Strategies for selection of thermochemical processes for the valorisation of biomass

Pyrolysis using several regional feedstocks has been carried out under nitrogen and hydrogen atmosphere and different biomass feedstocks were also liquefied using sub/supercritical solven

Strategies for selection of thermo-chemical processes for the

valorisation of biomass

Rawel Singha,b, Bhavya B Krishnaa,b, Garima Mishraa,b, Jitendra Kumara,

Thallada Bhaskara,b,*

a Thermo-catalytic Processes Area(TPA), Bio-Fuels Division (BFD), CSIR-Indian Institute of Petroleum (IIP), Dehradun 248005, India

b Academy of Scientific and Innovative Research (AcSIR), New Delhi, India

a r t i c l e i n f o

Article history:

Received 10 January 2016

Received in revised form

3 March 2016

Accepted 4 March 2016

Available online xxx

Keywords:

Pyrolysis

Hydrothermal liquefaction

Kinetic analysis

Lignocellulosic biomass

Aquatic biomass

Algae

a b s t r a c t Research on biomass conversion has been gaining a lot of interest as biomass is renewable and sus-tainable in nature Products from biomass can be obtained by different methods amongst which thermo-chemical route has a very high potential Biomass is generally available in a localised manner in varying quantities and qualities throughout the year and hence, region specific technologies have to be devel-oped considering the end user requirement Pyrolysis is a very versatile technique with the above considerations The process parameters can be tweaked to necessity to produce more bio-oil or bio-char Thermogravimetric analysis is essential for understanding the decomposition behaviour of the feedstock before the lab scale pyrolysis is carried out Pyrolysis using several regional feedstocks has been carried out under nitrogen and hydrogen atmosphere and different biomass feedstocks were also liquefied using sub/supercritical solvents This review aims to provide a comparison of the results obtained using various processes This helps in the decentralised processing of biomass (dry biomass using pyrolysis and wet biomass by hydrothermal liquefaction) to produce bio-crude which can be upgraded to produce fuels/ chemicals/petrochemical feedstocks in an environmental friendly manner

© 2016 Elsevier Ltd All rights reserved

1 Introduction

Fossil resources derived fuels have played the most important

role in the rapid technological progresses over the past few

cen-turies It is estimated that more than 85% of the world's energy

requirements are obtained from conventional fuels [1] Energy

scenarios project that world's annual energy consumption will

in-crease steeply from current value of 500 to 1000e1500 Exa Joules

per annum by 2050[2e4] Use of fuels derived from fossil resources

leads to global warming due to high levels of CO2emission in

at-mosphere Renewable, sustainable and environment friendly

alternate resources are required to address these issues Solar

ra-diation, winds, tides and biomass are renewable resources and

whilefirst three resources can be used to obtain energy, biomass

can be used to produce energy, chemicals and materials[5] Need

for a secure source of transportation fuels and chemicals make it

essential to explore bio-fuels/bio-based hydrocarbons as

alternatives to hydrocarbons derived from fossil resources[6] The transition from the current fossil-based to bio-based carbon econ-omy is expected to evolve continuously in the coming decades and

a continuous changeover to more complex bio-renewable feed-stocks like agricultural residues, industrial wastes, green plants, wood, or algae will occur[7]

2 Types of biomass feedstocks Biomass is a plant matter of recent (no geologic) origin or ma-terial derived there from and can be used to produce various useful chemicals and fuels[8,9] Biomass contains variety of plant species with varying morphology and chemical composition Low hydrogen to carbon ratio and high oxygen to carbon ratio in biomass suggests that biomass can be utilised for the production of fuels as well as functional chemicals[7] Depending on the nature

of biomass used different biomass generation are shown inFig 1 First-generation bio-fuels are derived from edible feedstock from the agricultural sector such as corn, wheat, sugarcane, and oilseeds First generation biofuels have limitation of food versus fuel issue Second-generation bio-fuels are non-edible and

* Corresponding author Thermo-catalytic Processes Area(TPA), Bio-Fuels Division

(BFD), CSIR-Indian Institute of Petroleum (IIP), Dehradun 248005, India.

E-mail addresses: tbhaskar@iip.res.in , thalladab@yahoo.com (T Bhaskar).

Contents lists available atScienceDirect Renewable Energy

j o u r n a l h o m e p a g e :w w w e l s e v i e r c o m / l o c a t e / r e n e n e

http://dx.doi.org/10.1016/j.renene.2016.03.023

0960-1481/© 2016 Elsevier Ltd All rights reserved.

Renewable Energy xxx (2016) 1e12

comprise of raw materials derived from lignocellulosic biomass and

crop waste residues from various agricultural and forestry

pro-cesses [10,11] Lignocellulosic biomass has three major

compo-nents: cellulose, hemicellulose and lignin The agricultural residues

can be classified as field and seed crop, fruit and nut crop, vegetable

crop and nursery crop[12] The residues generated from the forest

products industry can be divided into two categories: (1) logging

residues-generated from logging operations, e.g., fromfinal fellings

and (2) industrial by-products- generated by the forest industries

during processing of timber, plywood, particleboard, pulpwood,

etc.[13,14] Energy crops are specifically grown to produce some

form of energy Energy crops are generally divided into two types:

herbaceous and woody Herbaceous energy crops are mostly types

of grasses, which are harvested like hay Perennial grasses, such as

switchgrass, miscanthus, bluestem, elephant grass, and wheatgrass

could all potentially be grown as energy crops[15] Third

genera-tion bio-fuels are based on algal matter (micro- and macro algae)

and cyanobacteria, which yield carbohydrates, proteins, vegetable

oils (lipids), and, subsequently, biodiesel and hydrogen gas, are

gaining considerable interest The term algae can refer to

micro-algae, cyanobacteria (the so called“blue-green algae”), and macro

algae (or seaweed) The differences between microalgae and macro

algae are shown inFig 2

3 Thermochemical conversion of biomass

There are several methods of conversion of biomass viz:

