Membrane integrity monitoring – Direct methods

Bubble point test (BPT)

The purpose of this test is to determine the largest pore (or defect) size on membrane surface through which the first bubble leaks by overcoming the surface tension. Leak point can be also located under certain circumstances, if it can be identifiable visually. The size of the largest pore (or defect) that can be estimated using the bubble point pressure using the Eq. (1) (Côté, 2002).

ddddddd           ————————–  (1)

BP  = bubble point of the defect ( Pa )
B = capillary constant ( 0-1)
d = defect diameter ( m )
γ = surface tension at the air-liquid interface ( 0.0728 N/m at 20 oC )
θ = contact angle as defined here (radian)

Bubble point (BP) can be calculated using the following calculator, where capillary constant was assumed 1 and contact angle π/8 (or 22.5o). As can be seen in the graph, ~3 micron defect can burst at around 100 kPa (or 1 bar), which means at least 100 kPa is required to detect 3 micron defect. (Note: There are two opposite definitions of contact angle depending on which side of the angle is measured in the figure shown here.)

Bubble point tests are performed by gradually pressurizing either membrane lumen or shell while monitoring bubbles in other side as illustrated in Fig. 1. This method is suitable for hollow fiber membranes in pressure vessel, but this method is also applicable for immersed hollow fiber membranes and flat sheet membranes with self-supporting structures, e.g. BIO-CEL®.

image004 (1)aasdfg

Fig. 1. Bubble point test to estimate the largest pore (or defect) size.

The leaking hollow fibers in pressure vessel can be located using BPT. Pressurized air is supplied to shell side of wet module while one of the two openings of fiber lumen is closed. The unclosed lumen is positioned upward and filled up with water. If any fiber leaks, bubbles come out from the leaking fiber’s bore in the top. After plugging the leaking fiber’s bore, same procedure is performed for the other side of the lumen after flipping the module upside down. Finally both sides of the leaking fibers are plugged and the module can be reused.

Pressure decay test (PDT)

Pressure decay tests (or pressure holding tests) are the most common test methodology used in field, which are performed in situ at below the bubble point by reversing the permeate flow with pressurized air. Once all the water filled in membrane lumen permeates back through the membrane, membrane lumen is filled up with pressurized air. Due to the capillary suction pressure illustrated in Fig. 2, air does not leak through the membrane as long as pressure is below the bubble point. After isolating the membrane from pressure source, lumen pressure is monitored for the predetermined time (5-20min typical). If there are defects on membrane surface, air can permeate through the defects and the pressure drops quickly. If there are no defects, only small amount of air will be lost by the diffusion through membrane pores. However, it is not perfectly clear where to set a line between normal pressure decay rate and abnormal pressure decay rate.

Pressure decay tests are particularly useful for the hollow fiber membranes with integrated skin with support layer. It can be also used for the hollow fiber membranes with non-integrated skin layer and flat sheet membranes, but maximum allowable air pressure may not be high enough to detect the breaches smaller than, for example, 3-5 micron.

image006zxcvbFig. 2. Capillary suction phenomenon in hydrophilic membrane. The water inside membrane pore resists against air pressure due to the capillary suction pressure.

The pressure decay rate (PDR) can be modeled as Eq. (2) assuming the air diffusion through wet pores follow Fick’s first law (Farahbakhsh, 2004)

image007hgfd     ————————- (2)

dP/dt = pressure decay rate (PDR, Pa/s  )

β      = ratio of saturation concentration of air in permeate to the saturation concentration of air in pure water ( – )
κ      = experimental membrane parameter (m-1)
D      = diffusion coefficient of air ( m2/s )
H      = Henry’s constant ( mol/m3/atm )
P1     = pressure in pressurized side ( atm )
P2     = pressure in unpressurized side ( atm )
R       = universal gas constant ( J/mol/K )
T       = absolute temperature ( K )
V       = air volume in pressurized spaces ( m3 )

In practical situation, experimentally measured pressure decay rate (PDR) is compared with the predetermined PDR of the intact new membranes with same membrane area at the same pressure. If the measured PDR is within the preset criteria, membranes are considered intact. One complexity is that the PDR is affected by temperature due to the diffusivity of air, which requires multiple standard PDRs for different temperatures.

