Determination of Total Mercury and Mercury Thermospecies in Cement and Cement Raw Materials

Abstract

Cement manufacturing is the second largest anthropogenic source of Hg emissions in the environment. Therefore, the establishment of analytical methodologies that can be utilized in the determination of Hg concentration from cement raw materials and cement is of great importance. The total Hg and Hg thermospecies in cement raw materials and cements were determined by thermal desorption techniques with a Zeeman Hg analyzer. No chemical pre-treatment of samples is required for this technique prior to analysis. An optimum single-stage temperature program was applied to determine total Hg at an optimum heating rate of approximately 5 °C s⁻¹ while Hg thermospecies were determined over four stages at an optimum heating rate of approximately 0.2 °C s⁻¹ per stage from ambient temperature to 720 °C. Total mercury concentrations in cement raw materials ranged between 2.19 ng g⁻¹ and 395 ng g⁻¹, while in cement, concentrations ranged between 1.32 ng g⁻¹ and 31.0 ng g⁻¹. The highest Hg contents were found in dust return (580 ng g⁻¹ and 679 ng g⁻¹). Hg thermospecies determination showed that cement raw materials and cements contain one Hg thermospecies that is released at 20–180 °C while dust return contained one to four Hg thermospecies that could be released at 20–180 °C, 180–360 °C, 360–540 °C, and/or 540–720 °C, thus indicating that new Hg compounds are formed during cement production.

1. Introduction

Mercury (Hg) is a hazardous element of major public health concern due to its high toxicity, high volatility, and long residence time in the environment [1,2,3]. It occurs naturally on the Earth’s crust and is emitted to the atmosphere through various anthropogenic sources associated with the combustion of fossil fuels such as coal combustion at electrical power stations, metal smelters, waste incineration, and cement production plants [4,5,6].

Cement production is the second largest source of anthropogenic Hg emissions in the environment worldwide, with Europe accounting for 11% of global anthropogenic Hg emissions [7]. Hg enters the cement manufacturing process as a trace element through raw materials (limestone, slag, and coal) that are utilized during cement production. Hg concentration in natural raw materials varies from deposit to deposit, depending on the origin and source [8,9,10,11]. A typical cement manufacturing process involves three fundamental stages, viz., preparation of feedstocks, clinker production, and cement production.

Feedstock preparation includes quarrying, crushing, and milling of raw materials. After milling and homogenization in a silo, the raw materials are transferred into the pre-heaters (kiln feed) before entering the kiln operating at maximum temperatures of 1450 °C, where they are burned to form cement clinker. Clinker is produced inside the kiln, and its nodules are then milled and mixed with other materials to produce cement. Hg desorption and emissions occur in the cement kiln due to its high heating temperature of approximately 1450 °C [12,13,14,15,16]. Due to Hg being very toxic and harmful to human life upon inhalation and ingestion, studies involving the quantification of Hg in cement-making industries are essential, as limited information on the actual concentrations from cement raw materials is available [17].

There are four official standard methods of solid sample analysis, viz., bomb combustion [18], acid extraction [19,20], wet oxidation [19], and thermal decomposition [21,22]. The main limitation of the first three methods is that sample preparation is long, taking up to 8 h, depending on the solid type, e.g., coal requires a longer digestion time to transfer Hg into solution. The US EPA method 7473 [22] is based on thermal decomposition with amalgamation of Hg vapours on golden wire and quantitative determination of Hg by AAS, but the reliability of the method is affected by interferences that occur due to the presence of SO₂ and Cl⁻ ions (common constituents of most coals), thus affecting the precision and accuracy of the analysis results. Therefore, the applications of these recommended methods often result in contradictory results. This is why additional studies of the determination of Hg in solid matrices are necessary to eliminate the shortcomings of the recommended methods described above.

