Bioresource Technology 223 (2017) 307–311

Contents lists available at ScienceDirect

Bioresource Technology
journal homepage: www.elsevier.com/locate/biortech

Short Communication

Biofilm-based photobioreactor absorbing water and nutrients by

capillary action
Hayato Hamano a,1, Shun Nakamura a,1,2, Jumpei Hayakawa a, Hideaki Miyashita b,c, Shigeaki Harayama a,⇑
a
Department of Biological Sciences, Faculty of Science and Engineering, Chuo University, 1-13-27 Kasuga, Bunkyo-ku, Tokyo 112-8551, Japan
b
Graduate School of Global and Environmental Studies, Kyoto University, Kyoto 606-8501, Japan
c
Graduate School of Human and Environmental Studies, Kyoto University, Kyoto 606-8501, Japan

h i g h l i g h t s

 A simple biofilm-based culture system of the unicellular green alga Pseudochoricystis ellipsoidea strain Obi was developed.
2 1
 The footprint biomass productivities achieved with this system were 8–10 g m day .
 This system requires neither irrigation nor supply of CO2-enriched air.

a r t i c l e i n f o a b s t r a c t

Article history: Cells of the unicellular green alga, ‘‘Pseudochoricystis ellipsoidea”, were uniformly spread on a cellulosic
Received 2 August 2016 sheet or on a polytetrafluoroethylene (PTFE) membrane sheet superimposed on a cellulosic sheet at a
Received in revised form 18 October 2016 density of 3.5–5.0 g dry weight per m2, and the sheet was adhered to an inverted V-shaped acrylic plate
Accepted 19 October 2016
of 10 cm in height. Several acrylic plates were placed side by side on a tray containing liquid medium at a
Available online 31 October 2016
depth of 0.6 cm, and illuminated from above with a light intensity of 300–340 lmol m 2 s 1. Water and
nutrients were supplied to cells by capillary action through the cellulosic sheet. Footprint biomass pro-
Keywords:
ductivities of cells grown in atmospheric CO2 on this photobioreactor were 8–10 g m 2 day 1. This culti-
Green alga
Biofilm-based photobioreactor
vation system is strongly energy- and labor-saving as it does not require mixing of culture fluid, irrigation
Biomass productivity of medium, and delivery of CO2-enriched air.
Capillary action Ó 2016 Elsevier Ltd. All rights reserved.

1. Introduction either by bubbling of CO2-enriched air or by bicarbonate supple-

Biofuels have attracted increasing attention as alternative bio-
fuel resources, because they have the potential to decrease CO2
emissions and improve energy security. Biofuels from microalgae,
which are called third-generation biofuels, have been declared to
have several advantages, including higher biomass/oil productivi-
ties and little competition with food crops for arable land (Zhu,
2015).
Microalgae are generally cultivated in liquid as suspended cul-
tures; for their rapid growth in such cultures, CO2 must be supplied
⇑ Corresponding author.
E-mail addresses: harayama_lab02@bio.chuo-u.ac.jp (H. Hamano), naka-
shun6@gmail.com (S. Nakamura), jhykw@kc.chuo-u.ac.jp (J. Hayakawa), miyashi-
ta.hideaki.6v@kyoto-u.ac.jp (H. Miyashita), harayama@bio.chuo-u.ac.jp (S.
Harayama).
1
Hayato Hamano and Shun Nakamura contributed equally to this work.
2
Present address: Kurita Water Industries Ltd., Nakano Central Park East, 4-10-1
Nakano, Nakano-ku, Tokyo 164-0001, Japan.
mentation. Cells in suspended cultures must also be mixed to pro-
vide uniform light exposure and sufficient gas exchange. These
requirements for the cultivation of microalgae in liquid increase
operational costs and energy consumption. Moreover, cells must
be harvested and dewatered before oil extraction, and these pro-
cesses add to the levels of energy consumption and cost (Pahl
et al., 2013).
As an alternative to suspension cultivation methods, the culti-
vation of algal cells on the surface of solid or semi-solid materials
is now becoming an increasing focus. Such cultivation methods,
commonly called ‘‘attached cultivation” or ‘‘biofilm-based cultiva-
tion,” have been classified by Berner et al. (2015) into three irriga-
tion procedures: constantly submerged, intermittently submerged,
and perfused systems. Among these systems, we are most inter-
ested in perfused systems because cells in the constantly sub-
merged systems cannot use atmospheric CO2, and the
intermittently submerged systems require energy-intensive mech-
anization. The advantages of the perfused systems can be summa-

