13.3 Chemical Diagenesis
The relationship between porosity and permeability of
the rock matrix (fractures are neglected here) is an important but very complex topic. This relationship initially depends, among other factors, on the size of the
particles and the grain size distribution or sorting of
the sediment which control the geometry of individual
pores and their connection with neighboring pores. F or
example, well sorted sands of a certain grain size have
higher porosities and permeabilities than poorly sorted
sands with the same medium grain size. These differences are enhanced and modified by compaction and
various types of cementation. Authigenic clay minerals
in the pores reduce the permeability of sandstones.
The two diagrams in Figure 13.21 show the overall
trend of a rapidly decreasing permeability with the loss
of porosity. F or example, a decrease in the porosity of
sandstones from 20% to 10% may bring about a re duction in permeability by two to three orders of magnitude. Grain overgrowths and beginning ofpore-filling
first narrow and then close pore throats and fine pores,
whereas wider pores, vugs and molds still remain
open. With prograding cementation, however, the residual pore space can become very irregular. For this
reason, certain pore types have been defined (Fig.
13.21c) which allow better estimates for the variable
relationship between porosity and permeability (Fig.
13.21 b). In many carbonate rocks (e.g. reef-type structures) this relationship is even less clear.
Mudrocks maintain a relatively high total porosity
of 5-10% at burial depths of about 5 km, but their
permeabilities are extremely low at this state of compaction.
The theoretical background of these problems is treated in
detail, e.g., by Dullien (1992) and Howard (1992). Further
information is found in textbooks on hydrocarbon exploration and production (cf. Chap. 14). A classification ofpore
types in relation to reservoir characteristics has been proposed by Ehrlich et a!. (1991).
Plug sampies of sandstones (Middle Jurassic, burial depth
around 4000 m) recovered from two oilfield wells in the
North Sea, one from the oil-filled crest and the other from
the water-filled flank, have shown a much higher permeability of the oil-filled pore system (Ehrenberg and Boassen
1993). This is caused by differences in the pore-size distribution and intergranular macroporosity, whereas the bulk composition of the sandstones, their degree of diagenetic alteration and porosity are about the same. In the lower-permeability weil, the illite is more widely distributed in the pore
system than in the oil-filled pores where illite grew only in
the residual water-filled pores.
The great differences in the evolution of porosity and permeability ofvarious carbonate rocks is not only a function of
the diagenetic realms but is also largely controlled by the
nature ofthe carbonate sediments (e.g. McIlreath and Morrow 1990). While the behavior of chalks under increasing
sediment load is relatively weil predictable (e.g. Brasher and
Vagle 1996), that of reefs is highly variable. In many cases
fractures only provide rock permeability. Dolomitization can
both increase and reduce porosity and permeability. However, dolomites lose porosity more slowly than limestones
683
during deep buria!. They are therefore the principal reservoir
rocks in many ancient carbonate rocks (e.g. Su 1995). Salier
and Henderson (1998) have described a case study from the
Permian of Texas, where the porosity and permeability of
dolomites are strongly reduced due to the effect of
hypersaline waters at the platform interior, while at the
basinward platform margin high values have been maintained
(Fig. 13.21 b). Mowers and Budd (1996) have determined the
specific surface of the pore system of dolomites to quantity
porosity and permeability reduction resulting from calcite
cementation. Whereas limited calcite cementation in moldic
pores (Fig. 13.22e) generated little change in permeability,
abundant late calcite cement, occupying both intercrystalline
and moldic pores, caused a 10 to 1000-fold decrease in permeability.
Many sandstones exhibit secondary porosity in contrast to shales, which commonly maintain their state of
compaction and cementation they have achieved during diagenesis. Secondary porosity is frequently observed in thin sections as voids after the leaching of
clastic grains (e.g., feldspar, bioclastic carbonate) or
carbonate cement (Fig. 1322). This dissolution process commonly takes place in a hydrologically open
system with fluid flow. This also applies to pervasive
dolomitization which can create substantial secondary
porosity. Dissolution, recrystallization, and. transformation of mineral phases in place can in fact change
the distribution and geometry of pores, but these proces ses do not significantly enhance the overall porosity. The same is true of fracturing by overpressure
(Sect. 13.2) or tectonic movements. Dolomitization of
calcium carbonate can create both an increase in porosity by 12% (cf. Fig. 13.19, No. 8) or a porosity reduction. The second possibility applies to diagenetic systems in which not only Mg++, but in addition C0 3 "" can
react with CaC0 3 and thus enhance the mineral volurne. It appears that in many cases both reactions take
place and therefore do not allow any predictions for
the porosity evolution.
One of the most important processes for providing CO 2 for
the dissolution of carbonate or dolomitization in deep-burial
diagenesis appears to be the maturation ofkerogen (Schmidt
and McDonald 1979; Bj0rlykke 1988). It is, however, only
the humic type ofkerogen with a high OIC ratio, which enables substantial leaching of carbonate. Other mechanisms
for the dissolution of carbonate and possibly other types of
cement are the circulation of acid meteoric water and convective flow (Fig. 13.l6f).
For the migration and storage of hydrocarbons it is
crucial that sufficient primary (or secondary) porosity
and permeability are preserved at eritieal burial depths
(approximately 3 to 5 km). Thus, the time and depth
range of pore spaee reduction (or seeondary increase in
pore spaee) control the quality of sands as reservoir
rocks. Onee the pores have beeome filled with oil (Fig.
