Biomass fly ash geopolymer monoliths for effective methylene blue removal from wastewaters

Biomass fly ash geopolymer monoliths for effective methylene blue removal from wastewaters

Accepted Manuscript Biomass fly ash geopolymer monoliths for effective methylene blue removal from wastewaters Rui M. Novais, Guilherme Ascensão, Davi...

9MB Sizes 23 Downloads 91 Views

Accepted Manuscript Biomass fly ash geopolymer monoliths for effective methylene blue removal from wastewaters Rui M. Novais, Guilherme Ascensão, David M. Tobaldi, Maria P. Seabra, João A. Labrincha PII:

S0959-6526(17)32371-5

DOI:

10.1016/j.jclepro.2017.10.078

Reference:

JCLP 10870

To appear in:

Journal of Cleaner Production

Received Date: 8 June 2017 Revised Date:

6 October 2017

Accepted Date: 8 October 2017

Please cite this article as: Novais RM, Ascensão G, Tobaldi DM, Seabra MP, Labrincha JoãA, Biomass fly ash geopolymer monoliths for effective methylene blue removal from wastewaters, Journal of Cleaner Production (2017), doi: 10.1016/j.jclepro.2017.10.078. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT 1

Biomass fly ash geopolymer monoliths for effective methylene blue

2

removal from wastewaters

3 Rui M. Novais a,*, Guilherme Ascensão a, David M. Tobaldi a, Maria P. Seabra a, João

5

A. Labrincha a

6

a

RI PT

4

Department of Materials and Ceramic Engineering / CICECO- Aveiro Institute of

Materials, University of Aveiro, Campus Universitário de Santiago, 3810-193 Aveiro,

8

Portugal

M AN U

9

SC

7

10

*Corresponding author: Tel.: +351234370262; fax: +351234370204

11

E-mail address: [email protected] (Rui M. Novais)

AC C

EP

TE D

12

1

ACCEPTED MANUSCRIPT Abstract

2

For the first time biomass fly ash geopolymer monoliths were used as adsorbents for the

3

removal of methylene blue from synthetic wastewaters. Highly porous and lightweight

4

fly ash-based geopolymers were produced and then evaluated as methylene blue

5

adsorbents. The geopolymers’ porosity strongly affects the dye extraction, a threefold

6

increase (from 5.4 to 15.4 mg/g) being observed when the porosity rises from 40.7 to

7

80.6%. The maximum uptake reported here (15.4 mg/g) surpasses several other

8

powdered adsorbents, which demonstrates the interesting potential of this innovative

9

adsorbent. Moreover, these monolithic adsorbents can be used directly in packed beds

SC

RI PT

1

as membranes, this being a major advantage over powdered adsorbents.

11

Furthermore, these adsorbents were successfully regenerated and reused (up to five

12

cycles) without compromising the performances. In fact, enhanced methylene blue

13

uptake (up to 20.5 mg/g) was observed after regeneration. Additionally, an unexplored

14

waste stream was used as raw material which mitigates the waste environmental

15

footprint contributing towards a circular economy.

TE D

M AN U

10

16

18

Keywords: inorganic polymer; porosity; adsorption; waste; dye.

EP

17

1. Introduction

20

Water scarcity is one of the most pressing concerns of our society, as recently

21

highlighted by the World Economic Forum report (World Economic Forum, 2016). By

22

2050, over 40% of the world population is estimated to live in areas with extreme water

23

paucity (OECD, 2012). This phenomenon cannot be dissociated from the overwhelming

24

growth of world population and the unsustainable industrial developing, which have

25

increased the water demand and the water contamination to unprecedented levels. The

AC C

19

2

ACCEPTED MANUSCRIPT latter exerts pressure on the available resources to cope with demand. In this context,

2

wastewater treatment is of foremost importance as it may increase the water supply.

3

The presence of trace amounts of dyes (<1 ppm) in industrial wastewaters is extremely

4

noticeable and unwanted (Rafatullah et al., 2010). One of the most commonly used dyes

5

is methylene blue (MB), a cationic dye, which is known to cause blindness, respiratory

6

distress and abdominal disorders (Khan et al., 2015; Rafatullah et al., 2010). For these

7

reasons its removal from wastewaters is mandatory. Several methods have been

8

employed for MB removal (e.g. photodegradation, ion exchange, ultrafiltration), still

9

adsorption is considered one of the most effective, simple and low cost technique.

10

Although activated carbon presents very high adsorption capacity (Kannan and

11

Sundaram, 2001), its widespread use is restricted by the high production cost

12

(Rafatullah et al., 2010). As a result, new and low cost alternatives are pursued.

13

Geopolymers, amorphous to semi-crystalline structures (Davidovits, 1994), are formed

14

by chemical reaction between Si- and Al- rich materials with alkaline (or acidic)

15

activators. They are intrinsically porous with a negatively charged aluminosilicate

16

network balanced by cations (e.g. Na+, K+), suggesting the feasibility of being used as

17

adsorbents (Luukkonen et al., 2016). Nevertheless, the existing literature is surprisingly

18

scarce on this application. Nonetheless their use as granules (Al-Zboon et al., 2011) and

19

monoliths (Novais et al., 2016e) for heavy metal extraction and as powders for dye

20

removal (Liu et al., 2016) from wastewaters have been reported. Despite the relevance

21

of these investigations it should be highlighted that all the studies considering the

22

removal of MB from wastewaters focussed on using powders which have crucial

23

technical limitations: these powdered adsorbents cannot be directly used in packed beds

24

and their recovery after use is not easy. The use of monolithic bodies, as proposed here,

AC C

EP

TE D

M AN U

SC

RI PT

1

3

ACCEPTED MANUSCRIPT may allow their direct use in packed beds as membranes which would significantly

2

simplify the process.

3

Nowadays our society is facing a paradigm shift from linear to circular economy. Waste

4

valorisation is one of the pillars of this new concept. Geopolymer technology allows the

5

use of distinct waste streams as raw materials such as fly ash (FA) (Zhuang et al., 2016)

6

and glass waste (Novais et al., 2016b), which reduces the consumption of metakaolin

7

(MK) – benchmark raw material – besides obvious environmental advantages. Here,

8

biomass FA collected from a Portuguese paper pulp plant (where wood forest biomass

9

is used as fuel) was used to partially replace MK in the geopolymers production. The

10

volume of this type of FA, less common than coal FA, has sharply increased in the last

11

decade (Tarelho et al., 2015), which has raised concerns regarding its management

12

strategies. In fact, this solid waste is currently being mainly disposed mainly in landfills,

13

which is unsustainable in the circular economy concept. Therefore, the incorporation of

14

66.6 wt.% of this waste in the geopolymer production instead of MK, as proposed here,

15

may reduce/prevent landfilling practices.

16

In this investigation, FA-based geopolymer monoliths, exhibiting distinct porosity, were

17

developed and used as MB adsorbents. This is the first ever investigation reporting the

18

use of geopolymer bodies to remove MB from wastewaters. MK-based geopolymers

19

were also produced for comparison purpose. The influence of parameters such as

20

contact time, MB initial concentration, nature of the binder (MK- or FA-based), and

21

porosity of the bodies on the MB adsorption by the monoliths was evaluated.

