WITHDRAWN: Adsorption kinetics of volatile organic compounds (VOCs) onto super activated renewable carbon (SARC) in low-carbon composites (LCCs)

doi:10.21203/rs.3.rs-3579746/v1

Download PDF

Research Article

WITHDRAWN: Adsorption kinetics of volatile organic compounds (VOCs) onto super activated renewable carbon (SARC) in low-carbon composites (LCCs)

https://doi.org/10.21203/rs.3.rs-3579746/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this older preprint version

Read the latest preprint version →

The full text of this preprint has been withdrawn, as it was submitted in error. Therefore, the authors do not wish this work to be cited as a reference. Questions should be directed to the corresponding author.

Editorial notes are used to provide important context regarding the topic of a preprint or to alert readers to potential issues concerning that preprint or a downstream publication associated with it. For more information on editorial notes, see our Editorial Policies.

This is the first study of adsorption kinetics of volatile organic compounds (VOCs) onto super activated renewable carbon (SARC), with the aim of maximizing the removal efficiency of odor during the compounding of low-carbon composites (LCCs). Three different micrometer-sized SARC samples were prepared by alkali (NaOH) modification of renewable carbon, named as small sized SARC (S-SARC), medium sized SARC (M-SARC) and large sized SARC (L-SARC), respectively. The morphology and geometrical properties of these prepared SARC were investigated via scanning electron microscopy (SEM) and Brunauer-Emmett-Teller (BET) analysis, and their surface groups were determined by Boehm titration. Molecular dynamics (MD) simulations were applied to elucidate the migration behavior of SARC particles, with the aim of homogenizing the particles’ distribution and increasing their contact opportunities with VOCs molecules. The performance of SARC samples employed as adsorbents in LCCs had been exploited, and their adsorption capacity were evaluated by gas chromatography-mass spectroscopy (GC/MS) upon identifying and quantifying the volatile compounds. The Langmuir and Freundlich adsorption isotherms were applied, and the results revealed Freundlich was better fitting for the adsorption kinetics. The influence of various factors, e.g. chemical and physical properties of SARC, on the adsorption of VOCs were compared, and the results confirmed the dominant chemical adsorption (Freundlich Isotherm). Thus, using SARC, especially S-SARC can significantly promote the removal of odorous VOCs, and open up new horizons of full utilization of LCCs as substitutes for traditional high-carbon plastics in broader industries.

Super activated renewable carbon (SARC)

Low-carbon composites (LCCs)

Volatile organic compounds (VOCs)

Odor removal

Molecular dynamics simulation

Adsorption kinetics

Low-carbon composites (LCCs), as a type of biodegradable material, have been recognized as potential alternatives for petroleum derived plastics with high carbon content. They have been applied in a wide variety of fields, including automotive, aerospace, construction, and furniture. Typical LCCs consist of a continuous thermoplastic or thermoset matrix, and lignocellulosic fibers as reinforcement fillers ¹. Upon compounding, the polymer matrix transmits external loads to lignocellulosic fibers via interfacial shear stresses, thus protecting fibers from excessive damage. Meanwhile, the use of reinforcing lignocellulosic fillers, e.g., wood fibers or agrowaste flours, can impart strength to LCCs and offer additional benefits such as easy availability and renewability. Moreover, compared with other reinforcement fillers, e.g., glass fibers or carbon fibers, lignocellulosic fillers can minimize the abrasive wear of the processing equipment and avoid the skin irritations in workers during handling ². In recent years, there has been a growing trend to increase the proportion of lignocellulosic fillers, in exchange for further reducing the use of nonbiodegradable plastics. This advances the biodegradability of LCCs to meet the stringent government environmental regulations; however, it will also exacerbate the issue of odor emission which is mainly sourced from the thermally degraded filler-fibers during compounding.

Volatile organic compounds (VOCs), as the main contributors to odor, are a highly diverse class of chemicals, which have been recognized as pervasive pollutants and carcinogens. During LCCs compounding, VOCs can easily emit as gases due to their volatile nature, thus causing greenhouse effects and ozone depletion ³. Moreover, VOCs are toxic and some are considered to be mutagenic, carcinogenic, or teratogenic; the spreading of VOCs in the air has been reported to be responsible for many acute and chronic health issues through direct contacting the skin and respiratory tract ⁴. In view of these negative impacts on environment and human health, it is of practical importance to remove the odorous VOCs, and adsorption is a superior technique for this purpose.

Various adsorbents have been applied for the VOCs adsorption, including zeolites, carbon, activated alumina, clay and silica. Among these materials, renewable carbon (RC) is particularly promising due to its highly porous structure and considerable specific surface area. Many studies of RC in literature have focused on the filtration and adsorption of VOCs in the air. However, for LCCs compounding, the early adsorption of VOCs molecules prior to their coming off is more important for increasing the VOCs removal efficiency. Firstly, the VOCs concentration is relatively higher inside of the LCCs matrix, compared with the open space in air, thus providing more chances for VOCs to contact RC particles. Secondly, the intensive shear force during compounding can enhance the interaction between RC and VOCs, therefore making it easier to trap VOCs molecules into the porous channels of RC. In this regard, the migration and distribution of RC particles inside of the LCCs matrix will be critical for increasing the VOCs adsorption efficiency; however, they have not yet been thoroughly investigated at the molecular level, due to the intrinsic scale limit of experimental approaches. This is where Molecular Dynamics (MD) simulation is valuable, allowing models to be established before full investing into experiments. The cost of MD is much lower compared with the lab experiments, and can be completed quickly. In MD, individual RC particles are modeled, and Newton’s equations are integrated over time to give information on the particles’ position. The outcome of MD would provide good understanding of the RC migration dynamics inside of LCCs domain, and a more homogeneous distribution allows more RC particles to interact with VOCs molecules for a higher adsorption.

