3.1. Mineral Products Derived from CLO
The work of the CLO treatment line was evaluated in terms of content and level of material recovery (η) in the product (glass fraction). Material recovery is the quotient of its quantity in the product in relation to its content in the feed (CLO). The results are shown and discussed for the intermediate product after each device to show the effect of each step of the process carried out.
The mass of intermediate products obtained during the processing of CLOs from Marszów and entrusted from other MBT systems on the CLO processing line and their material composition are presented in
Figure 3.
Figure 4 presents the average levels of debris and stone recovery (inert) from CLOs in the subsequent stages of the glass concentrate production process (A—waste from Marszów, B—entrusted waste) and Figure 6 presents that of glass.
The VARIOMAT dual-level screen separates the CLO stream into three fractions: <10, 10–35 and >35 mm.
Fraction <10 mm constituted 50.9 ± 6.4 and 53.0 ± 13.5% of the mass of CLOs, respectively. It was a mixture of ash and sand, as well as, in large quantities, an organic fraction that was dried and largely ground to grain <10 mm during the unloading of chambers, transport of CLOs to the screen and screening operations. The high content of roasting losses confirms the high content of organic matter in this fraction (from 27.1 to 45.8% sm, 36.8% sm on average).
The 10–35 mm fraction directed to the production of glass concentrate constituted 39.7 ± 5.3% of the mass of CLOs from Marszów and 34.5 ± 6.7% of the mass of entrusted CLOs.
The fraction share > 35 mm in the mass of CLOs was small—on average 9.4 ± 2.2% of the sample from Marszów and 12.5 ± 9.9% of the sample entrusted. Mainly plastics passed into this fraction, on average at 57.6 and 63.7%, metals at 28.0 and 23.4%, other impurities at 21.0 and 21.6% and inerts (stones, debris) at 20.7 and 21.4%, respectively, and in the case of waste entrusted organic fraction loss, 30.6%.
The first mineral fraction is <10 mm (M-1). The mineral nature of this fraction was confirmed by industrial tests to separate this waste into organic and mineral fractions and to determine the actual amount of minerals for potential use. The first involved the use of air separation on Trenso separation tables.
Fraction <10 mm was fed to a vibrating screen driven by an eccentric. The inclination of the sieve can be adjusted steplessly. The air is introduced into the fractionated product by means of a pressure fan under the screening deck. This fluidizes the light elements and separates them from the heavy elements. The heavier parts are moved up the table, and the lighter parts are transported down towards the outlet of the light fractions, depending on the movement and the angle of inclination of the screening deck. Results obtained in the studies: 40.7% mineral fraction (including fraction <5 mm —32.9%, >5 mm stones—5.2% and glass—2.6%), light fraction 50.6% and dust 8.7%. Tests were also performed on Doppstadt water separators. The tests were carried out at BYŚ in Warsaw in July 2019. The share of the heavy mineral fraction in the <10 mm fraction was 51.2%.
Two more mineral fractions constitute ballast (M-2) and ballast (M-4), which in material terms are a mixture of stones and debris contaminated with various components. M-2 waste is a fraction separated on the LASER separator from the 10–35 mm fraction after removing paper, organics and plastics on the air separator. M-4 waste is a heavy fraction separated on an NIHOT air separator from the 30–80 mm fraction after removing metals from it on a magnetic separator.
Material balances of debris and stones (inert) in the treatment processes of the CLO produced in the system in Marszów and entrusted CLOs are presented in
Figure 4.
In the case of ZZO in Marszów, the M-2 fraction after the Laser autosorter (
Figure 5) was 11.5 ± 2.8% of the CLO weight. This waste constitutes debris and stones (49.5 ± 10.7%) and glass (20.8 ± 6.4%). Up to 62.7% of the stones and debris contained in the CLO passed through (
Figure 4). Contamination consists primarily of paper (8.1 ± 2.7%) and organic contamination (7.5 ± 2.5%) (
Figure 3). The waste showed a low humidity of 4.9 ± 1.2% and low roasting loss of 14.8 ± 9.7% (laboratory analysis). The share of minerals determined in industrial tests on water separators by Doppstadt was 70%. In the case of entrusted CLOs, the M-2 ballast constituted 11.3 ± 6.5% of the CLO mass. A total of 4% less stones and debris contained in the CLO passed into it than in the case of waste from Marszów. It contained more debris and stones (62.0 ± 6.6%), but significantly less glass (12.8 ± 5.3%). The paper content was 4.8 ± 3.6% and organics was 4.4 ± 3.3% (
Figure 3).