me-chanical, chemical, biochemical and thermochemical Mechanical

processes only perform a size reduction of feedstock Chemical

processes carry out a change in the chemical structure of the

molecule by reacting with other substances These processes

include the wide class of chemical reactions where a change in the

molecular formula occurs [16] Bio-chemical processes occur at

lower temperatures and most common types of biochemical

processes are fermentation and anaerobic digestion The

fermentation uses microorganisms and/or enzymes to convert a

fermentable substrate into recoverable products (usually alcohols

or organic acids)[17] Anaerobic digestion involves the bacterial breakdown of biodegradable organic material in the absence of oxygen over a temperature range from about 30 to 65
C The main end product of these processes is biogas (a gas mixture made of methane, CO2 and other impurities) [16,18] An overview of thermochemical and biochemical processes during biorefinery is shown inFig 3

Thermochemicals processes are carried out in the presence of heat and can also use catalyst Thermo-chemical methods utilize the entire biomass without any pre-treatment steps to produce value added hydrocarbons In comparison to the biochemical processes, thermochemical processes occur faster in the range of few seconds, minutes or hours when the former takes time in the range of days to complete The other advantages of thermo-chemical methods of conversion are that they are not feed-stock specific and can also process moisture-rich/aquatic feed-stocks The micro-organisms are feed specific and even the slightest of change could lead to its non-functionality This poses

a major risk in the commercialisation of the process at an

First generation

biofuels

Edible parts of agricultural crops and forest trees

Competes with human and animal food Food vs fuel issue

Second generation biofuels

Non edible parts of crops, forest residues, energy crops

No food vs fuel issue

Food production for human population is essential-hence availability of biomass is plenty

Third generation biofuels

Macro algae, micro algae

Does not compete with agricultural land

Utilises CO 2 from atmosphere or from industrial emissions for growth

Expected higher yield per hectare

Fourth generation biofuels

Modified organisms for better yield

Yet to be used in large scale

Fig 1 Different biofuel generations depending on biomass type.

Fig 2 Comparison of microalgae and macro algae.

R Singh et al / Renewable Energy xxx (2016) 1e12

industrial level since the biomass availability in terms of quality

and quantity keeps varying all through the year Various

ther-mochemical processes for biomass conversion are shown in

Fig 4

This review article focuses on the strategies for selection of

thermochemical processes for valorisation of diverse biomass

feedstocks and need to have decentralised units that may be the

most immediate solutions for introduction of bio-based energy

systems For the implementation of these units most important

thing is to do the fundamental research on diversified feedstocks

to know the thermal decomposition behaviour, kinetics of biomass and effective heat management through thermodynamic data Based on the fundamental studies from TG/DTG studies pyrolysis approaches for dry biomass have been proposed for centralised and decentralised biorefineries For effective uti-lisation of wet biomass hydrothermal liquefaction has been pro-posed and effect of various parameters has been discussed The utilisation of biomass components viz cellulose and lignin to valuable chemicals has also been discussed using appropriate catalytic and thermal methods

Biomass

Pre-treatment

Holocellulose

Fermented to ethanol, butanol, xylitol etc by biochemical routes

Conversion to high value chemicals by thermochemical routes

Lignin

Only burnt in biochemical processes

Thermo-chemical route can be used for the production

of aromatics

Fig 3 Overview of thermochemical and biochemical processes for biomass conversion.

Biomass

Combustion Heat/ electricity Centralised heating, electricity by IGCC

Gasification

Syn gas FT to form fuels/ chemicals

Steam reforming to hydrogen

Pyrolysis/

hydrothermal liquefaction

Bio-oil

Upgraded to fuels/fuel blends

Upgraded to produce chemicals or petrochemical feedstocks

Bio-char

Adsorbents, catalysts, electrodes, soil management and C-sequestration etc

Fig 4 Thermochemical processes for biomass conversion.

R Singh et al / Renewable Energy xxx (2016) 1e12

4 TG/DTG studies of different feedstocks

“Thermogravimetric analysis” is a technique in which the mass

of a substance is measured as a function of temperature, while the

substance is subjected to a controlled temperature programme The

thermal behaviour of biomass is determined using TGA The TG and

derivative thermogravimetric (DTG) curves observed from TGA can

be used to determine the apparent weight loss of samples Various

thermodynamic parameters and kinetic studies of different

feed-stocks have been carried out and discussed below

4.1 Thermodynamic parameters

Pyrolysis behaviour and kinetic study of different lignocellulosic

biomass feedstocks (Cerus, Cheed, Cokad, Sagwan and Shimbal)