In addition to the migration behavior of RC particles, their physical and chemical properties are crucial for the elimination of odorous VOCs during LCCs compounding. The limited surface area and poor functional groups of virgin RC may restrict its adsorption efficiency towards VOCs; however, its geometrical structures and surface properties can be upgraded by various modification processes. In this study, a simple alkali (NaOH) modification was applied to functionalize the RC particles into an innovative super activated renewable carbon (SARC). The NaOH can expand micropores and increase specific surface area via the reaction with carbon, meanwhile generating new active groups to promote the VOCs adsorption.

The aim of this work was to study the adsorption behavior of three different micrometer-sized SARC (S-SARC, M-SARC and L-SARC) towards VOCs, thus maximizing the removal efficiency of odor from LCCs. The impact of SARC particles’ properties on their adsorption capacity was investigated. Bleached chemical thermo-mechanical pulp (BCTMP) was used for preparing the LCCs by compounding with polypropylene (PP), a widely used polymer with excellent thermal properties, mechanical strength, and ease of processing. The surface functionalities and physical properties of three clusters of SARC were determined using Boehm titration, scanning electron microscopy (SEM), and Brunauer-Emmett-Teller (BET) surface area analysis. Gas chromatography-mass spectroscopy (GC/MS) was applied to identify and quantify the odorous VOCs, and the adsorption kinetics were analyzed via Langmuir and Freundlich adsorption isotherms.

2.1. Material and reagents

Commercial carbon made from coconut shell with ash content < 4% was purchased from the General Carbon Corp. (USA). Polypropylene homopolymer with MFI 40 g/10 min was supplied by Braskem. Bleached chemical thermo-mechanical pulp (BCTMP) was used as lignocellulosic fiber and was provided by Tembec (now known as Rayonier Advanced Materials Inc.). The obtained BCTMP was stored in a sealed chamber under temperature of 23.0 ± 1.0°C and relative humidity of 50.0 ± 2.0%. GC/MS pre-assembled glass vials were purchased from MilliporeSigma with volume of 15 mL and tan PTFE/silicone septum. All other chemicals of analytical grades are purchased from MilliporeSigma as well, and used without further purification.

2.2. Sieving and alkali modification of super activated renewable carbon (SARC)

The purchased renewable carbon was purified in boiling water for 3 h and then oven-dried at 105°C for 6 h. The obtained carbon powder was then sieved in a RETSCH Analytical Sieve Shaker AS 200 and was further separated by spinning disks into three granulometric fractions: small size (S-RC), medium size (M-RC), and large size (L-RC). Each fraction of carbon was soaked in NaOH solution (concentration of 10 M) at 70°C for 2 h, and then was placed on a rotation vibrator at 35°C for 24 h. The remained solution was filtered out and each fraction of samples was washed thoroughly by distilled water before oven-dry at 105°C for 4 h. Correspondingly, the three clusters of resulted super activated renewable carbon were labelled as S-SARC, M-SARC, and L-SARC, and stored in desiccator until further use.

2.3. Brunauer-Emmett-Teller (BET) analysis

The specific surface area and pore size distribution of S-SARC, M-SARC, and L-SARC were analyzed by Brunauer-Emmett-Teller (BET) method. 130 mg of samples were degassed under vacuum at 150°C for 10 h. The test was conducted in Quantachrome Instruments Autosorb iQ Station 1 with AsiQwin automated gas sorption data using nitrogen adsorption-desorption analysis at 77.35 K. The instrumental parameter “equilibration time” was set to 60 s and pressure tolerance was 0.1.

2.4. Scanning electron microscopy (SEM) and energy dispersive X-ray spectroscopy (EDS)

The powders of S-SARC, M-SARC, and L-SARC were dispersed in a mixed solution of distilled water, ethanol and acetone (80:15:5, v/v/v), and dried on a glass plate at 55°C for 24 h. The surface of the sample was gold coated for conductivity, and the accelerating voltage of SEM was set as 20 kV ⁵. The tests were conducted by a field emission scanning electron microscope coupled with an energy dispersive X-ray spectroscopy (EDS), mod. JEOL-6610LV (JEOL, Japan).

2.5. Molecular dynamics simulations

The coordinates, size and angles of S-SARC, M-SARC, and L-SARC were created using Visual Molecular Dynamics (VMD) software. The data was exported and saved as LAMMPS input file using the TK Console in the extensions tab of VMD active window. The modelling temperature was set as 473K (same as the compounding temperature in automotive industries). An energy minimization was applied as modelling periodic boundary during the simulation for all X, Y and Z directions.

2.6. Boehm titration

The Boehm titration works on the principle that oxygen groups from SARC have different degrees of acidity and can be neutralized by three bases: Na₂CO₃ is used to neutralize carboxylic and lactonic groups, NaHCO₃ neutralizes only carboxylic acids, and NaOH neutralizes all Brønsted acids containing carboxylic, lactonic, and phenolic groups. The samples were first dried in a vacuum oven at 105°C for 12 h. 2 g of sample was placed in 25 ml of 0.05 M solutions of Na₂CO₃, NaHCO₃, NaOH and HCl, respectively, in sealed vials. The vials were shaken in a vibrator (150 rpm) for 24 h and then filtered. An aliquot of 5 mL of each filtrate was pipetted, and back titrated by standardized HCl for basic solutions and by standardized NaOH for acidic solutions⁶.

