Professional Documents
Culture Documents
Waste Management
journal homepage: www.elsevier.com/locate/wasman
a r t i c l e i n f o a b s t r a c t
Article history: Vacuum evaporation consists in the boiling of a liquid substrate at negative pressure, at a temperature
Received 21 September 2012 lower than typical boiling temperature at atmospheric conditions. Condensed vapor represents the so
Accepted 15 February 2013 called condensate, while the remaining substrate represents the concentrate.
Available online 4 April 2013
This technology is derived from other sectors and is mainly dedicated to the recovery of chemicals from
industrial by-products, while it has not been widely implemented yet in the field of agricultural digestate
Keywords: treatment. The present paper relates on experimental tests performed in pilot-scale vacuum evaporation
Anaerobic digestion
plants (0.100 and 0.025 m3), treating filtered digestate (liquid fraction of digestate filtered by a screw-
Digestate treatment
Biogas
press separator). Digestate was produced by a 1 MWe anaerobic digestion plant fed with swine manure,
Vacuum evaporation corn silage and other biomasses. Different system and process configurations were tested (single-stage
Concentration and two-stage, with and without acidification) with the main objectives of assessing the technical feasi-
bility and of optimizing process parameters for an eventual technology transfer to full scale systems.
The inputs and outputs of the process were subject to characterization and mass and nutrients balances
were determined.
The vacuum evaporation process determined a relevant mass reduction of digestate.
The single stage configuration determined the production of a concentrate, still in liquid phase, with a
total solid (TS) mean concentration of 15.0%, representing, in terms of mass, 20.2% of the input; the
remaining 79.8% was represented by condensate. The introduction of the second stage allowed to obtain
a solid concentrate, characterized by a content of TS of 59.0% and representing 5.6% of initial mass.
Nitrogen balance was influenced by digestate pH: in order to limit the stripping of ammonia and its
transfer to condensate it was necessary to reduce the pH. At pH 5, 97.5% of total nitrogen remained in
the concentrate. This product was characterized by very high concentrations of total Kjeldhal nitrogen
(TKN), 55,000 mg/kg as average.
Condensate, instead, represented 94.4% of input mass, containing 2.5% of TKN. This fraction could be
discharged into surface water, after purification to meet the criteria imposed by Italian regulation. Most
likely, condensate could be used as dilution water for digestion input, for cleaning floor and surfaces of
animal housings or for crop irrigation.
The research showed the great effectiveness of the vacuum evaporation process, especially in the two
stage configuration with acidification. In fact, the concentration of nutrients in a small volume deter-
mines easier transportation and reduction of related management costs. In full scale plants energy con-
sumption is estimated to be 5–8 kWhe/m3 of digestate and 350 kWht/m3 of evaporated water.
Ó 2013 Elsevier Ltd. All rights reserved.
0956-053X/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.wasman.2013.02.023
1340 A. Chiumenti et al. / Waste Management 33 (2013) 1339–1344
Fig. 2. Schematics of the first stage (upper scheme) and of the second stage of the system (source: Italschell).
total Kjeldhal nitrogen (TKN) (APHA, AWWA, WPCF Standard 3. Results and discussion
methods, 1992) (Table 1).
Preliminary analyses performed on digestate allowed excluding 3.1. Process parameters
the presence of nitrites and nitrates, allowing to consider TKN as an
indication of total nitrogen content. The concentration units were set to operate at a temperature of
Acid consumption, when applicable, was determined manually 35 °C and at a relative pressure of 96 kPa, equivalent to an
by mass and volume determinations. absolute pressure of 5.3 kPa. Pressure and temperature were
Table 1
Synthesis of monitored parameters and methods. (TS = total solids, VS = volatile solids, TKN = total Kjeldhal nitrogen).
constant during the tests and did not vary significantly from the set Digestate from the biogas plant was characterized by TS content
values. from 5.6% to 5.7%, TKN varying from 4700 to 3510 mg/kg and pH
The first unit operated with treatment cycles of about 40–45 h, from 7.6 to 7.9.
