Durability Index Testing Procedure Manual
Durability Index Testing Procedure Manual
Durability Index Testing Procedure Manual
Procedure Manual
2018
(Ver 4.5.1, April 2018)
CONTENTS
PREFACE ............................................................................................................................................... 1
PART 1: STANDARD PROCEDURE FOR PREPARATION OF TEST SPECIMENS ......................... 2
1.1 SCOPE ..................................................................................................................................... 2
1.2 TEST SPECIMENS .................................................................................................................. 2
1.3 APPARATUS ............................................................................................................................ 2
1.4 PREPARATION OF SPECIMENS FROM CUBES .................................................................. 3
1.5 PREPARATION OF SPECIMENS FROM SITE ELEMENTS .................................................. 4
1.6 REFERENCES ......................................................................................................................... 4
1.7 REVISIONS .............................................................................................................................. 5
PART 2: STANDARD PROCEDURE FOR OXYGEN PERMEABILITY TEST .................................... 6
2.1 SCOPE ..................................................................................................................................... 6
2.2 APPARATUS ............................................................................................................................ 6
2.3TEST SPECIMENS ................................................................................................................... 8
2.4 CONDITIONING OF SPECIMENS........................................................................................... 8
2.5 TESTING OF SPECIMENS...................................................................................................... 8
2.6 CALCULATIONS .................................................................................................................... 10
2.7 REPORTING .......................................................................................................................... 13
2.8 REFERENCES ....................................................................................................................... 13
2.9 REVISIONS ............................................................................................................................ 14
APPENDIX A (Part 2) ................................................................................................................... 16
APPENDIX B (Part 2) ................................................................................................................... 17
PART 3: STANDARD PROCEDURE FOR WATER SORPTIVITY AND POROSITY TEST .............. 22
3.1 SCOPE ................................................................................................................................... 22
3.2 APPARATUS .......................................................................................................................... 22
3.3 TEST SPECIMENS ................................................................................................................ 23
3.4 CONDITIONING OF SPECIMENS......................................................................................... 23
3.5 TESTING OF SPECIMENS.................................................................................................... 24
3.6 CALCULATIONS .................................................................................................................... 26
3.7 REPORTING .......................................................................................................................... 29
3.8 REFERENCES ....................................................................................................................... 30
3.9 REVISIONS ............................................................................................................................ 30
PART 4: STANDARD PROCEDURE FOR CHLORIDE CONDUCTIVITY TEST ............................... 32
4.1 SCOPE ................................................................................................................................... 32
4.2 APPARATUS .......................................................................................................................... 32
4.3 PREPARATION OF THE CHEMICAL SOLUTION (5M NaCl) ............................................... 34
4.4 TEST SPECIMENS ................................................................................................................ 34
4.5 CONDITIONING OF SPECIMENS......................................................................................... 34
4.6TESTING OF SPECIMENS..................................................................................................... 35
4.7 CALCULATIONS .................................................................................................................... 37
4.8 REPORTING .......................................................................................................................... 38
4.9 REFERENCES ....................................................................................................................... 38
4.10 REVISIONS .......................................................................................................................... 39
APPENDIX A (Part 4) ................................................................................................................... 41
ii
PREFACE
This version of the Durability Index Testing Procedure Manual contains a number of
new aspects. The most important is that Parts 1, 2, and 4 have been formalised as
SANS Test Methods: SANS 3001-CO3-1:2015, SANS 3001-CO3-2:2015, and SANS
3001-CO3-3:2015, respectively. Part 3, the Water Sorptivity and Porosity Test
Procedure, must still be formalised through the SABS processes. It is important to
note that if there are discrepancies between this Manual and the SANS Tests, the
SANS Tests will govern. However, the SANS tests will themselves undergo a review
for improvements and clarifications in the future (2018).
There are no substantive changes in the test methods compared with earlier
versions, but the wording has been improved to make them clearer, tighter
restrictions have been placed on the CCI test (time for taking the measurements),
the figures have been improved, and other detail added. Illustrative calculations
have also been added as an Appendix in the OPI Test Method, and precision data
have been added where appropriate.
The other change to note is the inclusion of porosity as an important parameter in
the water sorptivity test. While determining porosity has always been a part of the
test, this parameter is now being realised as important in its own right, and water
sorptivity cannot be viewed in isolation of porosity. Ideally, a potentially durable
concrete should have both low water sorptivity and low porosity values.
Please report any comments or errors to the Civil Engineering Department at UCT.
1
CONCRETE DURABILITY INDEX TESTING MANUAL
1.3 APPARATUS
a) A water-cooled diamond-tipped core barrel, with a nominal internal diameter of
70 mm, attached to a suitable coring drill.
2
b) A holding device in which cubes can be clamped firmly and securely to ensure
they remain in position while coring takes place.
c) A water-cooled moveable bed diamond saw.
