Materials and Structures (2013) 46:337–344

DOI 10.1617/s11527-012-9968-1

RILEM TECHNICAL COMMITTEE

Recommendation of RILEM TC 178-TMC: Testing

and modelling chloride penetration in concrete*
Methods for obtaining dust samples by means of grinding concrete in order
to determine the chloride concentration profile

Prepared by Ø. Vennesland, M. A. Climent, C. Andrade

RILEM Technical Committee 178-TMC (Carmen Andrade)**

Published online: 6 December 2012

 RILEM 2012

1 Scope applicable both for laboratory cast concrete speci-

mens, for concrete cores taken from structures and for
This RILEM Technical Recommendation intends to testing on site. These sampling procedures may be
give a general description of methods of sampling for applied for obtaining concentration profiles of any
obtaining chloride concentration profiles in concrete, other chemical species present in concrete.

2 Significance and use

*The text presented hereafter is a draft for general
consideration.
The chloride concentration profile in concrete from the
Comments should be sent within 6 months of publication to the
exposed surface inwards is a valuable tool for
TC Chairlady, Dr. Carmen Andrade, Instituto de Ciencias de la
Construcción ‘‘Eduardo Torroja’’, Serrano Galvache n8 assessing the risk of corrosion of reinforcement steel
4 – Apartado 19.002, 28033 Madrid, Spain. in concrete structures exposed to marine or de-icing
Phone: ?34 91 302 04 40; Fax: ?34 91 302 07 00, salt environments. From the knowledge of the Cl-
E-mail: [email protected]
profile of a structure at a given age, one can get some
qualitative information about the rate of ingress of
**TC Membership:
Chairlady: C. Andrade, Spain. such ions into the structure, and then predictions about
Secretary: J. Kropp, Germany. the reinforcement corrosion onset time can be formu-
Full members: S. Ahmad, R. Antonsen, V. Baroghel-Bouny, lated. These assessments are highly improved when
M. P. A. Basheer, N. S. Berke, M. Carcasses, M. Castellote,
several Cl- profiles are obtained at different times for
C. Cavlek, Th. Chaussadent, M. A. Climent, S. Helland,
F. Fluge, R. François, J. M Frederiksen, M. Geiker, J. Gulikers, the same structure. Furthermore several test methods
D. Hooton, A. Legat, M. Maultzsch, S. Meijers, L. O. Nilsson, for obtaining Cl- transport parameters in non-steady-
C. Page, P. Pedeferri , K. P. Pettersson, R. Polder, A. state conditions are based on fitting the experimental
Raharinaivo, M. Salta, L. Tang, M. Thomas, J. Tritthart,
Cl- content profile to a particular equation. These
O. Troconis de Rincon, Ø. Vennesland.
transport parameters can be used as input parameters
RILEM Technical Committee 178-TMC of mathematical models intended to predict the
(Carmen Andrade) (&) corrosion initiation time. The first step for obtaining
Instituto de Ciencias de la Construcción ‘‘Eduardo the profile is sampling, that may be done on-site or at
Torroja’’, CSIC, Serrano Galvache n8 4, Apartado 19.002,
28033 Madrid, Spain laboratory on cores previously taken from the struc-
e-mail: [email protected] ture or concrete specimen.
338 Materials and Structures (2013) 46:337–344

