Muthumani and Shi
EFFECTIVENESS OF LIQUID AGRICULTURAL BY-PRODUCTS AND
SOLID COMPLEX CHLORIDES FOR SNOW AND ICE CONTROL
Anburaj Muthumani, M. Sc.
Research Associate
Winter Maintenance & Effects Program
Western Transportation Institute
Montana State University
PO Box 174250, Bozeman, MT 59717
Phone: (406) 994-6782; Fax: (406) 994-1697
Email: [email protected]
Xianming Shi, Ph.D., P.E.*
Associate Professor
Department of Civil and Environmental Engineering
Washington State University
Sloan 101, P.O. Box 642910
Pullman, WA 99164
Phone: (509) 335-7088
Email: [email protected]
*Corresponding Author
Prepared for the TRB 2014 Annual Meeting and Transportation Research Record
Sponsoring committee: Committee on Winter Maintenance (AHD65)
Total words: 5,000
Figures & Tables = 6 *250 = 1,500
Submitted on August 1, 2014
1
Muthumani and Shi
2
ABSTRACT
Agro-based products and complex chlorides/minerals (CCM) based products are increasingly
employed in snow and ice control operations, either used alone or more commonly as additives
for chloride-based products. Recent studies have shown that agro-based or CCM based products
have the potential to improve the overall deicing and/or anti-icing performance and reduce the
corrosion and environmental impacts. However, the effectiveness of such products has been
limited to qualitative field observations and their specific role in snow and ice control is poorly
understood. This work consists of a systematic laboratory investigation, with a focus on the
thermal properties, ice melting behavior, and corrosivity of four agro-based deicers and two
CCM based deicers. First, CCM based deicers do not exhibit significantly better ability to lower
the freezing point of water when compared with NaCl, but they feature slightly better ice melting
capacity at 15oF than NaCl. Second, agro-based additives seem to significantly lowered the
freezing point of 23 wt.% NaCl brine but did not significantly improve the ice melting capacity
at 15oF or 25oF, implying their role as ‘cryoprotectants’. Third, CCM and agro-based deicers do
not exhibit significantly lower characteristic temperature than reagent-grade NaCl. A very strong
positive linear relationship exists between the eutectic temperature (Te) and the characteristic
temperature (Tc) of the tested liquids, indirectly confirming the validity of using DSC
thermograms to assess liquid deicers. The gravimetric method reveals that CCM based deicers
exhibit slightly lower corrosivity to carbon steel than NaCl and agro-based additives exhibit
significant benefits in reducing the corrosivity of 23 wt.% NaCl brine. The electrochemical
method reveals that while the beet-based additives do not significantly alter the corrosion
potential of carbon steel, the other type of additives moved the corrosion potential to a much
more positive level, implying anodic type inhibitor at work.
Muthumani and Shi
3
INTRODUCTION
More effective and less corrosive snow and ice control chemicals could result in significant
economic, environmental, and societal benefits. Approximately 70% of US roads are located in
snowy regions, with nearly 70% of the US population living in these regions [1]. The most
common freezing point depressants used for highway winter operations are sodium chloride
(NaCl), magnesium chloride, (MgCl2), calcium chloride (CaCl2), and potassium acetate (KAc).
Among them , chloride salts are the most readily available and widely used in either a solid or
liquid form [2]. Chloride salts are effective over a wide range of temperatures [3], and their
baseline performance and corrosivity have been recently reported [4]. The last two decades have
seen increased use of chloride based deicers and continued concerns over their deteriorating
effects on motor vehicles [5, 6], transportation infrastructure [7, 8] and the environment [9]. KAc
is more expensive than chloride salts but generally considered to be non-corrosive to carbon steel
and more benign on surrounding soils and ecosystems [10]. However, recent studies have found
that KAc can be corrosive to galvanized steel [11] and increase the emulsification risk of asphalt
concrete [12].
The last two decades have also seen the continued introduction of agro-based chemicals into
snow and ice control operations, either used alone or more commonly as additives for chloridebased products [15]. They have emerged since the late 1990s, often produced through the
fermentation and processing of beet juice, molasses, corn, and other agricultural products [16,
17]. Janke and Johnson Jr (1999) developed a deicer from a by-product of a wet milling process
of corn (steepwater). The deicer formulation is noncorrosive, inexpensive, water soluble, and
readily available in large quantities. Tests have shown that successful inhibition is achieved with
the addition of these steepwater solubles to chloride salts [18]. Recently, glucose/fructose and
unrefined sugar have been mixed in sand to prevent freezing and added in salt brine for anti-icing
[19]. Taylor et al. (2010) evaluated the brines made of glycerol, NaCl, MgCl2, and commercial
deicers individually and in combination and concluded that the blend of 80% glycerol with 20%
NaCl showed the greatest promise in good laboratory performance and low negative impacts
[20]. Agro-based additives increase cost but may provide enhanced ice melting capacity, reduce
the deicer corrosivity, and/or last longer than standard chemicals when applied on roads [21, 22].