3
1.5 PREPARATION OF SPECIMENS FROM SITE ELEMENTS
Note: This section only describes the procedure of preparation of test specimens from site
concrete elements. The project specifications should indicate frequency and number of
cores per exposed surface area of concrete elements.
a) Coring of the specimen from site concrete elements shall take place between
28 d and 35 d after casting, unless otherwise required by the project
specifications.
b) Place and firmly secure the core barrel perpendicular to the surface of the
concrete.
c) Core to a depth of between 80 mm and 100 mm. Ensure that the sides of the
core are parallel and within 5° perpendicular to the face.
d) Break off the core from the concrete face with a hammer and chisel, ensuring
that the 35 mm nearest the surface is undamaged.
e) Mark each core with a reference number, place in a sealed bag, and send to the
laboratory for further preparation.
f) After coring, the specimen(s) shall be kept at ambient conditions in the
laboratory for a maximum of 3 days before cutting. The durability index
conditioning shall be started immediately after cutting.
g) Cut the surface 5 mm from the exposed face of the core and discard. Cut the
required thickness (30 ± 2 mm) of the test specimen from the core.
h) Where a specimen is damaged during this process, for example where
aggregate excessively chips from the surfaces to be tested, the specimen shall
not be used for testing.
i) Cores from site elements must be protected from conditions of adverse drying
and damage on site, and during transport to a laboratory. These conditions may
include, inter alia, high drying temperature and/or very low humidity, rough
handling and impact, etc. It is good practice to wrap samples in plastic-wrap or a
sealed plastic bag and transport them in a container that protects them from
shock, damage, and high temperatures.
1.6 REFERENCES
(1) Alexander MG, Ballim Y, Mackechnie JM, ‘ Concrete durability index testing manual’ Research
Monograph No. 4, Departments of Civil Engineering, University of Cape Town and University of
the Witwatersrand, March 1999.
(2) Gouws S, ‘Durability Index Approach – Method Statements.’ Document submitted to the
Durability Index Test Method working group (under the auspices of the C&CI Technical
Committee), University of the Witwatersrand, 14 August 2003.
(3) Gouws SM, ‘Durability Index Approach – Progress Report 1: Method Statements Summary
document of major amendments to Durability Index Test Methods as agreed upon by Durability
Index Test Method working group (Under the auspices of the C&CI Technical Committee,
University of the Witwatersrand, 19 August 2003.)
(4) Gouws SM, ‘Durability Index Approach – Progress Report 2: Method Statements. Summary
document of major amendments to Durability Index Test Methods as agreed upon by Durability
4
Index Test Method Working Group (Under the auspices of the C&CI Technical Committee,
University of the Witwatersrand, 7 October 2003.)
1.7 REVISIONS
Note: Revisions below may refer to older versions with different clause numbers to this version.
Revisions Description Date
MGA
Cl. 5 c) Core to depth of 80-100 mm
J February 2009
Cl. 5 i) new clause added, to cover transport of cores from site
Cl. 5 j) original Cl 5 i)
SG as agreed with MGA
K May 2010
2.d, 3 and 5 – The option of using the facing machine was omitted
5
CONCRETE DURABILITY INDEX TESTING MANUAL
2.2 APPARATUS
a) An oven capable of maintaining a temperature of 50 ± 2ºC.
Note: Most laboratory ovens are of the forced draft, ventilated type. If, however, the oven
being used is of the closed (unventilated) type, then the relative humidity inside the oven
must be maintained by the inclusion of trays of saturated calcium chloride solution. The trays
should provide a total exposed area of at least 1 m2 per 1 m3 of volume of the oven and
should contain sufficient solid calcium chloride to show above the surface of the solution
throughout the test.
b) Permeability cell as shown in Figure 2.1. The permeability cell should have a
volume of 5 L with a tolerance of ± 5%, and of construction such that it does not
expand or contract in the pressure range 0 kPa to 120 kPa. The cell should be
housed in a room where the temperature is controlled at 23 ± 2 ºC. The
airtightness of the equipment needs to be tested regularly using impermeable
blank specimens, manufactured from, for example, rigid PVC. A drop of 0 kPa
in pressure from an initial permeability cell pressure of 100 kPa over a 24 hour
period is required.
c) Compressible rubber collars with Shore hardness 39A, as shown in Figure 2.2,
for each cell, that allow a tight fit around the specimen to eliminate any leakage
of oxygen, except through the pores of the specimen. The collars shall be free
of cracks and tears.
6
Figure 2.1. Permeability Cell Arrangement
7
f) Vernier calliper, capable of reading to 0.02 mm.
g) Desiccator, containing anhydrous silica gel as the desiccant, with the relative
humidity controlled at a maximum of 60%.
8
Figure 2.3: Different parts of the specimen assembly
Figure 2.4: (a) Permeameter setup, (b) a close-up of specimen assembly, and (c) cell
without specimen assembly
c) Partially tighten the top screw on the cover plate to ensure that it is centred.
Once the specimen has been centred, tighten the apparatus adequately to
ensure no leakage of gas.
9
d) Open the oxygen inlet and outlet valves of the permeability cell. Open the valve
of the oxygen supply tank to between 100 and 120 kPa, and allow oxygen to
flow through the permeameter cell for 5 seconds. This will purge the test
chamber of gases other than oxygen.
e) Close the outlet valve of the permeability cell, ensuring that there are no leaks.
f) Increase the pressure in the permeability cell to 100 ± 5 kPa and close the inlet
valve.
g) After 5 min, record the time, t0, to the nearest minute, as the initial time, and
initial exact pressure P0, to the nearest 0.5 kPa. Use t0 and P0 as such in the
calculations. Thereafter, take at least eight readings at intervals corresponding
to a pressure drop rate of 5 ± 1 kPa. A pressure drop of more than 5 kPa/min
might be an indication of leakage. In such a case, release the pressure in the
chamber, check that the sample fits tightly in the collar, and restart the test
immediately, starting at paragraph (d).
h) Terminate the testing when the pressure has dropped to 50 ± 2.5 kPa or after 6
hours ± 15 min, whichever occurs first. A minimum of 8 readings is required.
