice of reference temperature.

If the reference temperature is chosen such that Eer = 0,
 
A B
CV Tr = er + ∆e − V0 exp(−R1 vr ) + exp(−R2 vr ) , (15)
R1 R2

5
then the expressions for the temperature and pressure simplify
 
A B
CV T (V, e) = e + ∆e − V0 exp(−R1 v) + exp(−R2 v) , (16a)
R1 R2
ω
P (V, T ) = A exp(−R1 v) + B exp(−R2 v) + CV T . (16b)
V
In this case, for large V  V0 , the temperature and pressure reduce to

CV T (V, e) = e + ∆e ,
ω
P (V, T ) = CV T ,
V
and the JWL equation of state limits to an ideal gas equation of state.
Equations (8b), (3b) and (16a) imply that T = 0 on the C = 0 isentrope.
In particular, at V0 the temperature is zero at P = A exp(−R1 )+B exp(−R2 ).
This can be unrealistically high for detonation products. It is partly due to
violating the thermodynamic condition that CV → 0 as T → 0.
For the EOS of explosive products, A, B > 0. A feature of the simplified
form of the JWL EOS, Eq. (16), is that the pressure is positive for all T ≥ 0.
In fact, the domain of the EOS is the entire quadrant T ≥ 0 and P ≥ 0.
Whereas in the (V, e)-plane, the domain is the region above the C0 = 0
isentrope; i.e., eC0 (V ) ≥ 0.
When used for explosive reactants, A > 0 and B < 0. Then even for
C ≥ 0, the pressure on an isentrope goes negative for sufficiently large V .
This restricts the domain of the EOS in the low temperature expansion re-
gion.

Domain of EOS
The domain of an EOS is limited by the conditions T ≥ 0 and P ≥ 0. From
Eq. (3) and Eq. (8b), the isentrope parameters for the reference temperature
Cr and for the T = 0 isentrope C0 are related by

CV Tr = eCr (Vr ) − eC0 (Vr )

Cr − C0
= .
ωvrω

6
The reference state (Vr , er ) determines Cr . Increasing Tr requires decreasing
C0 . If Tr is larger than the value in Eq. (15) then C0 < 0, and from Eq. (13b),
Eer > 0. For isentropes with C < 0, it follows from Eq. (3a) that the pressure
is negative for sufficiently large V . Thus, raising the reference temperature
can restrict the low temperature isotherms in expansion.
The domain of an EOS is also limited by thermodynamic stability, which
requires that the bulk modulus to be positive. From Eq. (14), the isothermal
bulk modulus is
 ∂P 
KT (V, T ) = −V
∂V T
 
= v AR1 exp(−R1 v) + BR2 exp(−R2 v) + ωCV T /V (17)
− ω(ω + 1)v −ω−2 E
er /V0 .

As the temperature of the reference state is raised, E er increases. For suf-

ficiently large Tr , Er > 0. Then at sufficiently low temperature, even with
e
A, B > 0, KT becomes negative in a portion of the expansion region.
Domain limits may have consequences for simulations. For example, P -T
equilibrium of a partly burned explosive will fail to have a solution outside
the EOS domain. This can occur in a hydro simulation when a detonation
wave hits a free surface and the hot partly burn explosive in the reaction
zone expands beyond the initial explosive density.

Heat of detonation
Detonation waves are typically modeled with a single step irreversible re-
action; reactants → products. The heat of detonation is defined as the
change in the heats of formation at standard pressure and temperature,
(P0 , T0 ) = (1 bar, 25 C);

∆Hdet = ∆Hprod − ∆Hreact . (18)

Since the reaction preserves stoichiometry, this can be re-expressed in terms

of the enthalpy (H = e + P V ) of reactants and products;

∆Hdet = Hprod (P0 , T0 ) − Hreact (P0 , T0 ) . (19)

