Professional Documents
Culture Documents
discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/274960587
CITATIONS READS
0 29
8 authors, including:
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
All content following this page was uploaded by Guillaume Da on 14 April 2015.
Introduction Results
The environmental impacts of the transformation of cassava roots into Energy and water consumption varied widely between technologies
starch were assessed for three contrasting technologies at small (ST1, (figures 2 & 3). The large scale technology required 702kWh/t starch,
ST2) and very large scale (VLT), meaning 1-2 and 100-200t starch per day, mainly (75%) from fuel oil used for the drying operation, but was the most
respectively. The goal of the study was to assess energy and water use for efficient in terms of water use (10m3/t starch) due to the practice of water
each unit operation, so as to identify high usage hotspots. The Life Cycle recycling between unit operations. The two small scale technologies were
Assessment (LCA) framework was applied in order to test its relevance as similar in terms of electricity use (59kWh/t starch), and were able to rely on
a tool to analyze unit operations environmental performance, for process solar energy for drying, due to the small volumes of production. In contrast,
eco-engineering purposes. water consumption varied from 20 to 60m3/t starch between the two small
scale technologies due to differences in the design of the rasping and
starch recovery (extraction) operations.
Fresh Cassava Roots
100% 100%
WASHING 3.6
28.4
80% 18.3
RASPING 80% 7.4
33.0 33.2 34.7 Drying - Fuel oil
Separation - Recycled
Drying - Elec Separation
60% 540.3
EXTRACTION Separation Extraction - Recycled
Factory
Figure 1: Unit operations of Figure 2: Energy consumption by unit operations of VLT, Figure 3: Freshwater consumption by unit operations of
the cassava starch ST1 and ST2 technologies (large, small 1 and small 2 VLT, ST1 and ST2 technologies. The figures within the
extraction process. The respectively). The figures within the bars and on the bars and on the horizontal axis indicate actual water use
dotted line represents the horizontal axis indicate actual energy use per operation per operation and total water use respectively, in m3/t
selected system boundaries. and total energy use respectively, in kWh/t starch. starch.
Methods VLT
80%
60%
ST1
80%
60%
ST2
80%
60%
te rm y
Fr erre nisin orm n
hw ria g ra tion
n
Te e e oph tion
es ria hic n
la d up y
tra cu ion
at rm n
n
ss ple n
ep n
rtic al o um eple e
te an to n
te rm y
Fr erre nisin orm ion
at l a dia n
n
Te e e oph tion
es ria hic n
at co tion
ur in ot ity
N Urb lan co icity
la d up ity
tra ccu tion
at rm n
n
ss ple n
ep on
H de nge
te an tox n
att rm ity
n
Te e ph tion
es ria hic n
y
tra ccu on
ate rm n
n
ss ple n
ep n
n
it
ult ar ec icit
it
T Io r f atio
Fr est op atio
M er e cidif tio
a co tio
W sfo atio
M r de tio
Fo l de tio
il d tio
tio
H d ng
ula xid an tio
hw tria g ra tio
M er e cid tio
W sfo atio
M r de tio
Fo l de tio
tio
ra la cc icit
T Io r f atio
Fr est op atio
ula xid an tio
M r e cidif tio
a co tio
W sfo atio
M r de tio
Fo l de tio
il d tio
tio
m t fo xic
m t fo xic
ult ar ec ic
ur n d o toxic
il d ti
m t fo ic
ult ar ec ic
nd o pati
nd oc at
T Io r f at
a
a
at l a dia
in utr ica
hw l e a
a
a ple
le
in utr ica
hw l e a
a
le
le
ne cha
e x
x
ne cha
al e e ox
ric M er tox
rtic l o um ple
a
ate l a dia
in utro ica
hw l e a
a
a ple
le
ne cha
e x
tu an d o ox
n p
ic
n p
n to
at an d to
rr utr ic
n p
p
l la nd u
if
al lan cc
O ate
O ate
O a te
nd o
a
e
Drying Drying
e
Drying
lim
lim
lim
et
et
u
et
a
zo
zo
zo
at
at
C
defined as one ton of loose, dried cassava starch Centrifugation Sedimentation Sedimentation
e
es st
es st
es s
al
al
te
at a
ar
ar
ar
Separation Separation
a
Separation
Pa mic
Pa mic
Pa ic
m
ur
he
he
Ag
Ag
oc
oc
oc
ot
ot
Ph
Ph
Washing Washing
Ph
Washing
weight basis, wwb), obtained at the end of the Figure 4: Environmental impacts characterization of the transformation of cassava roots into starch for the
manufacturing process. very large scale (VLT) and two small scale (ST1, ST2) technologies.
Primary data for ST1 and ST2 factories were
obtained from the authors own measurements in
collaboration with one cassava starch factory
based in Vietnam (suburban Hanoi region) and
Results (continued) Conclusions
one based in Colombia (Cauca department). Data The LCA characterizations (figure 4), Among the three technologies assessed, markedly
for the VLT factory and data related to water and using the ReCiPe method, indicated different levels of energy and water consumption were
chemicals consumption (e.g. alum, SO2) were that the main impact contributions were identified for some unit operations, such as extraction or
based on measurements published by at the drying operation for the large root washing, which indicates the potential for technology
Piyachomkwan et al. (2005), Sriroth (1996) and scale technology, and at the extraction transfers and improvements of the overall environmental
Sriroth et al. (2000) in Thailand. The factories operation for the small scale performance of the cassava starch industry.
surveyed were representative of the standard technologies, mainly because of energy This work demonstrates the feasibility and usefulness of
technology used in the three regions of the study. use, as well as water use in the case of applying LCA concepts for integrating environmental
the most water-intensive technology. performance indicators in the design and evaluation of
transformation processes for agro-industrial products. The
Acknowledgments
environmental dimension should come in addition to other
Stakeholders of the cassava starch industry in Vietnam and Colombia, and staff and students at HUST and aspects including process yields, microbial safety and
CIAT, who made possible the data collection. The support of Le Thanh Mai from HUST regarding the electricity product quality.
www. post er session. com
mix in Vietnam is gratefully acknowledged. The authors thank CIRAD and CIAT for financial support.