The M-4 ballast (
Figure 5) separated from the 35–80 mm fraction constituted 3.2 ± 0.9% of the mass of processed CLO from ZZO in Marszów. Up to 20.5% of the stones and debris contained in the CLO passed through (
Figure 3) and their share in waste was 62.2 ± 6.6%. The waste contained 12.8 ± 5.3% glass and 10.4 ± 1.2 metals. Plastics constituted 4.8 ± 3.6% of its mass and organics constituted 4.4 ± 3.3% (
Figure 3). The waste showed a low humidity of 8.4 ± 5.9% and low roasting loss of 13.3 ± 8.1% (laboratory analysis). In the case of CLO M-4 ballasts after the autosorter, the laser constituted 4.2 ± 3.9% of the CLO mass. A total of 24.9% of all stones and debris passed into it. This waste contained 68.3 ± 1.8% of debris and stones and only 3.6 ± 3.8% of glass. Among organic contamination, the share of plastics was 12.4 ± 3.1% (
Figure 3).
The material balance of glass in the treatment processes of the CLO produced in the system in Marszów and entrusted CLOs are presented in
Figure 6.
The glass fraction (M-3) was 11.7± 3.4% of the CLO mass in the case of waste from ZZO in Marszów, and 7.8 ± 2.8% in the case of entrusted CLOs.
Figure 7 presents photos of CLO and glass fraction after subsequent stages of the glass recovery process.
The processing of CLO on the line allowed to recover from 49.8 to 74.3% of the glass it contained (on average 69.4 ± 7.0%) for samples from Marszów and from 38.3 to 72.1% (on average 58.3 ± 14.2%) for samples from other MBT systems. The concentrates contained practically no paper, plastics or metals. Organic contamination was only present in two samples from Marszów in trace amounts of 0.1 ± 0.1% (
Figure 3). Fraction <5 mm was present in four samples from Marszów in an amount of 0.2 to 2.5% and in one sample from other MBT systems (5.8%). The main impurity in the glass concentrate found in all samples was “other ingredients”. Their share was on average 0.9 ± 1.0% in samples from Marszów and 2.2 ± 2.0% in entrusted samples. The content of glass in the final product ranged from 93.0 to 99.5 (on average 98.0 ± 1.9%) of the fraction weight in the case of CLOs from Marszów and from 92.8 to 99.1% (on average 96.8 ± 2.8%) in the case of CLOs from other MBT systems (
Figure 3). In 14 out of 30 measuring series carried out, the share of glass was 99%. Glass contents lower than 98.0% were found only in six studies. These were samples obtained from high-hydration CLOs.
3.2. Application of the Mineral Fraction Obtained from CLO
The 0–5 mm fraction was tested for the possibility of using it for winter road maintenance. The grain composition was tested by a screening aggregate for concrete according to PN-EN 933-1 [
17] and the following results were obtained: dust <0.063 mm—12.5%, main fraction/0–2.0 mm/ = 89.6%, oversize = 10.4%. Based on the current requirements for tenders based on European standards for sand for winter maintenance, the screening is correct. Unfortunately, this aggregate was too light, it was impossible to test the pen indicator and it contained a lot of organic parts. Comparison of the costs of purchasing rinsed sand meeting the requirements and preparing the material from the CLO was more favorable for sand. Therefore, further research in this direction was abandoned, also taking into account the problem in acquiring potential customers who would like to use the material obtained from waste for winter road maintenance.
A C25/30 concrete mix formula was prepared, Portland cement CEM I 42,5 according to PN-EN 206:2014 [
18], in which the primary aggregate with granularity of 5–10 mm was replaced by mineral waste of the same fraction, but originating from CLO. At the beginning, the prepared formula was tested in a laboratory, testing the compressive strength of concrete according to PN-EN 12390-3:2011 [
19] along with PN-EN 12390-3:2011/AC:2012 [
19]. Samples after 56 days showed an average compressive strength of 40.5 MPa. The obtained result was used to assess the identity of the concrete class according to PN-EN 206+A1:2016-12 [
18], which showed that the developed formula for a mixture of concrete with an aggregate derived from CLO meets the standards of class C25/30. Concrete absorbability tests according to PN-88/B-06250 and tests of water permeability through concrete according to PN-88/B-06250 [
20] were also carried out, which confirmed that the concrete mix meets the standards for class C25/30. The formula prepared in this way was handed over to a local concrete company and a trial batch was commissioned in the amount of 100 concrete blocks for retaining walls and boxes. The blocks made were used to make storage boxes at ZZO Marszów (
Figure 8). As part of these activities, the possibility of using this mineral fraction from CLO on an industrial scale was confirmed, providing the possibility of rational use of natural resources in the form of an aggregate.
