close

Вход

Забыли?

вход по аккаунту

?

acssuschemeng.7b03344

код для вставкиСкачать
Subscriber access provided by READING UNIV
Article
Characterization and Catalytic Upgrading of Aqueous
Stream Carbon from Catalytic Fast Pyrolysis of Biomass
Anne Katherine Starace, Brenna A. Black, David D. Lee, Elizabeth Palmiotti, Kellene A. Orton, William E.
Michener, Jeroen ten Dam, Michael J Watson, Gregg T Beckham, Kimberly A. Magrini, and Calvin Mukarakate
ACS Sustainable Chem. Eng., Just Accepted Manuscript • DOI: 10.1021/
acssuschemeng.7b03344 • Publication Date (Web): 23 Oct 2017
Downloaded from http://pubs.acs.org on October 25, 2017
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
ACS Sustainable Chemistry & Engineering is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.
Page 1 of 36
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Sustainable Chemistry & Engineering
Characterization and Catalytic Upgrading of Aqueous Stream
Carbon from Catalytic Fast Pyrolysis of Biomass
Anne K. Starace,1 Brenna A. Black,1 David D. Lee,1 Elizabeth C. Palmiotti,1 Kellene A. Orton,1
William E. Michener,1 Jeroen ten Dam,2 Michael J. Watson,2 Gregg T. Beckham,1 Kimberly A.
Magrini1 and Calvin Mukarakate1*
1
National Renewable Energy Laboratory, 15013 Denver West Parkway, Golden, Colorado
80401, USA
2
Johnson Matthey Technology Centre, PO Box 1, Belasis Avenue, Billingham, Cleveland TS23
1LB, U.K.
Corresponding author
*Email: calvin.mukarakate@nrel.gov
ABSTRACT Catalytic fast pyrolysis (CFP) of biomass produces a liquid product consisting of organic
and aqueous streams. The organic stream is typically slated for hydrotreating to produce hydrocarbon
biofuels, while the aqueous stream is considered a waste stream, resulting in the loss of residual biogenic
carbon. Here, we report the detailed characterization and catalytic conversion of a CFP wastewater stream
with the ultimate aim to improve overall biomass utilization within a thermochemical biorefinery. An
aqueous stream derived from CFP of beech wood was comprehensively characterized, quantifying 53
organic compounds to a total of 17% organics. The most abundant classes of compounds are acids,
aldehydes, and alcohols. The most abundant components identified in the aqueous stream were C1-C2
organics, comprising 6.40% acetic acid, 2.16% methanol, and 1.84% formaldehyde on wet basis. The
1
ACS Paragon Plus Environment
ACS Sustainable Chemistry & Engineering
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Page 2 of 36
CFP aqueous stream was catalytically upgraded to olefins and aromatic hydrocarbons using a Ga/HZSM5 catalyst at 500°C. When the conversion yield of the upgraded products was measured with fresh, active
catalyst, 33% of the carbon in the aqueous stream was recovered as aromatic hydrocarbons and 29% as
olefins. The majority of the experiments were conducted using a molecular beam mass spectrometer and
separate GC-MS/FID experiments were used to confirm the assignments and quantification of products
with fresh excess catalyst. The recovered 62% carbon in the form of olefins and aromatics can be used to
make coproducts and/or fuels potentially improving biorefinery economics and sustainability. Spent
catalysts were collected after exposure to varying amounts of the feed, and were characterized using
multipoint-Brunauer–Emmett–Teller (BET) adsorption, ammonia temperature programmed desorption
(TPD), and thermogravimetric analysis (TGA) to monitor deactivation of Ga/HZSM-5. These
characterization data revealed that deactivation was caused by coke deposits, which blocked access to
active sites of the catalyst and spent catalysts regained total activity after regeneration.
KEY WORDS: CFP aqueous stream, wastewater treatment, HZSM-5, BTX, coke formation
2
ACS Paragon Plus Environment
Page 3 of 36
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Sustainable Chemistry & Engineering
INTRODUCTION
Catalytic fast pyrolysis (CFP) is a promising approach for converting non-food biomass, such as
agricultural and forestry residues, energy crops, municipal solid waste, and others, into
transportation fuels and chemicals.1–9 During this process, biomass vapors are either produced
and partially deoxygenated in the presence of a catalyst, a configuration denoted in situ CFP, or
are separated from solid biomass and deoxygenated over a catalyst bed in a separate reactor,
which is termed ex situ CFP.7,10,11 Dehydration is a prevalent deoxygenation mechanism during
both pyrolysis processes, wherein a hydroxyl group and a hydrogen cation are removed to
produce water. Thus, CFP produces a liquid product containing a significant amount of water,
and the condensed product typically self-separates into an aqueous stream and an organic
stream.12 The organic stream contains the majority of the desired fuel precursors and
consequently has been the predominant focus of CFP research, while the aqueous stream is
typically sent for wastewater treatment, thus comprising a potentially large biorefinery operating
cost.13
Previous CFP studies, in Figure 1, show that a significant amount of carbon (3-14 wt% of the
original biomass carbon) is retained in the aqueous stream.3–8 Paasikallio et al. reported that in
situ CFP of pine using a spray dried HZSM-5 catalyst produced an aqueous stream containing 32
wt% organics accounting for 14 wt% of the original biomass carbon.4 Iisa et al. reported that ex
situ CFP of pine using two HZSM-5 catalysts with clay and silica binders generated aqueous
streams with 8 wt% (clay) and 15 wt% (silica) organics.6 The organic content in the aqueous
stream from the clay binder catalyst accounted for 3 wt% of the biomass carbon and those from
silica binder catalyst accounted for 7 wt% of the biomass carbon. Aqueous streams were further
3
ACS Paragon Plus Environment
ACS Sustainable Chemistry & Engineering
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Page 4 of 36
generated from in situ and ex situ CFP of pine using commercial HZSM-5 catalysts (Nexceris
and Johnson Matthey) with comparable acid strengths.7 The in situ CFP aqueous stream
exhibited approximately 14 wt% organics accounting for 6-7 wt% of the biomass carbon, while
the ex situ stream had about 11 wt% organics accounting for 5 wt% of the biomass carbon.
Dayton et al. generated an aqueous stream containing 10 wt% organics accounting for 5 wt%
carbon of the biomass from in situ CFP of pine.8 The organic compounds characterized in these
various CFP aqueous fractions contained mostly typical carbohydrate pyrolysis products such as
acids, aldehydes, ketones, and anhydrosugars; additionally, simple phenols and small amounts of
lignin derived methoxyphenols were also observed.3,6,7 A recent detailed characterization of fast
pyrolysis and CFP aqueous streams from a range of process conditions reported greater than
75% mass closure.3 These aqueous streams contained additional compounds and different
proportions of organics than those normally observed in pyrolysis vapors and bio-oils. In
addition to containing biogenic carbon, these studies shows that the CFP aqueous streams
contains 57-62 wt% of the total liquid yield.5–9 This is a large volume to send to wastewater
treatment, as is typically proposed in biorefinery concept models.13 An alternative approach is to
treat the CFP aqueous stream as a revenue stream by recovering and/or upgrading organics to
produce coproducts or fuels. Hence, there has been recent interest in characterizing and
valorizing aqueous streams to improve overall carbon conversion efficiencies and offset
biomass-derived fuel production cost in a biorefinery context.3 The amount of biomass carbon
retained in the aqueous streams depends on the CFP process, with the general trend that as the
amount of deoxygenation increases, the amount of carbon in the aqueous stream and the organic
yield decrease. Since the organic product yield needs to be balanced with the organic product
4
ACS Paragon Plus Environment
Page 5 of 36
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Sustainable Chemistry & Engineering
oxygen content, the optimum will be determined by the tradeoff between these two competing
factors.
