Accepted Manuscript

Biomass Proximate Analysis using Thermogravimetry

Roberto Garca, Consuelo Pizarro, Antonio G. Lavn, Julio L. Bueno

PII: S0960-8524(13)00601-9
DOI: http://dx.doi.org/10.1016/j.biortech.2013.03.197
Reference: BITE 11657

To appear in: Bioresource Technology

Received Date: 24 July 2012

Revised Date: 27 March 2013
Accepted Date: 29 March 2013

Please cite this article as: Garca, R., Pizarro, C., Lavn, A.G., Bueno, J.L., Biomass Proximate Analysis using
Thermogravimetry, Bioresource Technology (2013), doi: http://dx.doi.org/10.1016/j.biortech.2013.03.197

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
 Biomass Proximate Analysis using Thermogravimetry

Roberto Garca, Consuelo Pizarro(*), Antonio G. Lavn, Julio L. Bueno

Department of Chemical Engineering and Environmental Technology, Faculty of

Chemistry, University of Oviedo. Julin Clavera 8, 33006, Oviedo, Spain

ABSTRACT

This work proposes a 25 minutes-last thermogravimetric method as a tool to determine

biomass samples proximate analysis data (moisture, ash, volatile matter and fixed

carbon contents) just by direct measure of weight changes on each samples TG chart.

Compared with international standards commonly used to that aim, TG is a faster and

easier to develop technique. Obtained results were satisfactory, with AEE under 6% for

moisture and volatile matter, close to 10% for fixed carbon determination and AAD of

1.6 points for ash content.

Key words
Biomass, proximate analysis, thermogravimetry

* Representing Author: Tel +0034 985 103438; Fax +0034 985 103434
E-mail address: pizarroconsuelo@uniovi.es (C. Pizarro)
E-mail addresses: garciafroberto@uniovi.es (Roberto Garca),
pizarroconsuelo@uniovi.es (Consuelo Pizarro), agl@uniovi.es (Antonio G. Lavn),
jlbueno@uniovi.es (Julio L. Bueno)
 1. Introduction

As widely reported, biomass constitutes an important feedstock in the current world

energetic scenario (McKendry, 2002; Muthuraman et al., 2010; Jorquera et al., 2010). It

presents some environmental advantages, such as gaseous CO2 emissions neutrality (Gil

et al., 2010; Munir et al., 2009), or low NOx and SO2 emissions (Li et al., 2009; Qian et

al., 2011). In addition to this, it has social advantages too, providing a source of wealth

in rural areas, avoiding its depopulation (Bahng et al., 2009). Finally it is cheap and

easy to produce. In that way, biomass is an autonomous resource, which partly avoids

dependence on fossil fuels, which are produced in only a limited number of countries.

In addition to this, the low price of raw materials and the development of biomass-

consuming energy systems have made them economically competitive with traditional

fossil fuels.

Because of that, biomass appears as an important role-playing fuel in several national

and international policies (Gaska and Wandrasz, 2008; Rosendahl et al., 2007), such as

the European Union White Papers on energy saving, or the PER (Renewable Energy

Plan) in Spain (Lapuerta et al., 2004).

As can be seen, the future of biomass energy conversion appears to be quite optimistic,

but its energetic use also presents some problems, the most important one being related

to its own nature, even though it is a highly heterogeneous fuel, with plenty of different

origins, from a huge variety of industrial wastes to wood transformation industry-

residues or energy crops. Taking this into account, biomass characterization is required,

to reliably predict its behaviour as a fuel. When considering biomass thermal

conversion, proximate analysis is one of the most important characterization methods.

This consists in determining moisture, ash, volatile matter and fixed carbon contents of

the raw biofuel. These values are essential ascertain moisture, volatile matter and fixed
carbon affect both on the combustion behaviour and the plant design. In that way, high

moisture values decrease the combustion yield, while high volatile matter/fixed carbon

ratios are related with the fuels reactivity. On the other hand, ash deeply influences the

transport, handling and management costs of the process. It is also influential in

corrosion and slag formation. Traditionally, these measures are developed following

different national and international normative, such as ASTM E-871 for moisture,

ASTM E-830, D-1102 or UNE-EN 14775 for ash or ASTM E-872 and ASTM E-1755

for volatile matter determination (Demirbas, 2004; Khalil et al., 2008). Fixed carbon is

usually determined by difference. All those methods are time-consuming and tedious

(Karatepe and Kckbayrak, 1993), and the success of the operation is highly

influenced by the operators skills. In that way, a fast, simple, reliable and highly

accurate method for routine tests, would be desirable. Since all the above mentioned

standards basically establish heating a sample under different specific conditions

(Mayoral et al., 2001), and a weight difference determination, then a thermobalance

and a conveniently adapted thermogravimetric study appears as an effective tool

(Warne, 1991), resulting in both time-saving, from several hours to just a few minutes

in each experiment (Sadek and Herrell, 1984) and sample quantity reduction, as it

requires matter weight in the range of milligrams (Beamish, 1994).

