Biomass and Bioenergy 19 (2000) 229–244

Bamboo: an overlooked biomass resource?

J.M.O. Scurlocka,∗ , D.C. Daytonb , B. Hamesb
a Environmental Sciences Division, Oak Ridge National Laboratory, Bethel Valley Road, P.O. Box 2008, Oak Ridge,
TN 37831-6407, USA
b National Renewable Energy Laboratory, 1617 Cole Boulevard, MS 3311, Golden, CO 80401, USA

Received 4 January 2000; accepted 16 May 2000

Abstract
Bamboo is the common term applied to a broad group (1250 species) of large woody grasses, ranging from 10 cm to
40 m in height. Already in everyday use by about 2.5 billion people, mostly for ber and food within Asia, bamboo may
have potential as a bioenergy or ber crop for niche markets, although some reports of its high productivity seem to be
exaggerated. Literature on bamboo productivity is scarce, with most reports coming from various parts of Asia. There is little
evidence overall that bamboo is signicantly more productive than many other candidate bioenergy crops, but it shares a
number of desirable fuel characteristics with certain other bioenergy feedstocks, such as low ash content and alkali index. Its
heating value is lower than many woody biomass feedstocks but higher than most agricultural residues, grasses and straws.
Although non-fuel applications of bamboo biomass may be actually more protable than energy recovery, there may also be
potential for co-production of bioenergy together with other bamboo processing. A signicant drawback is the diculty of
selective breeding, given the lack of knowledge of owering physiology. Further research is also required on propagation
techniques, establishment and stand management, and mechanized harvesting needs to be developed. Published by Elsevier
Science Ltd.

Keywords: Bamboo; Bioenergy; Biofuel; Fuel analysis; Productivity

1. Introduction — they may have been considered and dismissed 20

years ago, or they may be already in widespread use
It may be observed cynically that every 5 –10 years outside the industrialized world. Bamboo is such a
a new “magic feedstock” appears on the bioenergy candidate, carrying with it a degree of Eastern mys-
scene, a new Philosopher’s Stone about which great ticism: yet remarkably little is known about this en-
claims are made concerning its productivity, its ease tire sub-family of tall graminaceous plants, despite
of integration into existing markets for growing and its everyday utilization, mostly for ber and food, by
fuel supply, its life-cycle analysis, and so on. Some of about 2.5 billion people – over 40% of the world’s
these novel candidate feedstocks may not be so new population.
This review evaluates bamboo as a potential bio-
∗ Corresponding
energy feedstock, and tackles some of the myths and
author. Tel.: 001-865-241-5926; fax:
facts surrounding it — its “prodigious” productiv-
001-865-574-4665.
E-mail address: [email protected] (J.M.O. Scurlock). ity, its “disastrous” owering, its multiple uses, etc.

0961-9534/00/$ - see front matter Published by Elsevier Science Ltd.

PII: S 0 9 6 1 - 9 5 3 4 ( 0 0 ) 0 0 0 3 8 - 6
230 J.M.O. Scurlock et al. / Biomass and Bioenergy 19 (2000) 229–244

A limited range of bamboo species are characterized Thyrsostachys siamensis — native to Thailand.
with respect to fuel quality, and a number of research Used for construction in both rural and urban areas
recommendations are concluded. of Thailand; also cultivated for edible shoots.
Phyllostachys pubescens — sometimes described
as Phyllostachys edulis. Originally from China, where
2. What is bamboo? it occurs extensively (20; 000 km2 or 60% of total
bamboo cover); introduced to Japan in about 1750.
Bamboo is the vernacular or common term for The largest of the Phyllostachys genus, this species is
members of a particular taxonomic group of large harvested for both poles and edible shoots throughout
woody grasses (subfamily Bambusoideae, family South-East Asia. This species requires a climate with
Andropogoneae=Poaceae). Bamboos encompass 1250 precipitation of 1200 –1800 mm, mean annual temper-
◦
species within 75 genera, most of which are rela- ature of 13–20 C and monthly mean minimum tem-
tively fast-growing, attaining stand maturity within peratures not lower than freezing [4].
ve years, but owering infrequently. Dwarf bam- Phyllostachys bambusoides — native to China, but
boos may be as little as 10 cm in height, but stands extensively cultivated in Japan since 1866. The largest
of tall species may attain 15 –20 m, and the largest and most commercially valuable of this genus af-
known (e.g. Dendrocalamus giganteus) grow up to ter Phyllostachys pubescens, producing good-quality
40 m in height and 30 cm in culm (stem) diameter. wood. Hardier and more cold-tolerant than the
Bamboos are distributed mostly in the tropics, but latter [6].
occur naturally in subtropical and temperate zones of Bamboo has been neglected or ignored in the past
◦
all continents except Europe, at latitudes from 46 N by tropical foresters, who tend to concentrate on tim-
◦
to 47 S and from sea level to 4000 m elevation [1–3]. ber trees at the expense of traditional multi-purpose
Asia accounts for about 1000 species, covering an woody species such as bamboo and rattan [1]. Lit-
area of over 180; 000 km2 (the size of Missouri, half erature on the dynamics and productivity of natural
the size of Germany, or about 2% of US total land bamboo stands is meagre, and reports from plantation
area). Most of this comprises natural stands of native stands are almost non-existent [5]. Bamboo has been
species rather than plantations or introductions. China used for handicrafts and building material in India and
alone has about 300 species in 44 genera, occupying China for thousands of years, yet its potential contri-
33; 000 km2 or 3% of the country’s total forest area bution to sustainable natural resource management has
[4]. Another major bamboo-producing country is In- only recently been recognized. Unfortunately, most
dia, with 130 species covering 96; 000 km2 or about bamboo is harvested from forest stands at a rate which
13% of the total forested area [5]. Other nations with exceeds natural growth, so current utilization is any-
signicant bamboo production and utilization include thing but sustainable [1,2].
Bangladesh, Indonesia and Thailand.
The taxonomy of bamboos is still poorly under-
stood, at least in part because of the infrequent ow- 3. Commercial applications of various bamboo
ering of many species (at intervals of 30 – 60 years). species
Major economic species include the following:
Dendrocalamus strictus — native to India. Solid Many Asian species of bamboo have strong, light
culms, of greatest economic importance in India, and exible woody stems, which lend themselves to
where only about 10 out of more than 100 bamboo applications as a construction material — one of the
species are commercially exploited. Used mostly for most notable modern uses being temporary scaolding
paper-making and construction. poles which are often seen surrounding the most mod-
Dendrocalamus asper — thought to be native to ern of high-rise buildings in Asian countries. Bamboo
Thailand. Thailand intends to propagate plantlets of utilization in South America is modest by compari-
this species since much of the present edible bamboo son, except in certain local areas where indigenous
shoot production is from natural forests and not sus- species have been used for centuries, and where some
tainable [1]. Asian bamboos have been introduced (notably an
 J.M.O. Scurlock et al. / Biomass and Bioenergy 19 (2000) 229–244 231

