-
- News
- Books
Featured Books
- pcb007 Magazine
Latest Issues
Current IssueThe Sustainability Issue
Sustainability is one of the most widely used terms in business today, especially for electronics and manufacturing but what does it mean to you? We explore the environmental, business, and economic impacts.
The Fabricator’s Guide to IPC APEX EXPO
This issue previews many of the important events taking place at this year's show and highlights some changes and opportunities. So, buckle up. We are counting down to IPC APEX EXPO 2024.
Getting to Know Your Designer
In this issue, we examine how fabs work with their design customers, educating them on the critical elements of fabrication needed to be successful, as well as the many tradeoffs involved. How well do you really know your customer? What makes for a closer, more synchronized working relationship?
- Articles
- Columns
Search Console
- Links
- Events
||| MENU - pcb007 Magazine
Comparison of halogen-free base materials (Oct 2000)
October 21, 2000 |Estimated reading time: 15 minutes
Comparison of halogen free base materials Abstract
Multek Europe and the Bromine Science and Environmental Forum (BSEF) have conducted tests comparing several halogen-free materials with established halogen-based materials Standard FR-4 and Standard FR-4-180. Two categories of tests were carried out: product technical performance and environmental impact.
The product technical performance tests carried out were: thermomechanical (solder shock testing, Cu peel strength, Tg- and z-axis-expansion measurements); electrical investigations (dielectric constant and dissipation factor measurements both in the low and high frequency range, surface resistance, volume resistivity) of the laminates; characterisation of the prepregs (flow measurements, rheology); and processing tests.
The environmental impact tests assessed the following criteria: potential consumer exposure (offgasing), smoke toxicity, and end-of-life issues (incineration characteristics, potential to leach from landfills).
This paper summarizes the initial findings from these tests.
Introduction
FR-4 type Epoxy laminates based on Tetrabromobisphenol-A (TBBPA) as flame retardant have been the industry standard for many years. Negative perceptions of brominated flame retardants (BFR) have recently led to the development of various halogen-free materials. At this stage, however, little is known about the true comparison of such materials with the industry standard products from the perspective of environmental, technical or economic performance.
This study is looking at the technical properties and the environmental impact of four new halogen free materials. Commercially available FR-4 and High-Tg-FR-4 material are used as a standard for comparison. The experiments are assessing key criteria with regard to the technical performance such as thermomechanical properties, electrical properties, prepreg characterization, and manufacturability. The environmental performance of the materials is evaluated by assessing possible consumer exposure (offgasing), smoke toxicity, and end-of-life issues (incineration characteristics, leaching from landfill). Results from the technical testing and preliminary results from the environmental testing are reported.
1. Technical Performance 1.1 Thermomechanical characterization of the investigated materials 1.1.1 Cu Peel Strength
The Cu peel strength is a measure of the adhesive strength of the Cu structures to the PCB outerlayers, which is important for the repair of assembled boards. Over that very low Cu peel strength values can be a hint at potential delamination problems of multilayer boards. To measure this value, laminates with 0.56 mm (22 mil) dielectric thickness and 1 oz / 1 oz Cu of all investigated materials were used. The Cu was structured using a conventional print and etch method to receive 4 mm wide Cu stripes. 8 of these stripes for each material were peeled from the surface measuring the necessary strength (Test method: IPC-TM-650; 2.4.8). A minimum of 1.0 N/mm shall be reached. The results are given in table 1.
Halogen Free Base Materials Cu Peel Strength after Conditioning Unit FR-4 FR-4-180 1 2 3 4 "as received" N/mm 1.72 - 1.97 1.21 - 1.57 1.19 - 1.34 1.43 - 1.66 1.18 - 1.43 0.92 - 1.12 2h / 125°C N/mm 1.76 - 2.00 1.21 - 1.58 1.19 - 1.38 1.47 - 1.68 1.14 - 1.47 0.92 - 1.11 Tg 1 (TMA) °C 133 160 135 133 141 143 Tg 2(TMA) °C 134 164 136 130 143 144 Tg 1 (DMA) °C 149 189 153 153 185 171 Tg 2(DMA) °C 145 191 151 155 188 177 Table 1: Cu Peel Strength Results
For halogen flame retarded base materials the Cu peel strength decreases with raising Tg. There are laminates, which show a strong decrease of Cu peel strength after thermal stress (conditioning) – for instance cyanate ester, but obviously not the halogenated epoxies. The effect of decreasing Cu peel strength with raising Tg can also be found with the halogen free base materials – material “3” and “4” with higher glass transition temperatures tends to have a reduced Cu adhesion. The material “4” showed Cu peel strength values even below the specified minimum. Compared to “2” the material “1” has a lower Cu peel strength with nearly similar glass transition temperature. For all the investigated materials there is obviously no influence of thermal stress on Cu peel strength.