Despite their advantages, there are concerns over the toxicity of agro-based chemicals to the
aquatic ecosystems, high cost, and quality control issues [21, 23].
In addition to agro-based deicers, another developing class of deicers features the unique synergy
of complex chlorides and mineral products. These products are usually, but not limited to, mined
and evaporated solid salt products with naturally occurring chloride and non-chloride
constituents.
Prior to this study, the effectiveness of agro-based or complex chlorides/minerals (CCM) based
products has been limited to qualitative field observations and their specific role in snow and ice
control is poorly understood. There remains an urgent need to systematically examine their
Muthumani and Shi
4
overall effectiveness, corrosion and environmental impacts, and potential benefits in enhancing
anti-icing and/or deicing. In particular, there is contradictory information over the “modes of
action” by which such products provide benefits. A variety of manufacturer claims have been
made about these products, such as: (1) lowering the freezing point temperature; (2) improving
the ice melting capacity; and (3) reducing the corrosiveness to metals. Some of these claims, if
proven true, would aid in the prevention of ice formation or refreeze, reduce the use of traditional
products, and reduce the environmental impacts of snow and ice control operations without
sacrificing the level of service on winter roads. Fu et al. [24] conducted more than 100 hours of
friction readings over nine snow events and concluded that two beet molasses-based materials
worked effectively as pre-wetting and anti-icing agents. These liquid organic by-products
exhibited better performance than regular salt and salt brine, due to unknown mechanism(s). Yet,
laboratory testing under a Transportation Pooled Fund study [25] revealed that the organic
corrosion inhibitor packages used in three chloride-based products have little benefit in
suppressing the effective temperature or in providing ice melting capacity. Another laboratory
study [26] revealed that blending sugar beet-based organic liquids into a 23% salt brine at the
volume ratio of 20:80 not only significantly reduced the corrosiveness of the brine to carbon
steel and reduced the mass loss of PCC specimens after the salt scaling test, but also led to lower
splitting tensile strength of the PCC specimens after the salt scaling test. Such blending also
significantly reduced the brine’s 60-min ice melting capacity at 15F.
In this context, this work reports some experimental results of a systematic laboratory
investigation, with a focus on the thermal properties, ice melting behavior, and corrosivity of
four agro-based deicers and two CCM based deicers identified by the sponsor, the Clear Roads
Pooled Fund (www.clearroads.org). The main hypotheses tested include: these deicers feature
lower freezing point than their counterpart (solid or liquid NaCl); these deicers feature higher ice
melting capacity than their counterpart; and these deicer feature significantly lower corrosivity to
carbon steel than their counterpart.
Muthumani and Shi
5
METHODOLOGY
Materials of Interest
Seven agro-based deicers and two CCM based deicers were identified by the project panel, as
they are readily available and representative of the commonly used products of the same type on
the market. The agro-based deicers were prepared by mixing the vendor-provided “concentrates”
with a 23.3 wt.% NaCl aqueous solution, at either 70:30 or 80:20 volume ratio, depending on
the vendor specification. In contrast, the CCM based deicers were used as received. Depending
on the specific test, a rock salt (white pellets from North American Salt Company, Overland
Park, Kansas) or reagent grade NaCl powder were employed as control. For the tests in this
work, each deicer or control was tested at least in triplicates. Following a chemical titration
method (Mohr’s method using 0.01 M silver nitrate solution), the chloride concentration of asreceived samples was measured. CCM 1 and CCM 2 deicers were found to feature a Cl content
of 58.9% and 59.6% (by atomic weight), respectively. These are slightly lower than reagent
grade NaCl, which features a theoretical Cl content of 60.7%. Agro 1, Agro 2, Agro 3 and Agro
4 “concentrates” feature a chloride concentration of 0.25 M, 2.05 M, 0.05 M and 0.55 M,
respectively. Note that a 23.3 wt.% NaCl would feature a chloride concentration of 3.99 M. Agro
1, Agro 2, and Agro 4 “concentrates” all contain beet sugar based byproducts and likely have a
certain amount of MgCl2 or CaCl2 added to enhance their ice melting performance. Agro 3
“concentrate” contains a non-beet-sugar byproduct and only trace amounts of chloride. Note that
Mohr's method is prone to interferences from compounds like sulfides, phosphates, etc.; as such,
the chloride concentrations will be further validated via calibrated silver/silver chloride sensors.