international project for bamboo housing in Costa Food — bamboo shoots of a number of species are
Rica). African use of bamboo is more limited and a well-known feature of Chinese and other Asian cui-
recent, since there are few native species except in sine, generally imported into the USA in canned form
Madagascar, although indigenous and introduced (one estimate suggests 30; 000 t year −1 in the early
bamboo has been used in Kenya for soil stabilization, 1990s). Exports from Taiwan are worth $50 million
construction and fuel [1] and in Tanzania for water annually, and those from Thailand $30 million, with
pipes [7]. Worldwide commercial bamboo utiliza- much of this going to meet Japanese demand.
tion is reported to be 20 million ton per annum. It is Combustion and other bioenergy applications —
unclear whether this gure represents dry weight or a preliminary literature search found no references to
(more likely) harvested weight at about 15% moisture the use of bamboo as an energy feedstock, although
content — but this number is considered unreliable anecdotal descriptions of bamboo as a fuel are com-
since about 80% of bamboo is used locally and statis- monplace. Molini and Irizarry [11] proposed the use
tics are hard to obtain. More than half of this amount of bamboo as a fuel for power generation in Puerto
is harvested and utilized by poor people in rural areas. Rico in preference to sugar cane, since its lower mois-
Total revenues from bamboo and its products were ture content at harvest obviates the need for drying,
estimated in the 1980s at $4.5 billion [1]. but they provide few data in support of their case.
Approximately 1500 commercial applications of Limited experience has been gained using de-lignied
bamboo have been identied — mostly in Asia, ex- bamboo pulp as a substrate for ethanolic fermentation
cept where noted below. They may be divided up into [12]. Early work on preparing a diesel-like fuel from
the following broad categories: bamboo culms [13] is cited by Tewari [2]; the process
Construction and reinforcing ÿbers — these in- appears to have been the pyrolysis of “black liquor”
clude agricultural and shing tools, handicrafts, musi- from bamboo pulping, but does not seem to have pro-
cal instruments, furniture, civil engineering (bridges, gressed beyond the laboratory scale [13].
scaolding poles), domestic building (house frames,
walls, window frames, roofs, interior dividers).
Paper, textiles and board — (including rayon, 4. Physiological characteristics
plywood, oriented strand board, laminated ooring).
Bamboo bers are relatively long (1.5 –3.2 mm) and Bamboos vary across a wide range of physiolog-
thus ideal for paper production [8]. Paper production ical characteristics, so the following information is
in China dates back 2000 years, whilst in India, 2.2 merely illustrative, and mostly concerns the tall bam-
million ton of bamboo per year are processed into boos of likely interest as biomass resources. In general,
pulp, making up about two-thirds of total pulp pro- bamboos can be classied as sympodial (clumped) or
duction [1,9]. At least eight North American suppli- monopodial (spreading): tropical bamboos are sym-
ers are importing and marketing tongue-and-groove podial, whereas temperate species can be of either cat-
ooring made from laminated bamboo, which is said egory [8]. Although some bamboos can adapt to vary-
to be as hard, durable and dimensionally stable as ing environments, most require relatively warm and
oak or other hardwood ooring (e.g. Plyboo America humid conditions (e.g. mean annual temperature of
◦
Inc., Kirkville, NY). Bamboo culms are sliced into at least 15 –20 C and annual precipitation of at least
strips, which are boiled to remove starch, dried, and 1000 –1500 mm).
laminated into solid boards using urea–formaldehyde Shoot buds appear as swellings on the side of the
adhesives. The boards may be treated with preserva- underground rhizomes, which generally occupy the
tives such as boric acid, before or after laminating, top 30 –50 cm of soil and may spread for tens of me-
or both, and a darker amber color may be produced tres. With the onset of warm spring weather, the bud
by pressure-steaming the bamboo to carbonize it. lengthens and develops into a compact upright shoot
Although the adhesive tends to emit formaldehyde which forms a sharp point and penetrates the ground
for a long time after production, the amount of urea– surface. After emergence there is little radial growth
formaldehyde resin in a laminated product is much of the shoot, with “growth” taking the form of massive
less than in a panel board product [10]. elongation of internodes, as much as 0:5 m week −1 in
232 J.M.O. Scurlock et al. / Biomass and Bioenergy 19 (2000) 229–244

the case of tall bamboos, until the shoot is approxi- example, P. bambusoides collected from China in
mately of the same height as the rest of the stand. At 1926 and grown at Byron, Georgia, USA, began to
this point, sheaths are shed and leafy branches emerge ower in 1989, but recovered with vigorous vegetative
from the internodes near the top of the shoot. Further growth by 1993 [16]. The threat of catastrophic ow-
growth over the next few years comprises thickening ering need not pose an economic problem for bamboo
of the walls of the stem and increases in wood density growers, as long as uneven-aged propagation material
[14]. Compared with other tall owering plants, this is maintained, and entire stands are replaced before
pattern of growth may give a misleading impression of they approach owering age.
high productivity — in fact, all that may be observed Unlike other highly productive members of
is an (albeit rapid) re-distribution of previously stored the Andropogoneae=Poaceae family (e.g. sugar
reserves. cane, switchgrass, miscanthus), the entire bamboo
The reported leaf area index of mature stands is sub-family (Bambusoideae) lacks the C4 photosyn-
generally high, e.g. 8.02 for P. pubescens and 11.6 thetic pathway and anatomy [17]. In the absence of
for P. bambusoides [4,15]. Such a dense canopy may this feature (which may lead to higher water-use and
absorb up to 95% of incident solar radiation [4]. nutrient-use eciencies under high light conditions),
The reported pattern of leaf fall for most bamboos the maximum possible productivity of bamboos such
is semi-deciduous, with leaves shed at the end of the as P. pubescens is unlikely to greatly exceed that of
growing season (e.g. Bambusa sp. growing at Auburn, other bioenergy crops with C3 photosynthesis such as
Alabama [14]) or during the following growing sea- short-rotation willow coppice.
son. P. pubescens renews its leaves on a 2-year cy-
cle, with most leaf fall taking place in the spring, be-
ginning with the second growing season after shoot 5. Fuel characteristics
emergence. This biennial pattern of leaf renewal may
be re ected in a biennial pattern of shoot emergence, 5.1. Description of samples
with alternating “good” and “poor” years of new shoot
production [3,4]. Nine bamboo samples were obtained from
The pattern of owering in bamboos varies with G.R. Lovell, Bamboo Curator for the US Depart-
species. A few (e.g. Bambusa atra, native to the An- ment of Agriculture Agricultural Research Service
daman Islands of the eastern Indian Ocean) are known (USDA-ARS) at the University of Georgia Plant
to ower frequently, even annually. Others, such as Genetic Resources Conservation Unit in Grin,
Bambusa vulgaris, ower a few culms at a time. How- Georgia. Three dierent ages of three dierent bam-
ever, the majority — for example, Dendrocalamus boo species were sampled for analysis to determine
strictus — display gregarious owering, whereby an the fuel characteristics of the bamboo samples in
entire clump at one location produces owers and then relation to other biomass feedstocks with potential
dies back over the course of 2–3 years. This happens for electricity and fuel production, and to determine
typically every 30 – 40 years (more than 60 years in how the fuel properties change as the plant grows
some cases), and is therefore observed quite rarely, so under carefully controlled conditions. What follows
the physiology of owering is still little understood. is a description of each of the bamboos analyzed. All
Clear-cutting does not appear to halt stand mortality, bamboo stands were established on Norfolk Loamy
◦
although some species can be induced to develop new ne sand soils with slight slope (¡ 5 ), following
shoots for a year or two before nally dying altogether. their transfer from Savannah, Georgia, around 1978.
As a consequence of the rarity of owering, the tax- Phyllostachys nigra, cultivar Henon (Fig. 1): These
onomy of bamboos is still confused and based largely rhizomes originated in Nagasaki, Japan and were ob-
on vegetative features such as leaf anatomy, arrange- tained in 1909 from a Japanese bamboo grower by
ment of vascular bundles in leaf sheaths and culms, W.D. Hills, an agricultural explorer. This monopodial
etc. [2,3]. bamboo grows to a maximum height of 20 m with an
A few genera, e.g. Phyllostachys and Arundinaria, 11 cm culm diameter, and is a much larger cultivar of
have been reported to recover after owering [2]. For the common black bamboo species whose culms do
 J.M.O. Scurlock et al. / Biomass and Bioenergy 19 (2000) 229–244 233

Fig. 1. Stand (clonal repository) of Phyllostachys nigra, cultivar Henon, at the USDA-ARS= University of Georgia Plant Genetic Resources
Conservation Unit in Grin, Georgia, USA. Maximum culm diameter is about 11 cm, although this size has not been achieved in the
stand pictured. Photograph taken August 1999 by G.R. Lovell, Bamboo Curator.