3 Definitions paste and aggregate zones, whose contributions to the

chloride content may be quite different. For this reason
The concentration profile of a chemical species in the minimum amount of sample is related normally with
concrete can be defined as the spatial distribution of this the maximum size of aggregates [1]. Care must be taken
species in concrete. This distribution is a function of the also in gaining the test samples from the concrete
exposure conditions and time [1]. The chloride profile in element in order not to affect the chloride content [1].
the reinforcement cover zone of concrete is a set of data The modifications to the original Cl- content of the
pairs (depth from surface-concentration) obtained by samples can be due to contamination with portions of
chemical analysis of chloride on representative samples material coming from another sampling point [3]; to
taken from the concrete specimen or structure. Usually partial washing of the soluble chlorides if water is used
the profile is presented in tabular or graphical form. The as a lubricant during drilling or cutting the samples; or
depth from surface of each sample is expressed in length to redistribution of Cl- ions during the sampling due to
units, and the units of the concentrations must be clearly moisture transport. This latter situation may be pro-
established, i.e. % chloride referred to concrete mass. duced for the case of cores taken from a structure, to be
Concrete contains free chloride in its pore solution, cut and crushed later at laboratory, if they are not
bound chloride (chemically or physically bound or adequately protected from drying.
trapped by solid phases of concrete). This suggests that If a realistic concentration profile must be estab-
several chloride profiles can be considered or obtained, lished from experimental analyses, samples must be
(total, free or bound chloride), depending on the fraction taken from the concrete element at different and known
considered, or on the analytical method used for depths from the surface, i.e. each concrete sample for
determining the chloride content (acid soluble fraction, the Cl- profile must be associated to a particular depth
water soluble fraction, etc.) [2]. [4]. A high spatial resolution is required if especial
Regarding the sampling procedures and in order to effects are to be detected, such as surface wash-out, or if
avoid confusions the following definitions are proposed: transport parameters are to be derived from the profiles
[1]. The determination of Cl- profiles for high perfor-
A) Dry drilling method: Sampling by drilling with a
mance impervious concretes, with a high resistance to
rotary hammer and a masonry bit, working
Cl- penetration, may require also the selection of a
directly on the surface of a concrete specimen
sampling procedure with a high spatial resolution.
or real structure. This procedure allows obtaining
Other factors which should be considered when
powdered samples corresponding habitually to
selecting the sampling procedure are the costs and time
depth steps of about 5 mm.
consumptions, associated with the sampling and analysis
B) Grinding method: Sampling by grinding the
of the samples; and the damages to the appearance and
surface of a concrete specimen or real structure.
integrity of the structure, which may be important for
C) Cutting and crushing method: Sampling by saw
instance when performing extensive coring. The more
cutting concrete slices of adequate thickness,
accurate are the methods of sampling, the higher are the
(usually not less than 10 mm), from a specimen
costs, time consumptions and eventual damage, since the
or core, and posterior crushing of the slices
high resolution procedures usually need high diameter
manually (mortar and pestle) or mechanically
cores or sampling areas. This calls for a compromise
(ball or ring mills), to get the powdered sample.
between the interest of sampling at several different
points of a structure, to get representative information of
the behaviour of the whole structure, and the need of
4 General considerations of the sampling obtaining detailed profiles, in order to increase the
procedures reliability of the derived transport parameters.
In the case of sampling on site, a previous step will
When sampling one of the most important needs is consist in the selection of the areas where the chloride
getting representative samples. The size of the samples profiles will be determined or cores drilled for so. This
must be large enough to fulfil the sample requirement of selection is out of the scope of present Recommenda-
the analytical chemistry procedure; and to avoid errors tion as it will depend on the purpose of the inspection
due to local heterogeneity, for instance between cement or particular test.
Materials and Structures (2013) 46:337–344 339

