Science of the Total Environment 668 (2019) 743–759

Contents lists available at ScienceDirect

Science of the Total Environment

journal homepage: www.elsevier.com/locate/scitotenv

The effect of coal alternative fuel from municipal solid wastes employing
hydrothermal carbonization on atmospheric pollutant emissions
in Zimbabwe
Andile B. Maqhuzu, Kunio Yoshikawa, Fumitake Takahashi ⁎
Department of Transdisciplinary Science and Technology, School of Environment and Society, Tokyo Institute of Technology, 4259, Nagatsuta, Midori-ku, Yokohama 226-8703, Japan

H I G H L I G H T S G R A P H I C A L A B S T R A C T

• Effects of waste-to-energy on air pollut-

ant emissions in Zimbabwe were
assessed.
• 289,300 Mg of coal-alternative fuel can
be produced from collected household
wastes.
• N2.2% and 4.2% reductions in green-
house gas and SOx are possible.
• Waste-to-energy increases NOx emis-
sion by 18% for specific power genera-
tion scenarios.
• Waste-to-energy will be promising only
when appropriate investments are
done.

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

Article history: The vast increase of municipal solid waste (MSW) generated in Zimbabwe coupled with a severe energy crisis have
Received 13 December 2018 made waste-to-energy technology more attractive and necessary. Coal-alternative solid fuel production from MSW
Received in revised form 14 February 2019 though hydrothermal carbonization can play a critical role to improve both waste management and energy supply.
Accepted 4 March 2019
Moreover, MSW conversion to a carbon neutral solid fuel that can be burnt in existing coal-fired power stations
Available online 05 March 2019
might reduce greenhouse gas (GHG) emissions despite GHG releases from waste collection, waste conversion to
Editor: Daniel CW Tsang fuel, and fuel transportation processes. The purpose of this paper is to investigate present MSW generation in
Zimbabwe, its characteristics as a fuel source, and the impact of coal-alternative solid fuel production from MSW
Keywords: using hydrothermal carbonization technology on GHG and other air pollutant emissions. Four different scenarios
Municipal solid wastes based on the balance between fuel supply and demand were tested in this paper. The results suggest 0.54 ± 0.14
Waste to energy kg/capita/day of MSW generation in Zimbabwe and about 1051.7 ± 270.7 Gg of annual MSW generation from
Hydrothermal carbonization the current urban population. 289.3 Gg of coal-alternative solid fuel production was expected from domestic
Atmospheric pollutant emission MSW collectable in urban areas. The model predicted that co-burning of alternative fuel in coal-fired power plants
Zimbabwe
could reduce the methane potential of household waste from 62,200 to 15,800 Mg CH4 per year. Under the best pos-
sible scenario, it could reduce SOx emissions by 4.2%, CH4 emissions by 4.5%, CO2 emissions by 3.1%, and Global
Warming Potential by 2.2%. On the other hand, NOx emissions would increase by 18%. If without additional instal-
lation of air pollutant control devices in power plants, waste-to-energy generates a trade-off between global
warming and acid rain. In addition, geological locations generate a large demand/supply gap of alternative fuel
and regulate maximum available consumption of alternative fuel.
© 2019 Elsevier B.V. All rights reserved.

⁎ Corresponding author.
E-mail address: takahashi.f.af@m.titech.ac.jp (F. Takahashi).

https://doi.org/10.1016/j.scitotenv.2019.03.050
0048-9697/© 2019 Elsevier B.V. All rights reserved.
744 A.B. Maqhuzu et al. / Science of the Total Environment 668 (2019) 743–759
1. Introduction
Zimbabwe is a non-coastal, low income republic situated in southern
Africa with an estimated population of 15.6 million in 2015. Zimbabwe's
economy is anchored by mining, tourism and agriculture and the Gross
Domestic Product (GDP) per capita in 2017 was pegged at US$1333.39
(World Bank, 2019). Waste management still remains a key social and
engineering concern due to the health hazards associated with poor
waste treatment. Although municipal solid waste (MSW) disposal is
shifting from mere landfilling to more integrated and sustainable sys-
tems such as waste-to-energy (WTE) in developed nations, the instiga-
tion of such viable waste management practices in developing nations
such as Zimbabwe is still confronted by a lack of political willpower,
weak financial capacity and a lack of technical expertise. Zimbabwean
municipalities, in particular, are incapacitated as most urban areas are
teeming with uncollected waste, adversely affecting the aesthetics and
more importantly, making the inhabitants more susceptible to health
risks. Local authorities are confronted by unprecedented levels of
MSW generation compounded by financial constraints which are damp-
ening their efforts in curbing the rising problem (Masocha, 2006;
Mudzengerere and Chigwenya, 2012; Mwanza and Phiri, 2013). This
has taken a toll on the existing waste management and sanitation sys-
tems as demonstrated by the unusually high fatality rates in urban
areas during a 2008 cholera epidemic (Mbiba, 2014; W.H.O., 2008).
Against this backdrop, Zimbabwe is confronted by a severe energy
crisis characterized by insistent load shedding and power blackouts
which undermine the development of the economy and people's liveli-
hood. An estimate of the cost of electricity supply outages in Zimbabwe
in 2009, was about US$1.8 billion (Kaseke, 2014). With a waste manage-
ment crisis on the one hand and an energy deficit on the other, appro-
priate waste management based on social, economic, and
environmental considerations is essential for the environment-
friendly development of Zimbabwe. As a reliable supply of electric
power is also crucial from the viewpoints of economic growth, hydro-
thermal carbonization (HTC), a WTE technology, could be a promising
waste management system in Zimbabwe. Proponents of hydrothermal
technology cite the following advantages; versatility in feedstock and
its preparation (dewatering is not needed), fugitive greenhouse emis-
sions are low, solid energy content is enhanced, moderate process con-
figuration, possibility of integrated energy recovery, thermal
degradation of hazardous substances is possible, results in high
grindability and drying capacity of the product and superior dechlorina-
tion capabilities that impede corrosion in downstream applications
(Hwang et al., 2012; Jin et al., 2013; Lu et al., 2012; Zhao et al., 2014).
Opponents of the process cite the low heating value of the hydrochar
leading to low combustion efficiency, unpredictability in ignition and
combustion behaviour and need to adapt firing equipment for the fuel
(Lu et al., 2011; Muthuraman et al., 2010a). Jin et al. (2013) and
Muthuraman et al. (2010b) report improved emissions for blends of
up to 20% and heterogeneous burning profiles beyond that limit. The
objective of this paper is to assess the environmental impact of hydro-
thermal treatment of Zimbabwe MSW (ZMSW) for alternative fuel pro-
duction and waste management. The potential maximum amount of
consumable Zimbabwe MSW (ZMSW) derived hydrochar for co-firing
in power plants, its environmental impact (CO2, SO2 and NOx emis-
sions) and others will be investigated under some scenarios. These re-
sults will be helpful assessing the feasibility of hydrothermal
treatment of ZMSW and to find the necessary conditions for an appro-
priate waste management system.
2. Waste generation in Zimbabwe
The collection of ZMSW is confined to the urban population of about
5.38 million people, albeit with meagre resources available for a major-
ity of local authorities, the waste management system still leaves a lot to
be desired. Local authorities' capacity to collect, process and dispose of
waste is limited because of inadequate infrastructure, political priority
and awareness (Makwara and Magudu, 2013). Refuse compactors, for
example, are constantly in a state of disrepair and more often than
not, councils have to resort to the use of inappropriate equipment
such as tractors to ferry waste to dumpsites (Jerie, 2014). It is worrying
to note that despite council's contention with illegal dumpsites, ob-
servers claim to have witnessed council tippers divert waste to open
gravel pits in close by farms instead of legal dumpsites (Mandevere,
2015). Moreover, a study by Muchandiona (2013) reveals that the
final disposal of ZMSW is not ecologically sound as 98% of all legally op-
erated landfills are not sanitary. These ‘landfills’ are not suitably main-
tained and due to a lack of daily cover, are exposed to the wind and
rodents which scatter the waste. They are also a fire hazard; a serious
fire broke out in October 2013, burning for two weeks and leading to
one fatality (Kharlamova et al., 2016).
There are contentions between urban dwellers and municipalities
due to the growing complexity of ZMSW composition. In the second
largest city, Bulawayo, local authorities refuse to collect disposable dia-
pers from residents (Mbiba, 2014). Another emerging issue is the cur-
rent state of the healthcare sector's waste disposal practices.
Incinerators at both the public and private clinics are either absent or
dysfunctional (Kühling and Health, 2014). Dumping in rivers is com-
monly opted for by Zimbabwe's industrial sector, further worsening
the negative impacts on the environment (Lopes, 1996). The collection
of ZMSW in urban areas is mostly irregular with most high-density
areas going for long periods without the collection of their waste. This
inevitably prompts more illegal dumping and rampant burning of
waste as the locals take matters into their own hands (Makwara and
Magudu, 2013). Mandevere (2015) reveals that inhabitants of low-
density areas in Harare, the capital, have resorted to privately hiring col-
lectors or alternatively transporting their own waste to a local disposal
site independently as they have the financial capacity to do so. As a re-
sult, the full brunt of the substandard waste management system is felt
more by urban dwellers in high density areas compared to their low-
density counterparts. Scenes of heaps of illegally dumped litter
compounded by public urination, open defecation and sewage sludge
from burst pipes have become a trademark for these low-income
high-density areas (Huvengwa, 2012; Makwara and Magudu, 2013;
Mangundu et al., 2013).
To assess the impact of an HTC waste management strategy for
Zimbabwe, it is necessary to identify the nature and composition of
the country's waste. There is spatial and temporal variation in MSW
composition and generation rates. Therefore, as part of our attempt to
implement an HTC waste management system, the statistical analysis
of the reported MSW characteristics spanning different locations, in-
come levels and time is an essential step. Other researchers have fo-
cused on the environmental impact of Zimbabwe's waste (Chifamba,
2007; Makoni et al., 2004) and possible solutions alluded to by
Kharlamova et al. (2016), Makura (2014), Makwara and Magudu
(2013), Mandevere (2015), Mubaiwa (2006), and Zamba (2014). Esti-
mations on the energy potential of Zimbabwe's MSW have been given
by Jingura and Matengaifa (2009), Jingura et al. (2013), and Scarlat
et al. (2015) who have gone on further to suggest its exploitation via
biogas technologies (either anaerobic digestion or landfill gas recovery).
Mbohwa and Zvigumbu (2007) suggested incineration while Gwenzi
et al. (2015) focused on biochar production via pyrolysis. None of the
studies have considered the impact of these technologies on the envi-
ronment at a national level.
Among researchers there is a clear consensus about the positive cor-
relation between MSW generation rates and economic status for devel-
oping nations (Shekdar, 2009). An improvement in Zimbabwe's
economic growth alongside rapid urbanization is likely to surge MSW
generation rates even higher and an alternative strategy to avert the im-
minent social and environmental problems associated with rising MSW
quantities will be needed. Hydrothermal treatment of waste could be an
effective solution for converting this waste into a resource thereby
 A.B. Maqhuzu et al. / Science of the Total Environment 668 (2019) 743–759
creating a much cleaner environment. Berge and Flora (2015), Hwang
et al. (2012), Jin et al. (2013), Lu et al. (2011), Muthuraman et al.
(2010a), Muthuraman et al. (2010b), Prawisudha (2011), and Zhao
et al. (2014) conducted extensive research on MSW conversion to
hydrochar. There is widespread knowledge on the chemical nature
and economic viability of this process (Funke and Ziegler, 2010; Reza
et al., 2014). There are also reports on real world applications of the
technology (EASME, 2016; Novianti et al., 2014; Safril et al., 2017).
However, a gap exists in the scientific literature on the environmental
impact of a large-scale HTC system.
3. Materials and method
3.1. System scope of modelled coal-alternative fuel production from ZMSW
The process of production of coal-alternative fuel (ZMSWH) from
ZMSW is executed within several operations as shown in Fig. 1. The
first stage involves the conversion of ZMSW to an organics fraction
after the removal of metals, glass, ceramics (incombustibles), etc. This
is essentially a heavy fraction of small-sized particles containing or-
ganics, plastics, wood, rubber, etc. This fraction, ZFOMSW is placed
into the hydrothermal carbonization reactor and converted into
ZMSWH (hydrochar). Within the discussed system of ZMSWH produc-
tion from ZMSW, the environment, in terms of direct pollutant emis-
sions, is affected by the following processes: collection and
transportation of ZMSW and ZMSWH (diesel fuel combustion in vehi-
cles), shredding of ZMSW (diesel consumption), ZOFMSW hydrother-
mal carbonization (energy coal-fired boiler) and receiving area
operations (diesel fuel combustion). Other unit operations consume
electricity and pollutant emissions are mainly from power generation
Fig. 1. Overview of coal-alternative fuel production.
745
at the coal-fired stations. The methodology of the research involved
the collection of data on ZMSW characteristics, general MSW properties
and Zimbabwe's thermal power generation from previous studies for
the development of a numerical model for estimating a) the quantity
of ZMSW generated and its composition b) the quantity and composi-
tion of ZOFMSW c) the potential amount of ZMSW derived hydrochar
(ZMSWH) available for co-firing d) emissions from the hypothetical
co-burning of ZMSWH in Zimbabwe's thermal power stations. This
entailed the estimation of emissions related to ZMSW collection, its con-
version to ZMSWH, coal-mining and hypothetical co-combustion with
ZMSWH so that a full environmental impact of this waste to biocoal sys-
tem could be ascertained.
3.1.1. Model limitations
Some comments are warranted on the limitations of our model.
Given the scale and complexity of an HTC based waste-to-energy sys-
tem our simplified model cannot fully capture all parameters. For exam-
ple, HTC produces gases and aqueous chemicals, in addition to the
desired solid product. A commercial HTC plant would have to recircu-
late and/or properly dispose of its process water to reduce its impact
on the environment. This prerequisite adds to the cost, complexity,
and emissions of any waste-to-energy system that employs HTC tech-
nology. We did not make any estimations on either the amounts or
quantities of other product streams. Coal usage and its associated toxic
emissions (trace metals like mercury) and trace organics
(e.g., dioxins) were not modelled. Despite the envisaged environmental
benefits, the capital costs could possibly be a hurdle, particularly for HTC
which is not yet widely established. An economic analysis, however,
was not included in this study.
746 A.B. Maqhuzu et al. / Science of the Total Environment 668 (2019) 743–759

