Content uploaded by Camilo Urbina
Author content
All content in this area was uploaded by Camilo Urbina on Nov 16, 2015
Content may be subject to copyright.
Applied Engineering in Agriculture
Vol. 31(4): © 2015 American Society of Agricultural and Biological Engineers ISSN 0883-8542 DOI 10.13031/aea.31.9945 1
TECHNICAL NOTE:
PILOT PLANT FOR BORON REMOVAL FROM NATU R AL LY
CONTAMINATED IRRIGATION WATER
C. A. Urbina, L. F. Figueroa, N. A. Lara, H. A. Escobar, Y. A. Zapata
ABSTRACT. A pilot plant for removing boron from irrigation water was installed in a commercial farm located in the
Lluta River Valley (Arica, Chile). Water prior treatment contained dissolved boron concentration ranging from 17.98 to
23.76 mg L-1. After treatment, boron concentration substantially dropped to 1.28 to 1.79 mg L-1. Treatment did not affect
either pH or electrical conductivity of treated water. Over a period of 8 months, 7,440 m3 ha-1 (898.55 mm) of the treated
water were used for irrigation of a commercial field of tomato which produced 130.4 Mg ha-1 and over 75% of the highest
value commercial fruit size. The average operational cost of the pilot plant was US$0.46 m-3. This operational cost, which
amounts to 6.5% of the tomato production cost in the area where the study was conducted, is compatible with a profitable
agricultural activity under the conditions of Lluta Valley.
Keywords. Boron, Cost, Toxicity, Water treatment.
oron is both an essential element for plant
nutrition and a toxic element within a narrow
margin of concentrations in soil solution for
most plant species, and particularly for crops of
economic importance (Maas, 1990) albeit there is a
considerable variation of sensitivity to the toxicity across
species and between cultivars (Van der Leeden et al.,
1990). However, even relatively tolerant crops might
experience a detriment of the yield and/or quality of the
harvest due to the need of synthesizing compounds to
immobilize boron at the cytoplasm level that uses
metabolic energy that therefore is not available for growth
(Nable et al., 1997).
The northern portion of the Atacama Desert of Chile,
has at least three valleys with relatively important
horticultural crop production, yet two of them (Lluta and
Camarones) are affected by the natural occurrence of high
boron concentrations in both surface and groundwater
sources of irrigation water (Figueroa, 2006). The Lluta
River, located 15 km north of the city of Arica, is known
for having a Boron concentration varying from 10 to
40 mg L-1, while subterranean waters of this basin can have
Boron concentrations that range from 25 to 100 mg L-1.
Total salt content is also relatively high, which results in
electric conductivities (EC) in the range of 1.8 to 4 dS m-1
in surface waters, and from 3.5 to 10 dS m-1 in subterranean
waters. Even with these detrimental factors present in the
irrigation water, the availability of this resource (average of
2,000 L s-1 for Lluta River) and the subtropical weather of
this region, have allowed the historic development of a
regionally important level of agriculture, which is
associated with the cultivation of pre-incaic highly tolerant
ecotypes of maize and alfalfa, and the production of
modern cultivars of onion, garlic, and beetroot although of
limited yield and quality (Figueroa et al., 2000).
As a way to assess the impact of Boron in crop produc-
tivity in economic terms, an example exists in the same
geographical area: the Azapa Valley, which, albeit being
relatively brackish (EC ranging from 1 to 4 dS m-1) is
relatively low in Boron (ranging from 0.5 to 2 mgB L-1),
and due to this, using the same cultural practices and the
same growing season that in Lluta Valley, crop producers
obtain higher yields and cultivate a broader diversity of
profitable off season crops. This situation is summarized in
the value of the land, which is between 5 to 20 times higher
in Azapa Valley than in Lluta Valley (Torres and Acevedo,
2008).
The production of horticultural crops is less affected by
the salinity of irrigation water than by high boron levels.
Submitted for review in September 2012 as manuscript number NRES
9945; approved as a Technical Note for publication by the Natural
Resources & Environmental Systems Community of ASABE in April
2015.
The authors are Camilo A. Urbina, Research Associate, Leonardo F.