There have been relatively few reports presented in the literature on the valorization of CFP
aqueous streams. One example is from Wilson et al., wherein phenol, cresol, and alkyl phenols
were separated from a CFP aqueous stream and used to produce bioderived resol resin.14 Another
example is steam reforming CFP aqueous stream to produce hydrogen. Work by Kechagiopoulos
et al. achieved approximately 60% hydrogen yield from catalytic steam reforming of a CFP
aqueous stream. This yield was lower than that obtained with model compounds (90%), although
the authors noted that the low yield likely resulted from the deposition of coke, and thus higher
yields would be expected after process improvements are made.15
Conversely, more current studies have addressed valorizing fast pyrolysis (FP) aqueous streams.
FP oil consists of biomass pyrolysis products and is typically an emulsion, but can be readily
separated into an organic fraction and an aqueous fraction by adding water and mixing. Pollard et
al. have developed a condensation train to separate pyrolysis oil into five different fractions so
that the addition of water is not necessary to produce an aqueous stream, and a low water content
stream is also produced.16 Since the approaches to valorizing both CFP and FP aqueous streams
are similar, we also briefly review previous work to valorize FP aqueous streams, which includes
many different approaches. The aqueous stream can be used as both the steam source and a
hydrocarbon source for catalytic steam reforming to produce hydrogen.17 An acid-rich aqueous
stream, produced from the fractional condensation of FP pine vapors, has also been used as a
pretreatment technique to wash biomass, to reduce its inorganic content, prior to pyrolysis.18
Another method used a series of separation techniques to remove valuable organic components
5
ACS Paragon Plus Environment
ACS Sustainable Chemistry & Engineering
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Page 6 of 36
in the aqueous stream. For example, Teella et al. used nanofiltration and reverse osmosis to study
the feasibility of separating carboxylic acids and sugars in simulated aqueous streams.19 Another
approach is to catalytically convert the organics in the aqueous stream to produce fuels and
chemicals.9,20–22 Vispute et al. partitioned an aqueous stream from FP oil, and catalytically
converted the organic component of the aqueous stream into polyols using Ru/C in a batch
reactor.21 Similarly, Abnisa et al. upgraded the aqueous fraction of FP oil from palm shell by
passing it over HZSM-5 from 405-555°C. Between 20 and 30% of the biomass carbon was
retained in the upgraded oil stream 9 Mukarakate et al. recently used HZSM-5 with a silicon-toaluminum ratio (SAR) of 30 at 500°C to upgrade an oak FP aqueous stream to aromatics and
phenols, recovering 40% of the carbon from the aqueous stream.23
To date, there has been no work reported on direct catalytic conversion of a CFP aqueous stream
without preprocessing a priori. We employed an aqueous stream sample generated from in situ
CFP of beech wood using a pilot plant at the Center for Research and Technology Hellas
(CERTH) in Greece.5 During this process, beech wood was pyrolyzed in a circulating fluidizedbed reactor using HZSM-5 at a catalyst-to-biomass ratio of around 16 to generate an aqueous
stream containing 20% organics, accounting for 10% of the biomass carbon. This sample was
comprehensively characterized prior to upgrading over a Ga-modified catalyst at 500°C. The
resulting products and subsequent catalyst deactivation were analyzed in real time with a
molecular beam mass spectrometer (MBMS). Pre-and post-use catalysts at three feed-to-catalyst
ratios were characterized to determine the extent of catalyst coking during this process.
Additional catalytic upgrading experiments with a GC-MS/FID system were conducted to
support the assignments used during the mass-spectrometry based method (MBMS) used to
6
ACS Paragon Plus Environment
Page 7 of 36
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Sustainable Chemistry & Engineering
quantify products. The GC-MS/FID studies were conducted using fresh excess catalyst. Our goal
was to conduct an in-depth characterization of the aqueous stream and perform experiments to
determine the feasibility of recovering the waste carbon via catalytic conversion to value-added
co-products, benzene, toluene, and xylene (BTX), with no preprocessing. These data are
necessary to develop rigorous techno-economic models of CFP processes to determine the yields
and products required for catalytic upgrading of the aqueous stream. Valorization of the aqueous
stream, instead of sending it to wastewater treatment, may in turn reduce the overall
thermochemical biorefinery operating costs, and potentially, reduce the minimum biofuel selling
price.
MATERIALS AND METHODS
Materials. The aqueous stream used in this work was produced from in situ CFP of beech wood
with HZSM-5 catalyst, silica-to-alumina ratio (SAR) of 50, at CERTH, as described
previously.5,24 This aqueous stream will be referred to as “CFP aqueous stream” for the
remainder of the article. The unmodified HZSM-5 catalyst with a SAR of 30, silica binder, and 1
mm particle size was received from Johnson Matthey (Chilton, UK). The Ga/HZSM-5 catalyst
was also produced by Johnson Matthey by adding 5 wt% Ga to the unmodified HZSM-5 catalyst.
Ga was loaded to HZSM-5 using incipient wetness and inductively coupled plasma mass
spectrometry analysis showed that the Ga loading was 5.0%
Aqueous stream characterization. Carbon, hydrogen, and nitrogen content were measured with
a LECO TruSpec CHN module (LECO Corp., St. Joseph, MI) via high temperature combustion
7
ACS Paragon Plus Environment
ACS Sustainable Chemistry & Engineering
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Page 8 of 36
followed by infrared (IR) and thermal conductivity detector (TCD) analysis. Detailed description
of this process is given in the Supporting Information (SI).
The water content of the aqueous stream was measured via Karl Fisher titration according to the
standard ASTM E203-08 method. Analysis of the inorganic content of the aqueous stream was
conducted by digesting the aqueous stream in 72 wt% nitric acid at 200°C for 15 minutes,
filtering, and analyzing the resulting solution using inductively coupled plasma atomic emission
spectroscopy (ICP-AES). Total organic carbon (TOC) analysis was measured by a Shimadzu
TOC-LCSH analyzer (Shimadzu, Columbia, MD) via a combustion catalytic oxidation method
after sample acidification with concentrated hydrochloric acid.
Quantification of individual compounds contained in the organic portion of the aqueous stream
was performed using a suite of methods developed by Black et al.3 Briefly, sugars, organic acids,
select aldehydes, aromatics, phenolics, and higher molecular weight ketones were analyzed using
liquid chromatography (LC) coupled in a system consisting of a Shodex SZ5532 LC column, 6
mm i.d. × 150 mm (Showa Denko America Inc., New York, NY) with evaporative light
scattering detection/mass spectrometry (MS); an Aminex HPX-87H LC column, 7.8 mm i.d. ×
300 mm column (Bio-Rad Laboratories, Hercules, CA) with refractive index detection/MS; or a
YMC C30 LC column, 4.6 mm i.d. × 150 mm column (YMC America, Allentown, PA) with
diode array detection/MS. The remaining compounds were analyzed using gas chromatography
(GC)-MS with a Stabilwax, 30 m × 0.25 mm i.d., 0.25 µm film thickness (J & W Scientific Inc.,
Folsom, CA); HP 1 or HP 5 column. Detailed methods using the latter two GC columns have not
been previously published and are as follows: an Agilent 7890A GC and Agilent 5975C mass-
8
ACS Paragon Plus Environment
Page 9 of 36
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Sustainable Chemistry & Engineering
selective detector (Agilent Technologies Inc., Santa Clara, CA) was used for the characterization
of analytes in the aqueous stream. Using a splitless injection, 1 µL sample volume was
introduced onto a 30 m × 0.25 mm i.d., 0.25 µm film thickness HP 5-MS or HP 1-MS capillary
column (J & W Scientific Inc., Folsom, CA) at 280°C. The helium flow was kept constant at 1
mL/min with an oven program as follows: the initial column temperature of 35°C was held for 5
min and then increased to 225°C at 5°C/min, and lastly, to 325°C at 15°C/min with a hold time
of 10 min. Electron impact ionization was used at 70 eV electron energy and a mass scan range
of m/z 25 – 575. An Agilent Environmental ChemStation G1701DA version D.00.00.38 was
used for data analysis.