Note 1: It is possible to automate the readings. In this case, pressure readings shall be
recorded by the data logging device at 15 minute intervals until the pressure drops to 50 ± 5
kPa or up to 6 hours ± 15 min, whichever occurs first. All the data points so generated shall
be used in the calculation.
Note 2: Specimens can be re-tested if an obvious error in testing or measurement has been
made. However, this should be within 30 min of the end of the initial test, to ensure that the
moisture condition is not adversely affected.
i) The same specimens that were used in the oxygen permeability test can also
be used in the water sorptivity test. For details of the procedure, please refer to
Part 3: ‘Standard Procedure for Water Sorptivity and Porosity Test’.
2.6 CALCULATIONS
NOTE: A standard spreadsheet has been developed to perform the calculations described
below, and it is strongly recommended that this spreadsheet be utilized. A copy of this
spreadsheet is a free download from
http://www.theconcreteinstitute.org.za/durability in the ‘Concrete Tools’ Section.
10
Po is the initial pressure at start of test (at time t0) to the nearest 0.5 kPa,
in kilopascals (kPa);
Pt is the pressure reading at time t, measured from t0, to the nearest 0,5
kPa, in kilopascals (kPa).
b) The coefficient of correlation (r2) should be greater than 0.99. Where the
correlation is less than 0.99 a re-test should be done on the same specimen. (If
the conditions of Note 2 in 2.5 h) above cannot be met, place the specimen in
the 50 ºC oven overnight prior to cooling in the desiccator as per 2.4 (c) above,
in order to re-test). If the subsequent test of the specimen also has a correlation
coefficient of less than 0.99, this specimen should be discarded and another test
specimen prepared.
Note 1: Every reading recorded as described in section 2.5 shall be used in the regression
analysis. No data points shall be excluded in the determination of the correlation coefficient.
No additional manipulation or exclusion of data points is allowed in order to improve the
correlation coefficient. If the correlation coefficient is less than 0.99, the sample shall be re-
tested.
Note 2: Notwithstanding Note 1 above, discretion should be exercised whether to always
discard a specimen as above. It is possible to have very impermeable or alternatively very
permeable specimens, where the r2 may be less than 0.99, but will generally achieve 0.98.
Note 3: The slope of the linear regression line forced through the (0,0) point can be
calculated from the equation (2.1):
P 2
∑ �ln � 0 ��
P t
z= P (2.1)
∑ �ln � 0 � t�
Pt
where
z is the slope of the linear regression;
P0, Pt, and t are defined as in 2.6 a).
c) The correlation coefficient, r2, can be calculated from the equation (2.2):
2
2
∑�ti -tp,i �
r =1- (2.2)
∑ t2i -( ∑ ti )2�
n
where
ti is the time at any given pressure reading, recorded to the nearest
minute, in seconds (s);
tp,i is the predicted time at the same pressure reading (based on the linear
regression), in seconds (s);
n is the number of data points being considered.
11
d) The value of tp,i can be calculated from:
ln( P0 ⁄Pt )
tp,i = (2.3)
z
where P0, Pt and z are as defined above.
The ‘slope’ and ‘rsq’ functions available in Excel CANNOT be used, as they do
not force the line through the zero point.
e) The D’arcy coefficient of permeability may be calculated from:
ω ×V ×g ×d×z
k= (2.4)
R×A×T
where
k is the coefficient of permeability of the test specimen in metres per
second (m/s);
ω is the molecular mass of oxygen (i.e., 0.032 kg/mol), in kilograms per
mole (kg/mol);
V is the volume of the permeability cell, recorded to the nearest 0.01 litre
or 0.00001 m3. The volume of the permeability cell includes the volume
of the cell up to the lower face of the specimen. The volume shall be
determined by dimensional measurement, accurate to the nearest mm,
or by the volume of water contained at 23 ± 2 ºC.
g is the gravitational acceleration (i.e., 9.81 m/s2), in metres per second
squared (m/s2);
d is the average specimen thickness, to the nearest 0.02 mm, in metres
(m);
z is the slope of the linear regression line forced through the (0,0) point,
in reciprocal seconds (s-1);
R is the universal gas constant (8.313 Nm/Kmol), in newton metres per
Kelvin mole (Nm/Kmol);
A is the cross sectional area of the specimen, in square meters (m2);
T is the absolute temperature in Kelvin (K).
f) The coefficient of permeability k is calculated for each specimen. The oxygen
permeability index (OPI) shall be given as the average of the individual OPI
values of the specimens (i.e., the geometric mean), which for four specimens is:
12
Where one specimen has been discarded, the coefficient of permeability may
be calculated from the average of at least three valid test specimens using the
following formula:
2.7 REPORTING
The test report shall include the following information:
a) The individual coefficient of permeability (k) of each specimen to three decimal
places;
b) The individual oxygen permeability index (OPI) of each specimen, to two
decimal places;
c) The average oxygen permeability index (OPI) of all specimens, to two decimal
places;
d) The identification mark of the specimens;
e) A detailed description of the specimens, including flaws such as visible cracks,
honeycombing defects or visible bleed paths. This is particularly important in
this test since the test is stated to be indicative of macro-structural problems.
