Subscriber access provided by NEW YORK UNIV
Article
Optimal Operation Strategy for Biohydrogen Production Ehecatl Antonio Del Río-Chanona, Pongsathorn Dechatiwongse, Dongda Zhang, Geoffrey C. Maitland, Klaus Hellgardt, Harvey Arellano-Garcia, and Vassilios S. Vassiliadis Ind. Eng. Chem. Res., Just Accepted Manuscript • DOI: 10.1021/acs.iecr.5b00612 • Publication Date (Web): 02 Jun 2015 Downloaded from http://pubs.acs.org on June 10, 2015
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.
Industrial & Engineering Chemistry Research 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 30
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
Industrial & Engineering Chemistry Research
Optimal Operation Strategy for Biohydrogen Production Ehecatl Antonio del Rio-Chanona,† Pongsathorn Dechatiwongse,‡ Dongda 1
Zhang,† Geoffrey C. Maitland,‡ Klaus Hellgardt,‡ Harvey Arellano-Garcia,¶ and Vassilios S. Vassiliadis∗,† E-mail:
[email protected] Phone: +44 (0) 1223 330142. Fax: +44 (0) 1223 334796
Abstract
2
3
Hydrogen produced by microalgae is intensively researched as a potential alterna-
4
tive to conventional energy sources. Scaling-up of the process is still an open issue, and
5
to this end accurate dynamic modeling is very important. A challenge in the devel-
6
opment of these highly nonlinear dynamic models is the estimation of the associated
7
kinetic parameters. This work presents the estimation of the parameters of a revised
8
Droop model for biohydrogen production by Cyanothece sp. ATCC 51142 in batch and
9
fed-batch reactors. The latter reactor type results in an optimal control problem in
10
which the influent concentration of nitrate is optimized which has never been consid-
11
ered previously. The kinetic model developed is demonstrated to predict experimental
12
data to a high degree of accuracy. A key contribution of this work is the prediction ∗
To whom correspondence should be addressed Department of Chemical Engineering and Biotechnology, University of Cambridge, Pembroke Street, Cambridge CB2 3RA, UK ‡ Department of Chemical Engineering, Imperial College London, South Kensington Campus, London SW7 2AZ, UK ¶ School of Engineering, University of Bradford, Richmond Road, Bradford, Yorkshire BD7 1DP, UK †
1
ACS Paragon Plus Environment
Industrial & Engineering Chemistry Research
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
13
that hydrogen productivity can achieve 3365 mL/L through an optimally controlled
14
fed-batch process, corresponding to an increase of 116% over other recently published
15
strategies.
16
Introduction
17
Global warming has been directly linked to the release of carbon dioxide (CO2 ) by burning
18
carbon-based energy resources 1 . Aside from the environmental impact, it does not seem to
19
be sensible to permanently rely on limited and non-renewable conventional fuels for energy
20
supply 2 . To reduce the production of CO2 and fulfill the increasing demands on energy
21
supply, seeking novel sustainable and environmentally friendly energy resources is actively
22
researched internationally.
23
Currently, hydrogen (H2 ) produced by microorganisms is considered one of the fuels of
24
the future with great potential for sustainability and environmental friendliness 3 . Differ-
25
ent microorganisms such as green algae, cyanobacteria and purple non-sulfur bacteria are
26
known to generate hydrogen. Chlamydomonas reinhardtii, an outstanding representative
27
of green algae, produces biohydrogen by photosynthesis and utilizes water as the hydrogen
28
source 4 . Instead of releasing CO2 , C. reinhardtii fixes CO2 for its growth, indicating that
29
the hydrogen generated in this process could be carbon-neutral 5 . Cyanobacteria are capable
30
of producing hydrogen via different metabolic pathways including both photosynthesis and
31
nitrogen-fixing pathways 6 . In particular, Cyanothece sp. ATCC 51142 attains the highest
32
hydrogen production rates compared to other natural species, and CO2 can also be chosen as
33
the carbon source in its photo-autotrophic growth period 7 . Purple non-sulfur bacteria have
34
the advantage of continuously generating hydrogen after deprivation of a nitrogen source 8 .
35
Additionally, their anoxygenic photosynthesis pathway significantly facilitates the commer-
36
cialization of biohydrogen production, as hydrogen is only generated in anaerobic conditions
37
regardless of the species of microorganisms 4,6,9 .
38
Extensive studies have been conducted to improve the productivity of hydrogen by dif2
ACS Paragon Plus Environment
Page 2 of 30
Page 3 of 30
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
Industrial & Engineering Chemistry Research
39
ferent microorganisms. Suppression of methanogenic activity for hydrogen production using
40
mixed bacterial cultures has been investigated 10 . Effects of light intensity, temperature
41
and nutrient ratio on microorganism growth and hydrogen production have also been ex-
42
amined 7,11–14 . Different processes have also been designed to extend hydrogen production
43
periods and improve biomass concentration 15,16 . Despite these efforts, there are unresolved
44
problems that still seriously prevent industrialization of biohydrogen production processes.
45
For example, in laboratory scale research, recent studies have concluded that light atten-
46
uation in photobioreactors (PBRs) can strongly limit algal and cyanobacterial growth and
47
hydrogen production 17,18 . The configuration of a PBR is also found to affect the uniformity
48
of culture mixing, light transmission and cell growth rate 19,20 . Furthermore, finding the
49
optimal operating conditions such as incident light intensity, temperature and nutrient ratio
50
is very demanding if it is carried out purely by experiments. On the other hand, recent
51
research demonstrated that different reactor operation modes such as well-mixed mode and
52
non-mixed mode can largely influence the productivity of biohydrogen production process 21 .