2.7. LCCs-SARC compounding

25 wt% of BCTMP was blended with polypropylene pellets and three different sized SARC, respectively (Table 1). 2 wt % of processing additives, such as coupling agents, processing aids and UV stabilizers were used in each sample. After drying at 80°C for 6 h in a convection oven, the three samples were melt-blended in a Brabender at heating temperature of 200°C and rotation speed of 50 rpm. The extrudates were pelletized using an inline pelletizer for further use.

Table 1

Formulation of SARC embedded LCCs
PP (wt%)	BCTMP (wt%)	SARC (wt%)	Additives (wt%)
72.65	25	0.35	2

2.8. Gas chromatography-mass spectroscopy (GC/MS) analysis

GC/MS analysis was performed by a Thermo ISQ-7000 GC/MS operating in electron ionization (EI, 70 eV) mode. Separation was achieved using a 30 m × 0.25 mm i. d., 0.25 µm, Rxi-5 ms capillary column (Restek). The oven temperature was initially set at 50°C and held for 0.5 min, then was increased at 11°C/min to 230°C (held for 1 min). The oven equilibration time was 3 min. The front inlet gas flow rate was controlled at 1 mL/min. The GC inlet temperature was 230°C, with a split ratio of 20:1. The ion source temperature was set as 270°C. The scanned mass range of spectrometer was from 30 m/z to 1100 m/z. The injected volume of sample was 10 µL.

2.9. Langmuir monolayer adsorption isotherm

Langmuir isotherm model assumes monolayer VOCs adsorption onto the SARC particles, and the linear form of Langmuir isotherm equation is given as:

\(\frac{{C}_{e}}{{Q}_{e}}=\frac{{C}_{e}}{{Q}_{m}}+\frac{1}{{K}_{L}{Q}_{m}}\) Eq. (1)

where C_e is the equilibrium concentration of VOCs (mg/L), Q_e is the adsorbed VOCs amount per gram of SARC at equilibrium (mg/g), Q_m is the maximum monolayer coverage capacity (mg/g), and K_L is the Langmuir isotherm constant (L/mg). Of which the value of Q_m and the K_L could be computed from the fitting curve.

2.10. Freundlich multilayer isotherm

Freundlich isotherm assumes the heterogeneous surface energies and its typical logarithmic form is given by Eq. (2):

\(log{Q}_{e}=log{K}_{F}+\frac{1}{n}log{C}_{e}\) Eq. (2)

where Q_e is the adsorbed VOCs amount by SARC at equilibrium (mg/g), C_e is the equilibrium concentration of VOCs (mg/L), K_F (mg/g (l/mg)^1/n) is the adsorption capacity of SARC which can be defined as the adsorption or distribution coefficient and represents the quantity of VOCs adsorbed onto SARC for a unit equilibrium concentration, n is the Freundlich constant which indicates how favorable is the adsorption process, the value of 1/n smaller than 1 indicates a normal Langmuir isotherm while 1/n larger than 1 is the cooperative adsorption.

3.1. Properties of different sized SARC

The SEM images of S-SARC, M-SARC, and L-SARC are shown in Fig. 1 (c), (d), and (e), which exhibit different surface morphology after NaOH modification. L-SARC samples have uneven surfaces with unique equiaxed pores in cross directions, thus exponentially amplifying their adsorption capacity. Their internal pore structure is clearer and orderly, and the surface pores are more open with thinner walls. These thin walls may increase the surface area to some degree, however, making them vulnerable to collapse. As shown in Fig. 1 (e), a certain number of destructed lamellar pores are visible upon the corrosion of NaOH. In some areas, adjacent macropores are merging into super large pores. One of the possible reactions may be attributed to NaOH corrosion was proposed by Tryk et al. ⁷:

\(\text{C}+6{\text{O}\text{H}}^{-}= {\text{C}\text{O}}_{3}^{2-}+3{\text{H}}_{2}\text{O}+4{\text{e}}^{-} ; {{\Delta }\text{G}}^{^\circ }=-70.72 \text{k}\text{c}\text{a}\text{l}/\text{m}\text{o}\text{l}\text{e}\) Eq. (3)

It is also known that:

\(4{\text{H}}^{+}+{\text{O}}_{2}+4{\text{e}}^{-}=2{\text{H}}_{2}\text{O}; {{\Delta }\text{G}}^{^\circ }=-113.38 \text{k}\text{c}\text{a}\text{l}/\text{m}\text{o}\text{l}\text{e}\) Eq. (4)

Combining both equations:

\(\text{C}+2{\text{O}\text{H}}^{-}+{\text{O}}_{2}= {\text{C}\text{O}}_{3}^{2-}+{\text{H}}_{2}\text{O} ; {{\Delta }\text{G}}^{^\circ }=-184.1 \text{k}\text{c}\text{a}\text{l}/\text{m}\text{o}\text{l}\text{e}\) Eq. (5)