◦
not turn black. It is cold hardy down to −20 C, has is named after D. Bisset, the Superintendent of the
dense dark green foliage and a gray waxy lm on many USDA Plant Introduction Station at Savannah, Geor-
culms that makes this species a striking ornamental gia, from 1924 to 1957. 1, 2, and 4-year-old samples
plant. It also has straight, strong culms that make it of this bamboo species were obtained.
useful for construction purposes. 1, 2, and 4.5-year-old
samples of this bamboo species were obtained. 5.2. Sample preparation
Phyllostachys bambusoides, cultivar White Crook-
stem: Rhizomes were collected near Yeunguk in the The following preparation was performed to obtain
Lungtau Mountains of the Peoples Republic of China samples which were representative of each bamboo
in 1926 by F.A. McClure, an agricultural explorer stand. Each of the nine bamboo samples comprised
with the US Bureau of Plant Industry. This monopo- 6 –9 representative culms with 2–3 nodes. Culms were
dial bamboo grows to a maximum height of 11 m with split into pieces 1.3 cm wide using a knife or wood
◦
a 5 cm culm diameter, and is cold hardy to −18 C. chisel, and cut into blocks about 7.5 cm long to facil-
This cultivar diers from the common form in hav- itate milling. Approximately one-half of each sample
ing culms that are curved in a serpentine manner at was processed whole, meaning that the nodes and in-
the base. It also has deposits of white powder on the ternodes were milled together. The other half of the
culms which persists and often completely obscures sample was separated into node samples and internode
the green color in older culms. 1, 2, and 4.5-year-old samples prior to milling. The node samples included
samples of this bamboo species were obtained. the complete node plus less than one inch of culm on
Phyllostachys bissetii: Rhizomes were collected either side of the node. Each internode sample con-
from Szechuan Province, Peoples Republic of China, tained all of the pieces of culm not used in the node
in 1941 by John Tee-Van of the New York Zoo- sample. Overall, 27 analyses were performed, repre-
logical Society. This bamboo grows to a maximum senting whole bamboo, nodes and internodes for each
height of 7 m with a 2.5 cm culm diameter. With a of the nine samples listed earlier. Milling and homog-
◦
cold hardiness of −23 C, it is one of the most cold enization was performed to assure that the chemical
hardy samples cultivated by the USDA-ARS, and is analysis was performed on material which represented
the rst to send up shoots in the spring — it is also the that average of each sample. Samples were milled us-
fastest growing and the most invasive. This bamboo ing a Standard Whiley knife mill with a 2 mm screen.
234 J.M.O. Scurlock et al. / Biomass and Bioenergy 19 (2000) 229–244

Material ¡ 2 mm was sieved for 5 min using a 20 and temperature that reduce thermal NOx formation.
mesh sieve. Any material ¿ 20 mesh was milled until Likewise, the sulfur content of the bamboo samples
it passed the 20 mesh screen. The ¡ 20 mesh material is very low compared to coal, and like many woody
was homogenized by cone and quartering four times. biomass materials, is also lower than many herba-
ceous biomass feedstocks, grasses, and straws (Tables
5.3. Fuel analyses 1 and 2). Changes in the %C, %H, %N, and %S do
not correlate with the dierent bamboo samples or
The proximate, ultimate and elemental ash analy- maturity of the samples.
ses for the nine bamboo samples collected for this There does appear to be some variation in the
study are summarized in Table 1. The moisture con- composition of the ash between each of the bam-
tents of the samples ranged from 8 to 23% with no boo samples and with the maturity of the individual
correlation between the amount of moisture and the species. The silicon content of the P. nigra samples
sample or the age of the sample. The moisture con- appears to increase with the maturity of the sample
tent of the P. nigra and P. bissetii samples increased and while the overall ash content of the 4.5-year-old
with age of the harvested plant, while the moisture P. nigra is lower than the younger samples, the sil-
content of the P. bambusoides sample decreased as icon content of the ash is greater by a factor of 4.
the age of the harvested plant increased. The ash con- A similar increase in ash silicon content is also ob-
tents of all of the bamboo samples were 1% or less served for the P. bisetii samples, and a more gradual
and no correlation between ash content and bamboo increase for the P. bambusoides samples. The alu-
sample or age of sample was apparent. Compared to minum content of all the bamboo samples is very
other biomass feedstocks, these ash contents are com- similar and in all cases the amount of titanium in the
parable to what is found in woody biomass materi- samples was below the detection limits of the mea-
als. Many herbaceous biomass materials, grasses and surements. The iron oxide contents of the bamboo
straws have higher ash contents. The volatiles content samples are also quite low. The amount of alkaline
of the bamboo samples ranged from 63 to 75% with earth oxides, CaO and MgO, in the bamboo samples
no correlation between volatiles content and bamboo appears to increase as the bamboo plants mature.
sample or sample age. The remainder of the bamboo This is evident for all three samples. Conversely,
samples was xed carbon. The xed carbon content the amount of alkali metal oxides appears to de-
of the bamboo samples ranged from 12 to 17% of crease as the bamboo plants mature. The amount of
the samples “as received”. All of the bamboo samples K2 O in the bamboo ashes ranges from 30 to 50%,
had very similar higher heating values, ranging from however, the overall ash contents of the bamboo
19.09-19.57 GJ t−1 on a dry basis (Tables 1 and 2). samples is low and this amounts to only 0.2– 0.6%
These heating values are comparable to, but slightly K2 O in the dry samples. The phosphorus content of
lower than, most woody biomass feedstocks and the bamboo ashes is relatively high, however, given
higher than most grasses and straws [18]. The carbon the low ash contents of the samples the phospho-
and hydrogen contents of the bamboo samples were rus content of the dry materials ranges form 0.1 to
all very similar, at about 52% C and 5% H. The vari- 0.2%. The percentage of phosphorus in the bam-
ation in the nitrogen content of the bamboo samples boo ash appears to increase as the bamboo plants
was larger, ranging from 0.2 to 0.5%. From a com- mature.
bustion perspective, this is a very low N content and The low ash and chlorine contents of the bam-
would be benecial in terms of minimal fuel-bound boo samples make them attractive for use in biomass
nitrogen conversion to NOx if bamboo were used as combustion applications for electricity production. As
a boiler fuel. If bamboo were used as a co-ring fuel, a feedstock for combustion applications, the potas-
this could oset some of the input fuel-bound N, sium contents of these bamboo samples are also quite
because the bamboo samples tested here had lower low. In fact, the alkali index of these samples (de-
N contents than many coals used for power produc- ned as kg alkali oxide per GJ energy content) ranges
tion. Co-ring with bamboo may also have other from 0.1 to 0.3, generally below the empirical limit of
NOx benets caused by dierences in ame structure 0.17– 0.34 kg GJ−1 known to cause adverse fouling
 J.M.O. Scurlock et al. / Biomass and Bioenergy 19 (2000) 229–244 235

Table 1
Fuel analyses for selected bamboo samples

Phyllostachys nigra Phyllostachys bambusoides Phyllostachys bissetii

1 year 2 year 4.5 year 1 year 2 year 4.5 year 1 year 2 year 4 year
Proximate analysis (% as received)

Moisture 8:42 8:79 13:62 22:61 12:92 9:54 8:52 10:66 21:97
Ash 0:86 0:87 0:41 0:66 0:84 0:53 1:14 0:78 0:9
Volatiles 73:94 73:66 72:27 62:93 70:51 75:55 73:18 72:24 64:99
Fixed carbon 16:78 16:68 13:7 13:8 15:73 14:38 17:16 16:32 12:14

Higher heating value (GJ t −1 dry basis)

19.57 19.48 19.27 19.49 19.53 19.09 19.42 19.50 19.51

Alkali index (kg alkali oxide GJ−1 )

0:21 0:24 0:08 0:20 0:21 0:09 0:30 0:19 0:20

Ultimate analysis (% dry matter)

