Bandosz2015-Spent Coffee-Based Activated Carbons
Bandosz2015-Spent Coffee-Based Activated Carbons
Bandosz2015-Spent Coffee-Based Activated Carbons
34
Spent Coffee-Based Activated Carbons:
Production, Properties, and Environmental
Applications
Teresa J. Bandosz, Karifala Kante
Department of Chemistry, The City College of New York, New York, NY, USA
SBET, specific surface area; Vmic, volume of micropores; Vt, total pore volume; NA, not available.
In the work by Kante and co-workers, the ratios of on the surface. Their chemistry depends on the carbon-
spent coffee to ZnCl2 were 0.5, 1, and 2.16 The mixtures ization temperature; after treatment at 800 °C, pyridinic
were homogenized and carbonized at 800 °C with a 1-h nitrogen, pyridine-N-oxides, and quaternary nitrogen
hold time. A marked increase in the surface area and should exist on the surface.19 These species are known to
porosity was found. While the carbon with the ZnCl2/ contribute to carbon basicity,6,20 decrease the band gap,21
spent coffee ratio 0.5 had a surface area 645 m2/g and the and also affect the electronic properties of the surface by
total pore volume 0.328 cm3/g, the sample obtained with changing the charge on the neighboring carbon atoms.10
highest activation ratio (2) had a surface of 1121 m2/g A more complex chemical activation of spent coffee
and pore volume of 0.954 cm3/g (Table 34.1). Interest- was proposed by Namane and co-workers.17 In their
ingly, acid washing had a mixed effect on porosity. Thus, approach, the mixture of ZnCl2 and H3PO4 (equal molar
the surface area and pore volume of the former sample concentration of 1 mol/L) was used as an activation agent.
slightly increased (about 10%); for the latter, a 17% and The spent coffee was saturated with that solution, dried,
30% decrease in the surface area and pore volume was and then carbonized at 600 °C for 45 min. The resulting
found. These changes were attributed to the removal of carbon was washed with diluted HCl to remove the solu-
residual zinc salt. Apparently, its deposition in the pore ble salt and the excess of the dehydrating agent. The final
system of the sample with high activation ratio (∼2 wt% product had a surface area of 640 m2/g and pore volume
zinc) contributed to the formation of small pores. The of 0.95 cm3/g (Table 34.1). The surface pH was 6.33.
surface pH values of the unwashed samples were Boonamanuayvitaya and coworkers subjected cof-
between 6 and 7. Acid washing resulted in an increase in fee residue to a range of activation methods, including
the acidity and the pH decrease was 1 to 2 units. The car- chemical activation with ZnCl2, and CO2 and steam acti-
bon had a significant content of oxygen (up to 24 wt%, vation of residue derived chars.22 In the case of the chemi-
which increased with an increase in the activation ratio) cal activation, the ZnCl2-to-precursor ratio was 3:1. In all
and 2–3 wt% nitrogen. The content of the latter decreased procedures, the final temperature was 600 °C with a 4-h
with an increase in the activation ratio. The presence of holding time. Besides the physical activation of chars, the
nitrogen was considered to be an important asset of the chemically activated carbons were further subjected to
coffee-derived carbon. This nitrogen has its origin in caf- a high temperature activation with CO2 and steam. The
feine. Using a nitrogen-rich precursor is expected to lead most porous carbon was chemically activated with ZnCl2
to the uniform distribution of nitrogen functionalities and then reactivated with CO2. Its surface area reached
900 m2/g with a total pore volume of 1 cm3/g (Table 34.1). TABLE 34.2 H2S Breakthrough Capacities, Surface pH Values
Other activation methods led to materials with surface and Ash Contents
areas ranging from 10 to 470 m2/g. The adsorbent with the Ash content H2S Brth H2S Brth
highest surface area was obtained with chemical activation Sample (%) (mg/g) (mg/cm3) pHin pHexh
in at least one step and it contained from 5% to 11% ash.
COFAC-0.5 3.78 81.3 15.3 6.81 3.55
A second activation step doubled the amount of ash as a
result of burn-off. Fourier transform infrared spectroscopy COFAC-1 2.29 127.0 22 6.27 2.65
showed strong bands from various oxygen groups, sug- COFAC-2 4.98 18.3 0.53 6.54
gesting a relatively high degree of surface hydrophilicity.