Figueroa, Associate Professor, Nelson A. Lara, Associate Professor, and
Yubinza A. Zapata, Research Associate, Departamento de Química,
Facultad de Ciencias, Universidad de Tarapacá. Av. General Velásquez,
Arica, Chile; and Hugo A. Escobar, Director, Laboratorio de Cultivo de
Tejidos Vegetales, Facultad de Ciencias Agronómicas, Universidad de
Tarapacá, Velásquez, Arica, Chile. Corresponding author: C. A. Urbina,
Departamento de Química, Facultad de Ciencias, Universidad de
Tarapacá. Av. General Velásquez 1775, Arica, Chile; phone: 56 58
205000; e-mail: camilo.urbina.alonso@gmail.co
m
.
B
2 APPLIED ENGINEERING IN AGRICULTURE
The yield response to increasing levels of salinity is
gradual, while the response of yield to Boron concentration
increase is abrupt for sensitive crops (Ferreyra et al., 1997).
Tomatoes are sensitive to boron toxicity showing a
decrease in the quality and volume of the yield after a
certain threshold (Oyinlola, 2005; Kaya et al., 2009). The
tomato crop in Lluta Valley is reported to produce yields as
low as 30 Mg ha
-1 and up to 60 Mg ha-1 due to boron
toxicity, while for the same level of management practices
and productive inputs tomato yields reported for Azapa
Valley is of 180 Mg ha-1. (Albornoz et al., 2007). These
authors also found that yield and quality of an experimental
crop of indeterminate tomato is not affected by high levels
of salinity (3.1 and 396 dS m-1) but is significantly and
adversely affected by a high level of Boron (0.8 vs
7.1 mgL-1), and also, that the same high salinity water, once
treated for removal of boron to a concentration below 1 mg
L-1, shows no significant difference with the low salinity
low boron irrigated control.
The aim of this work was to study the economic aspects
of employing a specific boron removal ion exchange resin
for treating irrigation water, with the hope of providing
better economic prospects to the local producers.
MATERIAL AND METHODS
A pilot plant designed to remove boron from irrigation
water was constructed at a commercial farm on the Lluta
river Valley, at coordinates 18°24’26.90”S,
69°59’38.48”W, to investigate the technical and economic
feasibility for use in tomato production. The water to be
treated was obtained from the “Boca Negra” channel,
which delivers water from the Lluta River, and was
accumulated in a reservoir with a capacity of 1,000 m3. The
treated water was stored in a second reservoir with a
capacity of 300 m3, prior to be used for irrigation. An
evaporation pit of 300 m2 with 0.4 m of depth was built for
evaporation of discarded regenerant solutions. All the
reservoirs and evaporation pit were lined with 1 mm thick
high-density polyethylene (HDPE) lining.
TREATMENT PLANT DESCRIPTION AND OPERATION
The basic configuration of the pilot plant includes two
main reservoirs, one that will serve to store the water prior
to treatment, and the other for accumulation of the treated
water. The treatment plant includes a system that pumps the
water trough a quartz sand filter and a vessel containing the
boron-specific ion exchange resin. The treated water is
metered on the treatment plant water outlet. The pilot plant
also includes special tanks for preparation of the
regeneration solutions and a chemical corrosion resistant
pump for circulation of the regeneration solutions through
the resin column. An additional evaporation pit is required
for the discard of the spent regeneration solutions. A
separate pumping station takes the treated water from the
reservoir and distributes it to the drip irrigation system
(fig. 1).
The resin vessel was custom made of fiberglass material,
with a diameter of 0.92 m and a height of 2.1 m. All piping
and connections were built using commercial grade
hydraulic PVC piping with an external diameter of 32 mm.
The resin vessel was partially filled with 600 L of an
industrial grade specific boron resin with N-methyl D-
glucamine functional groups (Xi-nan Fine Chemical Co.
Ltd., Sichuan, China). The raw water pump was a
commercial centrifugal irrigation pump (Model SM 100,
Reggio, Reggo Emilia, Italy). The regenerant solution
pump was a commercial centrifugal pump with all its parts
in contact with the regeneration solutions made of chemical
resistant plastics and/or stainless steel (ESPA, model
Tifon). Dosing pumps were used to supply regenerant
Figure 1. Schematic layout of the pilot plant built for boron removal in water for irrigation purposes.
