Membranes possible key to recycling frac water

Gilmore studying membrane separation technology in Marcellus Shale drilling

BY ERIC LIDJI FOR GREENING OF OIL

One of the biggest environmental challenges in drilling for natural gas in eastern North America’s Marcellus Shale is the tremendous demand on local water supplies needed for hydraulic fracturing, a technique that releases gas from shale rock deep in the ground. Another challenge is handling the briny wastewater that comes out of wells during the drilling process.

(Start the conversation. See comments section at bottom of page.)

An obvious solution to both would be some way to re-use water, which is why a lot of research dollars are going toward studying technologies to treat and re-use flowback water at Marcellus sites.

Kevin Gilmore, a visiting assistant professor at the Department of Civil & Environmental Engineering at Bucknell University in Lewisburg, Pa., is studying the use of membrane separation technology in Marcellus development. Gilmore presented some of his findings at “The Science of Marcellus Shale” on Jan. 29 at Lycoming College in Williamsport, Pa.

After the summit, he spoke to Greening of Oil about his research.

About 13% comes back up

The most common way to develop natural gas trapped in shale formations is hydraulic fracturing, where drillers pump large amounts of sand and chemical infused water underground to crack and hold open rocks, allowing natural gas to come to the surface.

While most of that “slick” water stays underground, about 13 percent on average returns to the surface as “flowback water,” per the Susquehanna River Basin Commission, an interstate regulatory agency.

That flowback water is infused not only with the original sand and chemicals, but also with salty minerals like sodium, calcium and chloride picked up underground.

Most municipal wastewater treatment plants in Pennsylvania can’t handle water this salty, and even if they could, there isn’t enough regional treatment capacity to handle the tremendous amount of flowback produced by wells across the western half of the state, especially with the number of well permits tripling in the past year.

One technology being studied to manage this flowback water at well sites is membrane separation. Membranes are currently used to make water drinkable and to process municipal waste, and some think the technology may have some usefulness in shale gas development, where it is already being used commercially in limited amounts.

Through a small grant from the Pennsylvania Water Resources Research Center, an affiliate of the U.S. Geological Survey, Gilmore is looking at whether membrane separation technology can be effective in treating flowback water for discharge or re-use.

Membranes differ in filtration quality

Membranes differ in filtering capabilities, with pores measured in microns (a millionth of a meter) and nanometers (a billionth of a meter).

Gilmore described microfiltration as being “almost like an extremely fine cloth filter.”

Ultrafiltration is a step finer.

Nanofiltration and reverse osmosis are non-porous, meaning they don’t even have holes like a typical filter. Instead, the water dissolves into the membrane and diffuses across it, leaving the unwanted substances behind.

The goal in shale development is to reduce the amount of total dissolved solids, or TDS, in the flowback water.

Typical Marcellus wells produce flowback water with TDS in the range of 100,000 milligrams, or 100 grams, per liter or higher. (By comparison, a liter of drinking-quality water is 1,000 milligrams.)

Pennsylvania is considering regulations that would prohibit discharging water with greater than 500 milligrams per liter.

Gilmore said that so far, his testing hasn’t gotten flowback water that clean. The best results filtered flowback samples with 30,000 mg/L of TDS, down to 1,400 mg/L.

However, Gilmore said, those results came from “single stage” testing, using one filter size. Passing the water through a series of finer and finer filters might further reduce the TDS.

But are they economic?

The problem is whether those finer filters could be used economically.

“If money’s not an issue, membranes can do a pretty good job,” Gilmore said.

Membranes can use a lot of energy, though, especially nanofiltration and reverse osmosis, which require high pressures to dissolve the liquids into the membrane.

“The pressure is so high and the recovery is so low that it’s probably not going to be economically feasible to use reverse osmosis membranes as a treatment for disposal,” Gilmore said, adding that in wells with lower concentrations of TDS, or in cases where flowback is diluted by another water source, the recovery rates may be more economic.

That energy demand is one of the biggest obstacles facing membranes, not only because it can increase development costs, but also because it can potentially increase the carbon footprint of developing natural gas from shale, which is a cleaner energy source than oil or coal.

Limited, but important application

While membrane technology may not currently be able to economically get flowback water to meet the discharge quality standards proposed in Pennsylvania, it still might become very useful for recycling water. If drilling companies can re-use even a portion of the briny water that comes up from wells, it reduces the demand on local water systems and local wastewater treatment plants.

If those companies can recycle water on site, it keeps water-hauling trucks off the road, some of which take wastewater to underground injection wells in other states.

Fewer trucks on the road burning gasoline and diesel results in a smaller environmental footprint for shale gas production.

The system Gilmore uses in his lab is small, but he said the technology could be scaled up to meet the field volumes. While liquids must hit the surface of the membrane to get filtered, membranes don’t need to be flat. Creative configurations like tubes and spirals can condense a membrane system without reducing its surface area, he said.

Because membrane systems are contained, they work in rainy Pennsylvania, where evaporation techniques being used in the dry Barnett Shale of Texas aren’t appropriate.

Cleaning the residual stream

Another challenge facing membranes is waste.

While filtration produces clean (or cleaner) water, it also leaves behind the undesirable TDS filtered out of the flowback, and that “residual stream” must be handled, too.

Gilmore said that since the residual stream is still a highly concentrated liquid, and not a crystallized solid, it could technically be pumped into underground injection wells. Right now Pennsylvania doesn’t have many deep disposal wells, but some regulatory agencies are studying whether it’s feasible or desirable to add more to accommodate shale drilling.

“That, I think, is one of the areas where more research is needed as well: the ultimate fate of that liquid residual stream,” Gilmore said.

Links of interest

The Susquehanna River Basin Commission

National Energy Technology Laboratory on membrane technology

Contact Eric Lidji at ericlidji@mac.com