Eutectic Curves
A eutectic curve illustrates the freezing point temperature of an aqueous solution as a function of
its concentration. As such, solid products were made into solutions first before their eutectic
curves were obtained. In order to establish eutectic curves for deicer products, the test method
standardized by ASTM International for automotive coolants (ASTM D1177–07) was adopted
[19]. The test apparatus consisted of a plastic flask with deicer solution (100 mL), a stirrer made
up of stainless operated by wiper motor (60 to 80 stokes per min), and a thermostat coupled with
a data logger to measure temperature readings (for every second). The test apparatus was kept in
a state-of-the-art temperature-regulated environmental chamber and temperature of the room was
reduced constantly until the deicer solution freezes or supercools. The cooling rate of the
solution was approximately 0.5C/min. According to the standard test protocol, “the freezing
point is taken as the intersection of projections of the cooling curve and the freezing curve. If the
solution supercools, the freezing point is the maximum temperature reached after supercooling”
[19].
As shown in Figure 1a, at the eutectic point, there exists equilibrium between ice, salt and a
solution with a specific concentration. This specific concentration is called the eutectic
Muthumani and Shi
6
concentration and the temperature at which this equilibrium is found is called the eutectic
temperature. Above the eutectic concentration the excess deicing chemical crystalizes out due to
the saturation of liquid. In other words, the freezing point of the solution decreases with
increasing concentration up to the eutectic concentration. The freezing point of the solution
decreases with the increase in the concentration beyond the eutectic concentration [20]. As
shown in Figure 1a, the lowest freezing point (a.k.a., eutectic temperature) for NaCl is -6oF (21oC) at a concentration of 23.3% by weight of solution.
Modified SHRP Ice Melting Test
Modified SHRP ice melting tests were conducted in a Plexiglas chamber in a 12 ft.×14 ft. stateof-the-art temperature-regulated environmental chamber using de-ionized water [23]. The tests
were conducted at 15°F (-9°C) and 25°F (−4°C), respectively, with triplicate samples tested for
each combination of deicer type and temperature. For testing solid deicers, 4.170±0.005 g of
deicer was evenly applied over the ice sample. For testing liquid deicers, 3.8 mL of deicer is
applied evenly over the ice surface with a syringe. After 10, 20, 30, 45, and 60 min respectively,
the liquid volume is removed and volumetrically measured with a calibrated syringe. Solid rock
salt and 23.3% by weight of liquid rock salt was used as the control for CCM and agro-based
deicers respectively.
DSC Measurements
The differential scanning calorimetry (DSC) thermogram was obtained for each deicer to
quantify its thermal properties, using a Q200 apparatus (TA Instruments, Salt Lake City, Utah).
Solid deicers were made into 23.3 wt.% aqueous solutions first. Subsequently, these aqueous
solutions and the liquid deicers were further diluted by de-ionized water, at 1:2 volume ratio.
Subsequently, approximately 10-μL samples were pipetted into an aluminum sample pan and
hermetically sealed for DSC measurements. The DSC measures the amount of thermal energy
that flows into a deicer sample during the solid/liquid phase transition. The thermograms were
measured in the temperature range of 25 to -60°C (77 to -76°F) with a cooling/heating rate of
2°C (3.6°F) per minute. The first peak at the warmer end of the heating cycle thermogram was
used to derive the characteristic temperature of the liquid tested (Tc). In field practice, the
effective temperature is the lowest temperature limit at which the material remains effective
within 15-20 minutes of application and is the lowest temperature a deicer should be used to
achieve effective ice melting [26, 27]. Since the effective temperature can vary as a function of
road weather scenario and subjective observation, it is necessary to establish a more reproducible
and measurable alternative parameter to discriminate different products in terms of their lowest
working temperature. To this end, Tc is employed to indicate the “effective temperature” of
liquid deicers. This is based on the observation that Tc corresponds to the temperature threshold
below which ice crystals start to form in the diluted deicer solution and above which there is no
presence of ice crystal and thus no risk of slippery pavement. The enthalpy of fusion (H,
Muthumani and Shi
7
integrated surface area of the characteristic peak) is another parameter derived from the DSC
thermogram [28].
Corrosion to Carbon Steel
This work employed two different types of corrosion test methods, one of which was a
gravimetric method and the other was an electrochemical method. For both tests, solid deicers
and liquid deicers were made into their corresponding test solution, assuming a 3:100 dilution
ratio by weight and by volume, respectively. The gravimetric method followed the NACE
Standard TM0169-95 as modified by the PNS Association [27], but used de-ionized water in
place of distilled water. Three replicate 1.38″×0.56″×0.11″ ASTM F436, Type 1 TSI® steel
washers with a Rockwell Hardness of C 38–45 were used in each test solution and in the control
solutions (de-ionized water and a 3% NaCl aqueous solution) for testing. The average crosssection loss result in MPY (milli-inch per year) was translated into a percentage, or percent
corrosion rate (PCR, with no unit), in terms of the 72-h average corrosivity of the deicer solution
relative to solid salt (NaCl).