5 Methods of sampling from surfaces in all directions, even upwards from a

horizontal surface. The size of the concrete dust
5.1 Dry drilling method samples must be large enough to guarantee a repre-
sentative profile and to enable an accurate determina-
This procedure consists in collecting the concrete dust tion of the chloride content by the corresponding
obtained when drilling small diameter holes on the chemical procedure (minimum 5 g of dust for each
concrete surface, using a rotary hammer and a sample). The amount of dust obtained depends on the
masonry bit, see Fig. 1. The drilling is performed in diameter of the holes, hence on the diameter of the bit,
steps of selected thickness. The number of steps in the and on the thickness of the drilling steps. A minimum
profile and their thickness must be decided taking into bit diameter of 18–20 mm is recommended to avoid
account the estimated depth reached by the penetrating excessive contributions from large aggregates;
chloride front and the desired spatial resolution. although bits of lower diameter, (10–12 mm), could
Nevertheless, a minimum of 5–8 steps are needed be used for laboratory tests on mortar or cement paste
for obtaining a detailed enough profile, and the specimens. Another useful means for obtaining more
recommended minimum thickness of the steps is representative profiles, or for increasing the amount of
5 mm, although it is possible to obtain samples at steps concrete dust obtained for each sample, is to perform
of 2.5 mm. several drillings in the surroundings of each point to be
The dry drilling method may be considered as a studied in a structure, mixing the dust samples
simple, cheap and less destructive sampling method, as corresponding to each hole for a given depth step,
compared for instance with the grinding method, since see Fig. 2. The minimum number of drilled holes
it does not require taking cores of large diameter. All recommended can be established in function of the bit
these characteristics may allow considering the method diameter and the maximum aggregate size of the
as more suitable for performing extensive sampling concrete, for a certain thickness of the drilling steps [5]
campaigns. On the other hand the dry drilling is usually (see Table 1).
a sampling method of low spatial resolution, due to the The risk of modifying the chloride content of the
minimum thickness of the drilling steps (about 5 mm), concrete dust samples obtained is low since the
as compared with the grinding procedure. drilling is performed in dry conditions; provided that
The collection of concrete dust during dry drilling a proper mechanical fixation of the drilling machine is
is performed with specific dust collectors attached to ensured for obtaining a perfectly perpendicular
the concrete surface, see Fig. 1, or more conveniently
by using a vacuum attachment that allows sampling

Fig. 2 Drilling holes for collecting dust

Table 1 Recommended number of drill holes for collecting

concrete dust samples, in function of the hole diameter and the
maximum size aggregate, for a 10 mm thickness of the drilling
steps [5]
Max. aggregate Number of drill holes with diameters
size (mm)
20 mm 26 mm 32 mm 40 mm

8 1 1 1 1
16 2 1 1 1
Fig. 1 Disposition of drilling machine for sampling concrete 32 5 3 2 1
dust
340 Materials and Structures (2013) 46:337–344

drilling to surface [3]. Another possibility to decrease 4 The drill hole, drill bit and dust collector shall be
the risk of contaminating the concrete dust samples cleaned between each step interval by use of
due to lateral contact of the bit with the wall of the bore compressed air or air blower.
hole at lower depths, is the use of two bits of slightly 5 For each sample of dust collected, the depth is
different diameters, for instance 18 and 22 mm. After measured at the center of the hole by a slide gauge
obtaining each sample with the lower diameter bit, the or similar with an accuracy of at least 0.1 mm.
bore hole can be widened by passing the higher 6 The concrete dust collected in plastic bags is
diameter bit through the same depth step. Thus, the marked with structure denomination, location,
next sample can be taken without lateral contact of the depth interval and date.
bit (lower diameter) with the wall of the hole. 7 The total profile will consist of minimum 5–8
For most purposes each concrete dust sample can be steps.
associated to a depth from the concrete surface
equivalent to the mid point of its drilling step, when
plotting the concentration profile. Nevertheless, when 5.2 Grinding method
it is necessary a more careful calculation of the depth
of each sample, the geometry of the drilled hole must This procedure consists in collecting the dust obtained
be taken into account [4]. The procedure for these when grinding thin successive layers of concrete
calculations, which considers the geometry of the bit parallel to the exposed surface. The grinding is
tip, is outlined in the Annex of this recommendation. performed by using a turning lathe with an adequate
This may be of interest for instance when the working tool (on concrete specimens or cores taken
concentration profile is to be used as an input of a from structures), or with especially constructed com-
mathematical model for deriving transport parameters mercial devices equipped with a diamond bit rotating
or predicting the corrosion initiation time. horizontally, that may be used on concrete specimens
or cores and on real structures. This grinding method
5.1.1 Dry drilling equipment and procedure may be considered as a sampling procedure with high
spatial resolution, since it allows obtaining powdered
The minimum equipment needed is: samples corresponding to thin layers of 0.5–2 mm
thickness. The samples can be easily assigned to a
• Drilling machine.
depth from surface corresponding to the centre of the
• Drill bits of appropriate diameter (for concrete
grinding step. The grinding areas are usually higher
minimum 18 mm).
than 40 cm2, in order to get large enough samples
• Dust collector.
(minimum 5 g), thus avoiding heterogeneity
• Compressed air or air blower.
problems.
• Slide gauge or similar.
In both field and laboratory cases the procedure will 5.2.1 Grinding equipment and procedures
be the following:
5.2.1.1 Equipment In the case of real structures two
1a Field. Selection of the testing area. The locations
sampling strategies may be used: to take cores
shall be marked on a sketch of the structure. Paint
representative of the selected areas, which will be
or similar and any fouling must be removed from
ground later at laboratory (with a lathe or with a
the surface.
grinding device); or to use the especial grinding device
1b Laboratory. The area will be the top or bottom of
on-site. The adequate diameter of the cores or
the core or specimen. The selected area has to be
specimens is normally about 10 cm, although it
that within a diameter *10 mm less than the full
depends on the grinding device used.
diameter or size of the specimen (Fig. 2).
The equipment needed is:
2 Placing and fixing of the machine and dust
collector. • Grinding device or turning lathe.
3 For each depth interval, concrete dust is gathered • Appropriate grinding tools (bits).
in separated marked plastic bags. • Dust collection system.
Materials and Structures (2013) 46:337–344 341