3.2. Estimation of the quantity and composition of ZMSW basis for drawing comparisons between some scenarios in subsequent
sections of this paper. The equation for estimating methane potential
An estimation of ZMSW generation rate was determined by using is given by;
GDP per capita (current US$) data from 2012 (World Bank) and MSW
generation rates of 38 low income countries reported by Hoornweg Methane emissions ðGg=yearÞ
and Bhada-Tata (2012). The underlying assumption is the correlation ¼ ðMSWT  MSW F  MCF  DOC  DOC F  F  ð16=12Þ−RÞ
between GDP per capita and MSW generation rate. A linear relationship  ð1−OXÞ ð3Þ
between the two variables was assumed so that Zimbabwe's current
MSW generation could be ascertained. Regression analysis was carried where MSWT is total MSW generated (Gg/year), MSWF is fraction of
out on the data. This estimate was averaged together with the reported MSW disposed to solid waste disposal sites, MCF is methane correction
values shown Table 1. The collectable ZMSW was found by applying the factor (fraction), DOC is the degradable organic carbon (fraction)
formula: (kg C/kg SW), DOCF is the fraction DOC dissimilated, F is the fraction
of methane in landfill gas, 16/12 represents the conversion of C to
MSW collected annually CH4, R is the recovered CH4 (Gg/year) and OX is the oxidation factor.
¼ MSW generation per capita ðkg=capita=dayÞ Default values were used except for MSWT, MSWF, MCF and DOC
 Total Urban Population  365 days  Collection Efficiency ð1Þ (carbon content not originating from plastics but from organic biode-
gradable waste) which were determined from the model. R was as-
Mean collection efficiencies were reported to be 63.9% sumed to be 0 as the recovery of LFG is not practiced in Zimbabwe.
(Muchandiona, 2013) and 85.2% (Jingura and Matengaifa, 2009) hence MSWF was assumed to be equal to be 100% (due to predominantly
an average value of 74.6% was assumed. The composition of ZMSW poor solid waste management, landfills and open dumps are not so dif-
was ascertained by statistically analysing data from previous field stud- ferent) and MCF was taken as 0.4 for unmanaged, shallow sites (b5 m).
ies on ZMSW shown in Table 2. The total population of Zimbabwe is es-
timated to be 17.09 million with 31.4% living in urban areas (AFDB, 3.4. Estimation of the quantity and characteristics of ZOFMSW
2019). ZMSW is collected in urban areas only.
The process flow and equipment for converting ZMSW to ZOFMSW
3.3. Energy content, proximate and ultimate analysis of ZMSW is shown in Fig. 1. The mass of each ZMSW component, i, remaining
after its removal by each unit operation, jn, is calculated by multiplying
In order to determine the proximate analysis and ultimate analysis its total weight input by the separation efficiencies of each equipment jn
of ZMSW on a dry basis, an estimate of the moisture content was carried as shown in Eq. (4). The separation efficiencies are organized in a matrix
out. The moisture and energy content of ZMSW was calculated using as shown in Table 5.
mid-values of the typical range of values shown in Table 3. The proxi-  
mate and ultimate analysis of ZMSW was then predicted using mean W in ¼ miin  f j1  f j2 ⋯f jn ð4Þ
values reported in literature which are summarized in Table 4. An esti-
mation of the energy content of the ZMSW was obtained by using aver-
age heating values and verified with the Modified Dulong Formula.
Wi n is the mass of waste fraction i remaining after passing through
Energy content ðKJ=kgÞ; E ¼ 337C þ 1428 ðH–O=8Þ þ 95S ð2Þ equipment j1 to jn (Mg)
mi in is the initial mass of waste fraction i to be processed (Mg)
In order to validate our results, our simulated ZMSW energy content, fj n is the percentage of waste fraction remaining after passing
proximate and ultimate composition was compared to field data re- through equipment jn (%).
ported by Makarichi et al. (2018a). The methane potential of ZMSW
was determined by using the Intergovernmental Panel on Climate Once the mass components for ZOFMSW were calculated, the prox-
Change (IPCC) default method (Intergovernmental Panel on Climate imate, ultimate and energy content were predicted using Tables 3 and 4.
Change, 1996) which assumes that all potential methane is released in Emissions emanate from the shredders driven by a diesel powered
the year the MSW is discarded. This is unrealistic but it will provide a 600HP engine. In the receiving area, a mobile pre-shredder unit will
be used for the tearing pre-shredding of very large ZMSW components.
Fuel Consumption for a 600HP diesel engine power was calculated
Table 1 using Eq. (5).
Reported Zimbabwe (Africa) MSW waste generation rates.
Fuel consumption; L=h
MSW generation rate Reference
kg/capita/day ¼ ðspecific fuel consumption ðkg=ðkWhÞÞ  rated engine power ðkWÞÞ
 engine load factor=Fuel Specific Weight ð5Þ
0.700 (Simelane and Mohee, 2012)
0.500 (Mshandete and Parawira, 2009)
0.500 (Jingura et al., 2013) The specific weight of diesel is 0.84 kg/L (Khoshnevisan et al., 2014)
0.530 (Hoornweg and Bhada-Tata, 2012) and specific fuel consumption rate for many diesel engines lies between
0.311 (Practical Action, 1999) 0.21 and 0.26 kg/(kWh) (Klanfar et al., 2016). The shredders are as-
0.500 (Jingura and Matengaifa, 2009)
0.790 (Mohee and Simelane, 2015)
sumed to operate for 4380 h/year (12 h/day × 365 days). From the vol-
0.290 (IPCC, 2006) ume of diesel combusted, the following emission factors were used to
0.529 (Scarlat et al., 2015) estimate pollutant air emissions. For every litre of diesel combusted,
0.440 (Friedrich and Trois, 2011) 2.70 kg of CO2 (Palmer, 2007), 0.07 g of NOX (EPA, 2014), 4.72 g of
0.600 (Hoornweg et al., 2015)
SO2 (Olatunji et al., 2015) and 0.15 g of CH4 (EPA, 2014) are released.
0.42 (Muchandiona, 2013)
0.700 (Achankeng, 2003)
0.640 (International Energy Agency) 3.5. Estimation of the yield and composition of ZMSWH
0.480 (Makarichi et al., 2018b)
0.640 This study In this paper we wish to use a simple model to predict the quantity
0.54 ± 0.14 Overall average
and elemental composition of ZMSWH. This may seem at odds with the
 A.B. Maqhuzu et al. / Science of the Total Environment 668 (2019) 743–759 747

Table 2
Reported composition of ZMSW from literature.

Location MSW components (%)

Paper & Glass & Metals Plastics Leather & Textiles Yard waste & Wood Putrescibles Miscellaneous Reference
cardboard ceramics rubber sweepings

Harare
30.0 10.0 10.0 10.0 5.0 5.0 0.0 5.0 5.0 20.0 (Togarepi and Tsiko, 2012)
Sunningdale 11.7 0.8 0.8 6.8 0.3 2.1 0.0 0.6 75.9 0.8 (Togarepi and Tsiko, 2012)
Mabelreign 13.5 2.3 1.1 8.0 0.0 1.1 0.0 0.3 73.6 0.2 (Togarepi and Tsiko, 2012)
Borrowdale 23.1 6.5 4.0 11.7 0.4 3.8 0.0 0.7 47.7 2.3 (Togarepi and Tsiko, 2012)
Glenview Area 8 22.0 5.0 10.0 25.0 0.0 12.0 0.0 0.0 22.0 4.0 (Mangundu et al., 2013)
Mbare 6.0 2.0 12.0 11.0 0.0 4.0 16.0 0.0 28.0 21.0 (Zamba, 2014)
16.1 2.8 3.8 26.4 0.0 1.5 17.6 0.0 14.3 17.5 (Mbohwa and Zvigumbu,
2007)
Borrowdale 35.0 4.0 3.0 18 1.0 9.0 15.0 15.0 (Mandevere, 2015)
Gunhill 32.0 5.0 4.0 16 2.0 8.0 18.0 15.0 (Mandevere, 2015)
Mandara 33.0 4.0 3.0 16.5 2.5 8.0 17.0 16.0 (Mandevere, 2015)
Westlea 30.0 6.0 5.0 22 7.0 10.0 15.0 5.0 (Mandevere, 2015)
Milton Park 31.0 2.0 2.0 21 6.0 8.0 26.0 4.0 (Mandevere, 2015)
Waterfalls 29.0 4.0 3.0 20 9.0 7.0 23.0 5.0 (Mandevere, 2015)
Mbare 10.0 1.0 5.0 9 2.0 9.0 54.0 10.0 (Mandevere, 2015)
Budiriro 14.0 2.0 4.0 10 1.0 9.0 49.0 11.0 (Mandevere, 2015)
Highfield 13.0 3.0 3.0 8 2.0 9.0 55.0 7.0 (Mandevere, 2015)
13.0 4.0 4.0 23 1.0 10.0 12.0 0.0 28.0 5.0 (Makarichi et al., 2018a)

Epworth
3.3 4.0 1.9 12.4 0.1 5.1 2.3 0.0 46.4 24.5 (Makarichi et al., 2018a)

