The goal of this three-part series is to highlight advancements in Aquifer Storage and Recovery (ASR) well technology and to demonstrate how these advancements can improve the efficiency of ASR well systems.  In our introduction, we discussed the evolution of ASR well technologies and the current well recharge methods utilized in the industry; each with its advantages and disadvantages. Part One described the alternative recharge method we call “reverse siphon”, in which the hydraulic controls to prevent air entrainment and to initiate recharge are located at the surface at the wellhead, instead of connected at the pump assembly down in the well.  Part Two described the use of epoxy coating to prevent iron oxide particulates from clogging the well screen and filter pack interface. In this final installment of our series, we will discuss the use of glass beads as filter pack media to optimize well recharge rates and productivity.

History of Filter Pack Usage

First, let’s examine the purpose of a filter pack media installed between the borehole wall and well screen.  According to Groundwater and Wells (Driscoll, 1986), filter pack is installed to:

  • Retain most of the coarser-grained formational material behind the well screen
  • Reduce the amount of fine-to-medium grained sand entering the well (sand pumping)
  • Improve well hydraulics
Figure 1 – Damaged pump impellor due to pumping sand. Do you seal off the interval where sand is entering the well? Is the severity of sand pumping so great that the well has to be replaced?

Filter pack material should be well-rounded, well-sorted to provide good porosity and high hydraulic conductivity (improved flow to the well) near the well screen (Driscoll, 1986).  Traditional filter pack materials tend to be quartz and feldspar grains that are rounded by both wind and water weathering processes.  Grains should not be composed of rock fragments and/or calcium carbonate.  From time to time, we hear about wells that pump large volumes of sand, destroying pump equipment such as bowls and impellors (Figure 1).

Generally, the pumping of sand is due to one or more factors: poor installation of filter pack, poor well screen design, or poor selection of filter pack media. For this blog post, let’s assume that the well screen was properly designed and the filter pack was installed appropriately.  The big question is:

Is there a filter pack that could yield better hydraulics and reduce the potential of pumping sand particulates into the well?

Filter pack selection is particularly critical for Aquifer Storage and Recovery (ASR) wells since they tend to clog more frequently, resulting in declining injection and production rates over time.  To answer the question above, a series of tests were conducted that measured the degree of roundness (sphericity), sorting, and collapse strength of glass beads (2.4-2.9mm) and silica sand (6 x 9 mesh) (Figures 2 and 3).

Figure 2: Physical properties (sphericity, sorting, & packing) comparison between glass beads and silica sand.
Figure 3: Collapse strength analysis of glass beads versus silica sand. 21 samples of each media were crushed and measured in Newtons. Glass bead filter pack media has a collapse strength 7.4 times greater than silica sand filter pack media.

The collapse strength of the glass beads is 7.4 times greater than naturally occurring silica sand, which means glass beads can withstand more abrasion and maintain consistent pore space diameters over time. We also noted that the degree of roundness and sorting was better in the glass beads versus silica sand (Figure 2). The greater collapse strength and roundness of glass beads coupled with their higher degree of sorting results in a greater abundance of pore spaces (porosity) (Beard and Weyl, 1973 & Nagtegaal, 1978). Next, we explore whether these initial findings are valid and measurable.

Testing Hydraulics of Glass Beads versus Silica Sand

Figure 4: ASR well with glass beads and silica sand filter pack media. Tested intervals are 639-660 feet for glass beads and 685-706 feet for silica sand. Access tube terminates at 619 feet and the pump Intake is at 614 feet.

To test the difference in performance between these filter pack media, an ASR well was constructed with both glass beads (2.4-2.9mm) and naturally-occurring silica sand (6 x 9 mesh) (Figure 4). Twenty-foot thick intervals of glass beads and silica sand were selected based on similar hydraulic conductivities of the aquifer (36 feet/day in the glass beads and 28 feet/day in the silica sand).  Hydraulic conductivity of the aquifer was determined by conducting multiple slug injection tests.  The goal of the evaluation between the two media was to assess whether glass beads have better hydraulic properties than naturally-occurring silica sand.  Spinner log analyses were conducted over a one-year period of recharge and recovery operations.  The spinner logging tool measures the vertical productivity in a well and also measures variability in injection rates during recharge.  In other words, how much water is flowing out of and into the glass beads and silica sand?  Figures 5 and 6 represent the results in specific capacity (gpm/feet) during recovery and recharge operations.


Figure 5: Recharge specific capacity of glass beads versus silica sand filter pack media.
Figure 6: Recovery specific capacity of glass beads versus silica sand filter pack media.










The following conclusions were derived from the year-long ASR well performance test:

1. During recharge, the specific capacity of the glass beads was 3 to 9 times higher than the silica sand. This means that the injected water prefers to flow through the glass beads than the silica sand, which results in a higher volume of resources flowing through the glass beads (Figure 5).

2. The performance of the silica sand slowly degraded from 5.9 to 2.3 gpm/feet, which suggests pore spaces are either being clogged and or reducing in diameter due to abrasion. The glass beads retain their initial specific capacity of 10 gpm/feet or better after clogging events (Figure 5).

3. During recovery, the specific capacity of the glass beads was about 1.8 to 2.4 gpm/feet higher than silica sand (Figure 6). This means that water was being pumped out of the aquifer at an increased rate through the glass beads. Higher specific capacity results in shallower pumping water levels, which extends the life of the pump equipment and lowers pumping and maintenance costs over time.

4. For both recharge and recovery operations, the glass beads showed no decline in performance after backwashing events (unclogging events) over the testing period, indicating longevity and consistent well performance.

5. By evaluating and scaling up these findings, we developed a recharge operational cycle model (duration of injection cycles and frequency of backwashing) that compared the performance between an ASR well with glass beads to a silica sand ASR well. The glass beads were estimated to have a hydraulic efficiency of 45-55% greater than silica sand.

For those who manage well-fields and well recharge systems, these test results show tangible benefits in the form of both operational and maintenance costs and capital costs. If you would like a better understanding of how the technology advancements discussed in this blog series would apply to you, please contact me via email at

Additional glass bead performance research is being conducted that I hope to share later this year, so stay tuned for a subsequent blog on ASR well advancements.
Beard, D. C., and Weyl, P.K., 1973, Influence of texture on Porosity and Permeability of Unconsolidated Sand, The American Association of Petroleum Geologist Bulletin, Volume 57, No. 2, pp. 349-369.

Driscoll, F., 1986, Groundwater and Wells, 2nd Edition, by Johnson Screen, St. Paul, Minnesota, p. 1089.

Nagtegaal, P. J., 1978, Sandstone Framework Instability as a Function of Burial Diagenesis, Geological Society of London, Volume 135, pp. 101-105.

« Back to the Blog