The OH⁻ ions from NaOH are consumed during alkali modification, and react with carbon with the aid of oxygen. This process generates ionized carbon (CO₃²⁻), resulting in the collapse of carbon walls meanwhile generating more internal pores for L-SARC. On the other hand, the medium and small sized SARC particles (M-SARC and S-SARC) have relatively thicker pore walls, which can resist the excessive corrosion collapse caused by NaOH solution. As shown in Fig. 1 (d), M-SARC particles possess well defined micropattern in their porous channels. The honeycomb structure is visible among the larger M-SARC particles. Most M-SARC particles are crosslinked with each other, possibly due to the strong interactions among their remaining acidic groups. These groups are mostly oxygen-containing groups, e.g. carboxyl, lactone and hydroxyl groups, which can generate hydrogen bonding or electrostatic interactions between each other. While for S-SARC samples (Fig. 1 (c), their sizes are much smaller than those of the others, indicating that S-SARC consist of more micropores. These pores are composed of crystalline structure, usually featuring internal cavities of variable shapes and diameters which are able to adsorb VOCs molecules via capillary and partitioning mechanisms ⁸. The individual particles are evenly dispersed inside of the domain and have very minimum interference between particles. As shown in the size distribution histogram (Fig. 1 (f)), L-SARC particles are relatively uniform in their size with a narrow distribution, and d_av is about 175.1 µm. In contrast, M-SARC has a heterogenous size distribution with an average diameter of 80.5 µm. The majority of S-SARC particles have fairly homogeneous texture and size distribution. The average diameter of S-SARC is 28.6 µm, which is the smallest among the three SARC samples.

The micro- and meso-pore size distributions of three SARC samples were determined with BJH model. As shown in Fig. 1 (a), the pore size distribution was similar for all samples and relatively intensive. The average pore diameter increased as the size of SARC increased, and the peak values were 1.71 nm, 1.93 nm and 2.45 nm for S-SARC, M-SARC and L-SARC, respectively. According to the IUPAC (International Union of Pure and Applied Chemistry) classification, both S-SARC and M-SARC were microporous (< 2 nm), while L-SARC were mesoporous (2–50 nm)⁹.

With high resolution energy-dispersive X-ray spectroscopy (EDS), the elemental compositions of different sized SARC were determined and the analysis results are shown in Fig. 1 (b). Carbon (C) and oxygen (O) are the main elements for all three samples with total concentration of more than 99% (99.17% for S-SARC; 99.21 for M-SARC; 99.22 for L-SARC). As the size of SARC increases, the carbon content decreases, while the oxygen content increases. The majority of oxygen elements are sourced from acidic functional groups, and ratio of O/C can be used to characterize the degree of polarity and hydrophobicity. The high O/C ratio may result in a weakened affinity towards organic VOCs, thus negatively affecting the adsorption capacity. Alkali modification is able to reduce the oxygen-containing groups. As shown in Figure. 1(b), the smaller the diameter of SARC, the less the number of oxygen element and the ratio of O/C. This is because the smaller sized SARC particles are more accessible by NaOH solution due to their larger specific area, resulting in more acidic groups being modified.

3.2. Molecular dynamic simulation

Particle size, as an important factor, can significantly influence the migration of SARC particles inside of LCCs, thus affecting their adsorption towards VOCs molecules. Figure 2 (a-c) shows the molecular models of LCCs matrix consisting of different sized SARC particles in PP domain. The figures (a), (b), and (c) correspond to S-SARC, M-SARC, and L-SARC, respectively. Due to the arbitrary scales of MD simulation, magnitudes of these three SARC samples were relative. It can be observed that as the size decreases, the distribution of SARC becomes more homogeneous, suggesting better contact between SARC particles and polymer chains.

The migration behavior of SARC particles inside the domain was simulated by LAMMPS-VMD software. In the user-defined coordinate system, the elapsed time (fs) and the x- and y-coordinates of the individual particles were recorded for the calculation of particles’ migration velocity:

\(v=\frac{\sqrt{{\left({x}_{s}-{x}_{i}\right)}^{2}+{\left({y}_{s}-{y}_{i}\right)}^{2}}}{{t}_{s}-{t}_{i}}\) Eq. (6)

where v is the particles’ velocity, x_s is the particle’s x-coordinate in the stopping position, x_i is the particle’s x-coordinate in the initial position, y_s is the particle’s y-coordinate in the in the stopping position, y_i is the particle’s y-coordinate in the initial position, t_s is the time at the stopping position, and t_i is the time at the initial position.

Figure 2 (d), (e), and (f) are the migration profiles of S-SARC, M-SARC, and L-SARC, respectively, in 3D view (Ortho, Front, Left and Perspective). The flow of these particles may be driven by shear stress, Coulomb friction as well as inter-particle electrostatic interactions. The relaxed confining stress present at the surface of particles would allow them to readily arrange during compounding. The results were plotted in terms of particles’ velocity and elapse time. There was a significant decrease in particles’ velocity as their size increased, and the M-SARC and L-SARC particles migrated with velocities of only 10–40% of those of S-SARC particles. This may be because S-SARC meandered more easily through the fortuitous channels inside of the domain, thus having higher velocities. Moreover, the large fluctuation of velocity trajectory indicates S-SARC particles’ travel may be impeded by frictional interactions between particles, thus migrating in multiple modes such as shear flow, rolling and sliding with different speed. In contrast, a relative narrow range of velocities was observed for both M-SARC and L-SARC. This may be attributed to their larger size and the viscous PP matrix, which would have fewer particle-particle interactions to affect their rearrangement. The higher migration velocity of particles allows them to contact the domain more frequently and distribute homogeneously, thus being more effective in capturing VOCs.