Treated water
Regenerant
Solution Tanks
Treated water Reservoir
Treated water
pump
Farm Irri
g
ation S
y
ste
m
Untreated
water pump
Quartz Filtration
Untreated
water
Resin Column
Concentrated Boron
Brine
evaporation pit
Re
g
enerant Solution
31(4): 3
solution tanks with the concentrated regenerant chemicals
(DOSIVAC, Model Millennium 300, Loma Hermosa, San
Martin , Argentina).
Preparation of acid regenerant solution was performed
using 18 M industrial grade sulphuric acid (Oxiquim S.A.,
Iquique, Chile) diluted to form a 0.25 M acid solution,
prepared with untreated irrigation water. Alkaline
regenerant solution was prepared with 18 M industrial
grade sodium hydroxide (Oxiquim S.A., Iquique, Chile) to
form a 0.5 M diluted solution with untreated water. Spent
regenerant solutions are discarded in the evaporation pit
and periodically the salts are removed. The plant operation
steps are described in table 1.
W
ATER ANALYTICAL PROCEDURES AND
DETERMINATIONS
Triplicate samples of 75 mL were taken monthly from
untreated irrigation water during the 8 months the pilot
plant was under study. Once every two months, treated
irrigation water was sampled during a single cycle of
operation every 5 cubic meters, in order to assess the resin
exhaustion curve. The treated irrigation water was also
sampled monthly by triplicate from the reservoir in order to
check the quality for irrigation purposes. All these samples
were subject to determination of dissolved Boron, pH and
electric conductivity. The Boron content in irrigation water
was determined by the Azomethine H spectrophotometric
method (Shanina et al., 1967) using a UV/Vis Spectropho-
tometer (Spectronics, model Genesys 6), pH was
determined with a calibrated pH meter (WTW Instruments,
model 330i). Electric conductivity was determined with a
temperature calibrated conductivity meter (WTW
instruments, model 330i).
PILOT SCALE TOMATO CROP SET UP AND PRODUCTIVE
DATA REGISTRATION
A field of 0.99 ha was prepared for cultivation, with
92 crop rows spaced at 1.5 m, measuring 60 m each, for a
total effective surface of 0.828 ha. Each row was planted
with 260 tomato (Lycopersicon esculentum Mill.) seedlings
of cv. Naomi (ENZA ZADEN Export B.V.) for a total
amount of 23,920 plants at start of the study. The planting
was performed by hand on 17 April 2010. Irrigation was
provided by means of drip irrigation tape (Toro Aqua
Traxx® PC, Bloomington. Minnesota, USA) with emitters
spaced at 0.2 m and a emitter flow rate of 1 L h-1, using
60 m of tape per crop row. Plants were irrigated daily with
water treated by the Pilot Plant for boron removal. Water
was applied at a rate to maintain proper soil-water content
levels as assessed by visual inspection of plant turgidity,
which required 2.41 to 9.06 mm d-1 of treated water during
the season. Harvest was performed by hand 3 times per
week starting mid-July 2010 and continuing until end of
November 2010. Production of each harvest (total fruit
yield and yield per specific commercial classification of the
fruits) was recorded. Fruit were classified following the
general Chilean commercial criteria described by Albornoz
(2007) as: extra (> 250 g), 1° (150-250 g), 2° (100-149 g),
and 3° (80-99 g). Classification as “Discard” included fruit
below 80 g, cracked fruit, or with other kind of visible
damage (mainly due to insects or birds).
ASSESSMENT OF WATER BORON REMOVAL OPERATIONAL
COSTS
As costs involved in fruit production are of interest to
producers, the operational cost of the irrigation water
treatment pilot plant was determined by considering all
chemical inputs and labor. All prices used for the
calculations are the average of prices paid for the resources
required for the operation of the pilot plant during the
study. Prices are expressed in US$ at the rate of change for
the Chilean currency (1US$= 480 Clp).
RESULTS AND DISCUSSION
IRRIGATION WATER CHARACTERIZATION
“Boca Negra” irrigation channel water quality, with
respect to Boron concentration, pH and salinity (measured
by Electric Conductivity) was moderately variable
throughout the study period, which was expected (table 2),
while all the values of this parameter measured during the
study period were in the more phytotoxic range following
the classification of “non usable for irrigation” even for
tolerant crops under the categorization of Van der Leeden
(1990). Minimum dissolved boron concentration measured
was during April 2010 at 17.98 mg L-1 and the maximum
concentration was during November 2010, at 23.76 mg L-1.