C 51:89 51:19 51:39 52:28 51:84 50:85 51:22 51:7 51:07
H 5:21 5:29 5:25 5:09 5:18 5:40 4:90 5:00 4:51
N 0:4 0:29 0:21 0:59 0:6 0:38 0:55 0:3 0:32
S 0:04 0:03 0:03 0:05 0:05 0:04 0:05 0:03 0:05
Cl 0:19 0:14 0:05 0:06 0:06 0:04 0:07 0:03 0:06
Ash 0:94 0:95 0:47 0:85 0:96 0:59 1:25 0:87 1:15
O∗ 41:52 42:12 42:61 41:10 41:33 42:75 41:98 42:11 42:91

Ash elemental∗∗ (% dry matter)

SiO2 0:045 0:046 0:077 0:039 0:047 0:045 0:057 0:061 0:158
Al2 O3 0:004 0:005 0:002 0:004 0:004 0:003 0:005 0:004 0:011
TiO2 0:000 0:000 0:000 0:000 0:000 0:000 0:000 0:000 0:000
Fe2 O3 0:019 0:002 0:002 0:004 0:002 0:002 0:002 0:002 0:006
CaO 0:015 0:018 0:024 0:018 0:021 0:037 0:029 0:024 0:073
MgO 0:021 0:023 0:031 0:027 0:030 0:075 0:045 0:041 0:076
Na2 O 0:003 0:001 0:003 0:003 0:000 0:001 0:002 0:001 0:002
K2 O 0:415 0:475 0:158 0:378 0:411 0:180 0:578 0:371 0:382
P2 O5 0:130 0:105 0:097 0:125 0:180 0:123 0:155 0:107 0:181
SO3 0:028 0:024 0:011 0:018 0:022 0:019 0:052 0:025 0:033
Cl 0:034 0:027 0:003 0:009 0:008 0:002 0:025 0:006 0:007
CO2 0:072 0:112 0:016 0:060 0:062 0:028 0:129 0:094 0:056
Undetermined 0:155 0:110 0:046 0:166 0:174 0:075 0:171 0:133 0:166
◦
Ash elemental (% ash, 600 C)
SiO2 4:74 4:89 16:30 4:61 4:87 7:64 4:52 7:04 13:74
Al2 O3 0:40 0:56 0:42 0:42 0:42 0:48 0:43 0:49 0:99
TiO2 ¡ 0:01 ¡ 0:01 ¡ 0:01 ¡ 0:01 ¡ 0:01 ¡ 0:01 ¡ 0:01 ¡ 0:01 ¡ 0:01
Fe2 O3 2:06 0:23 0:42 0:43 0:16 0:39 0:16 0:25 0:48
CaO 1:59 1:90 5:06 2:17 2:17 6:34 2:33 2:74 6:35
MgO 2:22 2:44 6:66 3:12 3:10 12:70 3:62 4:71 6:62
Na2 O 0:27 0:15 0:71 0:35 ¡ 0:01 0:21 0:16 0:10 0:19
K2 O 44:20 50:00 33:60 44:50 42:80 30:50 46:20 42:70 33:20
P2 O5 13:80 11:02 20:70 14:69 18:74 20:80 12:36 12:25 15:71
SO3 2:94 2:51 2:32 2:13 2:27 3:18 4:17 2:93 2:83
Cl 3:63 2:87 0:59 1:10 0:84 0:26 2:03 0:65 0:58
CO2 7:68 11:83 3:42 7:00 6:47 4:77 10:35 10:83 4:89
Undetermined 16:47 11:60 9:80 19:48 18:16 12:73 13:67 15:31 14:42
236 J.M.O. Scurlock et al. / Biomass and Bioenergy 19 (2000) 229–244

Table 2
Comparison of selected fuel properties of bamboo with other bioenergy crops

Fuel property Bamboo “Bamboo” “Bamboo” Miscanthus Switchgrass

(range of three (species not (species not (Miscanthus x (Panicum
Phyllostachys species)a given)b given)c giganteus)d virgatum)e

Gross heating 19.1–19.6 15:85 18:96 17.1–19.4 18.3

value
(dry; GJ t −1 )
Moisture content (%) 8.4 –22.6 10.4 2.94 15 15
(samples as (at harvest?) (after drying?) (at harvest) (at harvest)
received after
shipping)
Ash content (%) ¡ 1:0 3:98 2:04 1.5 – 4.5 4.5 –5.8
Sulfur content (%) 0.03– 0.05 N=A 0:15 0:1 0.12
a Present study.
b Ref. [8, Table 3:8, p. 30].
c C.E. Pugh (1981) laboratory report for Northeast Louisiana University, cited in [38].
d Data for miscanthus from [39].
e Data for switchgrass from [40]. See also [18] for a review of bioenergy fuel characteristics.

and slagging in combustion systems [19,20]. Chlorine # 8491 was analyzed with each batch of bamboo
in biomass has been shown to increase the volatility samples, as a quality-assurance reference for the de-
of alkali metals during combustion [21], and the low termination of extractives, lignin and carbohydrates.
chlorine content of the bamboo samples suggests that Aqueous and ethanolic extractions were performed
the potassium that is present is unlikely to be volatile, in duplicate on the milled, homogenized samples.
and therefore not problematic in terms of ash depo- Each extractives-free sample was analyzed in dupli-
sition. The low chlorine content of the bamboo sam- cate for lignin and carbohydrate content, giving a
ples means that burning them is unlikely to enhance total of four separate determinations for each bamboo
high temperature corrosion in biomass combustion sample.
systems. Table 3 shows the complete wet chemical analy-
sis of the 27 bamboo samples. Total mass closure for
5.4. Wet chemical analysis all samples was greater than 99%, indicating that the
analysis is complete and no signicant biomass frac-
All bamboo samples were analyzed using American tions were excluded. The values obtained are similar
Society for Testing and Materials (ASTM) standard to those reported by Higuchi [22] for samples of ma-
methods for whole biomass analysis, as listed below: ture bamboo (Table 4). Direct measurement of the
ASTM E1690 Test Method for the Determination protein content of these samples was not performed,
of Ethanol Extractives in Biomass. but the protein content can be estimated from the ni-
ASTM E1721 Test Method for the Determination trogen elemental analysis [23] and varies only slightly
of Acid Insoluble Residue in Biomass. with age in the range 1.4 –3.8%. Interference from
ASTM E1756 Test Method for the Determination acid-insoluble protein may introduce a slight addi-
of Total Solids in Biomass. tional error in the lignin analysis. The lignin values
ASTM E1757 Preparation of Biomass for Compo- of 25 –30% place bamboo at the high end of the nor-
sitional Analysis. mal range of 11–27% reported for non-woody biomass
ASTM E1758 Standard Method for Determination [24], and more closely resemble the ranges reported
of Carbohydrates in Biomass by High-Performance for softwoods (24 –37%) and hardwoods (17–30%)
Liquid Chromatography. [25,26]. This would suggest that bamboo should have
US National Institute of Standards and Technol- similar physical properties and uses to conventional
ogy (NIST) Bagasse Standard Reference Material softwoods and hardwoods [27]. Indeed, its high lignin
 J.M.O. Scurlock et al. / Biomass and Bioenergy 19 (2000) 229–244 237

Table 3
Wet chemical analysis of selected bamboo samples

Sample Fraction (%) (%) Ethanol (%) Total (%) (%) (%) (%) (%) Total
ash extractives lignin glucan xylan mannan arabinan galactan

P. bissetii 1 yr Whole 1.2 1.4 28.2 42.2 24.0 0.5 1.1 1.1 99.7
P. bissetii 1 yr Nodes 0.6 2.1 28.5 41.9 24.5 0.0 2.1 2.1 101.8
P. bissetii 1 yr Internodes 1.4 1.6 27.9 43.7 24.6 0.5 1.0 1.0 101.8
P. bissetii 2 yrs Whole 1.0 3.1 27.9 40.7 23.6 0.6 1.1 1.2 99.2
P. bissetii 2 yrs Nodes 0.7 1.9 28.4 42.3 25.7 0.0 2.0 2.1 103.1
P. bissetii 2 yrs Internodes 0.9 1.5 28.3 43.6 25.9 0.5 1.0 1.0 102.7
P. bissetii 4 yrs Whole 1.2 2.2 27.1 41.9 22.8 0.8 1.2 1.2 98.3
P. bissetii 4 yrs Nodes 0.8 3.0 27.0 41.3 23.3 0.0 2.0 2.0 102.3
P. bissetii 4 yrs Internodes 1.1 1.4 26.2 47.2 24.1 0.4 1.1 1.1 102.7
Average 1.0 2.0 27.7 42.8 24.3 0.4 1.4 1.4 101.3
Standard deviation 0.3 0.6 0.8 1.9 1.0 0.3 0.5 0.5 1.7

