CMWR, June 4th - 7th 2018 www.hydrosys.uni-stuttgart.de

Institute for Modelling Hydraulic and Environmental SystemsDepartment for Hydromechanics and Modelling of Hydrosystems

Modeling field-scale applications ofmicrobially induced calcium carbonateprecipitationJohannes Hommel∗, Adrienne J. Phillips, Robin Gerlach, Alfred B. Cunningham, RainerHelmig∗, Holger Class∗∗University of Stuttgart, Montana State University,

Motivation and SetupIn 2014, the first field-scale application of MICP to mitigate leakage wasperformed in a fractured sandstone formation at a depth of 340.8 m [3].

Figure 1: Schematic cross section of the well and formation for the field-scaleMICP application (left) and the radial simulation domain and grid with theinitial and boundary conditions (right).

Three simulation domains of various sizes were investigated to asses theinfluence of the domain size and the high-permeable layer extent:

• small: 2.4 m×2.4 m domain with a 1.6 m radius high-permeable layer

• large: 8 m×8 m domain with a 4 m radius high-permeable layer

• extended: 8 m×50 m domain with a 2 m radius high-permeable layer

Model ConceptThe MICP model is discussed in detail in Hommel et al. [2]. It includesreactive two-phase multi-component transport including two solid phases.

solid phases:∂

∂t(φλρλ) = qλreactions

solutes:∑α

[∂

∂t(φραxκαSα) +∇· (ραxκαvα)−∇· (ραDpm,α∇xκα)

]= qκreactions

• processes determining the distribution of biomass:- growth: rgrowth = µ ρbiofilm φbiofilm

CO2w

CO2w +KO2

C substratew

C substratew +Ksubstrate

,

- decay: rdecay = kdecay ρbiofilm φbiofilm,- attachment: rattachment = (ca,1 φbiofilm + ca,2) Sw φ C bacteria

w ,- detachment: rdetachment = cd,1 (Sw φ |∇pw|)0.58 + cd,2 µ,

• (bio-) chemical reactions:- microbially catalyzed ureolysis: CO(NH2)2 + 2H2O

urease−−−→ 2NH3 + H2CO3,rurea = kurease kub ρbiofilm φbiofilm

murea

murea+Kurea.

- influence of NH3 on the pH: NH3 + H+ ←−→ NH +4 ⇒ increase in pH,

- precipitation (and dissolution) of calcite: Ca2+ + CO 2 –3 ←−→ CaCO3 ↓ ,

rprecipitation = kprecipitation Asw (Ω− 1)nprecipitation,

which is depended on the calcite saturation state Ω = [Ca2+][CO 2−3 ]

Kspand the water-solid

surface area Asw.

• clogging: φ = φ0 − φcalcite − φbiofilm ⇒ K = K0

(φ−φcrit

φ0−φcrit

)3

Figure 2: Model development prior to the field-scale modeling [1].

ResultsField observations [3]:• significant decrease in injectivity ≈45 h after the first injections

• significant decrease in pressure decay after well shut in

• CaCO3 detected in cores 1.8 m above the injection point

• total of 24 mineralization and 6 microbial injections during 4 days

Model results [1]:• significant decrease in permeability close to the well for all scenarios

investigated

• significant increase in injection pressure ≈48 h after the first injections

• CaCO3 reaches about 1 m above the injection point

• the most recent model calibration (to 1D column results, [2])underestimates the permeability reduction

1 2 3 4 5 6 7 80

0.3

0.6

0.9

1.2

1.5

1.8·10−12

Radius [m]

Per

mea

bilit

y[m

2 ]

K0 8 m×8 m2014 8 m×8 m2014 2.4 m×2.4 m2018 8 m×50 m, 2014 param.2018 8 m×50 m, [2] param.

5 15 25 35 45 55 65 753.2

4.2

5.2

6.2

·106

Time [h]In

ject

ion

Pre

ssur

e[P

a]

Pressure boundary condition at r = 50 m

Figure 3: Final and initial permeability along the radius through thehigh-permeable layer as predicted by simulations for various grids, domain sizes,and injection strategies (left). Injection pressure for the 2018 simulations andthe pressure boundary condition (at 50 m) over time (right). Note that theinitial permeability on the left is only shown for the “large” 8 m×8 m scenario.

−342 −341 −3400

0.5

1

1.5

Depth [m]

Cal

cite

Vol

.F

rac.

[%]

1 2 3 4 5 6 7 80.1

0.11

0.12

Radius [m]

Por

osit

y[-

]

2014 8 m×8 m2014 2.4 m×2.4 m2018 8 m×50 m, 2014 param.2018 8 m×50 m, [2] param.

Figure 4: Calcite volume fractions at the inner radius over depth predicted bysimulations on various simulation domain sizes (left) and radial porositydistribution at the height of the high-permeable layer (right).

More detailed results and discussions will be published in Cunningham et al. [1].

Literature[1] Cunningham, A. B., Class, H., Ebigbo, A., Gerlach, R., Phillips, A. J., and Hommel, J.

(2018). Field-scale modeling of microbially induced calcite precipitation. ComputationalGeosciences. submitted.

[2] Hommel, J., Lauchnor, E. G., Phillips, A. J., Gerlach, R., Cunningham, A. B., Helmig, R.,Ebigbo, A., and Class, H. (2015). A revised model for microbially induced calciteprecipitation: Improvements and new insights based on recent experiments. WaterResources Research, 51(5):3695–3715.

[3] Phillips, A. J., Cunningham, A. B., Gerlach, R., Hiebert, R., Hwang, C., Lomans, B. P.,Westrich, J., Mantilla, C., Kirksey, J., Esposito, R., and Spangler, L. H. (2016). FractureSealing with Microbially-Induced Calcium Carbonate Precipitation: A Field Study.Environmental Science & Technology, 50:4111–4117.

Institute for Modelling Hydraulic and Environmental SystemsDepartment for Hydromechanics and Modelling of Hydrosystems

Modeling field-scale applications ofmicrobially induced calcium carbonateprecipitationJohannes Hommel∗, Adrienne J. Phillips, Robin Gerlach, Alfred B. Cunningham, RainerHelmig∗, Holger Class∗∗University of Stuttgart, Montana State University,