7
 For a high explosive, the initial pressure and temperature are small com-
pared to the values behind the detonation front; i.e., CJ state. To a good
approximation, one can take P0 = 0 and T0 = 0. We use this approximation
to derive a simple expression for ∆Hdet .
For the reactants, the initial enthalpy reduces to Hreact (P0 , T0 ) = e0 where
e0 = ereact (P0 , T0 ). For the products, the JWL EOS goes to an ideal gas in
the limit of low density; i.e., from Eq. (14), V P (V, T ) → ωCv T as V → ∞.
Hence, the limit V → ∞ gives P = 0. From Eq. (13), e(V, T ) → Cv T − ∆e
as V → ∞. Consequently, Hprod (P0 , T0 ) = −∆e. The heat of detonation is
then given by
∆Hdet = −∆e − e0 . (20)
Typically, the energy origin of the reactant EOS is chosen such that e0 = 0.
In this case, ∆Hdet = −∆e. For an exothermic reaction, ∆e is the heat
release.
It is instructive to develop a geometric interpretation for ∆Hdet . Equa-
tion (19) can be re-expressed as

∆Hdet = [Hprod (Vcj , Pcj ) − Hreact (V0 , P0 )] − [Hprod (Vcj , Pcj ) − Hprod (Scj , P0 )] .
(21)
We note that Eq. (8a) implies the temperature on all isentropes goes to 0
as V → ∞ or equivalently P → 0. Hence, Hprod (Scj , P0 ) = Hprod (P0 , T0 ).
(More generally, Hprod (Scj , P0 ) − Hprod (P0 , T0 ) = [T (Scj , P0 ) − T0 ] CP , where
CP is the specific heat at constant pressure.) The Hugoniot jump condition
can be express in terms of enthalpy. For a detonation wave,

Hprod (Vcj , Pcj ) − Hreact (V0 , P0 ) = 12 (V0 + Vcj )(Pcj − P0 ) . (22)

Since (∂P H)S = V , the change in enthalpy along the CJ release isentrope is
Z Pcj
Hprod (Vcj , Pcj ) − Hprod (Scj , P0 ) = dP V (P, Scj ) . (23)
P0

Therefore
Z Pcj
1
∆Hdet = 2
(V0 + Vcj )(Pcj − P0 ) − dP V (P, Scj ) . (24)
P0

In the (V, P )-plane, the first term corresponds to the area between the P -axis
and the segment of the Rayleigh line from P0 to Pcj . Similarly, the second

8
 PBX9502-Shaw
50

40 CJ isentrope
Rayleigh line

P (GPa)
30 CJ

20

10
0
0
0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
3
V (cm /g)

Figure 1: Shaded red area in (V, P )-plane between the CJ release isentrope
and the Raleigh line corresponds to heat of detonation when the release
isentrope is extended to P = 0 as V → ∞.

term corresponds to the area between the P -axis and the portion of the
release isentrope between P0 and Pcj . Thus, ∆Hdet corresponds to the area
between the Rayleigh line and the CJ isentrope; as shown in figure 1.
To get the full heat of detonation out of an explosive, the product have
to expand to a very low pressure and hence a very large specific volume. In
practical applications, the volume of expansion is limited. As a rule of thumb
Vf inal /V0 is 7 to 10. For a high explosive this corresponds to Pf inal ∼ 0.1 GPa.
This may be substituted in Eq. (24) to obtain the effective enthalpy that can
be extracted from an explosive. It is also worth noting that the data available
to calibrate a products EOS extends down to about the same pressure. Thus
the full ∆Hdet depends on extrapolating the EOS, which is likely to introduce
a significant error.

Illustrative examples of domain limitations

To illustrate some of the domain issues, we use the JWL parameters cali-
brated to PBX 9502 in table 1. For the reactants, the choice of ∆e shifts the
energy origin to give the ambient state, P0 = 0 and T0 = 298.15 K (25 C),
with e0 = 0 and the simplified EOS form of Eq. (16). For the products, ∆e
corresponds to the heat of detonation and is often denoted Edet .

