Tiered Leak Detection Methods Using Bubble Ammonia and Halogen Techniques

Literature Overview and Technical Context

Leak detection is a critical quality gate in the fabrication of clad and bimetal pressure vessels, heat exchangers, and piping systems where even microliter-scale leakage can lead to catastrophic environmental or safety consequences. The literature entry under review describes a graduated sensitivity system that combines three conventional methods—bubble testing for coarse screening, ammonia detection for intermediate sensitivity, and halogen leak detection for high-sensitivity verification—with helium mass spectrometry as the ultimate reference standard. This tiered approach is not merely a collection of techniques but a structured engineering philosophy that matches detection sensitivity to the hazard classification of the service medium, thereby optimizing inspection cost and schedule without compromising safety.

In practice, I have observed that many fabrication shops default to helium mass spectrometry for every leak check regardless of the application, which is both expensive and time-consuming. The tiered system described here advocates a more rational approach: use bubble testing to eliminate gross leaks, ammonia detection to identify moderate leakage paths, and halogen detection for fine micro-cracks and interfacial defects in clad layers. This graduated methodology aligns well with the risk-based inspection philosophy embedded in standards such as GB/T 150, ASME VIII Div.1, and API 570.

Technical Principles and Sensitivity Comparison

Each method in the tiered system operates on a distinct physical principle, which determines its detection limit, applicable surface conditions, and operational constraints. Understanding these differences is essential for selecting the appropriate method at each stage of fabrication.

Method	Detection Principle	Typical Sensitivity	Applicable Surface	Key Advantage	Key Limitation
Bubble Test	Soap solution film rupture at leak point	~10^-2 to 10^-1 Pa·L/s	Smooth, clean surfaces	Simple, low cost, rapid	Cannot detect micro-cracks or hidden leaks
Ammonia Detection	Ammonia diffusion through leak, color change on indicator paper	~10^-3 to 10^-2 Pa·L/s	Smooth, accessible surfaces	Moderate sensitivity, portable	Ammonia toxicity requires ventilation; false positives from residual moisture
Halogen Leak Detection	Halogen gas diffusion through leak, haloscope indicator color change	~10^-4 to 10^-3 Pa·L/s	Smooth, clean surfaces	High sensitivity, non-toxic	Requires halogen gas supply; surface must be free of contaminants
Helium Mass Spectrometry	Helium ionization and mass filtering	~10^-6 to 10^-8 Pa·L/s	Any surface condition	Highest sensitivity, quantitative	Expensive equipment, requires vacuum system

The bubble test, while the simplest method, remains indispensable as a first-pass screening tool. A soap solution applied to the joint or weld surface creates a thin film that ruptures at any leak point, producing visible bubbles. This method is particularly effective for detecting gross leaks from incomplete welds, porosity clusters, or manufacturing defects exceeding approximately 0.1 mm in effective leak area. However, its sensitivity drops sharply for defects smaller than this threshold, making it unsuitable for detecting interfacial debonding in clad layers or micro-cracks in weld overlays.

Ammonia detection occupies the middle tier of the sensitivity spectrum. The technique involves pressurizing the test component with an ammonia-containing gas mixture, typically 5% ammonia in nitrogen, and applying an indicator paper strip impregnated with boric acid and a pH-sensitive dye across the suspected leak area. When ammonia diffuses through a leak, it reacts with the boric acid to produce a visible color change, typically from yellow to purple. The sensitivity of this method is approximately one order of magnitude better than bubble testing, making it suitable for detecting fine porosity chains and small interfacial cracks in clad welds. However, the method has practical limitations: ammonia is a hazardous substance requiring proper ventilation and personal protective equipment, and the indicator paper can produce false positives if residual moisture or acidic contaminants are present on the test surface.

Halogen leak detection represents the high-sensitivity tier among the three conventional methods. The technique uses a halogen gas, typically a mixture of fluorine and hydrogen or a specialized halogen gas blend, as the test medium. When halogen gas escapes through a leak, it interacts with a haloscope indicator—a specialized glass tube containing a halogen-sensitive chemical—that produces a color change proportional to the gas concentration. The sensitivity of halogen detection is approximately 10^-4 to 10^-3 Pa·L/s, which is sufficient to detect interfacial defects in weld overlay layers and fine micro-cracks that would be invisible to bubble or ammonia methods. The primary advantage of halogen detection is that halogen gases are non-toxic, eliminating the safety concerns associated with ammonia. However, the method requires a halogen gas supply and the haloscope indicator must be properly calibrated and maintained.

Sensitivity Grading and Medium Hazard Classification

The core engineering philosophy of the tiered system is the matching of detection sensitivity to the hazard classification of the process medium. This is not an arbitrary classification but a risk-based approach that considers the toxicity, flammability, and environmental impact of the service fluid.