22

These innovative adsorbents present a huge potential for treating industrial wastewaters

23

as they can be easily handled and, when exhausted, incorporated as filler in construction

24

materials, which are major advantages in comparison with the commonly used powder

25

adsorbents, Moreover the use of an unexplored waste stream as aluminosilicate source

AC C

EP

TE D

M AN U

SC

RI PT

1

4

ACCEPTED MANUSCRIPT 1

contributes to circular economy, besides decreasing the cost associated with

2

geopolymers production.

3 4

2. Experimental Conditions

6

RI PT

5 2.1. Materials

Commercial MK (Argical™ M1200S; Univar) and biomass fly ash waste were used as

8

aluminosilicate sources. The FA was produced in a Portuguese co-generation plant

9

(Novais et al., 2016c). The chemical composition of the raw materials is presented in

SC

7

Table S1 (as supplementary material).

11

As for the alkaline activators a mixture of sodium silicate (H2O = 62.1 wt.%;

12

SiO2/Na2O=3.15; Quimialmel) and 10 M NaOH solution (ACS reagent, 97%; Sigma

13

Aldrich) was used. Hydrogen peroxide (H2O2) was used as foaming agent.

M AN U

10

15

TE D

14 2.2. Geopolymers preparation

Two different batches of geopolymers were prepared: i) FA-based: using a mixture of

17

2/3 FA and 1/3 MK (in weight); and ii) MK-based: using MK as aluminosilicate source.

18

The second batch (MK-based) was used for comparison purposes. For each batch four

19

mixtures were prepared containing different H2O2 amount, which will induce distinct

20

porosity levels. For comparison a composition without blowing agent was also

21

prepared. Details of the mixture composition are presented in Table 1.

22

The mixing procedure was described elsewhere (Novais et al., 2016d), involving

23

homogenization of the alkaline activators; followed by mixture with the aluminosilicate

24

sources; and a third step where the blowing agent was added to the slurry and mixed for

AC C

EP

16

5

ACCEPTED MANUSCRIPT 1

120 s. After mixing the specimens were cured for 1 day at 40 ºC and 65% relative

2

humidity, and then demoulded an cured at ambient conditions until the 28th curing day.

3 4

2.3. Methylene blue adsorption tests For adsorption the monolithic bodies (cylindrical discs: d=22 mm and thickness =3 mm;

6

selected following a recent study by the authors (Novais et al., 2016e)) were immersed

7

in 200 mL of a solution containing a specified MB concentration (1–50 ppm) and

8

magnetically stirred during a predetermined period of time (30 h) at room temperature.

9

Aliquots from the solution were taken at regular time intervals, and the MB

10

concentration in the liquid was evaluated by determining the absorbance in a

11

spectrometer (Shimadzu UV-3100, JP) at a λ = 664 nm.

12

The amount of MB adsorbed by the monolithic bodies was calculated by using equation

13

1:

14

 =

15

where  is the quantity of MB uptake by the geopolymer (mg MB/g geopolymer),  is

16

the initial concentration of MB (mg/L),  is the remaining equilibrium MB

17

concentration (mg/L), is the solution volume (L) and  is the mass of geopolymer

18

(g).

19

The removal efficiency (E) of the dye was determined using equation 2:

20

E (%) =

21

Langmuir and Freundlich adsorption isotherm models were used to fit the experimental

22

results (Hajjaji et al., 2013; LeVan and Vermeulen, 1981). The Langmuir model

23

assumes homogeneous binding sites and equivalent sorption energies in the surface, and

24

that there is no interaction between the sorbed species (Kim, 2015). It is described by

25

the equation:

×

(1)

AC C

EP



TE D

(  )

M AN U

SC

RI PT

5

  

× 100

(2)

6

ACCEPTED MANUSCRIPT   

1

 =

2

In eq. (3) KL (L/mg) the affinity of the sorbate for the binding sites; qmax (mg/g) the

3

maximum adsorption capacity.

4

Once obtained the KL value, the Langmuir isotherm can be expressed by a separation

5

factor, RL, given by:

6

 =

7

If RL is between 0 and 1, then there is favourable adsorption.

8

The Freundlich isotherm model, on the other hand, has been interpreted as sorption onto

9

a heterogeneous surface, having sites with different affinity. In that model, one

10

presumes that the stronger binding sites are occupied first, and that the binding strength

11

decreases with the increasing degree of occupation (Kim, 2015). The model has the

12

form:

13

 = !"  /$

14

where KF is the Freundlich constant, and n is a parameter which represents the absence

15

of linearity of the adsorbed quantity in function of Ce. Equation (5) is usually converted

16

into alternative linear form, thus becoming:

17

%&' = %&'!" + %&' 

18

If the value of n is between 1 and 10, then there is favourable adsorption. Larger values

19

of n suggests, on the contrary, a stronger interaction between the surface of the

20

adsorbent and adsorbate; when 1/n is equal to 1, this means a linear adsorption, leading

21

to identical adsorption energies for all the sites (Febrianto et al., 2009).

(3)



RI PT

 

(4)

(5)

EP

TE D

M AN U

SC

 



(6)

AC C

$

22 23 24

7

ACCEPTED MANUSCRIPT 1

2.4. Methylene blue desorption tests Desorption tests were performed on the monolithic bodies used in the adsorption

3

experiments, but using only those that adsorbed the highest MB initial concentration,

4

i.e. 50 ppm. After the adsorption tests, specimens were immersed in 200 mL of

5

deionised water, and magnetically stirred at room temperature for 30 h. Then the

6

concentration of MB leached from the specimens was measured in the same

7

spectrometer as that used for the adsorption tests.

9

2.5. Geopolymer regeneration tests

SC

8

RI PT

2

The regeneration of the geopolymer monoliths after MB adsorption was also performed,

11

but only for the specimens that showed the highest MB absorption. Regeneration was

12

achieved by heating the specimens at 400 ºC for 2 h to promote MB thermal

13

decomposition. Afterwards the specimens were reused, and the adsorption of MB

14

(C0=50 ppm) was evaluated. This process was repeated up to five times.

16

TE D

15

M AN U

10

2.6. Materials characterisation

X-ray powder diffraction (XRPD) was used to evaluate the mineralogical compositions

18

of the precursors and the produced geopolymers using a Rigaku Geigerflex D/max-

19

Series instrument (Cu Kα radiation, 10–80 °2θ, 0.02 °2θ step-scan and 10 s per step),

20

and phase identification by PANalytical X’Pert HighScore Plus software. XRPD was

21

also used for quantifying both the crystalline and amorphous amounts in the specimens.

22

For this purpose, XRPD data for full quantitative phase analysis (FQPA) were recorded

23

using a θ/θ diffractometer (PANalytical X'Pert Pro, NL), equipped with a fast RTMS

24

detector (PIXcel 1D, PANalytical), with Cu Kα radiation (45 kV and 40 mA, 5–80 °2θ

25

range, with a virtual step scan of 0.02° and virtual time per step of 200 s). FQPA was

AC C

EP

17

8

ACCEPTED MANUSCRIPT assessed combining the Rietveld and reference intensity ratio (RIR) methods (Gualtieri,

2

2000; Gualtieri and Brignoli, 2004). More in detail: 10 wt.% α-alumina (NIST SRM

3

676a) – certified phase purity being 99.02±1.11 wt.% (Cline et al., 2011) – was added to

4

the sample, and considered as an additional phase in the refinements. In this fashion, the

5

refined weight fractions of each crystalline phase (Wic) were rescaled with regard to the

6

known weight fraction of α-Al2O3 (i.e. the added standard) (Ws), aiming at obtaining the

7

real crystalline phase weight fraction (Wi):

8

)* = + -.+ / 1 )*2 3

9

In eq. (7), Wsc is the refined weight fraction of the internal standard. Thus, knowing the

10

weight fractions of all the crystalline phases, it follows that the amorphous weight

11

fraction (Wa) is the difference between 1 and the “as-received” components (Wi):

12

)4 = 1 − ∑* )*

13

Furthermore, the errors associated with the amorphous and crystalline phase fractions

14

were determined following what reported by Madsen and Scarlett (2008).