9
Table 1: JWL parameters for PBX 9502 [Tarver & McGuire, 2002, table 1].

products reactants units

A 1361.77 6.3207×104 GPa
B 71.99 -4.472 GPa
R1 6.2 11.3 -
R2 2.2 1.13 -
ω 0.5 0.8938 -
ρ0 1.895 1.895 g/cc
V0 1/ρ0 = 0.5277 1/ρ0 = 0.5277 cc/g
e0 0.0 0.0 MJ/kg
∆e 6.9 ∗ V0 -0.246929 MJ/kg
Cv 1.0 × 10−3 ∗ V0 2.487 × 10−3 ∗ V0 MJ/(kg·K)

Table 2: CJ detonation state for JWL products EOS of PBX 9502.

V 0.3967 cc/g
ρ 2.5206 g/cc
e 1.8336 MJ/kg
P 28.0 GPa
D 7.716 km/s

The CJ detonation state is determined by the initial state of the reactants

(V0 , e0 ), the EOS of the products, and two conditions:

1. The state lies on the detonation locus;

ecj = e0 + 12 [P (Vcj , ecj ) + P0 ] · (V0 − Vcj ) . (25)

2. The sonic condition;

P (Vcj , ecj ) − P0
= (ρ c)2 (Vcj , ecj ) . (26)
V0 − Vcj

The CJ state is given in table 2. The isentrope, Eq. (3), through the CJ
state corresponds to the parameter C = 0.884 GPa. The CJ isentrope in

10
the (V, P )-plane is shown in figure 2. Also shown is the CJ isentrope for a
heat of detonation 50 per cent higher. In general, the CJ pressure and the
detonation speed increase with increasing ∆e.
With the choice of reference temperature given by Eq. (15), which leads to
the simplified EOS of Eq. (16), the temperature of the CJ state is Tcj = 2039 K.
The isotherm through the CJ state and the T = 0 isotherm are shown in
figure 3. Several points are noteworthy:

50
CJ isentrope
40
Rayleigh line
1.5∗Edet
P (GPa)

30 CJ

20

10
0
0
0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
3
V (cm /g)

Figure 2: CJ isentrope and Rayleigh line. The initial state and CJ states
are indicated by the circles.
50
T=2039
T=0
40
P (GPa)

30

20

10

0
0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
3
V (cm /g)

Figure 3: CJ isotherm and cold curve for products with simplified EOS,
Eq. (16).

11
 1. Explosive products are usually thought of as gaseous. However, the
T = 0 isotherm is what one would expect for the cold curve of a solid.
This is because the products are at high density and the EOS does not
correspond to that of an ideal gas.

2. At the CJ state, the density is greater than the ambient solid density.
One would expect the specific heat at such a high density to be close to
the classical limit of a solid, which for TATB (the explosive component
of PBX 9502) is 0.0023 MJ/(kg·K). Despite Cv being much smaller, the
CJ temperature is low; expect Tcj = 3000 to 4000 K. This is due to the
high pressure of the “cold curve” at the initial density.

3. The pressure difference between the CJ isotherm and the cold curve
is relatively small. This can cause problems with the P-T equilibrium
mixture EOS of a partly burned explosive.

When the CJ temperature is chosen to be higher, Eqs. (13, 14) rather

than Eq. (16), the low temperature isotherms are qualitatively different.
Figure 4 shows selected isotherms for the EOS when Tcj is chosen to be
3500 K. There are two anomlies. First, the lowest isotherms (T . 500 K) dip
below zero pressure. Second, the isothermal compressibility goes negative,
KT = −V (∂V P )T < 0, along portions of some isotherms. This occurs at
T = 500 K even when the pressure is positive. Both anomalies limit the
PBX 9502 isotherms

T=0
0.4 T=300
T=500
T=1000
P (GPa)

T=2000
0.2 T=3500

-0.2
0 2 4 6 8 10
3
V (cm /g)

Figure 4: Low temperature isotherms in the expansion region (V > V0 ) for

products EOS with Tcj = 3500 K.