Bulawayo
30.0 0.4 0.6 42.0 0.0 0.0 4.0 0.0 23.0 0.0 (Mwanza and Phiri, 2013)
20.0 5.0 10.0 10.0 3.0 2.0 0.0 5.0 10.0 35.0 (International Labour
Organization)

Gweru
CBD 5.0 0.0 3.0 37.0 0.0 0.0 10.0 0.0 42.0 3.0 (Mangizvo, 2007)
Monomotapa 10.4 3.5 30.6 5.0 5.0 9.0 0.0 10.4 10.7 15.4 (Jerie, 2014)
Shamrock Park 10.3 5.0 39.6 5.7 3.0 1.0 0.0 12.1 6.4 16.9 (Jerie, 2014)
Mkoba 17.1 5.0 18.3 4.5 4.0 8.0 0.0 13.0 20.7 9.4 (Jerie, 2014)
Kudzanai/Kombayi 14.8 3.0 6.1 7.2 1.0 2.0 0.0 1.9 57.1 6.9 (Jerie, 2014)
Ascot 18.6 3.3 5.0 6.4 5.6 4.0 0.0 17.0 31.6 8.5 (Jerie, 2014)

Masvingo
11.0 0.0 5.4 9.0 1.1 2.1 0.0 0.0 70.4 1.0 (Huvengwa, 2012)
Mucheke 30.0 5.0 4.0 40.0 0.0 0.0 0.0 0.0 15.0 6.0 (Mangizvo, 2008)

Victoria Falls
34.0 6.0 5.0 15.0 2.0 4.0 0.0 0.0 26.0 8.0 (Masocha, 2004)

Mutare
Sakubva 27.0 5.0 6.0 23.0 0.0 3.0 3.0 1.0 32.0 0.0 (Manyanhaire et al., 2009)

Chinhoyi
Low density 18.7 8.9 3.0 6.4 9.3 5.4 0.0 0.0 43.3 5.0 (Sango, 2010)
High density 10.3 4.0 1.0 3.0 7.7 2.1 0.0 0.0 56.1 15.5 (Sango, 2010)
12.0 15.0 28.0 4.0 40.0 1.0 (Nyamombe and Gwisai,
2015)

Chegutu
Mubaira CBD 20.0 13.3 0.0 25.4 0.0 0.0 33.3 0.0 8.0 0.0 (Takaedza, 2014)
Mubaira 17.1 0.0 0.0 28.6 0.0 0.0 14.3 0.0 40.0 0.0 (Takaedza, 2014)
residential area

Chitungwiza
4.0 3.0 1.0 10 0.0 11.0 11.0 0.0 40.0 20.0 (Makarichi et al., 2018a)

highly complex nature of hydrothermal carbonization. Even with the
vast amount of research on the topic, the mechanisms and kinetics in-
volved in the process are still not fully understood (Román et al.,
2018). After sifting through some papers in the literature, the work by
Li et al. (2015) and Lin et al. (2016) stood out. Through the development
of multiple linear regression and regression tree models (based on ex-
perimental data on the carbonization of MSW and other forms of bio-
mass), the researchers were able to model product characteristics
based on the process conditions and feedstock chemical properties
very well. Multiple regression analysis is particularly useful for simulat-
ing experimental data (Bach et al., 2013). Lin et al. (2016) developed a
multiple regression model for predicting MSW-derived hydrochar, iter-
ating the findings from Li et al. (2015) and Román et al. (2018) that
hydrochar properties depend on both process conditions (temperature
and residence time) and feedstock properties (e.g., carbon content). Re-
sults from experiments indicate that hydrochar and gas yields are more
strongly influenced by temperature than by residence time. Table 6 pro-
vides a summary of how ZMSWH properties were predicted. With a rise
in temperature, the H/C and O/C atomic ratios decrease causing the
hydrochar to exhibit similar characteristics to lignite. The lower H/C
and O/C atomic ratios presumably lead to lesser emissions, moisture
contents and energy losses in the combustion process (Liu et al., 2013).
748 A.B. Maqhuzu et al. / Science of the Total Environment 668 (2019) 743–759

Table 3 Emissions from this stage, emissions are expected from the coal-
Typical moisture content and energy content of MSW components (O'Leary and fired boiler producing steam entering the hydrothermal carbonization
Tchobanoglous, 2002).
reactor (Fig. 1). The formation of hydrochar was found to have a total
Component % moisture content Energy content (MJ/kg) energy requirement of 2 2 MJ/kg of MSW on average (Prawisudha,
Range Typical Range Typical 2011). This helped to estimate coal requirements and associated emis-
sions related to ZOFMSW conversion to ZMSWH. The quantity of CO2
Food wastes 50–80 70 3.5–7.0 4.7
Paper 4–10 6 11.6–18.6 16.7 and SO2 from ZMSWH combustion was predicted from its elemental
Cardboard 4–8 5 14.0–17.4 16.3 composition (e.g. C + O2 → CO2, S + O2 → SO2). The calculated fossil
Plastics 1–4 2 27.9–37.2 32.6 carbon and biogenic carbon ratio was used to determine amount of car-
Textiles 6–15 10 15.1–18.6 17.4 bon dioxide emitted.
Rubber 1–4 2 20.9–27.9 23.3
Leather 8–12 10 15.1–19.8 17.4
Garden trimmings 30–80 60 2.3–18.6 6.5 3.6. Emissions estimates from Zimbabwe's coal-fired power stations
Wood 15–40 20 17.4–19.8 18.6
Glass 1–4 2 0.1–0.2 0.1
Tin cans 2–4 3 0.2–1.2 0.7 Annual coal consumption was estimated and emission factors ap-
Nonferrous metals 2–4 2 0 0.0 plied to this estimate to quantify emissions from thermal power gener-
Ferrous metals 2–6 3 0.2–1.2 0.7 ation. Zimbabwe's thermal power plants are old and inefficient and
Dirt, ashes, brick, etc. 6–12 8 2.3–11.6 7 breakdowns are frequent; the installed and dependable capacities are
Municipal solid waste 15–40 20 9.3–15.1 10.5
shown in Fig. 2. Dependability is attributed to a combination of ageing
power plants which have suffered years of neglect and minimal mainte-
nance, and inconsistent coal supplies as well as problems with the
From Lin et al. (2016) multiple regression model, the yield of transmission and distribution network. The hours of operation are sig-
hydrochar, Ysolid, can be predicted by; nificantly reduced due to factors such as breakdowns and unreliable
coal supply. Co-firing would not be expected to improve the depend-
Y solid ð%; dbÞ ¼ −0:78 17T þ 0:001153T 2 –0:2561t þ 0:000475t 2 ability. In terms of installed capacity, we are assuming the plants are op-
þ 0:0004888Tt þ 192:8 ð6Þ erating closer to their actual nameplate capacity but with similar hours
of operation to dependable capacity. Emissions related to coal use were
estimated for two scenarios; dependable and installed capacity. A quick
Their model identifies reaction temperature (T) and residence time
estimate can reveal the quantity of coal required for producing 1 MW
(t) as the main influencers of solid yield. Li et al. (2015) have more pa-
per year. Energy conversion takes place in two stages; the first part of
rameters and is influenced by feed properties (ZOFMSW proximate and
the conversion depends on the efficiency of the boiler and combustion.
ultimate composition in our case) as well experimental conditions.
About 88% on a higher heating value basis is the normal range for a well-
optimized power plant. However, Murehwa et al. (2012) report a value
Solid yield ð%; dbÞ ¼ −1:12Ashfeed −1:22VMfeed −1:58FCfeed of 70% for a Zimbabwean plant. The second part is the steam cycle effi-
þ 1:63Cfeed −4 ciency. Modern Rankine cycle, adopted in coal-fired stations, have effi-
: 89Hfeed −0:73Ofeed þ 0:43Solidsinitial −0 ciencies that vary from 32 to 42%. This is governed mainly by the
: 21Tfinal þ 9:70 ðHT=tÞ þ 1:36HR−22:19VR þ 212:76 ð7Þ
steam parameters. Higher steam pressure and temperatures in the
range of 600 °C and 227 atm achieve a 42% efficiency. Taking the
where Cfeed = carbon content of the ZOFMSW (%); Hfeed = hydrogen lower value of 32%, the overall conversion efficiency (30% × 70%) is as-
content of ZOFMSW (%); Ofeed = oxygen content of ZOFMSW (%); sumed to be 21%. The plants typically use 20–30% of the energy input ac-
Ashfeed = ash content of ZOFMSW (%); VMfeed = volatile matter content cording to Murehwa et al. (2012). The heat rate or heat input required
of ZOFMSW (%); FCfeed = fixed carbon content of ZOFMSW (%); to produce one unit of electricity (1 kWh), if the energy conversion is
Solidsinitial = initial feedstock concentration (%, solid); Tfinal = final re- 100% efficient, would be 3600 kJ. After considering the overall conver-
action temperature (°C); t = reaction time (min); HT = heating time sion efficiency, 17,142 kJ/kWh (3600/21%) is required. Zimbabwe
(min); HT/t = heating time to reaction time ratio; HR = heating rate steam coal has an average heating value of 25.45 MJ/kg (refer to
(°C/min); V = volume (mL); VR = volume ratio. Table 7), therefore to produce 1 kWh, 0.673 kg of coal
We assumed experimental conditions of t = 60.00 min, HR = 7.00 (17.14 MJ/25.45 MJ/kg) is required. A specific coal requirement of
°C, HT = 70.00 min, HT/t = 1.17, V = 250.00 mL, VR = 0.90. HTC exper- 0.673 Mg/MWh is our estimate. On a yearly basis this is 5892 Mg/MW.
imental results are representative of those obtained at a different reac- However, according to Kaseke (2014) electricity is available from the
tor scale when scaling up (Hoekman et al., 2014). grid only 62.75% of the time (out of 8760 h/year) due to plant

Table 4
Mean values for proximate and ultimate analysis of MSW components.

Component Proximate composition (dry basis) Ultimate composition (dry basis)

Ash VM FC C H N Cl S O P F Br

Paper & cardboarda 7.19 71.29 19.26 43.72 6.03 2.87 0.13 0.13 40.00 0.00 0.00 0.00
Metal, glass & ceramicsb 100.00 0.00 0.00 0.00 0.0 0.0 0.00 0.00 0.00 0.00 0.00 0.00
Plasticsc 2.88 90.28 2.60 71.97 9.38 0.50 5.71 0.20 7.28 0.00 0.00 0.00
Leather, textiles & rubberd 7.10 79.76 9.93 55.48 7.19 3.31 0.69 0.52 31.16 0.00 0.00 0.00
Yard waste & sweepingse 5.65 55.03 13.51 47.10 5.96 2.31 0.37 0.21 37.75 0.00 0.00 0.00
Woodf 1.53 77.23 14.54 49.07 6.02 0.31 0.10 0.08 42.88 0.00 0.00 0.00
Putresciblesg 4.61 60.97 7.61 52.14 7.51 2.06 0.80 0.23 31.53 0.00 0.00 0.00
Miscellaneoush 20.77 58.14 9.96 39.08 4.76 1.58 0.63 0.42 27.75 0.40 0.01 0.00

Sources: a, c, dChandrappa and Brown (2012), Gidarakos et al. (2006), Liu and Liptak (1999), Roberts (2015), and Williams (2013); bLiu and Liptak (1999); e, fChandrappa and Brown
(2012), Liu and Liptak (1999), Roberts (2015), and Williams (2013); gChandrappa and Brown (2012), Gidarakos et al. (2006), Hussain et al. (2014), Liu and Liptak (1999), Roberts
(2015), and Williams (2013); hKathirvale et al. (2004), Katiyar et al. (2013), Liu and Liptak (1999), Omari et al. (2014), Oumarou et al. (2012), Valkenburg et al. (n.d.), and Williams (2013).
 A.B. Maqhuzu et al. / Science of the Total Environment 668 (2019) 743–759 749

Table 5
Percentage of waste component retained (i) after passing through unit operation (j).