15

The Rietveld refinements were accomplished by means of GSAS-EXPGUI software

16

suite (Larson and Von Dreele, 2004). Instrumental broadening, obtained from the

17

refinement of LaB6 standard (NIST SRM 660b), was used to avoid measurement related

18

artefacts. The Rietveld refinements were achieved according to this strategy: scale-

19

factors, zero-point, the background was fitted using the shifted Chebyshev function (15

20

to 20 coefficients), unit cell parameters were refined. The profile was modelled using

21

the Thompson-Cox-Hasting formulation of the pseudo-Voigt function (Thompson et al.,

22

1987), and two Lorentzian (LX and LY) terms, peak correction for asymmetry, as well as

23

sample displacement effects, were refined. The Gaussian parameter (GW, an angle

24

independent term) of the pseudo-Voigt profile shape function of the phases constituting

25

the samples was instead constrained to the value gained for α-Al2O3 (NIST SRM 676a).

+

(7)

SC

/0

(8)

AC C

EP

TE D

,

M AN U



RI PT

1

9

ACCEPTED MANUSCRIPT The water absorption of the porous bodies was determined by using the Archimedes’

2

principle, while the bulk density was calculated by measuring the specimens mass and

3

volume.

4

The geopolymers’ true density was evaluated by using a helium pycnometer

5

(Multipycnometer, Quantachrome): 1.87 g/cm3 for the MK-based composition and 2.01

6

g/cm3 for the FA-based one (both prepared without blowing agent). Then the porosity of

7

the specimens was determined (Novais et al., 2016c).

8

The geopolymers’ compressive strength (cylindrical discs: d=22 mm and length =48

9

mm) was measured at the 28th day using a Universal Testing Machine (Shimadzu,

10

model AG-25 TA). The tests were performed at 0.5 mm/min. Three specimens were

11

tested for each composition.

12

The geopolymers’ microstructure was evaluated using scanning electron microscopy

13

(SEM - Hitachi S4100 equipped with energy dispersion spectroscopy, EDS – Rontec),

14

while the geopolymers’ morphology was investigated using optical analysis (Leica

15

EZ4HD microscope).

16

The temperature evolution of the geopolymer slurries upon curing in the first hours of

17

reaction was measured by using a quasi-adiabatic calorimeter (Novais et al., 2016a),

18

which was placed inside a climatic chamber (at 40 ºC). To provide a better insight an

19

infrared camera (Flir E4) was also used for monitoring the temperature evolution.

20

The chemical composition of the raw materials (FA and MK) was determined by X-ray

21

fluorescence (Philips X´Pert PRO MPD spectrometer).

22

Particle size distribution was determined by laser diffraction (Coulter LS230 analyzer).

23

Fourier-transform infrared spectroscopy (FTIR) measurements of the adsorbent, before

24

and after MB adsorption, were carried out with a Bruker Tensor 27 spectrometer, in

AC C

EP

TE D

M AN U

SC

RI PT

1

10

ACCEPTED MANUSCRIPT 1

attenuated total reflectance (ATR) mode in the wavenumber range of 4000–350 cm–1 –

2

256 scans and 4 cm–1 in resolution.

3

Zeta potentials were determined by Zetasizer Nano ZS (Malvern) at ambient

4

temperature.

6 7

RI PT

5 3. Results and discussion 3.1. Slurries calorimetric characterisation

The temperature evolution in the first 10 h of reaction for the MK- and FA-based

9

slurries is shown in Fig. 1, while the maximum temperature (Tmax) and the time to peak

10

(time required to reach the Tmax) are presented in Table S2 (provided as Supplementary

11

Material). Significant differences are observed between the compositions, showing that

12

both the aluminosilicate source and the water amount in the compositions affect the

13

geopolymerization rate. For the MK-based slurry (prepared without H2O2) the Tmax was

14

64.2 ºC, and it was reached after 76 min. As for the FA-based slurry, the Tmax was

15

significantly reduced (around 11.4 ºC), while the time to peak increased by

16

approximately 34 min. These data indicates that a weaker and slower geopolymerisation

17

is reached when using FA, instead of MK, as main aluminosilicate source. There are

18

two possible explanations for these results: i) the much lower amorphous content in FA

19

(shown by the XRPD patterns) corresponds to a lower number of reactive species (in

20

the dissolution step), thus reducing the polymerisation degree; and ii) their much coarser

21

particle size distribution. In fact the mean particle size of FA (63 µm) is almost 13 times

22

higher than those of MK (5 µm). The raw materials particle size distribution crucially

23

affects their reactivity (Nazari et al., 2011); larger particle size is expected to hinder the

24

dissolution

25

geopolymerisation rate and extension.

AC C

EP

TE D

M AN U

SC

8

of

the

amorphous

silica

and

alumina,

therefore

reducing

the

11

ACCEPTED MANUSCRIPT In this work the compositions were intentionally prepared using distinct water and

2

hydrogen peroxide contents (see Table 1 for details) to promote high porosity values.

3

Therefore, the comparison between compositions (e.g. MK_0.11 and MK_0.22) can

4

only provide the combined effect of water and hydrogen peroxide contents. However,

5

recently the authors have reported that the influence of the hydrogen peroxide content

6

on the slurries’ thermal behaviour upon curing was minor (Novais et al., 2016a), in

7

comparison with the water amount. In this sense, the observed decrease on the

8

maximum temperature (up to 16 ºC), and the increase on the time to peak (up to 63 min)

9

detected between the MK-based compositions, can be mainly attributed to the increase

10

on the water content, which is detrimental to the polycondensation reactions. The

11

impact of the water amount in the temperature evolution of FA-based compositions was

12

similar.

13

The thermal behaviour of geopolymeric slurries upon curing was also characterised

14

using an infrared camera. Fig. 2 presents the surface thermal map for two compositions

15

(MK_0.00 and FA_0.00) with curing time. These results support those collected with

16

the thermocouple (shown in Fig. 1).

18

SC

M AN U

TE D

EP

17

RI PT

1

3.2. Geopolymers characterisation The XRPD patterns for the FA- and MK-based geopolymer are presented in Fig. 3. The

20

patterns for the FA-based geopolymers show that the position of the diffraction peaks

21

matches those of the precursor (MK). This statement is equally valid for the MK-based

22

compositions whose XRPD pattern remained unchanged for the distinct compositions.

23

This explains why FQPA analyses were carried out only on selected specimens (i.e.

24

MK_0.11 and FA_0.11). FQPA results are listed in Table S3, whilst an example of a

25

Rietveld refinement is presented in Fig S1. MK_0.11 is composed of α-quartz (2.1

AC C

19

12

ACCEPTED MANUSCRIPT wt.%), mica (5.4 wt.%), and anatase (0.9 wt.%); the amorphous phase accounts for 91.7

2

wt.%. On the other hand, FA_0.11 was shown to be made of 69.9 wt.% amorphous

3

phase; the crystalline phases of FA_0.11 are: α-quartz (4.8 wt%), calcite (10.8 wt%),

4

microcline (11.6 wt%), mica (2.8 wt%), together with traces of lime (0.1 wt%). These

5

FQPA results agree well with, and validate the above hypothesis about a weaker and

6

slower geopolymerisation in FA, because its much lower amount in amorphous phase

7

(i.e. 69.9 vs 91.7 wt.% in MK_0.11), generating a lower number of reactive species.