The electrochemical method was established to allow rapid determination of corrosion rate of
metals and to reveal information pertinent to the corrosion and inhibition mechanisms. Corrosion
of the CCM and agro-based deicers to ASTM A36 mild steel coupons were measured using a
Gamry Instruments® Potentiostat and a conventional three-electrode system. The steel coupon, a
platinum mesh, and a saturated calomel electrode (SCE) were employed as the working
electrode, counter electrode, and reference electrode, respectively. At 24 h of continuous
immersion, the potentiodynamic polarization curve of three to five steel specimens in each
diluted deicer was taken respectively. The current-potential plot of the steel in deicer solution
was measured as an external potential signal (DC perturbation) was applied within ±150 mV
range of its open circuit potential at a sweeping rate of 0.2 mV/s. The resulted polarization curve,
potential (E, in mV) as a function of logarithm of current density (i, in μA/cm2), was then used
to derive the corrosion potential (Ecorr) of the steel in the specific solution and its instantaneous
corrosion rate in terms of corrosion current density (icorr). These parameters were taken from the
point where the anodic current density (ia) equals the cathodic current density (ic) on the working
electrode (i.e., mild steel).
Muthumani and Shi
8
(a)
30
25
-4
-9
15
CCM 2
10
-14
5
CCM 1
0
-19
NaCl (reagent
grade)
-5
-10
-24
-15
-20
0%
5%
10%
15%
20%
25%
30%
Solution Concentration (% by weight)
(b)
35%
-29
40%
Temperature (oC)
Temperature (oF)
20
Muthumani and Shi
9
30
25
-4
-9
15
10
-14
5
Agro 3
0
-19
Agro 4
-5
Temperature (oC)
Temperature (oF)
20
Agro 1
-10
-24
Agro 2
-15
NaCl (reagent
grade)
-20
0%
5%
10%
15%
20%
25%
30%
35%
-29
40%
Solution Concentration (% by weight)
(c)
Figure 1: Eutectic curve for: a) sodium chloride (NaCl) aqueous solution; b) complex
chloride/mineral based deicers; and c) agro-based deicers
RESULTS AND DISCUSSION
Ice Melting Behavior and Thermal Properties
Figure 1b and Figure 1c illustrate the eutectic curves of CCM and agro-based deicers,
respectively. For the agro-based deicers, each weight concentration was prepared as follows. The
starting solution was an alternative to 23.3% NaCl, by replacing 20% or 30% of the 23.3% NaCl
brine with agro-based “concentrate”, depending on the vendor specification. Then the solutions
of other concentrations were made by assuming a given dilution or evaporation ratio of this
starting solution. This was designed to mimic the scenario that occurs on the pavement after the
application of the liquid deicer. Reagent grade NaCl (Fisher Scientific) was used as a control for
both solid and liquid deicers.
The experimental results in Figure 1b and Figure 1c reveal that CCM based deicers did not show
significant benefits in depressing the freezing point (relative to NaCl) whereas the use of agrobased additives in place of 23.3% NaCl brine at 20% or 30% volume ratio significantly depress
its freezing point. As shown in Figure 1b, the lowest freezing point of CCM 1 and CCM 2 are -
Muthumani and Shi
10
6.61oF and -6.7oF) at a concentration of 27% and 25% by weight of solution, similar to the
freezing point of 23% NaCl brine (-6oF). Figure 1c shows that the measured freezing point of
agro-based products ranges between -18.4oF (-28.13oC) and -9.52oF (-23.07oC), i.e., significantly
lower than the freezing point of 23% NaCl brine (-6oF (-21oC). These findings are consistent
with the previous findings that agro-based products act as a freezing point depressants [21, 22].
Figure 2 shows the average icemelt per gram of CCM or per mL of agro-based deicers at 15°F (9°C) and 25°F (−4°C). At 60 min, all the unit icemelt values were higher than 1 g/g for solid
deicers and higher than 1 mL/mL for liquid deicers, confirming that the deicers used were
effective and did not refreeze at 15°F (-9°C) or higher temperatures. CCM based deicers
produced more icemelt than the agro-based deicers, regardless of the testing temperature. This is
reasonable considering that the melting power of liquid deicers had been diluted by the water in
them. All liquid deicers achieved most of their melting potential within 10 minutes of application
onto the ice. In contrast, all solid deicers need sufficient time (more than 60 minutes) to achieve
its full potential which is consistent with the previous findings [24]. It can be noted that CCM
based deicers produced slightly high volume of ice melt at 15°F (-9°C) relative to rock salt and
the differences are not always statistically significant. Such benefits diminished at 25°F (−4°C).