COFAC-0.5A 1.02 26 5.97 5.20 2.81
pH of the surface for all carbons was over 10, the amount are arranged on the surface. Moreover, the electron-
of basic groups on the surface decreased with an increase acceptor properties of weakly acidic groups were indi-
in the activation ratio. It was accompanied by an increase in cated as attractive forces directing the phenol molecules
the porosity. The adsorption of salicylic acid was tested at to the most favorable positions on the carbon surface. The
room temperature. One gram of carbon was used and the adsorption efficiency for hydroquinol, phenol, catechol,
concentration of salicylic acid was 30 mg/L (pH 3). The car- and resorcinol were 80%, 80%, 58%, and 48% respec-
bon obtained with a low activation ratio was found to be tively. The performance of spent-coffee derived carbon
the best adsorbent (97% removal). It had the highest ratio of (640 m2/g) was compared to that of commercial unspeci-
basic to acidic groups, as determined by Boehm titration,19 fied activated carbon (950 m2/g) (Figure 34.3). Fitting the
and had the highest degree of microporosity. The efficiency experimental data to the Langmuir model indicated that
of the salicylic acid removal on the other three carbons was the former sample adsorbed 3.22 mg of phenol/g and the
only slightly lower (93.5–95.4), suggesting that the devel- latter adsorbed 3.70 mg/g. These results suggested com-
opment of porosity compensates for the decrease in the petitiveness of the spent coffee-based activated carbons
number of basic groups, which are apparently adsorption with the commercial products.
centers for salicylic acid via specific acid-base interactions. Namane and coworkers also evaluated the suitabil-
On the same series of carbons as those used for salicylic ity of their carbon to remove acidic and basic dyes, acid
acid adsorption, the adsorptive removal of phenols was blue and basic yellow.17 The conditions of the adsorption
evaluated.27 As in the case of salicylic acid, the carbon experiments were the same as those for phenols addressed
obtained with the activation ratio of 0.25 was found to be above. While on the commercial carbon, 4 mg/g of basic
the best adsorbent, on which 68% phenol was removed yellow were adsorbed; on the spent coffee-derived car-
from its 30 mg/L solution (1 g of carbon, 25 °C, pH 3). The bon, the capacity was 2.5 times higher (10 mg/g). The
range of phenol concentrations used to measure adsorp- opposite trend was found for acid blue: on the latter car-
tion isotherms was from 10 to 170 mg/L. The pH values bon, 3.57 mg/g was adsorbed, whereas 8.33 mg/g was
of the initial solution were adjusted by adding either 1 M adsorbed on the former. The good performance of the
NaOH or 0.5 M H2SO4. At equilibrium conditions, the coffee-derived carbon for the removal of basic yellow
carbon adsorbed approximately 85 mg of phenol/g from was linked to the weakly acidic surface of carbon and the
a phenol concentration equal to 170 mg/L. From a con- smaller size of this molecule than that of acid blue.
centration of 90 mg/L, 49 mg of phenol/g was adsorbed.
The peculiar plateau on the isotherms was explained by
the fact that phenol molecules replaced water molecules 34.5 APPLICATIONS OF COFFEE-
adsorbed on the carbon surface at a high phenol concen- DERIVED CARBONS AS
tration. The thermodynamic parameters of adsorption SUPERCAPACITORS
were also evaluated; a heat of adsorption lower than
40 kJ/mol indicated the physical adsorption processes. Rufford and coworkers investigated the application of
Adsorption of hydroquinol, phenol, catechol, and spent coffee-derived carbon as supercapacitors in acidic
resorcinol on spent coffee-derived carbon was also electrolytes.15 The electrochemical behavior of their mate-
investigated by Namane and co-workers.17 The initial con- rials was tested with scan rates between 5 and 100 mV/s.
centration of phenols was 20 mg/L. It was found that the The charge/discharge profiles were measured with current
quantity adsorbed depends on the position of hydroxyl densities between 0.5 A/g and 5 A/g. The carbon obtained
group in the phenol molecule. Because a substitution in using ZnCl2:spent coffee with an activation ratio of 1:1
the para position, as that in hydroquinol, results in a pla- exhibited an energy density up 20 Wh/Kg in 1 M H2SO4
nar structure and a small cross-section area, the surface and a specific capacitance measured in a two-electrode
of carbon can be more efficiently used when molecules cell reached 368 F/g. The cyclic voltammetry curves