Pyrolysis technology coupled with atomic absorption spectrometry or atomic fluorescence spectrometry was used by researchers worldwide in the determination of Hg content in solid samples. Mercury removal from powdered samples by mild pyrolysis before combustion was originally proposed by Merriam et al. (1990) [23], and further studies were conducted by other researchers who assessed the effects of temperature, residence time, and volatile matter content on Hg removal [24,25,26,27,28,29]. Further development in the analytical methodology is necessary for the identification and quantification of mercury species as a function of volatility temperature in cement and cement raw materials to enhance the selection of cement raw materials and coal (fuel) suitable for combustion during cement production. The main objective of this study is to determine the total mercury concentration and mercury thermospecies in cement raw materials and cement after thermal decomposition by a Hg analyzer.

2. Materials and Methods

2.1. Instrumentation

A Model RA-915+ Zeeman Mercury analyzer (Lumex, St. Petersburg, Russia) with a PYRO-915 attachment was used for Hg measurements. The PYRO-915 attachment enables the determination of Hg in samples having complex matrices such as soils, sediments, oil products, and foodstuffs by utilizing the pyrolysis technique incorporated within the instrument [30,31,32].

The operation of the Hg analyzer is based on the release of Hg from solid samples during thermal decomposition. The concentration of Hg is measured by an atomic absorption spectrometer at a 253.7 nm Hg⁰ resonance line, and any background absorption is corrected using a Zeeman Effect correction system [30].

2.2. Calibration Standards

The RA-915+ Zeeman Mercury analyzer was calibrated using SARM 20 certified reference material of coal (Sasolburg) with a certified Hg value of 250 ± 30 ng g⁻¹. The fish protein certified reference material for trace metals, DORM 3 (Industrial Analytical, Johannesburg, South Africa), with a certified Hg concentration of 0.382 ± 0.06 mg kg⁻¹ covering the lower concentration range, and the dogfish liver certified reference material for trace metals, DOLT-4 (Industrial Analytical, South Africa), with a certified Hg value of 2.58 ± 0.22 mg kg⁻¹ covering the higher concentration range, were used in the validation of the analytical methods.

2.3. Reagents and Solutions

South African standard reference materials of coal, SARM 20 (Council for Mineral Technology, MINTEK, Randburg, South Africa), with a certified Hg concentration of 250 ± 30 ng g⁻¹, and DOLT-4 (Industrial Analytical, South Africa), with a certified Hg value of 2.58 ± 0.22 mg kg⁻¹, together with fish protein certified reference material for trace metals, DORM 3 (Industrial Analytical, South Africa), with a certified Hg concentration of 0.382 ± 0.06 mg kg⁻¹, NCSZC 78001 (coal fly ash, China, 39 ± 0.9 ng g⁻¹), were either used for the calibration of the instrument or validation of the results.

2.4. Sampling of Raw Materials and Cement

Cement raw materials (limestone, fly ash, slag, and gypsum) were sampled from different Pretoria Portland Cement (PPC) plants, viz., PPC Jupiter, PPC Hercules, and PPC De Hoek, on different stockpiles over a period of three months ranging from January 2019 to March 2019. The final product (cement) was sampled from the storage silos of each plant. Proper personal protective equipment, like a dust mask, a work suit, safety shoes, goggles, and a helmet or hard hat, was always worn by the sampler during sampling to prevent injury.

Two types of cement, viz., CEM I 52.5N and CEM II 42.5N, were sampled at each PPC plant. The difference between these two cements is that CEM II 42.5N is extended with fly ash and slag or slag alone, depending on the plant preference, while in CEM I 52.5N, extenders are not added. The role of the added extenders is to reduce the amount of clinker used during cement production as a cost-saving intervention. Gypsum is added to both cements on the finish mill as a minor additional constituent (MAC) and acts as a setting time regulator and strength enhancer. It controls the hydration process and prevents rapid setting, allowing for proper mixing, transportation, and placement of the concrete. CEM II 42.5N produced at PPC De Hoek is extended with slag only, while CEM II 42.5N produced at PPC Hercules is manufactured with slag and fly ash as extenders, but synthetic ALI gypsum is used instead of natural gypsum, while PPC Jupiter uses OMV synthetic gypsum (phosphate waste product) that is supplied by the Oranje Mynbou en Vervoer (OMV) ready mix concrete plant.

