Non-Destructive UV Inspection of Coatings Based on Enhanced Fluorescent Response

Stephen E. Cantor

The use of fluorescent materials to provide a non-destructive method for on-line inspection has been an area of increased importance to the coatings industry. Significant improvements in productivity have been realized by employing ultraviolet, UV, light curing to produce conformal coatings for computer boards.

This UV curing technique is also applicable to clear coatings applied to the surface of metals. Enhanced fluorescent response is a method to determine thickness of the coating, as well as a means of discovering voids that would be potential sites for corrosion.

However, the fluorescing agents used for inspection of the final assembly absorb UV light in the same region of the spectrum that is required to cure the coating. This filtering, or blocking phenomenon, limits the quantity of the fluorescing agent, which can be incorporated in the coating formulation. When the levels of the fluorescent materials are increased in an attempt to improve the brightness of the coating (as measured by exposure to long wavelength UV ("black light "), the coating may not cure properly. Therefore, coatings as thin as 1-3 mils may display a soft undercoat of wet, uncured material. This problem has impeded the acceptance of UV technology for systems that demand a bright fluorescent response for their inspection.

Conformal Coatings

Conformal coatings are designed to protect delicate electronic circuits against environmental elements such as dust, humidity, and other airborne contaminants. Solvent-free UV curing coatings are ideally suited to provide optimum circuit assembly prorect ion by increasing processing speed, decreasing processing costs, and assuring compliance with federal standards for air quality and worker safety. Because they are free of solvents, UV-cured coatings do not contain ozone-depleting chemicals.1–3


In the last decade, the incorporation of a fluorescent compound to provide a non-dest ructive method for this on-line inspection has been an area of much inrerest.4–6 However, unlike the solvent-based lacquers—which require no cure mechanism to form a film—the UV-based systems require that the correct wavelength of light strike the photoinitiator(s) to produce the free-radical catalysts needed to polymerize the ingredients and create the desired polymeric film.

Unfortunately, many fluorescing agents absorb radiation in the same region of the spectrum as the photoinitiators. As previously noted, this filtering or blocking phenomenon limits the amount ofthe fluorescing agent that can be incorporated into the coating (Fig. 1). Furthermore, high-intensity UV exposure can degrade these fluorescent dyes and actually fade them so that visual inspection is difficult. This limitation has impeded the acceptance of UV technology for systems that demand bright fluorescent response for their inspection.

Generally, levels of 0.02-0.04% of the fluorescing agent can be incorporated without significant detrimental effect on cure depth. However, when the levels of the fluorescent materials are increased in an attempt to improve the brighrness of the coating, as measured by exposure to black light, the coating will not cure properly.

Photoinitiators, which respond to the visible part of the spectrum (i.e., red shift), are one method to obviate the filtering effect of the fluorescing agent, but, in general, they impart a red or dark yellow color to the resulting coating. The preferred technique, described in this article, is to use photoinitiators that absorb in the visible part of the spectrum and, thus, are not affected by increased quantities of the fluorescent agent. At the same time, they provide a method to allow complete depth of cure without imparting a dark color to the cured coating. The acyl phosphine oxide, "photoiniriator #2," is a commercially available product (Ciba Corp).

The rapidly increasing industrial use of UV coating has created requirements for on-line inspection, non-destructive measurement, as well as off-line monitoring of cure and cure depth. Significant emphasis has been directed at developing scanners that can examine documents or packages with UV-cured inks or coatings containing fluorescent compounds. These automatic systems are important for prevention of counterfeit currency, as well as barcode identification, testing for faulty welds, control of overcoat on clear labels, and leak detection.

UV-cured coatings that are produced with enhanced levels of fluorescing agents improve the sensitivity of electro-optical scanners. Therefore, electro-optical devices can analyze more parts, at higher speeds, and with increased accuracy because the enriched fluorescence is more readily detectable.

Another attribute of this enhanced fluorescence is the use of less materials. It would also follow that the thickness of the coating, containing higher levels of the fluorescing agent, can be reduced while still allowing the electro-optical device to respond. The use of thinner coatings will permit excess heat generated by surface mounts or integrated circuits to be dissipated more readily and with lower coating costs.