http://dx.doi.org/10.1016/j.biortech.2016.10.088
0960-8524/Ó 2016 Elsevier Ltd. All rights reserved.
308 H. Hamano et al. / Bioresource Technology 223 (2017) 307–311
rized as follows: (i) algal cells can be harvested from a substratum
surface by scraping biofilm from the surface, and the water content
of the harvested biomass is comparable to that of those collected
by centrifugation; (ii) cells on a solid surface have direct contact
with the air and can thus absorb CO2 directly from the atmosphere,
allowing rapid growth under an atmospheric CO2 concentration;
and (iii) the surface plane of the substratum can be oriented in
the best direction to capture the maximum amount of solar radia-
tion, and inclined at the optimal angle for the highest photosyn-
thetic quantum yield; this maximizes biomass productivity per
horizontal footprint area (hereinafter referred to as ‘‘footprint bio-
mass productivity”).
Recent review articles have summarized the current state of
biofilm-cultivation technology (Berner et al., 2015; Gross et al.,
2015). The footprint biomass productivities achieved in most of
the reported perfused systems were less than 10 g m 2 day 1, but
very high productivities were reported in a few cases, as shown
below.
Cells of Botryococcus braunii FACHB 357 (B race) were collected
on a nitrate–cellulose or cellulose–acetate membrane, and the
membrane was adhered to a moist filter paper attached to a verti-
cally orientated glass plate to make a three-layered bioreactor:
(membrane)–(filter paper)–(glass plate). This reactor was continu-
ously illuminated from one side at an intensity of 500 lmol m 2 -
s 1 at 25 °C with a flow of air containing 1% (v/v) CO2. Culture
medium was supplied from above the filter paper with an appro-
priate flow rate, and the medium dripping from the lower end of
the filter paper was recirculated. The footprint productivity of
the biomass with the bioreactor was 50 g m 2 day 1 (Cheng
et al., 2013). Using essentially the same cultivation system, the
footprint biomass productivities of Acutodesmus obliquus (formerly
Scenedesmus obliquus) and Spirulina platensis were determined to
be 80 and 60 g m 2 day 1, respectively (Liu et al., 2013; Zhang
et al., 2015). Halochlorella rubescens was grown in another cultiva-
tion system called the twin-layer photobioreactor under different
light intensities and CO2 concentrations. The footprint biomass
productivity with this bioreactor under optimal conditions was
30 g m 2 day 1 (Schultze et al., 2015).
In these two cultivation systems, the production- and opera-
tional costs of the bioreactors may be high because circulation of
cultural fluids and CO2 delivery are two major elements of the
operational costs besides the costs of nutrients, water, and labor
(Rogers et al., 2014; Slade and Bauen, 2013).
In this paper, we report a simpler and less expensive biofilm-
based culture method of a unicellular green alga without the need
for the mixing of culture fluid or CO2 delivery.
2. Materials and methods
2.1. Algal strain and culture conditions
The unicellular green alga, ‘‘Pseudochoricystis ellipsoidea” strain
Obi belonging to the class Trebouxiophyceae, has been described
(Satoh et al., 2010), and deposited in NITE Biological Resource Cen-
ter (NBRC Japan) as NBRC 112353. Prior to biofilm cultivation, cells
of strain Obi were precultured in 200-mL MA5 medium (Imamura
et al., 2012) in a round-bottomed Roux-type culture flask with a
light path of 3 cm. Air containing 1% (v/v) CO2 was sparged in
the flask at a flow rate of 70 mL/min at 25 °C, and light from a
light-emitting diode (LED) source (Vegefarm VEFA140WZG, peak
wavelengths at 440, 470, 640, 660, and 740 nm; Altrader company,
Yokohama, Japan) was illuminated at a right angle to the surface of
the flask. The light intensity on the flask surface was
300 lmol photons m 2 s 1. When the optical density at 750 nm
(OD750) of the culture measured with a spectrophotometer (Hita-
chi, U-3900) exceeded 10, the cultured cells were harvested by
centrifugation (5000g, 10 min) at 4 °C, washed once with the half
strength MA5 medium (hereinafter referred to as ‘‘1/2-MA5 med-
ium”), and resuspended in 1/2-MA5 medium to an OD750 of
approximately 50.
For biofilm cultivations, 10 ml of the cell suspension was
spread, either on a nonwoven cellulosic sheet (15-cm wide by
20-cm long) or on a polytetrafluoroethylene (PTFE) membrane
(POREFLON HPW-010-30, pore size = 0.1 lm, thickness = 30 lm;
Sumitomo Electric Industries, Osaka, Japan) of the same size, to a
density between 0.10 and 0.15 g dry cell weight per sheet area
(300 cm2). Three different cellulosic sheets were used: (i) Water-
absorbing Wiper (WAW) [c  s (si-by-es), Yokohama, Japan] made
of cellulose fibers, 30% of them being cotton fibers; (ii) BenchkoteÒ
(BK) made of Whatman cellulose paper (thickness = 0.32 mm), one
side of which is laminated with a polyethylene layer (GE Health-
care); and (iii) Benchkote PlusÒ (BKP) made of Whatman cellulose
paper (thickness = 0.7 mm), one side of which is laminated with a
polyethylene layer (GE Healthcare). Cells were spread on a PTFE