2.5. Sample Preparation

Samples were first dried at 105 °C to remove moisture that might be present. A jaw crusher (model: Retsch GmbH; type: BB200 Mangan) was used to crush limestone and raw coal to a particle size of approximately 4.75 mm. Limestone, gypsum, and raw coal were further pulverized using the LM2 Lab pulverising mill to a smaller particle size of approximately 2.00 mm; then, all samples were ground to pass through a 212 µm sieve using a Siebtechnik TEMA mill.

2.6. Analysis of Samples

A sample weighing approximately 20 to 300 mg was transferred into a sample boat and recorded using the RAPID software of the instrument. The integration of the analytical cell was switched on, and the sample containing the boat was inserted into the PYRO 915+ attachment, which consists of two chambers that are heated to 750 °C and 800 °C, respectively. The Hg vapour produced was passed through the analytical cell, where Hg levels were processed, and a signal was produced and integrated to give the final Hg content in ng g⁻¹. The background absorption signal was eliminated using a high-frequency Zeeman correction system [33]. Every sample was analyzed in triplicate, and the results were reported as the mean ± standard deviation of the mercury concentration.

The concentration of Hg in the sample was abstracted from the calibration curve plotted as the integrated analytical signal (arbitrary units) versus the absolute mass of Hg (ng) [34]. The integration was considered complete when the analytical signal declined to baseline. The absorption of 254 nm resonance radiation by mercury atoms was measured using the Zeeman correction for background absorption. The certified reference material SARM 20 was used to check the accuracy of the instrument, and blank samples were run after every ten samples and after samples with high values of Hg.

2.7. Determination of Hg Thermospecies by Zeeman Hg Analyzer

Samples ranging from 20 to 300 mg were weighed into sample boats, and each sample was inserted into the furnace of the Zeeman Hg analyzer. The atomizer furnace was gradually heated from ambient temperature to 720 °C at an optimized heating rate over five temperature stages to obtain thermoscanning peaks that represent Hg thermospecies in cement raw material samples and dust return based on volatility temperatures.

3. Results

3.1. Optimization of the Heating Rate in the Evaporation Chamber

The heating rate of the evaporation chamber was optimized to obtain well-separated analytical integrated absorbance peaks of Hg. The total Hg concentration was achieved by heating the raw material and cement samples from ambient temperature to 720 °C at a slower heating rate (

Table 1. Temperature program applied in the determination of Hg in samples.

Analyte	Heating Stage Temperature Range
Total Hg determination	Single Stage: Ambient T–720 °C
	Stage 1: Ambient T–340 °C
	Stage 2: 344–378 °C
Hg thermospecies	Stage 3: 387–485 °C
	Stage 4: 510–617 °C
	Stage 5: 641–720 °C
	Stage 6: 720 °C–Ambient T (Cooling)

).

The optimum heating rate ranges between 3.60 and 7.20 °C s⁻¹ and takes approximately 150 s. To quantify Hg thermospecies, samples were heated from ambient temperature to 720 °C over four different stages, viz., 20–180 °C, 180–360 °C, 360–540 °C, and 540–720 °C, at the optimum heating rate of 0.144 °C s⁻¹ stage⁻¹, and this took approximately 1000 s (Table 1). Stage 6 involves the cooling of the evaporation chamber to ambient temperature in preparation for an analysis of the next sample.