• Compressed air or air blower. 2) Preparation of the specimen or core for grinding.
• Slide gauge or similar. Usually the specimen must be cut to fit into the
• General equipment for extracting cores from a grinding device or lathe housing. It is preferably
concrete structure, and equipment for dry cutting to perform all these cuttings in dry conditions,
concrete specimens, (in the case of using a lathe or without using water as a lubricant, in order to
when it is decided to grind at laboratory). prevent the washing out of the soluble chlorides.
3) Fixing of the specimen to the grinding device or
lathe. This ensures a perpendicular grinding to
5.2.1.2 Procedures In field sampling (grinding on-
surface.
site), the steps will be:
4) Preparation of plastic bags for collection of dust.
1) Selection of testing areas, paint or similar and any 5) Grinding of samples in the predetermined layers
fouling must be removed from the surface. or depth steps (it must be ensured that at least 5 g
2) Fixing of the grinding equipment to the concrete of dry concrete dust is collected from each layer).
structure surface. This ensures a perpendicular 6) Between the grinding of each layer, the grinding
grinding to surface. equipment and grinding area must be thoroughly
3) Preparation of plastic bags for collection of dust. cleaned by use of compressed air or air blower.
4) Grinding of samples in the predetermined layers 7) For each sample of concrete dust collected, the
or depth steps (it must be ensured that at least 5 g depth step below the surface is calculated as the
of dry concrete dust is collected from each layer). average of four measurements of penetration
5) Between the grinding of each layer, the grinding depth with a slide gauge or similar with an
equipment and grinding area must be thoroughly accuracy of least 0.1 mm.
cleaned by use of compressed air or air blower. 8) The concrete dust of each layer is collected in
6) For each sample of concrete dust collected, the marked plastic bags to be analyzed for chloride
depth step below the surface is calculated as the concentration. The marks of each bag will consist
average of four measurements of penetration on the structure denomination, specimen denom-
depth with a slide gauge or similar with an ination, depth interval and date.
accuracy of least 0.1 mm.
The procedures for extracting cores from a concrete
7) The concrete dust of each layer is collected in
structure and for cutting concrete cores or specimens are
marked plastic bags to be analyzed for chloride
not described since they are routinely performed by all
concentration. The marks of each bag will consist
control laboratories specialized in the study of struc-
on the structure denomination, location, depth
tures. An important precaution when extracting the
interval and date.
cores is to protect them immediately from drying, and to
In laboratory sampling, (on specimens or cores proceed as soon as possible to the grinding process, in
taken from structures), the steps will be: order to avoid chloride spatial redistributions.
1) The area of the grinding will be the top or bottom
face of the core or specimen. The grinding will be 5.3 Cutting and crushing method
performed within a diameter *10 mm less than
the full diameter or size of the sample, see Fig. 3. This procedure can be applied only at laboratory, since
it starts with a concrete cast specimen or with a core
taken from a real structure. The specimen is dry cut in
slices of adequate thickness, which will be later
crushed into powder to get the samples to be analyzed
for their chloride contents, see Fig. 4. The number and
thickness of the slices must be decided by estimating
the penetration depth reached by the chlorides into the
structure, and by considering the desired spatial
Fig. 3 The grinding area must be slightly smaller than total resolution. The normal thickness of the slices is about
face, in the case of a concrete core or specimen 5–10 mm, although it is possible to cut thinner slices
342 Materials and Structures (2013) 46:337–344

