Environmental Engineering Reference
In-Depth Information
8.3.2 Pilot-Scale Testing
Demonstration and analysis of Osorb treatment effectiveness at a scale relevant to actual
ield deployment is important in determining the practical capabilities of the technol-
ogy. Many nanoengineered technologies fail to bridge the gap between laboratory and
practice because of the inability to scale-up quickly or do so economically. One of the
key challenges in moving Osorb to pilot-scale testing was manufacturing enough media.
Manufacturing methods were thus developed to produce the lexible, nanoengineered,
organosilica matrix in large reactors to facilitate a substantial amount of pilot work. Two
produced water treatment systems were able to be ielded due to the success of the manu-
facturing development:
1. A 4 gpm skid-mounted ixed bed design, “Skid 1”
2. A 60 gpm luidized bed design, “PWUnit#1”
The 4 gpm skid unit, Skid 1, was a straightforward system that included two bag ilter
vessels acting as packed columns. Skid 1 was designed to operate the bag ilter vessels
either in a series or in a parallel coniguration. Each bag ilter vessel was itted with illed,
ixed media ilter bags of Osorb that could be replaced. Each ilter bag contained a 4-in.
layer of ilter sand and 1.5 kg Osorb packed in a manner to provide room to swell, if neces-
sary. Skid 1 was tested in conjunction with David Burnett, Director of Technology of the
Global Petroleum Research Institute at Texas A&M. The Skid 1 ixed-bed treatment sys-
tem was used to treat produced water at the Texas A&M testing facilities. This water was
pretreated, but contained amounts of BTEX, dispersed oil, and grease. The Skid 1 ixed
bed treatment system treated 100 gal of produced water at a rate of 2 gpm during testing.
The samples were analyzed by an independent laboratory. The results show a substantial
decrease in BTEX, oil, and grease (Table 8.4).
It was determined through testing and modeling that when moving to a produced water
low rate >10 gpm, a luidized bed treatment method was preferential to a packed col-
umn design. A luidized bed design (PWUnit#1) was built to operate at a 60 gpm capacity
and be trailer-mounted for transportation to sites with access to produced water. Contact
between the Osorb and the contaminated water occurred in a pair of 175-gallon stainless-
steel tanks. Inside these tanks, a pair of eductors were used to provide suficient turbu-
lence and mixing for absorption (Figure 8.11). Eductors are nozzles designed to create an
TABLE 8.4
Extraction of Produced Water Hydrocarbons by 2 gpm Cartridge System
Concentration (ppb)
Percent
Reduction
Solute
Input
25 gal
50 gal
75 gal
100 gal
Benzene
4200
114
139
194
191
95
Ethylbenzene
94
1.6
2.2
2.5
2.7
97
Naphthalene
10
5.2
5.9
7.0
7.3
27
Toluene
244
46
55
88
83
62
1,2,4-Trimethylbenzene
10
1.1
1.4
1.9
1.9
81
1,2,4-Trimethylbenzene
3.3
0.3
0.4
0.6
0.4
84
m
+
p
-Xylene
39
6.3
8.6
10
11
64
o
-Xylene
23
5.1
6.8
8.0
8.6
69
Oil and grease
11,500
-
-
-
0
100