For all three solid deicers tested, the unit icemelt exhibited a very strong linear relationship with
time. At 15°F (-9°C), CCM 1, CCM 2, and rock salt featured a melt rate of 0.898, 0.828, and
0.728 mL/g/min, respectively. At 25°F (−4°C), CCM 1, CCM 2, and rock salt featured a melt
rate of 1.471, 1.411, and 1.471 mL/g/min, respectively.
For agro-based deicers, the volume of ice melt revealed mixed results when compared with its
control at both temperatures. For instance, agro-based deicer 3 produced slightly more ice melt
and agro-based deicer 2 produced slightly less ice melt than the rock salt brine at 15°F (-9°C)
and 25°F (−4°C). There is no significant difference between agro-based deicer 1, agro-based 4
and rock salt brine in terms of ice melting capacity. These results suggest that the agro-based
additives may act as ‘cryoprotectants’, which tend to inhibit freezing without melting the ice
[20]. It remains unclear whether the agro-based additives provide other benefits to the snow and
ice control operations, such as weakening the microstructure of ice formed on pavement or
improving the longevity of deicer on pavement. These potential mechanisms merit further
investigation in future work. A recent study demonstrated that anti-icers not only depress the
freezing point of the solution on pavement but also physically weaken the ice on pavement [25].
Table 1 presents the characteristic temperature and enthalpy of fusion of the CCM and agrobased deicers, both derived from the DSC thermograms. The results indicate that CCM and agrobased deicers do not exhibit significantly lower characteristic temperature than reagent-grade
NaCl. The Tc of agro-based deicer ranges between 30oF (-1oC) and 25oF (-4oC), which is
consistent with a previous study which found that Tc of one agriculturally based product to be
23°F (-5°C) [2]. Note that the coefficient of variance (COV) for Agro 2 and Agro 4 were high,
implying the challenge in obtaining consistent and uniform samples from these two beet-based
Muthumani and Shi
11
liquid deicers. The same quality assurance issue applies to CCM 1, a solid deicer. The enthalpy
of fusion, H, ranges from 89 J/g to 176 J/g, for CCM and agro-based deicers, all of which are
lower than that of reagent grade NaCl (197 J/g). This suggests that the amount of thermal energy
corresponding to the aqueous brine solution’s liquid/solid phase transition is reduced; in other
words, it is thermodynamically easier to freeze a solution with lower H value. Note that the least
powerful deicer (deionized water) would feature a high H value of 345 J/g [28].
Figure 3 helps to further explore the correlations between various thermal properties and ice
melting parameters. The results reveal that the solid deicers disrupt all the potential correlations.
As shown in Figure 3a, a very strong positive linear relationship exists between the eutectic
temperature (Te) and the characteristic temperature (Tc) of the liquids, indirectly confirming the
validity of using DSC thermograms to assess liquid deicers. The solids deviate from this linear
relationship since they were made into liquids first before being tested for their Tc. Figure 3b
suggests that ice melting capacity (IMC) at 25F, 60 min decreases exponentially with the H
value of liquid deicers. Once removing solid deicers from Figure 3c, a very strong positive linear
relationship exists between the IMC at 25F, 60 min and the IMC at 15F, 60 min. Figure 3d
suggests that ice melting capacity (IMC) at 15F, 60 min fails to exhibit a strong correlation with
the Te value of liquid deicers.