8

SEM micrographs of two compositions (FA- and MK-based) prepared with 0.11 wt.%

9

H2O2 are shown in Fig. 4. The EDS spectrum was also included in the figure. An

SC

RI PT

1

10

amorphous

11

geopolymerisation occurrence. Nevertheless, the gel composition was distinct between

12

the samples reflecting the nature of the raw materials: i) mainly composed by Si, Al and

13

Na for the MK-based geopolymers; ii) while in the FA-based ones the third prevailing

14

element was Ca.

15

The geopolymers microstructure (at the 28th day) is illustrated by the optical and SEM

16

micrographs shown in Fig. 5. Significant differences in terms of number, volume and

17

size distribution of the produced pores are perceived between the samples. The

18

geopolymers microstructure is affected not only by the water and H2O2 content, but also

19

by the binder composition. As expected, an increase on the blowing agent amount

20

prompts a rise of the pore area and volume for both systems. Nevertheless, the pore size

21

distribution and the type of pores (open or closed) are remarkably distinct between the

22

compositions. In the former, a rather homogeneous pore size distribution is observed

23

when the H2O2 content is 0.11 wt.%. At this incorporation level pores are mainly round-

24

shaped, while at higher incorporation contents pores lose their roundness due to pore

25

coalescence that begins with 0.22 wt.% H2O2. For this system extensive pore

was

observed

in

all

M AN U

matrix

samples,

suggesting

AC C

EP

TE D

geopolymer

13

ACCEPTED MANUSCRIPT coalescence is reached at 0.57 wt.% H2O2. In the latter (FA-based) the pore size

2

distribution at low H2O2 content (between 0.11 and 0.22 wt.%) is non-uniform and the

3

produced pores (round-shaped) seem to be mainly closed which is in line with the low

4

water absorption of this specimen (25.5 wt.%). When the H2O2 content increases from

5

0.41 to 0.57 wt.% a remarkable change on pores connectivity is observed, opening the

6

geopolymers’ structure. In fact, the latter is supported by the twofold increase on the

7

geopolymers’ water absorption (see Table 2).

8

The influence of the H2O2 amount on the physical properties (apparent density and total

9

porosity) of the produced geopolymers is shown in Fig. 6. For the MK-based

10

geopolymers a steady reduction on apparent density and an increase on total porosity

11

are observed when the blowing agent content rises. In this system the apparent density

12

ranges between 0.93 and 0.56 g/cm3. In the FA-based specimens the same behaviour

13

(apparent density reduction and total porosity increase) is also observed, yet the

14

fluctuation of these parameters, when rising the blowing agent amount, differs. When

15

the blowing agent is below 0.22 wt.%, the apparent density values are always higher

16

than those observed for the MK-based ones, while above 0.22 wt.% H2O2 the opposite

17

occurs. Indeed, in the FA-based geopolymers a sharp fluctuation (approximately 45%)

18

in the apparent density is visible between 0.22 and 0.41 wt.% H2O2. For that reason the

19

apparent density of the composition FA_0.57 is around 30% lower than that of

20

MK_0.57, while the total porosity is approximately 10% higher in the FA-based

21

specimens. The high total porosity values (up to 80.6 %) observed for these

22

geopolymers is expected to enhance the MB adsorption, in comparison with the less

23

porous geopolymers, due to the increase on the adsorption sites for MB.

24

Fig. 7 presents the influence of the binder nature and H2O2 content on the compressive

25

strength of the produced specimens. The compressive strength for the FA-based

AC C

EP

TE D

M AN U

SC

RI PT

1

14

ACCEPTED MANUSCRIPT geopolymers is always superior in comparison with their MK-based counterparts. Still

2

differences tend to decrease as the blowing agent content rises.

3

A remarkable decrease around 44 times on the compressive strength, of the FA-based

4

samples, is observed when the H2O2 content rises from 0.11 to 0.57 wt.%. Despite the

5

low mechanical resistance of these highly porous samples (180 kPa) the specimens kept

6

their integrity after the adsorption, desorption and regeneration tests.

7 8 9

3.3. MB adsorption tests 3.3.1. Influence of contact time

SC

RI PT

1

These tests were assessed on the specimens made with the lowest and highest H2O2

11

amounts, i.e. with 0.11 and 0.57 wt.%, respectively. The adsorption rates were followed

12

by measuring the adsorbed MB dye by the geopolymers as a function of contact time; a

13

typical pattern of the removal efficiency versus contact time is shown in Fig. 8. The

14

removal efficiency increases with the sorption time, until a plateau is reached, at about

15

30 h, this being the time at the equilibrium. The time dependence of MB adsorption has

16

been previously investigated (Khan et al., 2015; Li et al., 2006; Yousef et al., 2009).

17

The equilibrium time here reported is smaller than that of FA-geopolymers (100 h) (Li

18

et al., 2006), similar to kaolin-geopolymers (30-48 h) (Yousef et al., 2009), and higher

19

than that of phosphoric acid geopolymers (90 min) (Khan et al., 2015). Nonetheless,

20

these studies were focused on geopolymer powders that cannot be directly used in

21

packed beds. On the contrary, the geopolymer monoliths can be directly assembled in

22

packed beds as membranes simplifying the wastewater treatment process.

23

EDS maps before and after MB adsorption are shown in Fig. 9. The FA used here

24

contain around 3.5 wt.% of SO3 (see their chemical composition in Table S1). Indeed,

25

the EDS map presented in Fig. 9b, corresponding to the geopolymer monolith before

AC C

EP

TE D

M AN U

10

15

ACCEPTED MANUSCRIPT 1

MB adsorption, shows the presence of small amounts of sulphur. After the adsorption

2

test, a significant increase on the sulphur content is evident (see Fig. 9d), which

3

confirms the MB adsorption by the geopolymers – MB formula being: C16H18ClN3S.

5

3.3.2. Influence of MB initial concentration

RI PT

4

Fig. 10a and 10b illustrate, respectively, the uptake capacity and the removal efficiency

7

of the distinct geopolymer monoliths as a function of the initial MB concentration (C0).

8

The MB uptake by the porous geopolymers significantly increases with the rise of C0

9

for all the specimens. For example a 38-fold increase on the MB uptake by the

10

specimen coded as FA_0.51 was observed when C0 jumped from 1 to 50 ppm. The very

11

high MB uptake (15.4 mg/g) of this specimen suggests a strong interaction between the

12

dye molecules, which are available in solution, and the geopolymers’ active sites. On

13

the contrary, the removal efficiency drops with the increase of C0 for all compositions.

14

At low C0 (i.e. ≤ 5 mg/L) a high number of active sites is available in the specimens,

15

favouring the interaction with the dye molecules, either with MK or FA. When the MB

16

concentration increases the number of available sites drops, since some of them are

17

already filled up, thus explaining the decrease on the removal efficiency.

18

The geopolymers’ porosity also affected the MB removal, more porous geopolymers

19

exhibiting higher MB adsorption this being particularly evident at high C0. In fact, when

20

the C0 was 50 mg/L a remarkable threefold increase from 5.4 to 15.4 mg/g was

21

observed when the porosity of the FA-based geopolymers rises from 40.7 to 80.6%.