3.2. Calibration of Zeeman Hg Analyzer

The calibration curve of Hg determination was generated as the absolute mass of Hg (ng) versus the atomic absorption peak area (peak area, arbitrary units). The absolute mass of Hg (m_Hg) was determined from the relationship between the certified value of Hg concentration (C_Hg) and the mass of the CRM (m_CRM) subsamples taken for analysis:

$${\mathrm{m}}_{\mathrm{Hg}}\left(\mathrm{ng}\right)\to {\mathrm{C}}_{\mathrm{Hg}}\left(\mathrm{ng}{\mathrm{mg}}^{-1}\right)\times {\mathrm{m}}_{\mathrm{CRM}}\left(\mathrm{mg}\right)$$

(1)

A plot of Hg standard signals against absorbance was drawn to check whether the instrument is well optimized (Figure 1). The calibration curve with a correlation coefficient of approximately 1 was obtained, thus confirming a well-optimized instrument. Therefore, it can be concluded that sample matrices did not have any effect on the release of Hg and can be used for the creation of a single universal calibration curve for the determination of mercury. Such data is stored in the PC memory and used for the automatic calculation of Hg concentration during analysis.

3.3. Validation of Analytical Results

Certified reference materials (CRMs) were analyzed regularly to ensure the validity and reliability of the results. The accuracy of the applied method is confirmed when there is good agreement between the found and certified values. To check the validity of the applied analytical method, CRMs were analyzed. The choice of CRMs was based on including CRMs with lower Hg content (NCSZC 78001) and CRMs with higher Hg content (SARM 20) so that a wide range of Hg values were represented. The CRMs containing Hg content in the ranges of those found in cement and cement raw products were analyzed, viz., NCSZC 78001 (coal fly ash, China, 39 ± 0.9 ng g⁻¹) and SARM 20 (MINTEK, South Africa, 250 ± 30 ng g⁻¹), to check the validity of the method. The results of the analysis of reference materials indicated that the applied analytical method is reliable, as there was good agreement between the certified and found values at a 95% level of confidence using a two-tailed t-test (

Table 2. Results of Hg determination in standard and certified reference material (ng g⁻¹).

	SARM 20		NCSZC 78001
Instrument	Certified Value	Found Value	Certified Value	Found Value
RA-915+	250 ± 30	247 ± 4.0	39 ± 0.9	38 ± 1.0
DMA-80	250 ± 30	251 ± 5.0	39 ± 0.9	38 ± 0.8

).

3.4. Results of the Determination of Total Hg in Cement Raw Materials

The results for the determination of total Hg (ng g⁻¹) in cement raw materials and coals that are utilized during combustion are summarized in

Table 3. Results of the determination of total Hg concentration (ng g⁻¹) in cement raw materials.

Raw Materials	PPC Jupiter, ng g−1	PPC De Hoek, ng g−1	PPC Hercules, ng g−1
Limestone	2.40 ± 0.28	3.02 ± 0.32	2.19 ± 0.32
Fly Ash	160 ± 2.55	N	89 ± 25
Slag	2.56 ± 0.41	3.23 ± 0.36	1.58 ± 0.19
Natural Gypsum	5.75 ± 0.47	2.84 ± 1.19	2.30 ± 0.33
OMV Gypsum	395 ± 4.83	N	N
ALI Gypsum	N	N	262 ± 7.38
Raw Coal	28.7 ± 2.63	87 ± 2.24	142 ± 19

N: not added during cement production.

. The analyzed mined limestone utilized at PPC for cement production contained the lowest concentration of Hg (2.40 ± 0.28 ng g⁻¹ − 3.02 ± 0.32 ng g⁻¹, with a mean concentration of 2.54 ± 0.43 ng g⁻¹), followed by natural gypsum (2.30 ± 0.33 ng g ⁻¹ − 5.75 ± 0.47 ng g⁻¹, with a mean concentration of 3.63 ± 1.86 ng g⁻¹) and coal (28.7 ± 2.63 ng g⁻¹ − 142 ± 19 ng g⁻¹, with a mean concentration of 86 ± 57 ng g⁻¹). In contrast, the concentrations of Hg in fly ash (89 ± 25 − 160 ± 2.55 ng g⁻¹, with a mean concentration of 124 ± 50 ng g⁻¹), ALI synthetic gypsum (262 ± 7.38 ng g⁻¹), and OMV synthetic gypsum (395 ± 4.83 ng g⁻¹) were higher than those in naturally mined raw materials (limestone, gypsum, and coal).