Fig. 2 shows experimental PDR with one membrane module with 20,000 polysulfone fibers. PDR without broken fibers is 0.16 psi/min (or 1.1 kPa/min), but it increases as the number of breached fiber increases.


Fig. 2. Pilot scale pressure decay test result with 20,000 polysulfone hollow fiber membranes in a pressure vessel (114 m2) from Polymem (Hugaboom, 2004).

To use PDT, the exact test condition must be defined for a specific case considering the level of sensitivity required, membrane surface area, types of membrane, etc. This is because the natural PDR by air diffusion is case specific depending on test pressure, the ratio of membrane area to lumen volume, membrane pore size, membrane porosity, etc. More details are found in ASTM D6908 (2010).

Advantages of the PDT in immersed and pressurized membranes are: (EPA, 2005)

  • Ability to meet the resolution criterion of 3 µm under most conditions
  • Ability to detect integrity breaches on the order of single fiber breaks and small holes in the lumen wall of a hollow fiber, depending on test parameters and system-specific conditions
  • Standard feature of most MF and UF systems
  • High degree of automation
  • Widespread use by utilities and acceptance by States
  • Simultaneous use as a diagnostic test to isolate a compromised module in a membrane unit in some cases

Limitations of the pressure decay test are:

  • Inability to continuously monitor integrity
  • Calculation of test method sensitivity requires measurement of the volume of pressurized air in the system
  • Potential to yield false positive results, if the membrane is not fully wetted (which may occur with newly installed and hydrophobic membranes that are difficult to wet, or when the test is applied immediately after a backwash process that can fill pores with air)
  • Difficult to apply to membranes that are oriented horizontally as a result of potential draining and air venting problems

Diffusive air flow (DAF) test

DAF test is basically same as PDT, but DAF measures either air flow rate or displaced water flow instead of pressure decay rate. Air pressure can be maintained constant during the test, but it can be allowed to vary by closing air pressure valve, too. Since DAF is affected by temperature, multiple baselines are required for various temperatures to compare with the actually measured DAF.

If DAF test is compared to  the PDT, DAF test is more sensitive to membrane integrity breach than the PDT especially when the displaced water volume is measured instead of air flow rate (Trimboli et al., 2001)

 Vacuum decay test, VDT (EPA, 2005)

VDT is opposite of PDT. This test is applicable for pressurized hollow fiber modules and spiral wound modules. If this test is applied for immersed membranes, membrane modules must be taken out from mixed liquor to contact air with membrane surface. Procedures are as follow

  1. Drain the water from one side of the membrane. Typically, the filtrate side of a spiral-wound NF or RO membrane is drained. Shell side liquid is drained for hollow fiber modules.
  1. Apply a vacuum to the drained side of the membrane.A vacuum of 20-26 inch Hg (or 0.7-0.9 bar) is applied during the test. For the compliance with LT2ESWTR requirements, the applied vacuum must be sufficient to meet the resolution criterion of 3 µm. If membranes are exposed to the air during the test, care must be taken in order not to dry the membrane to prevent pores from drying out.
  1. Isolate the vacuum source and monitor the vacuum decay for a designated period of time.If there are no leaks in the membrane, then the vacuum should decay marginally over the duration of the test. Typically, this test is monitored over a period of 5 to 10 minutes. The rate of pressure decay should be compared to the pre-establish upper control limit, UCL (or any lower control limit, LCLs, that may also be established).

Advantage of vacuum decay test is the ability to test spiral-wound membranes or other systems that cannot be pressurized on the filtrate side of the membrane.

Disadvantages of the vacuum decay test are:

  • Inability to continuously monitor integrity
  • Not widely used for full-scale systems in current practice
  • Difficulty in removing entrained air after the test has been completed
  • Calculation of test method sensitivity requires measurement of the volume of air under vacuum in the system
  • Inability to meet the resolution criterion of 3 µm under most conditions primarily because the maximum deliverable vacuum pressure is limited at 0.7-0.9 bar in practical situation