Assessment of the Use of Hand Warmer for Nitrate Retardation in Porous Media

ABSTRACT This paper assessed the possible reuse of used hand warmer contents (HWCs) as a nitrate retardation material in porous media. Under saturated and unsaturated flow conditions, column experiments with a pulse source of potassium nitrate solution were conducted in silica sand and Andisols with a mixture of HWCs. Mixture of HWCs and flow velocities in a flow field were varied in order to investigate the relation between the degree of retardation and the amount of HWCs under the approximately one order range of Reynolds number. Experimental breakthrough curves measured by capillary electrophoresis were analyzed using temporal moments to estimate the dispersion coefficient, retardation factor and distribution coefficient. The results of column experiments showed that the increase in the amount of HWCs added resulted in the increase of the first two temporal moments and retardation factor in both soils, particularly in silica sand. Reuse of HWCs would be useful to alleviate an immediate leaching of nitrate with high concentration to groundwater in a sandy medium rather than in an aggregated medium.

Assessment of the Use of Hand Warmer for Nitrate Retardation in Porous Media

Kazuya INOUE*, Ikko IHARA*, Akari YOSHINO**, Tsutomu TANAKA*

*Graduate School of Agricultural Science, Kobe University, Hyogo 657-8501, Japan

**Graduate School of Agricultural and Life Science, The University of Tokyo, Tokyo 113-8657, Japan

ABSTRACT

This paper assessed the possible reuse of used hand warmer contents (HWCs) as a nitrate retardation material in porous media Under saturated and unsaturated flow conditions, column experiments with a pulse source of potassium nitrate solution were conducted in silica sand and Andisols with a mixture of HWCs Mixture of HWCs and flow velocities in a flow field were varied in order to investigate the relation between the degree of retardation and the amount of HWCs under the approximately one order range of Reynolds number Experimental breakthrough curves measured by capillary electrophoresis were analyzed using temporal moments to estimate the dispersion coefficient, retardation factor and distribution coefficient The results of column experiments showed that the increase in the amount of HWCs added resulted in the increase of the first two temporal moments and retardation factor in both soils, particularly in silica sand Reuse of HWCs would be useful to alleviate an immediate leaching of nitrate with high concentration to groundwater in a sandy medium rather than in an aggregated medium

Keywords: column experiment, hand warmer, nitrate, retardation

INTRODUCTION

Overloading agricultural soils with nitrogen leads to rising nitrate concentration in groundwater, while enhancing crop growth by the application of nitrogen gives economic benefits in many parts of the world Several strategies controlling nitrate levels in subsurfaces have been taken into consideration due to links with infantile

methemoglobinemia and a risk of developing gastric cancer (Forman et al., 1985)

However, because nitrate can readily be transported beneath the soil zone without ion exchange or adsorption to soil surface, high concentrations of nitrates can be found in groundwater systems, especially in sandy aquifers Nitrate contamination has necessitated reliance on the development of several treatment processes including ion exchange, chemical denitrification and reverse osmosis with varying degrees of efficiency and cost (Kapoor and Viraraghavan, 1997)

Hand warmers have been used for approximately 50 years in Japan, while the use of hand warmers during cold weather has recently become increasingly popular in Europe and other areas of the world Iron powder, activated carbon, vermiculite, sodium chloride and water are the main components of hand warmers, and almost all of which have been incinerated as combustible wastes In 2008, approximately 1.5 × 109 hand warmer sheets made in Japan were sold Under the assumption that the mean weight of one hand warmer is 50 g, approximately 7.3 × 104 tons of hand warmers were disposed annually To our knowledge, there are no attempts of the reuse of the contents of used hand warmers for the purpose of the retardation of nitrate transport in soils and the

Table 1 - Specifications of soils used in column experiments

(g/cm3)

Mean particle size (cm)

Uniformity coefficient

Hydraulic con- ductivity (cm/s)

Porosity

Andisol 2.40 0.076 2.74 0.012 0.67

reduction of its peak concentration However, a few studies reported other technologies

to be applied for the nitrogen reduction (Ohe et al., 2003; Moreno et al., 2005) This

paper explored the possible application of used hand warmer contents (HWCs) toward delaying nitrate migration and avoiding nitrate leaching with a high concentration through column experiments Specifically the effect of the ratio of HWCs to soils on the degree of retardation associated with a soil type was investigated using temporal moment approaches