MS was incorporated with all quantitative analyses as often as possible and individual
compounds were often characterized by multiple analytical methods to corroborate calculated
concentrations. All analyses were performed in triplicate independent experiments and all
quantitative standard curves were maintained with an R2 value of ≥ 0.995 with five or more
points of reference ranging between concentrations of 1 to 100 µg/mL. Individual authentic
standards were obtained for quantitation in the highest purity available (Table S1) for the
construction of each external calibration curve and select internal standards were added to adjust
for chromatographic and detector response shift, as outlined in previous work.3
Vertical Reactor-MBMS. A vertical reactor coupled with an MBMS (Extrel CMS, Pittsburgh,
PA) was used to upgrade the aqueous stream and characterize the resulting products in real time.
A detailed description of this apparatus was reported previously.23 A schematic diagram of the
reactor is shown in Figure S1A. Briefly, the aqueous stream was metered into the reactor using a
9
ACS Paragon Plus Environment
ACS Sustainable Chemistry & Engineering
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Page 10 of 36
syringe pump at a feed rate of 56 µL/min. To create a uniform flow of the aqueous stream, it was
fed through a 120 kHz ultrasonic nozzle with a sheath of carrier gas (helium at 600 sccm). The
bottom view of the initial design of this sheet is shown in Figure S1B. The theory and design of
the sheath, shown in Figure S1C, can be found elsewhere25 and it provides more carrier gas
shielding between the liquid in the center of the reactor and the walls of the reactor. The center
of the reactor was kept at 500°C by a surrounding tube furnace. 0.5-1.0 g of un-reduced catalyst
was held in place in the reactor by two plugs of quartz wool. After passing over the catalyst, the
vapors were mixed with helium diluent gas before sampling and analysis by the MBMS. The
upgraded products were quantified using response factors calculated from standards as described
in the SI. Figure S2 shows typical calibration curves and response factors from the MBMS.
GC-MS/FID system.
A tandem microreactor system (Rx-5030TR, Frontier Laboratories,
Japan) coupled to a Gas Chromatography Mass Spectometer –Flame Ionization DetectorThermal Conductivity Detector (GC-MS/FID) system was used to identify and quantify products
from catalytic conversion of organics in the aqueous stream. A detailed description of this
system has been reported in previous studies.26–28 For the current study one microliter aliquots of
the aqueous stream samples were injected into the reactor instead of biomass. Briefly, this
system consists of two reactors connected in series; the first reactor was used for volatilizing
organics in the aqueous stream and the second reactor was used for catalytic upgrading. Both
reactors were set to 500 °C and 20mg catalyst was loaded in the second reactor. After upgrading
the products were measured using the GC-MS/FID system, except for CO, which was quantified
using TCD. The GC oven was held at 40 °C for 3 min and heated to 300 °C at a ramp rate of 10
°C·min-1. Standards comprising 25 representative compounds (8 aromatic hydrocarbons, 10
10
ACS Paragon Plus Environment
Page 11 of 36
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Sustainable Chemistry & Engineering
oxygenates, 5 olefins, and CO and CO2) were used to calculate response factor for quantifying
the upgraded products. Compounds without standards were quantified using response factors
from compounds with similar functional groups and similar molecular weights.
Catalyst characterization. In experiments with varying aqueous feed-to-catalyst ratios, the
post-use catalyst was cooled under a flow of inert gas prior to removal from the reactor for coke
analysis. The surface area of the coked, as well as the fresh, catalyst was performed using
nitrogen physisorption at 77 K with a Quadrasob SI surface area and pore size analyzer from
Quantachrom Instruments (Boynton Beach, FL). More detail on this method is given in the SI.
Catalyst coke content was measured by heating in an atmosphere of 20% oxygen in nitrogen to
700°C at a ramp rate of 20°C/min. Mass loss up to 250°C was from water loss and mass loss
between 250-700°C was coke loss via combustion. The amount of coke was reported on a g per
g of dry catalyst basis. The number of active sites on the fresh and coked catalyst was measured
via ammonia TPD on an AMI-390 micro activity test system (Altamira Instruments, Pittsburg,
PA), for which the method details are provided in SI.
RESULTS AND DISCUSSION
Characterization of the aqueous stream. A summary of the compositional and chemical
characterization of the aqueous stream is shown in Figure 2 and the quantification of individual
organic compounds and inorganic elements identified in this stream are shown in Tables 1 and 3,
respectively. As can be seen in Figure 2, the water content of the aqueous stream was 83.03 ±
1.18% by weight and the inorganic content was negligible at 0.030 ± 0.001%. The total organic
content was calculated in two ways; (1) by applying a factor of 2.5 (out of a range of 1.7-2.5) to
11
ACS Paragon Plus Environment
ACS Sustainable Chemistry & Engineering
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Page 12 of 36
the TOC measurement, as suggested by Iglesias et al.29 and Bader,30 resulting in 16.90% organic
content, and (2) by difference from the Karl Fisher water measurement resulting in 17.09%
organic content. This resulted in a total mass balance range from 99.96% to 100.15%. Figure 2
also shows that acids were the largest class of compounds in this stream comprising 52.9 wt% of
the organic fraction on a dry basis, followed by aldehydes (18.5 wt%), alcohols (15.3 wt%),
ketones (7.54 wt%), and aromatics (4.6 wt%). Interestingly, only six compounds were identified
as acids compared to 24 compounds identified as aromatics, almost all of them phenols. Table 1
shows that on a wet basis, the most abundant compound present was acetic acid, which
comprised 6.40 wt% of the solution, followed by methanol (2.16 wt%), formaldehyde (1.84
wt%), and formic acid (1.01 wt%).The carbon and hydrogen content of the CFP aqueous stream
is shown in Table 2. Values are an average of triplicate measurements plus or minus two
standard deviations. The carbon and hydrogen content were re-measured three months after the
original measurement, verifying that they had not changed significantly.