the second unit was characterized by 10–15 h of operation: tests Filtered digestate used for the tests was characterized by
were interrupted when the production of condensate stopped, limited variations of the TS content (3.3–3.4%), while relevant
indicating that the evaporation ceased. variations were detected in terms of volatile solids content,
The treatment capacity was of about 500 kg of filtered digestate 54.2–70.0%TS, and of TKN, ranging from 3380 to 4550 mg/kg. The
per cycle for the first unit and 20–25 kg/cycle for the second unit. value of pH varied from 7.6 to 7.9. The variability of TS and TKN
The relevant difference is mainly related to the continuous load of in digestate and, hence, in filtered digestate, was detected in
the first unit and the discontinuous of the second. several monitored plants, depending on several factors related to
the management of the biogas plant, such as quality and quantity
3.2. Mass balance of feedstocks and mixing level in digesters (Chiumenti et al., 2010).
Filtered digestate quality, furthermore, varies from plant to
In terms of mass balance a significant effect was determined by plant depending on the type of separation equipment. Also in this
the configuration of the plant. In case of single stage treatment, in particular case the feeding of the monitored plant was subject to
fact, it was possible to obtain a concentrate characterized by being variations in the 2 years’ time span of the study, but the variations
liquid, even if dense. In terms of mass, concentrate represented of digestate quality did not affect the vacuum evaporation process.
20.2% of input, while the condensate represented the remaining The first stage of the process determined an increase of TS up to
79.8%. 15.0%, as average. The resulting concentrate was a dense liquid
The introduction of the second stage determined a significant fraction.
increase of the concentration rate, so that the final product re- This result is related to the configuration of the system, in par-
sulted solid. The reduction of mass was also increased: the solid ticular to the static conditions of the product (not stirred) and to a
concentrate represented 5.6% of initial mass, while condensate progressive reduction of the efficiency of the heat exchangers dur-
from the second stage resulted of 14.6% of the input, with a com- ing the process, due to the formation of deposits over the surface of
prehensive production (both from first and second stage) repre- exchangers.
senting 94.4% of input mass. Experimental tests carried out by other authors, with single
stage systems, demonstrated that the highest content of solids that
3.3. Chemical characteristics can be achieved in the concentration of animal manure was be-
tween 20% and 25% TS (ten Have and Chiappini, 1993).
Tables 2 and 3 report the most representative characteristics of The two stage process allowed to obtain a solid concentrate
input and outputs of the process, in relation to test series n.1 and with an average TS content of 59.0%. Other authors underline that
n.2 (single stage treatment, with and without acidification) and the vacuum evaporation process is not capable to achieve complete
to test series n.3 (two stage process with acidification). drying of the product (Bonmatí and Flotats, 2003).
In particular, the evolution of TS with time was monitored to
evaluate the performance of the process: in Fig. 3 representative
Table 2 plots are depicted. The first unit reached the highest concentration
Average characteristics of input and output streams in case of test series n.1 and n.2:
of TS after 27–32 h of operation. The second unit determined a
single-stage process with and without acidification (n.d. = not determined). (TS = total
solids, VS = volatile solids, TKN = total Kjeldhal nitrogen).
steep increase of the concentration rate, reaching the highest TS
concentrations (63%) in 10–15 h of operation (Fig. 3).
Sample TS (%) VS (%TS) pH TKN
The chemical characteristics of concentrate and of condensate
(mg/kg)
(first stage), resulted strongly influenced by pH of filtered dige-
Test series n.1: single stage – without acidification state. Digestate itself, in fact, was characterized by a pH in the
Digestate 5.7 53.8 7.6 4700
Filtered digestate 3.4 54.2 7.6 4550
range of 7.5–7.9, resulting in chemical conditions favorable to
Condensate (without acidification) n.d. n.d. 7.7 3730 the stripping of gaseous ammonia. In these conditions, in fact, con-
Concentrate 15.1 75.8 8.6 5350 densate showed significant concentration of TKN, 3730 mg TKN/kg,
Test series n.2: single stage – with acidification compared to 4550 mg TKN/kg of filtered digestate; the
Digestate 5.6 71.0 7.9 3780
Filtered digestate 3.3 70.0 7.9 3660
Condensate (with acidification) n.d. n.d. 3.2 10
Concentrate 15.0 84.7 4.6 18,700
Table 3
Average characteristics of input and output streams in case of test series n.3: two-
stage process with acidification (n.d. = not determined). (TS = total solids, VS = volatile
solids, TKN = total Kjeldhal nitrogen).