membrane adhered to a cellulosic sheet, or directly on a cellulosic
sheet, and the sheet was folded in the middle and placed onto the
surface of an inverted V-shaped acrylic plate [Supplementary
material 1 (A)]. Several inverted V-shaped acrylic plates were
placed side by side onto a plastic tray containing medium at a
depth of 0.6 cm to form a ‘‘corrugated photobioreactor” [Supple-
mentary material 1 (B)], and illuminated from above with an LED
source either continuously or under a 12-h light/12-h dark cycle.
The light intensity at the top of the acrylic plates was between
300 and 340 lmol photons m 2 s 1. As water evaporated from
the substratum of the photobioreactor, the cellulosic sheets contin-
uously drew the medium upward by capillary action. Air flow rate
at the top of the acrylic plates was measured with a hot wire
anemometer (TM-4001; Tenmars, Taipei, Taiwan), and was below
0.1 m/s. The temperature of the bioreactor space was controlled
at 24 °C–25 °C.
2.2. Determination of algal growth
Dry weight of cells grown on a PTFE membrane sheet or a cel-
lulosic sheet was measured as follows: cells on the sheet were
scraped off with a scraper, and those remained on the sheet after
the scraping were washed out into 100 ml of deionized water.
The algal suspension thus obtained was centrifuged (5000g,
10 min) at 4 °C, washed twice with Milli-Q water, and dried at
105 °C for 1 day to determine the dry weight of the cells
gravimetrically.
2.3. Microscopy
Cells of strain Obi grown on different substrata were observed
by the hanging drop method using a differential interference con-
trast microscope (Olympus BX51). In this method, a PTFE mem-
brane or a BK sheet on which cells grew was cut into a small
piece with scissors, and the piece was placed on a coverslip. The
coverslip was then inverted; the cells under the coverslip were
microscopically observed with either a 20  or a 40  objective.
3. Results and discussion
3.1. Biofilm growth of strain Obi on various substrata
Dry weights of cells of strain Obi on PTFE membranes or cellu-
losic sheets of the corrugated photobioreactor were determined
after their growth for 0, 3 and 7 days, and footprint biomass densi-
ties were calculated by dividing the dry weights by the footprint of
 H. Hamano et al. / Bioresource Technology 223 (2017) 307–311
the photobioreactor (3.5  15 cm2). Subsequently, average foot-
print biomass productivities [g m 2 day 1] between day 0 and
day 3, and between day 3 and day 7 were calculated from the foot-
print biomass density values. The means and standard deviations
of the footprint biomass densities and footprint biomass productiv-
ities were obtained from the data of two to six independent exper-
iments, and are shown in Fig. 1(A). The footprint biomass
productivities between day 0 and day 3 on the PTFE membranes
adhered to three different cellulosic sheets (WAW, BKP and BK)
were similar to each other, being approximately 10 g m 2 day 1.
These productivities were higher than that of strain Obi grown in
air-sparged liquid medium which was 4.5 g m 2 day 1 (data not
shown). The footprint biomass productivities on the PTFE mem-
branes between day 3 and day 7 were, however, much lower than
those between day 0 and day 3. This decrease may have been
because of several factors: (i) After the incubation of the biofilm-
based photobioreactor for more than 3 days, small air bubbles
were generated in the space between the PTFE membrane and
the cellulosic sheet, which disrupted the supply of water and nutri-
ents from the cellulosic sheet to algal cells on the PTFE membrane.
Since this gas generation was not observed when the PTFE mem-
brane bioreactor was incubated without an inoculum of strain
Obi, the gas was expected to be oxygen generated through photo-
synthesis. (ii) The moisture content of cellulosic sheets decreased
with time, especially in the case of BK: the upper 1 cm of the PTFE
membrane adhered to the BK sheet almost dried out after incuba-
tion for 7 days.
The growth of strain Obi directly on a WAW sheet was the slow-
est among the substrata examined. We observed that, during the
incubation period, algal cells flowed through the WAW sheet into
the medium-containing tray. It is likely that cells of strain Obi
did not attach strongly onto the WAW matrix, and descended into
the tray during the incubation period. The growth of strain Obi on a
BKP sheet between day 0 and day 3 was slower than that between
day 3 and day 7. We observed that a significant number of cells
inoculated directly onto the BKP sheet penetrated inside the fabric,
where they would not receive sufficient light for photosynthesis.
The growth of strain Obi on a BK sheet was more rapid in the first
3 days than in the later period, probably because the water supply
was insufficient by the end of the incubation period.
3.2. Microscopic observation of cells grown on substrata
Cells of strain Obi grown on a PTFE membrane attached to a BK
sheet and those grown directly on a BK sheet were observed under
a microscope. On the PTFE membrane at day 3, cells grew as sev-