The ICP-AES analysis of 13 elements, shown in Table 3, found that sulfur was the most
abundant element at 304.2 ppm. The next most abundant was sodium (13.7 ppm) followed by
aluminum (9.8 ppm), phosphorous (9.4 ppm), calcium (7.1 ppm), iron (3.5 ppm), and
magnesium (1.7 ppm). In previous work, where elemental analysis of oils and aqueous streams
from FP followed by hydrotreating of eight feedstocks was performed, sulfur and sodium were
detected in the aqueous stream, but the amount of aluminum was below the detection limit.31
This result suggests that perhaps the aluminum measured in the CFP aqueous stream came from
dealumination of the zeolite catalyst used during the CFP process. This higher abundance of Al
12
ACS Paragon Plus Environment
Page 13 of 36
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Sustainable Chemistry & Engineering
in the CFP derived aqueous stream compared to in FP aqueous streams was also reported by
Black et al.3
CFP aqueous stream upgrading. The data in Table 1 and previous work3 show that the aqueous
stream generally contains a large number of organic species, each in small mass fractions. Thus,
our work on valorizing this stream was focused on catalytic upgrading, which can achieve a
relatively narrow product distribution from a complex feedstock. We first attempted to upgrade
the aqueous fraction from FP bio-oils (pine and oak) to olefins and aromatics, and the results
from this study are presented in Figure S3 in the SI. As can be seen in Figure S3, the Ga/HZSM5 produced a higher aromatic yield than the unmodified HZSM-5 for both the aqueous stream
from the FP of pine (denoted FP pine) and FP of oak (FP oak). Based on these results, we chose
to use the Ga/HZSM-5 SAR 30 catalyst for improved carbon recovery from the CFP aqueous
stream. Additionally, Ga/ZSM-5 has shown increased aromatic yield compared with unmodified
HZSM-5 as previously reported.32
A baseline mass spectrum of the CFP aqueous stream was obtained with the MBMS by passing it
over inert sand as a catalyst surrogate at 500°C (Figure 3A). Good agreement was observed
between the compounds identified in the baseline mass spectrum and those identified in the
detailed LC/MS and GC/MS chemical characterization in Table 1. Note that Figure 3A appear to
show fewer compounds compared to those observed in Table 1,however, this is simply because
many of the compounds are present in very small concentrations which result in low intensity
peaks that are not easily seen when the spectrum is scaled to view the largest peaks. Considering
the compounds identified in the chemical characterization, the species in Figure 3A can be
13
ACS Paragon Plus Environment
ACS Sustainable Chemistry & Engineering
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Page 14 of 36
assigned as: formaldehyde m/z 30, methanol m/z 32, acetic acid and hydroxyacetaldehyde m/z
60, cyclopent-2-en-1-one m/z 82, phenol m/z 94, methyl phenol m/z 108, catechol m/z 110, and
2-methoxyphenol and methylcatechol m/z 124. Additionally, the following fragment peaks could
be due to: acetaldehyde and formic acid m/z 29, methanol, butanol, and hydroxyacetic acid
fragments m/z 31, butanol fragment m/z 42, propanone and hydroxypropanone m/z 43, and
acetaldehyde fragment m/z 44. This agreement between the MBMS characterization at 500°C
and the LC/MS and GC/MS characterization at lower temperatures indicates that the chemical
composition of the aqueous stream does not undergo significant thermal decomposition at
500°C.
When the aqueous stream containing these oxygenated species (Figure 3A) was upgraded over
fresh Ga/HZSM-5, the mass spectrum changed to Figure 3B, which contains completely
deoxygenated species, primarily olefins (ethylene m/z 28, propylene m/z 42, butylene and methyl
propylenes m/z 56) and aromatic hydrocarbons (benzene and alkyl benzenes m/z 78, 92, 106,
indene and alkyl indene m/z 116, 130, naphthalene and alkyl naphthalenes m/z 128, 142, 156,
and anthracene and alkyl anthracenes m/z 178, 192, 206). Oxygen loss was achieved via
dehydration m/z 18, decarbonylation m/z 28, and decarboxylation m/z 44. The majority of these
assignments were confirmed using the GC-MS experiments discussed below. As the catalyst
aged, Figure 3C, the quantity of the completely deoxygenated products decreased and
breakthrough of compounds found in the aqueous stream, for example phenol m/z 94 and methyl
phenol 108 shown in blue were observed. The comprehensive characterization of the aqueous
stream prior to upgrading allowed us to identify which compounds were most likely due to
breakthrough and also to identify intermediates and/or side products, which are not the
14
ACS Paragon Plus Environment
Page 15 of 36
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Sustainable Chemistry & Engineering
completely deoxygenated products seen over the completely active catalyst or compounds
identified in the starting material. For instance, methyl benzofuran m/z 132, dimethyl benzofuran
m/z 146, naphthol m/z 144, and methyl naphthol m/z 158, also in blue, do not correspond to any
of the compounds identified in the CFP aqueous stream, as shown in Figure 3C. Benzofurans
(m/z 132 and 146) are partially deoxygenated products likely formed due to lack of accessible
active sites as the catalyst aged, and naphthols (m/z 144 and 158) as well as phenols have been
shown to be formed from reactions of water and coke precursors on the catalyst.26
Figure 4 shows the change in product distribution for selected compounds found in Figure 3
mass spectra, as more aqueous stream was fed over the fixed-bed of Ga/HZSM-5. The single
ring aromatics comprised benzene and alkyl benzenes, multi ring aromatics comprised
napthalene, alkyl naphthalenes, anthracene and alkyl anthracenes, side products comprised
napthol and alkyl naphthols, and breakthrough species comprised acetic acid, catechol, guaiacol,
and methyl guaiacol. At the beginning of the experiment, the active sites of the catalyst are
accessible and the organics available in the aqueous stream are almost completely deoxygenated
to produce olefins and aromatic hydrocarbons shown in mass spectrum in Figure 3B. The
catalyst maintains its activity for producing aromatic hydrocarbons for the first 10-15 minutes of
the experiment as shown by the profile for multi ring aromatics. The aromatic hydrocarbons then
decreases gradually until the end of the experiment.
Conversely, the single ring aromatic
compounds begin high and stay constant for the first 10 minutes and decreases rapidly until the
end of the experiment. Some of the starting materials breakthrough almost immediately and they
appear to increases at the same rate as the decrease for single ring aromatics. The breakthrough
species observed during the first 40 minutes comprised of phenol and methyl phenols as
15
ACS Paragon Plus Environment
ACS Sustainable Chemistry & Engineering
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Page 16 of 36
indicated by mass spectrum in Figure 3C. Phenol and cresols could also be produced as side
products through reactions of water and coke precursors as discussed earlier because they appear
in the mass spectrum at the same time as naphthols and alkyl naphthols (Figure 3).26 The mass
spectrum in Figure 3C does not show other breakthrough species such as acetic acid and catechol
because they were still being upgraded over the catalyst to produce aromatic hydrocarbons or
forming coke. Phenol could breakthrough the catalyst because it has been demonstrated to be
recalcitrant towards conversion over HZSM-5.33 It strongly adsorbs on acid sites causing
deactivation through both active site and pore blockage. The amount of side products formed is
relatively constant throughout the experiment. The change in product distributions observed in
Figure 4 may be due to coke deposits on the catalyst, which block active sites and this result was
supported by the catalysts characterization studies discussed below.
Three separate experiments were conducted to quantify products from upgrading organics in the
aqueous stream with Ga/HZSM-5. Each experiment used 56 µL/min of aqueous stream with a
carrier gas feed of 600 sccm He, but differed in the total length of the experiment. After each
experiment, the catalyst was cooled under a flow of inert gas and collected for further analysis,
resulting in coked catalyst samples at 0.5, 1, and 1.5 feed carbon-to-catalyst ratios (F:C) on a
mass to mass basis. Averaged spectra of these three experiments are shown in Figures S4-6.
Spectra were normalized to the argon tracer signal and m/z 18, from water, was truncated. The
mass spectra at the beginning of the experiment are similar for the three experiments and they
contain species shown in Figure 3B.