higher than that of the test with lower pH. The concentrate was
FILTERED DIGESTATE
characterized by high TKN concentration, 16,890 mg/kg, slightly 100.0% MASS
lower than the concentration obtained with lower pH. The second 100.0% N
stage determined the production of a condensate characterized by VACUUM
CONDENSATE
TKN concentration of 46 mg/kg. CONCENTRATION
1st STAGE 79.8% MASS
Considering the potential destination of the condensate, dilu- 2.3% N
tion of biomasses to be destined to anaerobic digestion, necessary CONCENTRATE
in case of high ST feedstocks, this result appears to be satisfactory. 20.2% MASS
97.7% N
For a discharge on surface water of this fraction a higher acidificat-
ion is preferable. VACUUM CONDENSATE
CONCENTRATION 14.6% MASS
The final product of the two stage process was a solid concen- 2nd STAGE 0.2% N
trate characterized by very high concentration of TKN, 55,000 mg
CONCENTRATE
TKN/kg. Other authors obtained a concentrate characterized by a
5.6% MASS
TKN between 18,100 and 41,200 mg/kg (Bonmatí and Flotats, 97.5% N
2003).
Acid consumption varied from 30 to 40 kg/m3. Other authors re- Fig. 4. Average mass and nitrogen balances in the single stage system, test series
n.1 and n.2 (upper scheme). Average mass and nitrogen balance for the two stage
ported a consumption of 3.0–3.3 kg/m3 of H2SO4 for the acidificat-
vacuum concentration process with acidification, test series n.3 (lower scheme).
ion of digestate, in laboratory tests, but no indication was given
about the concentration of acid and pH shift (Panvini et al., 2009).
3.4. Nitrogen balance Test series n.3 (two stage) were performed with a pH reduction
to 5.0 to achieve a compromise between acid consumption and
Nitrogen balance was not only influenced by the pH of the input stripping of ammonia.
product (filtered digestate) but also by the configuration of the sys- In this case the first stage determined a separation in the con-
tem (one or two stage process). densate of 2.3% of TKN, while the second stage determined a sep-
In the test series n.1 (single stage, without acidification), filtered aration of 0.2% of TKN. At the end of the two stage process,
digestate was characterized by a pH in the range 7.5–7.9, hence, the concentrate, representing 5.6% of input mass, contained
conditions favorable to the stripping of ammonia from the product. 97.5% of TKN (Fig. 4). Since the process was performed in enclosed
As a result 78.1% of nitrogen was removed from digestate by strip- reactors, the loss of nitrogen in the atmosphere was prevented.
ping, and was lately recovered in the condensate, representing These results were confirmed, by other Authors that performed
79.8% of input mass. The concentrated fraction contained the tests on digestate: with a correction of pH < 6.5 removal of 25% of
remaining 21.9% of nitrogen (Fig. 4). This configuration, hence, nitrogen from the feedstock was obtained, while with a pH of 8.4
determined the production of significant quantity of condensate more than 50% of nitrogen passed to condensate even at 33% of
containing most of TKN and did not determine a significant con- evaporated mass (Bonmatí et al., 2003; Bonmatí and Flotats,
centration of TKN in concentrate. 2003; Panvini et al., 2009).
These tests resulted useful for a first determination of the distri-
bution of TKN in the different output fractions and showed that to
achieve the main goal of concentrating nutrients in limited quanti-
ties of product the vacuum evaporation process must be performed 4. Conclusions
with acidification.
The acidification to pH 3.5 performed in test series n.2 (single The performed experimental tests allowed to have indications
stage) limited the stripping of ammonia by shifting the chemical on the possibility to implement the vacuum evaporation process
equilibrium between gaseous ammonia and soluble ammonium in anaerobic digestion plants for a sustainable management of
to the soluble form: 99.2% of nitrogen was maintained in the liquid digestate.
concentrate, while only 0.8% of N was detected in condensate The most evident outcome is represented by the significant
(Fig. 4). The results obtained with this level of acidification could reduction of the mass of digestate to be destined to agricultural
be considered satisfactory in terms of concentration of nutrients utilization. This result was more evident in case of a two stage pro-
and quality of condensate, since almost all input TKN was collected cess that transformed digestate in a solid product, containing 97.5%
in concentrate. of input TKN, in 5.6% of mass.
1344 A. Chiumenti et al. / Waste Management 33 (2013) 1339–1344