As is well known, thermal analysis can be defined as a group of methods for measuring

the property of a substance when subjected to a controlled temperature program. It is a

highly developed technique with many different uses when applied to biomass. Some

authors have used it to study the thermal behaviour of biomass and fuel blends in both

oxidative and inactive atmospheres (Ghetti et al., 1996; Varol et al., 2010; Wilson et al.,

2010). Others used this technique as a tool to thermally characterize fuels and ashes, by

studying its melting behaviour or structural changes (Biswas et al., 2011; Miranda et al.,
2011, Ross et al., 2008) while others used TG as a tool to develop kinetic modelization

(Ramajo-Escalera et al., 2006; Wilson et al., 2011). However, but not so many works

have been developed for biomass proximate analysis, whilst, on the other hand,

thermogravimetry is often used in that way for coals (Ottaway, 1982; Slaghuis and

Raijmakers, 2004).

In this context, the aim of this work is to determinate if TG analysis can be used as an

effective tool to approach ultimate analysis data of biomass fuels and propose a method,

based on modified-coal methods to reliably and accurately obtain these data.

2. Materials and methods.

Firstly, thirteen biomass samples were grinded and sieved to 500 m to guarantee its

homogeneity in a proximate analysis routine by using ASTM standards E 871, E 1755

and E872; for moisture, ash and volatile matter respectively. In addition to this, fixed

carbon content was calculated by difference using the balance:

% FC = 100 (% Ash + %VM ) [1]

where %FC, %Ash and %VM respectively mean the mass percentages of fixed carbon,

ash and volatile matter of the raw sample.

All these procedures were detailed and referenced in our previous works (Garca et al.,

2012). After this routine every sample was tested using a Perkin Elmer STA 6000

thermobalance, using 10 and 20 mg of sample, and 40 ml/min flow of flue gas, for both

nitrogen and air. Samples were selected all around Spain trying to track every possible

biomass origin, commercial fuels, agri-food industry wastes, forest wastes, energy crops

and cereals (Fernndez et al., 2012) with the aim of obtaining a general method, suitable

for a wide range of biomass fuels, with different characteristics and compositions.
The bibliographical review search for methods developed for coals showed two basic

groups, some simpler ones, which consist in a chosen continuous temperature ramp path

between room temperature and the final set-point (TR1 and SP1, as detailed in table 1).

In this group we may include works proposed by Mayoral et al., 2001(MAY3), Sadek

and Herrell, 1984) (SDK1) or (Lapuerta et al., 2004)(LAP). On the other hand, there are

some other proceedings consisting in continuous heating (TR1, 2 or 3 depending on the

number of this steps) combined with intermediate dwellings (DT1, 2 and 3) to reach

different intermediate or final set-point (SP1, 2 or 3). In this group we can include some

other Mayoral proposed methods (MAY1, MAY2 and MAY4 ), Ottaway(OTT)

(Ottaway, 1982), Sadek and Herrel (SDK2), Karatepe and Kckbayrak, 1993(KAR),

Warne (WAR) (Warne, 1991), and Beamish (BEA) (Beamish, 1994). Table 1

summarizes the conditions required in each of these works, with each temperature ramp

(TR), intermediate and final setpoints (SP) and dwelling times (DT) required when each

set-point is reached. Most of them are conducted in inactive atmosphere (using argon,

helium or nitrogen depending on the author) to measure moisture, volatile matter and

fixed carbon and with a final combustion time in oxidative environment (oxygen or air),

when set-point is reached, to measure ashes. On the other hand, Lapuertas work is

completely carried out in oxidative atmosphere. Some of these works explain how

relevant data must be obtained from Thermogravimetry. TG diagrams commonly

present a number of slopes, each kinetics showing different phase changes. A first,

normally, small one represents moisture release due to drying, occuring at a temperature

under 150C (Zheng and Kozinski, 2000). In the range between 200 and 900 C, a huge

mass loss is seen. In the context of different phenomena that can be observed. Thus,

between 200 and 600 C the biggest mass loss occurs due to the release of CO2 and

CH4, these gases coming from hemicelluloses (190-320 C), cellulose (280-400 C) and
lignin (320-450 C) decomposition (Strezov, 2004), and a later chemically bonded CO2

and chemically formed water release (450-600 C). Finally, from 600 to proximately

900 C mass loss rate decreases, due to the evolution of carbon-containing species (COx,

CxHy and tars) and char oxidation until constant weight is reached (Haykiri-Ama,

2003).