The test report shall also include the following information, if known:
f) The source of the specimens;
g) The location of the specimens (i.e., within the core or member);
h) The type of concrete, including binder type, water/cement ratio and other
relevant data supplied with the specimen;
i) The curing history;
j) A description of any unusual specimen preparation, for example, removal of
surface treatment;
k) A description of unusual features such as cracks, voids, and excessively
chipped edges;
l) The name of the test officer; and
m) The age of concrete at time of testing.
2.8 REFERENCES
(1) Alexander MG, Ballim Y, Mackechnie JM, ‘ Concrete durability index testing manual’ Research
Monograph No. 4, Departments of Civil Engineering, University of Cape Town and University of
the Witwatersrand, March 1999
13
(2) Mackechnie JR and Alexander MG, ‘Practical considerations for rapid chloride conductivity
testing.’ Proceedings of the Second International RILEM Workshop on Testing and Modelling
the Chloride Ingress Into Concrete, C. Andrade and J. Kropp, ed., 2000
(3) Gouws SM, ‘Durability index approach – method statements.’ Document submitted to the
Durability Index Test Method working group (Under the auspices of the C&CI Technical
Committee), University of the Witwatersrand, 14 August 2003.
(4) Gouws SM, ‘Durability Index Approach – Progress Report 1: Method Statements’ Summary
document of major amendments to Durability Index Test Methods as agreed upon by Durability
Index Test Method working group (Under the auspices of the C&CI Technical Committee,
University of the Witwatersrand, 19 August 2003.
(5) Gouws SM, ‘Durability Index Approach – Progress Report 2: Method Statements’ Summary
document of major amendments to Durability Index Test Methods as agreed upon by Durability
Index Test Method working group (Under the auspices of the C&CI Technical Committee,
University of the Witwatersrand, 7 October 2003.
(6) Mukadam Z, Alexander M.G., Beushausen H.D., ‘The effects of drying preconditioning on the
South African durability index tests.’ Cement and Concrete Composites (69): 1-8, 2016.
(7) Stanish, K., Alexander, M. G. and Ballim, Y., (2001), Assessing the repeatability and
reproducibility values of South African durability index tests, SAICE Journal, Vol. 48(2), pp. 10-
17.
(8) ASTM C 802-14, 2014, “Standard Practice for Conducting an Inter-laboratory Test Program to
Determine the Precision of Test Methods for Construction Materials,” ASTM International.
(9) ASTM C 670-90, 2015, “Standard Practice for Preparing Precision and Bias Statements for
Test Methods for Construction Materials,” ASTM International.
2.9 REVISIONS
Note: Revisions below may refer to older versions with different clause numbers to this version.
Revisions Description Date
14
Revisions Description Date
All MGA
Cl. 5 b): addition of definition of ‘z’.
J Cl. 5 c): insert ‘to the nearest 0.02 mm’ at end of 25 Feb. 2009
definition of ‘d’
M MGA / MO. Changed 1s% to Coefficient of Variation (CoV), App A 12 July 2017
SS / MGA: Better clarity in Cl. 2.5, 2.6, including revised Dec. 2017/April
O
illustrations Figure 2.3, 2.4. 2018
15
APPENDIX A (Part 2)
Typical ranges of within test coefficient of variation and multi-laboratory precision are
provided in table A.1. These values may be refined from time to time as additional
data become available. These data derive mainly from inter-laboratory test
programmes aimed at establishing repeatability and reproducibility data. In general,
in excess of 30 test results were available.
1 2 3
16
APPENDIX B (Part 2)
General
To illustrate the calculations given in Clause 2.6, typical calculations are given below
to determine the oxygen permeability index. All variables are as defined in 2.6 a) to
e).
Calculations
a) Illustrative time and pressure readings of four specimens that were obtained
from the same concrete are shown in table B.1, truncated for brevity, and
dimensions are given in table B.2. The calculation of z for specimen 1 is shown
in table B.3 and b).
1 2 3 4 5 6 7 8
Specimen number
1 2 3 4
Actual Pressure Actual Pressure Actual Pressure Actual Pressure
time Pt time Pt time Pt time Pt
hh.mm.ss kPa hh.mm.ss kPa hh.mm.ss kPa hh.mm.ss kPa
12:55:00 100,0 10:12:00 100,0 14:01:00 100,0 14:00:00 100,0
13:01:00 96,0 10:17:00 97,0 14:03:00 98,5 14:03:00 98,0
13:05:00 93,0 10:22:00 92,0 14:06:00 95,5 14:06:00 96,0
13:09:00 90,0 10:25:00 89,5 14:09:00 93,5 14:09:00 94,0
13:18:00 85,0 10:32:00 84,5 14:13:00 92,5 14:13:00 90,0
13:24:00 80,0 10:39:00 80,0 14:16:00 91,0 14:16:00 88,0
13:28:00 78,0 10:46:00 77,5 14:19:00 87,0 14:19:00 86,0
13:34:00 74,0 10:53:00 74,5 14:23:00 85,0 14:23:00 84,0
13:38:00 72,0 10:57:00 70,0 14:26:00 82,5 14:27:00 81,0
13:42:00 70,0 11:05:00 67,0 14:31:00 80,0 14:31:00 79,0
13:48:00 67,0 11:12:00 64,0 14:34:00 78,0 14:36:00 75,0
13:54:00 65,0 11:18:00 62,0 14:37:00 76,0 14:41:00 73,0
17
Table B.2: Dimensions of the specimens and parameters used
1 2 3 4 5 6
Specimen number
Parameter Measurement
1 2 3 4
1 3,16 3,02 2,90 3,06
1 2 3 4 5 6 7
18
b) The slope of the linear regression line of specimen 1, forced through the (0,0)
point may be calculated as follows:
P 2
∑ [ ln � 0 � ]
P t
z= P0
∑ [ ln � � ×t]
P t
P 2
From Table B.3: ∑ �ln � 0 �� = 0.828
P t
P
Therefore ∑ [ ln � 0 � ×t] = 6606.060
P t
0.828
Therefore z= 6606.060
z = 0.000125
c) The calculation of the correlation coefficient, r2, of the linear regression line of
specimen 1, forced through (0,0) is shown in Table B.4 and d).