eral layers on most of the membrane surface [Supplementary
material 2 (A)], whereas monolayered biofilms were also observed
in some areas. It is likely that cells initially applied to the surface of
the PTFE membrane underwent cell division forming monolayer
colonies. Later, the growth transitioned from two-dimensional to
three-dimensional expansion forming multilayered colonies.
In contrast to the growth on the PTFE membrane, cells of strain
Obi grew on the BK sheet by colonizing tangled cellulosic fibers
[Supplementary material 2 (B)]. There are two potential reasons
why colonization occurred so intensively at these locations. First,
water and nutrients were supplied through the water channel
within the cellulose fibers to cells associated with the fibers. Sec-
ond, cells that did not associate with the fibers may have flowed
down in medium stored in the bottom tray.
3.3. Footprint biomass productivity under light–dark conditions
Outdoor algal cultures are subjected to diurnal changes in light
intensity. We thus examined the footprint biomass productivities
of strain Obi under 12-h light/12-h dark cycles. As shown in
309
Fig. 1(B), the footprint biomass productivities under 12-h
light/12-h dark conditions were approximately half of those under
continuous light conditions.
3.4. Footprint biomass productivities of strain Obi at two different tilt
angles of substrata
The substratum surface of the corrugated photobioreactor can
be tilted to an optimal angle to obtain an optimum photon flux
density for the highest photosynthetic quantum yield. We then
examined the footprint biomass productivities on PTFE mem-
branes attached to WAW sheets at two different tilt angles. The
footprint area of one type was 3.5  15 cm2; thus, the light dilution
rate, namely the ratio of the substratum surface area (20  15 cm2)
to the footprint area was 5.7. This ratio was used in the experi-
ments shown in Fig. 1(A) and (B), and served as a control. The foot-
print area of the other type was 2.7  15 cm2 with a light dilution
rate of 7.5. Mean footprint biomass productivities with each of the
two photobioreactor types were determined from four indepen-
dent experiments, and the results are shown in Fig. 1(C). The foot-
print biomass productivities with a light dilution rate of 7.5 were
higher than those obtained in parallel control experiments with a
light dilution rate of 5.7. However, the footprint biomass produc-
tivities in the controls were smaller than the mean footprint bio-
mass productivities from previous experiments shown in Fig. 1
(A) and (B). Thus, the mean footprint biomass productivities with
a light dilution rate of 5.7 were recalculated including all of the
previous data, and compared with those with a light dilution rate
of 7.5. In this case, the footprint biomass productivities between
day 0 and day 3 were not significantly different between the two
different light dilution rates, whereas those between day 3 and
day 7 were significantly higher with a light dilution rate of 7.5.
These results indicate that the footprint biomass productivities
with a light dilution rate of 7.5 were not inferior to those of 5.7,
and we concluded that the optimal light dilution rate under the
current experimental conditions was higher than 5.7.
3.5. Toward the development of an industrial corrugated
photobioreactor
Although the footprint biomass productivities of 8 and
10 g m 2 day 1 were attained in the corrugated photobioreactor
with the BK sheet and PTFE membrane, respectively, further
improvements are required for the photobioreactor to be commer-
cially viable. First, more economical materials should be consid-
ered for substrata on which algae grow. PTFE resin is not so
expensive, at approximately US$5 per kg; however, PTFE mem-
brane is still expensive, the cheapest being more than US$1/m2.
In this study, we also used Whatman cellulosic paper as a cellulosic
sheet. This paper is manufactured for scientific uses and very
expensive. Inexpensive unwoven cellulosic or synthetic fabrics
are manufactured by many companies. We tested some of them,
but they did not exhibit uniformity of capillary action, which is
particularly important for obtaining reproducible results. Thus,
cheap but high-quality unwoven fabrics useful as substrata of the
corrugated photobioreactor should be developed. Before the defini-
tive selection of new substratum materials, it would be preferable
to study the mechanism of the attachment of strain Obi to various
substrata. Several studies on this issue in other microalgae have
been reported, which should provide some hints for future investi-
gations. For example, in their recent review article, Katarzyna et al.
(2015) discussed the importance of the surface properties of sub-
strata for the growth of microalgae.
Methods applicable to large-scale inoculation and harvesting of
microalgal biomass on the corrugated photobioreactor should also
be established. Mechanical scraping may be a preferred method for
310 H. Hamano et al. / Bioresource Technology 223 (2017) 307–311