22

When porosity increases, the quantity of active sites accessible for MB adsorption rises,

23

suggesting that further enhancement in adsorption can be achieved by increasing the

24

geopolymers porosity.

AC C

EP

TE D

M AN U

SC

6

16

ACCEPTED MANUSCRIPT Interestingly, the influence of porosity was less pronounced in the MK-based

2

geopolymers, with 34% increase on the MB uptake when comparing the higher porosity

3

(MK_0.57) with the lower porosity (MK_0.11) specimens using 50 mg/L MB. These

4

findings suggest that the binder nature, and not only the monoliths porosity, crucially

5

affects the MB removal. Indeed, the raw materials (MK or FA) used in geopolymers

6

preparation will affect the pH values of the water solution after geopolymers immersion

7

over time, due to distinct levels of free alkalis in the geopolymer composition (Novais et

8

al., 2017). A major fluctuation in the pH on the MB solution, from 5.9 to ~8.2, was

9

detected when the MK_0.57 bodies were immersed for 30 h in 50 mg/L MB solution.

10

As for the FA_0.57 specimen, an even higher pH fluctuation was observed, from 5.9 to

11

9.6. The pH of the MB solution affects the MB adsorption efficiency, higher values

12

favouring adsorption (Khan et al., 2015). Therefore, the higher adsorption uptake

13

exhibited by the FA_0.57, in comparison with their MK-based counterparts, may also

14

be attributed to the higher pH attained after monoliths immersion. One other

15

explanation for these results is the higher specific surface area of specimen FA_0.51

16

(19.4 m2/g) in comparison with that of MK_0.51 (7.9 m2/g) specimen.

17

The evolution of the adsorbents (MK_0.57 and FA_0.57) zeta potential with pH is

18

presented in Fig. S2. Results show that the adsorbents’ zeta potential is negative within

19

the studied pH interval (2 to 12), demonstrating that the geopolymers surface is

20

negatively charged. In the specimen MK_0.57 the zeta potential sharply decreases when

21

the pH rises from 3.6 to 5.6, while at higher pH values only minor fluctuations are

22

observed. As for the FA_0.57 a gentler decrease in the zeta potential was observed, still

23

this parameter decreases continuously up to pH=11. These results show that the

24

attraction between the geopolymers framework (negatively charged) and the cationic

25

dye is pH dependent, higher pH values required by the FA-based adsorbents to

AC C

EP

TE D

M AN U

SC

RI PT

1

17

ACCEPTED MANUSCRIPT maximize the attraction for the MB dye. These results suggest that the MB adsorption

2

mechanism is due to electrostatic interactions.

3

Indeed, FTIR spectra before and after MB adsorption for the specimen FA_0.51

4

(presented in Fig. S3) are rather similar. The strong asymmetric band, having its

5

maximum at around 3450 cm–1, shows the presence of adsorbed molecular H2O, that

6

also generated the feature near 1630 cm–1, due to H–O–H bending; the band located at

7

approximately 980 cm–1 belongs to the asymmetric Al–O–Al/Si–O–Si stretching

8

(Tobaldi et al., 2010). The band at ~1440 cm–1 is related to sodium carbonate, due to the

9

reaction of residual sodium content with atmospheric CO2 (Cheng-Yong et al., 2017).

10

Aside from these bands, the FTIR spectrum of that specimen after 30 h MB adsorption

11

shows a band centred at around 1610 cm–1, assigned to the stretching vibration of

12

aromatic rings (Coates, 2006) and methyl and methylene bending vibration (1355 and

13

1335 cm–1, and ~1410 cm–1, respectively) (Coates, 2006) providing evidence of

14

methylene blue adsorption.

15

To better understand the adsorption by the porous geopolymers, optical micrographs of

16

the monoliths’ surface and fracture surface (transversal direction) are provided in Fig.

17

11. Fig. 11 shows a homogenous MB adsorption on the monoliths’ surface for all

18

compositions. Nevertheless, in the specimen FA_0.11 several bright and black spots are

19

observed, which are most probably unreacted raw material particles that do not

20

contribute to adsorption (see Fig. 11c). The latter contributes to the poorer adsorption of

21

this specimen, in comparison with the other compositions. However, another reason is

22

the distinct MB diffusion into the geopolymers. As shown by the inset micrograph in

23

Fig. 11c MB does not fully penetrate this specimen (FA_0.11), hindering the maximum

24

adsorption level. This can be associated with the lower amount of open pores, in line

25

with the optical and SEM micrographs shown in Fig. 5. On the contrary, the MB was

AC C

EP

TE D

M AN U

SC

RI PT

1

18

ACCEPTED MANUSCRIPT homogeneously distributed throughout the higher porous FA specimen, whose pores

2

were found to be mainly open (see Fig. 5). Results demonstrates that both volume and

3

pores connectivity are crucial to ensure high adsorption values. Monoliths with high

4

open porosity can be produced by controlling both the blowing agent and the free water

5

content (which induces capillary pores after evaporation).

6

Moreover the monoliths’ size is also expected to alter the adsorption time and the MB

7

uptake, a decrease on the former and an increase on the latter being expected when

8

rising the monoliths’ surface area.

9

The MB diffusion into the MK-based samples was somewhat surprising. Indeed, in the

10

lower porous specimens, MB was detected throughout the entire sample, being less well

11

distributed in the higher porous sample (absent from its centre – see inset micrograph in

12

Fig. 11b). This unexpected result explains the lower MB adsorption (see Fig. 10)

13

observed for the MK_0.57 in comparison with that of MK_0.11 when using 35 mg/L

14

MB.

15

Table 3 compares the MB adsorption level here achieved with other literature studies.

16

The maximum MB uptake here achieved (15.4 mg/g) is higher than that of several

17

powdered geopolymer adsorbents and of fly ashes (treated and non-treated) (see Table

18

3), being slightly superior to Cu2O geopolymers (Falah et al., 2015a), and smaller than

19

other adsorbents, such as zeolites (Rida et al., 2013) or geopolymers (Liu et al., 2016).

20

Nevertheless the specific surface area of the monoliths (19.4 m2/g) is significantly

21

smaller than that reported by Liu et al. (2016), 67.6 m2/g, using powdered (d<150 µm)

22

coal FA geopolymer. These data demonstrate the huge potential of these innovative

23

monolithic adsorbents, also because the monoliths may be employed directly in packed

24

beds (Novais et al., 2016e).

AC C

EP

TE D

M AN U

SC

RI PT

1

25

19

ACCEPTED MANUSCRIPT 1

3.3.3. Isotherm model studies With the aim of getting insights into the adsorption mechanism, the data were fitted

3

using both the Langmuir and Freundlich models. Langmuir and Freundlich parameters

4

for MB adsorption on the studied samples are listed in Table 4; an example of the fitting

5

using both the models is reported in Fig. S4. Langmuir isotherm model assures a good

6

fitting, but the accuracy is lower than obtained by using the Freundlich model (R2 values

7

are higher in this case). The only exception to this trend was observed with specimen

8

MK_0.57, in which the Langmuir model promoted higher fitting. Anyway, considering

9

the separation factor RL, values among 0 and 1 were estimated for all samples,

10

independently of the initial MB concentration used. This means that adsorption is

11

favourable in all cases (see Table S4, of the supplementary information file). The same

12

conclusion is obtained by considering the Freundlich model: n > 1 in all cases (see

13

Table 4). There is strong probability of having multilayer adsorption of MB molecules

14

on the active surface sites (1/n values range between 0 and 1), due to a percolation

15

mechanism. The latter is in line with previous investigations (Li et al., 2006),

16

suggesting that heterogeneous adsorption sites are present in these geopolymers

17

adsorbents.