have been done The thermodynamic properties were calculated

using the kinetic triplet values determined using the model free

approaches The results obtained from thermal decomposition

processes indicate that there are three main stages in the pyrolysis,

i.e., dehydration, active and passive pyrolysis The value of apparent

activation energy calculated using isoconversional methods are

used to evaluate the thermodynamic properties such as enthalpy,

Gibbs free energy and entropy of biomass pyrolysis A shift in the

DTG curve, i.e increase in the DTG peak temperature is observed as

the heating rate increases, but this shift is not uniform for all the

biomass studied The estimated thermodynamic parameter values

are found to be different for biomasses all of which have forest as

their origin; but they are similar for a particular biomass at different

heating rates

The model-free approach (Friedman method, Kissinger Akahira

Sunose method and Flynn-Wall-Ozawa method) does not require

assumption of specific reaction models, and yields unique kinetic

parameters as a function of either conversion (isoconversional

analysis) or temperature (non parametric kinetics) Model free

ki-netic methods are conversional in evaluation of pre-exponential

factor and reaction mechanism and the constraints involved do

not permit a straight forward evaluation of the remaining kinetic

parameters A and f(a) Thus a combination of pyrolysis data from

TGA and model free (isoconversional) methods can be a potent tool

for predicting the reaction kinetics as well as the thermodynamic

parameters of the biomass pyrolysis process

The results obtained from the TGA studies of the feedstocks

showed that the apparent activation energy values calculated from

the isoconversional methods (150e170 kJ/mol) are found to be

similar for the all studied forest biomass except for Cokad which is

showing a relative high value of Ea (~200 kJ/mol) The values of the

pre-exponential factor are found to lie in the range of 108e1014s1

The value of A for Cokad is of the order of 1014from which it can be

attributed that the rotations of the active complex and the reagent

do not change during the reaction The reaction order values are

found to be high which can be attributed to the multiscale and

multiphase nature of biomass feedstock The estimated

thermo-dynamic parameter values are found to be different for studied

biomasses all of which have forest as their origin; but they are

similar for a particular biomass at different heating rates

4.2 Kinetic studies and reaction mechanism during pyrolysis

The thermal decomposition of biomass proceeds via a very

complex set of competitive and concurrent reactions with

forma-tion of over a hundred intermediate products and thus the exact

mechanism for biomass pyrolysis remains a mystery till date

Modelling pyrolysis reactions with its unrevealed reaction

mech-anism presents a great challenge The specific temperatures at

which various heterogeneous reactions occur, their reaction rates

and the energies involved in these reactions are valuable infor-mation useful for pyrolysis system design

The development of thermochemical processes for biomass conversion and proper equipment design requires the knowledge

of several process features which include a good understanding of the governing pyrolysis mechanisms, the determination of the most significant pyrolysis parameters and of their effect on the process and knowledge of the kinetics[19] Understanding both multiscale and multiphase complexities represents a vital step forward in optimizing pyrolysis and developing next-generation biofuel technologies[20] A precise conception of solid state py-rolysis kinetics is very crucial in designing and operating industrial biomass conversion systems The kinetic modelling studies of biomass pyrolysis assists in analysis and optimization of reaction conditions, process parameters and adaption of pyrolysis systems

to regionally differing surrounding conditions and diverse nature of biomass feedstock Kinetic studies form basis for development of a prototype model for energy provision to remote rural areas Fundamental research will lead to a ‘building-up’ approach whereby chemical mechanisms are integrated into particle models (accounting for transport phenomena) which are capable of pre-dicting global performance (i.e., bio-oil yield and composition) Kinetics is the study of the dependence of the extent or rate of a chemical reaction on time and temperature Study of kinetics in-volves using mathematical models that quantify the relationship between the rate of reaction, time, and temperature[21] Kinetic analysis is expected to be capable of

 Revealing complexities in the reaction kinetics and prompting some mechanistic clues

 Adequately describing the temperature dependence of the overall reaction rate

 Producing reasonably consistent kinetic characteristics from isothermal and non isothermal data[22]

A comprehensive kinetic analysis of a solid state reaction has four main stages: Stage 1: Experimental collection of data; Stage 2: Computation of kinetic characteristics for the data from stage 1; Stage 3: Validation of kinetic parameters estimated; Stage 4: Interpretation of the significance of any parameters evaluated in stage 2 The different stages are shown inFig 5

Modeling of pyrolysis implies the representation of the chemical and physical phenomena constituting pyrolysis in a mathematical form The inherent complexity of the pyrolysis process has posed formidable challenges to modelling attempts The pyrolytic decomposition involves a complex series of interlinked reactions, and consequently, changes in the experimental heating conditions

or sample composition and preparation may affect not only the rate

of reaction, but also the actual course of reactions[21] Besides the sheer extent and range of pyrolysis reactions, several other issues complicate the modeling of pyrolysis More often these issues are inter-linked making it extremely difficult to separate the influence

of one from another

Modeling biomass pyrolysis is a challenge because of the variety

of the raw materials involved and also because of the wide oper-ating conditions Shin et al.[23]observed that the accurate pre-dictions of gas species and aromatics from the pyrolysis of biomasses, and mainly the effect of different operating conditions, not only require the description of the released components, but also the definition of their successive gas phase reactions The py-rolysis of various lignocellulosic materials is differentiated by the various reaction rates and thefinal product distribution achieved The quantitative formulation of the pyrolysis of a single biomass particle involving all the above-mentioned effects is a task requiring considerable effort [21] In view of the importance of

R Singh et al / Renewable Energy xxx (2016) 1e12

kinetics in pyrolysis of a biomass it is necessary to know the values

of kinetic parameters of the biomass under a particular set of

conditions However, difficulty arises in studying the thermal

behaviour of biomass due to lack of exact knowledge of the course

of reactions and their degree of completion Moreover, the vast

number of products resulting from the thermal degradation of

biomass hinders a thorough understanding of the process [21]

Solid state reactions ordinarily demonstrate a tangled interplay of

various chemical and physical processes such as solid-state

decomposition, reaction of gaseous products with the solid,

subli-mation, polymorphous transitions, diffusion, melting, evaporation,

adsorption, desorption, etc Therefore, the effective activation

en-ergy of a solid state reaction is generally a composite value

deter-mined by the activation energies of various processes and by their

influence on the overall reaction rate Even if the temperature is

kept constant (single isothermal experiment), the relative

contri-butions of the elementary steps into the overall reaction rate vary

with the extent of conversion ultimately resulting in a dependence

of the effective activation energy on the extent of conversion

Additionally, the kinetics of solid state reactions are known to be

sensitive to pressure, size of crystals, gaseous atmosphere and

many other factors which are likely to change during the process

[24]

The relative importance of the internal heat transfer to the

external heat transfer is defined by the ratio of their respective

characteristic times:

tinternal

texternal¼hL

k ¼ Bi

This is the definition of the Biot number, a dimensionless number commonly used in thermal analysis Biot numbers larger than 10 characterise a heat transfer limited by the internal con-duction The internal pyrolysis number gives a measure of the relative importance of the internal conduction and the chemical reaction