16
ACS Paragon Plus Environment
Page 17 of 36
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Sustainable Chemistry & Engineering
The average carbon yields from conversion after 10 minutes time on stream, calculated as a
percent of the total carbon fed over the catalyst, are given in Table 4. Predominantly one–ring
aromatics were formed during the first 10 minutes (Figure 4), including toluene, which formed in
the highest quantity (~16.2% of the carbon fed), followed by xylenes (~6.5%), and then benzene
(~5.2%). This gives a total BTX (benzene, toluene, xylene) product selectivity of 86% of the
32.5% recovered from the initial carbon fed. The yield of olefins (the sum of ethylene,
propylene, and butylene) was approximately 29% of the initial carbon fed. This result is
consistent with upgrading FP aqueous stream over Ga/HZSM-5, which also found roughly equal
carbon yields for olefins and aromatics (see Figure S2). The amount of CO2 produced over fresh,
active catalyst was around 12% of the carbon fed and CO was 23%. This results in a carbon mass
balance of 96.5% without including coke yield. The amounts of olefins formed were surprisingly
higher than normally formed from catalytic upgrading of biomass with HZSM-5 based catalysts.
The high olefin yield could be due to the excess water in the feed, which has been demonstrated
to increase conversions of furan,34 and benzaldehyde.35 Gilbert et al. 34 found that increasing the
molar ratio of water to furan from 3.5 to 47.7 increased the furan conversion over ZSM-5 at
600°C from 51.8% to 84.8%. Conversion of benzaldehyde with Ga/ZSM-5 at 450°C was found
to increase from 55.8 to 67.52% with the addition of water.35 The increased conversion rated
could be due to the water reducing coke deposits through steam stripping and making active sites
accessible for reactions. Along with the increased conversion of furan with increasing water
content, the amount of olefins increased dramatically, while the amounts of aromatics decreased.
The presence of water also increased the selectivity to one-ring aromatics and decreased the
selectivity to multi ring aromatics. This result agrees with our data in Table 1 showing a much
17
ACS Paragon Plus Environment
ACS Sustainable Chemistry & Engineering
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Page 18 of 36
larger amount of olefins and more one ring aromatic hydrocarbons compared to two ring
aromatic hydrocarbons.
Since the MBMS is a mass spectrometry based analytical technique, additional experiments
were conducted using a GC-MS/FID system to complement the results. This system allows us to
identify and quantify products from upgrading organics in the aqueous stream using low feed-tocatalyst ratios. It is a batch system, and during the experiment three one microliter samples were
separately volatilized at 500 °C and the volatile compounds were upgraded over 20mg
Ga/HZSM-5. The upgraded products identified in the GCMS included olefins (ethylene,
propylene, 2-methylpropylene) and aromatic hydrocarbons (benzene, toluene, p-xylene, mxylene, 1-ethyl 3-methyl benzene, indene, naphthalene, and methyl naphthalenes). These species
are consistent with the products measured from the MBMS, except for the three ring aromatic
hydrocarbons which were not observed in the GC-MS studies. Table 4 shows that the amount of
olefins produced under these conditions were 43% and the aromatics yield was 19%. Propylene
and 2-methylpropylene were the major olefins produced as shown in Table S1. The yield of
aromatics from the GCMS is lower than that from the MBMS because the GCMS data was
collected at low feed carbon-to-catalyst ratio (0.03 vs 0.2). This could also be due to the GCMS
not able to chromatograph heavy aromatic compounds (two to three ring aromatic
hydrocarbons). Detailed GC-MS/FID results from three separate aqueous stream injections are
presented in Table S1. The carbon balance for the GC-MS/FID study was 87%, which was lower
than measured on the MBMS. As mentioned earlier, the missing carbon could be due to losses in
the column. In addition, coke yield is missing from these values because our attempt to
determine the coke contents on the catalyst was not successful because of the small amount of
18
ACS Paragon Plus Environment
Page 19 of 36
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Sustainable Chemistry & Engineering
feed and catalyst used in this system and the detection limit of the thermogravimetric analyzer
used.
Catalyst characterization. In order to evaluate the effect of coke on accessibility of active sites,
spent catalysts from the three experiments described above from the MBMS were collected for
analysis. The amount of coke on each spent catalyst is shown in the blue, checkered bars of
Figure 5 as a weight percent on a dry-coke free basis, i.e. the mass loss of coke divided by the
dry, coke-free catalyst mass and multiplied by 100. The uncertainty of the TGA measurement, as
measured with a standard, was less than ± 3%. The possibility of uneven catalyst coking was
minimized by mixing the entire coked sample before removing a portion for analysis. The coke
data were measured from multiple samples collected after mixing the catalyst. On average, the
0.5, 1.0, and 1.5 F:C post reaction catalysts contained 3.8, 6.0, and 6.7% coke yields. This
demonstrated, unsurprisingly, that the rate of coke deposition was highest on the fresh catalyst
and that the rate of coke deposition slowed after longer times on stream due to decrease in
accessible active sites.
The reduction in accessible active sites was confirmed by measuring the total acid sites on these
three spent catalysts using NH3-TPD. As can be seen by the black hashed bars in Figure 5, the
fresh catalyst contained 620 µmol/g of accessible acid sites, while the samples collected at F:C
ratios 0.5, 1.0, and 1.5 exhibit 450, 300, and 220 µmol/g of accessible acid sites, respectively.
Thus the drop in acid sites from F:C of 0.5 to 1 (150) was approximately as large as the drop
from fresh to F:C 0.5 (170). Conversely, the decrease in the number of acid sites from F:C of 1 to
1.5 was only 80, which was only slightly above the uncertainty of the measurement (about ± 50).
19
ACS Paragon Plus Environment
ACS Sustainable Chemistry & Engineering
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Page 20 of 36
Additional evidence can also be seen by the multi-point BET surface area results shown in the
solid red bars in Figure 5. Surface areas (m2/g) of the fresh, F:C 0.5, F:C 1.0, and F:C 1.5 catalyst
were 354, 280, 199, and 224, respectively. The decrease in surface area from fresh to F:C 0.5 is
about the same as the decrease from F:C 0.5 to F:C 1.0. Surface area measurements have the
highest amount of uncertainty of any of the coke characterization techniques, around ±10%.
With this large uncertainty, the surface area of the F:C 1.0 and F:C 1.5 coked catalyst samples
are statistically equivalent. Based on the overall results for coke characterization, it is clear that
most coke is deposited up to a F:C of 1.0, which rapidly decreased the surface area and number
of acid sites of the catalyst. After a F:C ratio of 1.0 has been achieved, additional feed passed
over the catalyst has little effect on the surface area and number of acid sites. Steam can cause
dealumination of zeolites resulting in irreversible deactivation of the catalyst. These samples
regained their total activity after regeneration indicating that these short runs did not cause
significant dealumination. However, dealumination will occur in large scale systems and can be
minimized by adding phosphorous to HZSM-5.36
This process recovered a total of 61.5% carbon in the form of olefins and aromatic hydrocarbons,
which if sold commercially can generate revenue to potentially improve biorefinery economics.
Future work will focus on concentrating organics by removing water from the aqueous stream in
a manner that does not add significant cost or energy to the process to reduce the size and cost of
the upgrading reactor. For example, Engtrakul et al. recently demonstrated the application of
high-performance architectured surface-selective (HiPAS) membranes for concentrating organics
from a mixture of water and biomass CFP vapors.37 A preconcentration step of this type will
likely be required to make this process economically feasible. Additionally, we will assess the
20
ACS Paragon Plus Environment
Page 21 of 36
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Sustainable Chemistry & Engineering
impact of catalyst type on products with the goal of tuning catalysts to achieve selective
production of biomass derived chemicals more valuable than BTX.