P. bambusoides 1 yr Whole 0.9 2.0 26.8 41.8 24.1 0.8 1.2 1.3 99.0
P. bambusoides 1 yr Nodes 0.3 1.8 27.0 40.9 23.9 0.7 1.3 1.3 97.3
P. bambusoides 1 yr Internodes 1.3 1.6 26.2 44.5 26.0 0.5 1.1 1.1 102.3
P. bambusoides 2 yrs Whole 0.9 1.9 26.8 40.8 24.1 0.7 1.2 1.2 97.5
P. bambusoides 2 yrs Nodes 0.4 1.6 27.4 40.3 24.9 0.8 1.3 1.4 98.0
P. bambusoides 2 yrs Internodes 1.3 1.3 26.2 44.4 27.0 0.4 1.2 1.2 102.9
P. bambusoides 4.5 yrs Whole 0.6 1.1 25.5 44.9 24.0 0.6 1.3 1.3 99.3
P. bambusoides 4.5 yrs Nodes 0.2 1.5 25.3 44.0 23.3 0.8 1.3 1.4 97.8
P. bambusoides 4.5 yrs Internodes 0.8 1.1 24.3 48.1 24.0 0.5 1.1 1.1 101.0
Average 0.7 1.5 26.2 43.3 24.6 0.6 1.2 1.3 99.4
Standard deviation 0.4 0.3 1.0 2.6 1.2 0.2 0.1 0.1 2.1

P. nigra 1 yr Whole 0.8 2.3 28.3 41.3 24.4 0.5 1.4 1.4 100.4
P. nigra 1 yr Nodes 0.2 1.7 27.9 40.7 23.1 0.7 1.4 1.5 97.2
P. nigra 1 yr Internodes 0.1 1.7 28.6 41.3 24.0 0.5 1.4 1.4 99.0
P. nigra 2 yrs Whole 0.8 2.2 27.8 42.7 24.2 0.6 1.2 1.2 100.7
P. nigra 2 yrs Nodes 0.9 1.4 27.9 42.8 24.6 0.6 1.3 1.3 100.7
P. nigra 2 yrs Internodes 0.6 1.6 27.5 43.1 23.4 0.5 1.2 1.2 99.0
P. nigra 4.5 yrs Whole 0.4 1.9 25.2 46.4 23.0 0.6 1.1 1.1 99.7
P. nigra 4.5 yrs Nodes 0.4 1.6 25.5 46.0 24.0 0.6 1.2 1.2 100.6
P. nigra 4.5 yrs Internodes 0.2 1.5 25.5 46.2 22.9 0.5 1.1 1.2 99.1
Average 0.5 1.8 27.1 43.4 23.7 0.6 1.3 1.3 99.6
Standard deviation 0.3 0.3 1.3 2.3 0.6 0.1 0.1 0.1 1.2

NIST reference material #8491 (bagasse)

Average 1.8 4.4 23.1 41.7 24.7 0.7 1.6 1.6 99.5
Standard deviation 0.3 1.0 0.5 1.2 0.9 0.1 0.2 0.2 2.0

content contributes to the relatively high heating value The average and standard deviation of 27 indepen-
of bamboo, and its structural rigidity makes it a valu- dent analyses of the standard reference material are
able building material. Similarly, the glucan contents shown at the foot of Table 3. The reproducibility and
of 40 – 48% compare to the reported cellulose content total mass closure indicate the precision that can rea-
of softwoods (40 –52%) and hardwoods (38–56%). sonably be expected from these methods of analysis.
Cellulose contents in this range make bamboo a use- Dierences between bamboo samples smaller than the
ful feedstock for paper production and for processes standard deviation of the reference material will not
which convert cellulose to fuels, chemicals and other be signicant. The average total mass closure for the
bio-based materials. NIST bagasse reference samples was 99:5 ± 1:2%,
238 J.M.O. Scurlock et al. / Biomass and Bioenergy 19 (2000) 229–244

Table 4
Previous chemical analysis of bamboo [22]

(%) (%) Ethanol–benzene (%) (%) (%) Total

ash extractives Lignin Cellulose Pentosan

Species
Phyllostachys heterocycla 1.3 4.6 26.1 49.1 27.7 108.82
P. nigra 2.0 3.4 23.8 42.3 24.1 95.61
P. reticulata 1.9 3.4 25.3 41.0 26.5 98.15

indicating that no signicant biomass fractions were fewer roots and buds present, do not propagate well
excluded from the analysis. [14]. Adamson et al. [9] used existing farm machin-
The lignin, glucan and xylan contents all showed ery (ploughs, etc.) to propagate bamboo from nursery
variances with age comparable to the 1–5% reported elds. Rows of 2-year-old bamboo were cut at a height
for softwood [27]. There is a slight trend toward lower of about 2 feet (60 cm), the rhizomes dug up with a
lignin with increased age (Table 3), although it is only turning plough, and then the plants separated by hand.
signicant in the 4 and 4.5 year old bamboo samples. February (early spring in the US South-East) was the
This trend is conrmed by a slight increase in the inde- optimum time for propagating. Propagating from en-
pendently determined glucan content. These changes tire plants (about six nodes of stem material and a foot
occur in both the node and internode fractions, be- (30 cm) of attached rhizome) was more successful
ing only slightly more pronounced in the internode than propagating from rhizomes alone. Fertilization
segments. However, the magnitude of the composi- and weed control in nursery elds and young stands is
tional dierences is small and is not expected to result desirable, but weed control with herbicides was prefer-
in signicant variations in heating value or physical able to disturbing the soil [14]. Adamson et al. [9] re-
properties. ported low multiplication rates (annual doubling): 1 h
Dierences between the composition of node and of 2-year-old nursery eld was only enough to plant
internode fractions are also small, 1– 4%, and com- a 4-h stand. Traditional methods for propagation also
pare to those reported for the natural variance in wood include branch and stem cuttings and seedlings, where
[27]. The small dierences in chemical composition available [1].
between node and internode samples for these bam- In India, tissue culture techniques have been used
boo samples suggest that neither the number of nodes for the rapid large-scale production of planting stocks
nor the length of internode segments would be crit- from selected vigorous seedlings of Dendrocalamus
ical to the utilization of bamboo for energy conver- strictus [28]. Cultured seedlings of several bamboo
sion, chemical production or as a building material species have also been induced to ower and seed
(Table 3). precociously using cytokinins and control of pH,
suggesting that hybridization may be possible in the
future. Nadgauda et al. [28] review 34 published
English-language reports of tissue culture in a variety
6. Stand establishment and management
of bamboo species.
Since most bamboos ower very infrequently, plan- Seven or eight years may be required after estab-
tation stands are usually established from vegetative lishing a stand of tall bamboo before the culms are
material rather than from seedlings, although the lat- large enough and suciently “woody” to be harvested.
ter approach is a possibility. Rhizome propagation Some weaker shoots may die back before reaching ma-
involves cutting 30 –50 cm lengths of rhizomes, 1–2 turity. The ultimate stand height may not be achieved
years old with nodes and buds present — however, low until 15 –20 years [14]. Adamson et al. [9] agree that
multiplication rates make it dicult to establish large yields are likely to be low during an initial establish-
plantations by this method [8]. Older rhizomes, with ment period of up to 8 years.
 J.M.O. Scurlock et al. / Biomass and Bioenergy 19 (2000) 229–244 239