AeE=RTrCPL2 The external pyrolysis number is the product of biot number and the internal pyrolysis number It gives the ratio between heat convection rate and chemical reaction rate and can be stated as follows,

Py0 ¼ h

AeE=RTrCPL Different solid state kinetic methods used for biomass pyrolysis are shown inFig 6

Studies of the kinetics of cellulose, hemicellulose and lignin separately revealed that the interactions between fractions are important, and the pyrolysis behaviour of biomass components is not completely additive like that in consecutive reaction model In the TG-DTG analysis of lignocellulosic material two or three peaks usually appear, that can be assigned to cellulose, lignin and hemi-cellulose, indicating that, although there are interactions between fractions, their identity is maintained Specific conclusions with respect to each feedstock are as follows[25,26]:

Microcrystalline Cellulose: The master plots method predicts Fig 5 Different stages of kinetic analysis.

R Singh et al / Renewable Energy xxx (2016) 1e12

the pyrolysis process of microcrystalline cellulose by an

autocata-lytic reaction mechanism Friedman method gave more stable

values than in comparison with Vyazovkin AIC method Distributed

activation energy model gave a goodfit to the kinetic data

Lignin: The present work shows that lignin pyrolysis is a

com-plex process and series of reactions occur rather than a simple

single step reaction The lignin pyrolysis takes place over a wide

range of temperature where three or more peaks can be seen in the

DTG curve indicating presence of more than one pseudo

compo-nents (sinapyl alcohol, coniferyl alcohol and guaiacyl alcohol) A

wide distribution of activation energy in DAE Model shows

pres-ence of complex process with multiple reactions

Rice Straw: The kinetic results for rice straw showed that the

reaction mechanism for rice straw pyrolysis can be kinetically

characterized by two successive reactions At conversion values less

than 0.3 the decomposition of rice straw is governed by diffusion

and it tends to third order rate equation at high conversion

(pre-dicted by compensation effect method)

Groundnut husk: The kinetic results for groundnut husk

pre-sent a complicated analysis, as at conversion values less than 0.1 the

decomposition process is governed by diffusion but no clarity in

mechanism is seen at conversion values greater than 0.1

Pine: In the case of pyrolysis process of pine wood at conversion

values less than 0.7 the pyrolysis process is governed by two and

three dimensional diffusion whereas at higher conversion values

(a> 0.7) the mechanism is controlled by reaction order mechanism

with order of reaction as 1.5

Deodar: Kinetic predictions similar to pine wood were observed

in case of deodar pyrolysis showing that at conversion values less

than 0.8 the pyrolysis process is governed by two dimensional diffusion whereas at higher conversion values (a> 0.8) the mech-anism is controlled by a third order reaction mechmech-anism The entire process for both pine and deodar wood pyrolysis is closer to diffusion controlled mechanism

Water hyacinth: Water hyacinth presented a complex pyrolysis kinetics which could not be easily modelled The model free kinetic methods gave a badfit for the case of water hyacinth showing its complicated pyrolysis behaviour DAE Model for three pseudo-components used for this case gave a reasonablefit but the model could not be validated at higher heating rates Hence, more com-plex DAE Models with more number of pseudocomponents can be used further for study of kinetics of water hyacinth

In addition to these feedstocks, kinetics of tamarind seed husk, another Indian biomass feedstock was studied Tamarind seed husk exhibited an abnormal behaviour giving very high activation en-ergy values ranging from 144.53 to 639.57 kJ/mol with the model free kinetic analysis methods

Modelfitting and model free methods both suffer from certain drawbacks and have certain advantages over the other A combi-nation of model free and modelfitting kinetic methods can help in providing much more informative and meaningful results rather than empirical values with no mechanistic meaning Model free kinetic methods can be used efficiently to give initial parameter estimates for modelling biomass pyrolysis using distributed acti-vation energy models It can be used to give preliminary insights into the kinetic triplet and to provide initial estimates for kinetic parameters which will help to avoid the extra computation time needed in optimization of objective function for prediction of these Fig 6 Different solid state kinetic methods.

