Agglomeration of coke particles by aliphatic binders: A hidden pollutant in effluent of steam crackers
Introduction
The traditional sources of microparticles are dusty cargo loaded and unloaded in the ports (Zhao et al., 2020). The amount of suspended solid particles in coastal waters around the loading ports (in Indonesia, China, Columbia and Australia), has been determined to be in the range of 10–511 ppm (Song et al., 2017; Benitez-Polo and Velasco, 2020). During handling and storage procedures, the coke and coal is placed as open piles at terminals leading to the release of coke fines into the coastal waters (Kroon et al., 2020). The wind has a significant impact on this issue when it moves up the coke particles from the open piles (Koval et al., 2018; Bai et al., 2020). Recently, the effect of micro particles such as coke fines on the marine organisms has been investigated (Gardon et al., 2018; Liu et al., 2018a; Pikula et al., 2020a, b). Coke and coal particles contain various inorganic species and organic compounds (with mutagenic and carcinogenic effects (Suyatna et al., 2017)) such as cadmium, chromium, lead and mercury as well as polycyclic aromatic hydrocarbons (PAHs) (George et al., 2020). The biological reactivity of coke particles in benthic ecosystem is explained by structure of the biological community, water temperature, strength of light, pH, oxygen content and salinity level of water (LaRowe et al., 2020; Middelburg, 2018). During sampling procedures around the port terminals thin layers of fine coke particles was detected in bottom sediments with different particle diameters in the range of <38 μm (Caballero-Gallardo et al., 2018), <40 μm (Benitez-Polo and Velasco, 2020), <63 μm (Berry et al., 2017) and ≤425 μm (Henley et al., 2015). Because of these changes in water composition of the oceans, a significant decrease (1% annually) in the population of phytoplankton has been reposted (Boyce et al., 2010; Whiteside and Herndon, 2019). On the other hand, zooplankton species accumulate PAHs when feeding through those exposed phytoplankton species (Hsieh et al., 2019). The effect of PAHs coming from coke particles on zebrafish was studied by Guerrero–Castilla et al. (Guerrero-Castilla et al., 2019). Moreover, it was shown that cadmium ions coming from coke particles are responsible for 50 % mortality in zebrafish (Renieri et al., 2017; Tretyakova et al., 2021).
Self–assembled monolayers of organic substrates are assigned the molecular assemblies adsorbed and organized spontaneously on a solid surface (Li et al., 2014). Such monolayers can act as the agglomeration function in some circumstances (Fameau and Zemb, 2014).
After immersion of nanoparticles and microparticles in biological fluids, a protein–based layer (protein corona) enshrouds the particles. This adsorbed layer is responsible for the upcoming interactions of suspended particles with the surrounding biological media rather than its original surface of particles (Polo et al., 2017; Mahmoudi, 2018). Based on this knowledge, the effect of some proteins was studied on the agglomeration of polystyrene nanoparticles (d<100 nm) and the results revealed the mechanism of molecular interactions occurring during agglomeration of polystyrene particles by the adsorbed layer of protein (Wang et al., 2019).
As with biological systems, the fundamental of suspended pyrolytic coke agglomeration (Pourabdollah, 2021a) and separation from aqueous media is the preferential wetting of lipophilic or hydrophobic properties of fine particles with respect to the organic binder or organic extractors (Pourabdollah, 2018). This potential allows the coke fines to be wetted by binder or oil phase while the other non–wet suspended materials such as corrosion products, remain in aqueous suspension (Pourabdollah, 2021b). Adequate amount of viscose hydrocarbons and sufficient mechanical agitation lead to form the oil coated coke particles agglomerate with each other to form submicron clusters (Mehrotra et al., 1983). From a thermodynamic point of view, the driving force for the wetting by such hydrocarbon binders and subsequent agglomeration of coke particles is decreasing of total surface energy of aqueous system. This process can be optimized if the surface energy of both coke–water and binder–water interfaces are high, and the surface energy of coke–binder interface be low.