Muthumani and Shi
12
A
5
10
4
8
3
Product A1
2
Product A2
1
Rock salt
6
Product A1
4
Product A2
2
0
Rock salt
0
10
20
30
45
Time (Min)
60
10
C
15F
20
30
45
Time (min)
60
D
25F
4
Product B1
2
Product B2
1
Product B3
Icemelt (mL)
3
Icemelt (mL)
B
25F
Icemelt (mL)
Icemelt (mL)/gm of deicer
15F
3
Product B1
Product B2
2
Product B3
1
Product B4
0
10
20
30
45
Time (min)
60
Rock salt
Product B4
0
10
20
30
45
Time (min)
60
Rock salt
Figure 2: Temporal evolution of deicer performance measured from a Modified SHRP Ice Melting Test: A) 15oF, B) 25oF –
complex chloride/mineral based deicers; C) 15oF, D) 25oF - agro-based deicers
Muthumani and Shi
13
Table 1: Comparisons between thermal property parameters obtained from DSC thermograms and eutectic parameters and ice melting
capacities
Characteristic Temperature
Peak
Enthalpy of fusion (J/g)
Eutectic Curve
Ice Melt
Eutectic
Concentrati
on (wt.%)
60 min
@ 15oF
60 min
@ 25oF
Deicer
Original
state
Average F
(oC)
COV
Average
COV
Eutectic
Temperature
o
F (oC)
CCM 1
Solid
28.0 (-2.2)
20%
162.2
8%
-6.6 (-21.4)
27%
4.5
7.1
CCM 2
Solid
22.9 (-5.0)
1%
89.4
4%
-6.7 (-21.5)
25%
4.2
7.2
Agro 1
Liquid
24.8 (-4)
3%
138.7
3%
-18.6 (-28.1)
27%
1.5
2.6
Agro 2
Liquid
30.4 (-0.9)
42%
156.1
7%
-9.5 (-23.1)
26%
1.4
2.4
Agro 3
Liquid
25.4 (-3.7)
4%
136.1
6%
-17.9 (-27.7)
24%
1.9
3.2
Agro 4
Liquid
28.1 (-2.2)
23%
176.1
4%
-15.4 (-26.3)
27%
1.5
2.5
NaCl (reagent)
Solid
23.5 (-4.7)
2%
197.7
3%
-6.3 (-21.3)
23%
3.9
-
Rock Salt (23.3 wt%)
Solid
-
-
-
-
-
-
1.6
2.6
Rock Salt
Solid
-
-
-
-
-
-
3.5
7.0
o
(ml/g for solid,
ml/ml for liquid)
Muthumani and Shi
14
(a)
(b)
(c)
(d)
Figure 3: Correlations between various thermal properties and ice melting parameters: (a)
Te vs. Tc; (b) H vs. IMC25F, 60 min; (c) IMC25F, 60 min vs. IMC15F, 60 min; (d) IMC15F, 60 min vs. Te.
Note only the products that deviate from a strong one-to-one relationship are labeled.
Corrosion to Carbon Steel
Table 2 shows the gravimetric and electrochemical corrosion test results for CCM and agrobased deicers. The gravimetric test revealed that the CCM deicers feature slightly lower
corrosivity to carbon steel than solid NaCl, whereas most agro-based deicers (except Agro 1)
Muthumani and Shi
15
feature much lower corrosivity to carbon steel than both solid NaCl and 23.3% NaCl brine.
Figure 3 presents some representative digital photos of steel washers after the cyclic exposure to
various deicer solutions, with the rustier steel surface generally corresponding to the more
corrosive deicer solution. The gravimetric test provides the average corrosion rate of ASTM
F436, Type 1 TSI® steel washers over the 72 hours of cyclic immersion, whereas the
electrochemical test provides the instantaneous corrosion rate of ASTM A36 mild steel coupons
at 24 hours of continuous immersion. As such, the corrosion rates measured via the
electrochemical method exhibit significantly different trends than those via the gravimetric
method. The latter features cyclic immersion and thus is more representative of the field scenario
of metallic corrosion.
Yet, the electrochemical data are useful to shed light on corrosion inhibition mechanism. For
instance, Table 3 presents the corrosion potential (Ecorr) of steel coupons in CCM and agro-based
deicers. While the beet-based additives do not significantly alter the corrosion potential of carbon
steel, the other type of additives (in Agro 3) moved the corrosion potential to a much more
positive level, implying anodic type inhibitor at work. Figure 4 illustrates the potentiodynamic
polarization curves of three steel coupons exposed to 3% NaCl solution and those exposed to 3%
Agro 3 deicer solution.
Muthumani and Shi
16
Table 2: Gravimetric and electrochemical test results for CCM and agro-based deicers
Gravimetric Test
Deicer
3% CCM 1
3% CCM 2
3% Agro 1
3% Agro 2
3% Agro 3
3% Agro 4
3% NaCl
3% of 23.3% NaCl
DI Water
Average
Percentage
Corrosion
Original state
Corrosion
Rate
Rate (%)
(MPY)
Solid
Solid
Liquid
Liquid
Liquid
Liquid
Solid
Liquid
Liquid
50.5
46.2
42.8
18.7
20.3
29.5
56.3
53.3
5.0
82.0
74.1
80.2
30.8
34.0
52.9
100
85.8
0
Electrochemical Test
Ecorr
(mV,
SCE)
-724.0
-748.3
-714.0
-748.0
-497.3
-727.0
-733.7
-
Average
icorr
Corrosion
(µA/cm2)
Rate
(MPY)
11.3
40.3
18.7
26.8
24.7
43.3
32.9
-
Figure 4: Typical digital photos of steel coupons after the gravimetric test
5.2
18.4
8.5
12.2
11.2
19.8
15.0
-
Muthumani and Shi
17
Figure 5: Potentiodynamic polarization curves of carbon steel coupons subjected to 3%
NaCl or 3% Agro 3 deicer, at 24 hr of continuous immersion
CONCLUSIONS


CCM based deicers do not exhibit significantly better ability to lower the freezing point
of water when compared with solid NaCl. However, they feature slightly better ice
melting capacity at 15oF (-9°C) than solid NaCl but the differences are not always
statistically significant.