MATERIALS AND METHODS

Column experiments using hand warmer contents

After using hand warmers (Kiribai Chemical Co., Japan), the contents were collected and washed with water to remove sodium chloride, which is not suitable for the production of food in a field For column experiments, silica sand with a low uniformity coefficient was selected in order to simulate a sandy aquifer In addition, Andisols, which are volcanic ash soils and have unique properties such as a low bulk density

(Maeda et al., 2008), were taken from a field of corn, dried at 110ºC and passed through

a 2-mm sieve Grain sizes less than 0.2 mm were excluded to avoid the adsorption of nitrate onto the surface of silt or clay and to clarify the effect of soils with HWCs on the nitrate retardation Physical properties for both soils (silica sand and Andisol) are listed

in Table 1

In column experiments, used HWCs were mixed with soil of interest in order to examine the relation between the amount of HWCs and the degree of retardation of nitrate From the practical viewpoint, mixture ratios of HWCs were set to 0%, 5% and 7.5% by weight for both soils As both soils had different porosities, only in Andisol column, additional experiment under the mixture ratio of 14.3% was conducted while the total amount of HWCs was equal to 7.5% mixture ratio in silica sand column These soils with HWCs were completely saturated before packing to avoid the entrance of air and filled into the column (30 cm in length and 10.6 cm in diameter) layer by layer in increments of 2 cm Each layer was compacted to adjust the dry density prior to filling the next layer, providing the porosity estimation in each experiment indirectly from the measurements of the particle density and the dry soil bulk density In Fig 1, the detailed column design is illustrated

After packing, as shown in Fig 1(a), water was applied to the column up to a specific level controlled by constant head reservoirs at the top and bottom of the saturated media, while maintaining the saturated condition of porous media Steady saturated flow field was established in the column when fluctuations in the observed drainage rate from the bottom reservoir became negligible A volume of 80 cm3 of KNO3 solution was applied

to the top of the column to produce a pulse input with an initial concentration of 300 g/m3 of NO3-N Prescribed solute source condition was designed to clarify the transient

Fig 1 - Schematic diagram of an experimental column system

concentrations of NO3-N and to effectively provide the parameter identification relevant

to solute transport using temporal moment approaches, which are stated in the following subsection Pore water samples at the end of the column were taken at specific intervals Nitrate ion in the pore water samples was analyzed by capillary electrophoresis (G1600A, Agilent technologies, USA) The basic anion buffer and a fused silica capillary with 104 cm in length and 50 μm internal diameter were obtained from Agilent technologies The temperature controlled cartridge for fused silica capillary was set at 15ºC

Column experiments under unsaturated conditions were carried out in a similar manner

to investigate the impact of water content on the nitrate behavior As shown in Fig 1(b)

in unsaturated experiments, instead of constant head reservoir, suction was applied at the bottom of the column to keep the moisture content inside the column uniform while water was sprinkled over the top of the medium until the outflow from the bottom of the column equaled the volume of the water input Prior to an application of solute pulse, samples were adjusted to steady state water flow condition with a Darcy flux of approximately 3.0 × 10-3 cm/s and 2.0 × 10-3 cm/s for silica sand and Andisol, respectively

Temporal moment analysis

The classical equilibrium model for one-dimensional solute transport during steady state flow in a homogeneous porous medium can be expressed in the form of advection and dispersion equation:

x

c v x

c v t

c R

∂

∂

−

∂

∂

=

∂

∂

2

2

where c is the concentration of solute, x is the coordinate, t is the time, R is the retardation factor, v is the average pore water velocity and α is the dispersivity

Temporal moment analysis is convenient to use as there is no need to solve the transport model in real time and space and is applied to characterize experimental breakthrough curves (BTCs) in terms of mean breakthrough time and degrees of spreading and asymmetry (Valocchi, 1985) The normalized temporal moments μn at a location x, are

defined as

=

=

n n

where M is the n-th order absolute temporal moment Dispersion coefficient D and n

dispersivity from temporal moments are calculated as (Valocchi, 1985)

v M v

⎟⎟

⎠

⎞

⎜⎜

⎝

⎛

−

=

1 2 μ ξ

where ξ is the distance between the source and the observation location Das and P Kluitenberg (1996) derived the following theoretical first temporal moment using Laplace formation of Eq (1)

v xR

5

μ (4)

where t0 is the pulse duration In all experiments, t0 was set to 60 seconds Under the

assumption that the sorption-desorption isotherms are reversible and single-valued, the retardation factor and the distribution coefficient K are given by: d

