Share

Greenhouse Study

A greenhouse column experiment was conducted as part of the overall research project to investigate use of spent foundry sands (SFS) to manufacture blended synthetic topsoil for landscaping, turf, and horticultural applications.

Introduction

The U.S Environmental Protection Agency has a more complete discussion spent foundry sand.

The overall research project has four fundamental goals:

  1. Determine the extent to which various contaminants and plant macronutrients could be leached from a blended topsoil material containing spent foundry sands representative of three major core binder systems, and identify soil characteristics and processes that control contaminant solubility and leaching.
  2. Determine the extent of plant uptake of SFS contaminants in a blended topsoil environment.
  3. Compare quality characteristics and plant growth performance of commercially available topsoils and SFS-containing synthetic topsoil blends, and
  4. Provide demonstration sites of beneficial utilization of SFS and technology and outcome transfer to the foundry industry, landscaping/turf/manufactured soil industries, and environmental regulatory agencies.

The greenhouse experiment was conducted to assess availability (plant uptake) and transport of possible inorganic and organic contaminants that might be present in spent foundry sands. The experiment also allowed us to assess the performance of SFS based manufactured topsoils, in particular nutrient availability and nitrogen mineralization from various organic materials that might be used in the topsoil blends. SFS based blended topsoil was made by combining SFS with widely available, unrestricted use organic materials and natural subsoil material.

Greenhouse Experiment Setup

Spent green sands from two iron and one aluminum foundry were utilized in the experiment. The sands were selected on the basis of the chemistry of the core mold binder systems used at these foundries. The three binder systems represented in the experiment are: phenolic urethane no-bake (PUNB, iron foundry), furfuryl alcohol no-bake (FNB, iron foundry), and Shell (aluminum foundry). Spent sands were collected from each foundry and brought back to Penn State University for analysis and use in the experiment. Spent sands from the two iron foundries were already single grained. However, the spent sand from the aluminum foundry contained some core butts, both pre- and post-casting. Prior to use in the greenhouse experiment all sands were passed through a 2 mm sieve in our laboratory. Any pieces of metal caught on the sieve were removed. Any sand aggregates and core butts caught on the sieve were reduced to single grain particles by grinding with a mortar and pestle and added to the bulk sand.

Three organic materials were selected for blending with SFS: yard trimmings compost, spent mushroom substrate (SMS), and biosolids compost. Although the spent mushroom substrate is not a completely composted material, in this report the term “compost” will sometimes be used to refer collectively to all three of these organic materials. These composts are widely available and may be sold or given away to the general public without any regulatory limitations on their use.

All the materials used in the soil blends were physically and chemically characterized.Results of these analyses are given in Tables 1–4.

Table 1. Particle size distribution in spent foundry sands.
SFS Sand
(2-0.05 mm)
Silt
(0.05-0.002 mm)
Clay
(<0.002 mm)
% by weight
PUNB
97.6
1.1
1.2
FNB
97.5
1.2
1.3
Shell
90.7
5.7
3.5
Hagerstown B
15.1
37.9
47.0

 

Table 2. Chemical characterization of organic materials used in manufactured soil blends.
Analyte Yard Trimmings Compost Spent Mushroom Substrate Biosolids Compost
pH
7.9
8.1
7.1
Soluble Salts (mmho/cm)
2.33
16.45
6.87
Organic Matter (%)
47.4
58.7
52.4
Total Nitrogen (%)
1.8
2.1
3.3
Organic Nitrogen (%)
1.8
2.1
2.8
Ammonium Nitrogen (mg kg-1)
4.5
5.3
5024.5
Organic Carbon (%)
28.7
29.9
29.0
Carbon/Nitrogen Ratio
15.7
14.2
8.7
Phosphorus (P2O5, %) 0.84
1.67
5.34
Potassium (K2O) 1.43
3.70
0.33

 

Table 3. Macroelement concentrations in spent foundry sands and organic materials used in the greenhouse experiment.