R Singh et al / Renewable Energy xxx (2016) 1e12

values Thus a combination of both these kinetic analysis methods

can be a powerful tool for predicting the reaction kinetics of

biomass pyrolysis process For effective valorisation of

lignocellu-losic biomass into energy products/chemicals, the study of role of

solid catalyst is essential In this direction, the kinetic studies for

selected feedstocks (rice straw, groundnut, pine, deodar) using the

conventional microporous catalytic materials (such as Mordenite,

y-zeolite, ZSM-5) has been initiated

5 Pyrolysis of different feedstocks

Pyrolysis is said to be the basis of all thermochemical methods of

conversion and it is defined as the heating of any material in the

absence of oxygen Pyrolysis processes produces bio-oil, bio-char

and non-condensable gases as products and the amount produced

varies on several factors The operating parameters for pyrolysis

depend on the process type and the end product requirements

Researchers have established the effect of pyrolysis temperatures,

particle size, gas flow rate and many other parameters through

their studies for the selected feedstock The process parameters

have been identified in each case and the products have been

characterised The end product utilisation is different in each case

and this dictates the development of indigenous technologies

depending on availability and costs involved with logistics Bio-oil

obtained from slow pyrolysis of biomass does not have any direct

high commercial value Hydropyrolysis experiments have been

carried out by various groups and some of the parameters that have

been tested are the effect of variations infinal temperature, gas

used, pressure, heating rate, and particle size Various reactor

configurations have been used, which give us information on the

heat and mass transfer effects There are conceptual articles which

give us an insight into the possibilities of using hydropyrolysis in

the transition stage of shifting from fossil-based economy to

renewable feedstock-based economy Basic fundamental data has

to be generated for both the processes using several feedstocks

available in the presence and absence of catalysts thereby

gener-ating a huge database of information which will be helpful to

un-derstand the process No single process can produce solutions to all

fossil resource utilisation problems and hence, different processes

have to be used on different feedstocks to get different products

5.1 Types of pyrolysis

There are various kinds of pyrolysis depending on the reactor

employed, gas atmosphere used and residence time inside the

reactor The classification of pyrolysis based on the residence time

generally is slow, intermediate, fast andflash pyrolysis Pyrolysis is

generally carried out in inert atmosphere like helium or nitrogen

etc and in cases where hydrogen atmosphere is used the process is

termed as hydropyrolysis It can also be carried out under vacuum

Based on the reactor used, the pyrolysis processes are ablative,

rotating cone, screw, auger, bubblingfluidised bed or circulating

fluidised bed and microwave pyrolysis The products and ratios in

which they are formed vary depending upon the reaction

param-eters such as environment, reactor used, pyrolysis temperature,

rate of heating and source of heat Longer vapour residence time

favours the production of bio-char Moderate temperatures and

short vapour residence time are optimum for producing liquids

One of the most important parameters in pyrolysis is the residence

time of the solid phase which can vary from seconds to days Fast

pyrolysis is characterised by high heating rates and short residence

times Fast pyrolysis generally requires the feedstock to be supplied

as fine particles; and the reactor design must facilitate rapid

removal of the hot vapours from the presence of the hot solids In

fast pyrolysis, liquid fuel called bio-oil condenses from the vapours

and aerosols; the process also yields non-condensable gases of medium calorific value Other pyrolysis techniques include inter-mediate pyrolysis and flash pyrolysis In intermediate pyrolysis, reaction occurs at controlled heating rates thus avoiding tar for-mation Interestingly, the size and shape of the feed particles are less critical than in fast pyrolysis, which allows a wider variety of biomass feedstock Flash pyrolysis occurs with very fast heating rates of1000
C/s and uses even shorter solid residence times

(<0.5 s) than fast pyrolysis The typical operating temperature for flash pyrolysis is 800e1000
C and the biomass is supplied in the

form of dust[27] This process gives a similar product distribution

as fast pyrolysis A higher amount of liquid products can be pro-duced through fast, intermediate andflash pyrolysis; whereas slow pyrolysis produces a higher amount of the solid fuel (char)[28] A comparison of pyrolysis under nitrogen and hydrogen environment

is shown inFig 7

5.2 Mechanistic differences of pyrolysis under nitrogen and hydrogen environment

In the process of hydropyrolysis, as the reaction takes place under the pressures of hydrogen, the formation of free radicals is hindered The amount of unsaturated hydrocarbons reduces thereby increasing the quality of the bio-oil formed The probable reaction mechanism under nitrogen environment could be the random thermal cracking of the large biomass molecule which leads to an increased variation in the products formed However, in the presence of hydrogen it could be the systematic bond cleavage (initiation with hydrogenolysis/hydrogenation, bond breaking in the order of increasing bond strength and stability of formed components etc.) in the macromolecular structure which leads to the formation of few products with higher selectivity

In case of rice straw, pyrolysis under hydrogen and nitrogen environments showed that the reaction mechanism is different in both cases as the product portfolio was different in two atmo-spheres The bio-oils obtained under hydrogen atmosphere showed more selectivity towards phenolic compounds More amount of bio-oil was obtained under nitrogen atmosphere but the selectivity towards phenolic compounds was less[29]

Mechanistic differences between slow and hydropyrolysis are presented asFig 8

Pyrolysis

Presence of nitrogen Endothermic

Possibility of free radicals and olefins which can repolymerise leading to less stable bio-oil More yields of bio-oil and bio-char

Suitable in decentralised units

Presence of hydrogen Exothermic

Presence of hydrogen produces more saturated comounds in bio-oil increasing shelf life of bio-oil When combined with in situ hydroconversion, process is self sustainable Centralised unit is more energy

efficient

Fig 7 Comparison of pyrolysis under nitrogen and hydrogen environment.

R Singh et al / Renewable Energy xxx (2016) 1e12

5.3 Differences in behaviour of different biomass decomposition in

slow pyrolysis

5.3.1 Agricultural residues

Slow pyrolysis of rice straw under nitrogen atmosphere has

been carried out at 300, 350, 400 and 450
C[29] The yield of

bio-oil increased with increase in temperature up to 400
C beyond

which it reduced It was concluded that 400
C is the optimum

temperature for slow pyrolysis of rice straw giving high yields of

desirable products namely the bio-oil and bio-char In case of wheat

straw, the conversion was a maximum at 400
C In case of wheat

husk, the bio-oil yield was almost similar at around 23 wt.% at 300,

350 and 400
C but increased to 28.5 wt.% at 450
C The biochar

yield in case of thermal pyrolysis of wheat husk was 35.1 wt.%[30]

Cotton residue is an interesting feedstock for thermo-chemical

methods of conversion to produce value added hydrocarbons in a

sustainable manner and for slow pyrolysis under nitrogen

envi-ronment, 400
C was found to be the optimum pyrolysis

temper-ature to obtain maximum bio-oil yield The bio-oil yield observed

was 38 wt.% and biochar yield was 30 wt.% at 400
C[31]