Fig. 1. Footprint biomass densities and productivities of strain Obi grown in atmospheric air at 25 °C on PTFE membranes or cellulosic sheets moistened with 1/2-MA5
medium. (A) Strain Obi was grown on different substrata under continuous light. The light dilution rate (LDR), namely the ratio of the substratum surface area to the footprint
area, was 5.7. PTFE: polytetrafluoroethylene; WAW: Water-absorbing Wiper; BK: BenchkoteÒ; BKP: Benchkote PlusÒ. PTFE + WAW indicates that PTFE was superimposed on
WAW. (B) Strain Obi was grown on PTFE + WAW either under 12-h light/12-h dark cycles (12-h L/12-h D) or under continuous light (24-h L) with LDR of 5.7. (C) Strain Obi
was grown on PTFE + WAW under continuous light with two different LDR (5.7 and 7.5). The results of the parallel experiments are labeled as ‘‘LDR = 7.5” and ‘‘LDR = 5.7
(Control)”. Since the footprint biomass productivities for ‘‘LDR = 5.7 (Control)” were smaller than those from previous experiments shown in (A), the mean footprint biomass
densities and productivities with LDR of 5.7 including all of the previous data are also shown as ‘‘LDR = 5.7 (All)”.

harvesting microalgal biomass since cells harvested by scraping 8 g m 2 day 1, respectively. This photobioreactor does not require
were shown to have a water content similar to that harvested by the mixing of culture fluid, the irrigation of medium, or the deliv-
centrifugation. Biomass remaining after scraping can serve as an ery of CO2-enriched air. Thus, this cultivation system is strongly
inoculum for regrowth (Johnson and Wen, 2010). energy-saving and labor-saving.

