Cold Regions Science and Technology 87 (2013) 1–5

Contents lists available at SciVerse ScienceDirect

Cold Regions Science and Technology

journal homepage: www.elsevier.com/locate/coldregions

Use of tabletting & coating accelerator for the prevention of early-frost of concrete in
cold weather
Jae-Suk Ryou, Yong-Soo Lee ⁎
Department of Civil & Environmental Engineering, Hanyang University, 17 Haengdang-dong, Seongdong-gu, 133-791 Seoul, Republic of Korea

a r t i c l e i n f o a b s t r a c t

Article history: The use of accelerators prevents early-frost by developing strength of concrete in cold weather. However, no
Received 27 July 2012 security of workability occurs because early hydration makes them to react rapidly. Thus, the accelerators via
Accepted 30 November 2012 tabletting were coated with polyvinyl alcohol (PVA) of water-soluble polymer substance. The discussion in-
cludes the following: mortar setting time, workability by elapsed time, early strength to assure the develop-
Keywords:
ment of adequate strength, and freezing–thawing resistance. As a result, workability can be secured as well as
Tabletting
Accelerator
the development of early strength to prevent early-frost. The porosity and progress of hydration affecting the
Polyvinyl alcohol (PVA) concrete were also found to be superior by analyzing the mercury intrusion porosimetry (MIP) and scanning
Pan-coating method electron microscopy (SEM) results.
Workability © 2012 Elsevier B.V. All rights reserved.
Early-frost

1. Introduction The coating, whose objective includes the protection of contents,

When concrete is placed under low temperature in cold weather,
it is anticipated to have serious damage in quality due to freezing.
This is called early-frost, which indicates strength degradation,
spalling, and cracking due to freezing in the initial hardening
stage, and repeated freezing–thawing cycle after placement of con-
crete in low temperature (Mehta and Monteiro, 2006; Suprenant,
1985). The basis to prevent early-frost will assure the development
of adequate strength to prevent freezing of concrete in initial stage
(ACI 306R, 2010). When cold-weather concrete is utilized, the
methods of hot air curing mix proportion adjustment, and others
are used to prevent the concrete from being damaged due to
early-frost. Nowadays, the hybrid method by admixtures and hot air
curing has been used. In cold-weather, the use of accelerators among
admixtures has advantages, particularly in terms of preventing early
frost and improving early strength. Workability, however, is hard to se-
cure because early hydration of accelerators makes them to react rap-
idly (Brook et al., 1988; Hama and Kamada, 1999; Prudencio, 1998).
Therefore, the control technique of the accelerator's reaction time
in cold weather concrete will be required. Although the control
accelerator's reaction time conducted tabletting method in previous
research, the concrete with tablet, as part of the tablets crashed due
to friction between the aggregates and the mixer blade in early
mixing, reacted without securing workability. If the tablet is placed
after mixing other materials, the desired performance may not
occur due to poor tablet distribution (Ryou and Lee, 2012).
⁎ Corresponding author. Tel.: +82 2 2220 4323.
E-mail address: imcivil@hanyang.ac.kr (Y.-S. Lee).
exterior improvement, and control of work performance, and others,
is used mainly in foods and medicines. Main coating methods in this
field were pan-coating method, fluidized-bed coating method, press-
coating method, and others (Dale, 1959; Loyd et al., 2005; Yang
et al., 2007). The pan-coating method is most common, as shown in
Fig. 1.
In this study, tablets were used as accelerators, which have the ad-
vantage of preventing early-frost and in developing early strength.
The tablets were then coated with water-soluble polymer substance.
Whether the operation time can be secured by adjusting the reac-
tion time in the early stage and whether early strength can be devel-
oped due to accelerator's function by dissolving water-soluble coating
material in the following stage are evaluated by testing.
2. Experiment
The surface of the accelerators via tabletting was coated with
water-soluble polymer substance, which used polyvinyl alcohol
(PVA). The PVA as coating materials can withstand friction between
the aggregates and the mixer blade in early mixing due to stiffness.
Water-solubility can be adjusted depending on multiple grades (degree
of deacetylation) (Korsmeyer et al., 1985; Peppas, 1997; Ravve, 2003).
The coating method of the tablet is pan-coating method. The
blending solution of PVA and purified water was sprayed to the tablet
of pan-inside, which simultaneously blew hot air. The tablet surface
was then bonded with PVA by evaporation of purified water. Repeat-
ing this process can increase coating thickness (Heinämäki et al.,
1997; Ruotsalainen et al., 2003). The PVA coating is 0.01–0.05 mm
in thickness when the process is repeated at least twice. The features