3.5. Results of Total Hg Determination in Different Grades of Cements

The results for the determination of total Hg (ng g⁻¹) in cements are shown in

Table 4. Total Hg concentration in different grades of cement.

Cement Plant	Total Hg (ng g−1)
PPC Jupiter	6.56 ± 0.72 a
	31.0 ± 4.14 b
PPC De Hoek	1.20 ± 0.020 a
	2.93 ± 0.28 b
PPC Hercules	1.32 ± 0.83 a
	8.13 ± 2.50 b

^a CEM I 52.5N; ^b CEM II 42.5N.

. CEM I 52.5N is the cement that has high early strength, while CEM II 42.5N is the cement that has low early strength. The results of Hg determination in cements showed that CEM I 52.5N contains less Hg concentration than CEM II 42.5N. The level of Hg is higher in CEM II 42.5N due to the utilization of both synthetic gypsum and fly ash during cement milling, as these materials are not added during CEM I 52.5N production.

Fly ash and slag are added in PPC Jupiter as extenders, and OMV gypsum is added as a minor additional constituent (MAC); hence, a high Hg concentration of 31 ± 4.14 ng g⁻¹ is observed in CEM II 42.5N compared to cements produced in PPC De Hoek and PPC Hercules. Fly ash and slag are added in PPC Hercules as extenders, and ALI gypsum is added as MAC; hence, a lower concentration of Hg (8.13 ± 2.50 ng g⁻¹) is observed than in PPC Jupiter. A lower Hg concentration is also observed in PPC De Hoek CEM II 42.5N cements because slag is added as an extender and natural gypsum is added as a MAC.

3.6. Results of Total Hg Determination in By-Product of Cement Production

The results for the determination of total Hg (ng g⁻¹) in dust return are shown in Figure 2. Dust return is a by-product of combustion that is released from the raw meal material at the pre-heaters that operate at a temperature of up to 1000 °C and is captured at the dust collectors. Therefore, the components of the dust collectors also include the air blown from the clinker cooler through the pre-heaters.

3.7. Results of the Determination of Hg Thermospecies

Mercury thermospecies were determined on the raw materials utilized during cement production. Hg thermospecies are chemical forms of Hg that are released from materials at variable volatility temperatures. A summary of these results is shown in

Table 5. Results of the determination of total Hg and Hg thermospecies in materials utilized during cement production.

Material	Total Hg, ng g−1	Hg Thermospecies Concentration, ng g−1
		20–180 °C	180–360 °C	360–540 °C	540–720 °C
OMV Gypsum	208 ± 257	200 ± 252	-	-	-
ALI Gypsum	262 ± 5.37	228 ± 30	-	-	-
Fly Ash	126 ± 48	108 ± 48	-	-	-
Coal	121 ± 26	106 ± 22	-	-	-

and

Table 6. Results of the determination of total Hg and Hg thermospecies in dust return.

Cement Plant	Sample ID	Total Hg, ng g−1	Hg Thermospecies, ng g−1
			20–180 °C	180–360 °C	360–540 °C	540–720 °C
PPC De Hoek	DH14	235 ± 6.7	194 ± 4.0	–	–	–
	DH15	179 ± 1.8	71.4 ± 0.8	12.9 ± 0.10	20.9 ± 0.5	0.5 ± 0.02
	DH16	1663 ± 19.8	1223 ± 12	390 ± 2.0	–	–

.