3.3. Calculation of adsorption energy

Shown in Fig. 3 (a), (b), and (c) are the adsorption isotherm of N₂ on the three clusters of SARC (insert), which can be categorized as Type I (Langmuir isotherm), where nitrogen molecules are adsorbed physically by micropore filling, according to IUPAC classification (IUPAC, 1985). The adsorption energy is further calculated based on the isotherm profiles. As is known, adsorption is driven by the adsorption force field on the surface of SARC. This unique force field generates an adsorption potential that is determined by the adsorbate, the adsorbent, and the distance between the adsorbate molecules and adsorbent surface. The Dubinin-Radushkevich (D-R) equation is able to adequately characterize these parameters, and the adsorption energy of SARC can be calculated by D-R equation via linearly fitting the adsorption isotherm. The D-R equation can be expressed as follows ^10,11:

\(W={W}_{0}exp{\left[\frac{-A}{{\beta E}_{0}}\right]}^{2}={W}_{0}exp{\left[\frac{-A}{E}\right]}^{2}\) Eq. (7)

\(lnW=ln{W}_{0}-2.303{\left(\frac{RT}{E}\right)}^{2}\) Eq. (8)

where W is the equilibrium adsorption amount under P/P₀, mL/g; W₀ is the micropore adsorbed volume, mL/g; A is the adsorption potential, which equals to RTln (P₀/P); β is an affinity coefficient; E₀ and E are the normal N₂ adsorption energy and characteristic adsorption energy, respectively; P₀ and P are the saturation vapor pressure and equilibrium pressure, respectively; R is the conventional gas constant; and T is the absolute temperature.

The linear correlation of the N₂ adsorption isotherm for the three clusters of SARC was determined according to the D-R equation, and the relationship between ln²(P₀/P) and lnW was obtained. The slope of the fitting curve in Fig. 3 (a-c) is 2.303(RT/βE₀)², and the affinity coefficient of nitrogen is 0.33. Thus, the adsorption energies of S-SARC, M-SARC, and L-SARC were calculated as 25.29 kJ/mol, 24.80 kJ/mol, and 23.98 kJ/mol, respectively.

Acidity and basicity of three SARC samples are presented in Fig. 3 (d). Boehm titration confirmed the lowering of total acidic groups as the size of SARC decreased, with values of 0.178 mmol/g, 0.165mmol/g, and 0.125 mmol/g for L-SARC, M-SARC, and S-SARC, respectively. This result is in good agreement with the trend of oxygen element and the ratio of O/C in the previous section of 3.1. Properties of different sized SARC. While for total alkaline groups, an opposite trend is presented, with S-SARC having the highest alkaline groups on its surface, followed by M-SARC and L-SARC.

3.4. Identification and quantification of VOCs by GC/MS

GC/MS spectroscopy was applied for measuring the odorous VOCs. Both lignocellulosic fibers and PP can emit VOCs due to the thermal degradation of lignocellulosic fibers, and the reaction of PP with oxygen present in the extruder barrel. However, at the temperature range for LCCs compounding (180–220°C), lignocellulosic fibers are the main producers for VOCs. Figure 4 presents the various emission levels of VOCs from LCCs embedded with S-SARC, M-SARC, and L-SARC, respectively. A complicated composition of VOCs can be observed, which includes more than 120 compounds. Most of these compounds possess unpleasant odors, and some of them are carcinogens. Therefore, it is necessary to adsorb the volatile compounds and prevent them emitting into the environment. Worth mentioning, these volatiles can also be useful for other purposes, e.g., pesticide or pharmacology, upon proper collection and treatment. Moreover, as the size of SARC decreases, both the numbers and intensity counts of the GC peaks - which stand for the compounds escaped from the adsorption, are reduced, indicating that the smaller sized SARC particles are more efficient in the capture of VOCs.

MS analysis was applied to identify the GC peaks, and the detected compounds were categorized in Table 2. The concentration of compounds was calculated based on the integrated chromatographic area of GC peaks. The main types of VOCs were phenols, aldehydes, ketones, acid derivatives, furfural and furan derivatives, hydrocarbons, and others. Among which, the most abundant compounds were acid derivatives. These acids were mainly from fiber-pyrolysates, and their high content indicated that a large amount of cellulose and hemicellulose were degraded during the LCCs compounding. Similarly, the furfural and furan derivatives were produced from lignocellulosic fibers which were rich in pentosan, and phenolic volatiles were generated from the pyrogenation of lignin. Aldehyde, ketones, and other molecules from wood-resin, were among the least emitted volatiles, however, usually had the strongest odor. For these identified as hydrocarbons, e.g., Toluene and Benzene, the minority of them were from PP, while the majority ones were generated by lignocellulosic fibers. This result is in good agreement with previous studies which concluded that wood sawdust was the main pollution source because the unpleasant odor was mostly generated from sawdust rather than from PP during LCCs manufacturing ¹².

Table 2

GC/MS analysis of VOCs and their chemical nature.
Compounds	Contents ((µg/L)			Acidity	Hydrophobicity	Polarity
Compounds	S-SARC	M-MARC	L-SARC	Acidity	Hydrophobicity	Polarity
Phenols	21.09	28.57	34.44	Weak acid	Mostly hydrophilic	Polar
Aldehydes	12.15	11.41	12.55	Neutral	Hydrophilic	Polar
Ketones	13.37	11.83	15.31	Neutral	Hydro-neutral	Polar
Acid derivatives	36.25	41.65	42.99	Acid	Hydrophilic	Polar
Furfural and furan derivatives	19.70	22.08	22.25	Weak base	Hydrophobic	Polar
Hydrocarbons	16.98	18.01	17.40	Neutral	Hydrophobic	Nonpolar
Others^*	15.47	8.37	2.15	Mixed	Mixed	Mixed

^*Others include all detected minor compounds with or without odor.

