Critical nuclear reactors operate at constant power. But reactors are not always critical.

It is
necessary, for a reactor to be supercritical in order to start it up or raise its power level while
it must be subcritical to shut it down or reduce power. The study of the behavior of the
neutron population in a noncritical reactor is called reactor kinetics.
Most of the neutrons emitted in fission appear virtually at the instant of fission; these are the
prompt neutrons. A small fraction of the fission neutrons appear long after the fission event;
these are delayed neutrons. The time behavior of a reactor depends upon the various
properties of these two types of neutrons. The prompt neutrons will be considered first.
Prompt Neutron Lifetime
Following their emission, the prompt fission neutrons slow down as the result of elastic and
inelastic collisions with nuclei in the system. In a fast reactor, they do not slow down very
much—only to the order to tens or hundreds of keV—before they are absorbed or leak out of
the core. In thermal reactors, however, most of them succeed in reaching thermal energies
without being absorbed or escaping from the system. As thermal neutrons, they diffuse about
in the reactor; some eventually are absorbed and some leak out. The average time between
the emission of the prompt neutrons and their absorption in a reactor is called the prompt
neutron lifetime and is denoted by lp.
Consider first the value of lp in an infinite thermal reactor. It can be shown theoretically, and
it has been found experimentally, that the time required for a neutron to slow down to thermal
energies is small compared to the time that the neutron spends as a thermal neutron before it
is finally absorbed. The average lifetime of a thermal neutron in an infinite system is called
the mean diffusion time and is given the symbol td- It follows therefore, that
l p ≃ td
for an infinite thermal reactor.
(7.1)

It is an easy matter to calculate t d. A thermal neutron of energy E travels, on the average, an

absorption mean free path λ a (E) before being absorbed. It survives, therefore, the time
t (E)=λ a (E)/v (E)
(7.2)

where v ( E ) is the neutron speed corresponding to the energy E. The mean diffusion time is
then the average value of t(E), i.e.,

t d=t(E)
Since λ a (E)=1/ Σa (E)where Σ a (E) is the macroscopic absorption cross section at
(7.3)
the energy E, Eq. (7.2) can be written as
1
t (E)=
Σ a ( E)v (E) If the absorption is 1/v , as it is at least approximately in thermal reactors,
(7.4)
Σ a ( E )=Σ a ( E 0) v 0 /v (E), where E0 = 0.0253 eV and v 0 = 2200 m/sec, and Eq. (7.4) becomes
 1
t (E)= v
Σ a ( E 0 ) v 0 Thus, t(E) is a constant, independent of E, and Σ́ a= √ π Σ a ( E0 ) T
2 v0
(7.5)

= √
1 π
t d=
Σ a ( E0 ) v 0 2 Σ́ a v T
(7.6)

where use has been made of Eqs. (5.60) and (5.64) with ga(T) = 1. Values of t d for several
moderators are given in Table 7.1.

If the reactor consists of a mixture of fuel and moderator, then Σ́ a = Σ́ a F + Σ́ a M and Eq. (7.6)
becomes

t d= √π ,
2 v T ( Σ́ a F + Σ́ a M )
(7.7)
which can also be written as

t d= √π Σ́ a M
2 v T Σ́ a M Σ́ a F + Σ́ a M

The first factor in this equation is the mean diffusion time for the moderator while the second
factor is equal to 1 — f, where f is the thermal utilization (see Eq. 6.51). Thus, for the
mixture,
t d=t d M (1−f )
(7.8)

Example 7.1 Calculate the prompt neutron lifetime in an infinite, critical thermal reactor
consisting of a homogeneous mixture of 235U and unit density H2O at room temperature.
Solution:
Since the reactor is critical,
k ∞=ηT f =1 ,

so that f =1/ηT . From Table 6.3. ηT =2.065 and therefore f = 0.484. According to Table 7.1, t d
for water is 2.1 ×10−4 sec, and so using Eqs. (7.1) and (7.7),
−4 −4
l p ≃ t d =2.1 ×10 (1−0.484 )=1.08 ×10 sec. [Ans.]

The above results pertain only to thermal reactors. Prompt neutron lifetimes are considerably
shorter in fast reactors than in thermal reactors since the neutrons never have an opportunity
to reach thermal energies. In a fast reactor the value of l p is of the order of 10-7 sec.
Reactor with No Delayed Neutrons
As noted at the beginning of this section, the delayed neutrons play an important role in
reactor kinetics. This is a remarkable fact since so very few fission neutrons are delayed—
less than one percent for thermal fission in 235U. To understand the importance of the delayed
neutrons, it is helpful to consider first the kinetics of a reactor in the absence of delayed
neutrons—that is, assuming that all neutrons are emitted promptly in fission. For the moment,
the discussion will be restricted to the infinite thermal reactor. Since the eventual absorption
of a fission neutron begins a new generation of fission neutrons, it is clear that in the absence
of delayed neutrons, l p is also equal to the time between successive generations of neutrons in
the chain reaction. This time is also called the mean generation time. It follows from the
definition of that the absorption of a neutron from one generation leads to the absorption, l p
sec later, of neutrons in the next generation. Thus, if NF(t) is the number of fissions (which,
of course, is proportional to the number of neutron absorptions) occurring per cm 3/sec at the
time t, then the fission rate l p sec later will be
N F ( t+l p ) =k ∞ N F (t)
The first term in this equation can be expanded as
(7.9)
d N F (t )
N F ( t+l p ) ≃ N F ( t )+l p ,
dt
and when this is substituted back into Eq. (7.9), the result is
d N F ( t ) k ∞−1
≃ N F (t ) .
dt lp
(7.10)
The solution to this equation is

N F ( t ) =N F ( 0 ) exp
( k ∞ −1
lp )t
.where N F ( 0 )is the fission rate at t = 0. Equation (7.11) can also be
(7.1 1)
written as
t
N F ( t ) =N F ( 0 ) e .in which
T

(7.1 2)
lp
T=
k ∞ −1
(7.13)
is called the reactor period—in the absence of delayed neutrons.