Waste component (in) Equipment (jn)

Bag rippera (j1) Manual sortingb (j2) Magnetc (j3) Eddy current separatord (j4) Shreddere (j5) Air classifierf (j6)

Paper & cardboard (i1) 100 95 100 100 100 85–99

Metal (ferrous) (i2) 100 95 12 100 100 2–10
Metal (non-ferrous) (i3) 100 95 100 13 100 45–65
Glass & ceramics (i4) 100 95 100 100 100 2–10
Plastics (i5) 100 95 100 100 100 85–99
Leather, textiles & rubber (i6) 100 95 100 100 100 85–99
Yard waste & sweepings (i7) 100 95 100 100 100 85–99
Wood (i8) 100 95 100 100 100 85–99
Putrescibles (i9) 100 95 100 100 100 85–99
Miscellaneous (i10) 50 95 80 80 100 85–99

a – Assume 50% of components in the miscellaneous category are combustible and remaining is left in the yard after inspection.
b – Assume 5% of each component of ZMSW is out of specifications or not suitable for conversion as determined during manual sorting.
c, d – Based on mixed-waste material recovery facilities (MRF) separation efficiencies reported by Pressley et al. (2015) in Appendix A. Supplementary data. Assume equal distribution of
ferrous and non-ferrous metals in the ‘metals’ category.
e – Assume no loss of material in this unit operation.
f – Based on the report by Savage and Diaz (1986). ZMSW is separated into a light fraction (“lights”) that contains an abundance of combustible material, ZOFMSW, and a heavy fraction
(“heavies”) that is mostly inorganic in nature.

shutdowns, meaning 3552 Mg of coal per MW are required per year.
Using this figure, coal consumption estimates under installed and de-
pendable capacity can be ascertained.
The emissions estimates are in three parts; coal mining, transporta-
tion and combustion. Emissions from coal transport were based on
emission factors, average fuel economy of diesel trains shown in
Table 8 and haulage distances. Coal is conveyed to Hwange station by
an overland conveyor belt which is 6 km long from Hwange Colliery
hence it was not regarded in this estimate. Coal is hauled by rail from
Hwange Colliery to Munyati power station, approximately 618 km
away. Harare Power station and the colliery are a distance of 812 km
apart. Bulawayo power station is roughly 314 km away. Emissions re-
lated to the mining of coal were based on IPCC guidelines. Estimates of
emissions from coal combustion (CO2, SO2 and NOX) were based on
emission factors and Zimbabwe coal chemistry. From the elemental
analysis of the Zimbabwe steam coal, the percentage of carbon and sul-
phur is known. From stoichiometry, 12 kg of carbon requires 32 kg of
oxygen to form 44 kg of carbon dioxide or 1 kg of carbon produces
Table 6
Model and assumptions for predicting the properties of ZMSWH.
3.67 kg of carbon dioxide when it burns. 32 kg of sulphur requires
32 kg of oxygen to form 64 kg of sulphur dioxide or 1 kg of sulphur pro-
duces 2 kg of SO2 when burnt. Therefore
CO2 emitted per year ¼ Total carbon in coal  3:67
¼ coal consumed per year  ð%CÞ  3:67 ð8Þ
SO2 emitted per year ¼ Total sulphur in coal  2
¼ coal consumed per year  ð%SÞ  2 ð9Þ
Turchi et al. (2011) report a coal NOx emission factor of
3.4 kg NO2/ton. Coal nitrogen content is generally not a good predictor
of NOx emissions hence the choice of a NOx emission factor. A simple
correlation between coal nitrogen content and NOx emissions has not
been found despite the extensive research in this area (Hachenberg,
2014). Methane emissions and Global Warming Potential (GWP) were
based on IPCC guidelines (IPCC, 2006; Solomon et al., 2007).


Emissions ðGg CH4 Þ ¼ Emissions Factor m3 CH4 =t
 Ton of Coal Produced
 
 Conversion Factor Gg=106 m3 ð10Þ

Component Basis for prediction
C Regression tree model associated with
hydrochar carbon content (%, dry wt.)
H 23% reduction in H/C atomic ratio from
MSW to hydrochar at 210 °C and
60-minute residence time
O 45% reduction in O/C atomic ratio from
MSW to hydrochar at 210 °C and
60-minute residence time
N Assume 50% reduction from solids
S Assume 37% reduction from solids
Cl 80% of initial Cl can be removed to
produce a solid fuel with low chlorine
content
Ash By difference: Ash% = 100% − C% − H%
− N% − S% − O%
Calorific 1. Regression tree model associated with
value hydrochar energy content (kJ/g dry
solids)
2. Dulong Formula
Hydrochar 1. Regression tree model associated with
yield solid yield (%, dry wt.)
2. Li et al. equation (see Eq. (7))
3. Lin et al. equation (see Eq. (6))
Average value of the results from these 3
equations was used as the yield in our
model
Reference
(Li et al., 2015)
(Lin et al., 2016)
(Lin et al., 2016)
(Lin et al., 2016)
(Lin et al., 2016)
(Prawisudha et al., 2012)
(Li et al., 2015)
(Li et al., 2015)
(Li et al., 2015)
(Lin et al., 2016)
where the conversion factor is given as 0.67 Gg/106 m3.
The emission factor for fugitive methane from underground coal
mining at Hwange in Zimbabwe was reported to be
4.3–10 m3 CH4/Mg of coal (Lloyd and Cook, 2005; Zhou et al., 2009). A
midrange value of 7.3 m3 CH4/Mg was assumed in this study. The Global
Warming Potential (GWP) of greenhouse gases was based on the find-
ings of the IPCC's Fourth Assessment report wherein the recommended
GWP values of CO2 and CH4 are 1 and 25 (Solomon et al., 2007) respec-
tively. Our model assumes the milling of ZMSWH pellets and the firing
of the mill product through the existing pipework and burners at
Zimbabwe's coal-fired power stations as shown in Fig. 1. The coal used
is transported by conveyor belts to boiler bunkers for storage. From
the bunkers, the coal is fed into volumetric feeders, which supply the
coal mills (pulverizing mill). A more detailed description of the process
is given by Ogbu et al. (1995).
3.7. Emissions related to ZMSW, ZMSWH and coal transportation
Emissions factors were used to estimate the impact of transporting
ZMSW, ZMSWH and coal for power generation. Nearly 29.7% of the
total urban population resides in Harare (World Bank). This city with
1,501,363 dwellers is served by 47 refuse compactors. However, about
60 refuse compactors would be ideal (Makura, 2014). Based on the
750 A.B. Maqhuzu et al. / Science of the Total Environment 668 (2019) 743–759

About 1.7 % of the

urban population is
located near Hwange
power station. Roughly MUNYATI POWER STATION
13.6 Gg/yr of ZMSW can 120 MW (INSTALLED)
be collected 30 MW (DEPENDABLE)
HARARE POWER STATION
About 13.8 % of the urban 120 MW (INSTALLED)
population is located near 30 MW (DEPENDABLE)
Munyati power station.
Roughly 108.1 Gg/yr of
ZMSW can be collected

About 65.6% of the

urban population is
HWANGE POWER STATION located near Harare
920 MW (INSTALLED) power station. Roughly
780 MW (DEPENDABLE) 514.5 Gg/yr of ZMSW
can be collected

BULAWAYO POWER STATION

120 MW (INSTALLED)
25 MW (DEPENDABLE)

About 18.9 % of the urban

population is located near
Bulawayo power station.
Roughly 147.9 Gg/yr of
ZMSW can be collected

Fig. 2. Installed and dependable capacities of Zimbabwe's thermal power stations and relative distribution of the urban population.
Source: Power plant capacities taken from Murehwa et al. (2012) and ZMSW estimation based on our calculations.

Harare assessment, the rest of the urban population of 3,550,184 people 4. Results and discussion
would ideally require about 142 refuse compactors to give a grand total
of about 202 compactors for the whole system as it exists now. The de- 4.1. Quantity and characteristics of ZMSW
sired number of refuse compactors for 16 urban councils (including
Harare and Bulawayo) was found to be about 104 (Muchandiona, From Fig. 3, MSW generation rate for low income nations might be
2013). An estimate of 202 trucks is reasonable for 32 locations. In predicted from the equation y = 0.0004(GDP / capita) + 0.32. Our esti-
order to estimate the emissions from the collection of waste, data on mate for ZMSW generation rate is 0.64 kg/capita/day. As data from de-
Table 8 was used. In our model, ZMSWH would be transported by rail. veloping countries is generally unreliable, data would not be expected
Emission rates for carbon dioxide and methane were not locomotive- to fit well into this model hence a low R2 value. Our estimate was aver-
specific, but were based on fuel properties. They are largely indepen- aged together with reported values (Table 1) to allow for a better esti-
dent of engine parameters and are primarily dependent on fuel proper- mation of the generation rate. The average value of ZMSW generation
ties hence the choice of using emission for diesel fuel. rate is 0.54 ± 0.14 kg/capita/day. For the current urban population
this equates to 1051.7 ± 270.7 Gg/year from which 784.2 ± 201.9 Gg/-
year is collected. According to one investigation, 860.0 Gg/year of ZMSW
Table 7
Zimbabwe steam coal properties (Murehwa et al., 2012; Zhou et al., 2009).
are produced and 705.2 Gg/year collected (Shonhiwa, 2013). Another
investigation by Scarlat et al. (2015) reveals that 989.0 Gg/year are gen-
Property Reported value Values used in our model erated and according to a 2014 study by Makura (2014), Harare City
Total moisture % 3.6–6.4 5.00 council estimates that around 920.0 Gg/year of household and industrial
Ash % 1.5–12.4 waste is produced in the capital. Mubaiwa (2006) asserts that about
Volatile matter % 19.3–42
2500.0 Gg/year of solid waste (industrial and household) are generated.
Fixed carbon % 23.8–66.7
Gross calorific valve (MJ/kg) 13.87–32.08, 24.2–29.9 25.45 There is considerable variation in the waste generation estimates largely
Net calorific value (MJ/kg) 22.9–28.4 due to the fact that specific data on ZMSW characteristics is not avail-
Carbon % 68.3–76.5 73.77 able from local authorities. Table 9 is a statistical summary of ZMSW
Nitrogen % 1.72–1.88 1.77 composition. Hoornweg and Bhada-Tata (2012) compares waste com-
Sulphur % 0.15–4.67 2.41
position by national income level. Despite being classified as a low-
 A.B. Maqhuzu et al. / Science of the Total Environment 668 (2019) 743–759 751