13.22c), the residual water is no longer effective as
means ofmass transfer. Then the reservoir can more or
less maintain its porosity and permeability, whereas
The relationship between porosity and permeability of
the rock matrix (fractures are neglected here) is an important but very complex topic. This relationship initially depends, among other factors, on the size of the
particles and the grain size distribution or sorting of
the sediment which control the geometry of individual
pores and their connection with neighboring pores. F or
example, well sorted sands of a certain grain size have
higher porosities and permeabilities than poorly sorted
sands with the same medium grain size. These differences are enhanced and modified by compaction and
various types of cementation. Authigenic clay minerals
in the pores reduce the permeability of sandstones.
The two diagrams in Figure 13.21 show the overall
trend of a rapidly decreasing permeability with the loss
of porosity. F or example, a decrease in the porosity of
sandstones from 20% to 10% may bring about a re duction in permeability by two to three orders of magnitude. Grain overgrowths and beginning ofpore-filling
first narrow and then close pore throats and fine pores,
whereas wider pores, vugs and molds still remain
open. With prograding cementation, however, the residual pore space can become very irregular. For this
reason, certain pore types have been defined (Fig.
13.21c) which allow better estimates for the variable
relationship between porosity and permeability (Fig.
13.21 b). In many carbonate rocks (e.g. reef-type structures) this relationship is even less clear.
Mudrocks maintain a relatively high total porosity
of 5-10% at burial depths of about 5 km, but their
permeabilities are extremely low at this state of compaction.
The theoretical background of these problems is treated in
detail, e.g., by Dullien (1992) and Howard (1992). Further
information is found in textbooks on hydrocarbon exploration and production (cf. Chap. 14). A classification ofpore
types in relation to reservoir characteristics has been proposed by Ehrlich et a!. (1991).
Plug sampies of sandstones (Middle Jurassic, burial depth
around 4000 m) recovered from two oilfield wells in the
North Sea, one from the oil-filled crest and the other from
the water-filled flank, have shown a much higher permeability of the oil-filled pore system (Ehrenberg and Boassen
1993). This is caused by differences in the pore-size distribution and intergranular macroporosity, whereas the bulk composition of the sandstones, their degree of diagenetic alteration and porosity are about the same. In the lower-permeability weil, the illite is more widely distributed in the pore
system than in the oil-filled pores where illite grew only in
the residual water-filled pores.
The great differences in the evolution of porosity and permeability ofvarious carbonate rocks is not only a function of
the diagenetic realms but is also largely controlled by the
nature ofthe carbonate sediments (e.g. McIlreath and Morrow 1990). While the behavior of chalks under increasing
sediment load is relatively weil predictable (e.g. Brasher and
Vagle 1996), that of reefs is highly variable. In many cases
fractures only provide rock permeability. Dolomitization can
both increase and reduce porosity and permeability. However, dolomites lose porosity more slowly than limestones
683
during deep buria!. They are therefore the principal reservoir
rocks in many ancient carbonate rocks (e.g. Su 1995). Salier
and Henderson (1998) have described a case study from the
Permian of Texas, where the porosity and permeability of
dolomites are strongly reduced due to the effect of
hypersaline waters at the platform interior, while at the
basinward platform margin high values have been maintained
(Fig. 13.21 b). Mowers and Budd (1996) have determined the
specific surface of the pore system of dolomites to quantity
porosity and permeability reduction resulting from calcite
cementation. Whereas limited calcite cementation in moldic
pores (Fig. 13.22e) generated little change in permeability,
abundant late calcite cement, occupying both intercrystalline
and moldic pores, caused a 10 to 1000-fold decrease in permeability.
Many sandstones exhibit secondary porosity in contrast to shales, which commonly maintain their state of
compaction and cementation they have achieved during diagenesis. Secondary porosity is frequently observed in thin sections as voids after the leaching of
clastic grains (e.g., feldspar, bioclastic carbonate) or
carbonate cement (Fig. 1322). This dissolution process commonly takes place in a hydrologically open
system with fluid flow. This also applies to pervasive
dolomitization which can create substantial secondary
porosity. Dissolution, recrystallization, and. transformation of mineral phases in place can in fact change
the distribution and geometry of pores, but these proces ses do not significantly enhance the overall porosity. The same is true of fracturing by overpressure
(Sect. 13.2) or tectonic movements. Dolomitization of
calcium carbonate can create both an increase in porosity by 12% (cf. Fig. 13.19, No. 8) or a porosity reduction. The second possibility applies to diagenetic systems in which not only Mg++, but in addition C0 3 "" can
react with CaC0 3 and thus enhance the mineral volurne. It appears that in many cases both reactions take
place and therefore do not allow any predictions for
the porosity evolution.
One of the most important processes for providing CO 2 for
the dissolution of carbonate or dolomitization in deep-burial
diagenesis appears to be the maturation ofkerogen (Schmidt
and McDonald 1979; Bj0rlykke 1988). It is, however, only
the humic type ofkerogen with a high OIC ratio, which enables substantial leaching of carbonate. Other mechanisms
for the dissolution of carbonate and possibly other types of
cement are the circulation of acid meteoric water and convective flow (Fig. 13.l6f).
For the migration and storage of hydrocarbons it is
crucial that sufficient primary (or secondary) porosity
and permeability are preserved at eritieal burial depths
(approximately 3 to 5 km). Thus, the time and depth
range of pore spaee reduction (or seeondary increase in
pore spaee) control the quality of sands as reservoir
rocks. Onee the pores have beeome filled with oil (Fig.
13.22c), the residual water is no longer effective as
means ofmass transfer. Then the reservoir can more or
less maintain its porosity and permeability, whereas