19

SC

M AN U

TE D

EP

3.4. MB desorption tests

AC C

18

RI PT

2

20

The dye fixation efficiency was evaluated in the specimens showing the highest MB

21

uptake (see Fig. 10). The percentages of MB leached from the monoliths were 0.25,

22

1.66, 1.65 and 22.80% respectively for MK_0.11, MK_0.57, FA_0.11 and FA_0.57.

23

Results show that the MB fixation ability of the specimens is affected by the

24

geopolymers’ porosity and binder nature (MK- or FA-based). The higher porosity

25

geopolymers (MK_0.57 and FA_0.57) showed quite lower MB retention in comparison

20

ACCEPTED MANUSCRIPT with their lower porous counterparts. Moreover the MK-based geopolymers showed

2

higher affinity to the cationic MB dye. These differences cannot be solely explained by

3

the porosity levels, and are most probably due to the highest retention ability exhibited

4

the MK-based geopolymers. Indeed, specimen MK_0.11 leached only 0.25% of the

5

adsorbed dye after 30 h, while specimen FA_0.11, which has lower porosity (see Table

6

2), leached 1.65%. This suggests a stronger fixation of the MB dye in the MK-based

7

geopolymers, which hinders its leaching in water medium. Anyway the high leaching

8

observed for the composition FA_0.57, suggest that the geopolymers regeneration

9

under stronger conditions (e.g. using acidic medium) is feasible.

SC

RI PT

1

3.5

12

Fig. 12 presents the MB uptake and removal efficiency for specimen FA_0.57 after

13

regeneration (up to five cycles). After the first regeneration a significant increase

14

(~33%) in the MB uptake was observed (from 15.4 to 20.5 mg/g), which was associated

15

with an increase in porosity caused by the thermal treatment. Indeed the removal of

16

physically adsorbed water up to 200 ºC, observed for FA-geopolymers by using

17

thermogravimetric analysis (Novais et al., 2016c), is expected to enhance the

18

geopolymers’ porosity and simultaneously the quantity of active sites accessible for MB

19

adsorption. Indeed the thermal treatment induces a weight loss in the specimens in line

20

with the removal of water from the samples. The exceptional performance of the

21

monolith adsorbent was preserved after the second cycle, while the third induced a

22

slight reduction on the MB uptake. Nevertheless the uptake after the 3rd cycle (16.6

23

mg/g) is still superior to that observed without regeneration. At this stage the

24

performance of the adsorbent stabilized and only minor changes were observed on the

25

MB uptake and removal efficiency up to the 5th cycle, clearly demonstrating the

26

possibility of reusing this innovative adsorbent without any performance compromise.

M AN U

10 11

AC C

EP

TE D

Geopolymer regeneration tests

21

ACCEPTED MANUSCRIPT 1

The observed increase in the MB uptake after the 1st regeneration cycle suggests that a

2

preliminary thermal treatment to the monoliths (before the adsorption tests) could

3

enhance their removal efficiency. This topic will be addressed in future work.

4 4. Conclusions

6

In this investigation for the first time extremely light and porous geopolymer monoliths

7

were used to extract MB from wastewaters. The MB uptake is affected by parameters

8

such as the monoliths’ porosity, nature of the binder (MK- or FA-based), contact time

9

and MB initial concentration. The combination of higher porosity levels with FA-

SC

RI PT

5

binders leading to higher MB extraction.

11

The maximum MB uptake here reported (15.4 mg/g), surpassing several other powdered

12

adsorbents (e.g. coal fly ash, acid treated fly ash, FA-geopolymer, Cu2O geopolymer)

13

demonstrates the remarkable potential of these innovative materials. Moreover an

14

increase on the geopolymers’ porosity (e.g. rising the blowing agent content) may

15

further enhance their MB uptake ability. These innovative materials may be directly

16

used in packed beds, while powdered adsorbents cannot, which enhances the simplicity

17

of wastewater treatment systems.

18

The extraction/desorption of MB from the porous bodies in water was affected by the

19

monoliths porosity, higher leaching observed for the higher porosity geopolymers. The

20

MB leaching in the MK-based geopolymers was always smaller than their FA-based

21

counterparts, suggesting a stronger MB fixation by the former.

22

Results also show the feasibility of reusing the FA-based monoliths (up to 5 cycles)

23

after thermal treatment without any performance compromise. In fact a 33% increase in

24

the MB uptake (from 15.4 to 20.5 mg/g) was observed after the first regeneration cycle.

AC C

EP

TE D

M AN U

10

22

ACCEPTED MANUSCRIPT The latter is of crucial importance, as it may allow the use of this innovative material for

2

several cycles before exhaustion.

3

Additionally, the FA-based geopolymers, containing 66.6 wt.% of FA instead of MK,

4

allow the reuse of an unexplored waste stream (biomass fly ash) in line with the circular

5

economy concept.

RI PT

1

6

Acknowledgements: This work was developed within the scope of the project

8

CICECO-Aveiro Institute of Materials, POCI-01-0145-FEDER-007679 (FCT Ref. UID

9

/CTM /50011/2013), financed by national funds through the FCT/MEC and when appropriate co-financed by FEDER under the PT2020 Partnership Agreement.

M AN U

10

SC

7

11 References

13

Al-Zboon K., Al-Harahsheh M.S., Hani F.B., 2011. Fly ash-based geopolymer for Pb

14

removal from aqueous solution. Journal of Hazardous Materials 188, 414-421.

15

Cheng-Yong H., Yun-Ming L., Abdullah M.M.Al B., Hussin K, 2017. Thermal

16

resistance variations of fly ash geopolymers: foaming responses. Scientific Reports 7,

17

45355.

18

Cline J.P., Von Dreele R.B., Winburn R., Stephens P.W., Filliben J.J., 2011. Addressing

19

the amorphous content issue in quantitative phase analysis: the certification of NIST

20

standard reference material 676a. Acta Crystallographica A67, 357-367.

21

Coates J., 2006. Interpretation of Infrared Spectra, A Practical Approach, in: R.A.

22

Meyers (Ed.). Encyclopedia of Analytical Chemistry. John Wiley & Sons Ltd,

23

Chichester, 2000, pp. 10815-10837.

AC C

EP

TE D

12

23

ACCEPTED MANUSCRIPT Davidovits J., 1994. Properties of geopolymer cements. Proceedings First International

2

Conference on Alkali Cements and Concretes, SRIBM, Ukraine, Vol. 1, 131-149.

3

Falah M., Mackenzie K.J.D., Hanna J.V., Page S.J., 2015a. Novel photoactive inorganic

4

polymer composites of inorganic polymers with copper (I) oxide nanoparticles. Journal

5

of Materials Science 50, 7374-7383.

6

Falah M., Mackenzie K.J.D., 2015b. Synthesis and properties of novel photoactive

7

composites of P25 titanium dioxide and copper (I) oxide inorganic polymers. Ceramics

8

International 41, 13702-13708.