1.1.2 Solder Shock Tests
The solder shock testing is a measure of the thermal resistance of the base material. The test samples again were made out of the laminate with 0.56 mm (22 mil) dielectric thickness and 1 oz / 1 oz Cu of all investigated materials. The sample size was 3.7 by 3.7” with the Cu etched off from one side to see any measling effects if there would occur. The test was done as a solder float laying the sample onto the solder bath (Test method: IPC-TM-650; 2.4.23) after drying them at 80°C for 10 hours. A minimum requirement of 1 x 288°C / 10 sec shall be passed with 4 samples of the material. Table 2 illustrates the test results.
Solder Shock Repetitions Tested samples FR-4 FR-4-180 Halogen Free Base Materials 1 2 3 4 288°C / 10 sec 1 4 4 passed 4 passed 4 passed 4 passed 4 passed 4 passed 288°C / 60 sec 1 4 4 passed 4 passed 4 passed 4 passed 4 passed 4 passed 288°C / 10 sec until failure 4 mean 425.00 15.75 18.00 7.00 32.50 8.75 minimum 24 14 15 2 27 7 Table 2 : Solder Shock Test Results
The minimum requirement (1 x 288°C, 10 sec) was met by all tested materials. With an extension of the testing time to 60 sec, 3 of the tested 4 samples of material “2” showed measling. The repeated solder shock showed even larger differences. With the FR-4-180 results as a minimum requirement, both the halogen free materials “2” and “4” failed. Only material “3” showed a thermal resistance better than FR-4.
1.1.3 Tg and z-axis-expansion
The glass transition temperature Tg has become a measure of how well a laminate resin system resists softening from heat /1/. When the Tg temperature is reached, the resin changes from its “glassy” state and causes changes in the laminate’s properties. To measure the Tg samples were cut out of the laminate with 0.56 mm (22 mil) dielectric thickness with the Cu etched off completely. The measurements were made according to the following specifications after drying the samples for 10 hours at 80°C: IPC-TM-650, Method 2.4.24C (TMA), IPC-TM-650, Method 2.4.25C (DSC), IPC-TM-650, Method 2.4.24.2 (DMA). Out of the TMA-curve not only the Tg can be determined, but also the CTE(z) below and above Tg for the investigated material. It is advantageous to have a material with a high glass transition temperature and with low CTE-values for an improved thruhole reliability of the PC boards (see figure 1).
Figure 1: Schematic TMA-plot of FR-4 and FR-4-180
Table 3 summarizes the Tg and CTE(z) measurement results.The halogen free materials “1” and “2” seems to be very similar to the halogenated FR-4 from a thermomechanical standpoint of view. Both the materials “3” and “4” show higher Tg’s, not reaching the values of FR-4-180, but significantly higher than Standard-FR4. Also the z-axis-expansion of those two materials is much lower than for FR-4-180 thus indicating an improved thruhole reliability (this would have to be checked by suitable reliability tests).
Halogen Free Base Materials Unit FR-4 FR-4-180 1 2 3 4 Tg 1(TMA) °C 133 160 135 133 141 143 Tg 2 (TMA) °C 134 164 136 130 143 144 Tg 1(DMA) °C 149 189 153 153 185 171 Tg 2 (DMA) °C 145 191 151 155 188 177 Tg 1 (DSC) °C 129 171 125 125 128 - Tg 2(DSC) °C 128 166 128 127 129 - CTE z (20°C - Tg) ppm/K 59 53 64 56 24 55 CTE z (Tg - 225°C) ppm/K 245 199 260 276 137 187 z-axis-expansion ppm 29 302 20 091 30 629 31 671 13 874 21 838 Table 3: Tg and CTE measurement results 1.2 Electrical characterization of the investigated materials 1.2.1 Dielectric Constant and Dissipation Factor
Dielectric constant is the property of a material that determines the relative speed that an electrical signal will travel in that material /1/. Signal speed is roughly inversely proportional to the square root of the dielectric constant. Performance applications that require high “clock rates” desire base materials with a Dk as low as possible and with as little change over temperature.The necessary samples for measuring both dielectric constant and dissipation factor were produced in a conventional print and etch process out of laminates of two different dielectric thicknesses, 0.56 mm (22 mil) and 0.1 mm (4 mil) each with 1 oz / 1 oz Cu. The exact value of the dielectric thickness which is required for the calculation of the Dk was determined by cross sections (5 measurements for each material). For the measurements according IPC-TM-650; method 2.5.5.2, an HP 4194A Impedance / Gain Phase Analyzer was employed. The results are given in table 4.