21
ACS Paragon Plus Environment
ACS Sustainable Chemistry & Engineering
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Page 22 of 36
ASSOCIATED CONTENT
Supporting Information.
Additional information for experimental conditions used for aqueous stream and catalyst
characterization and quantification of upgraded products.
ACKNOWLEDGEMENTS
The compositional analysis and catalysis efforts in this study were performed in collaboration
with the Chemical Catalysis for Bioenergy Consortium, an Energy Materials Network
Consortium funded by the U.S. Department of Energy’s Bioenergy Technologies Office Contract
No. DE-AC36-08GO28308 with the National Renewable Energy Laboratory (NREL). We thank
Johnson Matthey for providing the catalyst. We thank Konstantinos G. Kalogiannis and Angelos
A. Lappas from CERTH, Chemical Process and Energy Resources Institute (CPERI) for
providing us with the CFP aqueous stream. We also thank Steve Deutch for running ICP-OES
analysis, as well as Matthew Yung and Jonathan Wells for ammonia TPD analysis.
REFERENCES
(1)
Talmadge, M. S.; Baldwin, R. M.; Biddy, M. J.; McCormick, R. L.; Beckham, G. T.;
Ferguson, G. A.; Czernik, S.; Magrini-Bair, K. A.; Foust, T. D.; Metelski, P. D.; et al. A
perspective on oxygenated species in the refinery integration of pyrolysis oil. Green
Chem. 2014, 16, 407–453.
(2)
Huber, G. W.; Iborra, S.; Corma, A. Synthesis of transportation fuels from biomass:
chemistry, catalysts, and engineering. Chem. Rev. 2006, 106 (9), 4044–4098.
(3)
Black, B. A.; Michener, W. E.; Ramirez, K. J.; Biddy, M. J.; Knott, B. C.; Jarvis, M. W.;
Olstad, J.; Mante, O. D.; Dayton, D. C.; Beckham, G. T. Aqueous stream characterization
from biomass fast pyrolysis and catalytic fast pyrolysis. ACS Sustain. Chem. Eng. 2016, 4,
6815–6827.
22
ACS Paragon Plus Environment
Page 23 of 36
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Sustainable Chemistry & Engineering
(4)
Paasikallio, V.; Lindfors, C.; Kuoppala, E.; Solantausta, Y.; Oasmaa, A.; Lehto, J.;
Lehtonen, J. Product quality and catalyst deactivation in a four day catalytic fast pyrolysis
production run. Green Chem. 2014, 16 (7), 3549–3559.
(5)
Vasalos, I. A.; Lappas, A. A.; Kopalidou, E. P.; Kalogiannis, K. G. Biomass catalytic
pyrolysis : process design and economic analysis. WIREs Energy Environ. 2016, 5, 370–
383.
(6)
Iisa, K.; French, R. J.; Orton, K. A.; Budhi, S.; Mukarakate, C.; Stanton, A. R.; Yung, M.
M.; Nimlos, M. R.; Alumina, B. Á.; Clay, Á. S. Á. Catalytic Pyrolysis of Pine Over
HZSM-5 with Different Binders. Top. Catal. 2016, 59 (1), 94–108.
(7)
Iisa, K.; French, R. J.; Orton, K. A.; Yung, M. M.; Johnson, D. K.; ten Dam, J.; Watson,
M. J.; Nimlos, M. R. In Situ and ex Situ Catalytic Pyrolysis of Pine in a Bench-Scale
Fluidized Bed Reactor System. Energy & Fuels 2016, 30, 2144–2457.
(8)
Dayton, D. C.; Carpenter, J. R.; Kataria, A.; Peters, J. E.; Barbee, D.; Mante, O. D.;
Gupta, R. Design and operation of a pilot-scale catalytic biomass pyrolysis unit. Green
Chem. 2015, 17, 4680–4689.
(9)
Abnisa, F.; Daud, W. M. A. W.; Arami-niya, A.; Ali, B. S.; Sahu, J. N. Recovery of
Liquid Fuel from the Aqueous Phase of Pyrolysis Oil Using Catalytic Conversion. Energy
& Fuels 2014, 28, 3074–3085.
(10)
Wang, K.; Johnston, P. a; Brown, R. C. Comparison of in-situ and ex-situ catalytic
pyrolysis in a micro-reactor system. Bioresour. Technol. 2014, 173, 124–131.
(11)
Gamliel, D. P.; Du, S.; Bollas, G. M.; Valla, J. A. Bioresource Technology Investigation
of in situ and ex situ catalytic pyrolysis of miscanthus x giganteus using a PyGC – MS
microsystem and comparison with a bench-scale spouted-bed reactor. Bioresour. Technol.
2015, 191, 187–196.
(12)
Paasikallio, V.; Kalogiannis, K.; Lappas, A.; Lehto, J.; Lehtonen, J. Catalytic Fast
Pyrolysis: Influencing Bio-Oil Quality with the Catalyst-to-Biomass Ratio. Energy
Technol. 2016, 5, 94–103.
(13)
U.S. Department of Energy. Bioenergy Technologies Office Multi-Year Program Plan.
Washington, D.C. 2016, pp 1–258.
(14)
Wilson, A. N.; Price, M. J.; Mukarakate, C.; Katahira, R.; Gri, M. B.; Dorgan, J. R.;
Olstad, J.; Magrini, K. A.; Nimlos, M. R. Integrated Biorefining: Coproduction of
Renewable Resol Biopolymer for Aqueous Stream Valorization. ACS Sustain. Chem. Eng.
2017, 5, 6615–6625.
(15)
Kechagiopoulos, P. N.; Voutetakis, S. S.; Lemonidou, A. A.; Vasalos, I. A. Hydrogen
Production via Steam Reforming of the Aqueous Phase of Bio-Oil in a Fixed Bed Reactor.
Energy & Fuels 2006, 20 (12), 2155–2163.
(16)
Pollard, A. S.; Rover, M. R.; Brown, R. C. Characterization of bio-oil recovered as stage
fractions with unique chemical and physical properties. J. Anal. Appl. Pyrolysis 2012, 93,
23
ACS Paragon Plus Environment
ACS Sustainable Chemistry & Engineering
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Page 24 of 36
129–138.
(17)
Li, H.; Xu, Q.; Xue, H.; Yan, Y. Catalytic reforming of the aqueous phase derived from
fast-pyrolysis of biomass. Renew. Energy 2009, 34 (12), 2872–2877.
(18)
Oudenhoven, S. R. G.; Westerhof, R. J. M.; Aldenkamp, N.; Brilman, D. W. F.; Kersten,
S. R. A. Demineralization of wood using wood-derived acid : Towards a selective
pyrolysis process for fuel and chemicals production. J. Anal. Appl. Pyrolysis 2013, 103,
112–118.
(19)
Teella, A.; Huber, G. W.; Ford, D. M. Separation of acetic acid from the aqueous fraction
of fast pyrolysis bio-oils using nanofiltration and reverse osmosis membranes. J. Memb.
Sci. 2011, 378 (1–2), 495–502.
