Hydrological characteristics of some volcanic materials as affected by particle size distribution and internal porosity

Abstract

A number of internally porous inorganic amendments have been recently introduced to replace peat in rootzone mixes used for golf greens and sports fields. In Italy, volcanic rocks provide a large availability of internally porous materials such as pumice, lapillus and zeolite. Volcanic sands are currently used as substitutes of silica sand. In order to assess the suitability of these volcanic materials for sand amelioration and/or as construction materials, the present research was aimed at studying the water retention and release of these materials considering different particle size distributions and a non-porous silica sand as a reference. Water retention curves were constructed for the internally porous sands as: (a) raw material, (b) graded material with the same particle size distribution of a non-internally porous silica sand. Additionally, available water was determined on specific sand fractions namely coarse sand (0.5–1.0 mm) and medium sand (0.25–0.5 mm) of all the materials. The available water content of raw volcanic materials was higher than of graded sands made of the same materials, thus showing the expected effect of texture on water retention. The comparison of water retention properties of the internally porous materials and non-internally porous silica sand with the same particle size distribution, showed a significant contribution given by internal porosity to water retention. Volcanic coarse sand fraction showed higher water availability than volcanic medium sand fraction. Volcanic materials used as amendments could enhance the water availability of sand-based rootzones, but if used as raw materials the possibility exists for the final mix not to meet the specifications for sports use. The use of sole volcanic sands as rootzone material could have a major impact on water-holding capacity, and hence on management, although investigation on their particle stability is still needed.

Keywords:

• Lapillus
• porosity
• pumice
• water retention
• zeolite

Introduction

Due to their characteristics of maintaining sufficient drainage and aeration under intense traffic, sands are a major component in the construction of sports surfaces and their use is common for rootzone mixes, drainage backfilling and surface topdressing (Baker 1990, Petrovic et al. 1997). Size and uniformity of sand grains affect most aspects of the performance of a sand with particular reference to hydraulic conductivity and water retention (Baker 1990). Based on this assumption, specifications for acceptable particle size distribution have been published over the past decades in order to give turf managers a tool for the proper selection of sands with the required characteristics for a specific use in sports turfs (Baker 1990, USGA Green Section Staff 1993). Sands with a selected particle size distribution provide a good balance between water retention, stability and drainage, though the addition of amendments is frequently required to improve water and nutrient retention. Peat is commonly used for sand amelioration (Waddington 1992, Beard 2002) although, like many organic materials, it undergoes a natural decomposition that reduces its beneficial effects with time. To overcome this limit a number of internally porous inorganic materials are currently available on the market as alternatives to peat for sand amelioration. These are characterized by resistance to biodegradation, high porosity, good water retention and cation exchange capacity (Huang and Petrovic 1995, Waltz et al. 2003, Bigelow et al. 2004).

Pumice, zeolite and lapillus are internally porous volcanic materials largely available as natural deposits in some regions of Italy and sandy materials obtained from these naturally occurring rocks are increasingly used in sports surfaces construction instead of graded silica sands (Volterrani and Magni 2004, Magni et al. 2004). However a deeper knowledge of hydrological characteristics is needed in order to highlight the potentials of the aforementioned materials as soil amendments and/or as construction materials.

Most of the research work carried out on internally porous materials is related to their effect as amendments of sand-based rootzones. Huang and Petrovic (1995) reported hydraulic conductivity and plant-available water of a sand mix as being affected by the amendment particle size and incorporation rate. The amendment particle size was also found to significantly affect bulk density, total and aeration porosity of the mix. McCoy and Stehouwer (1998) found that water retention properties of two sand mixes were affected by sand texture, amendment type and incorporation rate. These same factors are also reported to affect capillary porosity and water retention of different sand mixtures (Bigelow et al. 2004).