1 2 3 4 5 6 7 8 9
13:01:00 360 96,0 1,042 0,041 328 320 1 024 129 600
13:24:00 1 740 80,0 1,250 0,223 1 784 −44 1 936 3 027 600
13:34:00 2 340 74,0 1,351 0,301 2 408 −68 4 624 5 475 600
13:38:00 2 580 72,0 1,389 0,329 2 632 −52 2 704 6 656 400
13:42:00 2 820 70,0 1,429 0,357 2 856 −36 1 296 7 952 400
13:48:00 3 180 67,0 1,493 0,401 3 208 −28 784 10 112 400
19
1 2 3 4 5 6 7 8 9
� ti = 21 360
n = 12
Therefore
(∑ ti )2
= 38 020 800
n
Therefore
29232
r2 = 1 -
52776000 - 38020800
Therefore
r2 = 0.998
Since r2 ≥ 0.99, it is not necessary to re-test the specimen.
ω×V×g×d×z
k=
R×A×T
Where
20
V is the volume of oxygen under pressure in the tank in which specimen 1 was
tested, = 4,9L = 0,0049 m3;
Therefore
k = 6.426×10-10 m/s
f) The coefficient of permeability, ki, (m/s) of each of the other four specimens is
calculated in a similar fashion for each specimen resulting in the following
values.
k1 = 6,426× 10-10 m/s OPI = 9,19
k2 = 6,370 × 10-10 m/s OPI = 9,20
k3 = 6,074 × 10-10 m/s OPI = 9,22
k4 = 6,474 × 10-10 m/s OPI = 9,19
g) The average oxygen permeability index is calculated as follows:
OPI = [(OPI1 + OPI2 + OPI3 + OPI4)/4]
OPI = 9,20
21
CONCRETE DURABILITY INDEX TESTING MANUAL
3.2 APPARATUS
a) An oven capable of maintaining a temperature of 50 ± 2ºC.
Note: Most lab ovens are of the forced draft, ventilated type. If, however, the oven being
used is of the closed (unventilated) type, then the relative humidity inside the oven must be
maintained by the inclusion of trays of saturated calcium chloride solution. The trays should
provide a total exposed area of at least 1 m2 per 1 m3 of volume of the oven and should
contain sufficient solid calcium chloride to show above the surface of the solution throughout
the test.
b) Vacuum saturation facility as shown in Figure 3.1.
c) Plastic or stainless steel tray 20 mm deep and large enough to hold as many
specimens as will be tested simultaneously.
d) Ten layers of absorbent paper towel. Alternatively, 2 small rollers or 4 pins can
be used to support the specimens tested.
e) Vernier calliper, capable of reading to 0,02 mm.
f) Measuring scale with accuracy to 0.01 g.
g) A solution of tap water saturated with calcium hydroxide, (3 grams of Ca(OH)2
per 1 litre of water), maintained at 23 ± 2°C.
22
Figure 3.1: Vacuum Saturation Facility
23
e) Seal the curved sides of the specimens using a sealant as detailed in 3.2 i)
above.
f) If the specimens have been previously tested in the oxygen permeability cells,
they shall be tested immediately upon removal. No additional drying is
necessary provided the specimens have not got wet or have not had an
opportunity to absorb moisture from the atmosphere. Alternatively, specimens
may be placed back in the 50 ºC oven overnight prior to cooling in the
desiccator as per 3.4 (c) above, if they cannot be tested immediately after the
OPI test.
24
Figure 3.2: Test setup using paper towels Figure 3.3: Test setup using supports
25
Figure 3.5: Illustration of recommended arrangement
of specimens for saturation
k) Re-establish the vacuum to between -75 and -80 kPa. This shall be maintained
for 1 hour ± 15 minutes. At no point during this time period shall the vacuum be
permitted to rise above -75 kPa.
l) After 1 hour ± 15 min, release the vacuum and allow air to enter. Allow the
specimens to soak for a further 18 ± 1 hours.
m) After 18 ± 1 hr soaking, remove the specimens from the solution, dry the surface
to a SSD condition with a paper towel, and immediately weigh to an accuracy of
0.01 g. Record this as the vacuum saturated mass Msv of the specimen.