(20)
Gayubo, A. G.; Aguayo, T.; Atutxa, A.; Prieto, R.; Bilbao, J. Deactivation of a HZSM-5
Zeolite Catalyst in the Transformation of the Aqueous Fraction of Biomass Pyrolysis Oil
into Hydrocarbons. Energy & Fuels 2004, 18 (6), 1640–1647.
(21)
Vispute, T. P.; Huber, G. W. Production of hydrogen, alkanes and polyols by aqueous
phase processing of wood-derived pyrolysis oils. Green Chem. 2009, 11, 1433–1445.
(22)
Vispute, T. P.; Zhang, H.; Sanna, A.; Xiao, R.; Huber, G. W. Renewable chemical
commodity feedstocks from integrated catalytic processing of pyrolysis oils. Science
2010, 330 (6008), 1222–1227.
(23)
Mukarakate, C.; Evans, R. J.; Deutch, S.; Evans, T.; Starace, A. K.; ten Dam, J.; Watson,
M. J.; Magrini, K. Reforming Biomass Derived Pyrolysis Bio-oil Aqueous Phase to Fuels.
Energy & Fuels 2017, 31, 1600–1607.
(24)
Paasikallio, V.; Kalogiannis, K.; Lappas, A.; Lehto, J.; Lehtonen, J. Catalytic Fast
Pyrolysis: Influencing Bio-Oil Quality with the Catalyst-to-Biomass Ratio. Energy
Technol. 2017, 5, 94–103.
(25)
Berger, H. L. Ultrasonic Liquid Atomization: Theory and Application, 1st ed.; Partridge
Hill Publishers: Hyde Park, NY, 1998.
(26)
Mukarakate, C.; McBrayer, J. D.; Evans, T. J.; Budhi, S.; Robichaud, D. J.; Iisa, K.; ten
Dam, J.; Watson, M. J.; Baldwin, R. M.; Nimlos, M. R. Catalytic fast pyrolysis of
biomass: the reactions of water and aromatic intermediates produces phenols. Green
Chem. 2015, 17 (8), 4217–4227.
(27)
Xu, M.; Mukarakate, C.; Iisa, K.; Budhi, S.; Menart, M.; Davidson, M.; Robichaud, D. J.;
Nimlos, M. R.; Trewyn, B. G.; Richards, R. M. Deactivation of Multilayered MFI
Nanosheet Zeolite during Upgrading of Biomass Pyrolysis Vapors. ACS Sustain. Chem.
Eng. 2017, 5, 5477–5484.
(28)
Murugappan, K.; Mukarakate, C.; Budhi, S.; Shetty, M.; Nimlos, M. R.; Roman-Leshkov,
Y. Supported molybdenum oxides as effective catalysts for the catalytic fast pyrolysis of
lignocellulosic biomass. Green Chem. 2016, 18, 5548–5557.
24
ACS Paragon Plus Environment
Page 25 of 36
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Sustainable Chemistry & Engineering
(29)
Iglesias Jiménez, E.; Pérez García, V. Relationships between organic carbon and total
organic matter in municipal solid wastes and city refuse composts. Bioresour. Technol.
1992, 41 (3), 265–272.
(30)
Bader, R. G. Use of Factors for Converting carbon or Nitrogen to Total Sedimentary
Organics. Science 1954, 120, 709–710.
(31)
Howe, D.; Westover, T.; Carpenter, D.; Santosa, D.; Emerson, R.; Deutch, S.; Starace, A.;
Kutnyakov, I.; Lukins, C. Field-to-Fuel Performance Testing of Lignocellulosic
Feedstocks: An Integrated Study of the Fast Pyrolysis–Hydrotreating Pathway. Energy &
Fuels 2015, 29, 3188–3197.
(32)
Liu, C.; Wang, H.; Karim, A. M.; Sun, J.; Wang, Y. Catalytic fast pyrolysis of
lignocellulosic biomass. Chem Soc Rev 2014, 43 (22), 7594–7623.
(33)
Graca, I.; Comparot, J.-D.; Laforge, S.; Magnoux, P.; Lopes, J. M.; Ribeiro, M. F.;
Ribeiro, F. R. Influence of Phenol Addition on the H-ZSM-5 Zeolite Catalytic Properties
during Methylcyclohexane Transformation. Energy & Fuels 2009, 23, 4224–4230.
(34)
Gilbert, C. J.; Espindola, J. S.; Conner, W. C.; Trierweiler, J. O.; Huber, G. W. The Effect
of Water on Furan Conversion over ZSM-5. Chem Soc Rev 2014, 6, 2497–2500.
(35)
Ausavasukhi, A.; Sooknoi, T.; Resasco, D. E. Catalytic deoxygenation of benzaldehyde
over gallium-modified ZSM-5 zeolite. J. Catal. 2009, 268, 68–78.
(36)
Caeiro, G.; Magnoux, P.; Lopes, J.; Ribeiro, F. R.; Menezes, S. M. C.; Costa, A.;
Cerqueira, H. Stabilization effect of phosphorus on steamed H-MFI zeolites. Appl. Catal.
A Gen. 2006, 314, 160–171.
(37)
Engtrakul, C.; Hu, M. Z.; Bischoff, B.; Jang, G. G. Surface-Enhanced Separation of Water
from Hydrocarbons: Potential Dewatering Membranes for the Catalytic Fast Pyrolysis of
Pine Biomass. Energy & Fuels 2016, 30, 8343–8348.
25
ACS Paragon Plus Environment
ACS Sustainable Chemistry & Engineering
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Page 26 of 36
Table 1 Chemical characterization of in situ CFP
aqueous stream presented in g kg-1 on a wet weight
basis
Compound
Acetic acid
64.04
Formic acid
10.09
Butanoic acid
0.13
Propanoic acid
3.27
Acrylic acid
3.89
Glycolic acid
7.37
Acetaldehyde
5.93
Glycoaldehyde
5.01
Crotonaldehyde
0.29
Tiglic aldehyde
0.26
Formaldehyde
18.35
Furfural
0.91
3-Furaldehyde
0.63
Acetone
4.90
Diacetyl
0.33
Acetylpropionyl
0.20
Hydroxyacetone
2.04
Methylolacetone
0.35
Adipic ketone
0.19
Cyclopent-2-en-1-one
3.41
2-Methylcyclopent-2-en-1-one
0.55
3-Methylcyclopent-2-en-1-one
0.29
2,3-Dimethylcyclopent-2-en-1-one
0.09
3-Methylcyclopentane-1,2-dione
0.01
2-Hydroxy-3-methylcyclopent-2-en-10.36
one
2-Acetylfuran
0.02
Phenol
1.93
Toluene
0.01
o-Cresol
0.54
m-Cresol
0.74
p-Cresol
0.40
2,5-Xylenol
0.25
2,6-Xylenol
0.01
3,4-Xylenol
0.01
2,3,5-Trimethylphenol
0.01
Guaiacol
0.16
Syringol
0.06
4-Ethylguaiacol
0.01
Creosol
0.03
2-Ethylphenol
0.22
3-Ethylphenol
0.03
Hydroquinone
0.60
Catechol
1.97
2-Methylhydroquinone
0.14
Vanillic acid
0.37
Homovanillic acid
0.12
Acetosyringone
0.03
Vanillin
0.10
Protocatechuic aldehyde
0.04
Benzofuran
0.01
2-Methyl-1-benzofuran
0.01
Methanol
21.59
Butanol
4.30
26
ACS Paragon Plus Environment
Page 27 of 36
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Sustainable Chemistry & Engineering
Table 2 Carbon and hydrogen content of CFP aqueous stream
date
Oct. 2016
Jan. 2017
wt% C
6.85 ± 0.46
7.15 ± 0.06
wt% H
10.28 ± 0.46
10.61 ± 0.1
Table 3 ICP-AES elemental analysis of
CFP aqueous stream presented in parts
per million (ppm) on a dry weight basis
Element
(ppm)
Al
Ca
Cr
Cu
Fe
K
Mg
Mn
Na
Ni
P
S
Zn
9.8
7.1
0
0
3.5
<1
1.7
0
13.7
0
9.4
304.2
0
27
ACS Paragon Plus Environment
ACS Sustainable Chemistry & Engineering
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Page 28 of 36
Table 4 Quantification of aromatic products formed before catalyst deactivation
Carbon yield (wt%)
Compound
MBMS
GCMS/FID*
Feed carbon to catalyst ratio
0.2
0.01
C2-C4 Olefins
29.0
42.8
CO
23.0
15.2
CO2
12.0
9.8
Benzene
5.18
4.2
Toluene
16.22
7.0
Xylenes
6.52
4.5
Other alkyl benzenes
2.0
Indene
0.49
0.4
Naphthalene
0.53
0.4
Methylnaphthalenes
0.60
0.6
Dimethylnaphthalenes
1.54
Anthracene
0.68
Methylanthracene
0.50
Dimethylantracene
0.29
Total aromatics yield
32.51
18.8
Total yield
96.5
87.3
28
ACS Paragon Plus Environment
Page 29 of 36
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Sustainable Chemistry & Engineering
Figure Captions
Figure 1. Distribution of carbon from catalytic fast pyrolysis of biomass. The data is taken from
references 4, 6, 7, and 8.