The objective of this laboratory study was to determine and compare water retention and release properties of some internally porous volcanic materials tested as pure materials. Since the retention of water in a sandy material is directly related to the size of the pore channels and therefore controlled by grain size (Baker 1990) water retention properties were determined on materials with different particle size. In particular materials were subject to analysis as raw materials, as graded sands with a manipulated particle size distribution and as sand fractions with uniform particle size. In order to highlight the contribution to water retention given by internal porosity a non-internally porous silica sand was included in the study.

Materials and methods

Sandy materials obtained from the industrial sieving of the volcanic rocks lapillus, pumice and zeolite chabasite (zeolite) were sampled in three selected Italian production sites respectively at Cellere (Viterbo), Pitigliano (Grosseto) and Tessennano (Viterbo). A graded silica sand having a particle size distribution meeting the United States Golf Association specifications for rootzone mixes (USGA Green Section Staff 1993) was included in the study as control. All the materials were kindly provided by Europomice Srl (Milan, Italy). The particle size distribution of the four materials was determined by dry sieving with a standard mechanical sieve shaker. Mid Particle Diameter and Gradation Index, as defined by Baker (1990), were also determined. Results of particle size analysis are reported in Table I.

Table I. Particle size distribution, Mid Particle Diameter (D₅₀) and Gradation Index (D₉₀/D₁₀) of volcanic raw materials and a silica sand meeting the USGA specifications for texture.

		Lapillus	Pumice	Zeolite	Sand
	mm	^_______________________________________ g kg^−1 ^_______________________________________
Gravel	>4	8	12	49	0
	2–4	168	150	340	0
Very coarse sand	1.0–2.0	114	241	254	8
Coarse sand	0.5–1.0	105	218	94	183
Medium sand	0.25–0.5	129	148	71	690
Fine sand	0.15–0.25	115	55	37	99
Very fine sand	0.106–0.15	76	29	32	12
	0.053–0.106	132	41	41	5
Silt + Clay	<0.053	153	106	82	3
Mid Particle Diameter (D50)		0.3	0.8	1.6	0.4
Gradation Index (D90/D10)		75.7	47.2	49.3	2.5

Volcanic materials with no manipulation of the particle size distribution were identified as ‘raw’ materials. In order to minimize the effect of particle size distribution across different materials, volcanic sands were sieved, single fractions isolated and then reassembled in the appropriate quantity in order to reproduce the same particle size distribution of the silica sand. Materials included in the study with the particle size distribution reproducing that of the non-internally porous silica sand were identified as ‘graded’ materials.

Single sand fractions both from volcanic sands and silica sand, namely coarse (particle diameter ranging from 0.5 to 1.0 mm) and medium (particle diameter ranging from 0.25 to 0.5 mm) fractions were also included in the study and identified as ‘coarse’ and ‘medium’ materials respectively. Materials under investigation and the relative abbreviations are given below:

Table

a. Lapillus	raw material	Lp-Raw
b. Pumice	raw material	Pu-Raw
c. Zeolite	raw material	Ze-Raw
d. Silica sand	graded sand	Ss-Graded
e. Lapillus	graded sand	Lp-Graded
f. Pumice	graded sand	Pu-Graded
g. Zeolite	graded sand	Ze-Graded
h. Silica sand	coarse sand fraction (0.5–1.0 mm)	Ss-Coarse
i. Lapillus	coarse sand fraction (0.5–1.0 mm)	Lp-Coarse
j. Pumice	coarse sand fraction (0.5–1.0 mm)	Pu-Coarse
k. Zeolite	coarse sand fraction (0.5–1.0 mm)	Ze-Coarse
l. Silica sand	medium sand fraction (0.25–0.5 mm)	Ss-Medium
m. Lapillus	medium sand fraction (0.25–0.5 mm)	Lp-Medium
n. Pumice	medium sand fraction (0.25–0.5 mm)	Pu-Medium
o. Zeolite	medium sand fraction (0.25–0.5 mm)	Ze- Medium