( t )v x K d (R ) b

R= μ1−0.50 / , = −1θ/ρ (5) whereθ is the volumetric water content and ρbis the soil bulk density

RESULTS AND DISCUSSION

Representative BTCs under different mixture ratios of HWCs as a function of pore

volume, which is calculated as vt/L where L stands for the column length, are shown in

Fig 2 In silica sand (Fig 2 (a)), observed BTCs exhibit the reduction of peak concentration and the retardation of peak time with the increase of the amount of HWCs while there is a slight change of the peak concentration in Andisol (Fig 2 (b)) except for the more pronounced tailing of the BTCs with the increase of HWCs The peak concentrations for Andisol are lower than those for silica sand This is because the degree of the solute dispersion in Andisol is larger due to the difference between the soil properties such as the uniformity coefficient and the mean particle diameter for both soils

As seen in Fig 2, the peaks of BTCs appear in the vicinity of the unity In order to clarify the effect of HWCs, temporal moment approaches are carried out to analyze BTCs in terms of the mean travel time and the solute dispersion for each flow condition

To compare the change of BTCs in the same flow condition of Reynolds number, the variation of first and second temporal moments as a function of the mixture ratio of

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35

0 0.5 1 1.5 2 2.5 3 3.5

Pore volume (-)

Andisol

0%

5%

7.5%

0

0.2

0.4

0.6

0.8

1

0 0.5 1 1.5 2 2.5 3 3.5

Pore volume (-)

Silica sand

0%

5%

7.5%

Fig 2 - Representative breakthrough curves for different mixture ratios of HWCs in

silica sand and Andisol columns

0

0.5

1

1.5

2

2.5

0 2 4 6 8 10 12 14 16

Mixture ratio of hand warmer contents (%)

(a)

Silica sand (Saturated) (Unsaturated) Andisol (Saturated) (Unsaturated)

0 2 4 6 8 10

0 2 4 6 8 10 12 14 16

Mixture ratio of hand warmer contents (%)

(b)

Silica sand (Saturated) (Unsaturated) Andisol (Saturated) (Unsaturated)

HWCs are plotted in Fig 3 (a) and (b), respectively In Fig 3, both temporal moments are normalized by a corresponding moment obtained in a column without HWCs Obviously, both moments tend to increase with the increase of HWCs, while the first and second temporal moments range from 1 up to 2.1 and 9.4, respectively This implies that for the purpose of inducing the spread and retardation of nitrate, the use of HWCs might be effective in soils

The variation of the dispersion coefficient as a function of Reynolds number for all cases as shown in Fig 4 confirms the spread of solute in both soils As expected, the dispersion coefficient increases as Reynolds number becomes larger Under the saturated conditions, the dispersion coefficient in Andisol indicates higher values than that in silica sand This feature reflects that the solutes spread larger during the course of transport in Andisol On the other hand, opposite results are shown in unsaturated columns since the degree of spreading of solutes depends on the water content as well

as the flow velocity

In order to quantify the degree of retardation, the retardation factor and the distribution coefficient were estimated based on Eq (5) The relation between the mixture ratio of HWCs and the retardation factor is shown in Fig 5 (a) The retardation factor tends to increase with increasing percentage of HWCs, demonstrating the potency of HWCs It

is inferred that nitrate retardation appears to be controlled by the activated carbon and vermiculite, which compose approximately 20% of the HWCs

10-3

10-2

10-1

Reynolds number (-)

Silica sand (Saturated) (Unsaturated) Andisol (Saturated) (Unsaturated)

Fig 4 - Relation between Reynolds number and dispersion coefficient

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

0 2 4 6 8 10 12 14 16

Mixture ratio of hand warmer contents (%)

(a) Silica sand (Saturated)(Unsaturated)

Andisol (Saturated) (Unsaturated)

-0.1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5

3 /g)

Retardation factor, R (-)

(b)

Silica sand (Saturated) (Unsaturated) Andisol (Saturated) (Unsaturated)