TREATED WAT ER CHARACTERIZATION
The boron concentration of the treated irrigation water
during the operation cycle of the resin column resulted in
an average exhaustion curve as shown in figure 2. The
Table 1. Description of the operational steps of the pilot plant for boron removal from irrigation water.
Operation Step Description
Treatment of irrigation water
for removal of boron
The water is pumped from the untreated water reservoir through the quartz filtration unit and the resin column. This is
performed at a flow rate from 12 to 18 resin bed volumes per hour. A total amount of 90 m3 of water is treated in each
treatment cycle of up to 15 h.
Removal of boron from the
column
A 0.25 M acid solution is prepared using industrial grade 18 M sulphuric acid diluted in 1 m3 of untreated water. This
solution is circulated at a flow rate of 120 L min-1 through the resin column for 15 min, later discarded to the
evaporation pit. This dilute acid solution removes all the boron retained in the resin. A rinse of 10 bed volumes of
untreated water at the same flow rate is performed to remove all the remaining acid solution in the column. The rinse
water is also discarded to the evaporation pit. This step takes a total time of 120 minutes.
Regeneration of the resin A 0.5 M alkaline solution is prepared using industrial grade 18 M sodium hydroxide diluted in 1 m3 of untreated water.
This solution is circulated at a flow rate of 120 L min-1 through the resin column for 15 min, and then discarded to the
evaporation pit. A rinse of 10 bed volumes of untreated water at the same flow rate is performed to remove all the
excess alkaline solution from the resin column. The rinse water is also discarded to the evaporation pit. The
regeneration step takes 120 min and the column is ready for another treatment cycle.
4 APPLIED ENGINEERING IN AGRICULTURE
curve corresponds to prior reports for specific boron
removal resins, in which initially boron is efficiently
removed until leakage is detected as the resin capacity
diminishes (De la Fuente, 2005; Jacob, 2007; Kabay et al.,
2008).
On the other hand, pH and electric conductivity were not
affected by the treatment, and the values were consistently
similar to the untreated water (table 3).
The content of dissolved boron in the treated water
reservoir was steady throughout the study period (table 4).
As can be seen, the use of the reservoir allows for a
homogenization of the boron content in the water, which is
reduced by at least 90% with respect to the original boron
content in the untreated water. All dissolved boron
concentrations measured in the treated water reservoir
during the study, during the study are classified as
“permissible for irrigation” for boron tolerant and
semitolerant crops (Van de Leeden, 1990) and “doubtful
for irrigation” for sensitive crops. Tomatoes are considered
semi-tolerant to boron in the scale of boron sensitivity of
crops by Maas (1990). In any case, the dissolved boron
concentrations achieved through the treatment of the water
in the pilot plant, are comparable to the values reported for
irrigation waters in Azapa valley by Torres and Acevedo
(2008), which is a location not considered to have
significant restrictions for agriculture due to boron toxicity.
IRRIGATION WATER USE AND LABOR REQUIREMENTS
FOR WATER TREATMENT
The average daily treated water by month ranged from
2.17 to 616 mm d-1 during the study period (table 5), with
daily values ranging between 0.97 and 8.57 mm d-1 during
the season. As can be seen, applied water increased steadily
Table 2. Selected Lluta river water quality
parameters during April to November 2010.
Dissolved B EC
Sample Date (mg L-1) pH (dS m-1 at 25°C)
10 April 2010 17.98 8.18 1.90
10 May 2010 18.87 8.34 1.93
10 June 2010 18.95 8.23 2.02
10 July 2010 19.16 8.45 2.18
10 August 2010 19.34 8.41 2.34
10 September 2010 20.28 8.49 2.31
10 October 2010 21.45 8.56 2.43
10 November 2010 23.76 8.22 2.18
Figure 2. Average specific boron resin exhaustion curve during the study.
Table 3. Average values of water sampled during
the boron removal cycle at pilot plant.