5.3.2 Forest residues

Slow pyrolysis of deodar under nitrogen atmosphere was

car-ried out at 300, 350, 400 and 450
C Maximum bio-oil yield of

46.5 wt.% was obtained at 350
C and with further increase in

temperature, the bio-oil yield decreased Biochar yield observed at

350
C was 46.5 wt.% Conversion was seen to increase with

in-crease in temperature due to the primary decomposition of

biomass The increasing temperature also had an effect on the

non-condensable gases yield which increased at higher temperatures

due to secondary cracking[32] Bio-oil under nitrogen atmosphere

was also majorly composed of phenolic compounds and aromatic

ethers The complete loss of functionality and the aromatic nature

of bio-char indicated the complete conversion of biomass

5.3.3 Defatted biomass

After the production of bio-diesel or jet fuel from non-edible

oils, the de-oiled cakes are generally thrown away or burnt These

cakes can be an effective and renewable source of valuable

hy-drocarbons as they do not compete for fodder also in many cases

Slow pyrolysis of jatropha seed deoiled cake (JSDC) was carried out

under nitrogen at various pyrolysis temperatures of 300, 350, 400,

450 and 500
C The yield of bio-oil was a maximum of 38.4 wt.% at

450
C The biochar yield observed was 37 wt.% at 450
C Phenolic

compounds and aromatic ethers were the major fraction of

compounds present in the organic fraction of the bio-oil Thefirst order rate equation was used to calculate the frequency factor and activation energy of the process and the lower values as compared

to thermo-gravimetric analysis indicated the efficient heat and mass transfer in the reactor system[33]

The slow pyrolysis of de-oiled microalgae has been carried out

at 350, 400 and 450
C Low amount of bio-oil yield also could be attributed to the very high ash content of the de-oiled microalgae It can thus be concluded that the conditions required for pyrolysis increase in terms of severity in the following order: forest residue< agricultural residue < defatted biomass The bio-oils were majorly composed of phenolic compounds and aromatic ethers in most cases The next in line were hydrocarbons, furans and carbonyl compounds Alcohols, acids and esters were found in small fractions In case of de-oiled algae, the bio-oil had consider-able quantities of nitrogen compounds due to the protein content of algae The bio-oil obtained by slow pyrolysis can only be tried to blend with transformer/bunker/furnace oil in best case scenario in terms of fuels Since it has been identified that it is rich in phenolic compounds, the bio-oil can be used for the production of very high value chemicals after the development of efficient separation processes

5.4 Behaviour of different biomass feedstocks under hydropyrolysis 5.4.1 Agricultural residues

Rice straw pyrolysis has been carried out in hydrogen environ-ments to understand the effect of reaction atmosphere on the py-rolysis products Optimum temperature was found to be 400
C and higher pressures of 30 bar were required to produce maximum yield of phenolic derivatives in the presence of hydrogen [29] Hydropyrolysis of cotton residue has been carried out and it has been observed that the optimum conditions in the used experi-mental set up are 400
C and 20 bar pressure of hydrogen[34] 5.4.2 Forest residues

Hydropyrolysis of deodar was carried out at 350 and 400
C at different pressures of hydrogen of 1, 10, 20, 30 bar At both the temperatures, the trend followed by bio-oil, bio-char and gas yields were similar The bio-oil yield at every pressure condition was higher at 400
C than at 350
C The bio-oil yield was a maximum at

1 bar pressure and with increase in pressure from 10 to 30 bar, the bio-oil yield reduces though the difference in yields at 20 and

30 bar is negligible It has been well established that the bio-oil quality produced under hydrogen pressure is better than the case with nitrogen[35] With increase in pressure, the biochar yield increases since the vapours are more in the vicinity of bio-char leading to more bio-char formation[36] At 30 bar pressure, sec-ondary cracking is seen to take place as the yield of non-condensable gases increases at the expense of oil and bio-char Thus, it can be concluded that 400
C and 20 bar pressure are optimum conditions for the hydropyrolysis of deodar

5.4.3 Defatted biomass The hydropyrolysis of jatropha seed de-oiled cake has been carried out at various hydropyrolysis temperatures (300, 350, 400 and 450
C) and pressures of hydrogen (1, 20, 40 and 52 bar) It has been observed that the liquid bio-oil yields have increased with pyrolysis temperature and pressure It was also found that 40 bar is the optimum value of hydrogen pressure for the hydropyrolysis reactions at 450
C to obtain maximum liquid yield (17 wt.%) in the given experimental setup The frequency factor of 98 s1 and activation energy of 29 kJ/mol are the lowest at 40 bar reaffirming the fact that heat and mass transfer limitations are the least at

40 bar pressure[37,38]

Hydrogen

• Selective bond cleavage is higher

• Tendency to form more vapours which repolymerise into char

Nitrogen

• Random cleavage

• More gaseous products

when secondary reactions

occur

Fig 8 Mechanistic differences between slow and hydropyrolysis.

R Singh et al / Renewable Energy xxx (2016) 1e12

5.5 Catalytic pyrolysis

5.5.1 Nitrogen atmosphere

Pyrolysis can be carried out in the absence (thermal

decompo-sition) or presence of catalyst (catalytic decompodecompo-sition) The basic

criterion to use the catalysts in any conversion process is to carry

out the reactions at lower temperatures compared to thermal runs

The decomposition behaviour and product profile of the thermal

and catalytic pyrolysis of wheat straw and wheat husk and the

ef-fect of position of catalyst for wheat straw and wheat husk has been

studied Solid phase contact (SPC) and vapour phase contact (VPC)

where catalyst is mixed with the feed and kept separately in a

holder respectively have been studied The degradation mechanism

in the SPC of feed is very much different than in the case of VPC of

feed and hence, the products formed are different in both the cases

From the above experiments, it can be seen that VPC has produced

higher and effective conversion of biomass to bio-oil This can be

useful in a way that it reduces the cost of catalyst regeneration as it

is easy to remove char from the catalyst used in VPC than in SPC In

the case where the catalyst is cheap e.g., as waste of some other

process industry such asfly ash etc it can be thrown without being

regenerated The catalyst used and its contact method has to be

modified according to the requirement of the end products Hence,

a detailed understanding of these mechanisms is very much

required to develop an optimised process for the pyrolysis of

biomass to produce value added hydrocarbons in a sustainable

manner[30]