9

Falah M., Mackenzie K.J.D., Knibbe R., Page S.J., Hanna J.V., 2016. New composites

10

of nanoparticle Cu (I) oxide and titania in a novel inorganic polymer (geopolymer)

11

matrix for destruction of dyes and hazardous organic pollutants. Journal of Hazardous

12

Materials 318, 772-782.

13

Febrianto J., Kosasih A.N., Sunarso J., Ju Y.-H., Indraswati N., Ismadji S., 2009.

14

Equilibrium and kinetic studies in adsorption of heavy metals using biosorbent: a

15

summary of recent studies. Journal of Hazardous Materials 162, 616-645.

16

Gualtieri A.F., 2000. Accuracy of XRPD using combined Rietveld-RIR method. Journal

17

of Applied Crystallography 33, 267-278.

18

Gualtieri A.F., Brignoli G., 2004. Rapid and accurate quantitative phase analyses using

19

a fast detector. Journal of Applied Crystallography 37, 8-13.

20

Hajjaji W., Ganiyu S.O., Tobaldi D.M., Andrejkovičová S., Pullar R.C., Rocha F.,

21

Labrincha J.A., 2013. Natural Portuguese clayey materials and derived TiO2-containing

22

composites used for decolouring methylene blue (MB) and orange II (OII) solutions.

23

Applied Clay Science 83-84, 91-98.

AC C

EP

TE D

M AN U

SC

RI PT

1

24

ACCEPTED MANUSCRIPT Kannan N., Sundaram M.M., 2001. Kinetics and mechanism of removal of methylene

2

blue by adsorption on various carbons – a comparative study. Dyes and Pigments 51,

3

25-40.

4

Khan M.I., Min T.K., Azizli K., Sufian S., Man Z., Ullah H., 2015. Effective removal of

5

methylene blue from water using phosphoric acid based geopolymers. RSC Advances 5,

6

61410-61420.

7

Kim S.-K., 2015. Springer handbook of marine biotechnology. Springer-Verlag Berlin

8

Heidelberg, ISBN: 978-3-642-53971-8.

9

Larson A.C., Von Dreele R.B., 2004. General Structure Analysis System (GSAS). Los

M AN U

SC

RI PT

1

Alamos National Laboratory Report LAUR 86-748.

11

LeVan M.D., Vermeulen T., 1981. Binary Langmuir and Freundlich isotherms for ideal

12

adsorbed solutions. The Journal of Physical Chemistry 85, 3247-3250.

13

Li L., Wang S., Zhu Z., 2006. Geopolymeric adsorbents from fly ash for dye removal

14

from aqueous solutions. Journal of Colloid and Interface Science 300, 52-59.

15

Liu Y., Yan C., Zhang Z., Gong Y., Wang H., Qiu X., 2016. A facile method for

16

preparation of floatable and permeable fly ash-based geopolymer block. Materials

17

Letters 185, 370-373.

18

Luukkonen T., Runtti H., Niskanen M., Tolonen E.-T., Sarkkinen M., Kemppainen K.,

19

Rämö J., Lassi U., 2016. Simultaneous removal of Ni(II), As(III), and Sb(III) from

20

spiked mine effluent with metakaolin and blast-furnace-slag geopolymers. Journal of

21

Environmental Management 166, 579-588.

AC C

EP

TE D

10

25

ACCEPTED MANUSCRIPT Madsen I.C., Scarlett N.V.Y., 2008. Quantitative phase analysis, in: R.E. Dinnebier,

2

S.J.L. Billinge (Eds.). Powder Diffraction: Theory and Practice. Royal Society of

3

Chemistry Publishing, pp. 208-331.

4

Nazari A., Bagheri A., Riahi S., 2011. Properties of geopolymer with seeded fly ash and

5

rice husk bark ash. Materials Science and Engineering A 528, 7395-7401.

6

Novais R.M., Ascensão G., Buruberri L.H., Senf L., Labrincha J.A., 2016a. Influence of

7

blowing agent on the fresh- and hardened-state properties of lightweight geopolymers.

8

Materials and Design 108, 551-559.

9

Novais R.M., Ascensão G., Seabra M.P., Labrincha J.A., 2016b. Waste glass from end-

10

of-life fluorescent lamps as raw materials in geopolymers. Waste Management 52, 245-

11

255.

12

Novais R.M., Buruberri L.H., Seabra M.P., Bajare D., Labrincha J.A., 2016d. Novel

13

porous fly ash-containing geopolymers for pH buffering applications. Journal of

14

Cleaner Production 124, 345-404.

15

Novais R.M., Buruberri L.H., Seabra M.P., Labrincha J.A., 2016e. Novel porous fly

16

ash-containing geopolymer monoliths for lead adsorption from wastewaters. Journal of

17

Hazardous Materials 318, 631-640.

18

Novais R.M., Seabra M.P., Labrincha J.A., 2017. Geopolymer spheres as novel pH

19

buffering materials. Journal of Cleaner Production 143, 1114-1122.

20

OECD (Organisation for Economic Co-operation and Development), 2012. OECD

21

Environmental Outlook to 2050: the consequences of inaction, Paris, OECD Publishing,

22

http://dx.doi.org/10.1787/9789264122246-en.

AC C

EP

TE D

M AN U

SC

RI PT

1

26

ACCEPTED MANUSCRIPT Rafatullah M., Sulaiman O., Hashim R., Ahmad A., 2010. Adsorption of methylene

2

blue on low-cost adsorbents: a review. Journal of Hazardous Materials 177, 70-80.

3

Rida K., Bouraoui S., Hadnine S., 2013. Adsorption of methylene blue from aqueous

4

solution by kaolin and zeolite. Applied Clay Science 83-84, 99-105.

5

Tarelho L.A.C., Teixeira E.R., Silva D.F.R., Modolo R.C.E., Labrincha J.A., Rocha F.,

6

2015. Characteristics of distinct ash flows in a biomass thermal power plant with

7

bubbling fluidised bed combustor. Energy 90, 387-402.

8

Thompson P., Cox D.E., Hastings J.B., 1987. Rietveld refinement of Debye-Scherrer

9

synchrotron X-ray data from Al2O3. Journal of Applied Crystallography 20, 79-83.

M AN U

SC

RI PT

1

Tobaldi D.M., Tucci A., Škapin A.S., Esposito L., 2010. Effects of SiO2 addition on

11

TiO2 crystal structure and photocatalytic activity. Journal of the European Ceramic

12

Society 30, 2481-2490.

13

Wang S., Boyjoo Y., Choueib A.A., 2015. A comparative study of dye removal using

14

fly ash treated by different methods. Chemosphere 60, 1401-1407.

15

World Economic Forum, 2016. The global risks report 2016, 11 Edition, Geneva,

16

Switzerland: World Economic Forum.

17

Yousef R.I., El-Eswed B., Alshaaer M., Khalili F., Khoury H., 2009. The influence of

18

using Jordanian natural zeolite on the adsorption, physical, and mechanical properties of

19

geopolymers products. Journal of Hazardous Materials 165, 379-387.

20

Zhang Y., Liu L., 2013. Fly ash-based geopolymer as a novel photocatalyst for

21

degradation of dye from wastewater. Particuology 11, 353-358.

AC C

EP

TE D

10

27

ACCEPTED MANUSCRIPT 1

Zhuang X.Y., Chen L., Komarneni S., Zhou C.H., Tong D.S., Yang H.M., Yu W.H.,

2

Wang W. H., 2016. Fly ash-based geopolymer: clean production, properties and

3

applications. Journal of Cleaner Production 125, 253-267.