Five methods originally developed, and described in the literature, for coals and cokes

proximate analysis determination, were tested. In this work once observed which of

them presents the best results, they were slightly adapted to biomass samples, aiming to

reduce the experimental time without impoverishing the obtained results. The tested

methods were LAP, MAY1, SDK and KAR. In addition to this a variation of OTT,

called OTT* consisting in a second dwelling of five minutes at 550C, was used.

Obtained results are compared with proximate analaysis data determined using

international standards, provided in Table 2 (Garca et al., 2012), experimental errors

are calculated and average absolute error values shown in Figure 1.

The error criteria used are the average experimental error (AEE) and the average bias

error (ABE) commonly used by several authors in this field (Ahmaruzzaman, 2008;

Majumder et al., 2008). In addition to this, the average absolute deviation (AAD) is also

used. Those criteria are defined as follows:

[2]

[3]

[4]

As can be seen in figure 1, the KAR method presents the best results for both moisture

and volatile matter determination, and so, because of this will be considered as a base
method to develop future variations, trying to improve accuracy in both of this

parameters and ash and fixed carbon contents.

3. Results and discussion

Once KARs method has been chosen and carried out as the base one, some changes are

developed therein, aiming to obtain a more suitable method for biomass proximate

analysis data determination. In that way, biomass is reportedly far more reactive than

coal, and so, faster heating ramps can be used without losing complete conversion

during combustion. In that way, some new methods are tested, proving that the best

results were found when beginning with a heating ramp of 50C/min, from room

temperature until obtaining an isothermal 120C for 3 minutes, then a new 100C/min

heating ramp is programmed until 950C.When this point is reached, a cooling process

with -100C/min ramp starts until reaching 450C. Until this set-point, the process is

developed using nitrogen, to guarantee a non oxidative environment, but, when 450C is

reached, flue gas is changed for air. Then, a new 100C/min heating ramp begins until

800C, which provides better results than previous methods, and is isothermally

maintained for 3 minutes, when the program is finished, totalling 25 minutes per

experiment, enabling two experiments per hour, including cooling and stabilization of

the experimental equipment, which entails a great time saving compared to several

hours taken for the moisture and ash determination. This method is also more

convenient than ASTM standard, and so, volatile matter determination implies working

with a furnace at 950C, involving physical risk for the operator when introducing and

withdrawing samples. This proposed method requires direct measure of moisture,

volatile matter, fixed carbon and ash by difference from slope to slope. Volatile matter
and fixed carbon values are thus obtained by applying Beamishs correction which

means adding values obtained with the next formulae to the measured value.

[5]

[6]

where VM, FC, VMTG and FCTG are, respectively, the obtained volatile matter and fixed

carbon and those measured in TG, and M and A, moisture and ash content read on the

TG profile. These corrections are proven to slightly improve the results.

Finally, results obtained for each of the samples are detailed in Table 3, and the average

values are compared (named as NEW MET) with the coal-developed methods in Figure

1. As can be seen, moisture and volatile matter experimental errors are satisfactory, with

AEE under 6 %, resulting in AAD of 0.5 and 4.4 points, respectively. On the other

hand, fixed carbon and especially ash results highly improve the values obtained using

coal-developed methods, offering values close to 10 and 50 %, respectively. Taking into

account that fixed carbon determination is obtained empirically by difference, with the

entailing precision and accuracy limits, an 11% average error and 2.0 average absolute

deviation points, can be considered as quite acceptable result. Regarding ash

determination, this involves a problem already referred to by other authors (Mayoral et

al., 2001). As can be seen in Table 3, there exists a big difference in measured EE

depending on the considered sample. In that way, beetroot pellets or wheat grain present

accurate results, with an experimental error close to 5%, while the same error in some

other samples such as hazelnut shell or both brands of wood pellets approaches 90 %.