Reactor with Delayed Neutrons

In an accurate reactor kinetics calculation, it is necessary to consider in detail the production
and decay of each of the six groups of delayed neutron precursors. The resulting
mathematical analysis is necessarily rather complicated. In order to simplify the discussion, it
will be assumed for the moment that there is only one group of delayed neutrons, which
appear from the decay of a single hypothetical precursor.
Consider then an infinite homogeneous thermal reactor which may or may not be
critical. Since the thermal flux must, be independent of position, the time-dependent diffusion
equation for the thermal neutrons is (see Eq. 5.21)
dn
sT − Σ́ a ϕ T =
dt where sT is the source density of neutrons slowing down into the thermal
(7.14)
energy region, and n is the density of thermal neutron’s. From Eq. (5.61), the one-group
thermal flux is then
2
ϕT = n vT ,
√π
and Eq. (7.14) becomes

sT − Σ́ a ϕ T =
√ π d ϕT
2 v T dt Dividing this equation through by and making use of Eq. (7.6) gives
(7.15)
sT d ϕT √π
−ϕ T =t d . t d=
Σ́ a dt 2 Σ́ a v T Finally, since t d ≃ l p, Eq. (7.16) can also be written as
(7.16)
sT dϕ
−ϕ T =l p T
Σ́ a dt If all of the fission neutrons were prompt, then, in view of the definition of k ∞
(7.17 )
, the source density would be (see Eq. 6.6)
sT =k ∞ × Σ́ a ϕT

However, the fraction of the fission neutrons are delayed and so only the fraction (1−β ) are
prompt. The contribution to sT due to the prompt neutrons is then

sT ( from prompt neutrons )=(1−β)k ∞ Σ́ 0 ϕ T

A delayed neutron slows down very rapidly after it is emitted by its precursor. It follows that
the contribution to the thermal source density from the delayed neutrons is equal to the rate of
decay of the precursor multiplied by the probability p that the delayed neutron escapes
resonance capture while slowing down. Thus
sT ( from prompt neutrons ) =pλC ,

where p is the resonance escape probability, λ is the decay constant of the precursor, and C is
the precursor concentration in atoms/cm3. Combining the above results gives

sT =(1−β) k ∞ Σ́ a ϕ T + pλC .
Introducing this expression into Eq. (7.17) yields the following
(7.18)
equation for the thermal flux:

pλC dϕ
(1−β )k ∞ ϕ T + −ϕ T =l p T .
Σ́ a dt
(7.19)

The equation determining the precursor concentration can be obtained from the
following argument. In Section 6.5 it was shown that the rate at which fission neutrons,
prompt and delayed, are produced is ηT f Σ́ a ϕT =k ∞ Σ́ a ϕ T / p . The rate at which delayed
neutrons are produced is then β k ∞ Σ́ a ϕT / p . Now, since each delayed neutron appears as the
result of the decay of a precursor, it follows that the rate of production of the precursor is also
equal to β k ∞ Σ́ a ϕT / p The precursor decays, of course, at the usual rate λC , so that C is
governed by the equation
dC β k ∞ Σ́ a ϕ T
= −λC
dt p Equations (7.19) and (7.20) are coupled differential equations that must
(7.20)
be solved simultaneously to determine ϕ T
Consider now a specific problem. Suppose that up to the time t = 0 the reactor is critical and
k ∞= 1. A step change is then made in k ∞that the reactor becomes either supercritical or
subcritical. It is required to determine ϕ T as a function of time after t = 0.
This problem can be solved by assuming solutions of the form
ωt
ϕ= A e
(7.21)

and
ωt
C=C 0 e ,
where A and C 0 are constants and ω is a parameter to be determined. Inserting
(7.22)
these functions into Eq. (7.20) gives
β k ∞ Σ́ a A
C 0= .
p(ω + λ)

When C 0 and Eqs. (7.21) and (7.22) are introduced into Eq. (7.19), it is found that the
constant A cancels, which leaves
 λβ k ∞
( 1−β ) k ∞ + −1=ω l p
ω+ λ
It is convenient to rewrite this equation in the following form:
k ∞ −1 ωlp ω β
= +
k∞ 1+ω l p 1+ ω l p ω+ λ The left-hand side of Eq. (7.23) is known as the reactivity, in
(7.23)
this case of the infinite reactor, and is denoted by the symbol ρ ; thus
k ∞−1
ρ= .
k ∞ For a finite reactor, the reactivity is defined as
(7.24)
k−1
ρ=
k In terms of the reactivity, Eq. (7.23) may be written as
(7.25)
ωlp ω β
ρ= +
1+ω l p 1+ ω l p ω+ λ This result is known as the reactivity equation, for one group of
(7.26)
delayed neutrons. Before considering the significance of the reactivity equation, it should be
noted that if a reactor is supercritical, then k > 1 and ρ is positive. In this situation, the reactor
is said to have positive reactivity. On the other hand, when the reactor is subcritical, k < 1, ρ
is negative, and the reactor is said to have negative reactivity. It should be noted that from Eq.
(7.25), p is restricted to values in the range — ∞ < ρ < 1.