0165-232X/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.coldregions.2012.11.005
2 J.-S. Ryou, Y.-S. Lee / Cold Regions Science and Technology 87 (2013) 1–5
Fig. 1. Process of pan-coating method.
of tablet and coating are shown in Table 1. Fig. 2 shows PVA-coating
tablet and close-up image.
The mortars for setting time were made with W/C= 0.5 and with a
cement-to-sand ratio of 1:3, according to ISO 679. For the characteristics
of concrete made with 10-level specimens, in the pilot test, the concrete
mixtures were found to have the following properties: W/C = 52.0%,
S/a= 48.5%, target slump =150 mm, and target strength= 24 MPa.
Table 2 shows the replacement ratio for each testing.
When water was added, the cement gradually lost its workability
with slight heat, and then it hardened because a new matrix was devel-
oped by the cement clinker which reacted with the water. This step is
called setting. The hardening after the end of the setting is a step on
the development of measurable strength. In general, setting and hard-
ening time were retarded in cold weather (Mindess et al., 2003).
In this study, the setting time was measured according to ASTM C
403. After curing in a temperature and humidity chamber, the pene-
tration resistance was measured depending on the elapsed time.
The setting time was measured according to the graph of the curved
hand fitting that was drawn after points were indicated using the
penetration resistance and the elapsed time (the initial and final
setting times were decided at 3.5 and 28 MPa, respectively). The
Table 1
Features of the tablet and the coating.
Features
Tablet Shape: cylindrical, diameter: 5 mm, length: 5 mm
Main ingredient: aluminate-based accelerator (Acc.)
Tabletting method: direct compress
Coating Polyvinyl alcohol (PVA)to water ratio = 1:12
Coating thickness: 0.01–0.05 mm
Coating method: Pan-coating
Fig. 2. PVA coating tablet and close-up image.
dissolved time of the PVA coating was then evaluated by the differ-
ence of the initial setting time depending on the coating's presence.
To determine the possibility of developing early strength and of
securing workability in fresh concrete, a slump test was conducted
at 60-minute elapsed time according to ASTM C 143. After curing
under the typical cold weather protection of 15 °C temperature and
60% humidity in the temperature and humidity chamber, the com-
pressive strength was checked after 24 h to see if it matched the min-
imum compressive strength of 5 MPa for early-frost prevention
according to ASTM C 39 (KCI, 2009).
Although for the early-frost prevention meet to compressive
strength, severe repeated freezing and thawing may cause frost dam-
age (Cai and Liu, 1998). Accordingly, procedure B of the standard test
method described in ASTM C 666 was conducted, and the long-term
frost resistance was evaluated based on the relative dynamic elastic
modulus. The test is continued (either continuously or intermittently)
for 300 cycles or until the relative dynamic elastic modulus has
reached 60% of its initial value, whichever occurs first.
Sampling was conducted after the hydration progress was stopped
by acetone on 24-hour specimens, which developed early strength.
The presence of harmful pores after PVA coating tablet reaction and
the presence of the hydrates affecting early strength were evaluated
by analyzing the MIP and SEM results. In particular, it has advantage
that the porosity and pore-size distribution can be easily determined
through mercury intrusion porosimetry (MIP) test (Kumar and
Bhattacharjee, 2004).
3. Results and discussion
In Fig. 3, the initial setting time was shortened depending on the
increased replacement ratio, as the main ingredient of the tablet is ac-
celerators, which were directly involved in the hydration and setting
 J.-S. Ryou, Y.-S. Lee / Cold Regions Science and Technology 87 (2013) 1–5 3

Table 2 Table 3
Replacement ratio of PVA coating tablet and tablet for each mix proportion. Criteria for each section (Ryou and Lee, 2012).

Type Replacement ratio Rating Criteria

Mortar Control I Security of operation time

W/C = 0.5, cement-to-sand Tablet (0.5%, 1%, 1.5%, 2%) Prevention of early frost
Ratio = 1:3 PVA coating (0.5%, 1%, 1.5%, 2%) II No security of operation time
Concrete Control, ACC. 0.5% Prevention of early frost
W/C = 52%, S/a = 48.5% Tablet (0.5%, 1%, 1.5%, 2%) III Security of operation time
Binder: 337 kg/m3 PVA coating (0.5%, 1%, 1.5%, 2%) No prevention of early frost
IV No security of operation time
No prevention of early frost

compressive-strength criterion of 5 MPa for early-frost prevention.