The equipment needed is:

STEP
• General equipment for extracting cores from a
concrete structure.
• Diamond saw to dry cut the concrete slices.
• Mechanical equipment for filing away the washed
peripheral zones of the cores.
• Crushing equipment, mortal and pestle, ball mill,
ring mill, etc.
Fig. 4 Slices to be cut from a specimen and crushed later for • Slide gauge or similar.
gaining the concrete dust samples

5.3.1.2 Procedure
of about 2 mm. Taking this into account the spatial
resolution of this method is comparable to that of the 1) Preparation of the specimen or core. If necessary
dry drilling method. On the other hand the samples can eliminate mechanically by dry cutting or filing
be more easily assigned to a depth corresponding to away the peripheral portions of the extracted
the centre of the slice. Another inconvenience of this cores.
method is the loss of material due to the cutting 2) Mark the desired points of cutting at the lateral
process. The habitual diamond saws used for dry face of the core or specimen, taking into account
cutting concrete specimens have a thickness of about the material eliminated by the saw at each
5 mm. This implies that 5 mm of concrete depth are cutting.
lost for each slice cut. The size of the concrete 3) Fixing of the specimen or core to the cutting
powdered samples must be large enough to guarantee machine. This ensures a perpendicular cutting to
a representative profile and to enable an accurate the specimen axis.
determination of the chloride content by the corre- 4) Preparation of plastic bags or containers for the
sponding chemical procedure (minimum 5 g of dust slices.
for each sample). The crushing of the slices is 5) Perform the dry cutting of each slice. Each slice
performed manually with a mortar and pestle or must be marked to allow its identification.
mechanically with ball or ring mills. 6) Measure the thickness of each slice, performing
The risk of modifying the real chloride content of four measurements with a slide gauge or similar,
the samples comes from the fact that the coring with an accuracy of 0.1 mm. Calculate the depth
procedure at structures is performed with water step corresponding to each slice, taking into
lubrication, which may wash out the soluble chlorides. account the depth lost in each cutting. Store the
For this reason it is recommended to eliminate slices in their bags or containers, marking them
mechanically by dry cutting or filing away the
peripheral portions of the extracted cores, which are
the zones more susceptible to be washed out, before
starting the cutting process.

5.3.1 Cutting and crushing equipment α

and procedures
φ
5.3.1.1 Equipment In the case of real structures it is h
mandatory to take cores representative of the selected
areas, which will be cut and crushed later at laboratory.
There is no limitation for the diameter of the cores or
specimens, but too small diameters and too thin slices
will produce samples of lower size than required.
Habitual core diameters are about 5–10 cm. Fig. 5 Ideal projection of a bore hole performed with a bit of
diameter /
Materials and Structures (2013) 46:337–344 343

with the structure denomination, location, depth drying, and to proceed as soon as possible to the
interval and date. cutting process, in order to avoid chloride spatial
7) Preparation of plastic bags for collection of redistributions.
dust.
8) Crushing of the slices. Normally this operation
is performed in at least two steps: a manual
6 Annex. Bit shape geometric considerations
rough crushing with a mortar and pestle; and a
for the assignment of depth to the dry drilled
posterior milling to the desired grain size using a
dust samples
ball or ring mill. The habitual chloride chemical
analysis methods require the sample to be milled
The accurate assignment of depth to each sample
to pass through 0.16 or 0.32 mm sieves. The
obtained by dry drilling needs a consideration of the
nature and composition of the milling elements
geometry of the drilled hole, which depends on the
and containers must be adequate for the hard-
geometry of the bit tip. While there are bits with
ness of the concrete under study.
almost flat tips, the more common masonry bits end
9) Between the crushing and milling of each slice,
with a conical tip, see Fig. 5. For the bits with flat tips
all the equipment must be thoroughly cleaned.
each dust sample can be assigned to the centre of the
10) The concrete dust of each slice is collected in
depth step. Nevertheless, for a conically shaped bit tip
marked plastic bags to be analyzed for chloride
the amount of dust removed is not the same all along
concentration. The marks of each bag will
the drilling depth step; and furthermore for the second
consist on the structure denomination, specimen
and subsequent samples, concrete dust from depths
denomination, depth interval and date.
lower than the nominal drilling step is obtained, see
An important precaution when extracting the cores Fig. 6a, b. This leads to the conclusion that each Cl-
from a structure is to protect them immediately from concentration in the profile should be assigned to the