In order to avoid any particle segregation sands were kept moist during laboratory handling. Water retention measurements were conducted using the pressure plate method (Klute 1986). A pressure chamber with a ceramic plate having an air entry pressure of 0.3 MPa (Soil Moisture Equipment Co., Santa Barbara, CA), was used to determine water retention at soil water potentials of −0.033, −0.05, −0.1 and −0.2 MPa. A second pressure chamber, with a ceramic plate having an air entry pressure of 1.5 MPa, was used to determine water retention at soil water potentials of −0.5, −1.0 and −1.5 MPa. All the materials were hand-packed into plastic rings (1 cm tall×5.2 cm internal diameter) and slowly saturated from the bottom up. After drainage of excess water, plates were placed in the pressure chamber at each intended pressure until no drainage flow was noticeable from the outlet. For the determination of gravimetric water content samples were removed from plates, weighed, oven dried at 105 °C for 24 h and allowed to cool in a desiccator before the determination of the dry weight. Bulk density was determined and volumetric water content was calculated as the product of the gravimetric water content and bulk density. Water retention curves were plotted from the matric potential applied and volumetric water content.

Available water content was calculated as the difference between the water held at −1.5 MPa, considered as permanent wilting point, and the water held at −0.033MPa, considered as field capacity. Water retention of ‘coarse’ and ‘medium’ textured materials was determined only at −0.033 and −1.5 MPa soil water potential and available water content calculated as above.

A completely randomized block experimental design with three replications was adopted. The results were tested using one way ANOVA and the least significant difference at P ≤ 0.01 was used to detect differences between means.

Results and discussion

None of the volcanic materials, as available on production site, had a particle size distribution meeting the recommendations issued for sports use. In particular, referring to Baker (1990), a Gradation index value of 8 is considered the upper limit for a sand to be sufficiently uniform for sports turf usage, therefore only the silica sand included in the study met this criterion (Table I). Referring to USGA specifications, unacceptable contents of gravel particles and excess of fines (silt and clay) were recorded. Furthermore, fractions falling in the range of coarse and medium sand did not add up to the minimum 60% by weight recommended by the same guidelines (USGA Green Section Staff 1993). In Lp-Raw particles were distributed with similar percentages in almost all sand fractions, with the highest fine (silt + clay) content (15%) among the materials in the study being recorded. About 62% by weight of Pu-Raw fell in the fractions ranging from gravel to coarse sand while, in the same particle size range, Ze-Raw reached 74% by weight. The Ss-Graded, taken as reference for particle size distribution, was dominated by the coarse and medium sand fractions with the sum of the two representing 87% by weight of the material.

For all water potentials applied, water retention of ‘raw’ materials followed the sequence: Ze-Raw > Pu-Raw > Lp-Raw (Figure 1). Water retained by Lp-Raw was gradually released with potential varying from −0.03 to −5 MPa while Ze-Raw and Pu-Raw released water for the full range of potentials applied with most of it being released between −0.03 and −0.2 MPa.

Figure 1.  Water retention curves (volumetric moisture content in m³ m⁻³) of silica sand (Ss) meeting USGA specifications for texture and volcanic materials: lapillus (Lp), pumice (Pu) and zeolite (Ze). Volcanic materials are reported as Raw (texture of the materials as available on production site) or Graded (obtained by reassembling the fractions in the same proportions of the reference Ss).

When comparing materials with the same particle size distribution, thereby including internally porous volcanic sands and the non-internally porous silica sand, water retention observed for volcanic sands followed the same sequence as for ‘raw’ materials at all the water potentials applied, while silica sand proved the least water-retentive material for each of the water potentials. Assuming that porosity due to particle spatial arrangement should be equivalent in sands having the same particle size distribution, the significantly higher volumetric water content retained by volcanic materials is likely to be the effect of internal pore space. The modification of the particle size distribution, from Raw to Graded materials, had the effect of significantly reducing the water retained at all the potentials in zeolite, while lapillus and pumice showed a significant reduction in water retention only at −0.03, −0.05 and −0.1 MPa, with minor or non-significant differences being recorded at the other water potentials. Content of fines and a wider range of particle size, allowing a higher degree of interpacking of particles, are probably the factors causing the raw materials to be more water retentive compared with graded materials. Also the modifications determined in the internal pore space are likely to be a concurrent factor in changing water retention of graded materials but the actual contribution cannot be isolated from the cumulative effect observed.