The detected compounds were further classified for their acidity, hydrophobicity and polarity. The emitted VOCs of S-SARC embedded LCCs had the lowest amount of acidic, hydrophilic, and polar compounds (about 42%, 51%, and 61%), followed by M-SARC (about 49%, 58%, and 66%), and L-SARC (about 53%, 61%, and 72%). This indicated S-SARC had a good adsorption capacity towards acidic, hydrophilic and polar VOCs. The reason is, as discussed in section 3.1. Properties of different sized SARC, S-SARC were more easily to be infiltrated by alkali (NaOH) and had more alkaline groups being introduced, thus exhibiting higher affinity towards VOCs with opposite nature. Similar results were reported in the previous studies upon investigating the influence of chemical properties of VOCs on their adsorption onto carbons with different surface functionalities ^13,14. The modification which weakened the surface acidity of carbons was able to enhance their hydrophobicity and affinity with VOCs of various polarity ¹⁵.

The Langmuir isotherm is valid for the adsorption with uniform energies on a surface containing a finite number of identical sites, and only one molecule can be adsorbed at each site. The adsorption fitted Langmuir isotherm is considered as monolayer adsorption with no transmigration of adsorbates in the plane of the adsorbent surface. Figure 5 (a) shows the fitting curve between C_e/q_e versus C_e, with a slope of 1/Q_m and an intercept of 1/K_LQ_m. As the particle size of the samples decreased, the maximum coverage capacity Q_m increased. Among the three samples, S-SARC had the best fitting of R² = 0.8847, and the highest VOCs adsorption with Q_m=0.058 mg/g. Similar influences of adsorbents’ particle size on their adsorption capacity have been observed previously, e.g., Alkherraz et al. studied the biosorption behavior of activated carbon towards metal ions including Pb(II), Zn(II), Cu(II), and Cd(II). Their activated carbon was prepared from olive branches with different particle sizes of 125, 500 and 800 µm, and the results showed an increase in the uptake of metal ions as the size of activated carbon decreased ¹⁶.

To compare, the adsorption data was further analyzed by Freundlich isotherm, which is applicable for multilayer adsorption on heterogeneous sites. A linear Freundlich isotherm model was applied by plotting LogQ_e against LogC_e, due to the fact that adsorptions isotherm at low adsorbate concentration, e.g., VOCs emitted from LCCs, are often linear. In comparison with Langmuir model, regression analysis of the data from all three samples fitted Freundlich adsorption isotherm better (Fig. 5 (b)), with R² of 0.9919, 0.9694, and 0.9253 for S-SARC, M-SARC and L-SARC, respectively. The intercept of the fitting curve is log(-K_f) and the slope is 1/n. The constant K_f is an approximate indicator of adsorption affinity for VOCs. The larger the K_f, the higher the maximum adsorption capacity. As for n, its value is a function of adsorption strength, and a favorable adsorption tends to have a n between 1 and 10. Of which, n = 1 implies a linear adsorption leading to identical adsorption energies for all sites, while the larger value of n (smaller value of 1/n) corresponds to a more heterogeneous adsorption, which usually results in an increased non-linearity. As shown in Fig. 5 (b), all SARC samples have n > 1; meanwhile S-SARC has the minimum n = 1.491 and highest K_f=0.106 mg/g, implies its strong linear Freundlich adsorption towards VOCs.

3.6. Influence of physical and chemical properties on VOCs removal capacity

The alkali (NaOH) modification provided renewable carbon with high porous structure as well as rich surface functionalities including both acidic and alkaline groups, e.g., oxygen-containing, nitrogen-containing, and aromatic groups. These unique physical and chemical properties allowed the prepared SARC particles adsorbing various VOCs as well as thermally degraded gases during LCCs manufacturing.

Figure 6 (a-b) presents the physical properties of SARC samples and their influence on VOCs removal capacity. The NaOH solution erodes the carbon surface, and the reaction between carbon and NaOH may form more pores from micro-size to macro-size (as shown in Eq. (3), (4) and (5)). These newly generated pores can increase the BET surface area (S_BET) and migration channels for VOCs. Specifically, S-SARC had the highest S_BET after NaOH modification, followed by M-SARC and L-SARC. The high S_BET was proposed to have stronger adsorption towards adsorbates due to higher surface energy in previous work ^17,18, which is in good agreement with the results in this study. Similarly, the total pore volume and micropore volume have the following order: S-SARC>M-SARC>L-SARC. Moreover, S-SARC has the maximum ratio of micropore/total pore volume (78.41%). There is a general agreement that micropores are favorable in the adsorption of small guest molecules, e.g., VOCs, while meso- and macropores provide fast and unobstructed channels for transporting the VOCs to the active sites in micropores ¹⁹. As the diameter of SARC increases, the ratio of micropore/pore volume decreases, which negatively affects the adsorption of VOCs as shown in Figure 6 (b).

Linear regressions were also performed on the chemical properties versus removal capacity. As shown in Fig. 6 (d), the effect of total alkaline groups on the removal capacity demonstrated a similar positive trend with that of physical properties (S_BET and pore/micropore volume), while the total acidic groups presented an opposite trend (Fig. 6 (c)). Compared with physical properties (R² = 0.9746 for S_BET, 0.9796 for total pore, and 0.9921 for micropore), the adsorption capacity of SARC was more sensitive to the changes in its surface functionalities (R² = 0.9845 for total acidic, and 0.9977 for total alkaline), indicating its superior chemical adsorption (Freundlich Isotherm). Similar results were report with other carbon-based materials, e.g. Lin et al. found that activated-carbon fiber (ACF) with less surface oxygen groups (< 900 µmol/g) tended to adsorb more nonpolar VOCs (benzene, toluene) rather than polar VOCs (acetaldehyde, acetone) ²⁰. Yu et al. compared the adsorption behavior of GO and rGO towards benzene and toluene at room temperature. Their results showed rGO exhibited higher adsorption capacities of benzene and toluene (276.4 and 304.4 mg/g) than that of GO (216.2 and 240.6 mg/g), possibly due to the more hydrophobic nature of rGO after the plentiful oxygen groups were removed, which favored the adsorption of nonpolar or weak polar VOCs ²¹. Li et al. compared the performance of various acids and bases for their surface reduction modification of coconut shell-derived granular activated carbon (AC), in order to improve the adsorption of hydrophobic carbon-based gases. Their results showed that alkali modification increased the specific surface areas and pore volumes of AC, and decreased its total content of oxygen functional groups. The modified AC exhibited a good adsorption towards o-xylene; in contrast, the acid-modified AC exhibited a weaker adsorption as its total surface oxygen-containing groups increased ²².