A 18 B 18

mm

9 9

0 0

0 5 10 0 5 10
DEPTH / mm DEPTH / mm

Fig. 6 a Ideal geometric projection of the first 5 mm step projection of the second 5 mm step sample, taken by drilling
sample, taken by drilling with a 18 mm diameter bit from 0 to with a 18 mm diameter bit from 5 to 10 mm depth from surface.
5 mm depth from surface. The solid point indicates the The solid point indicates the approximate position of the gravity
approximate position of the gravity centre. b Ideal geometric centre
344 Materials and Structures (2013) 46:337–344

Table 2 Positions of the

Bit diameter, BIT tip cone Drilling depth Gravity centre of solid sample (mm)
gravity centres, (depth from
/ (mm) height, h (mm) step, L (mm)
surface), of samples 1st step 2nd step 3rd step
obtained by dry drilling
12 3 5 1.6 5.5 10.5
10 4.0 13.0 23.0
18 4 5 1.4 4.8 9.8
10 3.7 12.3 22.3
20 8.7 27.3 47.3
20 4.5 5 1.3 4.5 9.5
10 3.6 12.0 22.0
20 8.5 27.0 47.0
22 5 5 1.3 4.2 9.2
10 3.4 11.7 21.7
20 8.4 26.7 46.7

gravity centre (GC) of the solid sample really taken in of the drilling step, L, is higher or equal to the bit cone
the drilling step, in order to avoid systematic errors by height, h. Equation (2) is valid if L is constant for all
excess in the derivation of the ionic diffusion coeffi- the drilling steps.
cients [4]. Fig. 6a and b show the positions of the GC Table 2 contains the positions of the GC of the
of the dust samples obtained by drilling with a concrete dust samples obtained by dry drilling with
conically shaped bit of 18 mm diameter. conically shaped bits of usual diameters, from 12 to
A method for calculating the GC of the samples 22 mm. These positions have been calculated with
obtained with a conically shaped drill bit was Eqs. (1) and (2).
suggested [4]. The input data for these calculations
are the thickness of the drilling steps (L) and the height
(h) of the cone that constitutes the bit tip, see Fig. 5.
The explicit formulae for the calculations of the References
positions of the GC of the first sample and the second
and subsequent, (nth samples), which were not given 1. Kropp J (1995) Chlorides in concrete. In: Performance cri-
teria for concrete durability, RILEM Report 12, E. and F.
before, are as follows:First sample: N. Spon, London
2  2. Poulsen E (1995) Chloride profiles. Analysis and interpreta-
L h2 2Lh
2 þ 4  3 tion of observations. AEC Laboratory, Vedbaek, Denmark
xGC ¼
for L  h: ð1Þ 3. Vennesland Ø (1999) Report on sampling for obtaining
L  2h
3 chloride profiles, Third Meeting of RILEM TC-178 TMC,
Graz, Austria
Second and subsequent samples (nth sample) 4. Climent MA, de Vera G, Viqueira E (2001) Bit shape geo-
L 2h metric considerations when sampling by dry drilling for
xGC ¼ ðn  1ÞL þ  ð2Þ obtaining chloride profiles in concrete. Mater Struct
2 3 34:150–154
5. Deutscher Ausschuss für Stahlbeton (German Committee for
where xGC is the depth from surface of the concrete
reinforced concrete, DAfStb) (1989) Guide to the determi-
dust sample, which coincides with the gravity centre nation of the chloride content of concrete. Schriftenreihe
of the sample taken by dry drilling. It is important to DAfStb Heft 401, Berlin
bear in mind that Eq. (1) is only valid if the thickness