Bourque and Rutledge studied the adsorption of a long–chain aliphatic alkane (n-pentacontane, C50H102) on graphene scaffold. The results revealed that the adsorption process of n-pentacontane was performed without polymerization. They extended the findings to other n-alkanes and they concluded that because of low rigidity of long–chain aliphatic alkanes, their heterogeneous accommodation and nucleation are improved on the surfaces of solid carbon scaffolds (Bourque et al., 2016, 2017). Carbon–based nano– and micro–structures have strong adsorption affinity for a variety of long–chain alkanes (Bourque and Rutledge, 2018; Haggenmueller et al., 2006; Trujillo et al., 2012; Zhuravlev et al., 2014). Interactions of eicosane (Chen et al., 2017) and tetracosane (Roth et al., 2016) with solid carbon structures have been studied and the results showed that adsorption or nucleation is the sole phenomena on the solid surfaces without polymerization. Yu et al. Yu et al. (2017) studied the simulated performance of solid carbon and silica surfaces by epitaxial self–assembled layers of a long chain normal alkane (n–hexatriacontane, C36H74). The results revealed that the binding energy of graphene to n–hexatriacontane is –76.7 to –79.8meVÅ–2 for monolayer to pentalayer, respectively. On the other hand, the binding energy of graphene to silica (α–quartz) was determined to be –58.8meVÅ–2 representing the stronger adsorption of well-ordered alkane layers to graphene than to silica surface.
Recent studies showed that short and long n-alkanes are adsorbed perpendicular and laid–parallel to the surface of solid carbon scaffold as a two-step process including adsorption and orientation. The simulation results revealed that the orientation of n-alkanes on the surface of solid carbon is controlled by temperature and interactions (n-alkane to n-alkane and n-alkane to carbon surface). Increasing the chain length resulted in raising the temperature required for n-alkanes orientation (Fang et al., 2019).
Design of experiments (DoE) by orthogonal array approach (Taguchi algorithm) is assumed to be an important and practical statistical tool for process optimization of many scientific fields such as coal agglomeration (Sapakal and Telsang, 2012; Hong, 2012; Fazeli et al., 2012; Sadeghi et al., 2012; Chang et al., 2011). Chary and Dastidar Chary and Dastidar (2012, 2010) used L9 orthogonal array to optimize the agglomeration process of an oil sample. They optimized the process parameters including agitation speed, binder density, oil dosage and agglomeration time in the first run; and coal particle size, type of oil, type of coal and binder density in the second run. Likewise, Aslan and Unal Aslan and Unal (2011) used L18 orthogonal array to form the maximum agglomerates by tuning the process parameters including agitation rate, agitation time, amount of binder and particle content.
The major interactions in sludge propensity of industrial effluents are including fluid–particle, particle–particle and surface–particle interactions, which are always present however their relative weights differ based upon the situation considered (Pourabdollah, 2017; Henry et al., 2012). The dependency of particle deposition on the surface–particle forces has been confirmed by single particle deposition tests even for 25nm (Antelmi and Spalla, 1999) and 499nm (Darbha et al., 2010) latex fines as well as 820nm (Sjollema and Busscher, 1990) polystyrene particles.
Based on DLVO theory, which was founded by Derjaguin and Landau Derjaguin and Landau (1941) and Verwey and Overbeek Verwey and Overbeek (1948), intra particle interactions are determined as the sum of electrostatic double layer and van der Waals interactions (Liang et al., 2007). This classical theory can easily explain the measurement of coarse interaction forces in the scale of micrometers; however, it is not able to account for fine interactions in the scale of nanometers (Liang et al., 2007). Furthermore, this classical theory was developed for a perfectly homogeneous and smooth surface and the role of surface roughness (ε) on the interactions were examined and some correlations were proposed (Martines et al., 2008; Zhao et al., 2008; Huang et al., 2010).
Shadow effect (Ko and Elimelech, 2000; Malysa et al., 1986) represents a coupling between inter-particle interactions and dynamic motion of particles. As one particle approaches, other particles encounter an electrostatic repulsion that forces them to move away from the approached particle. By moving the suspended particles towards the container walls, a shadow area is formed behind the pre-adsorbed particles (Adamczyk et al., 1995, 2002).
Semi-coke has been introduced as a solid carbon product (Liu et al., 2018b; Meng et al., 2021) prepared from low temperature carbonization of low metamorphic coal (Meng et al., 2020; Zhang et al., 2020a). They have been used for thickening and dewatering of waste activated sludge with some modifications on their particle size, dosage and ratio of semi-coke to polymer binder (cationic poly-acrylamide). Another application and field study of suspended coke particles is the adsorption of organic compounds such as quinoline (Wang et al., 2020), p-nitrophenol (Yang et al., 2019), elemental mercury (An et al., 2020), cyanide (Monda et al., 2019) and azo dyes (Hea et al., 2020) from aqueous solution onto the surface of coke particles. Production of three phase oil/water/coke slurries, two phase water/coal-coke slurries, and sludge/coke slurries has been documented in literature. With a decrease of particle size of suspended coke, the apparent viscosity of aqueous fluid is greatly increased leading to the improvement of fluid slurry-ability. In such conditions, the maximum amount of suspended coke ranging from 74 wt.% for the coke water slurry to 62 wt.% for coke emulsion slurry with oil to water ratio of 1:3 (Fan et al., 2019). Xu et al. Xu et al. (2021) investigated the effect of addition of sec-octyl and isoamyl alcohols on the viscosity of a coal water slurry sample and the results revealed that in the presence of both alcohols the surface tension and viscosity of the slurry are reduced. Industrial application of coke/water slurries comprising coal or coke particles (60–75 wt.%), water (24–39 wt.%) and modifier additives (1 wt.%) has been addressed by Li et al. Li et al. (2020) and Zhang et al. Zhang et al. (2020b).