Fig 5 - Relation between (a) mixture ratio of HWCs and retardation factor and (b)

retardation factor and distribution coefficient

In addition, both the range and value of retardation factor in Andisol are smaller than those in silica sand for the same mixture ratio of soil and HWCs This finding is attributed to not only the complex geometry of the soil particles but the aggregate structure in Andisol (Eguchi and Hasegawa, 2008) Hand warmer contents may distribute locally in Andisol due to aggregate structure of soil particles within which nitrate passes through but HWCs may be included in small amounts, while in silica sand that does not have the nature of aggregation, HWCs are homogeneously distributed The difference of distributions of HWCs in both soils leads to the reduction of the possibility that nitrate adsorbs onto the HWCs Furthermore, the solute dispersion in Andisol is larger than that in silica sand because of the difference of physical properties for both soils such as the uniformity coefficient and the mean particle diameter, providing a low peak concentration and a large distribution space of solute in Andisol Therefore, it is inferred that nitrate in silica sand is more strongly adsorbed and retarded, as the degree

of adsorption onto HWCs depends on the solute concentration In a practical situation, natural degradation of nitrate concentration and the subsequent reduction of nitrate loading to groundwater are expected due to denitrification by microorganisms as the travel time of nitrate from the ground surface to groundwater becomes longer Therefore,

it is likely that the use of HWCs as a retardation-induced material will be significant

On the other hand, the specific surface of soils might be responsible for the difference in the degree of retardation In Fig 5 (b), distribution coefficient variations are plotted as a function of the retardation factor There are significant differences in the values of the distribution coefficient of two soils Given the fact that the distribution coefficient depends on the specific surface of the medium, Andisols provide the diversity of transport pathways, resulting in the low retardation values Consequently, sandy soils such as silica sand appear to be suited for the application of HWCs to induce the adsorption of nitrates and to alleviate an immediate leaching of nitrates to groundwater

In contrast, simply mixing the HWCs with Andisols may be less effective to retard nitrate migration due to the aggregate structure and the geometry of the interfaces

CONCLUSIONS

This study focused on the possible reuse of used hand warmer contents (HWCs) for retarding and spreading nitrate transport in two types of soils to avoid a prompt nitrate leaching to groundwater with a high concentration The column experiments allowed the observation of the nitrate behavior at various flow rates under the different conditions of mixture ratios of HWCs Temporal moments associated with observed breakthrough curves were computed to quantify the change of the mean travel time and the solute dispersion according to the mixture ratio of HWCs Moreover, temporal moments were applied to estimate the dispersion coefficient, the retardation factor and the distribution coefficient Results indicated that the retardation factor increased with the increase in the amount of HWCs and reached up to 3.3 and 1.7 in silica sand and Andisol, respectively In particular, the effect of HWCs on nitrate retardation in silica sand appeared to be significant Accordingly, HWCs have a potency to be an effective adsorbent for the nitrate transport in sandy media whereas porous media such as Andisols with an aggregate structure might hinder the ability of HWCs

REFERENCES

Das B S and Kluitenberg G J (1996) Moment analysis to estimate degradation rate

constants from leaching experiments, Soil Sci Soc Am J., 60(6), 1724-1731

Eguchi S and Hasegawa S (2008) Determination and characterization of preferential

water flow in unsaturated subsoil in Andisol, Soil Sci Soc Am J., 72(2), 320-330

Forman D., Al-Dabbagh S and Doll R (1985) Nitrate, nitrite and gastric cancer in

Great Britain, Nature, 313, 620-625

Kapoor A and Viraraghavan T (1997) Nitrate removal from drinking water - review, J

Environ Eng., 123(4), 371-380

Ohe K., Nagae Y., Nakamura S and Baba Y (2003) Removal of nitrate anion by

carbonaceous materials prepared from bamboo and coconut shell, J Chem Eng

Japan, 36(4), 511-515

Maeda M., Ihara H and Ota T (2008) Deep-soil adsorption of nitrate in a Japanese

Andisol in response to different nitrogen sources, Soil Sci Soc Am J., 72(3),

702-710

Moreno B., Gómez M A., González-López J and Hontoria E (2005) Inoculation of a submerged filter for biological denitrification of nitrate polluted groundwater: a

comparative study, J Hazard Mater., 117(2-3), 141-147

Valocchi A J (1985) Validity of the local equilibrium assumption for modelling

sorbing solute transport through homogeneous soils, Water Resour Res., 21(6),

808-820