Al Ca Fe K Mg Mn Na P S
mg kg-1
FNB 425
200
1001
55.7
909
47.6
65.1
10.7
375
Shell 1925
1020
1402
363
569
16.8
138
17.7
203
PUNB 1278
708
47480
63.7
252
286
296
<6.0
299
SMS 3714
73799
4410
27463
13089
326
2648
6129
10156
Biosolids 8602
18636
71423
1910
3784
2777
444
21705
8536
Yard 6573
38248
11693
8867
7745
1022
536
3480
2666

 

Table 4. Trace element concentrations in spent foundry sands and organic materials used in the greenhouse experiment.

As Ba Cd Cr Cu Hg Mo Ni Pb Se Zn
mg kg-1
FNB <0.30
27.2
<0.40
5.25
3.90
<0.007
0.85
9.15
1.07
<0.51
7.54
Shell <0.30
5.63
<0.40 3.51
4.44
<0.007 <0.50
1.96
0.86
<0.51 11.9
PUNB 7.50
7.95
<0.40 51.8
137
<0.007 6.97
26.3
2.42
<0.53 5.32
SMS 24.8
78.4
<0.40 20.5
78.3
0.032
3.27
8.19
4.89
0.84
153
Biosolids 5.74
451
4.36
66.0
460
1.255
16.6
22.5
148
4.04
1158
Yard 6.88
145
0.61
25.1
55.7
0.188
2.10
13.2
73.5
1.04
200

Soil Blending

Synthetic topsoils were made by blending each type of SFS with each type of compost and with subsoil at a dry weight ratio of 6.5:1.5:2.0 (SFS:organic material:subsoil). The greenhouse experiment then consisted of these nine blends and also a natural topsoil control (Hagerstown silt loam A horizon). The manufactured topsoil blends had the characteristics listed in Tables 5 and 6.

Table 5. Textural analysis of manufactured topsoil blends used in the greenhouse experiment.
Soil blends Particle size distribution Textural class
Sand Silt Clay
%
PUNB
78.2
9.8
12.0
Sandy Loam
FNB
78.1
9.8
12.1
Sandy Loam
Shell
72.9
13.3
13.7
Sandy Loam

 

Table 6. Total carbon, total nitrogen and organic matter content of manufactured topsoil blends used in the greenhouse experiment.
Soil blends Total N Total C Organic Matter
%
SMS
0.32
4.49
8.06
Yard
0.27
4.31
7.11
Biosolids
0.50
4.35
7.86

Leachate Columns

Columns for the plant growth and leaching experiment were constructed using 15 cm diameter PVC pipe cut to 30 cm lengths. The PVC pipe was glued to a flat PVC base plate with a nipple in the center to allow collection of leachate. The inner surfaces of the columns were lined with Teflon to minimize potential interference from PVC constituents. A 5 cm layer of acid washed virgin sand was placed in the bottom of each column. Columns were then filled with a 21.5 cm depth of manufactured soil blend or with natural topsoil. Material was added in three “lifts” and columns were tamped following the addition of each lift to achieve the same extent of packing with each material. Because of differences in bulk density of the composts the total amount of material and the bulk density of material in the columns varied for the different treatments (Table 7). Columns were filled on April 22, 2004. The same amount of inorganic fertilizer was added to each treatment. Because one objective of this experiment was to determine the amounts of N, P, and K fertility supplied by the compost materials, inorganic fertilizer addition was kept to the minimum amount needed for plant growth in the natural topsoil.

Table 7. Total weight of material added to columns and bulk density for each treatment.
Treatment Total dry weight of soil material in column
(g)
Initial bulk density of material in column
(g cm-3
Natural Topsoil
4790
1.26
PUNB + Yard
4492
1.18
PUNB + SMS
3764
0.99
PUNB + Biosolids
4383
1.15
FNB + Yard
4579
1.21
FNB + SMS
3730
0.98
FNB + Biosolids
4443
1.17
Shell + Yard
4552
1.20
Shell + SMS
3523
0.93
Shell + Biosolids
4244
1.12

After filling each column the moisture content was adjusted to 80% of field capacity by adding de-ionized water. On April 23, 2004 each column was planted with 40 seeds of annual ryegrass (Lolium multiflorum Lam.). De-ionized water was added as needed to maintain sufficient moisture for ryegrass growth but not enough to cause any leaching. Ryegrass establishment was monitored by counting the number of live seedlings on May 5, 7, and 10, 2004.