53
As a result, it is essential to choose the suitable reactor type and to find the optimal operating
54
conditions when scaling-up biohydrogen production processes.
55
Developing dynamic models to simulate and optimize fermentation processes have been
56
widely accepted as the most effective method to solve the previously mentioned problems.
57
A variety of dynamic models including the Monod model and the Droop model have been
58
developed and modified by previous research for biohydrogen production 22–25 . With accurate
59
and reliable dynamic models available, different reactor types and operation can be simulated,
60
and the optimization of operating conditions such as nutrient ratio and dilution rate can be
61
explored so as to reveal their maximal productive capability.
62
The use of dynamic models requires the use of state-of-the-art techniques in simulation
63
and optimization so as to select the optimal reactor type and operation. The aim of the
64
present study is twofold: to develop an accurate biohydrogen production model via parameter
65
estimation, and the dynamic optimization of hydrogen production process models. The
3
ACS Paragon Plus Environment
Industrial & Engineering Chemistry Research
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
66
species investigated in the current work is Cyanothece sp. ATCC 51142 because of its
67
distinctively high hydrogen production rate 7 .
68
The key motivation for this work is to present original contributions in the area of mod-
69
eling and optimizing fermentation processes, with particular emphasis on biohydrogen pro-
70
duction, as follows:
71
72
1. Complete dynamic model identification via the use of rigorous dynamic optimization procedures
73
2. Verification of the derived models through comparison with experimental data
74
3. Use of the derived dynamic process model to predict the optimal operation of the
75
76
77
underlying process 4. Investigate the impact on productivity of treating some traditionally constant operating parameters as controls
78
Process Modeling and Optimization
79
Model Development
80
Various dynamic models have been developed to simulate different growth phases of microor-
81
ganisms 25 . Specific to the biohydrogen production process by green algae and cyanobacteria,
82
five growth phases have been observed by recent experimental studies 15,26 (i) the lag phase,
83
(ii) the primary growth phase, (iii) the secondary growth phase, (iv) the stationary phase,
84
and (v) the decay phase 21 . In the primary growth phase, cyanobacteria grow rapidly be-
85
cause of the presence of essential nutrients, including nitrate (nitrogen source) and glycerol
86
(carbon source). Once nitrate in the culture is consumed, the cyanobacterial growth phase
87
shifts to the secondary growth phase. In this phase, cells continue growing by consuming the
88
intracellular nitrogen source accumulated in the primary growth phase, and the activity of
4
ACS Paragon Plus Environment
Page 4 of 30
Page 5 of 30
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
Industrial & Engineering Chemistry Research
89
nitrogenase is recovered because of the absence of nitrate. The nitrogen-fixing process is also
90
stimulated due to the activation of nitrogenase, which initiates the generation of hydrogen.
91
After the rapid consumption of the intracellular nitrogen source, a brief stationary phase is
92
observed which is followed by the decay phase where hydrogen is mainly generated.
93
To construct an accurate dynamic model for cyanobacterial hydrogen production, all of
94
the growth phases except the lag phase have to be considered. As most of the present dynamic
95
models are designed to simulate a single specific growth phase of microorganisms, such as
96
the stationary phase or the first growth phase 14,22–24 , modifications have to be introduced
97
with the aim to produce a complete process model that is capable of predicting all phases
98
seamlessly. By comparing the characteristics of different dynamic models, a revised Droop
99
model constructed in previous research 21 is selected as it is capable of simulating the entire
100
set of growth phases of cyanobacteria, except the lag phase. The revised Droop model is
101
presented in Equations (1a)–(1i). Further details of model selection and construction can
102
be found in 21 . As hydrogen production rates were found to be proportional to biomass
103
concentration in our experiment work 15 , the yield ratio of hydrogen to biomass, YH/X , is
104
determined as 2.34 mL·g−1 and does not need to be estimated.
5
ACS Paragon Plus Environment
Industrial & Engineering Chemistry Research
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 30
dX kq C = µmax X 1 − − µd X2 dt q KC + C
(1a)
kq C dC = −YC/X µmax X 1 − + Fin CFed dt q KC + C
(1b)
dN N = −YN/X µmax X + Fin NF ed (t) dt KN + N
(1c)
kq N C dq = Yq/X µmax − µmax 1 − q dt KN + N q KC + C
(1d)
dO N = YO/X X − Yd X2 f (O) + Fin OFed F dt KN + N
(1e)
dH = YH/X X (1 − f (O)) f (N) dt
(1f)
0.5 (N − 100)2 − (N − 100) f (N) = 0.5 0.5 (N − 100)2 + 0.1 f (O) =
(O2
Fin =
O + 0.1)0.5
0.1 720 − T
(1g)
(1h)
(1i)
105
where N is nitrate concentration (mg · L−1 ), q denotes the normalized intracellular nitrogen
106
source concentration , kq represents the normalized minimum intracellular nitrogen source
107
concentration, C is glycerol concentration (mM), O is oxygen concentration (% of oxygen
108
saturation in the solution), H is hydrogen production (mL·L−1 ), X is biomass concentration
109
(g · L−1 ). f (N), and f (O) are the switch functions to stimulate the production of hydrogen. 6
ACS Paragon Plus Environment
Page 7 of 30
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
Industrial & Engineering Chemistry Research
110
When the culture is anaerobic (O = 0.0) and the nitrate concentration is lower than its
111
threshold (N