Experimental Conditions and Results

The functional coating composition comprises a polyurethane-acrylated oligomer, acrylate monomer(s), flow agents, photoinitiator(s), and leveling agents. A commercially available substituted oxazole fluorescing agent was selected for this study. This substance fluoresces or phosphoresces above 350 nm, particularly in the visible range (400–700 nm), when stimulated by exposure to black light.

A number of commercially available standard photoinitiators were selected to demonstrate the marked difference between them and the "phoroinitiator #2" in the formulation. Experiments were conducted in which the quantity of the fluorescent agent was significantly increased. These standard photoinitiators are referred to as "photoinitiator # 1."

Example 1

Table 1 describes three experiments in which the level of the fluorescing agent (oxazole) varied, while the level of the special photoinitiator #2 remained constant. The effect on cure depth was measured by pouring the liquid into a soft plastic bulb. The bulb was then placed into a cavity in an opaque block (Fig. 2). This method ensures that the UV light penetrates only from the top. After exposure to UV light (Fusion Lamp at 7.8 J/cm²), the cured plugs were allowed to cool. The solid region was measured in millimeters.

Formulation A
Ingredients Parts/100
Polyurethane-acrylate oligomer 48
High boiling acrylates 48
Photoinitiator #1 2


The results demonstrate that only at a minimum concentration of 0.02 parts of the fluorescent agent can a depth of cure of 7 mm be obtained. The resulting plugs displayed a "faded and dull" response upon exposure to UV black light. Note, in experiments 2 and 3, the level of the fluorescing agent was increased, resulting in a significant decrease in cure depth until the special photoinitiator #2 was added at a 1% level (see: I-A, 2-A, 3-A,Table 1).

Table 1
Formulation A (%) 1 1-A 2 2-A 3 3-A
98 98 98 98 98 98
Fluorescent agent (%) 0.02 0.02 0.10 0.10 0.45 0.45
Photoinitiator #2 (%) 1 1 1
Depth of cure (mm) 7 35 1* 21 1* 15
Response to UV "black light" Faded & dull Faded & dull Very bright Very bright Very bright Very bright
*Only surface cure. Wet and soft below the surface


Not only was the depth of cure excellent (15–35 mm) , but a very bright response was also observed. Incorporation of an additional five parts of photoinitiator #1 into the solution failed to provide a cure depth in excess of 2 mm. Thus, the extra amount of the standard photoinitiator did not improve the results. Only those formulations in which the special photoinitiator #2 was present displayed the enhanced fluorescent response while maintaining an excellent cure depth (see Fig. 3).

Example 2

Printed circuit boards coated with the formulations described in the previous examples were exposed to two passes under a Fusion UV lamp. The effect of the high-intensity exposure was quite evident. The boards that had been coated with the solution containing only 0.02% of the fluorescent agent were not very responsive to the UV black light analysis. In contrast, the solutions that contained high levels of the fluorescing agent, such as 2A and 3A, were readily observed under the UV black light. Although the fluorescing agent incorporated was reduced in its effectiveness by the destructive effect of the high-intensity exposure, they are still able to provide excellent response to the UV black light due to their initial higher concentration level.

1. Improved visual observation. UV curing conformal coatings are commercially available and contain limited levels of fluorescing agents. This restriction has markedly reduced their acceptance by the electronics industry because of the difficulty this creates for the inspection process. For example, parts must be removed from the line and one-by-one placed in a darkened area so that the weak fluorescence stimulated by exposure to the black light can be observed. Printed circuit boards coated with the formulations described in this article are so responsive to the UV black light that they can be allowed to remain on the line, in a well-lighted work area, and still provide excellent observation of the coated regions.

2. Non-contact measurement of coating thickness. Because the coating thickness on a printed circuit board
is quite thin in relation to the board itself, it is most difficult to measure it accurately by a mechanical
method.7,8 Therefore, the use of a non-contact method would be the route of choice. This is readily accomplished by utilizing the fluorescence associated with the coating. Thus, an optical scanner "tuned" to respond to this phenomenon can provide an extremely accurate measurement of the film thickness. A schematic representation of the instrument and its operation is shown in Figure 4.