The overall effect of texture on available water is shown in Figure 2. The different retentions in Raw and Graded materials discussed above resulted in higher availability of water for all the raw materials. Pu-Raw retained the highest volumetric content of available water (0.189 m³ m⁻³), while Lp-Raw recorded the lowest (0.069 m³ m⁻³). Lp-Graded also retained less available water (0.035 m³m⁻³) compared with the other internally porous sands (0.089 and 0.103 m³ m⁻³ for Pu-Graded and Ze-Graded respectively) but it was still more than four times more retentive than non-porous Ss-Graded (0.008 m³ m⁻³). In comparing graded materials the effect of internal pore space seems to justify most of the differences observed between volcanic sands and silica sand.

Figure 2.  Available water (m³ m⁻³) of lapillus, pumice, zeolite and silica sand. For each volcanic material four bars are displayed corresponding (left to right) to raw material (texture as available on production site), graded material (obtained by reassembling the fractions in the same proportions of the reference silica sand used in the trial), coarse sand fraction (0.5–1.0 mm particle size) and medium sand fraction (0.25–0.5 mm particle size). For the silica sand data bars refer respectively to USGA-compliant material for texture and coarse and medium sand fractions as defined above.

Volcanic materials with the particle size falling in the coarse sand fraction retained significantly less available water than the respective graded materials and more available water than the respective medium sand fractions. Regardless of internal pore space, interpacking of particles could again explain the difference between graded materials and materials with particles falling in a single sand fraction. Differences due to this effect are reduced to a minimum in comparisons made across materials with highly uniform texture.

Data referring to coarse and medium sand fraction strongly confirm the effect of internal pore space on water availability with available water recorded for Ss-Coarse and Ss-Medium being virtually the same and of negligible amount, if compared with that of porous materials.

Both for medium and coarse sand fraction, pumice retained more available water than zeolite and zeolite was more retentive than lapillus. All the three coarse materials retained more available water than the relative medium fractions. This is apparently in contrast with the expectation of a coarser material to be less water retentive compared with a finer one. The lower water availability observed in medium-textured volcanic materials was due to a higher retention capacity at −1.5 MPa and a slightly lower retention at −0.03 MPa compared with coarse-textured volcanic materials (data not shown). Medium sand fraction seems to store part of the water retained in smaller pores compared with the coarse sand fraction of each material. A possible explanation could lie in the higher specific surface that is associated with a finer texture. A larger number of internal small pores could in fact be linked to the exterior surface of particles and retain more water at −1.5 MPa potential.

Conclusions

Sandy material obtained from volcanic rocks such as lapillus pumice and zeolite have been found to retain a considerable amount of water available for plants. Nonetheless when considering raw volcanic materials their particle size distribution does not meet the specifications to which sands for sports turf construction should conform. Furthermore, due to the high content of fine particles, their use as amendments for graded sands could be viewed as detrimental.

When considering graded volcanic sands, the comparison with a non-internally porous sand with the same particle size distribution shows their peculiar characteristics. Most of the water is retained by internal porosity and is available to plants while the interparticle porosity gives a negligible contribution.

Graded volcanic sands with an acceptable texture still have desirable hydrological characteristics and have therefore the potential to be used both as construction materials and inorganic soil amendments for silica sand amelioration. However, mechanical resistance of particles and incorporation rate in sand/amendment mixes still need to be investigated.

The analysis of the hydrological properties of single sand fractions of the three volcanic materials in the study revealed the coarse sand fraction to be more effective in storing available water compared with the medium sand fraction and this result should be taken into consideration in the selection of the proper particle size for soil amendment.