Halogen Free Base Materials FR-4 FR-4-180 1 2 3 4 Property Conditioning Sample Dielectric Constant (1Mhz) 24h/23°C/50% 3x7 628 4.57 5.00 4.42 4.57 4.99 4.69 Dielectric Constant (1Mhz) 24h/23°C/50% 1x2 116 4.63 4.65 4.49 4.41 5.35 4.61 Dielectric Constant (1Mhz) 96h/35°C/90% 3x7 628 4.61 5.02 4.41 4.56 5.00 4.70 Dielectric Constant (1Mhz) 96h/35°C/90% 1x2 116 4.64 4.65 4.50 4.42 5.35 4.61 Table 4: Dielectric Constant at 1 MHz, room environment
The dielectric constant of nearly all investigated HF materials was measured to be in the range of Standard-FR-4 and Standard-FR-4-180. Only the halogen free material “3” showed increased values for the dielectric constant.
The dissipation factor is a measure of the percentage of the total transmitted power that will be lost as electrons dissipate into the laminate material. Performance applications operating above 800 MHz require base materials with Df < 0.01. Table 5 illustrates the results.
Halogen Free Base Materials FR-4 FR-4-180 1 2 3 4 Property Conditioning Sample Dielectric Constant (1Mhz) 24h/23°C/50% 3x7 628 0.01577 0.01668 0.01113 0.01007 0.00412 0.00565 Dielectric Constant (1Mhz) 24h/23°C/50% 1x2 116 0.01777 0.01931 0.01344 0.01245 0.00414 0.00654 Dielectric Constant (1Mhz) 96h/35°C/90% 3x7 628 0.01642 0.01728 0.01153 0.01088 0.00499 0.00658 Dielectric Constant (1Mhz) 96h/35°C/90% 1x2 116 0.01829 0.01988 0.01385 0.01282 0.00489 0.00747 Table 5: Dissipation Factor at 1 MHz, room environment
From an “electrical loss” viewpoint all investigated HF materials show advantageous values in comparison to Standard-FR-4 and Standard-FR-4-180. For both the halogen free materials “3” and “4” the measured dissipation factor is relatively low, indicating that these materials could be used for higher frequency applications, too.
1.2.2 Surface Resistance and Volume Resistivity
The surface resistance characterizes the resistivity between copper traces of one layer along the surface of the base material. The resistivity between two Cu layers is defined as volume resistivity. Both values should be as high as possible to ensure a complete isolation of Cu structures from each other. The measurements according IPC-TM-650, method 2.5.17, were done using the HP 4329A Ohm-meter. Table 6 summarizes the results.All the surface resistance values were to high to be measured with the available equipment. The volume resistivity of the halogen free base materials is in all cases higher than determined for the comparative materials.
Halogen Free Base Materials FR-4 FR-4-180 1 2 3 4 Property Unit Sample Surface resistance TOhm 3x7 628 n.m. n.m. n.m. n.m. n.m. n.m. Surface resistance TOhm 1x2 116 n.m. n.m. n.m. n.m. n.m. n.m. Volume resistivity TOhm/cm 3x7 628 7 342.2 6 629.4 18 479.3 18 364.8 7 476.5 17 006.2 Volume resistivity TOhm/cm 1x2 116 10 511.7 8 550.2 13 901.3 14 660.9 12 258.5 17 442.3 Table 6: Surface Resistance and Volume Resistivity Results 1.3 Prepreg characterization 1.3.1 Flow
When heating a prepreg under pressure the resin liquifies and flows /1/. There is a finite period of time which the resin remains fluid enough to flow freely. With increasing cross-linking degree the resin gels and hardens. This is the basis for the multilayer lamination process. The “Flow” is the quantitative measure of how well the resin melts and flows out during the lamination process. For the test a temperature of 192°C and a pressure of 2,85 bar was applied for 5 minutes. For the measurements prepregs with 2116 glass style of each material were used. The results are given in table 7.
Halogen Free Base Materials FR-4 FR-4-180 1 2 3 4 Flow % 29.75 23.69 32.16 37.90 26.36 33.12 Table 7: Flow Measurement Results
For the halogen free base materials a trend toward a higher flow can be seen. This would help to ensure the complete filling of etched innerlayer Cu structures. On the other hand a high flow can cause unstable dielectric thicknesses. Since this value can be adjusted by the base material suppliers and kept within tight tolerances obviously no problem arises.