Table 8 Modified Dulong formula predicts a higher heating value of

Summary of MSW transportation and coal haulage parameters. 19.9 MJ/kg. The two estimations from separate methods are within a
Parameter Average Reference percentage difference of 27% between each other implying that the es-
value timation here is reliable. From Table 11 we can see that our predictions
Average annual distance 40,234 km (Alternative Fuels Data Center, n. measure up well against real data. From the characterization investiga-
travelled by refuse truck d.-a, n.d.-b) tion, ZMSW could potentially provide about 8.52 million GJ of energy
Fuel economy of refuse truck 0.92 km/L (Agar et al., 2007; Alternative equivalent to the energy from about 320.0 Gg of Zimbabwe steam
Fuels Data Center, n.d.-a, n.d.-b;
coal. Our results concur with the estimates given by Scarlat et al.
Chandler et al., 2001; Nguyen and
Wilson, 2010) (2015) of 8.9 million GJ being potentially recoverable from ZMSW.
Fuel economy of diesel train 99 (Association of American Shonhiwa (2013) contrasts our findings by suggesting that only
(1980–2015) ton-km/L Railroads) 2.86 million GJ is recoverable. Plastics are expected to be the largest con-
Lifetime mileage-weighted 7.09 g/km (Cai et al., 2013) tributor in terms of energy at 3.8 million GJ due to their high calorific
average air pollutant emission for NOX
factor for diesel refuse trucks 0.09 g/km
value. Paper and cardboard have the potential to contribute about
for SOX 2.3 million GJ. The methane potential (MP) of ZMSW using the IPCC de-
0.02 g/km fault methodology was calculated to be 62.2 Gg/year or
for CH4 92.7 million Nm3 per year. This is a specific methane generation rate
Emission factors for diesel heavy 0.87 kg/L (Greenhouse Gas Protocol)
of about 59.1 kg CH4/Mg of MSW. Scarlat et al. (2015) reports a value
truck for CO2
Emission factors for diesel 0.04437 (National Pollutant Inventory) of 88.0 million Nm3 per year.
locomotives (uncontrolled) kg/L for
NOX
0.0000167 4.2. Quantity and characteristics of ZOFMSW and ZMSWH
kg/L for
SOx
Carbon dioxide emission factors 2.68 kg/L (U.S. Energy Information We estimate that 386.9 Mg/year of ZOFMSW can be recovered
for diesel fuels Administration (EIA)) (nearly 70% of ZMSW) on a dry basis, and is available for conversion
Methane emission factors for 0.18 g/L into ZMSWH. Table 11 compares the calorific value of ZOFMSW with re-
non-highway mobile
ported measurements. Generally, our results are in good agreement
combustion
with real world values. The predicted elemental composition of
ZOFMSW is shown in Fig. 4. We predict an ash content of 6.6%, volatile
income nation, Zimbabwe's MSW composition fits the profile of an matter content of 81.1% and fixed carbon content of 12.3%. From the re-
upper middle-income nation. The quantity of paper (18.8 ± 3.2%) and gression tree models and Eqs. (6) and (7), the average value for the ex-
plastics (16.1 ± 3.5%) found in the waste stream is unusually high for pected yield of ZMSWH from ZOFMSW is 75%. On a complete dry basis, a
a low-income nation. The relatively low organic fraction (32.9 ± 6.6%) yield of about 289.3 Gg/year of ZMSWH can be expected. From ZMSW to
is not atypical of a low-income nation. The high standard deviation in ZMSWH the overall yield is about 37%. From the regression tree model
each waste category mirrors the large amount of variation in waste associated with hydrochar energy content, a caloric value of
composition within the different urban areas of Zimbabwe. On a dry 22.47 MJ/kg is expected for ZMSWH. From the Dulong Formula, an en-
weight basis, 556.4 Gg/year of ZMSW can be collected annually. In ergy content of 25.34 MJ/kg is expected. Oxygen and hydrogen content
order to validate our model, we compared our simulated proximate reduction from ZOFMSW to ZMSWH are expected as dehydration and
and ultimate composition results with measured data as shown in decarboxylation occur during HTC with the removal of hydroxyl, car-
Table 10. With a few notable exceptions there is reasonable agreement boxyl and carbonyl groups. The feedstock (ZOFMSW) presented very
between our simulation and real-life data. Using typical heating values high H/C and O/C atomic ratios, which were 1.59 and 0.39. The H/C
of MSW components, the calorific value of ZMSW is 15.1 MJ/kg (dry and O/C atomic ratios were expected to decrease to 1.22 and 0.21
basis). However, from the elemental content of ZMSW (Table 10), the after HTC treatment.

Fig. 3. MSW generation and GDP (both per capita) of 38 low income countries. MSW generation rates reported by Hoornweg and Bhada-Tata (2012) and GDP data from World Bank. Red
dot represents Zimbabwe (generation rate in 2012). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
752 A.B. Maqhuzu et al. / Science of the Total Environment 668 (2019) 743–759

Table 9
Statistical analysis of ZMSW composition.

Statistical analysis Paper & Glass & Metals Plastics Leather & Textiles Yard waste & Wood Putrescibles Miscellaneous
cardboard ceramics rubber sweepings

Mean 18.8 4.0 6.5 16.1 1.9 3.8 4.4 4.1 32.9 7.5
Standard error 1.6 0.5 1.4 1.7 0.5 0.7 1.2 0.8 3.3 1.4
Median 17.1 4.0 4.0 12.1 0.4 2.6 1.5 0.7 28.0 7.0
Mode 30.0 5.0 3.0 10.0 0.0 0.0 0.0 0.0 15.0 5.0
Standard deviation 9.5 2.8 8.2 10.4 2.7 3.6 7.1 5.0 19.6 8.4
Sample variance 89.5 7.9 67.4 107.3 7.1 12.7 50.4 24.7 385.2 70.1
Kurtosis (1.2) 2.8 8.5 0.2 1.4 0.1 6.9 (0.5) (0.5) 0.9
Skewness 0.2 1.2 2.8 0.9 1.5 1.0 2.4 0.8 0.6 1.0
Range 31.7 13.3 39.6 39.0 9.3 12.0 33.3 17.0 70.9 35.0
Minimum 3.3 0.0 0.0 3.0 0.0 0.0 0.0 0.0 5.0 0.0
Maximum 35.0 13.3 39.6 42.0 9.3 12.0 33.3 17.0 75.9 35.0
Confidence level (95.0%) 3.2 1.0 2.8 3.5 1.1 1.4 2.4 1.7 6.6 2.8

4.3. Environmental impact of co-burning MSW derived hydrochar for
power generation
4.3.1. Environmental impact of coal-fired power generation
Zimbabwe's thermal power generation is fickle hence a distinction
between installed (assumes more efficient operations) and dependable
capacity (existing situation) showed a wide variation.
Hypothetically, if the plants had less breakdowns, the total coal con-
sumption for thermal power generation would be about 4700 Gg/year.
Hwange power station as the biggest station is expected to consume
greater quantities of coal at 2900–3400 Gg/year depending on the oper-
ating regime. Our own estimations reveal a total of 3190 Gg/year is con-
sumed as the system exists now. There is reasonable agreement with
reported value of 3290 Gg/year. Harare and Bulawayo stations require
700 Mg a day, Munyati demands 624 Mg a day while Hwange consumes
8000 Mg a day or about 180,000 Mg per month and requires a stockpile
of 30,000 Mg according to newspaper reports (Chronicle, 2013; Herald,
2015). With an average calorific value of 25.5 MJ/kg, this represents
about 120.1 million GJ and 81.5 million GJ input for installed and de-
pendable capacity respectively.
The environmental impact of coal transportation is expected to be
significantly lower relative to mining and combustion of coal. Signifi-
cant CH4 emissions stem mainly from coal mining operations. About
15.6–23.1 Gg/year of CH4 can be emitted from this source. CO2 and
NOX emissions from coal transportation stand at about 0.1–0.4 Gg/year
of NOX and about 7.0–23.0 Gg/year of CO2. Total annual CO2 emissions
and combined GWP (from mining, transport and combustion) stand at
approximately 8700 Gg/year and 9060 Gg/year CO2e respectively for
dependable capacity. Under installed capacity, total annual CO2 emis-
sions and combined GWP are 12,800 Gg/year and 13,400 Gg/year CO2e
respectively. This is depicted in Fig. 5. SO2 emissions stand at about
154–228 Gg/year with a significant proportion attributed to coal
combustion.
4.3.2. Environmental impact estimate of HTC
The estimated fuel used to collect waste is about 8.8 million L/year.
The most significant amount of emissions related to refuse collection is
CO2 at 7.3 Gg/year. Fig. 4 is a prediction of the elemental composition of
hydrochar. A reduction in chlorine and sulphur content is expected due
to dechlorination and desulphurization properties of HTC. Chlorine
mainly from PVC (organic chlorine) and salt (inorganic chlorine) is con-
verted into HCl and discharged with steam during hydrothermal carbon-
ization resulting in a product with reduced Cl content compared to the
original MSW. Sulphur and nitrogen are expected to follow a similar
trait whereby S and N partly remains in the ZMSWH as it is released
into the gas and liquid phase during hydrothermal treatment. Total SO2,
NOX, CH4, CO2 emissions and GWP are shown in Fig. 6. NOx emissions
are expected to originate mainly from combustion. ZMSWH contains
about 2.7 Gg of nitrogen based on the elemental composition. The con-
tributors of nitrogen are proteins in food wastes and textiles. These
sources of nitrogen comprise of chemically-bound nitrogen in organic
form e.g. amines and inorganic nitrogen compounds and upon combus-
tion, N2 or NO is formed. Therefore, the amount of fuel-NO emitted de-
pends on the ZMSWH quantity combusted. CH4 quantities are expected
to be very low as the source will mainly be from vehicles involved in
ZMSW collection and transportation. Projected total CO2 emissions are
the region of 431.8 Gg/year. In terms of sources of CO2 emissions, our
model predicts that the proportion of fossil carbon (e.g. carbon from plas-
tics) is about 37% and the proportion of biogenic carbon is 63% in
ZOFMSW. A maximum of 231.9 Gg/year CO2 is expected from the direct
combustion of the hydrochar. Estimates of the net emissions from the
co-burning of ZMSWH and coal will be given under different scenarios
in the following section.
4.4. Environmental impact of co-combustion under different scenarios
Due to the uneven distribution of the urban population (MSW source)
relative to the size and location of coal-fired stations as shown in Fig. 2,

Table 10 Table 11
Proximate and ultimate analysis of ZMSW as compared to measured values from different Energy content analysis of ZMSW and ZOFMSW compared with field results.
cities in Zimbabwe.
Combustible Full MSW
City/town Proximate Ultimate fraction stream

Ash VM FC MC C H N Cl S O Dry Wet Dry Wet

basis basis basis basis
Hararea 12.5 79.2 8.4 30.9 47.1 5.1 1.6 0.7 0.8 30.9
Chitungwizaa 8.3 82.9 8.8 36.0 44.6 4.8 2.5 0.6 0.8 37.3 Field studya Harare 22.26 14.62 12.87
Epwortha 8.6 84.0 7.4 34.4 47.9 4.9 2.5 0.4 0.8 30.8 Chitungwiza 18.09 10.70 8.34
Average (field 9.8 82.0 8.2 33.8 46.5 4.9 2.2 0.6 0.8 33.0 Epworth 21.03 12.96 9.07
study) Average 20.46 12.76 10.09
ZMSW (our 20.4 68.2 9.9 29.0 45.9 6.1 1.6 1.6 0.2 24.2 Our study ZMSW 20.03 13.10 15.15 10.75
study) ZMSW (using Dulong Formula) 19.92
Percentage 70.2 18.5 19.8 16.7 1.5 21.8 31.6 85.7 120.0 30.4 ZOFMSW 18.39 12.22
difference ZOFMSW (using Dulong Formula) 23.69
a a
Field study conducted by Makarichi et al. (2018a). Makarichi et al. (2018a).
 A.B. Maqhuzu et al. / Science of the Total Environment 668 (2019) 743–759 753

70.00

58.60
60.00
54.26

50.00

40.00
ZOFMSW
%
30.00 27.91
ZMSWH

20.00 17.67
16.41

10.00 7.26
5.96 6.58
1.87 1.89
0.93 0.24 0.22 0.18
0.00
C H N Cl S O Ash

Fig. 4. Percent elemental composition of ZOFMSW and ZMSWH.