Avoidance or control of owering may comprise an Above-ground biomass of this mature stand (age not
important but yet undeveloped aspect of management, given) was 138 t ha−1 (117 t ha−1 in culms).
since this generally results in the die-back of entire Above-ground woody biomass increments aver-
stands. A few rhizome buds may remain viable, but aging 7:7 t ha−1 year −1 (8.8 and 6:6 t ha−1 year −1
the re-establishment of a new stand from surviving in 1986 and 1987, respectively) are reported for
rhizomes or from seed would be required over the a semi-natural lowland stand of Phyllostachys
course of following 5 –10 years. Bamboo re-growing pubescens in Zhejiang Province, China (latitude
◦
after seeding requires thinning and selection of the 30 N; elevation 100 m; mean annual temperature
◦
strongest shoots [14]. and rainfall, 16 C and 1800 mm). This mature stand,
Recommended fertilizer levels based on Dendro- which was subject to harvesting of poles over 8 years
calamus trials in India (cited in [8]) are N : P : K old, had an above-ground biomass of 56 t ha−1 , and a
100 : 50 : 50 kg ha−1 , similar to the levels suggested maximum height of 15 –20 m [4].
below by Sturkie et al. [14]. Fertilization at the be- ANPP of Thyrsostachys siamensis in Thailand
◦
ginning of the growing season can result in up to a (14 N; elevation 60 m; mean annual temperature
◦
three-fold increase in above-ground productivity [8]. and rainfall, 28 C and 950 mm) ranged from 1.6
−1
Harvesting of traditionally grown bamboo is to 8:1 t ha year −1 according to site quality (mean
un-mechanized and labor-intensive. Traditionally, 4:3 t ha−1 year −1 ). Above-ground biomass of stands
only the most mature poles (about 8 years old) are varied from 11 to 54 t ha−1 , with a mean value of
removed selectively, although it is feasible to har- 32 t ha−1 , and mean stand height ranged from 5.5 to
vest a range of ages of poles, sorting them later on 9.9 m [30].
the basis of density. Research in India suggests that A tropical plantation of Bambusa bambos in south-
◦
clear-cutting does not signicantly damage bamboo ern India (approx. 11 N; elevation 540 m; mean
◦
stands [8], so it may be possible to use machinery annual temperature and rainfall, 31 C and 600 mm),
such as modied sugar-cane harvesters. grown from tissue culture stock with fertilization
Some problems have been reported in restricting the and irrigation, attained an above-ground biomass
spread of hardy monopodial bamboos [14]. of 286 t ha−1 in its rst 6 years. This is equiva-
lent to an average above-ground productivity of
47 t ha−1 year −1 [5].
◦
7. Productivity of bamboo in Asia and South At a site in northern India (approx. 25 N; mean
◦
America annual temperature and rainfall, 26 C and 830
mm), total NPP of Dendrocalamus strictus sa-
As was suggested in Section 2 above, literature on vanna was estimated at 15.8–19:3 t ha−1 year −1 .
bamboo productivity is scarce, with most reports com- However, about half of total ecosystem NPP was
ing from various parts of Asia. All the data reviewed accounted for by below-ground turnover, and after
here are reported as oven-dry biomass, and are sum- accounting for other woody and herbaceous species,
marized in Table 5. above-ground woody biomass increment of bamboo
Above-ground net primary productivity (ANPP), was only 2:2 t ha−1 year −1 for a mature stand and
including leaf turnover, for Phyllostachys bambu- 1:1 t ha−1 year −1 for a recently harvested stand [31].
◦
soides in central Japan (latitude 35 N) was reported at Veblen et al. [32] estimated the biomass and pro-
24:6 t ha−1 year −1 over a 6-year period [15]. Woody ductivity of a stand of montane Chusquea culeou
◦
biomass increment, excluding leaf turnover, was in Chile (latitude 39:5 S, elevation 700 m, annual
15:5 t ha−1 year −1 . Above-ground biomass of this rainfall 4000 mm). ANPP was 10 –11 t ha−1 year −1 ,
mature stand (which last owered 16 –22 years previ- and above-ground biomass about 160 t ha−1 , with
ously) was 131 t ha−1 , of which 112 t ha−1 comprised a maximum stand height of about 9 m. Lower pro-
culms. A related study on Phyllostachys pubescens ductivities and biomass are reported for montane
in Japan found ANPP of 18:1 t ha−1 year −1 ; al- forest understorey bamboos in central China (approx.
though total NPP was high (33 t ha−1 year −1 ), allo-
◦
32 N) by Taylor and Qin [33]. Here, the most pro-
cation to roots and rhizomes was substantial [29]. ductive species, Fargesia spathacea (now renamed
240 J.M.O. Scurlock et al. / Biomass and Bioenergy 19 (2000) 229–244

Table 5
Reported maximum above-ground productivity of bamboo stands

Location (latitude, Mean annual Total ANPPa Above-ground Special stand Ref.
where available) temperature (maximum wood features
and gure productivity (elevation,
precipitation, reported); (average management)
where t ha−1 year −1 reported); t ha−1 and other
available year −1 remarks
◦
Southern India 31 C; 600 mm 47.0 N=A highland (540 m);
◦
(11 N) fertilized and [5]
irrigated

Central Japan N=A 24.6 15.5 [15]

◦
(35 N)

Japan N=A 18.1 N=A [29]

Georgia, USA N=A 9.1 “sustainable

◦
(32 N) yield” [9]

central Chile 4000 mm 10.5 6.2 montane (700 m) [32]

◦
(40 S)
◦
Zhejiang, China 16 C;1800 mm 10.5 7.7 managed by [4]
◦
(30 N) harvesting

Alabama, USA N=A 7.4 Probably [14]

◦
(32 N) fertilized
◦ ◦
Thailand (14 N) 28 C; 950 8.1 N=A
mm [30]
◦
northern India 26 C; 830 7.7 2.2
◦
(25 N) mm [31]

central China 1200 mm 4.5 3.1 montane (2750 m);

◦
(32 N) ANPPa [33]
gure corrected
for grazing
a ANPP = above-ground net primary productivity. All data are reported as oven-dry biomass.

F. robusta), growing at 2500 –3000 m elevation with bambusoides in Japan, and 18 t ha−1 year −1 for
annual rainfall of 1200 mm, had ANPP of about managed trial plots of Dendrocalamus strictus in
3:6 t ha−1 year −1 (or 4:5 t ha−1 year −1 , corrected India.
for grazing losses to giant pandas). Above-ground
biomass was nearly 24 t ha−1 , with maximum stand
height about 2.5 –3.0 m. 8. Limited experience with bamboo in the USA
El Bassam [8] cites above-ground (air-dry)
yields of native bamboo stands ranging from 1.5 Since there is a paucity of literature on US experi-
–2:5 t ha−1 year −1 for Thyrsostachys siamensis in ence with bamboos, most of the available information
Thailand to 10 –14 t ha−1 year −1 for Phyllostachys has been gathered together in one section covering
history, applications and productivity.
 J.M.O. Scurlock et al. / Biomass and Bioenergy 19 (2000) 229–244 241