Columns were intentionally leached on 5/5/2004 and then approximately once each month thereafter (on 6/9, 7/7, 8/10, 9/7, and 10/7/2004 ) for a total of 6 leachings. Leaching was done by adding 50 ml of de-ionized water to the surface of each column every 30 minutes until approximately 500 – 700 ml of leachate had been collected. Leachates were collected in amber glass bottles. Immediately after leaching pH and electrical conductance were measured. Leachates were filtered through a 0.45 µm membrane filter. A 50 ml aliquote was preserved for trace element analysis by adding 1 ml concentrated nitric acid and was stored at 4 o C. Analysis for Al, Ba, Ca, Cd, Cr, Cu, Fe, K, Mg, Mn, Mo, Na, Ni, P, S and Zn was done by inductively coupled plasma emission spectroscopy (ICP). None-acidified leachates were analyzed colorimetrically for NO 3 and NH 4 using a Lachat autoanalyzer, and for total N and total C by combustion using a Carlo Erba analyzer. Mercury was analyzed in the first two leachings using EPA Method 7470 (cold vapor atomic absorption). A preserved aliquote was sent to USDA-ARS for analysis of organic chemicals. Organics analysis is not yet complete. As of this date only phenolic compounds have been analyzed following EPA 8270 methodology.

Ryegrass top growth was clipped whenever the grass began to flower (on 6/2, 6/17, 6/30, 7/22, 8/11, 8/30 and 9/14/2004 ). Clippings were dried for 48 h at 65 o C, weighed, and ground to pass a 1 mm screen. A 10 g composite tissue sample was prepared for analysis by determining the fractional contribution of each clipping to the total yield and adding the corresponding proportion from each clipping. When total yield was less than 10 g pot -1 all of each clipping was used for the composite sample. Tissue N was determined by combustion using a Carlo Erba analyzer and samples were analyzed for other nutrients and trace elements by ICP following microwave digestion.

Results

Plant Growth

Ryegrass grew well in all SFS-based manufactured soils and greatly exceeded growth on natural topsoil (see graph below).

Ryegrass Cumulative Yield

However it must be emphasized that growth on natural topsoil was limited by low fertility levels. As was previously stated, inorganic fertilizer addition was kept very low to determine the fertility contribution of the composts. Ryegrass growth was also clearly influenced by the type of compost. The greatest growth occurred with biosolids and SMS composts and less growth was obtained with yard trimmings compost. Nitrogen availability likely accounted for most of the difference in ryegrass growth. Ryegrass is very responsive to nitrogen, and biosolids and SMS clearly supplied more nitrate nitrogen than yard trimmings compost (see graph below).

Leachate Nitrate

Conclusions

The results of this greenhouse column experiment with spent foundry sand and compost based manufactured soils indicates no potential environmental concerns with such use of these types of SFS. There was very little evidence to suggest that SFS increased plant uptake or leaching loss of any trace metals of environmental concern. The increases observed were too small to be of environmental significance. The limited organics analysis completed to date indicates no concern with phenolic compounds. The primary environmental concern indicated by this experiment is the potential for excessive nutrient loss (N and P) which is related to the compost component of the manufactured soils, not the SFS component. These issues warrant further study in field experiments. There were also interactive effects between SFSs and composts with respect to nitrogen mineralization, nutrient leaching and uptake and the leaching of some metals. These effects need to be investigated further in field studies to insure that topsoils manufactured from SFS and composts meet high performance standards and are not sources of excess nutrients in the environment.