MSW transportation and logistics challenges may adversely affect the
viability of the proposed technology. For example, roughly 66% of
Zimbabwe's urban population is near Harare power station with a name-
plate capacity of 120 MW and b2% of the urban population is near
Hwange power station with an installed capacity of 980 MW. Hence it
would be necessary to transfer excess alternate fuel from Harare to
Hwange. Table 12 gives an estimate of the quantities of MSW generated
in each urban city based on national generation rate and relative urban
population size. This formed the basis of estimation of emissions under
different scenarios. The aims of the scenario analysis are:
• identify possible pathways to attain the goal of co-firing of ZMSWH in
Zimbabwe's power stations
• to identify the maximum possible blend ratio for each route
• evaluate the environmental impact of each route relative to current
emissions (baseline).
Under all scenarios, MSW would still be collected, however conver-
sion to hydrochar and subsequent co-burning in power stations would
depend on the scenario. The scenarios that shall be considered in this
section are:
Scenario 1: Collection of ZMSW from all urban areas and entire con-
version to ZMSWH
Scenario 2: Collection of ZMSW from all urban areas but conversion
to ZMSWH limited only to waste collected from major cities (re-
maining ZMSW is landfilled)

12,823 13,402
8,657 9,059
10,000.00

1,000.00

100.00
23
Emissions (Gg/yr)

16
10.00

1.00
154 228

0.10 11 16

0.01

0.00
Dependable Capacity Installed Capacity

Fig. 5. Total emissions related to coal mining, transport and combustion for power generation.
754 A.B. Maqhuzu et al. / Science of the Total Environment 668 (2019) 743–759

Fig. 6. Emissions related to ZMSWH formation, transport and combustion.

Scenario 3: Collection of ZMSW from all urban areas and conversion

to ZMSWH limited only to waste collected from cities in relatively
close proximity to coal-fired power plants (the rest is landfilled)
Table 12
Scenario 4: Collection of ZMSW from all urban areas and conversion
Estimated ZMSW generation by city/town. to ZMSWH limited to waste from close cities. ZMSWH requirement
Urban area ZMSW generation (Gg/year)
is determined by power plant capacity
Scenario 5: Collection of ZMSW from all urban areas and conversion
Harare 381.46 ± 52.33
Bulawayo 167.80 ± 23.02 to ZMSWH limited to waste from Harare.
Chitungwiza 91.65 ± 12.57
Mutare 48.19 ± 6.61 Fig. 7a–e, shows projections for all scenarios. In scenario 1, ZMSW is
Epworth 43.01 ± 5.90
collected from all urban centres, converted to hydrochar and then dis-
Gweru 40.55 ± 5.56
Kwekwe 25.92 ± 3.55
tributed to each coal-fired plant according to its capacity.
Kadoma 23.75 ± 3.26 289.3 Gg/year of ZMSWH is available for distribution to all thermal
Masvingo 22.57 ± 3.10 power plants. The maximum blend ratios are 9.0% and 6.1% for depend-
Chinhoyi 20.02 ± 2.75 able capacity (D.C.) and installed capacity (I.C.) operations. Compared to
Norton 17.36 ± 2.38
the emissions from the existing situation (baseline), Fig. 7a, under this
Marondera 15.92 ± 2.18
Ruwa 14.56 ± 2.00 scenario, GWP drops from 13.4 to 13.1 million for I.C. and 9.1 to 8.9 mil-
Chegutu 12.99 ± 1.78 lion CO2e for D.C. This is only a slight reduction of 2.2% for both I.C. and
Zvishavane 11.62 ± 1.59 D.C. respectively. CO2 emissions would decrease slightly by 3.1% and
Bindura 11.22 ± 1.54
3.7% for I.C. and D.C. respectively. NOx emissions are expected to rise
Beitbridge 10.82 ± 1.48
Redcliff 9.23 ± 1.27
under this scenario (18.8% for I.C. and 45.5% for D.C.). To counteract
Victoria Falls 8.67 ± 1.19 the rise in NOx emissions we recommend the installation of air pollu-
Rusape 7.79 ± 1.07 tion control devices (APCDs), in particular, selective catalytic reduction
Chiredzi 7.82 ± 1.07 (SCR) units, which can reportedly decrease NOx emissions (Ma et al.,
Kariba 6.79 ± 0.93
2016). Overall SOx emissions are expected to decline from
Karoi 7.35 ± 1.01
Chipinge 6.50 ± 0.89 228.1 Gg/year to 218.6 Gg/year (a 4.2% decline) for I.C. and from 154.1
Gokwe 6.18 ± 0.85 to 145.0 Gg/year (5.9% reduction) for D.C. per year. Methane will de-
Shurugwi 5.63 ± 0.77 crease by 4.5% and 6.6% for I.C. and D.C. respectively. Fig. 7e shows the
Gwanda 5.20 ± 0.71 methane potential of the remaining uncollected waste. The conversion
Hwange 9.64 ± 1.32
Chivhu 3.33 ± 0.46
of all collected ZMSW will reduce the methane potential of household
Plumtree 2.99 ± 0.41 waste from 62.2 to 15.8 Gg CH4/year. Under scenario 2, Fig. 7b, ZMSW
Mvurwi 2.71 ± 0.37 is collected from 10 major cities and a yield of 237.9 Gg/year of
Banket 2.53 ± 0.35 ZMSWH is expected. The alternate fuel is distributed to each station
 A.B. Maqhuzu et al. / Science of the Total Environment 668 (2019) 743–759 755

a
100000
12,823 12,423
8,333 8,657
10000
13,402 13,129
8,931 9,059

1000

Emissions Gg/yr
100

19
10 16 16
11

1
228 219 145 154

0.1

0.01 22 15 16
23

0.001
Baseline (I.C) 6.1% maximum 9.0% maximum Baseline (D.C)
blend (I.C) blend (D.C)
CO SO NO CH GWP

b 100000

12,823 12,559
8,393 8,657
10000
13,301
13,402
8,958 9,059

1000
Emissions (Gg/yr)

100

20 17
10 16
11

1
228 221 147 154

0.1

0.01 22 15 16
23

0.001
Baseline (I.C) 5.0% maximum blend 7.4% maximum blend Baseline (D.C)
(I.C) (D.C)
CO SO NO CH GWP

c 100000

12,823 12,809
8,643 8,657
10000
13,402 13,513
9,170 9,059

1000
Emissions (Gg/yr)

100

19
10 16 14
11

1
228 226 152 154

0.1

0.01 23 15 16
23

0.001
Baseline (I.C) 3.6% maximum 5.3% maximum Baseline (D.C)
blend (I.C) blend (D.C)
756 A.B. Maqhuzu et al. / Science of the Total Environment 668 (2019) 743–759

d 100000

12,823 12,659
8,603 8,657
10000
13,345
13,402
9,050 9,059

1000

Emissions (Gg/yr) 100

10 16 13 12
11

1
228 223 154

0.1

0.01
23 15 16
23

0.001
Baseline (I.C) Maximum blends (I.C) Maximum blends (D.C) Baseline (D.C)
Harare 20.0% Harare 20.0%
Bulawayo 10.4% Bulawayo 20.0%
Munyati 3.1% Munyati 12.3%
Hwange 0.11% Hwange 0.18%

GWP
e 100000

12,823 12,714
8,549 8,657
10000
13,402 13,369
9,026 9,059

1000
Emissions (Gg/yr)

100

10 18
16 14
11

1
228 225 151 154

0.1

0.01 23 15 16
23

0.001
Baseline (I.C) 2.2% maximum blend 3.3% maximum blend Baseline (D.C)
(I.C) (D.C)

CO SO CH GWP
f 70.00

62.18
60.00

51.42
50.00
Emissions (Gg CH₄ /yr)