Fig. 3. PVA coating and tablet's initial setting time.
of the cement. In addition, the initial setting time of non-coating tab-
let showed shorter than the PVA coating tablet, because non-coating
tablets react part of the crashed tablets due to friction between the
aggregates and the mixer blade. Therefore, the dissolved time of the
PVA coating showed to be about 45 min as evaluated by the differ-
ence of the initial setting time depending on the coating's presence.
Fig. 4 shows the possibility of securing workability within the
margin of error of ± 25 mm in 60 min, whether early strength be-
came 5 MPa for preventing early-frost or not. The criteria for each
section are shown in Table 3.
In Fig. 4(a), the control located in section III, the typical problem of
cold-weather concreting, is shown. On the other hand, tablet 0.5% and
1% are located in section I. As for the 0.5% acc., tablet 1.5 and 2%
are located in section II. This shows what satisfy the minimum
However, there was difficulty securing workability. In Fig. 4(b), all
PVA coating tablets were located in section I, as PVA coating may be
resistant to friction between the aggregates and the mixer blade in
early mixing. In addition, its compressive strength is higher than
that of the tablet, as this was distributed well in the PVA coating
tablets in early mixing with other materials. Therefore, the PVA coat-
ing tablet shows that workability was secured and early-frost was
prevented.
The degree of durability resistance on freezing and thawing is
evaluated by relative dynamic modulus of elasticity. A relative
dynamic modulus of elasticity of 100 indicates durable concrete. A
relative dynamic modulus of elasticity more than 60 is considered
“satisfactory”. A relative dynamic modulus of elasticity in the range
of 40 to 60 is not desirable. But if the relative dynamic modulus of
elasticity is less than 40, then the concrete has serious durability
problem (Zaharieva et al., 2004).
The freezing–thawing resistance results are shown in Fig. 5. The
concrete with PVA coating tablets show over 60% relative dynamic
modulus of elasticity. Therefore, this indicates that the concrete with
PVA coating tablets can resist frost damage under severe repeated
freezing and thawing.
After PVA coating tablets react, the pores are evaluated by MIP to
confirm the presence of harmful pores. In hardened concrete, the
pores of less than 0.03 μm and 0.003–10 μm range can be classified
into gel pores and capillary pores, respectively. The capillary pores,
which are reported to be a factor affecting compressive strength
and permeability of chloride ion, can be classified into intergel parti-
cle pores and the rest (Moon et al., 2006).
In Table 4, the PVA coating tablets, the total intrusion volume, and
porosity show low values compared with the control, and are similar
to those of ACC. 0.5%. In Fig. 6, intergel particle pores of capillary pores

a) Tablet b) PVA coating

Fig. 4. Relative slump and compressive strength of the concrete.

4 J.-S. Ryou, Y.-S. Lee / Cold Regions Science and Technology 87 (2013) 1–5

Fig. 7 shows the observed results in terms of the hydration state

via SEM in the 24-hour curing. Strength development may progress
due to the reduction of the pores that filled the hydrates. Particularly,
ettringite has a lot of influence in early strength (Arai, 1998; Schmitt,
1973). In the following images, the control shows to be more porous
than the PVA coating (0.5%). The PVA coating (0.5%), which had more
ettringite as characterized by the long slender needles (Silva and
Monteiro, 2007), shows to be much denser than the control.

4. Summary

The surface of the accelerators via tabletting was coated with

water-soluble polymer substance, which used polyvinyl alcohol
(PVA). Whether the operation time can be secured by adjusting the
reaction time in the early stage and whether early strength can be de-
veloped were evaluated through a test and analysis. The following
conclusions were made:
Fig. 5. Relative dynamic modulus of elasticity in each cycle.

1) In the setting test, the dissolved time of the PVA coating was about
45 min as evaluated by the difference of the initial setting time
Table 4 depending on the coating's presence. The setting time was short-
MIP analysis results of each type at 24-hour curing. ened, as compared with the aluminate-based accelerator (Acc.).
Type Total intrusion volume Average pore diameter Porosity 2) The concrete with PVA coating tablets confirmed what would
(mL/g) (μm) (%) satisfy workability and compressive strength for prevention of
Control 0.1197 0.0465 23.2265 early-frost, and showed over 60% relative dynamic modulus of
Acc. 0.5% 0.1032 0.0468 20.3599 elasticity for securing freezing and thawing resistance.
PVA coating 0.5% 0.1035 0.0458 20.7491 3) The results of MIP analysis showed low values of the total intru-
PVA coating 1.0% 0.1050 0.0437 21.0668
sion volume and porosity compared with the control, as con-
PVA coating 1.5% 0.1062 0.0461 21.2401
PVA coating 2.0% 0.1081 0.0436 21.5870 firmed similar to those of the aluminate-based accelerator (Acc.).
As a result, after PVA coating tablets react, the pore structure of
the hardened concrete was not affected. In the SEM analysis, it
are zones which affect shrinkage and compressive strength (Atahan confirmed that the concrete with PVA coating tablets is much
et al., 2008); all cases show similar pore volume in the zone of denser than the control.
intergel particle pores. Compared with the rest of the capillary
pores, the pore volume of the control shows higher results. Therefore, Therefore, it was found that it is possible to secure the operation
after PVA coating tablets react, the pore structure of the hardened time by adjusting the reaction time in the early stage and to develop
concrete was unaffected because of the development of outstanding early strength due to accelerator's function by PVA coating in the
early strength due to low porosity compared with the control. following stage.