5.5.2 Hydrogen atmosphere

Catalytic hydropyrolysis of jatropha seed de-oiled cake has been

carried out at optimum conditions of 450
C and 40 bar pressure of

hydrogen It can be observed that the feed quantity plays a major

role in the product yield When all the other conditions are

main-tained constant other than the feed weight, it has been observed

that the bio-oil yield has reduced by half with a corresponding

increase in the biochar and gas yield This indicates that the feed

weight is also an important parameter in the hydropyrolysis of

jatropha seed de-oiled cake Some of the other observations from

the experimental results are that the catalytic experiments have

produced more oil than the thermal runs In case of n-ZSM-5 and

feed of particle size 0.5e2 mm, plain support has given maximum

liquid yields and maximum conversion when 0.5 g catalyst was

mixed with 20 g feed When Mo was incorporated it is seen to have

triggered secondary reactions thereby producing more char

Parti-cle size is also a very important reaction parameter for the process

of hydropyrolysis as differences in product yields have been

observed with changes in the particle size of the feed used With

increase in the amount of n-ZSM-5 added to feed with particle size

of<0.149 mm, the conversion was seen to reduce from 54% to 52.5%

and then to 44% The incorporation of metal did not make any

significant difference in the yields of bio-oil in case of n-ZSM-5

though the conversion decreased Highest conversion of 54% was

found in case of n-ZSM-5 added in the order of 0.25 g In case of

hierarchical zeolites, the increase in the amount of FAPTMS

increased the bio-oil content When Mo was impregnated it did not

produce any significant increase in the product yields But when Ni

was impregnated, it is observed to have produced maximum

amount of bio-oil among all experiments in this feed This might be

attributed to the hydrogenation activity of Ni increasing the bio-oil

yields The yields of h-zeolite Beta (8%) and h-ZSM-5 with higher

FAPTMS are similar It can be concluded that the presence of

cata-lyst plays a key role in the hydropyrolysis of biomass The reaction

mechanism in the absence and presence of catalysts is different as

evident from the differences in the product profile and distribution

Thermal and catalytic hydropyrolysis has been carried out using

hierarchical and nanocrystalline zeolites for the first time A maximum bio-oil yield of 19.1 wt.% has been obtained using nanocrystalline zeolite In addition to pressure and temperature, particle size and feed weight are also important operating param-eters for the process of hydropyrolysis In the presence of catalysts, selective cracking is observed to have happened and majorly aliphatic type compounds have been observed in the process

6 Hydrothermal liquefaction Hydrothermal liquefaction is a process for obtaining fuels/ chemicals from biomass in the presence of sub/supercritical solvent

at moderate to high temperature (250e350
C) and pressure (5e25 MPa) The hydrothermal liquefaction can be applied to a variety of biomass feedstocks without any drying Superior to py-rolysis hydrothermal liquefaction produces better quality bio-oil containing various chemicals including vanillin, phenols, alde-hydes, and organic acids, etc Subcritical water has unique prop-erties different from water at ambient conditions that include a high ion product (Kw) and a low dielectric constant, which are favourable for promoting reactions without catalysts The changes and optimization of reaction parameters and catalysts can produce the functional hydrocarbons/specialty chemicals in a single step

[39] 6.1 Hydrothermal liquefaction of different feedstocks Hydrothermal liquefaction of high or low moisture containing lignocellulosic and algal biomass for effective and complete con-version of organic content (lignin, cellulose and hemicellulose) into value added hydrocarbons has been carried out to understand the structure activity relationship with respect to the different feed-stocks The various works carried out include:

 Hydrothermal liquefaction of lignocellulosic biomass compo-nents (Cellulose and lignin)

 Utilization of cellulose using hydrothermal approach to func-tional chemicals

 Lignin valorisation to substituted phenols and aromatic ethers using organic solvents

 Hydrothermal liquefaction of agricultural residues (Wheat Husk, Wheat Straw, Sugarcane Bagasse, Rice Straw and Tama-rind Seed Husk)

 Effect of reaction environment on hydrothermal liquefaction of rice straw for production of functional chemicals

 Hydrothermal liquefaction of forest biomass residue (Pine wood, Deodar)

 Hydrothermal liquefaction of aquatic biomass (water hyacinth, microalgae, macro algae)

 Effect of macro algae composition on product distribution and nature of products

 Effect of organic solvents on hydrothermal liquefaction of various second and third generation biomass

6.2 Behaviour of different biomass feedstocks under hydrothermal liquefaction

The work focused on the utilization of sub-critical water as a reaction medium as well as acid/base catalyst to depolymerise various biomass feedstocks to value added hydrocarbons A single step method for the liquefaction of biomass has been employed to get maximum liquid products yield The fundamental studies to understand the role of individual biomass components (cellulose and lignin) on the production of valuable hydrocarbons and its

R Singh et al / Renewable Energy xxx (2016) 1e12

comparison with real biomass has been carried out The reaction

conditions have been optimized in order to get maximum

conver-sion to liquid products as well as maximum bio-oil yield The

experimental studies of lignin and cellulose were carried out using

water and alkaline catalysts and characterization of the various

products was carried out Conversion of cellulose into high yields of

methyl glucosides ca.90% over sulfonated carbon based catalyst

was carried out The carbon catalyst was demonstrated to be stable

forfive cycles with slight loss in catalytic activity[40]