According to Sturkie et al. [14] and Adamson Sturkie et al. [14] report a cumulative above-ground
et al. [9] “oriental” bamboo was introduced to the “yield” of 121 t ha−1 dry matter (48 t acre-1) for
US in about 1860, and the former US Department of Phyllostachys rubromarginata at Auburn, Alabama
◦
Agriculture (USDA) collection at Savannah, Georgia, (latitude about 32 N), for stands aged 14 –20 years,
dates back to 1919. After a brief resurgence of inter- or an average of 6.1–8:6 t ha−1 year −1 . These gures
est in bamboo as a potential bioenergy crop during exclude leaf biomass, with branches accounting for
the 1970s, the Savannah station was closed around 14% of above-ground dry matter. This cumulative
1978. The national bamboo collection (including 21 yield is also reported as up to 54 short tons per acre
species of Phyllostachys and 17 other genera) is (1 short ton=2000 lb), which may account for some of
presently maintained at Byron, Georgia, with funding the misleading gures quoted for bamboo productiv-
from the Agricultural Experiment Station at Grin, ity in the USA. The maximum productivity recorded
Georgia (now the USDA Plant Genetic Resources for any natural ecosystem is 88 t ha−1 year −1 (for
Conservation Unit). The USDA concluded that many oodplain grass in the Amazon [34]), which is close
other crops would out-produce bamboo under the ma- to the maximum gure reported for crops (about 88–
jority of US conditions, and that most of the land not 91 t ha−1 year −1 for above-ground yield of fertilized
already used for food production in southern states sugar cane [35,36]). Both of these examples represent
such as Florida was unsuitable for bamboo. However, roughly the maximum possible plant productivity with
large-scale trials were never conducted [16]. the advantages of C4 photosynthesis and no nutrient
Interest in bamboo as a crop in Alabama dates back limitations, and are equivalent to about 39 – 40 short
to the rise of angling in the 1930s and attempts by tons (35 –36 imperial or long tons) per acre per year.
local growers to meet the demand for shing poles The most productive bamboo stand reported by
[14]. Trial plots of bamboo were initiated at the Agri- Sturkie et al. [14] was probably fertilized; recom-
cultural Experiment Station, Auburn, in 1933, using mended applications, assumed to be annual, were
nursery materials from the US Plant Introduction N:P:K 80 : 35 : 50 lb acre−1 (91 : 40 : 57 kg ha−1 ).
Station at Savannah, Georgia. Development of the Other Phyllostachys species achieved lower annual
south-eastern paper and pulp industry in the late 1950s yields (2.0 –5:8 t ha−1 year −1 , or 0.8–2:3 t acre−1
and 1960s led to more extensive USDA trials at Cam- year −1 , including up to 21% in the form of branches).
◦
den, Alabama (latitude 32 N), but these were discon- In trials at Camden, Alabama (1959 –1966, plots
tinued in 1965. Bamboos found in Alabama include 65 m × 65 m, plants spaced 2:4 m × 2:4 m), Phyl-
Bambusa sp. from the Gulf Coast, Phyllostachys sp. lostachys bambusoides out-performed Pinus taeda
(introduced “Japanese bamboo”) and the sole native (Loblolly pine) by a factor of nearly 2 : 1, in
Arundinaria sp. (sometimes described as two species; terms of above-ground dry matter production over
A. gigantea and A. tecta) which used to occur abun- the rst 7 years of growth (stems and branches,
dantly throughout the US South-East in stands known bark-free in the case of P. taeda). The bam-
as “cane-brakes”. Some hardy bamboos are reported boo produced 12:4 t acre−1 (1:8 t acre−1 year −1 or
4:4 t ha−1 year −1 ) compared with pine productivity of
◦
to grow as far North as Pennsylvania (40 –42 N), and
Phyllostachys bambusoides is also known to grow in 7:1 t acre−1 (1:0 t acre−1 year −1 or 2:5 t ha−1 year −1 ).
◦
central northern California (40 N). Adamson et al. [9] found that shoot production of
In the early 1960s, the USDA evaluated the pa- various Phyllostachys species at Savannah, Georgia
◦
per pulping characteristics of 21 bamboo species, in- (latitude 32 N), varied greatly between years in re-
cluding ve tropical sympodial (clumped) bamboos sponse to environmental factors. They estimated that
◦
from Puerto Rico (latitude 18 N). All were found to sustained yields of over 4 short tons acre−1 year −1
be suitable for paper-making using either the Kraft or (9:1 t ha−1 year −1 ) were attainable for established
the Raitt processes, with long bers (1.4 –2.3 mm) [9]. stands of P. bambusoides. In trials with cutting this
Some experience with applications of bamboo also species on 5, 6 and 7-year cycles, yields were 3.9 –
appears to have been obtained at the Engineering Ex- 4.4 short tons acre−1 year −1 (9.0 –9:9 t ha−1 year −1 ),
periment Station, Clemson College, South Carolina although one 3-year cycle produced 5.9 short tons
(quoted in [14]). acre−1 year −1 (13:5 t ha−1 year −1 ). Newly estab-
242 J.M.O. Scurlock et al. / Biomass and Bioenergy 19 (2000) 229–244

lished stands responded positively to N fertilizer, ence has been gained outside Asia since the 1960s
especially when applied in early spring. in selecting well-adapted species=genotypes, or in
estimating the productivity of the crop under US
9. Potential for new applications in the Americas conditions. Reports of high productivity appear to
emanate from small-scale trials in the 1950s–1960s
Assuming that markets can be identied and estab- in Alabama, but there is little evidence overall that
lished for domestically grown bamboo products (e.g. bamboo is signicantly more productive than other
fresh or canned bamboo shoots, bamboo poles, bam- candidate bioenergy crops. Bamboo has good ber
boo ooring) there may be some commercial poten- quality for paper-making, and it shares a number of
tial for certain bamboo species (notably Phyllostachys desirable fuel characteristics with certain other bioen-
spp.) in the warmer, wetter parts of the US. As indi- ergy feedstocks, such as low ash content and alkali
cated by previous USDA interest, the US South-East index. Its heating value is lower than many woody
is a likely candidate region, although the prevalence of biomass feedstocks but higher than most agricultural
frost in the southern Appalachian mountains may set a residues, grasses and straws. In common with certain
northern limit to the commercial growing of bamboo. other potential energy crops, non-fuel applications
Widespread growing for bioenergy production is un- of bamboo biomass may be actually more protable
likely, though, both for reasons of climate limitation than energy recovery, although these other applica-
and because bamboo productivity is no greater than, tions might be used as a means of supplementing the
and in many cases less than, other candidate bioenergy income of bamboo bioenergy growers. Alternatively,
crops. bioenergy might provide a market for utilization of
Based upon the gures available from overseas, as waste materials from thinning=harvesting of bamboo
well as the limited trials conducted in the US, inten- stands grown for other purposes. Drawbacks include
sively managed bamboo stands with fertilization may the near-impossibility of selective breeding, given the
be capable of producing over 10 tonnes ha−1 year −1 poor state of knowledge on bamboo reproduction.
(4 metric ton per acre per year), but a combination Further research is clearly required on propagation
of variable soils and uneven management could sig- techniques to increase multiplication rates, although
nicantly reduce this gure over large plantation ar- recent studies in India appear promising. Large-scale
eas. Nevertheless, bamboo appears to share a number trials are needed in order to develop recommen-
of desirable fuel characteristics with other bioenergy dations for cost-eective establishment and stand
feedstocks, including a low moisture content at the management, and mechanized harvesting needs to be
time of harvest (Table 2). The waste products from developed for countries with high labor costs. The
bamboo processing for other purposes may also be economics of bamboo production require thorough
suitable for energy recovery using established biomass evaluation, both for single-use and multiple-product
feedstock handling equipment. scenarios.
Elsewhere in the America, notably Central America,
there has been interest expressed in building upon lo-
cal agricultural experience with management of bam-
boo stands, in order to provide fuel for conventional Acknowledgements
steam-generation power plants [37]. Under these con-
ditions, the local skills base may prove more signi- Jonathan Scurlock was supported by the US Depart-
cant in determining project success than the potential ment of Energy under contract DE-AC05-96OR22464
productivity of the bioenergy feedstock. with Lockheed Martin Energy Research Corporation.
David Dayton and Bonnie Hames thank the Solar
Thermal, Biomass Power, and Hydrogen Technolo-
10. Conclusions gies Division of the US Department of Energy, Oce
of Energy Eciency and Renewable Energy, and Dr.
Bamboo may indeed have potential as a bioenergy Richard Bain and Dr. Ralph Overend of the National
or ber crop for niche markets, but little experi- Renewable Laboratory, for support of this work.
 J.M.O. Scurlock et al. / Biomass and Bioenergy 19 (2000) 229–244 243