Figure 2. The overall composition and chemical group characterization of the CFP aqueous
stream, on a wet and dry weight basis, respectively.
Figure 3. Averaged mass spectrum recorded after volatilizing the CFP aqueous stream in the
vertical reactor at 500°C, A) with sand, B) after 10 minutes time-on-stream (TOS) with
Ga/HZSM-5, and C) after 40 minutes TOS with Ga/HZSM-5.
Figure 4. Change in product distribution from upgrading organics in CFP aqueous stream using
Ga/HZSM-5. Reaction conditions: Aqueous stream volatilization and upgrading at 500°C, 56
µL/min of aqueous stream with 600 sccm He carrier gas over 0.5 g catalyst. m/z contributing to
each grouping: single ring aromatics 78, 91, 92, 106; multi ring aromatics 128, 142, 156, 178,
192, 206; side products 144 and 158; breakthrough 60, 110, 124, 138.
Figure 5. Characterization of fresh and coked catalysts. Red, solid bars: Multi-point BET surface
areas of fresh and spent catalysts. Error bars are ± 10%. Blue, checkered bars: Weight percent
coked on each of the three coked catalyst, presented on a dry, coke-free basis. Error bars ±3%,
which is largest mass error seen with standard. Black, striped bars: Micromole of ammonia
desorbed per g of sample during ammonia temperature programmed desorption experiments for
the fresh catalyst and coked catalyst at three different F:C. Error bars are ± the standard deviation
of replicate measurements of a catalyst used previously to benchmark the instrument.
29
ACS Paragon Plus Environment
ACS Sustainable Chemistry & Engineering
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Page 30 of 36
Figure 1. Distribution of carbon from catalytic fast pyrolysis of biomass. The data was taken
from references 4, 6, 7, and 8.
30
ACS Paragon Plus Environment
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Sustainable Chemistry & Engineering
100
Acids
Aldehydes
Ketones
Aromatics
Alcohols
Water
Organics
Inorganics
80
Weight %
Page 31 of 36
60
40
20
0
Wet weight basis
Dry weight basis
Figure 2. The overall composition and chemical group characterization of the CFP aqueous
stream, on a wet and dry weight basis, respectively. Inorganic content was negligible.
31
ACS Paragon Plus Environment
ACS Sustainable Chemistry & Engineering
A) Raw aqueous phase
43
10
30
5
60
110
82
94
124 138
0
10
B) 10 minute TOS with Ga/HZSM-5
92
10 Ion signal
28
5
78
106
42
6
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Page 32 of 36
44
142
116
56
128
156
178 192
206
0
10
C) 40 minute TOS with Ga/HZSM-5
92
5
28
78
44
56
70
106
132142
108
156
94
144 158
116 128
0
20
40
60
80
100
120
m/z
140
160
180
200
Figure 3. Averaged mass spectrum recorded after volatilizing the CFP aqueous stream in the
vertical reactor, A) with sand, B) after 10 minutes time-on-stream (TOS) with Ga/HZSM-5, and
C) after 40 minutes TOS with Ga/HZSM-5. Reaction conditions: Aqueous stream volatilization
and upgrading at 500°C, 56 µL/min of aqueous stream with 600 sccm He carrier gas over 0.5g
catalyst.
32
ACS Paragon Plus Environment
Page 33 of 36
7
single ring aromatics
multi ring aromatics
side products
breakthrough
6x10
5
8
1.5x10
ion signal
4
1.0
3
2
0.5
1
10
20
30
40
50
time, minutes
60
70
ion signal from single ring aromatics
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Sustainable Chemistry & Engineering
80
Figure 4. Change in product distribution from upgrading organics in CFP aqueous stream using
Ga/HZSM-5. Reaction conditions: Aqueous stream volatilization and upgrading at 500°C, 56
µL/min of aqueous stream with 600 sccm He carrier gas over 0.5g catalyst. m/z contributing to
each grouping: single ring aromatics 78, 91, 92, 106; multi ring aromatics 128, 142, 156, 178,
192, 206; side products 144 and 158; breakthrough 60, 110, 124, 138.
33
ACS Paragon Plus Environment
ACS Sustainable Chemistry & Engineering
2
Surface area (m /g)
Acid sites (µmol)/g
Coke contents (%)
Surface area and acid sites
600
6
500
400
4
300
200
2
100
0
Coke content, wt.% (dry, coke-free basis)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Page 34 of 36
0
fresh
F:C 0.5
F:C 1.0
F:C 1.5
Figure 5. Characterization of fresh and coked catalysts. Red, solid bars: Multi-point BET surface
areas of fresh and spent catalysts. Error bars are ± 10%. Blue, checkered bars: Weight percent
coked on each of the three coked catalyst, presented on a dry, coke-free basis. Error bars ±3%,
which is largest mass error seen with standard. Black, striped bars: Micromole of ammonia
desorbed per g of sample during ammonia temperature programmed desorption experiments for
the fresh catalyst and coked catalyst at three different F:C. Error bars are ± the standard deviation
of replicate measurements of a catalyst used previously to benchmark the instrument.
34
ACS Paragon Plus Environment
Page 35 of 36
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Sustainable Chemistry & Engineering
For Table of Contents Use Only
SYNOPSIS: This work will enable recovery and upgrading of biogenic carbon from biomassderived aqueous streams to products, improving biorefinery economics.
35
ACS Paragon Plus Environment
ACS Sustainable Chemistry & Engineering
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
This work will enable recovery and upgrading of biogenic carbon from catalytic fast pyrolysis derived
aqueous streams to products that may improve biorefinery economics.
125x77mm (300 x 300 DPI)
ACS Paragon Plus Environment
Page 36 of 36
Документ
Категория
Без категории
Просмотров
0
Размер файла
679 Кб
Теги
acssuschemeng, 7b03344
1/--страниц
Пожаловаться на содержимое документа