Also noticed quite a few literatures reported an opposite result that the acid-modified carbon had higher VOCs adsorption, which was reasonable considering the polarity, hydrophilicity and the limited types of compounds in their investigated system ²³. Moreover, there existed competitive adsorptions, in which weakly affinitive VOCs would be displaced by the strongly affinitive ones, though they had similar volatility and chemical nature ²⁴. In the present study, the VOCs from LCCs were a wide variety of organic mixtures (as shown in section 3.4. Identification and quantification of VOCs by GC/MS), and alkali treatment was demonstrated as a more effective approach in modifying SARC to enhance its removal capacity of VOCs.

The synergistic effect of physical characteristics and surface chemistry on the adsorption dynamic was illustrated in Fig. 7. During LCCs compounding, lignocellulosic fibers were thermally degraded into various volatiles, meanwhile generating certain amount of residual ash particles. These ash particles could adhere to the external surface of SARC, and some particles could even block the entrance of macropores once their size matched, thus hindering the diffusion of VOCs from surface into the internal pores. Therefore, the micropores connected with them could not adsorb VOCs and decrease the adsorption rate. As shown in the SEM images, macropore were more easily being blocked, while meso- and micropores were majorly structured with open pores. S-SARC contained a lower proportion of macropores, thus was less vulnerable to ash blockage. More open pores, higher micropore volume as well as larger surface area determined the superior performance of S-SARC towards VOCs adsorption.

As for the chemical properties, it was reported that the surface functionalities of adsorbent can make certain contributions to total surface area, thus enhancing adsorption as well ²⁵. On top of that, the interactions between functional groups and VOCs molecules can generate chemical bonds or electrostatic attractions, allowing the volatile molecules to be efficiently captured. The oxygen-containing groups are the among the most dominant species on the surface of pre-modified carbon, which can be divided into acidic, neutral, and alkaline groups. While for nitrogen-containing groups, the extra p-electrons of pyrollic and quaternary nitrogen at high energy state would promote the oxidation reaction by forming hydrophilic superoxide ions, thus generating abundant active adsorption sites ²⁶. The majority of these two groups prefer to adsorb the polar VOCs. On the other hand, aromatic and other minor groups are mostly neutral or alkaline, and have more affinity towards nonpolar VOCs. After NaOH modification, more alkaline groups on SARC were introduced, thus allowing it to adsob more VOCs from LCCs, which were majorly composed of nonpolar organic compounds.

This study showcased the excellent performance of super activated renewable carbon (SARC) particles in the odor removal, and their adsorption kinetics towards volatile organic compounds (VOCs) during the compounding of low-carbon composites (LCCs). Among the three different sized SARC prepared by NaOH modification (small sized SARC (S-SARC), medium sized SARC (M-SARC) and large sized SARC (L-SARC)), S-SARC exhibited the highest activity in capturing VOCs. Its fast migration velocity and better distribution inside the polymer domain were demonstrated via molecular dynamics (MD) simulation, which would increase its contact opportunities with VOCs molecules thus favoring the adsorption. Scanning electron microscopy (SEM) revealed the different porous structure of three SARC samples, and the dominant presence of alkaline groups on the S-SARC surface was confirmed by Boehm titration. More than 120 compounds were identified by GC/MS, and the adsorption kinetics were analyzed based on the quantified VOCs content from mass spectroscopy (MS). The obtained adsorption data were better fitted in the linear form of Freundlich isotherms, rather than Langmuir isotherms. Compared with physical properties, the adsorption capacity of SARC was more sensitive to the change in its chemical functionalities, especially the total alkaline groups. The finding in this study provides important information for deeper understanding of the adsorption dynamics of VOCs onto porous SARC. The results shed some light on the design of functional carbon adsorbent with tunable adsorption affinity for the efficient removal of VOCs during LCCs compounding.

Acknowledgements

The authors would like to acknowledge the financial support from Ford Canada.

Ethical Approval

Not applicable.

Funding

Funding was provided by Ford Canada. We acknowledge and appreciate it.

Availability of data and materials

Not applicable.