References [16] Lovell GR, University of Georgia, USA, personal

communication.
[1] Williams JT, Dranseld J, Ganapathy PM, Liese W, Nor SM, [17] Jones CA. C4 grasses and cereals: growth, development and
Sastry CB, editors. Research needs for bamboo and rattan to stress response. New York: Wiley, 1985. 419 pp.
the year 2000. Tropical Tree Crops Program, International [18] Nordin A. Chemical elemental characteristics of biomass
Fund for Agricultural Research=International Network for fuels. Biomass and Bioenergy 1994;6:339–47.
Bamboo and Rattan, Singapore, 1991. 81 pp. [19] Miles TR, Miles Jr. TR, Baxter LL, Bryers RW, Jenkins BM,
[2] Tewari DN. A monograph on bamboo. Dehra Dun, India: Oden LL. Boiler deposits from ring biomass fuels. Biomass
International Book Distributors, 1992. 498 pp. and Bioenergy 1996;10:125–38.
[20] Baxter LL, Miles TR, Miles Jr. TR, Jenkins BM, Milne T,
[3] Liese W. Bamboos: biology, silvics, properties, utilization.
Dayton D, Bryers RW, Oden LL. The behavior of inorganic
Eschborn, Germany: Deutsche Gesellschaft fur Technische
materials in biomass-red power boilers: eld and laboratory
Zusammenarbeit (GTZ), 1985. 132 pp.
experiences. Fuel Processing Technology 1998;54:47–78.
[4] Qiu GX, Shen YK, Li DY, Wang ZW, Hudang QM, Yang [21] Dayton DC, Milne TA. Laboratory measurements of alkali
DD, Gao AX. Bamboo in sub-tropical eastern China. In: Long metal containing vapors released during biomass combustion.
SP, Jones MB, Roberts MJ, editors. Primary productivity In: Baxter L, DeSollar R, editors. Application of advanced
of grass ecosystems of the tropics and sub-tropics. London: technologies to ash-related problems in boilers. New York:
Chapman & Hall, 1992, pp. 159 –88. Plenum Press, 1996. p. 161–85.
[5] Shanmughavel P, Francis K. Above ground biomass [22] Higuchi H. Biochemical studies of lignin formation, III.
production and nutrient distribution in growing bamboo Physiologia Plantarum 1957;10:633–48.
(Bambusa bambos (L.) Voss). Biomass and Bioenergy [23] Phillips M. Journal of the Association of Ocial Agricultural
1996;10:383–91. Chemists 1939;22:422.
[6] Chao CS. A guide to bamboos grown in Britain. Royal [24] Bagby MO, Nelson GH, Helman EG, Clark TF.
Botanic Gardens, Kew, UK, 1989. 47 pp. Determination of lignin in non-wood plant ber sources.
[7] Lipangile TN. Bamboo and wooden water pipes. Ambio Tappi 1971;54:1876–8.
1987;16:299–301. [25] Fengel D, Wegener G. Wood: chemistry, ultrastructure,
[8] El Bassam N. Energy plant species: their use and impact on reactions. Berlin: Walter de Gruyter Publishers, 1984.
environment and development. London: James and James, p. 56 –9.
1998. 321 pp. [26] Dence CW. The determination of lignin. In: Lin SY, Deuce
[9] Adamson WC, White GA, Derigo HT, Hawley WO. CW, editors. Methods in lignin chemistry. Berlin: Springer,
Bamboo production research at Savannah, Georgia, 1992. p. 35–57.
1956-77. USDA-ARS-S-176. US Department of Agriculture, [27] Panshin AJ, de Zeeuw C. Textbook on wood technology:
Agricultural Research Service, Savannah, Georgia, 1978. 17 structure, identication, properties, and uses of the
pp. commercial woods of the United States and Canada. New
[10] Environmental Building News. Bamboo ooring. York: McGraw-Hill, 1980. p. 268–9.
Citation from Environmental Building News, West River [28] Nadgauda RS, John CK, Joshi MS, Parasharami VA,
Communications, Brattleboro, Vermont, USA, Vol. 6(10), Mascarenhas AF. Application of in vitro techniques for
November 1997. Forest Products Journal 1999;49(1):6. bamboo improvement. In: Chapman GP, editor. The bamboos:
Proceedings of an International Symposium, London, 25-29
[11] Molini AE, Irizarry JG. Bamboo as a renewable energy
March 1996, Linnean Society Symposium Series No. 19. San
resource. Proceedings of the First Pan American Congress
Diego: Academic Press for the Linnean Society of London,
on Energy and Second National Conference on Renewable
1997. p. 163–77.
Energy Technologies, 1–7 August, 1982, San Juan, Puerto
[29] Isagi Y, Kawahara T, Kamo K, Ito H. Net production and
Rico. Report No. CONF-8308231, Center for Energy and
carbon cycling in a bamboo Phyllostachys pubescens stand.
Environment Research, San Juan, 1983. p. 1–18.
Plant Ecology 1997;130:41–52.
[12] Ram MS, Seenayya G. Production of ethanol from straw [30] Suwannapinunt W. A study on the biomass of Thyrsostachys
and bamboo pulp by primary isolates of Clostridium siamensis (Gamble) forest at Hin-Lap, Kanchanaburi. Journal
thermocellum. World Journal of Microbiology and of Bamboo Research 1983;2:227–37.
Biotechnology 1991;7:372–8. [31] Tripathi SK, Singh KP. Productivity and nutrient cycling in
[13] Piatti L. Flussige brennsto aus bambus (liquid fuel from recently harvested and mature bamboo savannas in the dry
bamboo). Schweizer Archiv fur Angewandte Wissenschaft tropics. Journal of Applied Ecology 1994;31:109–24.
und Technik 1947;13:370–6. [32] Veblen TT, Schlegel FM, Escobar B. Dry matter production
[14] Sturkie DG, Brown VL, Watson WJ. Bamboo growing in of two species of bamboo (Chusquea culeou and C. tenui ora)
Alabama. Bulletin No. 387, USDA Agricultural Experiment in south-central Chile. Journal of Ecology 1980;68:397–404.
Station, Auburn University, Alabama, 1968. 30 pp. [33] Taylor AH, Qin ZS. Culm dynamics and dry matter
[15] Isagi Y, Kawahara T, Kamo K. Biomass and net production production of bamboos in Wolong and Tanajiahe Giant
in a bamboo Phyllostachys bambusoides stand. Ecological Panda Reserves, Sichuan, China. Journal of Applied Ecology
Research 1993;8:123–33. 1987;24:419–33.
244 J.M.O. Scurlock et al. / Biomass and Bioenergy 19 (2000) 229–244

[34] Piedade MTF, Junk WJ, Long SP. The productivity of the [38] Woods JE. Bamboo: agribusiness opportunities. Proposals
C4 grass Echinochloa polystachya on the Amazon oodplain. presented to ORNL Bioenergy Feedstock Development
Ecology 1991;72:1456–63. Program by West Wind Technology, Inc., Athens, Tennessee,
[35] Alexander AG, Allison W, Gonzalez-Molina C, Ortiz-Velez 1996.
J, Ramirez G, Velez-Santiago J, Velez A, Chu TL. Production [39] Scurlock JMO. Miscanthus: a review of European experience
of sugarcase and tropical grasses and as renewable energy with a novel energy crop. ORNL Technical Memorandum
source. Final Report (01-Jun-77 to 31-May-82) to US TM-13732, Oak Ridge National Laboratory, Oak Ridge,
Department of Energy. Center for Energy and Environment Tennessee, 1999. 18 pp.
Research, University of Puerto Rico, Rio Pedras, 1982. [40] McLaughlin SB, Samson R, Bransby D, Wiselogel A.
[36] Giamalva MJ, Clarke SJ, Stein JM. Sugarcane hybrids for Evaluating physical, chemical and energetic properties of
biomass. Biomass 1984;6:61–8. perennial grasses as biofuels. Bioenergy ’96, Proceedings of
[37] Alger J, consultant; Fielden D, Commonwealth Development the 7th National Bioenergy Conference, Nashville, Tennessee,
Corporation; Wagner V, International Bamboo Development USA, September 1996. p. 1–8.
Company; personal communications.