The quantity of fluoresced light generated by the coating is proportional to three factors:

  1. Incident UV intensity
  2. UV spot size
  3. Coating thickness

UV intensity and spot area are effectively set so that the coating thickness can be determined by measuring the amount of light fluoresced by the coating. The relationship is linear for a thin film. The output is read in an analog manner and the data are expressed in millivolts. A computer can be attached to the system via an RS-232 port so that the data can be recorded and, with the ability to convert selected data into a bar chart. Differences in bar height should be proportional to the thickness of the coating.

The specimens were examined by using a commercially available scanner provided by Angstrom Technologies, Inc. (Erlanger, Ky.). They offer a series of electro-optical scanners used for labels, sorting, safety sealing, and adhesive tracing. Several sets of FR-4 boards were carefully coated using very accurate drawdown bars to produce 1-, 4-, and l0-mil coatings. Uncoated samples were also tested to provide a control and to represent regions of void formation. The scanner can be set to read a certain optimum level of millivolts and respond if that level is not reached (potential void), or to respond if the value is too high, which indicates that the coating is too thick. Thus, a go/no-go process is possible for each coated part.

The results can be expressed as a bar chart, which is shown in Figure 5. Note that the uncoated board does have a 35-mV reading, which can be considered as the background count. The values obtained for the three coating thicknesses are not proportional, but they are different enough to readily set the value required to control the production process.

Figure 6 is a schematic representation of the conformal coating applied to a printed circuit board with a number of surface mounts. A void region is shown that will produce a low reading signal from the scanner. If this is a reproducible defect, then the spray or dipping application method can be adjusted to remedy the problem.


The use of phosphine oxide photoinitiators enables the curing of acrylate formulations containing high levels of fluorescent agents, thereby enhancing the evaluation of the cured coating either by visual inspection (exposure to UV black light) or by on-line electro-optical devices for quality control. This method provides a measurement of coating thickness as well as void identification. This analysis is conducted without destruction of the coating.


  1. Cantor, S.E. Two part, room temperature, shadow curing UV conformal coatings. Radtech Proceedings, p 498, Nashville, Tenn., April (1996).
  2. Bachmann, A.G. Solvent-free, 100% solids aerobic acrylic adhesives and coatings hold nurnerous manufacturing benefits. Nepcon West '92, Conf. Papers, p 931 (1992).
  3. Woods, J.G. Radiation-curable adhesives. In: Radiation Curing: Science & Technology. Pappas SP, editor. New York, NY: Plenum Press (1992).
  4. 4.Marinello, D.A., et. al. U.S. Patent 5,418,855 (1995) to Angstrom Technologies, Inc.
  5. 5. Melancon, K.C., Tiers GVD. U.S. Patent 5,310,604 (1994) to Minnesota Mining and Manufacturing Company.
  6. 6. Neckers, D.C., Song, J.C. ACS Polymer Material Science. Vol. 71, p 69 (1994).
  7. 7. Chemistry and Technology of UV and EB Formulation for Coatings, Inks and Paints. Vol. 1–5. OIdring PKT, editor. London: SITA Technology Ltd. (1991): p 218.
  8. 8. May, J.T. Non-contact measurement of conformal coating thickness. Nepcon West '89, p 1721 (1989).


Dr. Stephen Cantor has published numerous papers and made many technical presentations. His credentials include: research chemist, Uniroyal, 1965–82; research scientist, American Cyanamid 1982–6; marketing director, Pfaltz & Bauer Div. of Aceto Corp., 1986–90; lab manager, Dymax Corp., 1990–2002; and consultant for Dymax, 2002–8. Dr. Cantor, who has received 20 U. S. Patents, earned his BS from Queens College in New York, his PhD from the Univ. of Rochester, and post-doc from the Univ. of Arizona.

Share this article

More services


This article is featured in:
Electronics  •  Industrial and Protective Coatings  •  Testing & Control


Comment on this article

You must be registered and logged in to leave a comment about this article.