As part of the laboratory tests of the mineral aggregate obtained from CLO with granulation of 0/31.5 mm, the following procedures were performed: determination of the grain composition according to PN-EN 933-1 [
17], he sand index according to PN-EN 933-8+A1:2015-07, determination of density and water content by the Proctor method according to PN-EN 13286-2:2010 [
21], determination of resistance to comminution by the Los Angeles method according to PN-EN 1097-2:2010 [
22] and determination of aggregate frost resistance according to PN-EN 1367-1:2007 [
23], in order to assess the suitability of the material for incorporation as a hardening layer on communal dirt roads and access roads after mechanical stabilization or as material bound with hydraulic binders. The results are shown in
Table 2. Due to the lack of general requirements for these type of roads, the scope of the tests was established on the basis of requirements for mixtures not bound to the improved subsoil, substructures and pavements presented in the following documents: WT-4 technical requirements “Unbound mixtures for national roads” and WT-5 technical requirements “Hydraulic binder mixtures for national roads” of the General Directorate for National Roads and Motorways (GDDKiA).
The screening practically met the requirements of the WT-4 GDDKiA document containing the requirements for aggregates for unbound aggregate pavement, as the sand index test was positive with a result above 35%. This material has low resistance to grinding and low resistance to freezing and thawing in the presence of water. With the right approach of investors to the expected parameters of the material, it can successfully fulfil its role in the intended applications.
Test mixes were also made with hydraulic binders: Portland cement CEM I 42.5R and metallurgical CEM III 32.5 N and a mixture of fly ash from the Konin power plant with the fraction of 0/1 mm. Formulas have been prepared without increasing the binder content compared with standard formulas with aggregates of natural origin. Compressive strength tests of bound mixtures based on various types of cements and cement–ash mixtures provided results as expected, allowing for obtaining the expected classes of compressive strength after 28 days of maintenance (in the range of 2.5–7.0 MPa). Tests of freezing and thawing strength of samples confirmed low aggregate frost resistance, giving unsatisfactory results, i.e., frost resistance index below 0.6. The foundation layer made of cement bound aggregate after several freezing–thawing cycles in the structure will degrade and begin to behave as an unbound aggregate.
At the beginning of 2019, research was also carried out to obtain an answer as to the possibility of using selected mineral fractions (0–10 mm) from CLO to obtain a waste cement mix useful for constructing road foundations using a standard amount of cement and compaction energy. The quality criteria set by the PN-S-96012:1997 [
24] standard were adopted as the assessment criteria when making and receiving the foundation made of cement-stabilized soils and the subsoil layer improved with cement for the road surface. The composition of the mix as well as the preparation and sample testing were carried out in accordance with the procedures contained in the PN-S-96012:1997 [
24] standard. The optimum composition of the mixture was determined through mixing waste with cement in the amount of 6 to 10% of the weight of waste and water, in an amount such that the waste cement mix has the ability to form, i.e., the ability to preserve the shape after compaction using a standard amount of compaction energy and maximum density after compaction. A 10% cement mix based on the weight of the waste and 10% water in the mix (based on the weight of the waste and cement) was considered optimum and a full range of compression strength tests were carried out using it. The density of the cement–waste mixture was defined with a determined humidity of 10% at the level of 1958 kg/m
3. The frost resistance index, defined according to PN-S-96012:1997 [
24], could not be determined as the ratio of the strength of 28-day-saturated samples subjected to 14 freezing and thawing cycles to the strength of 28-day-saturated samples, expressed as a decimal fraction. The reason was that such a large frost destruction of samples after four cycles makes it impossible to carry out strength tests. The properties required by the PN-S-96012:1997 [
24] standard and many years of road construction practice for layers used in road pavement construction, which could potentially be carried out using a waste cement mix, are presented in
Table 3.
The data in
Table 3 show that the waste-based mixture does not meet the required parameters. Better parameters of the waste cement mix can be achieved by increasing the mix’s ability to compact and wedge by modifying the shape and size of the waste grains, optimizing the grain composition of the waste mix and increasing the amount of cement. However, the required significant percentage of cement in mixtures, close to the maximum amount in terms of price and technological reasonability, disqualifies cement–waste mixtures from being used commercially.
In September 2018, a 900 m long municipal road connecting the village of Marszów with the system was rebuilt using the mineral fraction obtained from the CLO (
Figure 9). The reconstruction consisted of road trenching, fertilizing the mineral fraction from the 0–80 mm CLO with a thickness of 20 cm after compaction and then applying a top layer of 0–31.5 mm granite aggregate with a thickness of 10 cm after compaction.