The literatures dealing with the particulate deposits have not considered the following research fields including the pyrolytic coke particles, their agglomeration mechanism, the factors affecting fouling formation and their importance in industrial effluents. The novelty of this work is comprehensive investigation of agglomeration phenomena of pyrolytic coke particles in aqueous media and their effect on the formation of particle based deposits.
With the formation of coke based deposits, the concentration of suspended particulate matter below 5 μm (pm5) decreases in aqueous effluent of steam crackers. This process can be assumed as a separation method for remediation of effluent water from suspended coke particles of pm5 as pollutant. Next, the coke particles can be recovered from the separated deposits by downstream filtration (Tran et al., 2021; Lorenz et al., 2020), centrifugation (Azimi-Boulali et al., 2020; Gu et al., 2021) or settling (Farzam Rad and Moradi, 2020; Waldschlager et al., 2020) systems leading to the production of a solid fine coke paste with fuel capability.
In this paper the chemical composition of the self–assembled monolayer on the surfaces of pyrolytic coke particles was determined. Moreover, it was justified that how the elevated temperatures increase the agglomeration tendency of coke fines leading to the enhancement of sludge formation on marine surfaces. The chemical scaffold of self–assembled monolayer was determined and the behavior of suspended coke fine in aqueous phase was interpreted. The results explain this question that how the pyrolytic coke particles is considered as a potential source of pollution in oceans and rivers. Moreover, an adsorption mechanism was proposed for self–assembling of long–chain normal aliphatic alkanes (C14–C24) on the surface of coke microparticles at different temperatures.
Section snippets
Experimental method
The particles of pyrolytic coke were obtained by co–injection of dilution steam and ethylene feed to steam cracker apparatus at 800–820 °C and 606 kPa (Barazandeh et al., 2015; Pourabdollah et al., 2019). In dilution steam generator (DSG) unit, the pyrolytic coke fines suspended in quench water have adsorbed different volumes and species of long–chain normal aliphatic hydrocarbons in the range of tetradecane (C14) to tetracosane (C24). The particle agglomeration and fouling propensity of
Distribution of coke particles
The characteristic parameters of the produced particles of pyrolytic coke were determined at different operating conditions on the furnaces. Among characteristic parameters, the particle seize analysis revealed that the diameter range of coke particles were constant along with different operational conditions of furnaces. According to the trace of particle counting presented in Fig. 3, the submicron particles of pyrolytic coke were formed in the range of 0.2μm–5.0μm according to the operational
Conclusion
Sludge propensity of pyrolytic coke fines in submicron scale was investigated on the marine surfaces and the results revealed that a self–assembled monolayer including long–chain normal aliphatic alkanes is adsorbed on the micropore scaffold of pyrolytic coke particles. The chemical analysis and specifications showed the following conclusions:
- 1
GC–MS results showed that among a broad range of chemical substrates presented in the aqueous phase as solute, the heavier solutes are adsorbed on the
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
References (114)
- et al.
Flow-induced surface blocking effects in adsorption of colloid particles
J. Colloid Interface Sci.
(1995) - et al.
Deposition of particles in the impinging-jet cell for the high coverage regime
J. Colloid Interface Sci.
(2002) Factors affecting spherical oil agglomeration performance of coal fines
Int. J. Miner. Process.
(2002)- et al.
Regeneration performance of activated coke for elemental mercury removal by microwave and thermal methods
Fuel Process. Technol.
(2020) - et al.
Multi-response optimization of oil agglomeration with multiple performance characteristics
Fuel Process Technol.
(2011) - et al.
Investigation of coil outlet temperature effect on the performance of naphtha cracking furnace
Chem. Eng. Res. Des.
(2015) - et al.
Effects of suspended mineral coal dust on the energetic physiology of the Caribbean scallop Argopecten nucleus (born, 1778)
Environ. Pollut.
(2020) - et al.
Effects of coal contamination on early life history processes of a reef-building coral, Acropora tenuis
Mar. Pollut. Bull.
(2017) - et al.
Heterogeneous nucleation of an n-alkane on graphene-like materials
Eur. Polym. J.
(2018) - et al.
Coke formation in steam crackers for ethylene production
Chem. Eng. Process.
(2002)