Fig. 6. Pore volume distribution as a function of pore diameter in 24-hour curing.

 J.-S. Ryou, Y.-S. Lee / Cold Regions Science and Technology 87 (2013) 1–5
Fig. 7. SEM micrographs of hydrates in 24-hour curing (×3000).
Acknowledgment
This research was supported by a Research Program through
Hanyang University of Korea (HY-2009-N).
References
ACI 306R, 2010. Guide to Cold Weather Concreting. USA, pp. 2–22.
Arai, Y., 1998. Cement Materials Chemistry, second ed. Chonnam National University
Press, Korea, pp. 156–252.
Atahan, H.N., et al., 2008. The morphology of entrained air voids in hardened cement
paste generated with different anionic surfactants. Cement and Concrete Composite
30 (7), 566–575.
Brook, J.W., et al., 1988. Cold weather admixture. Concrete International 10 (10),
44–49.
Cai, H., Liu, X., 1998. Freeze–thaw durability of concrete: ice formation process in pores.
Cement and Concrete Research 28 (9), 1281–1287.
Dale, E.W., 1959. Air-suspension technique of coating drug particles. A preliminary
report. Journal of the American Pharmaceutical Association 48 (8), 451–454.
Hama, Y., Kamada, E., 1999. The properties of concrete containing a frost-resistant
accelerator. Concrete Journal 37 (11), 3–8.
Heinämäki, J.M., et al., 1997. Optimization of aqueous-based film coating of tablets
performed by a side-vented pan-coating system. Pharmaceutical Development
and Technology 2 (4), 357–364.
KCI, 2009. Korea Concrete Standard Specification — Cold Weather Concreting. Korea
Concrete Institute, pp. 170–175.
Korsmeyer, R.W., et al., 1985. Mechanisms of solute release from porous hydrophilic
polymers. International Journal of Pharmaceutics 15 (1), 25–35.
5
Kumar, R., Bhattacharjee, B., 2004. Assessment of permeation quality of concrete
through mercury intrusion porosimetry. Cement and Concrete Research 34 (2),
321–328.
Loyd, V.A., et al., 2005. Ansel's Pharmaceutical Dosage Forms and Drug Delivery
Systems, eighth ed. Lippincott Williams & Wilkins, USA, pp. 199–230.
Mehta, P.K., Monteiro, P.J.M., 2006. Concrete — Microstructure, Properties, and
Materials, third ed. McGraw Hill, USA, pp. 135–148.
Mindess, S., et al., 2003. Concrete, second ed. Prentice Hall, USA, pp. 193–220.
Moon, H.Y., et al., 2006. Relationship between average pore diameter and chloride dif-
fusivity in various concretes. Construction and Building Materials 20 (9), 725–732.
Peppas, N.A., 1997. Hydrogels and drug delivery. Current Opinion in Colloid & Interface
Science 2 (5), 531–537.
Prudencio Jr., L.R., 1998. Accelerating admixtures for shotcrete. Cement and Concrete
Composite 20 (2–3), 213–219.
Ravve, A., 2003. Principles of Polymer Chemistry. Plenum Press, USA, pp. 269–339.
Ruotsalainen, M.J., et al., 2003. A novel technique for imaging film coating defects in the
film–core interface and surface of coated tablets. European Journal of Pharmaceutics
and Biopharmaceutics 56 (3), 381–388.
Ryou, J.S., Lee, Y.S., 2012. Properties of early-stage concrete with setting-accelerating
tablet in cold weather. Materials Science and Engineering A 532, 84–90.
Schmitt, H.C., 1973. Effect of clinker composition of setting and early strength of
cement. Zem-Kalk-Gips 26 (2), 63–66.
Silva, D.A., Monteiro, P.J.M., 2007. Early formation of ettringite in tricalcium aluminate–
calcium hydroxide–gypsum dispersions. Journal of the American Ceramic Society
90 (2), 614–617.
Suprenant, B.A., 1985. Freezing concrete as a construction practice. Cold Regions
Science and Technology 11 (2), 195–197.
Yang, J.H., et al., 2007. Experimental Pharmaceutics. Shinil, Korea, pp. 268–277.
Zaharieva, R., et al., 2004. Frost resistance of recycled aggregate concrete. Cement and
Concrete Research 34 (10), 1927–1932.