Agro-based additives seem to significantly lower the freezing point of 23 wt.% NaCl
brine. However, they do not significantly improve the ice melting capacity at 15oF or
25oF. These results suggest that the agro-based additives may act as ‘cryoprotectants’,
which tend to inhibit freezing without melting the ice. Additional research is needed to
elucidate their specific working mechanism in anti-icing and deicing operations and to
better understand the observed difference between their thermodynamics and kinetics.
Muthumani and Shi




18
CCM and agro-based deicers do not exhibit significantly lower characteristic temperature
than reagent-grade NaCl. The enthalpy of fusion, H, ranges from 89 J/g to 176 J/g, for
CCM and agro-based deicers, all of which are lower than that of reagent grade NaCl (197
J/g). This suggests that it is thermodynamically more difficult to freeze the NaCl brine
containing trace amount of other chlorides or agro-based additives.
A very strong positive linear relationship exists between the eutectic temperature (Te) and
the characteristic temperature (Tc) of the tested liquids, indirectly confirming the validity
of using DSC thermograms to assess liquid deicers.
The corrosion rates measured via the electrochemical method exhibit significantly
different trends than those via the gravimetric method; the latter features cyclic
immersion and thus is more representative of the field scenario of metallic corrosion.
The gravimetric method reveals that CCM based deicers exhibit slightly lower corrosivity
to carbon steel than NaCl and agro-based additives exhibit significant benefits in
reducing the corrosivity of 23 wt.% NaCl brine. The electrochemical method reveals that
while the beet-based additives do not significantly alter the corrosion potential of carbon
steel, the other type of additives moved the corrosion potential to a much more positive
level, implying anodic type inhibitor at work.
ACKNOWLEDGMENTS
The authors would like to thank Minnesota Department of Transportation and Clear Roads for
funding this study. The authors acknowledge the guidance provided by the technical panel
members including Colleen Bos, Ron Wright, Tom Peters, Michael Lashmet, Kim Linsenmayer,
Tim Peters, Larry J. Gangl, Pat Casey and Mike Mattison. Also, special thanks to our students
Bryan Smith, Vikina Martinez, Yao Lin, Chinomso Emmanuel and Scott Jungwirth for their in
conducting laboratory experiments.
REFERENCES
[1] FHWA (2013). "Road Weather Management Program - Snow and Ice." Website last updated
July 2, 2013 http://ops.fhwa.dot.gov/weather/weather_events/snow_ice.htm.
[2] Fay, L., K. Volkening, C. Gallaway and X. Shi (2008). Performance and impacts of current
deicing and anti-icing products: User perspective versus experimental data. Proc., 87th
Annual Meeting of Transportation Research Board, Transportation Research Board
Washington, DC.
[3] Cuelho, E., J. Harwood, M. Akin and E. Adams (2010). "Establishing best practices for
removing snow and ice from California roadways." Presentation to the Winter
Maintenance Committee at the 89th.
Muthumani and Shi
19
[4] Shi, X. M., K. Fortune, R. Smithlin, M. Akin and L. Fay (2013). "Exploring the performance
and corrosivity of chloride deicer solutions: Laboratory investigation and quantitative
modeling." Cold Regions Science and Technology 86: 36-44.
[5] Johnson, J. (2014). "Cost of Corrosion - Motor Vehicles." CC Technologies Laboratories,
Inc., Dublin, Ohio http://corrosionda.com/pdf/motorvehicles.pdf(Last accessed on April
8, 2014).
[6] Li, Y., Y. Fang, N. Seeley, S. Jungwirth, E. Jackson and X. Shi (2013). "Corrosion by
Chloride Deicers on Highway Maintenance Equipment." Transportation Research
Record: Journal of the Transportation Research Board 2361(1): 106-113.
[7] Shi, X., L. Fay, Z. X. Yang, T. A. Nguyen and Y. J. Liu (2009). "Corrosion of Deicers to
Metals in Transportation Infrastructure: Introduction and Recent Developments."
Corrosion Reviews 27(1-2): 23-52.