4. Conclusions Acknowledgements

A simple biofilm-based cultivation system for growing a unicel- This work was supported by the Program for the Development
lular green alga, ‘‘P. ellipsoidea” strain Obi was developed; the foot- of Innovative Total Technologies for Carbon Sequestration, Biomass
print biomass productivities on a PTFE membrane or on a cellulosic Production, and Biomass Utilization by the Ministry of Agriculture,
sheet under continuous illumination in atmospheric air were 10 or Forests, and Fisheries, Japan.
 H. Hamano et al. / Bioresource Technology 223 (2017) 307–311
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.biortech.2016.10.
088.
References
Berner, F., Heimann, K., Sheehan, M., 2015. Microalgal biofilms for biomass
production. J. Appl. Phycol. 27, 1793–1804.
Cheng, P., Li, B., Gao, L., Zhang, W., Wang, J., Liu, T., 2013. The growth, lipid and
hydrocarbon production of Botryococcus braunii with attached cultivation.
Bioresour. Technol. 138, 95–100.
Gross, M., Jarboe, D., Wen, Z., 2015. Biofilm-based algal cultivation systems. Appl.
Microbiol. Biotechnol. 99, 5781–5789.
Imamura, S., Hagiwara, D., Suzuki, F., Kurano, N., Harayama, S., 2012. Genetic
transformation of Pseudochoricystis ellipsoidea, an aliphatic hydrocarbon-
producing green alga. J. Gen. Appl. Microbiol. 58, 1–10.
Johnson, M.B., Wen, Z., 2010. Development of an attached microalgal growth system
for biofuel production. Appl. Microbiol. Biotechnol. 85, 525–534.
Katarzyna, L., Sai, G., Singh, O.A., 2015. Non-enclosure methods for non-suspended
microalgae cultivation: literature review and research needs. Renew. Sust.
Energy Rev. 42, 1418–1427.
311
Liu, T., Wang, J., Hu, Q., Cheng, P., Ji, B., Liu, J., Chen, Y., Zhang, W., Chen, X., Chen, L.,
Gao, L., Ji, C., Wang, H., 2013. Attached cultivation technology of microalgae for
efficient biomass feedstock production. Bioresour. Technol. 127, 216–222.
Pahl, S.L., Lee, A.K., Kalaitzidis, T., Ashman, P.J., Sathe, S., Lewis, D.M., 2013.
Harvesting, thickening and dewatering microalgae biomass. In: Borowitzka, M.
A., Moheimani, N.R. (Eds.), Algae for Biofuels and Energy. Springer, Netherlands,
pp. 165–185.
Rogers, J.N., Rosenberg, J.N., Guzman, B.J., Oh, V.H., Mimbela, L.E., Ghassemi, A.,
Betenbaugh, M.J., Oyler, G.A., Donohue, M.D., 2014. A critical analysis of
paddlewheel-driven raceway ponds for algal biofuel production at commercial
scales. Algal Res. 4, 76–88.
Satoh, A., Kato, M., Yamato, K., Ishibashi, M., Sekiguchi, H., Kurano, N., Miyachi, S.,
2010. Characterization of the lipid accumulation in a new microalgal species,
Pseudochoricystis ellipsoidea (Trebouxiophyceae). J. Jpn. Inst. Energy 89, 909–
913.
Schultze, L.K., Simon, M.V., Li, T., Langenbach, D., Podola, B., Melkonian, M., 2015.
High light and carbon dioxide optimize surface productivity in a Twin-Layer
biofilm photobioreactor. Algal Res. 8, 37–44.
Slade, R., Bauen, A., 2013. Micro-algae cultivation for biofuels: cost, energy balance,
environmental impacts and future prospects. Biomass Bioenergy 53, 29–38.
Zhang, L., Chen, L., Wang, J., Chen, Y., Gao, X., Zhang, Z., Liu, T., 2015. Attached
cultivation for improving the biomass productivity of Spirulina platensis.
Bioresour. Technol. 181, 136–142.
Zhu, L., 2015. Microalgal culture strategies for biofuel production: a review. Biofuels
Bioprod. Biorefin. 9, 801–814.