Results of the assessment of the presence of Hg thermospecies in materials utilized during cement production showed that synthetic gypsums, viz., OMV gypsum (200 ± 252 ng g⁻¹) and ALI gypsum (228 ± 30 ng g⁻¹), fly ash (108 ± 48 ng g⁻¹), and coal (106 ± 22 ng g⁻¹) contain one Hg thermospecies at the 20–180 °C range, as mercury was not detected above this temperature range (Table 5). This means that all Hg compounds in these materials are emitted and captured in dust return during cement production as the cement mill kilns operate at a maximum of 1450 °C.

Results of the analysis show that dust return may contain one to four Hg species as indicated by the number of Hg compounds detected over the full temperature range, viz., 20–180 °C, 180–360 °C, 360–540 °C, and 540–720 °C (Table 6). Sample DH15 collected from the PPC De Hoek plant contained four Hg thermospecies, viz., 71.4 ± 0.8 ng g⁻¹ (20–180 °C), 12.9 ± 0.10 ng g⁻¹ (180–360 °C), 20.9 ± 0.5 ng g⁻¹ (360–540 °C), and 0.5 ± 0.02 ng g⁻¹ (540–720 °C). These concentrations are associated with different chemical forms of Hg in the dust return sample. It can therefore be concluded from these results that only dust return samples contain more than one Hg thermospecies, thus indicating that new Hg compounds of higher volatility temperatures are formed during cement production.

4. Discussion

The results of the determination of Hg in CRMs show that the calibration curve of Hg has a regression equation of y = 324x + 402 and a correlation coefficient r² = 0.9918, indicating that the calibration curve obtained was linear, as r² was approximately 1 and is used for quantification of Hg in cement and its raw materials. Based on this correlation coefficient, it is certain that this thermal decomposition to determine Hg content on solid samples is more favourable than chemical decomposition methods, which are usually accompanied by analyte loss or contamination introduced through chemical pre-treatment, as it is more rapid and offers high efficiency.

ALI synthetic gypsum, also known as phospogypsum, is a by-product formed during the production of phosphoric acid from natural phosphate rock by the wet process [35,36,37]. Phospogypsum is mainly composed of gypsum but also contains a high level of impurities such as phosphates, fluorides, and sulphates, naturally occurring radionuclides, heavy metals, and other trace elements, and therefore, Hg in ALI gypsum may have originated from natural phosphate rock. Similarly, the presence of high Hg in OMV gypsum could be attributed to it being a phosphate waste product, as phosphate ores are known to contain trace amounts of metals including mercury. Hg in fly ash comes from coal as it is a by-product of coal combustion [38,39,40].

Slag contained low Hg content that ranged between 1.58 ± 0.19 ng g⁻¹ and 3.23 ± 0.36 ng g⁻¹, with the lowest average of Hg concentration detected at PPC Hercules. A possible reason for the low concentration in slag is that slag is the waste matter separated from metals during the smelting or refining of ore at a blast furnace operating at temperatures as high as 1650 °C, which volatilizes all Hg compounds; thus, mercury is emitted and therefore does not form part of the final product [41,42]. Results of the analysis of slag samples of different origins by Moreda-Piñeiro et al. (2001) also showed a lower concentration of Hg in slag (0.09 ± 0.003 − 0.13 ± 0.001 ng g⁻¹), which is much lower than that established in this study [43].

Dust return shows a higher mercury content compared with all other materials that are utilized during cement production (Figure 2). A high concentration of Hg in dust return is due to the presence of Hg emitted from pre-heaters and kilns operated at 1000 °C and 1450 °C, respectively. This product poses environmental concerns due to its high Hg content if it must be disposed of into the environment, and because of that, it is reintroduced during the cement manufacturing process to minimize the amount of Hg that is released into the environment. Results of this study and others indicate that Hg compounds are released at 720 °C or at lower temperatures; thus, it can be concluded that all Hg compounds are released and captured by dust collectors during cement production with a higher degree of certainty at a 95% confidence level [34].