3.6 CALCULATIONS
NOTE: A standard spreadsheet has been developed to perform the calculations described
below, and it is strongly recommended that this spreadsheet be utilized. A copy of this
spreadsheet is a free download from http://www.theconcreteinstitute.org.za/durability in
the ‘Concrete Tools’ Section.
a) Determine the porosity (n) of each specimen, as a percentage, by applying the
following formula:
Msv -Ms0
n= ×100 (3.1)
Adρw
where
Msv is the vacuum saturated mass of the specimen to the nearest 0.01g, in
grams
Ms0 is the mass of the specimen at time t0 (start of the test) to the nearest
0.01g, in grams
A is the cross-sectional area of the specimen to the nearest 0,02 mm2 in
millimetres squared
26
d is the average specimen thickness to the nearest 0,02 mm, in
millimetres
ρw is the density of water, 10-3 g/mm3, in grams per millimetres cubed
b) Determine and plot the mass gain (Mwt) versus the square root of time, by
applying the following formula:
2
⎡ � wt ) ⎤
∑ (�ti -T)(Mwti - M
r2 = ⎢ ⎥ (3.4)
⎢ 2 2⎥
� �
⎣ ∑��ti -T� ∑�Mwti - Mwt � ⎦
where
Mwti is the mass gain as calculated in 3.6 b) at any given time, in grams
ti is the time corresponding to the mass gain Mwti, in hours
and
∑ Mwti
� wt =
M (3.5)
n
27
and
∑ �ti
T= (3.6)
n
where
n is the number of data points.
d) If the coefficient of correlation is less than 0,98, discard the last (25 min) value
from the analysis, and re-determine the correlation coefficient, adjusting the
value of n, the number of data points, in the calculation as relevant.
e) If the coefficient of correlation is still less than 0,98, discard the next value (i.e.
20 min) from the analysis, and re-determine the correlation coefficient, adjusting
the value of n, the number of data points, in the calculation as relevant.
f) Repeat the procedure until a coefficient of above 0,98 is achieved, or there are
less than 5 data points remaining.
g) If a correlation coefficient of 0,98 cannot be obtained with a set of five or more
values, regard the specimen as unsuitable for the determination of the sorptivity.
However, record the range of data able to give a correlation coefficient of above
0,98.
h) Using the values obtained from procedures contained in c) to f) (inclusive),
determine the slope of the line of best fit (F) by linear regression analysis:
� wt )
∑ (�ti -T)(Mwti -M
F= 2 (3.7)
∑��ti -T�
where
Mwti is the mass gain as calculated in 3.6 b) at any given time, in grams,
equation (3.3).
ti is the time corresponding to the mass gain reading Mwti, in hours and
�𝑤𝑤𝑤𝑤 and T are given in equations (3.5) and (3.6) respectively.
𝑀𝑀
Fd
S= (3.8)
Msv - Ms0
where
F is the slope of the best fit line (equation (3.7)), in grams per square root
of hour
d is the average specimen thickness to the nearest 0,02 mm, in mm
28
Msv is the vacuum saturated mass to the nearest 0,01 g of the specimen, in
grams
Ms0 is the mass to the nearest 0,01 g of the specimen at the initial time (t0),
in grams
j) The procedure 3.6 a) through 3.5 g) is carried out separately for each specimen.
The sorptivity index is given as the average of the water sorptivity of the valid
individual test determinations.(4)
Note: Where one specimen has been deemed unsuitable in 3.6 g), it should not be used in
determining the average water sorptivity. At least three valid test specimens should be used
to determine the average water sorptivity.
Note: A simple way to calculate the slope of the regression line is by entering the data in a
Microsoft Excel range and use the function SLOPE {data range of Mwt; data range √t}. The
correlation coefficient can be obtained by using the RSQ {data range Mwt; data range of √t}
function.
3.7 REPORTING
Report the following:
a) Identification number of specimen.
b) Description of specimen.
c) The porosity of each specimen (in percent) to the nearest 1 decimal place.
d) The water sorptivity of each individual specimen (in mm/√h), to the nearest 1
decimal place.
e) The water sorptivity index (in mm/√h) to the nearest 1 decimal place.
f) The range of data used in the calculations.
The following shall also be reported if known
g) Source of the specimen.
h) Location of specimen within cube, core or member
i) Identification mark of each specimen
j) Type of concrete, including binder type, water/cement ratio and other relevant
data supplied with the specimen.
k) Curing history.
l) Unusual specimen preparation e.g. removal of surface treatment.
m) Unusual features such as cracks, voids, excessively chipped edges, etc.
n) Test operator.
o) Age of concrete at time of testing.
Note on interpretation of test values for water sorptivity (WS) and porosity: The measured
values of water sorptivity and porosity are inter-related. For example, a low WS value may
be due to a high porosity value, and vice versa (although this is not always the case).
Therefore, care should be taken in reporting and interpreting these values.
29
3.8 REFERENCES
(1) Alexander M.G., Ballim Y., Mackechnie J.M., ‘Concrete durability index testing manual’
Research Monograph no. 4, Departments of Civil Engineering, University of Cape Town and
University of the Witwatersrand, March 1999
(2) Mackechnie JR and Alexander MG, ‘Practical considerations for rapid chloride conductivity
testing.’ Proceedings of the Second International Workshop on Testing and Modelling the
Chloride Ingress Into Concrete,
(3) Nilsson L.O., Andersen A, Luping T, ‘Chloride Ingress Data from Field Exposure in a Swedish
Environment’, Proc. 2nd International RILEM Workshop “Testing and Modelling the Chloride
Ingress into Concrete”, Eds Andrade C, Kropp J, Paris, 11-12 September 2000.
(4) Gouws SM, ‘Durability index approach – method statements.’ Document submitted to the
Durability Index Test Method working group (Under the auspices of the C&CI Technical
Committee), University of the Witwatersrand, 14 August 2003.