RI PT

4 5 6 7

SC

8

M AN U

9 10 11 12

16

EP

15

AC C

14

TE D

13

28

SC

RI PT

ACCEPTED MANUSCRIPT

M AN U

1

Fig. 1 Influence of binder composition and hydrogen peroxide content on the

3

temperature evolution of the produced geopolymers.

AC C

EP

TE D

2

29

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

1 2

Fig. 2 Thermal map of MK-based and FA-based geopolymer slurries (prepared without

3

H2O2) upon curing (in a quasi-adiabatic calorimeter). For interpretation of the references

4

to colour in this figure legend, the reader is referred to the web version of this article. 30

1

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

2

Fig. 3 XRD patterns of a) FA-based and b) MK-based geopolymers produced with

3

distinct H2O2 content.

31

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

1 2

Fig. 4 SEM micrographs and EDS spectrum of a) MK-based and b) FA-based

3

geopolymers produced with 0.11 wt.% H2O2.

AC C

EP

TE D

4

32

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

1

Fig. 5 Influence of the blowing agent content in the geopolymers microstructure: optical

3

(a and c) and SEM (b and d) micrographs.

EP

5

AC C

4

TE D

2

33

1

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

2

Fig. 6 Apparent density and total porosity of MK-based and FA-based geopolymers

3

produced with distinct H2O2 content.

7

EP

6

AC C

5

TE D

4

34

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

1

Fig. 7 Compressive strength of the MK-based and FA-based geopolymers (cured for 28

3

days). The insert graph illustrates the compressive strength for the higher porosity

4

geopolymers.

7

EP

6

AC C

5

TE D

2

35

1

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

2

Fig. 8 Effect of the initial MB concentration (C0) on the time dependent removal

3

efficiency (%), specimen MK_0.11.

EP AC C

5

TE D

4

36

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

1

Fig. 9 EDS maps of specimen FA_0.57 before (a-b) and after (c-d) MB adsorption

3

(contact time 30 h, Co = 35 ppm) revealing changes in the sulphur content. For

4

interpretation of the references to colour in this figure legend, the reader is referred to

5

the web version of this article.

AC C

EP

TE D

2

37

1

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

2

Fig. 10 Effect of the initial concentration of MB on the removal efficiency and uptake

3

of this pollutant by the distinct geopolymer monoliths (contact time: 30 h).

38

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

1

Fig. 11 Optical microscopy micrographs of the surface of a) MK_0.11, b) MK_0.57, c)

3

FA_0.11 and d) FA_0.57 geopolymer specimens immersed during 30 h in a MB

4

solution (35 mg/L). The inset micrographs (transversal section) illustrate the distinct

5

MB diffusion throughout the specimens. For interpretation of the references to colour in

6

this figure legend, the reader is referred to the web version of this article.

AC C

EP

TE D

2

39

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

1 2

Fig. 12 MB uptake and removal efficiency for specimen FA_0.57 after multiple

3

regeneration cycles.

7 8 9 10 11

EP

6

AC C

5

TE D

4

12 13 14 15

40

ACCEPTED MANUSCRIPT 1

Table 1 Geopolymer preparation: mixture composition.

2

Sample name

Mixture proportion (wt.%) MK

Sodium silicate

NaOH

H2O

H2O2

MK_0.11

-

42.25

37.56

16.43

3.64

0.11

MK_0.22

-

40.72

36.20

15.84

7.02

0.22

MK_0.41

-

37.98

33.76

14.77

13.10

0.41

MK_0.57

-

35.58

31.62

13.83

18.40

0.57

FA_0.11

28.17

14.08

37.56

16.43

3.64

0.11

FA_0.22

27.15

13.57

36.20

15.84

7.02

0.22

FA_0.41

25.32

12.66

33.76

14.77

13.10

0.41

FA_0.57

23.72

11.86

31.62

13.83

18.40

0.57

4 5

10 11 12

EP

9

AC C

8

TE D

6 7

SC

M AN U

3

RI PT

FA

13 14

41

ACCEPTED MANUSCRIPT 1

Table 2 Water absorption, apparent density and total porosity of FA-containing

2

geopolymers cured for 28 days.

3

7 8 9

density

(wt.%)

(g/cm3)

MK_0.11

35.9 ± 0.6

0.93 ± 0.01

MK_0.22

44.5 ± 0.6

0.80 ± 0.01

MK_0.41

58.5 ± 1.1

0.64 ± 0.03

MK_0.57

83.0 ± 2.5

0.56 ± 0.01

FA_0.11

25.5 ± 1.7

1.19 ± 0.02

FA_0.22

31.0 ± 1.1

0.99 ± 0.03

50.6

FA_0.41

51.0 ± 0.9

0.55 ± 0.04

72.6

FA_0.57

111.5 ± 5.4

porosity (%) 50.3 57.3 65.8 70.1 40.7

0.39 ± 0.02

80.6

AC C

EP

TE D

10

Total

RI PT

6

absorption

Sample name

SC

5

Apparent

M AN U

4

Water

42

ACCEPTED MANUSCRIPT

3 4

qe (mg/g)

Reference

0.7

(Zhang and Liu, 2013)

3.0

(Khan et al., 2015)

3.8

(Wang et al., 2015)

8.0

(Wang et al., 2015)

14.8

(Falah et al., 2015a)

14.8

(Falah and Mackenzie, 2015b)

powder (120
Phosphoric acid MK-based geopolymer

powder

Coal fly ash

powder

Acid treated coal fly ash

powder

Cu2O-geopolymer

powder

Cu2O /TiO2 geopolymer

powder

Biomass FA-geopolymer a

cylindrical monolith

15.4

-

Cu2O /TiO2-CTAB geopolymer

powder

19.7

(Falah et al., 2016)

Kaolin geopolymer

powder (250
25.6

(Yousef et al., 2009)

Zeolite

powder

33.5

(Rida et al., 2013)

Coal fly ash geopolymer

powder

38.4

(Li et al., 2006)

Coal fly ash geopolymer

50.7

(Liu et al., 2016)

a

SC

Coal fly ash geopolymer

TE D

powder (d<150 µm)

The font in bold identifies the results obtained in this work.

AC C

2

Adsorbent shape

EP

Material

RI PT

Table 3 Methylene blue adsorption capacity of various adsorbents.

M AN U

1

5

43

ACCEPTED MANUSCRIPT

Table 4 Langmuir and Freundlich parameters for MB adsorption on studied samples.

2 Langmuir qm (mg/g)

KL (L/mg)

Freundlich

R2

2.630

2.634

0.969

MK_0.57

3.479

1.283

0.982

FA_0.11

2.403

FA_0.57

4.499

n

R2

1.512

1.868 0.978

1.390

1.882 0.969

1.244

0.983

0.934

1.849 0.996

2.593

0.965

2.628

1.732 0.983

AC C

EP

TE D

3

KF (L/g)

SC

MK_0.11

M AN U

Sample

RI PT

1

44

ACCEPTED MANUSCRIPT Highlights •

Novel porous monolithic adsorbents for methylene blue extraction were developed. The geopolymers were produced using 66.6 wt.% biomass fly ash (unexplored

RI PT



waste).

Methylene blue uptake by the monolithic bodies up to 20.5 mg/g was observed.



The removal efficiency can be controlled by pores’ volume and connectivity.



Adsorbents can be used in packed beds, regenerated and reused (at least 5

SC



AC C

EP

TE D

M AN U

times).