11 hampel joint project us-g 7

Joint Project on Constitutive Models:

Conclusions from phases I – III and introduction of project WEIMOS

Washington, DCSeptember 7-9, 2016

Dr. Andreas Hampel

7th US/German Workshop on Salt Repository Research, Design, and Operation

Dr. Andreas Hampel 2 / 187th US/German Workshop on Salt Repository Research, Design, and Operation

Washington DC, Sept. 07-09, 2016 Joint Project on Constitutive Models

In 2000, Udo Hunsche (BGR): Let’s perform a comparison of the models!

JP I (2004-2006): Development of a procedure

1. Perform systematic lab test series with certain salt types investigate (well-controlled in the lab) the various deformation phenomena

and their dependencies on in-situ relevant boundary conditions (~ seq, s3, T, de/dt)

2. Recalculate the lab tests determine a unique set of parameter values for each type of salt check the ability of the models to describe the various deformation phenomena

and the dependencies=> constitutive model is prepared and valid to model various potential in-situ situations

3. Simulate real underground structures that display the considered phenomena check the suitability of the models compare the results with each other and with in-situ measurements

History (≈ 1980s – 1990s):

Constitutive models were developed by several groups (mostly independently)based on numerous lab tests with rock salt and in-situ measurements in salt

-> How?



Joint Project series

Joint Project(funded by)

Period

II(BMBF)

2007 – 2010Suitability to perform 3-D simulations of real underground structures, incl. temporal extrapolations, calculation of permeability in the DRZ of a highly-loaded pillar

III(BMWi)

2010 – 2016

Modeling of the• temperature dependence of deformation (-> HLW)

a) rock salt from Asse mine (domal salt),b) rock salt from WIPP (bedded salt)

• damage reduction and healing (-> long-term integrity)a) rock salt from Asse mine

I(BMBF)

2004 – 2006Modeling of the basic deformation phenomena in rock salt: transient & steady-state creep, evolution of damage & dilatancy, creep failure, post-failure behavior, residual strength

Main objectives: document, investigate and compare constitutive models and modeling procedures (par. determ., numerical calc.)



Joint Project I

Partner Constitutive Model Num. Program

BGR Hannover CDM JIFE

A. Hampel, Mainz CDM FLAC3D

IfG Leipzig Günther/Salzer Model, FLAC, FLAC3DMinkley Model

KIT Karlsruhe KIT Model ADINA

Leibniz Universität Hannover (LUH), Lubby-MDCF ModelFLAC3D

Technische Universität Clausthal (TUC), Lux/Wolters Model FLAC3D

Technische Universität Braunschweig (TUBS) TUBSsalt FLAC3D,ANSYS

Sandia National Laboratories, Albuquerque MD Model (creep) Sierra Mechanicsand Carlsbad, NM, USA Code Suite

II



Recalculate a systematic series of lab tests with one salt type (example: clean salt from WIPP)

Creep tests at different temperatures and stress differences

Strength tests at different confining stresses

All tests were recalculated with a unique set of parameter values

, temperatures, and deformation rates



the unique set of parameter valuesapplied stresses

TUCtest1

TUCtest2

Perform and recalculate systematic lab tests with one salt type (example: Speisesalz from Asse mine)

JP III: damage & dilatancy reduction and healing

dilatancy calculated with

individual parameter values

dilatancy calculated with



JP I: Simulation of simple example structures

a) cylindrical specimen b) single drift c) room-pillar system



JP II: a) Discretization study and b) 3-D simulation of an Angersdorf mine section

6x6 20x10

-6

-5

-4

-3

-2

-1

0

1

0 2 4 6 8 10 12 14 16 18 20

Pillar thickness [m]

Min

imum

prin

cipa

l str

ess

[MPa

]

In-situ frac test

AH simulation

IfG-GS simulation

IfG-Mi simulation

IUB simulation

TUC simulation

s3 [MPa]

d

example:TUC



JP III: Simulations of in-situ temperature influence and damage reduction & healing

a.1) IFC:isothermal free

convergence

a.2) HFCP:heated borehole

(Asse mine)

b) Bulkhead(Asse mine)

6

7

8

9

10

11

12

13

14

0 0,01 0,02 0,03 0,04 0,05 0,06 0,07

heig

ht a

long

bor

ehol

e w

all [

m]

displacement [m]