45.37

39.46
40.00
35.18

30.00
24.05

20.00
15.82

10.00

0.00
Scenario 1 Scenario 2 Scenario 3 Scenario 4 Scenario 4 Scenario 5 Baseline
(I.C) (D.C)
 A.B. Maqhuzu et al. / Science of the Total Environment 668 (2019) 743–759
according to capacity. A maximum blend of 5.0% and 7.4% for I.C. and
D.C. is possible in this scenario. Under scenario 3, ZMSW from 6 urban
areas (Harare, Bulawayo, Kwekwe, Kadoma, Victoria Falls and Hwange)
is converted to yield about 168.4 Gg/year of ZMSWH. This fuel is hauled
to each station according to the capacity of each respective plant. To use
up all the fuel, a maximum blend of 3.6% and 5.3% is possible for I.C. and
D.C. respectively as shown in Fig. 7c. Under scenario 4, Fig. 7d, waste is
collected from 6 cities near power stations but in this case (like Scenario
3), however, the quantity of ZMSW converted depends on the quanti-
ties the city/town can provide. Harare can potentially provide about
104.9 Gg/year of ZMSWH meaning Harare power station can theoreti-
cally accept a blend ratio beyond 20% for either I.C. or D.C. Hwange
power station, however, is situated next to towns that can provide a
maximum of 3.7 Gg/year of ZMSWH. Under scenario 5, ZMSW from
Harare is converted to yield about 104.9 Gg/year of ZMSWH. This fuel
is hauled to each station according to the capacity of each respective
plant. To use up all the fuel, a maximum blend of 2.2% and 3.3% is possi-
ble for I.C. and D.C. respectively as shown in Fig. 7e. In all scenarios, SOX
emissions are expected to be reduced. Greater NOX emissions have to be
expected when burning ZMSWH. The partial substitution of coal using
ZMSWH (6.1 to 9.0% of total coal consumption) is not enough to offset
the NOX emissions emanating from ZMSWH combustion. Although
our model predicts increases in NOX emissions it must be understood
that this may not be so in practice. Despite the expected decrease of
NOx with increase in biomass proportion, pilot- and full-scale testing
showed mixed results according to Niksa et al. (2003). Overall, in either
one of these scenarios, the greenhouse emissions rates relative to the
baseline are expected to be more or less the same. The most suitable
route for Zimbabwe from an environmental perspective would be the
one with relatively significant reductions in methane potential. Using
this reasoning, Scenario 1 is the likely candidate, as it can potentially re-
sult in less impact on the environment. This route would likely be the
most expensive and not economically viable. However, this is conjec-
ture at this point, without an economic feasibility study we cannot
draw conclusions on the actual economic viability. The financial viabil-
ity of each route would have to be carefully examined in order to
draw up conclusions on the optimally balanced scenario (environmen-
tally friendly and economically efficient).
5. Conclusion
Environmental issues raise questions about the application of HTC in
developing countries. This study set to determine life cycle emissions
from MSW-derived hydrochar co-combustion in Zimbabwe's coal
power stations. Five scenarios differing in the amount of MSW con-
verted based on the spatial distribution of MSW and stations were de-
veloped to assess the environmental impact of the technology. Results
confirm that greenhouse gas emissions from unsanitary MSW landfills
and coal-fired power plants can be abated by HTC. Roughly 1051.7 ±
270.7 Gg/year of MSW are generated and a yield of 289.3 Gg/year of
hydrochar is possible. This can substitute between 6.1 and 9.0% of coal
used in thermal power plants with coal consumptions rates of 3200–-
4700 Gg/year. Under the best scenario, slight reductions in overall direct
emissions from the co-burning route; 4.2% for SOX, 4.5% for CH4, 3.1% for
CO2, 2.2% in Global Warming Potential (GWP) and a 18.0% rise for NOX
can be expected. The diversion of MSW from landfills can significantly
reduce the methane potential of household waste from 62.2 to
15.8 Gg CH4/year under this scenario. Financial and technical resources
for developing hydrothermal carbonization technology are essential for
Zimbabwe's future. It is recommended that future research efforts focus
on the economic evaluation of the process in developing nations. From
the foregoing study, it would be necessary for relevant authorities to
take immediate action by adopting the technology to prevent additional
Fig. 7. a. Projections for total direct emissions under scenario 1. b. Projections for total direct emissions under scenario 2. c Projections for total direct emissions under scenario 3. d.
Projected total emissions under scenario 4. e. Projected total emissions under scenario 5. f Methane potential of unconverted ZMSW under different scenarios.
757
irrecoverable damage to the environment. The financial viability of each
route would have to be carefully examined in order to draw up conclu-
sions on the optimally balanced scenario (environmentally friendly and
economically efficient).
Acknowledgements
This study was financially supported by JSPS KAKENHI Grant Num-
ber 26550059 and 15H04067. The authors appreciate it greatly.
References
Achankeng E. Globalization, urbanization and municipal solid waste management in
Africa. African Studies Association of Australasia and the Pacific, 2003 Conference
Proceedings - African on a Global Stage 2003: 1–22.
African Development Bank Group (AFDB). AFDB Socio Economic Database. [Online] Avail-
able from: http://dataportal.afdb.org/ [Accessed 31 January 2019]. 2019.
Agar, B.J., Baetz, B.W., Wilson, B.G., 2007. Fuel consumption, emissions estimation, and
emissions cost estimates using global positioning data. J. Air Waste Manage. Assoc.
57, 348–354.
Alternative Fuels Data Center, d. Average annual vehicle miles traveled of major vehicle
categories.Available from:. http://www.afdc.energy.gov/data/10309, Accessed date:
5 August 2017 (Online).
Alternative Fuels Data Center, d. Average fuel economy of major vehicle categories.Avail-
able from:. http://www.afdc.energy.gov/data/10310, Accessed date: 5 August 2017
(Online).
Association of American Railroads, d. The environmental benefits of moving freight by
rail.Available from:. https://www.aar.org/wp-content/uploads/2018/07/AAR-Envi-
ronmental-Benefits-Movig-Freight-by-Rail.pdf, Accessed date: 5 August 2017
(Online).
Bach, Q.-V., Tran, K.-Q., Khalil, R.A., Skreiberg, Ø., Seisenbaeva, G., 2013. Comparative as-
sessment of wet torrefaction. Energy Fuel 27, 6743–6753.
Berge ND, Flora JR. Energy Source Creation From Diverted Food Wastes via Hydrothermal
Carbonization. Environmental Research and Education Foundation, 3301 Benson
Drive, Suite 301 Raleigh, North Carolina 27609, 2015.
Cai, H., Burnham, A., Wang, M., 2013. Updated Emission Factors of Air Pollutants From Ve-
hicle Operations in GREETTM Using MOVES. Systems Assessment Section, Energy
Systems Division, Argonne National Laboratory.
Chandler, K., Norton, B.P., Clark, N., 2001. Waste management's LNG truck fleet: final re-
sults. DOE Report, January.
Chandrappa, R., Brown, J., 2012. Solid Waste Management: Principles and Practice:
Springer Science & Business Media.
Chifamba, P.C., 2007. Trace metal contamination of water at a solid waste disposal site at
Kariba, Zimbabwe. Afr. J. Aquat. Sci. 32, 71–78.
Chronicle. ZPC to upgrade thermal power stations. The Chronicle http://www.chronicle.
co.zw/zpc-to-upgrade-thermal-power-stations/, Bulawayo, Zimbabwe, 2013.
EASME, 2016. European Commission Executive Agency for Small and Medium-sized
Enterprises, Up-scaling, demonstration and first market application of Loritus'
patented hydrothermal carbonisation as an eco-efficient and cost-effective organic
waste processing technology. European Commission, Community Research and
Development Information Service, Available from:. http://cordis.europa.eu/project/
rcn/198801_en.html, Accessed date: 5 October 2017 (Online).
EPA U. Emission factors for greenhouse gas inventories. Stationary Combustion Emission
Factors. US Environmental Protection Agency 2014.
Friedrich, E., Trois, C., 2011. Quantification of greenhouse gas emissions from waste man-
agement processes for municipalities – a comparative review focusing on Africa.
Waste Manag. 31, 1585–1596.
Funke, A., Ziegler, F., 2010. Hydrothermal carbonization of biomass: a summary and dis-
cussion of chemical mechanisms for process engineering. Biofuels Bioprod. Biorefin.
4, 160–177.
Gidarakos, E., Havas, G., Ntzamilis, P., 2006. Municipal solid waste composition determi-
nation supporting the integrated solid waste management system in the island of
Crete. Waste Manag. 26, 668–679.
Greenhouse Gas Protocol, d. Calculating CO2 emissions from mobile sources.Available
from:. http://www.ghgprotocol.org/calculation-tools, Accessed date: 5 August 2017
(Online).
Gwenzi, W., Chaukura, N., Mukome, F.N.D., Machado, S., Nyamasoka, B., 2015. Biochar
production and applications in sub-Saharan Africa: opportunities, constraints, risks
and uncertainties. J. Environ. Manag. 150, 250–261.
Hachenberg, N., 2014. Predictions of NOx Emissions in Pulverized Coal Combustion. Elec-
tronic Theses and Dissertations, Paper 557. The University of Louisville.
Herald. Coal mining concerns threaten electricity generation in Hwange. The Herald
http://www.herald.co.zw/coal-mining-concerns-threaten-electricity-generation-at-
hwange/, Harare, Zimbabwe, 2015.
Hoekman, S.K., Broch, A., Robbins, C., Purcell, R., Zielinska, B., Felix, L., et al., 2014. Process
development unit (PDU) for hydrothermal carbonization (HTC) of lignocellulosic bio-
mass. Waste and Biomass Valorization 5, 669–678.
Hoornweg D, Bhada-Tata P. What a Waste: A Global Review of Solid Waste Management.
2012: Annex J.
758 A.B. Maqhuzu et al. / Science of the Total Environment 668 (2019) 743–759
Hoornweg, D., Bhada-Tata, P., Kennedy, C., 2015. Peak waste: when is it likely to occur?
J. Ind. Ecol. 19, 117–128.
Hussain, F., Chaudhry, M.N., Batool, S.A., 2014. Assessment of key parameters in municipal
solid waste management: a prerequisite for sustainability. International Journal of
Sustainable Development & World Ecology 21, 519–525.
Huvengwa, I., 2012. Solid Waste Management: An Urban Challenge, a Case of Masvingo
Town. Bindura University of Science Education, MSc Thesis.
Hwang, I.-H., Aoyama, H., Matsuto, T., Nakagishi, T., Matsuo, T., 2012. Recovery of solid
fuel from municipal solid waste by hydrothermal treatment using subcritical water.
Waste Manag. 32, 410–416.
Intergovernmental Panel on Climate Change (IPCC). Report of the Twelfth Season of the
Intergovernmental Panel on Climate Change. Mexico City, 11–13 September n.
d.1996.
Intergovernmental Panel on Climate Change (IPCC). Guidelines for National Greenhouse
Gas Inventories. Eds; Simon Eggleston, Leandro Buendia, Kyoko Miwa, Todd Ngara,
Kiyoto Tanabe 2006; 5.
International Energy Agency, d. Annex I: municipal solid waste potential in cities.Avail-
able from:. https://www.iea.org/media/etp/etp2016/AnnexI_MSWpotential_web.
pdf, Accessed date: 6 August 2017 (Online).
International Labour Organization, d. Start your waste recycling business technical hand-
outs.Available from:. http://www.gdrc.info/docs/waste/008.pdf, Accessed date: 5 Au-
gust 2017 (Online).
Jerie, S., 2014. Analysis of enterprise profile and composition of solid waste generated in
the informal sector of Gweru. Zimbabwe. Journal of Waste Management 2014, 1–13.
Jin, Y., Lu, L., Ma, X., Liu, H., Chi, Y., Yoshikawa, K., 2013. Effects of blending hydrothermally
treated municipal solid waste with coal on co-combustion characteristics in a lab-
scale fluidized bed reactor. Appl. Energy 102, 563–570.
Jingura, R.M., Matengaifa, R., 2009. Optimization of biogas production by anaerobic diges-
tion for sustainable energy development in Zimbabwe. Renew. Sust. Energ. Rev. 13,
1116–1120.
Jingura, R.M., Musademba, D., Kamusoko, R., 2013. A review of the state of biomass energy
technologies in Zimbabwe. Renew. Sust. Energ. Rev. 26, 652–659.
Kaseke, N., 2014. A comparative cost assessment of electricity outages and generation ex-
pansion in Zimbabwe. International Journal of Advanced Research in Management
and Social Sciences 3, 20.
Kathirvale, S., Yunus, M.N.M., Sopian, K., Samsuddin, A.H., 2004. Energy potential from
municipal solid waste in Malaysia. Renew. Energy 29, 559–567.
Katiyar, R.B., Suresh, S., Sharma, A., 2013. Characterisation of municipal solid waste gener-
ated by the city of Bhopal. India. International Journal of Chem. Tech. Research 5 (2),
623–628.
Kharlamova, M., Mada, S.Y., Grachev, V., 2016. Landfills: problems, solutions and decision-
making of waste disposal in Harare (Zimbabwe). Biosci., Biotech. Res. Asia 13 (1),
307–318.
Khoshnevisan, B., Rajaeifar, M.A., Clark, S., Shamahirband, S., Anuar, N.B., Mohd Shuib, N.L.,
et al., 2014. Evaluation of traditional and consolidated rice farms in Guilan Province,
Iran, using life cycle assessment and fuzzy modeling. Sci. Total Environ. 481, 242–251.
Klanfar, M., Korman, T., Kujundžić, T., 2016. Fuel consumption and engine load factors of