As can be observed with the absolute deviation, those values are not high at all; in fact

the AAD is just over that of the moisture, presenting a 1.6 % value. However, as

expected, ash values are really low in biomass, mainly in woody samples. Low absolute
errors imply really high deviations, in relative bases, like AEE, may be due to the

formation of highly specific weight oxides when the oxidative environment is reached.

4. Conclusions

This work proposes a suitable method to determine proximate analysis data of biomass.

Obtained results, with average experimental errors under 6 % for moisture and volatile

matter and close to 10 % for fixed carbon greatly improve coal-developed methods, for

the same tested samples. Ash determination entails average absolute deviation of 1.6

points, but the low expected values imply higher experimental errors. Other important

advantages are time saving and simplicity, as this method requires 25 minutes to obtain

all four data sets while only moisture or ash determination take several hours each,

using the standard normative.

ACKNOWLEDGMENTS

PSE-ARFRISOL, Ref. PS-120000-2005-1, is a science-technology project qualified as

Strategic by 2004-07 Spanish National Research Plan, Development and Diffusion, co-

financed by European Regional Development Funds and the Spanish Ministry of

Science and Education. We wish to acknowledge all members of the PSE-ARFRISOL

partnership for their cooperation.

We would also like to thank a number of companies such as Pellets Asturias, Factor

Verde, Molygrasa, Dibiosur, Enfosur, Acciona, Nutral Arroceras Dorado, CarsanBio,

Parque Verde, Vinos Viadecanes, Cooperativa Agrcola de Cangas del Narcea, Vino

de la Tierra de Cangas, Gebio, Bioenerga Aragonesa, Cafs el Gallego, Biomasas

Herrero, Ecowarm and Carbones Lamuo for their generous collaboration in supplying

most of the required samples.

References

Sorted alphabetically
 Ahmaruzzaman, M., [34] 2008. Proximate analyses and predicting HHV of chars

obtained from cocracking of petroleum vacuum residue with coal, plastics and biomass.

Bioresour. Technol. 99, 50435050.

Bahng, M.K., Mukarakate, C., Robichaud, D.J., Nimlos, M.R., [8] 2009. Current

technologies for analysis of biomass thermo-chemical processing: A review. Anal

Chim Acta. 651, 117-138.

Beamish, B.B., [18] 1994. Proximate analysis of New Zealand and Australian coals

by thermogravimetry. New Zealand J. Geol. Geoph. 37, 387-392.

Biswas, A.K., Umeki, K., Yang, W., Blasiak, W., [22] 2011. Change of pyrolysis

characteristics and structure of woody biomass due to steam explosion pretreatment.

Fuel Process. Technol. 92, 18491854.

Demirbas, A., [12] 2004. Combustion characteristics of different biomass fuels.

Prog. Energy Comb. Sci. 30, 219-230.

Fernndez, R.G., Garca, C.P., Lavn, A.G., Bueno, J.L., [30] 2012. Study of main

combustion characteristics for biomass fuels used in boilers. Fuel Process. Technol.

103, 1626.

Garca, R., Pizarro, C., Lavn, A.G., Bueno, J.L., [29] 2012. Characterization of

Spanish biomass wastes for energy use. Bioresour. Technol. 103, 249-258.

Gaska, K., Wandrasz, A.J., [9] 2008. Mathematical modeling of biomass fuels

formation process. Waste manag. 28, 973-985.

Ghetti, P., Ricca, L., Angelini, L., [19] 1996. Thermal analysis of biomass and

corresponding pyrolysis products. Fuel. 75(5), 565573.

Gil, M.V., Oulego, P., Casal, M.D., Pevida, C., Pis, J.J., Rubiera, F., [4] 2010.

Mechanical durability and combustion characteristics of pellets from biomass blends.

Bioresour. Technol. 101, 8859-8867.

 Haykiri-Ama, H., [33] 2003. Combustion characteristics of different biomass

materials. Energy Convers. Manag. 44, 155-162.

Jorquera, O., Kiperstok, A., Sales, E.A., Embiruu, M., Ghirardi, M.L., [3] 2010.

Comparative energy life-cycle analyses of microalgae biomass production in open

ponds and photobioreactors. Bioresour. Technol. 101, 1406-1413.

Karatepe, N., Kckbayrak, S., [14] 1993. Proximate analysis of Turkish lignites by

thermogravometry. Thermochim. Acta. 213,147-150.

Khalil, R.A., Meszaros, E., Grnli, M.G., Vrhegyi, G., Mohai, I., Marosvolgyi, B.,

Hustad, J.E., [13] 2008. Thermal analysis of energy crops Part I: The applicability of a

macro-thermobalance for biomass studies. J. Anal. Appl. Pyrolysis. 81, 5259.

