Chapter 10 NMR
Chapter 10 NMR
Chapter 10 NMR
Applications of NMR
NMR provides
Mineralogy-independent
porosity
T2 distributions
Use of T2 distributions
Pore-size distributions
Irreducible water
saturations
Permeability
NMR Principle
Gives unprecedent information about both porosity and pore size
distribution that is used to successfully derive continuous permeability
logs, notably in siliciclastic formations.
The physical principle called nuclear magnetic resonance refers to the
response of atom nuclei to externally applied magnetic fields.
Many atom nuclei have a magnetic moment, i.e. behave like tiny spinning
magnets. These spinning nuclei can interact with a magnetic field,
producing detectable signals. For most elements, nevertheless, the
measured signals are weak, but hydrogen that is abundant in both water
and hydrocarbons contained in the pore space of rocks, has a relatively
large magnetic moment.
The initial NMR signal amplitude is thus proportional to porosity; its overall decay
is the sum of the individual decays, which reflects pore size distribution. Separating
out ranges of T2 values by a mathematical inversion process produces the T2
distribution curve.
NMR - Theory
The area under the curve represents the
porosity and the curve shape the distribution
of pore sizes. This inversion process normally
requires stacking, in order to improve the
signal-to-noise ratio, which slightly degrades
the vertical resolution
NMR
Principles
NMR counts # hydrogen protons
aligned with an external magnetic
field (EMF)
Monitors the dephasing of protonss
spin with time
# protons aligned with EMF is
proportional to porosity
Change of amplitude of subsequent
echoes, known as the transverse
relaxation, is characterized by a
exponential decay with a
characteristic time (T2)
the initial NMR signal amplitude
depends only on the H proton density
in the pore space.
To measure T2 decay, a series of RF
pulses are used to rephase the
transverse magnetizaton.
These RF pulses recall the NMR signal
(echo)
1/T2=(1/T2bulk)+(/T2surf).(S/V)+(1/T2D)
Where:
T2bulk = relaxation time of the bulk fluids in the pore space
T2surf = relaxation time of the fluid in a few molecular layers next to the pore surface
= thickness of the surface fluid layer
S, V = pore surface and pore volume
T2D = relaxation time due to diffusion
If diffusion effects are negligible, and because T2bulk for water is on the order of seconds,
while the observed T2 is only a few hundred milliseconds or less, the above equation can be
approximate by:
1/T2= s (S/V)
Where:
s = surface relaxivity (0.003-0.03 cm/s) for clastics and (<0.003 cm/s) for carbonates.
The T2 spectrum, derived from the echo data, measures the porosity fraction
associated with each T2.
The integral of the T2 spectrum is the NMR porosity.
The advantages of T2 spectrum is the partition into fast decaying fluids (bound
water; 4-32 ms for clastics) and free fluids (> 32 ms, for clastics).
T2 time separating moveable and irreducible water is referred as T2 cut-off time.
kNMR= C. T2gm2.4 ( Kenyon et al. 1988); overestimate k in HC zones
K=(100 /C)4 (Smov/Swir)2 (Coates et at 1991); unaffected by HC
De-phasing of nuclei
Pulse Echo
Relaxation theory
Relaxation times
Decay distributions
Schlumberger CMR
Timur (1968)
K 1/2 = 2.25/Swirr
work OK in clean sandstones only
Free fluid
porosity
NMR in Carbonates
The interpretation model assuming that, in
water-saturated reservoir rocks, the T2 and poresize distributions are directly related.
It explains why NMR T2 curves are successfully
used to characterize sandstones containing mixed
pore-size distributions.
However, there is some concern within the oil
industry that NMR does not work as well in
carbonate reservoirs. First, NMR responses in
carbonates differ from those in sandstones.
NMR in Carbonates
Pore surfaces in carbonates are not equally effective in relaxing
hydrogen nuclei and carbonates are about three times less efficient
in relaxing the nuclear magnetism than sandstones.
For carbonates, relaxation times therefore tend to be three times
longer and a 100 msec cutoff was proposed for free-fluid porosity.
This cutoff value has often to be locally adapted. For instance, in
the Thamama formations of Abu Dhabi, permeable grainstones
could be distinguished from lower permeability packstones and
mudstones with a 225 msec cutoff.
But, while carbonate formations contain mixed pore-size
distributions, e.g., intergranular porosity and vugs, NMR logging
data in these formations nevertheless frequently yield unimodal T2
distributions, which often results in inconsistent T2 cutoff values to
distinguish bound and free fluids, and leads to unreliable
permeability predictions.