12
domain of the EOS. In particular, the room temperature products isotherm
(T ≈ 300 K) is not entirely within the domain.
The anomalous behavior of the isotherms at low temperatures is not
special to the JWL EOS. Rather it is typical of a Mie-Grüneisen form of
EOS with a constant specific heat. Physically, the specific heat and the
entropy should go to zero at T = 0. The Mie-Grüneisen EOS is compatible
with a non-constant specific heat; see for example [Menikoff, 2009].
For the reactants, selected isotherms in the expansion region are shown in
figure 5. Even at 700 K, when the pressure is positive, the expansion region
is limited by the isothermal compressibility going negative. Qualitatively,
the isotherm displays a van der Waal loop. Most EOS models for solids have
similar issues in the expansion region. Physically, the difficulty is that the
tacit assumption of a single phase EOS model breaks down; i.e., sublimation
leads to a mixed solid-vapor phase.
For an explosive, P -T equilibrium is frequently used for partly burned HE.
Since P -T equilibrium requires the component EOS are thermodynamically
stable, these examples show domain issues with the JWL EOS that can
cause a P -T equilibrium solution to fail to exist; typically, for moderate
temperature in expansion.
For a constant volume burn, the thermodynamic state corresponds to
that of the products at (V0 , e0 ). The energy on the CJ isentrope at V0 is
less than e0 by an amount equal to the area of the region in the (V, P )-

2
T=300
1.5 T=500
T=700
T=1000
1
P (GPa)

0.5

-0.5

-1
0.5 1 1.5 2 2.5 3
3
V (cm /g)

Figure 5: Selected isotherms in the expansion region (V > V0 ) for reactant

EOS.

13
 plane between the CJ isentrope and the Rayleigh line over the interval Vcj ≤
V ≤ V0 . Consequently, the constant volume burn temperature and pressure
are greater than the values on the CJ isentrope at V0 , which we denote
by Tcvb and Tcj (V0 ), respectively, and similarly for the pressure. For the
example, Tcvb = 2828 K and Pcvb = 12.2 GPa compared to Tcj (V0 ) = 1768 K
and Pcj (V0 ) = 11.6 GPa. It is anomalous for Tcvb to be so much greater than
the CJ temperature of 2039 K. This indicates that Tcj is too low. However,
raising Tcj would lead to the low temperature anomaly shown in figure 4.
The trade-off in anomalies point to an inherent weakness in the thermal
component of a Mie-Grüneisen EOS with a constant specific heat.
Typically, JWL parameters for explosive products are calibrated to data
for the CJ release isentrope, and for explosive reactants to shock Hugoniot
data. At pressures above the data, the JWL EOS represents an extrapolation
and can loss accuracy. Figure 6 compares the shock locus and the overdriven
detonation locus determined from the JWL EOS with experimental data. An
anomalous feature of the plot is that the loci cross at a pressure just above
the von Neumann spike state of a CJ detonation. Double shock experiments
utilize an HE property, known as shock desensitization, to get higher pressure
reactants data [Aslam, 2011]. These are compatible with a reactants EOS
for which the reactants shock locus remains below the detonation locus.

PBX 9502 PBX 9502

100 10
detonation locus
80 shock locus 8 CJ
Rayleigh line
us (km/s)
P (GPa)

60 6

40 4
VN
CJ
detonation locus
20 2 c0
shock locus
0
0 0
0.3 0.35 0.4 0.45 0.5 0.55 0 1 2 3 4
3 up (km/s)
V (cm /g)

Figure 6: Comparison of shock locus and detonation locus from JWL EOS
with data for PBX 9502. Shock data from [Dick, 1988; Gustavsen, 2006],
and overdrive detonation data from [Tang, 1998; Green, 1985].

14
Exercise for the student
In the section on ‘heat of detonation,’ it was shown that the area between
the CJ release isentrope and the Rayleigh line is the heat of detonation,
which corresponds to the parameter ∆e. With ∆e set to 0 for the explosive
products used in the previous section, figure 7 shows the CJ release isentrope
and the Rayleigh line. Compared to figure 2 the detonation speed (slope of
Rayleigh line) and the CJ pressure are lower. But the area does not appear
to be zero. Explain the paradox.