[8] Shi, X., Y. Liu, M. Mooney, M. Berry, B. Hubbard and T. A. Nguyen (2010). "Laboratory
investigation and neural networks modeling of deicer ingress into Portland cement
concrete and its corrosion implications." Corrosion Reviews 28(3-4): 105-154.
[9] Fay, L. and X. Shi (2012). "Environmental impacts of chemicals for snow and ice control:
State of the knowledge." Water, Air, & Soil Pollution 223(5): 2751-2770.
[10] Shi, X., J. M. Staples and O. Stein (2005). Managing winter traction materials on roadways
adjacent to bodies of water: challenges and opportunities. Environmental Stewardship in
Transportation through Waste Management, Materials Reuse and EMS: 2005 Summer
TRB Committee ADC60 Conference.
[11] Fay, L. and X. Shi (2010). "Laboratory investigation of performance and impacts of snow
and ice control chemicals for winter road service." Journal of Cold Regions Engineering
25(3): 89-114.
[12] Pan, T. Y., X. D. He and X. M. Shi (2008). "Laboratory Investigation of Acetate-Based
Deicing/Anti-Icing Agents Deteriorating Airfield Asphalt Concrete." 2008 Journal of the
Association of Asphalt Paving Technologists, Vol 77 77: 773-793.
[15] Nixon, W. A. and A. D. Williams (2001). A Guide for Selecting Anti-Icing Chemicals,
Version 1.0, IIHR Technical Report.
Muthumani and Shi
20
[16] Albright, M. (2005). "Changes in water quality in an urban stream following the use of
organically derived deicing products." Lake and Reservoir Management 21(1): 119-124.
[17] Cheng, K. C. and T. F. Guthrie (1998). "Liquid Road Deicing Environment Impact.
Levelton Engineering Ltd." Prepared for the Insurance Corporation of British Columbia,
File number 498-0670.
[18] Janke, G. A. and W. D. Johnson Jr (1999). Deicing composition and method, U.S. Patent
No. 5635101.
[19] Hallberg, S.-E., A. Gustafsson, A. Johansson and E.-L. Thunqvist (2007). Anti-Skid
Treatment Tests with Glucose, Fructose, and Unrefined Sugar. Transportation Research
Board 86th Annual Meeting.
[20] Taylor, P., J. Verkade, K. Gopalaakrishnan, K. Wadhwa and S. Kim (2010). "Development
of an Improved Agricultural-based Deicing Product." Institute for Transportation, Iowa
State University.
[21] Fischel, M. (2001). "Evaluation of selected deicers based on a review of the literature."
[22] Kahl, S. (2004). "Agricultural By-Products for Anti-Icing and De-Icing Use in Michigan."
In: Transportation Research Board (ed.), Proc. 6th Intl. Symposium on Snow Removal
and Ice Control Technology. Transportation Research Circular E-C063: Snow and Ice
Control Technology SNOW04-009, pp. 552-555.
[23] Pilgrim, K. (2013). "Determining the Aquatic Toxicity of Deicing Materials." Final Report,
Clear roads. December 2013 http://www.clearroads.org/downloads/Determine-theToxicity-of-Deicing-Materials_Final%20Report_12-30-2013.pdf.(Last accessed-July 30,
2014).
[24] Fu, L. P., R. Omer and C. Z. Jiang (2012). "Field Test of Organic Deicers as Prewetting and
Anti-Icing Agents for Winter Road Maintenance." Transportation Research
Record(2272): 130-135.
[25] Shi, X., L. Fay, K. Fortune, R. Smithlin, M. Johnson, M. Peterson, A. Creighton, Z. Yang
and D. Cross (2011). "Investigating Longevity of Corrosion Inhibitors and Performance
of Deicer Products under Storage or after Pavement Application." Final report prepared
for the Washington State Department of Transportation and the Transportation Pooled
Fund led by the Pacific Northwest Snowfighters Association. June 2011.
Muthumani and Shi
21
[26] Shi, X., M. Akin, J. Huang, Y. Zhang, S. Jungwirth, Y. Fang, A. Muthumani and P. Yi
(2013). "Evaluation and Analysis of Liquid Deicers for Winter Highway Maintenance
Operations." Final report prepared for the Western Transportation Institute. Bozeman,
MT. December 2013.
[27] PNS (2010). "Pacific Northwest Snowfighters Snow and Ice Control Chemical Products
Specifications and Test Protocols for the PNS Association of British Columbia,
Colorado, Idaho, Montana, Oregon and Washington."
[28] Akin, M. and X. Shi. Development of Standard Laboratory Testing Procedures to Evaluate
the Performance of Deicers. ASTM Journal of Testing and Evaluation, 2012, 40(6),
1015-1026.