Lapuerta, M., Hernndez, J.J., Rodrguez, J., [11] 2004. Kinetics of devolatilisation

of forestry wastes from thermogravimetric analysis. Biomass Bioenerg. 27, 385 391

Li, Z., Zhao, W., Li, R., Wang, Z., Li, Y., Zhao, G., [6] 2009. Combustion

characteristics and NO formation for biomass blends in a 35-ton-per-hour travelling

grate utility boiler. Bioresour. Technol. 100, 2278-2283.

Majumder, A.K., Jain, R., Banerjee, P., Barnwal, J.P., [35] 2008. Development of a

new proximate analysis-based correlation to predict calorific value of coal. Fuel. 87,

30773081.

Mayoral, M.C., Izquierdo, M.T., Andrs, J.M., Rubio, B., [15] 2001.Different

approaches to proximate analysis by thermogravimetry analysis. Thermochim. Acta

370, 91-97.

McKendry, P., [1] 2002. Energy production from biomass (Part 1): overview of

biomass. Bioresour. Technol. 83, 37-46.

 Miranda, M.T., Arranz, J.I., Romn, S., Rojas, S., Montero, I., Lpez, M., Cruz,

J.A., [23] 2011. Characterization of grape pomace and Pyrenean oak pellets. Fuel

Process. Technol. 92, 278283.

Munir, S., Daood, S.S., Nimmo, W., Cunliffe, A.M., Bibbs, B.M., [5] 2009.

Thermal analysis and devolatilization kinetics of cotton stalk, sugar cane bagasse and

shea meal under nitrogen and air atmospheres. Bioresour. Technol. 100,1413-1418.

Muthuraman, M., Namioka, T., Yoshikawa, K., [2] 2010. A comparison of co-

combustion characteristics of coal with wood and hydrothermally treated municipal

solid waste. Bioresour. Technol. 101, 2477-2482.

Ottaway, M., [27] 1982.Use of thermogravimetry for proximate analysis of coals

and cokes. Fuel. 61, 713-716.

Qian, F.P., Chyang, C.S., Huang, K.S., Tso, J., [7] 2011. Combustion and NO

emission of high nitrogen content biomass in a pilot-scale vortexing fluidized bed

combustor. Bioresour. Technol. 102, 1892-1898.

Ramajo-Escalera, B., Espina, A., Garca, J.R., Sosa-Arnao, J.H., Nebra, S.A., [25]

2006. Model-free kinetics applied to sugarcane bagasse combustion. Thermochim.

Acta. 448, 111116.

Ross, A.B., Jones, J.M., Kubacki, M.L., Bridgeman, T., [24] 2008. Classification of

macroalgae as fuel and its thermochemical behaviour. Bioresour. Technol. 99, 6494

6504

Rosendahl, L.A., Yin, C., Kaer, S.K., Friborg, K., Overgaard, P., [10] 2007.

Physical characterization of biomass fuels prepared for suspension firing in utility

boilers. Biomass Bioenerg. 31, 318-325.

Sadek, F.S., Herrell, A.Y. [17] 1984. Methods of proximate analysis by

thermogravimetry. Thermochim Acta. 81, 297-303.

 Slaghuis, J.H., Raijmakers, N., [28] 2004. The use of thermogravimetry in

establishing the Fischer tar of a series of South African coal types. Fuel. 83, 533536.

Strezov, V., Moghtaderi, B., Lucas, J.A., [32] 2004. Computational calorimetric

investigation of the reactions during thermal conversion of wood biomass Biomass.

Bioenerg. 27, 459-465.

Varol, M., Atimtay, A.T., Bay, B., Olgun, H., [20] 2010. Investigation of co-

combustion characteristics of low quality lignite coals and biomass with

thermogravimetry analysis. Thermochim. Acta. 510,195-201.

Warne, S.S.J., [16] 1991. Proximate analysis of coal, oil shale, low quality fossil

fuels and related materials by thermogravimetry. Trends in Anal. Chem. 10(6), 195-

199.

Wilson, L., John, G.R., Mhilu, C.F., Yang, W., Blasiak, W., [21] 2010. Coffee husks

gasification using high temperature air/steam agent. Fuel Process. Technol. 91,1330

1337.

Wilson, L., Yang, W., Blasiak, W., John, G.R., Mhilu, C.F., [26] 2011. Thermal

characterization of tropical biomass feedstocks. Energy Conserv. Manag. 52,191-198.