50

40 CJ isentrope
Rayleigh line
P (GPa)

30
CJ
20

10
0
0
0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
3
V (cm /g)

Figure 7: CJ isentrope and Rayleigh line for ∆e = 0.

15
Resolution of paradox
For the CJ state, the isentrope parameter C is negative. Figure 8 shows
the extension of the isentrope to the expansion region. For V & 1.1 cc/g,
the pressure is negative. For the expansion portion of release isentrope with
P < 0 the signed area between the isentrope and the P -axis is negative.
Even though the negative pressure is small compared to the CJ pressure, the
V interval with P < 0 is much larger than that with P > 0, and the positive
and negative areas cancel.
If the domain of the EOS is limited to be the region of the (V, e)-plane
with P (V, e) > 0, then the effective value of the heat release, −∆Hdet from
Eq. (24) with Pf inal = 0, is greater than 0, and is consistent with figure 7.

4 CJ isentrope
C=-2.38 GPa
3
P (GPa)

Rayleigh line

1
0
0

-1
0.5 1.5 2.5 3.5 4.5
3
V (cm /g)

Figure 8: CJ isentrope for ∆e = 0, expansion region.

16
Acknowledgement
The author thanks Josh Coe for his careful proof reading and suggestions to
improve the clarity of the presentation.

References

T. D. Aslam, “Double shock experiments and modeling of PBX 9502 re-

actants,” presentation at conference Shock Compression of Condensed
Matter (2011). http://www.osti.gov/scitech/servlets/purl/1078411
J. J. Dick, C. A. Forest, J. B. Ramsay, and W. L. Seitz, “The Hugoniot
and shock sensitivity of a plastic-bonded TATB explosive PBX 9502,”
J. Appl. Phys., 63 (1988) pp. 4881–4888.
L. Green, E. Lee, A. Mitchell, and C. Tarver, “The supra-compression of
LX-07, LX-17, PBX-9404, and RX-26-AF and the equation of state of
the detonation products,” in Eighth Symposium on Detonation (1985),
pp. 587–595.
R. L. Gustavsen, S. A. Sheffield, and R. R. Alcon, “Measurements of shock
initiation in the tri-amino-tri-nitro-benzene based explosive PBX 9502:
Wave forms from embedded gauges and comparison of four different ma-
terial lots,” J. Appl. Phys., 99 (2006) 114907/17.
J. W. Kury, E. L. Lee, H. C. Hornig, J. L. McDonnel, D. L. Ornellas, M. Fin-
ger, F. M. Strange and M. L. Wilkins, “Metal Acceleration by Chemical
Explosives,” in Fourth Detonation Symposium (1965), pp. 3–13.
E. L. Lee, H. C. Hornig and J. W. Kury, “Adiabatic Expansion of High
Explosive Detonation Products,” Technical Report LLNL, UCRL-50422
(1968).
R. Menikoff, “Complete EOS for PBX 9502,” technical report, LA-UR-09-
06529, LANL, 2009. http://www.osti.gov/bridge/servlets/purl/
1043481/
R. Menikoff, “Complete Mie-Grüneisen Equation of State,” technical re-
port, LA-UR-12-22592, LANL, 2012. http://www.osti.gov/bridge/
servlets/purl/1045381/

17
P. K. Tang, W. W. Anderson, J. N. Fritz, R. S. Hixson, and J. E. Vorthman,
“A study of the overdriven behaviors of PBX 9501 and PBX 9502,” in
Eleventh Symposium on Detonation (1998), pp. 1058–1064.

C. M. Tarver and E. M. McGuire, “Reactive Flow Modeling of the Interaction

of TATB Detonation Waves with Inert Materials,” in Twelve Detonation
Symposium (2002), pp. 641–649.

18