(5) Gouws SM, ‘Durability Index Approach – Progress Report 1: Method Statements’ Summary
document of major amendments to Durability Index Test Methods as agreed upon by Durability
Index Test Method working group (Under the auspices of the C&CI Technical Committee,
University of the Witwatersrand, 19 August 2003.
(6) Gouws SM, ‘Durability Index Approach – Progress Report 2: Method Statements’ Summary
document of major amendments to Durability Index Test Methods as agreed upon by Durability
Index Test Method working group (Under the auspices of the C&CI Technical Committee,
University of the Witwatersrand, 7 October 2003.
(7) Mukadam Z, Alexander M.G., Beushausen H.D., ‘The effects of drying preconditioning on the
South African durability index tests.’ Cement and Concrete Composites (69): 1-8, 2016.
3.9 REVISIONS
Note: Revisions below may refer to older versions with different clause numbers to this version.
Revisions Description Date
30
Revisions Description Date
(MGA)
All MGA
31
CONCRETE DURABILITY INDEX TESTING MANUAL
4.2 APPARATUS
a) An oven capable of maintaining a temperature of 50 ± 2ºC.
Note: Most lab ovens are of the forced draft, ventilated type. If, however, the oven being
used is of the closed (unventilated) type, then the relative humidity inside the oven must be
maintained by the inclusion of trays of saturated calcium chloride solution. The trays should
provide a total exposed area of at least 1 m2 per 1 m3 of volume of the oven and should
contain sufficient solid calcium chloride to show above the surface of the solution throughout
the test.
b) Vacuum saturation facility as shown in Figure 3.1, Part 3.
c) Conduction cell as shown in Figure 4.1 with anode and cathode parts
permanently marked on the outside of the cell, and with flexible rubber collars
free of cracks or tears.
Note: Two designs of the conduction cell are available (see Figures 4.1 a) and b))
d) Stabilized DC power supply, 0 V to 12 V, 0 A to 1 A.
32
a) Simple cell arrangement
33
e) Digital voltmeter and ammeter (two multimeters), capable of displaying four-
digits, 0 V to 20 V range, 0 mA to 300 mA, and a rated accuracy of 0.1 %.
f) Measuring scale, of accuracy at least 0.01 g.
g) Vernier calliper, capable of reading to 0.02 mm.
h) CP grade NaCl, of 99% purity.
i) Desiccator, containing anhydrous silica gel as the desiccant, with the relative
humidity controlled at a maximum of 60 %.
Note: Due to the highly corrosive solutions used during the test, all equipment must be
cleaned thoroughly with warm soapy water after each use. The copper electrodes and
banana plugs need to be cleaned with sandpaper or an acidic solution. Replacement
electrical connections are necessary from time to time. To protect electrical test equipment, it
is advisable to place such equipment on a shelf above the bench on which the conductivity
cell is placed.
34
perimeter of the specimen. Calculate and record, to the nearest 0.02 mm, the
average of each set of four readings.
j) With the flexible rubber collar in the central ring portion of the cells, place a
concrete specimen within the collar, ensuring that it is placed with one face
against the plastic lip of the rigid ring, as in Figure 4.4.
35
Figure 4.4: Properly placed specimen in collar
k) Screw the anode and cathode sections of the cell into the central portion.
Tighten both parts sealing the specimen, and ensure that there are no signs of
leakage. See Figure 4.5.
l) Place the assembled test rig (anode, cathode and central portion) to stand
upright on a horizontal surface, and completely fill both the anode and cathode
compartments in turn with the 5 M NaCl solution through the holes in each
compartment (see Figure 4.1 a) and b). Seal the holes with the cap-screws and
ensure that there are no signs of leakage.
m) Connect the ammeter and the voltmeter, as shown in figure 4.1, and adjust the
DC power supply until the voltage applied across the specimen (capillary
voltage) is approximately 10 V.
Note: The voltage across the specimen is read from the voltmeter; it is not the voltage
indicated in the DC power supply.
n) Simultaneously record the current and voltage readings from the ammeter and
voltmeter respectively. Upon switching on the power supply in the test circuit, the
capillary voltage should be quickly adjusted to approximately 10 V and the
current and corresponding capillary voltage across the specimen recorded within
10 seconds. Switch off the circuit (i.e. power supply) as soon as possible.
36
o) Testing should be completed within 15 min of removing a specimen from the
NaCl solution.
Note: Specimens can be re-tested within 30 minutes of the first test, but should be discarded
thereafter. Specimens should be stored in the NaCl solution before retesting.
4.7 CALCULATIONS
b) Determine the chloride conductivity of each specimen by applying the following
formula:
id
σ= (4.1)
VA
where:
σ is the chloride conductivity of the specimen (mS/cm)
i is the electric current (mA)
d is the average thickness of specimen (cm)
V is the voltage difference (V)
A is the cross-sectional area of the specimen (cm2)
b) Determine the chloride conductivity index as the average of the chloride
conductivity of the four test specimens. Where one specimen has been
discarded, the chloride conductivity index may be calculated from the average
of at least three valid test specimens.
c) Determine the chloride solution porosity of the specimen by applying the
following equation:
(Ms -Md )
n= ×100 (4.2)
Adρs
where:
n is the porosity as a fraction of the volume of the specimen that is
occupied with the solution, as a percentage (%).