equipment in quarrying of crushed stone. Tehnički vjesnik/Technical Gazette 23, 163.
Kühling J-G, Health E. Rapid Assessment: Healthcare Waste Component of Global Fund
HIV, TB and Malaria Projects in Zimbabwe. UNDP, 2014.
Li, L., Flora, J.R.V., Caicedo, J.M., Berge, N.D., 2015. Investigating the role of feedstock prop-
erties and process conditions on products formed during the hydrothermal carboni-
zation of organics using regression techniques. Bioresour. Technol. 187, 263–274.
Lin, Y., Ma, X., Peng, X., Yu, Z., 2016. Forecasting the byproducts generated by hydrother-
mal carbonisation of municipal solid wastes. Waste Manag. Res. 35, 92–100.
Liu DH, Liptak BG. Environmental Engineers' Handbook on CD-ROM: CRC Press, 1999.
Liu, Z., Quek, A., Kent Hoekman, S., Balasubramanian, R., 2013. Production of solid biochar
fuel from waste biomass by hydrothermal carbonization. Fuel 103, 943–949.
Lloyd, P., Cook, A., 2005. Methane release from South African coalmines. Journal of the
South African Institute of Mining and Metallurgy 105, 483–490.
Lopes, C., 1996. Balancing Rocks: Environment and Development in Zimbabwe: Nordic
Africa Institute.
Lu, L., Namioka, T., Yoshikawa, K., 2011. Effects of hydrothermal treatment on characteristics
and combustion behaviors of municipal solid wastes. Appl. Energy 88, 3659–3664.
Lu, X., Jordan, B., Berge, N.D., 2012. Thermal conversion of municipal solid waste via hy-
drothermal carbonization: comparison of carbonization products to products from
current waste management techniques. Waste Manag. 32, 1353–1365.
Ma, Z., Deng, J., Li, Z., Li, Q., Zhao, P., Wang, L., et al., 2016. Characteristics of NOx emission
from Chinese coal-fired power plants equipped with new technologies. Atmos. Envi-
ron. 131, 164–170.
Makarichi, L., Kan, R., Jutidamrongphan, W., Techato, K-a., 2018a. Suitability of municipal
solid waste in African cities for thermochemical waste-to-energy conversion: the
case of Harare Metropolitan City, Zimbabwe. Waste Manag. Res. 37, 83–94.
Makarichi L, Techato K-a, Jutidamrongphan W. Material flow analysis as a support tool for
multi-criteria analysis in solid waste management decision-making. Resources, Con-
servation and Recycling 2018b; 139: 351–365.
Makoni, F., Ndamba, J., Mbati, P., Manase, G., 2004. Impact of waste disposal on health of a
poor urban community in Zimbambwe. East Afr. Med. J. 81, 422–426.
Makura, R., 2014. Analysing the Challenges Besetting Solid Waste Management in
Zimbabwe's Urban Areas: A Case Study of Warren Park Suburb, Harare. University
of Zimbabwe, MSc Thesis.
Makwara, E.C., Magudu, S., 2013. Confronting the reckless gambling with people's health
and lives: urban solid waste management in Zimbabwe. European Journal of Sustain-
able Development 2, 67.
Mandevere, B., 2015. An Investigation Into Effectiveness of Household Solid Waste
Management Strategies in Harare, Zimbabwe. University of South Africa, MSc Thesis.
Mangizvo, R.V., 2007. Challenges of solid waste management in the Central Business Dis-
trict of the City of Gweru in Zimbabwe. Journal of Sustainable Development in Africa
9, 134–145.
Mangizvo, R.V., 2008. Management practices at the Mucheke municipal solid waste dis-
posal site in Masvingo City, in Zimbabwe. Journal of Sustainable Development in
Africa 10, 147–164.
Mangundu, A., Makura, E.S.M.S., Mangundu, M., Tapera, R., 2013. The importance of inte-
grated solid waste management in Independent Zimbabwe: the case of Glenview,
Area 8, Harare. Global Journal of Biology, Agriculture and Health Sciences 2, 85–92.
Manyanhaire, I.O., Sigauke, E., Munasirei, D., 2009. Analysis of domestic solid waste man-
agement system: a case of Sakubva high density suburb in the city of Mutare,
Manicaland Province, Zimbabwe. Journal of Sustainable Development in Africa 11,
127–140.
Masocha, M., 2004. Solid Waste Disposal in Victoria Falls Town: Spatial Dynamics, Envi-
ronmental Impacts Health Threats and Socio-economic Benefits. University of
Zimbabwe, M Phil Thesis.
Masocha, M., 2006. Informal waste harvesting in Victoria Falls town, Zimbabwe: socio-
economic benefits. Habitat International 30, 838–848.
Mbiba, B., 2014. Urban solid waste characteristics and household appetite for separation
at source in Eastern and Southern Africa. Habitat International 43, 152–162.
Mbohwa, C., Zvigumbu, B., 2007. Feasibility study for waste incinerator plant for the pro-
duction of electricity in Harare, Zimbabwe. Adv. Mater. Res. 18-19, 509–518.
Mohee, R., Simelane, T., 2015. Future directions of municipal solid waste management in
Africa. Oxford: African Books Collective. Project MUSE 1–250.
Mshandete, A.M., Parawira, W., 2009. Biogas technology research in selected sub-Saharan
African countries – a review. Afr. J. Biotechnol. 8, 116–125.
Mubaiwa A. Community based waste management in urban areas. Proceedings From the
2nd International Conference on Appropriate Technology, Bulawayo, Zimbabwe,
2006, pp. 99.
Muchandiona, A., 2013. Challenges and Opportunities in Solid Waste Management in
Zimbabwe's Urban Councils. University of Zimbabwe, MSc Thesis.
Mudzengerere, F.H., Chigwenya, A., 2012. Waste management in Bulawayo City Council in
Zimbabwe: in search of sustainable waste management in the city. Journal of Sustain-
able Development in Africa 14, 228–244.
Murehwa, G., Zimwara, D., Tumbudzuku, W., Mhlanga, S., 2012. Energy efficiency im-
provement in thermal power plants. International Journal of Innovative Technology
and Exploring Engineering (IJITEE) 2, 20–25.
Muthuraman, M., Namioka, T., Yoshikawa, K., 2010a. Characteristics of co-combustion
and kinetic study on hydrothermally treated municipal solid waste with different
rank coals: a thermogravimetric analysis. Appl. Energy 87, 141–148.
Muthuraman, M., Namioka, T., Yoshikawa, K., 2010b. A comparative study on co-
combustion performance of municipal solid waste and Indonesian coal with high
ash Indian coal: a thermogravimetric analysis. Fuel Process. Technol. 91, 550–558.
Mwanza, B., Phiri, A., 2013. Design of a waste management model using integrated solid
waste management: a case of Bulawayo City Council. International Journal of Water
Resources and Environmental Engineering 5, 111–118.
National Pollutant Inventory, d. Emission estimation technique manual for Railway yard
operations.Available from:. http://www.npi.gov.au/system/files/resources/
d8838f77-fc0d-7e14-dd35-e8a9f84a53fe/files/frailway.pdf, Accessed date: 5 August
2017 (Online).
Nguyen, T.T., Wilson, B.G., 2010. Fuel consumption estimation for kerbside municipal
solid waste (MSW) collection activities. Waste Manag. Res. 28, 289–297.
Niksa S, Liu G, Felix L, Bush PV, Boylan DM. Predicting NOX emissions from biomass
cofiring. 28th Int. Technical Conf. on Coal Utilization and Fuel Systems, 2003.
Novianti, S., Biddinika, M.K., Prawisudha, P., Yoshikawa, K., 2014. Upgrading of palm oil
empty fruit bunch employing hydrothermal treatment in lab-scale and pilot scale.
Procedia Environ. Sci. 20, 46–54.
Nyamombe WC, Gwisai RD. Assessing the source separation and recycling potential of
household waste. A case study of Gadzema suburb in Chinhoyi, Zimbabwe. Open
Journal of Applied & Theoretical Environmental Sciences (OJATES) 2015; Vol. 1:
56~71.
Ogbu MO, Oyeyinka BO, Mlawa HM. Technology Policy and Practice in Africa: IDRC, 1995.
Olatunji, S., Fakinle, B., Jimoda, L., Adeniran, J., Adesanmi, A., 2015. Air emissions of sul-
phur dioxide from gasoline and diesel consumption in the southwestern states of
Nigeria. Pet. Sci. Technol. 33, 678–685.
O'Leary, P.R., Tchobanoglous, G., 2002. Landfilling. In: Tchobanoglous, G., Kreith F., Hand-
book of Solid Waste Management. (2nd Ed.). McGraw-Hill, USA.
Omari, A., Said, M., Karoli, N., Geoffrey, J.R., 2014. Energy recovery routes from municipal
solid waste: a case study of Arusha-Tanzania. The Journal of Energy Technologies and
Policy 4 (5), 1–7.
Oumarou, M., Dauda, M., Abdulrahim, A., Abubakar, A., 2012. Characterization and gener-
ation of municipal solid waste in north central Nigeria. International Journal of Mod-
ern Engineering Research 2, 3669–3672.
Palmer A. The Development of an Integrated Routing and Carbon Dioxide Emissions
Model for Goods Vehicles. Cranfield University. PhD Thesis 2007.
Practical Action, 1999. Department for International Development Contract: Research
Development and Design of a Simple Solid Waste Incinerator.
Prawisudha, P., 2011. Chlorine-free Solid Fuel Production From Municipal Solid Waste by
Hydrothermal Treatment for Co-combustion With Coal. Environmental Science and
Technology. PhD, Tokyo Institute of Technology.
Prawisudha, P., Namioka, T., Yoshikawa, K., 2012. Coal alternative fuel production from
municipal solid wastes employing hydrothermal treatment. Appl. Energy 90,
298–304.
Pressley, P.N., Levis, J.W., Damgaard, A., Barlaz, M.A., DeCarolis, J.F., 2015. Analysis of
material recovery facilities for use in life-cycle assessment. Waste Manag. 35,
307–317.
 A.B. Maqhuzu et al. / Science of the Total Environment 668 (2019) 743–759
Reza, M.T., Andert, J., Wirth, B., Busch, D., Pielert, J., Lynam, J.G., et al., 2014. Hydrothermal
carbonization of biomass for energy and crop production. Applied Bioenergy 1.
Roberts, D., 2015. Characterisation of chemical composition and energy content of green
waste and municipal solid waste from Greater Brisbane, Australia. Waste Manag. 41,
12–19.
Román, S., Libra, J., Berge, N., Sabio, E., Ro, K., Li, L., et al., 2018. Hydrothermal carboniza-
tion: modeling, final properties design and applications: a review. Energies 11, 216.
Safril, T.S., Safril, B.I., Yoshikawa, K., 2017. Commercial demonstration of solid fuel produc-
tion from municipal solid waste employing the hydrothermal treatment. DEStech
Transactions on Environment, Energy and Earth Sciences.
Sango, I., 2010. Analysis of the solid waste management practices in Chinhoyi: bridging
the missing link. Economia. Seria Management 13, 332–347.
Savage, G.M., Diaz, L.F., 1986. Key issues concerning waste processing design. Proceedings
of National W aste Processing Conference, ASME, New York, USA, pp. 361–373.
Scarlat, N., Motola, V., Dallemand, J.F., Monforti-Ferrario, F., Mofor, L., 2015. Evaluation of
energy potential of Municipal Solid Waste from African urban areas. Renew. Sust.
Energ. Rev. 50, 1269–1286.
Shekdar, A.V., 2009. Sustainable solid waste management: an integrated approach for
Asian countries. Waste Manag. 29, 1438–1448.
Shonhiwa, C., 2013. An assessment of biomass residue sustainably available for thermo-
chemical conversion to energy in Zimbabwe. Biomass Bioenergy 52, 131–138.
Simelane T, Mohee R. Future Directions of Municipal Solid Waste Management in Africa.
Africa Institute of South Africa. Briefing No 81. 1–5 2012.
Solomon, S., Qin, D., Manning, M., Chen, Z., Marquis, M., Averyt, K., et al., 2007. Contribu-
tion of Working Group I to the Fourth Assessment Report of the Intergovernmental
Panel on Climate Change. Cambridge University Press, Cambridge.
Takaedza, R.J., 2014. The Solid Waste Management in Mubaira Growth Point, Chegutu
Rural District Council. Midlands State University, BSc Thesis.
Togarepi, S., Tsiko, R., 2012. A situational analysis of waste management in Harare,
Zimbabwe. Am. J. Sci. 8, 692–706.
759
Turchi CS, Langle N, Bedilion R, Libby C. Solar-augment potential of US Fossil-Fired power
plants. Technical Report NREL/TP-5500-50597. National Renewable Energy Labora-
tory 2011: 641–651.
U.S. Energy Information Administration (EIA), d. Voluntary reporting of greenhouse gases
program fuel emission coefficients.Available from:. http://www.eia.gov/oiaf/1605/
coefficients.html#tbl2, Accessed date: 5 August 2017 (Online).
Valkenburg, C., Gerber, M., Walton, C., Jones, S., Thompson, B., Stevens, D.J., d. Municipal
solid waste (MSW) to liquid fuels synthesis, volume 1: availability of feedstock and
technology.Available from:. https://www.energy.gov/sites/prod/files/2014/03/f14/
PNNL-18144_0.pdf, Accessed date: 5 September 2017 (Online).
W.H.O Cholera in Zimbabwe: Epidemiological Bulletin. Number 3. World Health Organi-
sation, Harare, 2008.
Williams, P.T., 2013. Waste Treatment and Disposal: John Wiley & Sons.
World Bank, d. GDP per capita (current US$).Available from:. http://api.worldbank.org/
v2/en/indicator/NY.GDP.PCAP.CD?downloadformat=excel, Accessed date: 10 Janu-
ary 2019 (Online).
World Bank, d. World development indicators.Available from:. http://api.worldbank.org/
v2/en/country/ZWE?downloadformat=excel, Accessed date: 6 August 2017
(Online).
Zamba BF. Community Participation in Sustainable Solid Waste Management in High
Density Suburbs. The Case of Mbare Harare, Zimbabwe. BSc Thesis, Midlands State
University 2014.
Zhao, P., Shen, Y., Ge, S., Chen, Z., Yoshikawa, K., 2014. Clean solid biofuel production from
high moisture content waste biomass employing hydrothermal treatment. Appl. En-
ergy 131, 345–367.
Zhou, P.P., Yamba, F.D., Lloyd, P., Nyahuma, L., Mzezewa, C., Kipondya, F., et al., 2009. De-
termination of regional emission factors for the power sector in Southern Africa. Jour-
nal of Energy in Southern Africa 20, 11–13.