Hydrother-mal conversion of lignin to substituted phenols and aromatic ethers

was performed using methanol and ethanol at various

tempera-tures (200, 250 and 280
C) and residence times of 15, 30 and

45 min FTIR and 1H NMR showed the presence of substituted

phenols and aromatic ethers in liquid products[41] Hydrothermal

liquefaction of rice straw and tamarind seed husk was also carried

out using organic solvents Thermal and catalytic hydrothermal

liquefaction of water hyacinth was performed at temperatures from

250 to 300
C under various water hyacinth:H2O ratio of 1:3, 1:6

and 1:12 under various residence times (15e60 min The use of

alkaline catalysts significantly increased the bio-oil yield Bio-oil

with high aliphatic carbon content was observed [42] De-oiled

microalgae cake obtained after the lipid extraction from

micro-algae was utilized through hydrothermal liquefaction to produce

some value added hydrocarbons A comparative study on

hydro-thermal liquefaction of various macro algae feedstocks viz Ulva

fasciata (MA’UF), Enteromorpha sp (MA’E) and Sargassum

ten-errimum (MA'ST) was carried out to understand the effect of the

compositional changes of macro algae samples on product

distri-bution and nature of products Maximum conversion of 81% was

observed with MA’UF FTIR and NMR spectra showed high

per-centage of aliphatic functional groups for all bio-oils[43]

Hydro-thermal liquefaction of U fasciata (MA’UF) was also carried out

using organic solvents (CH3OH and C2H5OH) The use of alcoholic

solvents significantly increased the bio-oil yield and the acids

ob-tained in bio-oil to the corresponding methyl and ethyl esters The

results showed that liquefaction with supercritical alcohols is an

effective way to produce functional hydrocarbons for chemical

feedstock[44] The preliminary studies on hydrothermal

liquefac-tion of rice straw using solid catalysts have been performed There

was no significant change in the product distribution using solid

catalysts However, the use of solid catalysts increased the TOC

content in aqueous products compared to thermal case indicating

the maximum carbon presence in aqueous products

It was observed that addition of alkali enhances the inherent

function of high temperature water, leading to high yields of

desired products (bio-oil) Appropriate use of bases/acids and

oxi-dants based on an understanding of the reaction mechanism can

lead to an excellent yield of the desired product Properties of sub/

supercritical solvent abolish the effect of biomass particle size or

heating rates on bio-oil yield since it can act both as a heat transfer

medium and as an extractant during liquefaction Solvent under

sub/supercritical conditions help to overcome the heat transfer

limitations that makes hydrothermal liquefaction relatively

inde-pendent of the size of biomass particle or heating rates The

hy-drothermal liquefaction of carbohydrates (cellulose and

hemicelluloses) generally leads to a mixture of water soluble

hy-drocarbons Conversion products include acetic acid, formic acid,

lactic acid, levullinic acid, 5-hydroxymethyl-furaldehyde, and

2-furaldehyde etc Hydrothermal liquefaction of lignin is a

prom-ising way of recovering phenolics rich bio-oils Both aromatic

aldehyde and phenolic compound are important chemical

in-termediates It was also observed during this work that

hydro-thermal liquefaction using organic solvents is beneficial for high

liquid products Temperature, type of biomass, solvent used (water

or organic) are major parameters that influence yield and

composition of bio-oil It was observed that compositional varia-tions in biomass types cause changes in product distribution and nature of products obtained from hydrothermal liquefaction since lignin, hemicelluloses, and cellulose behave differently during hy-drothermal liquefaction and the interaction between these com-ponents and products derived from the can greatly influence the product distribution and nature of products

Catalytic hydrothermal liquefaction of wheat husk using alkali catalysts gave high yield of light hydrocarbons (ether fraction) Alkaline catalysts promote the decomposition of lignin fraction to low molecular weight phenolic compounds[45] From hydrother-mal liquefaction of rice straw carried out using different reaction environments viz N2, O2 and CO2 it was observed that inert at-mosphere (N2) gave highest bio-oil yield (17 wt%) as well as con-version (78%) compared to O2and CO2on basis of gravimetric yields

as well as on organic carbon basis GC/MS showed that bio-oil ob-tained under different reaction conditions was composed of phenol, guaiacol, catechol, syringol etc and their derivatives[46]

To see the effect of organic solvents on whole biomass, hydro-thermal liquefaction of rice straw was also carried out using methanol and ethanol at different conditions of temperature and residence time Similar to the biomass components, high liquid product yield (47 wt%) was obtained using organic solvents Liquid products were mainly composed of alkyl derivatives of phenol and guaiacol[47]

To understand the effect of different biomass with varying composition, hydrothermal liquefaction study of agricultural and forest biomass residues was carried out The agricultural biomass (wheat straw and sugarcane bagasse) exhibited higher conversion under thermal and catalytic conditions compared to forest biomass (pine wood and deodar) Alkaline catalysts (KOH and K2CO3) have found an important effect on the decomposition of both agricul-tural and forest biomass residue in terms of both bio-oil yield and conversion

K2CO3showed higher catalytic activity in terms of both bio-oil yield as well as conversion for agricultural (wheat straw and sug-arcane bagasse) biomass compared to forest (pine wood and deodar) biomass Varying biomass composition had a significant effect on bio-oil yield as well as conversion Alkali catalysts also showed different activity for agricultural and forest biomass res-idue[48]

7 Centralised and decentralised biorefinery systems in Indian scenario

Bio-refineries are seen as the best future option for the pro-duction of energy in a sustainable and environment friendly manner Biomass is a much dispersed resource and its availability is also too varied depending on several factors Several region specific processes have to be developed and deployment of the same has to

be carried out based on supply and demand scenario With this background, slow pyrolysis can be carried out in decentralised lo-cations using resources from a group of villages or so The products can be used at that location or be transported to a centralised unit for further upgradation In case where centralised facilities can be built, hydropyrolysis can be a better option for the production of renewable hydrocarbons

8 Future prospects

It has been well established that there is a requirement to move towards sustainable sources of energy and hydrocarbons Fuels and chemicals can be effectively produced from biomass which is a renewable source of energy The thermo-chemical methods have the maximum potential to produce value added hydrocarbons in an

R Singh et al / Renewable Energy xxx (2016) 1e12