Treated Water Dissolved Boron EC
(m3) (mg L-1) pH (dS m-1 at 25°C)
5 0.00 8.33 2.19
10 0.00 8.36 2.21
15 0.00 8.27 2.17
20 0.00 8.42 2.34
25 0.00 8.35 2.28
30 0.00 8.22 2.56
35 0.00 8.17 2.45
40 0.00 8.26 2.36
45 0.00 8.67 2.41
50 0.15 8.23 2.37
55 0.18 8.12 2.23
60 0.33 8.18 2.29
65 0.53 8.31 2.18
70 1.01 8.39 2.32
75 1.86 8.26 2.27
80 3.68 8.37 2.19
85 7.47 8.78 2.23
90 11.84 8.65 2.39
Table 4. Selected Treated water quality parameters
during April 2010 to November 2010.
Dissolved B EC
Sample Date (mg L-1) pH (dS m-1 at 25 °C)
10 April 2010 x1.47 8.34 2.01
10 May 2010 1.63 8.19 1.97
10 June 2010 1.78 8.13 2.03
10 July 2010 1.69 8.19 2.21
10 August 2010 1.35 8.23 2.19
10 September 2010 1.29 8.33 2.34
10 October 2010 1.33 8.20 2.18
10 November 2010 1.48 8.31 2.27
Average resin capacity exhaustion curve measured in the pilot
plant during the study
0
1
2
3
4
5
6
7
8
9
10
11
12
13
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90
Treated Water (m
3
)
Dissolved Boron mg l
-1
31(4): 5
through the growing season as the crop developed and
average temperatures rose during spring and summer.
During the season, the pilot plant required an average of 5
hours of labor per day, which was mainly required for the
exhausted resin regeneration process.
CROP YIELD
Nearly 75% of the tomatoes harvested during the study
were in the highest commercial value category, with 17%
in the middle category, 7% in the lowest category and only
1% discarded. Fruit quality, as based upon fruit size, is of
particular interest for producers as local fresh market
tomatoes have a 30% to 50% price premium for the two
higher classes (fig. 3). Total production recorded was
108 tons in the field used for the study (130.4 Mg ha-1). The
production of tomato observed during the study is high,
both in total production and also in commercial value,
compared to the values reported by Albornoz (2007).
ECONOMICAL ASSESSMENT OF WATER TREATMENT
COSTS
The total cost of operation of the boron removal pilot
plant was US$0.46 m3 (table 6). The factors that accounted
for more than the half of this cost were labor and sodium
hydroxide. The operational cost recorded is relatively high
for irrigation water considering that normally in Lluta
Valley producers have the water available nearly for free.
However, the total direct cost of the water for the crop
during the study (US$3,421), given that it allows an
increase of both total amount and value of the production,
is profitable within the variable cost budget of tomato for
fresh consumption in the Chilean market. For reference, the
total cost budget for the pilot-scale tomato crop was
US$52,000, from which the water treatment was 6.5%, and
the total income was US$83,333. On the other hand, the
irrigation water treatment pilot plant was built at a cost of
US$31,250.
CONCLUSIONS
The results of the present study indicate that, under the
conditions of the Lluta River Valley, Arica, Chile, is
feasible to implement the use of specific boron removal
resin systems for treatment of irrigation water, neutralizing
most of the phytotoxic effects of this element on the
cultivation of tomato for the fresh market. The values of
dissolved boron observed for water prior to treatment,
ranging from 17.98 to 23.76 mg L-1 are incompatible with
use for irrigation of even boron tolerant plants. The values
of dissolved boron observed in the treated water, ranging
from 1.33 to 1.78 mg L-1, are compatible with the use for
irrigation of tomato and several other boron semi tolerant
and tolerant horticultural crops. The observed cost of the
treated water, which was US$0.46 m-3, while high, is
affordable within the normal budget for off season
indeterminate tomato production, which obtains high
market prices in Chile.
ACKNOWLEDGMENTS
The authors wish to thank for financial support of this
research to funding from the Performance Agreement
between University of Tarapaca and the Chilean Ministry
of Education (Convenio de Desempeño Universidad de
Tarapacá – MINEDUC).
REFERENCES
Albornoz, F., Torres, A., Tapia, M. L., & Acevedo, E. (2007).
Hydroponic tomato (Lycopersicon esculentum Mill.) crop with
desalted and deborified water in Lluta Valley. IDESIA, 25(2),
73-80. http://dx.doi.org/10.4067/S0718-34292007000200010.