The identity of the Hg compounds in raw materials and cement is related to their temperature of decomposition. The compounds were identified as a function of volatility temperatures of their pure Hg compounds in others studies, viz., Hg₂O (100 °C), HgCl₂ (304 °C), Hg₂Cl₂ (383 °C), HgSO₄ (450–500 °C), Hg₂SO₄ (335–500 °C), HgO (500 °C), HgS (544 °C), HgF₂ (645 °C), and Hg₂F₂ (570 °C) [44,45,46]. It can therefore be concluded that mercury compounds are dominated by mercury compounds bound to oxygen. Mercury compounds that are formed in dust generated during cement production are chlorides, fluorides, sulphates, and sulphides.

The average total Hg concentration in fly ash in samples collected from PPC cement (149 ± 1.40 ng g⁻¹) is much lower than the average Hg concentration reported by Du et al. (2018) of 550 ± 28.7 ng g⁻¹ from different coal-fired power plants in China [47]. Mercury concentrations in limestone reported in the literature ranged between 4.70 ± 1.40 ng g⁻¹ and 34.6 ± 19.5 ng g⁻¹ [17,48,49,50]. These Hg concentrations are higher than those found in PPC plants (2.49 ± 0.50 ng g⁻¹), thus indicating that limestone inputs utilized during cement production have low Hg content. The Hg concentrations of 4.7 ± 1.40 ng g⁻¹ and 141 ± 6.70 ng g⁻¹ found by Cui et al. (2021) in limestone and coal, respectively, are higher when compared to the average Hg concentration of 2.49 ± 0.5 ng g⁻¹ and 121 ± 1.20 ng g⁻¹ found at PPC cement plants [50].

The Hg concentration in gypsum from PPC cement (228 ± 30 ng g⁻¹) is lower than that obtained by Kogut et al. (2021) of 265 ± 1.2 ng g⁻¹ [17]. Coal, a fuel that is utilized for heating during PPC plant cement production, contained 121 ± 1.20 ng g⁻¹, which falls within the range of 31.0 ± 5.60 ng g⁻¹ to 158 ± 9.0 ng g⁻¹ reported in [17,48,50,51].

The content of Hg in dust return or dust filter bags found in PPC plants (1663 ± 19.8 ng g⁻¹) compares well to those established in other studies of 1690 ng g⁻¹ ± 3.80 − 2448 ± 604 ng g⁻¹ [17,49]. In general, PPC plant raw materials have lower Hg content than those utilized in other cement plants.

Hg thermospecies obtained from PPC cement coals were detected at a single temperature, as compared to those obtained from the literature. Coal from PPC shows that coal contains single Hg species at low temperatures, as compared to coal samples analyzed by Mathebula et al. (2020) and Mashyanov et al. (2017) [34,51]. Reports of Hg thermospecies content in dust return, cement, or raw materials were not found in studies in the literature.

5. Conclusions

Mined raw materials utilized in PPC plants showed low concentration of Hg (1.58 ± 0.19 − 5.75 ± 0.47 ng g⁻¹) except for coal (28.7 ± 2.63 − 142 ± 19 ng g⁻¹) when compared to synthetic gypsum (395 ± 4.83 ng g⁻¹). The Hg concentration in CEM II 42.5N cement is higher when compared to CEM I 52.5N since synthetic gypsum and fly ash are added during cement milling. Dust return has the highest concentration of Hg as it contains Hg released from the combustion of materials in pre-heaters and kilns. Results of this study indicate that most Hg compounds are released at a temperature of 720 °C or lower. Therefore, it can be concluded that all Hg is released and captured by dust collectors during cement production. It was also found that dust return contains Hg thermospecies with volatility temperatures ranging from 20 °C to 720 °C, thus indicating that Hg compounds of high volatility temperatures are formed during cement production.