Zheng, G., Kozinski, J.A., [31] 2000. Thermal events occurring during the

combustion of biomass residue. Fuel. 79,181-192.

Sorted numerically

[1] McKendry, P., 2002. Energy production from biomass (Part 1): overview of

biomass. Bioresour. Technol. 83, 37-46.

[2] Muthuraman, M., Namioka, T., Yoshikawa, K., 2010. A comparison of co-

combustion characteristics of coal with wood and hydrothermally treated municipal

solid waste. Bioresour. Technol. 101, 2477-2482.

 [3] Jorquera, O., Kiperstok, A., Sales, E.A., Embiruu, M., Ghirardi, M.L., 2010.

Comparative energy life-cycle analyses of microalgae biomass production in open

ponds and photobioreactors. Bioresour. Technol. 101, 1406-1413.

[4] Gil, M.V., Oulego, P., Casal, M.D., Pevida, C., Pis, J.J., Rubiera, F., 2010.

Mechanical durability and combustion characteristics of pellets from biomass blends.

Bioresour. Technol. 101, 8859-8867.

[5] Munir, S., Daood, S.S., Nimmo, W., Cunliffe, A.M., Bibbs, B.M., 2009.

Thermal analysis and devolatilization kinetics of cotton stalk, sugar cane bagasse and

shea meal under nitrogen and air atmospheres. Bioresour. Technol. 100,1413-1418.

[6] Li, Z., Zhao, W., Li, R., Wang, Z., Li, Y., Zhao, G., 2009. Combustion

characteristics and NO formation for biomass blends in a 35-ton-per-hour travelling

grate utility boiler. Bioresour. Technol. 100, 2278-2283.

[7] Qian, F.P., Chyang, C.S., Huang, K.S., Tso, J., 2011. Combustion and NO

emission of high nitrogen content biomass in a pilot-scale vortexing fluidized bed

combustor. Bioresour. Technol. 102, 1892-1898.

[8] Bahng, M.K., Mukarakate, C., Robichaud, D.J., Nimlos, M.R., 2009. Current

technologies for analysis of biomass thermo-chemical processing: A review. Anal

Chim Acta. 651, 117-138.

[9] Gaska, K., Wandrasz, A.J., 2008. Mathematical modeling of biomass fuels

formation process. Waste manag. 28, 973-985.

[10] Rosendahl, L.A., Yin, C., Kaer, S.K., Friborg, K., Overgaard, P., 2007.

Physical characterization of biomass fuels prepared for suspension firing in utility

boilers. Biomass Bioenerg. 31, 318-325.

[11] Lapuerta, M., Hernndez, J.J., Rodrguez, J., 2004. Kinetics of devolatilisation

of forestry wastes from thermogravimetric analysis. Biomass Bioenerg. 27, 385 391
 [12] Demirbas, A., 2004. Combustion characteristics of different biomass fuels.

Prog. Energy Comb. Sci. 30, 219-230.

[13] Khalil, R.A., Meszaros, E., Grnli, M.G., Vrhegyi, G., Mohai, I.,

Marosvolgyi, B., Hustad, J.E., 2008. Thermal analysis of energy crops Part I: The

applicability of a macro-thermobalance for biomass studies. J. Anal. Appl. Pyrolysis.

81, 5259.

[14] Karatepe, N., Kckbayrak, S., 1993. Proximate analysis of Turkish lignites by

thermogravometry. Thermochim. Acta. 213,147-150.

[15] Mayoral, M.C., Izquierdo, M.T., Andrs, J.M., Rubio, B., 2001.Different

approaches to proximate analysis by thermogravimetry analysis. Thermochim. Acta

370, 91-97.

[16] Warne, S.S.J., 1991. Proximate analysis of coal, oil shale, low quality fossil

fuels and related materials by thermogravimetry. Trends in Anal. Chem. 10(6), 195-

199.

[17] Sadek, F.S., Herrell, A.Y. 1984. Methods of proximate analysis by

thermogravimetry. Thermochim Acta. 81, 297-303.

[18] Beamish, B.B., 1994. Proximate analysis of New Zealand and Australian coals

by thermogravimetry. New Zealand J. Geol. Geoph. 37, 387-392.

[19] Ghetti, P., Ricca, L., Angelini, L., 1996. Thermal analysis of biomass and

corresponding pyrolysis products. Fuel. 75(5), 565573.