De la Fuente García-Soto, M. N., & Muñoz Camacho, E. (2005).
Boron removal in industrial wastewaters by ion exchange: An
analytical control parameter. Desalination, 181, 207-216.
http://dx.doi.org/10.1016/j.desal.2005.01.015.
Ferreyra, R. E., Aljaro, A. U., Ruiz, R. S., Rojas, L. P., & Oster, J.
D. (1997). Behavior of 42 crop species grown in saline soils
with high boron concentrations. Agric. Water Mgmt., 34(2), 111-
124. http://dx.doi.org/10.1016/S0378-3774(97)00014-0.
Figueroa, L. (2006). Reducción de la concentración de boro en
aguas del río Lluta y su utilización en riego del suelo salino y
boratado. Sociedad Chilena de la Ciencia del Suelo, Universidad
de Tarapacá, Boletín No. 22, 148-150.
Table 5. Average daily treated water applied by month.
Irrigation Water Applied
Month (mm day-1) (mm month-1)
April 2010 2.17 65.22
May 2010 2.66 82.37
June 2010 2.66 79.71
July 2010 2.90 89.86
August 2010 3.74 116.06
September 2010 3.99 119.57
October 2010 5.19 160.99
N
ovember 2010 6.16 184.78
Total 898.55
Figure 3. Tomato fruit size distribution (% of each commercial
classification) obtained in the crop irrigated with the treated water.
Table 6. Components of operational costs
of the pilot plant used in the study.
Average Value Average Cost
Variable Unit (Units m-3) (US$ m-3)
Sulphuric Acid kg 0.11 0.08
Sodium Hydroxide kg 0.21 0.16
Electric energy kW 0.34 0.08
Labor man hours 0.07 0.14
Total Cost (US$ m-3) 0.73 0.46
6 APPLIED ENGINEERING IN AGRICULTURE
Figueroa, L. T., Bastías, E., Escobar, H., & Torres, A. (2000).
Niveles de boro en el agua de riego utilizada en oliviculura del
norte de Chile. In Gestión de la Calidad del Agua y Control de
la Contaminación en América Latina y el Caribe (pp. 167-188).
Santiago, Chile: Oficina Regional de la FAO Para América
Latina y El Caribe.
Jacob, C. (2007). Seawater desalination: Boron removal by ion
exchange technology. Desalination, 205, 74-52.
http://dx.doi.org/10.1016/j.desal.2006.06.007.
Kabay, N. S., Yuksel, M., Kitis, M., Koseiglu, H., Arar, O., Bryjak,
M., & Semiat, R. (2008). Removal of boron from SWRO
permeates by boron selective ion exchange resins containing N-
methyl glucamine groups. Desalination, 223, 49-56.
http://dx.doi.org/10.1016/j.desal.2007.01.199.
Kaya, C., Tuna, L., Dikilitas, M., Ashraf, M., Koskeroglu, S., &
Guneri, M. (2009). Supplementary phosphorus can alleviate
boron toxicity in tomato. Sci. Hort., 121(3), 284-288.
http://dx.doi.org/10.1016/j.scienta.2009.02.011.
Maas, E. V. (1990). Crop salt tolerance. In K. K. Tanji (Ed.),
Salinity Assessment and Management. New York, N.Y.:
American Society of Civil Engineers.
Nable, R. O., Bañuelos, G. S., & Paull, J. G. (1997). Boron toxicity.
Plant Soil, 193, 181-198.
http://dx.doi.org/10.1023/A:1004272227886.
Oyinlola, E. Y. (2005). Distribution of boron and its uptake in the
plant parts of two tomato varieties. Chem. Class J., 2, 77-80.
Shanina, T., Gelman, N., & Mikhailovskaya, W. (1967).
Quantitative analysis of heterorganic compounds
spectrophotometric microdetermination of boron. J. Anal. Chem.
USSR (Eng. Translation), 22, 663-667.
Torres, A., & Acevedo, E. (2008). The salinity problem of the water
and soil resources in Lluta and Azapa valleys in northern Chile.
IDESIA, 26(1), 31-44.
Van der Leeden, F., Troise, F. L., & Todd, D. K. (1990). The Water
Encyclopedia. Chelsea, Mich.: Lewis Publishers.