ECN data

Hampel

Sandia

IfG-GS

LUH

KIT

TUC

TUBS

heated

zone

t_therm = 19 d (end of heating)HFCP simulation



JP III: Simulations of in-situ temperature influence

c.1) Room D (WIPP: bedded salt)unheated -> at natural rock temperature

Simulated with unique parameter values fromrecalculations of lab tests with WIPP salt



JP III: Simulations of in-situ temperature influence and damage reduction & healing

c.2) Room B (WIPP, bedded salt)heated after 354 days

Simulated with unique parameter valuesfrom lab tests with WIPP salt

heated

heated



Conclusions from Joint Project phases I – III

We have developed a procedure to compare different constitutive models

performance and recalculation of systematic lab tests

simulations of typical in-situ situations

Constitutive models with a unique salt-type-specific set of parameter values

are appropriate to describe various thermo-mechanical deformation phenomena and their dependencies on in-situ relevant boundary conditions in a wide rangetransient and steady-state creep, evolution of damage and dilatancy,creep failure and short-term strength, post-failure behavior and residual strength,(damage and dilatancy reduction and healing)

are prepared to model various in-situ situations in rock salt

some aspects need more experimental investigation and further development of the modeling

-> new Joint Project WEIMOS

~ seq, s3, T, de/dt



WEIMOS: Detailed investigation and further development of the modeling

convergence rates of Room B

I. Deformation at small, in-situ relevant deviatoric stresses

steady-state creep rates of WIPP salt

creep test with WIPP saltat 4 and 2 MPa most in-situ conditions

WEIMOS: triaxial creep tests at RT and small Ds very stable boundary conditions strain measurement with very high resolution-> WIPP salt




II. Stress and temperature dependence of damage reduction and healing

until now: only two TUC healing tests (Asse-Speisesalz)with high resolution dilatancy measurements

Test 1 Test 2

?? healing rate ~ seq, s3, T, evol ??

=>

bulkhead simulation

WEIMOS: more healing tests at different T and Ds dilatancy measurements with very high resolution-> WIPP salt




III. Deformation resulting from tensile stresses

Tensile stressess3 > 0

Tensile stressess3 > 0

IfG-G/S

strong temperature influence on dilatancy (WIPP Rooms D and B)-> influence of tensile stresses

Vergleich

Room D – Room B300 K – erhitzt

t = 1354 Tage

(Room B)

0,0%

0,2%

0,4%

0,6%

0,8%

1,0%

1,2%

0 0,5 1 1,5 2

Dila

tanz

dur

ch S

chäd

igun

g

Spurlänge [m]

Spur T2 horizontal nach rechts

Room DRoom B

DRZ

WEIMOS:started with basic calculation studies:

bending beam Brazilian test




IV. Modeling of layer boundaries and interfaces

Munson et al. (1990): Sandia Report SAND89-2671

WEIMOS:

• influence on convergence ?(e.g. WIPP: sliding on clay seams)

• influence on damage and dilatancy in the DRZ ?

o lab: shear tests on layered salt specimens / blocks (Sandia / RESPEC ?)

o in-situ experiments ?

improved understanding of shear stresses & strains

improved rock-mechanical modeling

reduced uncertainties



Demonstration of improved modeling in WEIMOS

V. Virtual Demonstrator

Simulation of a complex model to demonstrate the improved modeling of the various investigated phenomena

small deviatoric stresses

damage reduction and healing

influence of interfaces/layer boundaries

influence of e.g. thermally inducedtensile stresses

Simulation: step 1: open drift

step 2: installation of dam & backfill

step 3: post-operational phase and long-term behavior

rock salt

sealing systemcrushed salt

crushed salt interface

e.g. main drift



Summary: Joint Project WEIMOS (April 2016 – March 2019)

Identified needs for further development:

1. Deformation behavior at small deviatoric stresses

2. Deformation behavior resulting from tensile stresses

3. Influence of inhomogeneities (layer boundaries, interfaces) on deformation

4. Influence of temperature and stress state on damage reduction

“Further Development and Qualification of the Rock Mechanical Modeling for the Final HLW Disposal in Rock Salt”

Procedure:

Laboratory tests, microstructural investigations, optional: in-situ measurements.

Recalculations of the lab tests, simulations of basic examples and real in-situ structures.

Further development of the rock mechanical modeling.

Comparison of results, validation and qualification of the models and modeling procedures.

Main goal: Improved analysis and proof of long-term integrity of the geological barrier rock salt (CRZ: Containment providing Rock Zone)

Technology

11 hampel joint project us-g 7