[20] Varol, M., Atimtay, A.T., Bay, B., Olgun, H., 2010. Investigation of co-

combustion characteristics of low quality lignite coals and biomass with

thermogravimetry analysis. Thermochim. Acta. 510,195-201.

 [21] Wilson, L., John, G.R., Mhilu, C.F., Yang, W., Blasiak, W., 2010. Coffee husks

gasification using high temperature air/steam agent. Fuel Process. Technol. 91,1330

1337.

[22] Biswas, A.K., Umeki, K., Yang, W., Blasiak, W., 2011. Change of pyrolysis

characteristics and structure of woody biomass due to steam explosion pretreatment.

Fuel Process. Technol. 92, 18491854.

[23] Miranda, M.T., Arranz, J.I., Romn, S., Rojas, S., Montero, I., Lpez, M., Cruz,

J.A., 2011. Characterization of grape pomace and Pyrenean oak pellets. Fuel Process.

Technol. 92, 278283.

[24] Ross, A.B., Jones, J.M., Kubacki, M.L., Bridgeman, T., 2008. Classification of

macroalgae as fuel and its thermochemical behaviour. Bioresour. Technol. 99, 6494

6504

[25] Ramajo-Escalera, B., Espina, A., Garca, J.R., Sosa-Arnao, J.H., Nebra, S.A.,

2006. Model-free kinetics applied to sugarcane bagasse combustion. Thermochim.

Acta. 448, 111116.

[26] Wilson, L., Yang, W., Blasiak, W., John, G.R., Mhilu, C.F., 2011. Thermal

characterization of tropical biomass feedstocks. Energy Conserv. Manag. 52,191-198.

[27] Ottaway, M., 1982.Use of thermogravimetry for proximate analysis of coals

and cokes. Fuel. 61, 713-716.

[28] Slaghuis, J.H., Raijmakers, N., 2004. The use of thermogravimetry in

establishing the Fischer tar of a series of South African coal types. Fuel. 83, 533536.

[29] Garca, R., Pizarro, C., Lavn, A.G., Bueno, J.L., 2012. Characterization of

Spanish biomass wastes for energy use. Bioresour. Technol. 103, 249-258.
 [30] Fernndez, R.G., Garca, C.P., Lavn, A.G., Bueno, J.L., 2012. Study of main

combustion characteristics for biomass fuels used in boilers. Fuel Process. Technol.

103, 1626.

[31] Zheng, G., Kozinski, J.A., 2000. Thermal events occurring during the

combustion of biomass residue. Fuel. 79,181-192.

[32] Strezov, V., Moghtaderi, B., Lucas, J.A., 2004. Computational calorimetric

investigation of the reactions during thermal conversion of wood biomass Biomass.

Bioenerg. 27, 459-465.

[33] Haykiri-Ama, H., 2003. Combustion characteristics of different biomass

materials. Energy Convers. Manag. 44, 155-162.

[34] Ahmaruzzaman, M., 2008. Proximate analyses and predicting HHV of chars

obtained from cocracking of petroleum vacuum residue with coal, plastics and biomass.

Bioresour. Technol. 99, 50435050.

[35] Majumder, A.K., Jain, R., Banerjee, P., Barnwal, J.P., 2008. Development of a

new proximate analysis-based correlation to predict calorific value of coal. Fuel. 87,

30773081.
Figure 1 Average experimental errors observed for the tested methods.
 INACTIVE ATMOSPHERE OXIDATIVE ATMOSPHERE
METHOD TR1 SP1 DT1 TR2 SP2 DT2 CT TR3 SP3 DT3
MAY 1 20 105 5 20 900 - 20 - - -
MAY 2 20 900 5 -25 450 - - 20 700 -
MAY 3 80 1020 10 - - - - - - -
MAY 4 20 105 3 80 850 - - - - -
OTT 250 110 1-5 250 900 1-5 1 - - -
SDK 1 100 1000 3 - - - 3 - - -
SDK 2 100 135 2 100 1000 3 3 - - -
KAR 20 110 30 40 950 7 - -20 750 5
WAR 100 110 2 100 950 3 3 - - -
BEA 50 110 5 50 950 7 42 - - -
TR1 SP1 DT1 TR2 SP2 DT2 CT TR1 SP1 DT1
LAP - - - - - - 18-108 10-60 1100 -
Table 1 Summary of the studied analysis methods developed for coals and methods
assayed for biomass.