Medium Hazard Level	Example Media	Required Detection Sensitivity	Recommended Method	Rationale
Low	Water, air, inert nitrogen	~10^-2 Pa·L/s	Bubble test	Gross leak detection sufficient; low consequence of minor leakage
Medium	Hydrogen, ammonia, light hydrocarbons	~10^-3 to 10^-2 Pa·L/s	Ammonia detection	Moderate sensitivity needed; flammability and toxicity concerns
High	Chlorine, sulfuric acid, toxic chemicals	~10^-4 to 10^-3 Pa·L/s	Halogen detection	High sensitivity required; severe environmental and health hazards
Critical	Radioactive materials, highly toxic agents	~10^-6 to 10^-8 Pa·L/s	Helium mass spectrometry	Maximum sensitivity required; zero-tolerance for any leakage

This grading system is consistent with the classification approaches found in GB/T 150.1-2011, ASME BPV Code Section VIII, and API 570. For example, a hydrogenation reactor containing hydrogen gas at elevated pressure and temperature would be classified as a medium-to-high hazard application, warranting at least ammonia detection and preferably halogen detection for final verification. Conversely, a water-cooled heat exchanger shell side carrying cooling water would only require bubble testing for routine leak checks.

Engineering Practice and Defect Analysis

In my experience with clad pressure vessel fabrication, the tiered leak detection system is most valuable during the final quality assurance phase after all welding, forming, and post-weld heat treatment operations are complete. The following workflow illustrates the practical application of this system in a typical clad vessel fabrication sequence:

After mechanical forming of the clad shell, perform bubble testing on all longitudinal and circumferential welds to detect gross defects from the forming process.
After post-weld heat treatment (PWHT), repeat bubble testing to identify any defects that may have been introduced or revealed by the thermal cycle.
For medium-hazard service vessels, perform ammonia detection on all welds and the clad interface area to identify moderate leaks that bubble testing would miss.
For high-hazard service vessels, perform halogen detection on all critical welds and the clad interface to ensure detection of fine interfacial defects.
For critical service applications, perform helium mass spectrometry as the final verification step.

A common defect encountered in clad vessels is interfacial debonding between the clad layer and the base metal, which can occur due to inadequate weld heat input, hydrogen-induced cracking, or thermal mismatch during cooling. These defects are typically in the range of 0.05 to 0.5 mm in effective leak area and are invisible to bubble testing but detectable by ammonia or halogen methods. In one documented case, a 316L stainless steel clad pressure vessel fabricated for a sulfuric acid service showed no leaks during bubble testing but revealed multiple interfacial debonding sites during ammonia detection. Subsequent radiographic testing confirmed that the debonding was caused by incomplete fusion at the clad-base metal interface, which was traced back to insufficient preheat temperature during the overlay welding operation.

Another important consideration is the effect of surface preparation on leak detection accuracy. For bubble testing, the surface must be clean and free of oil, grease, and paint, as these contaminants can suppress bubble formation and mask leaks. For ammonia and halogen detection, the surface must be smooth and free of roughness that could cause false color changes on the indicator. Surface roughness exceeding Ra 3.2 μm can significantly reduce the reliability of ammonia and halogen detection, requiring surface polishing or grinding prior to testing.

Key Questions and Reflections

The tiered leak detection system raises several important engineering questions that deserve careful consideration. First, what is the appropriate transition threshold between sensitivity levels? The literature suggests a general framework, but in practice, the transition should be determined by the specific hazard classification of the service medium, the design pressure and temperature, and the applicable code requirements. Second, how should the results of multiple detection methods be reconciled? If bubble testing shows no leaks but ammonia detection reveals moderate leaks, the vessel should be considered defective and subject to repair before retesting. Third, what is the role of quantitative leak rate measurement in this tiered system? While bubble, ammonia, and halogen methods are primarily qualitative or semi-quantitative, helium mass spectrometry provides quantitative leak rate data that can be compared against code-specified acceptance criteria.

Reflecting on my own engineering practice, I have found that the tiered system is most effective when integrated into a comprehensive quality plan that defines the detection method, sensitivity requirement, and acceptance criteria for each fabrication stage. The system also serves as a useful tool for root cause analysis: if a defect is detected at a higher sensitivity level, it provides valuable information about the defect size and location that can guide the investigation into the manufacturing process failure.

Summary

The tiered leak detection system comprising bubble testing, ammonia detection, and halogen detection represents a rational and cost-effective approach to leak verification in clad and bimetal pressure vessel fabrication. By matching detection sensitivity to the hazard classification of the service medium, this system ensures adequate quality assurance without unnecessary expenditure of time and resources. The key to successful implementation lies in a thorough understanding of the principles, limitations, and practical constraints of each method, as well as a disciplined approach to surface preparation and result interpretation. Engineers should view leak detection not as an isolated inspection step but as an integral part of the overall quality assurance framework, where each detection method provides specific information that contributes to the confidence in the integrity of the fabricated component.