Ms is the vacuum saturated mass of the specimen determined in section
4.6 h) to the nearest 0.01 g, in grams (g).
Md is the mass of the dry specimen determined in section 4.6 a) to the
nearest 0.01 g, in grams (g).
A is the cross-sectional area of the specimen to the nearest 0.02 mm2 in
square millimetres (mm2).
d is the average specimen thickness to the nearest 0.02 mm, in
millimetres.
ρs is the density of salt solution (i.e., 1.19 x 10-3 g/mm3), in grams per
cubic millimetre (g/mm3).
37
Note 1: It has been found that the porosity determined from the chloride conductivity test is
normally lower than that determined in the sorptivity test.
Note 2: The repeatability and reproducibility of chloride conductivity tests are given in
Appendix A.
4.8 REPORTING
The test report shall include the following information:
a) Identification mark of the specimen.
b) A detailed description of the specimen.
c) The chloride conductivity of each individual specimen.
d) The chloride conductivity index to the nearest 2 decimal places.
e) The porosity of each specimen expressed as a percentage to two decimal
places.
The test report shall also include the following information, if known:
f) The source of the specimen.
g) The location of specimen within cube, core or member
h) Identification mark of each specimen
i) The type of concrete, including binder type, water/cement ratio and other
relevant data supplied with the specimen.
j) Curing history.
k) Unusual specimen preparation, for example removal of surface treatment.
l) Unusual features such as cracks, voids, and excessively chipped edges.
m) The name of the test officer.
n) The age of concrete at time of testing.
4.9 REFERENCES
(1) Alexander MG, Ballim Y, Mackechnie JM, ‘Concrete durability index testing manual’ Research
Monograph no. 4, Departments of Civil Engineering, University of Cape Town and University of
the Witwatersrand, March 1999
(2) Mackechnie JR and Alexander MG, ‘Practical considerations for rapid chloride conductivity
testing.’ Proceedings of the Second International Workshop on Testing and Modelling the
Chloride Ingress Into Concrete, C. Andrade and J. Kropp, ed., 2000.
(3) Gouws SM, ‘Durability index approach – method statements.’ Document submitted to the
Durability Index Test Method working group (Under the auspices of the C&CI Technical
Committee), University of the Witwatersrand, 14 August 2003.
(4) Gouws SM, ‘Durability Index Approach – Progress Report 1: Method Statements’ Summary
document of major amendments to Durability Index Test Methods as agreed upon by Durability
Index Test Method working group (Under the auspices of the C&CI Technical Committee,
University of the Witwatersrand, 19 August 2003.
38
(5) Gouws SM, ‘Durability Index Approach – Progress Report 2: Method Statements’ Summary
document of major amendments to Durability Index Test Methods as agreed upon by Durability
Index Test Method working group (Under the auspices of the C&CI Technical Committee,
University of the Witwatersrand, 7 October 2003.
(6) Otieno M, Alexander MG, ‘Chloride conductivity testing of concrete – past and recent
developments.’ Journal of the South African Institution of Civil Engineering 57(4): 55-64,
December 2015.
(7) Mukadam Z, Alexander M.G., Beushausen H.D., ‘The effects of drying preconditioning on the
South African durability index tests.’ Cement and Concrete Composites (69): 1-8, 2016.
(8) Stanish, K., Alexander, M. G. and Ballim, Y., (2001), Assessing the repeatability and
reproducibility values of South African durability index tests, SAICE Journal, Vol. 48(2), pp. 10-
17.
(9) ASTM C 802-14, 2014, “Standard Practice for Conducting an Inter-laboratory Test Program to
Determine the Precision of Test Methods for Construction Materials,” ASTM International.
(10) ASTM C 670-90, 2015, “Standard Practice for Preparing Precision and Bias Statements for
Test Methods for Construction Materials,” ASTM International.
(11) Otieno, M., 2018, “Sensitivity of the rapid chloride conductivity index test to concrete quality and
changes in various test parameters. Cement and Concrete Composites 86 (2018) 110 -116.
4.10 REVISIONS
Note: Revisions below may refer to older versions with different clause numbers to this version.
Revisions Description Date
D Skipped
J All MGA
New Cl 4 b) inserted; other clause numbers in Cl. 4. changed 25 Feb 2009
accordingly
39
Revisions Description Date
Cl. 6 a): ‘t’ changed to ‘d’ for specimen thickness; ditto Cl. 6 c)
K SG with MGA
Cl. 1 j): Must cool in desiccators only Cl. 2 a):
Changed ‘tap’ to ‘potable’ May 2010
Cl. 6 c): change t to d in the calculation to be
consistent
M MGA / MO. Changed 1s% to Coefficient of Variation (CoV), App A 12 July 2017
40
APPENDIX A (Part 4)
Typical ranges of within test coefficient of variation and multi-laboratory precision are
provided in Table A.1. These values may be refined from time to time as additional
data become available. These data derive mainly from inter-laboratory test
programmes aimed at establishing repeatability and reproducibility data. In general,
in excess of 30 test results were available.
1 2
Repeatability and reproducibility CCI
Repeatability (Coefficient of Variation (%)) CoV (%)a
Laboratory data 5,0 – 10,0
Ready mix concrete data 5,0 – 10,0
Site data 10,0 – 15,0
Reproducibility (Coefficient of Variation (%)) CoV (%)b
Laboratory data 21,1
a
Single operator coefficient of variation
b
Between laboratory coefficient of variation
41