Improved Detection Technique for Solvent Rinse Cleanliness Verification
The NASA White Sands Test Facility (WSTF) has an ongoing effort to reduce or eliminate usage of cleaning solvents such as CFC-113 and its replacements. These solvents are used in the final clean and cleanliness verification processes for flight and ground support hardware, especially for oxygen systems where organic contaminants can pose an ignition hazard. For the final cleanliness verification in the standard process, the equivalent of one square foot of surface area of parts is rinsed with the solvent, and the final 100 mL of the rinse is captured. The amount of nonvolatile residue (NVR) in the solvent is determined by weight after the evaporation of the solvent. An improved process of sampling this rinse, developed at WSTF. requires evaporation of less than 2 mL of the solvent to make the cleanliness verification. Small amounts of the solvent are evaporated in a clean stainless steel cup. and the cleanliness of the stainless steel cup is measured using a commercially available surface quality monitor.
The effectiveness of this new cleanliness verification technique was compared to the accepted NVR sampling procedures. Testing with known contaminants in solution, such as hydraulic fluid, fluorinated lubricants, and cutting and lubricating oils, was performed to establish a correlation between amount in solution and the process response. This report presents the approach and results and discusses the issues in establishing the surface quality monitor-based cleanliness verification.
Background
Cleaning operations at WSTF encompass support for materials, components, propulsion test systems, and a sizeable shuttle depot activity. As part of these operations, components and piece parts are cleaned for a variety of services including oxygen, hydrogen, hydrazines, and nitrogen tetroxide. Most of these parts are constructed of stainless steel and are resistant to corrosion in aqueous media. For precision cleaned parts, the standard cleaning process consists of two major elements: an aqueous precleaning process and a solvent-based final cleaning and cleanliness verification. Cleanlinesss levels are specified by a number that has an associated number of particles at this maximum size in micrometers followed by a letter to specify the maximum residue allowed on 1 ft` of surface area. The letters A. B. C. and D refer to 1, 2, 3. and 4 mg of residue, respectively. Cleanliness levels to 50A are required for high-pressure oxygen service. Following the aqueous precleaning. the parts are dried, visually inspected, and if passed, are transferred to the Class 100 clean room. The accepted process for solvent-based final cleaning are rinsed with HFE 7100 in the vapor degreaser followed immediately by a sampling rinse of approximately 100 mL/ft` of component surface area. This rinse is then filtered for a particulate count and then submitted for NVR determination. The components are then blown dry with nitrogen, packaged, and labeled for use.
Figure 2 shows the steps in obtaining the NVR from the solvent rinse. First the solvent is transferred to a boiling flask, and the volume of the solvent is reduced to 10 to 15 mL by distillation. The solvent trapped in the distillation by a condenser is recycled the remaining 10 to 15 mL is transferred to a tared weighing pan and placed in a 100° C oven for 30 min. to complete the drying. The weighing pan is allowed to cool in a desiccator before weighing again to determine the NVR. For HFE 7100. the process takes approximately 40 min. It should also be noted that some of the more volatile contaminants could be volatilized, especially during the 30 min drying in the oven. Light hydrocarbons can show 40% per NVR recovery or less from spiked solvent samples.