Where: TR 1, 2 and 3 are the first, second and third temperature ramps respectively,

measured in C/min. SP is the set-point for each temperature ramp, measured in C and

DT is the dwelling time after each ramp, measured in minutes. CT is the combustion

time in oxidative atmosphere when set-point is reached.

 Sample %M %A % VM % FC
Apple tree leaves 9.3 12 71.9 16.1
Beetroot pellets 12.5 9 76 15
Chestnut tree leaves 8.2 4.9 72.41 22.69
Hazelnut shell 8.74 2.2 77 20.8
Miscanthus 7.53 9.6 79 11.4
Nectarine stone 8.2 1.1 76 22.9
Peach stone 8.55 0.5 75.6 23.9
Pine and pine apple leaf pellets 8.25 3.2 74.5 22.3
Pistachio shell 8.75 1.3 82.5 16.2
Soya grain 10.4 4.8 77 18.2
Wheat grain 10.3 2.8 80 17.2
Wood pellets 1 7.96 1.3 82 16.7
Wood pellets 2 7.53 0.66 84 15.34
Table 2 Data obtained for moisture (M), ash (A), volatile matter (VM) and fixed
carbon (FC) content of biomass samples using ASTM normative, measured in mass
percentage.
 MOISTURE ASH VOLATILE MATTER FIXED CARBON
M MV EE BE AD MV EE BE AD MV EE BE AD MV EE BE AD
Apple tree leaves 13.3 8.3 11.0 -11.0 1.0 6.45 46.2 -46.2 5.5 67.3 6.4 -6.4 4.6 16.6 3.1 3.1 0.5
Beetroot pellets 12.1 11.0 11.8 -11.8 1.5 8.43 6.3 -6.3 0.6 62.5 17.8 -17.8 13.5 10.7 28.8 -28.8 4.3
Chestnut tree leaves 13.0 8.1 1.4 -1.4 0.1 2.90 40.9 -40.9 2.0 70.8 2.2 -2.2 1.6 20.2 10.8 -10.8 2.5
Hazelnut shell 14.1 8.5 2.5 -2.5 0.2 0.15 93.4 -93.4 2.1 73.5 4.5 -4.5 3.5 23.3 11.8 11.8 2.5
Miscanthus 11.8 7.9 5.5 5.5 0.4 6.37 33.7 -33.7 3.2 69.6 11.8 -11.8 9.4 13.5 18.6 18.6 2.1
Nectarine stone 18.6 7.9 4.1 -4.1 0.3 0.52 53.1 -53.1 0.6 80.7 6.2 6.2 4.7 21.1 7.7 -7.7 1.8
Peach stone 19.5 7.9 7.4 -7.4 0.6 0.73 46.8 46.8 0.2 81.8 8.1 8.1 6.2 20.4 14.7 -14.7 3.5
Pine and pine apple leaf pellets 13.0 8.8 6.8 6.8 0.6 1.81 43.3 -43.3 1.4 70.5 5.4 -5.4 4.0 21.3 4.6 -4.6 1.0
Pistachio shell 14.7 8.2 6.1 -6.1 0.5 0.53 59.1 -59.1 0.8 79.7 3.4 -3.4 2.8 17.5 7.8 7.8 1.3
Soya grain 19.1 9.4 9.3 -9.3 1.0 2.78 42.0 -42.0 2.0 76.6 0.5 -0.5 0.4 18.1 0.7 -0.7 0.1
Wheat grain 17.0 10.2 0.7 -0.7 0.1 2.95 5.3 5.3 0.1 77.6 3.0 -3.0 2.4 13.0 24.3 -24.3 4.2
Wood pellets 1 14.1 7.8 2.5 -2.5 0.2 0.21 84.0 -84.0 1.1 79.8 2.7 -2.7 2.2 18.4 10.1 10.1 1.7
Wood pellets 2 13.4 8.0 5.7 5.7 0.4 0.08 88.3 -88.3 0.6 81.9 2.5 -2.5 2.1 15.4 0.4 0.4 0.1
Average values 5.8 -3.0 0.5 49.4 -41.4 1.6 5.7 -3.5 4.4 11.0 -3.1 2.0
Table 3 Values obtained for each of the chosen samples.

Where M is the mass of sample (mg), Mv the measured value (%), EE, BE and AD are respectively experimental error (%), the BIAS error and

the absolute deviation for each individual sample. The average values are presented in the last row.
- A method to determinate proximate analysis data in biomass by TG is proposed
- Coal developed methods are compared and improved for biomass use.
- Satisfactory results are obtained, main on moisture and volatile matter values.