Zhou, H. et al. Recycling end-of-life WPC products into ultra-high-filled, high-performance wood fiber/polyethylene composites: A sustainable strategy for clean and cyclic processing in the WPC industry. Journal of Materials Research and Technology 18, 1–14 (2022).
Olakanmi, E. O. & Strydom, M. J. Critical materials and processing challenges affecting the interface and functional performance of wood polymer composites (WPCs). Materials Chemistry and Physics 171, 290–302 (2016).
Li, B. & Mi, C. On the adsorption of volatile organic compounds on hydroxyl-functionalized carbon nanotubes in aqueous solution. Diamond and Related Materials 125, 108994 (2022).
Zhao, Q., Wang, Q., Li, Y., Ning, P. & Tian, S. Influence of volatile organic compounds (VOCs) on pulmonary surfactant monolayers at air-water interface: Implication for the pulmonary health. Colloids and Surfaces A: Physicochemical and Engineering Aspects 562, 402–408 (2019).
Sun, B. et al. Interaction of a spirooxazine dye with latex and its photochromic efficiency on cellulosic paper. Carbohydrate polymers 95, 598–605 (2013).
Boehm, H. P. Chemical identification of surface groups. in Advances in catalysis vol. 16 179–274 (Elsevier, 1966).
Chiang, H.-L., Huang, C. & Chiang, P. The surface characteristics of activated carbon as affected by ozone and alkaline treatment. Chemosphere 47, 257–265 (2002).
Lin, S. et al. Marine macroalgae-derived multielement-doped porous biochars for efficient removal of sulfamethoxazole from aqueous solution: Sorption performance and governing mechanisms. Journal of Analytical and Applied Pyrolysis 171, 105963 (2023).
Zdravkov, B., Čermák, J., Šefara, M. & Jank\uu, J. Pore classification in the characterization of porous materials: A perspective. Open Chemistry 5, 385–395 (2007).
Rand, B. On the empirical nature of the Dubinin—Radushkevich equation of adsorption. Journal of Colloid and Interface Science 56, 337–346 (1976).
Dubinin, M. Fundamentals of the theory of adsorption in micropores of carbon adsorbents: Characteristics of their adsorption properties and microporous structures. Carbon 27, 457–467 (1989).
Wang, S. & Zhang, A. Chemical characterization of smoke from the production process of wood-plastic composites. Forestry Studies in China 9, 57–62 (2007).
Lillo-Ródenas, M., Cazorla-Amorós, D. & Linares-Solano, A. Benzene and toluene adsorption at low concentration on activated carbon fibres. Adsorption 17, 473–481 (2011).
Baur, G. B., Beswick, O., Spring, J., Yuranov, I. & Kiwi-Minsker, L. Activated carbon fibers for efficient VOC removal from diluted streams: the role of surface functionalities. Adsorption 21, 255–264 (2015).
Qiu, W. et al. Hierarchical pore structure of activated carbon fabricated by CO2/microwave for volatile organic compounds adsorption. Chinese Journal of Chemical Engineering 26, 81–88 (2018).
Alkherraz, A. M., Ali, A. K. & Elsherif, K. M. Removal of Pb (II), Zn (II), Cu (II) and Cd (II) from aqueous solutions by adsorption onto olive branches activated carbon: equilibrium and thermodynamic studies. Chem. Int 6, 11–20 (2020).
Jiang, Y.-H. et al. Characteristics of nitrogen and phosphorus adsorption by Mg-loaded biochar from different feedstocks. Bioresource Technology 276, 183–189 (2019).
Sun, B., Hou, Q., He, Z., Liu, Z. & Ni, Y. Cellulose nanocrystals (CNC) as carriers for a spirooxazine dye and its effect on photochromic efficiency. Carbohydrate polymers 111, 419–424 (2014).
Zhang, Y.-J., Xing, Z.-J., Duan, Z.-K., Li, M. & Wang, Y. Effects of steam activation on the pore structure and surface chemistry of activated carbon derived from bamboo waste. Applied Surface Science 315, 279–286 (2014).
Lin, C.-L., Cheng, Y.-H., Liu, Z.-S. & Chen, J.-Y. Adsorption and oxidation of high concentration toluene with activated carbon fibers. Journal of Porous Materials 20, 883–889 (2013).
Yu, L. et al. Adsorption of VOCs on reduced graphene oxide. Journal of Environmental Sciences 67, 171–178 (2018).
Li, L., Liu, S. & Liu, J. Surface modification of coconut shell based activated carbon for the improvement of hydrophobic VOC removal. Journal of hazardous materials 192, 683–690 (2011).
Yu, X. et al. Insight into the significant roles of microstructures and functional groups on carbonaceous surfaces for acetone adsorption. RSC advances 8, 21541–21550 (2018).
Lashaki, M. J. et al. Adsorption and desorption of a mixture of volatile organic Compounds: Impact of activated carbon porosity. Separation and Purification Technology 314, 123530 (2023).
Zhu, L., Shen, D. & Luo, K. H. A critical review on VOCs adsorption by different porous materials: Species, mechanisms and modification methods. Journal of hazardous materials 389, 122102 (2020).
Figueiredo, J. L. Functionalization of porous carbons for catalytic applications. Journal of Materials Chemistry A 1, 9351–9364 (2013).

No competing interests reported.

Download PDF

Version 1

posted

You are reading this older preprint version

Read the latest preprint version →

WITHDRAWN: Adsorption kinetics of volatile organic compounds (VOCs) onto super activated renewable carbon (SARC) in low-carbon composites (LCCs)

Status:

Version 1

Editorial Note

Abstract

Figures

1. Introduction

2. Materials and Methods

2.1. Material and reagents

2.2. Sieving and alkali modification of super activated renewable carbon (SARC)

2.3. Brunauer-Emmett-Teller (BET) analysis

2.4. Scanning electron microscopy (SEM) and energy dispersive X-ray spectroscopy (EDS)

2.5. Molecular dynamics simulations

2.6. Boehm titration

2.7. LCCs-SARC compounding

2.8. Gas chromatography-mass spectroscopy (GC/MS) analysis

2.9. Langmuir monolayer adsorption isotherm

2.10. Freundlich multilayer isotherm

3. Results and discussion

3.1. Properties of different sized SARC

3.2. Molecular dynamic simulation

3.3. Calculation of adsorption energy

3.4. Identification and quantification of VOCs by GC/MS

3.6. Influence of physical and chemical properties on VOCs removal capacity

4. Conclusions

Declarations

References

Additional Declarations

Status:

Version 1