1.3.2 Rheological Investigations
Prepregs with 1080 glass of each material was investigated rheologically to evaluate the dependence of the dynamic resin viscosity on the temperature. Due to the progressive cross-linking of the epoxy the heating rate effects the viscosity curve. The investigations were made using temperature ramps of 2, 4 and 6 K/min, since this is normally the range of heating rates within a multilayer press. Figure 2 shows the viscosity curve in dependence on the temperature for a 2 K/min heating rate as an example. Table 8 summarizes the measured minimum viscosity values and the corresponding temperatures
Halogen Free Base Materials FR-4 FR-4-180 1 2 3 4 Temperature of minimum viscosity with 2K/min °C 128.6 115.4 118.3 119.3 119.7 114.8 Minimum viscosity with 2K/min Pa*s 16 800 1 995.0 5 582.7 3 216.0 19 130 26 210 Temperature of minimum viscosity with 4K/min °C 141.1 125.7 125.9 119.9 128.0 118.8 Minimum viscosity with 4K/min Pa*s 7 086 825.6 1922.0 2821.0 16 200 1 586.9 Temperature of minimum viscosity with 6K/min °C 153.9 126.8 132.9 124.0 144.5 111.8 Minimum viscosity with 6K/min Pa*s 3 696 678.8 1 105.2 4163.0 13 710 2111.1 Table 8: Results of the Rheological Investigation
Both the viscosity curves and the comparison of minimum viscosities and corresponding temperatures show that there is not a large difference in the rheology of the standard and the halogen free base materials. So it could be assumed that the halogen free materials are compatible to the FR-4 press cycles employed yet. Nevertheless during the qualification of any of the halogen free materials for a PCB production this issue would have to be checked.
Figure 2: Dynamic viscosity in dependence on temperature of investigated materials 1.4 Manufacturability tests
The halogen free base materials were checked with regards to their compatibility to a number of PCB process steps:
- compatibility to reduced blackoxide
- compatibility to drilling with standard FR-4 drilling parameters
- compatibility to desmear including determination of NMP-absorption
- compatibility to electroless copper process.
To test blackoxide compatibility, a number of structured laminates (innerlayers) of each material received conventional reduced blackoxide processing using NaOH / NaClO2 to oxidize the Cu, and DMAB to reduce the CuO to Cu2O. The innerlayers underwent a multilayer press cycle using a conventional FR-4 program. After etching off the outerlayer Cu, the multilayers were inspected visually regarding the blackoxide appearance. None of the investigated materials showed any problems.The compatibility to the other processes mentioned above was investigated by processing a number of multilayers up to and including electroless Cu. Cross sections of thruholes were taken to evaluate the drilling quality (performance measures: hole wall roughness, nailheading, burr, micro delaminations within the thruhole area), desmear quality (performance measure: smear residues on innerlayer pads or on base material) and electroless Cu quality (performance measure: complete coverage with e’less Cu, obvious hole wall separations). The results from the investigation conclude that drill bit age (hit count) is a more influential parameter than material for the drilling process for all material types. The employed permanganate desmear (Atotech chemistry) removed the smear residues completely in the case of the investigated hal.free materials. Also, no problems resulting from electroless Cu were apparent – the inspected thruholes (cross sections of about 45 thruholes with different via diameters of each material were made and evaluated) showed a complete coverage of the hole walls and no evidence of hole wall separations. To test the behaviour of the halogen free materials within the desmear process the NMP absorption was measured after a 1 hour dip of the samples into pure NMP with subsequent GC / MS. The NMP absorption of the halogen free materials was determined to be higher than for FR-4-180 but much lower than the value for FR-4 (see table 9). Additionally, to reveal any other possible differences between the materials, the weight loss of drilled material samples was measured after undergoing 4 desmear cycles. The results of these tests can be seen in table 9.
Halogen Free Base Materials FR-4 FR-4-180 1 2 3 4 Property Unit Test vehicle NMP absorption ppm 1.6 mm core, Cu etched off 2 200 <10 130 320 60 140 Weigth loss during desmear % 1.6 mm core, Cu etched off 0.34 0.33 0.44 0.58 0.88 0.99 Weigth loss during desmear % 1.6 mm core, Cu cladded 0.47 .39 0.45 0.37 0.44 0.45 Table 9: NMP absorption and weight loss during desmear
The results from the weight loss tests performed on Cu cladded drilled samples showed nearly similar behaviour of the halogenated and the halogen free base materials. For the samples without outerlayer Cu the weight loss of the halogen free base materials is higher after desmear than for conventional FR-4 or FR-4-180. These results indicate a stronger attack of the desmear chemistry on the hal.free materials. Nevertheless, this result does not show a weak point of the new base materials; the test conditions were more extreme than the worst case that can realistically be assumed to occur in practice.
2. Sample Characterisation
In a preliminary experiment the composition of the halogen free materials and of Standard FR-4 was analysed. This information will be needed in combustion and incineration tests that are planned at a later stage.
2.1 EDX/Elemental analysis
In order to get qualitative indications as to their elemental composition, the samples have been subjected to an energy-dispersive X-ray fluorescence analysis. Besides the expected elements, the analysis also found considerable levels of Bromine in samples “1”,”3”, and “4”.
This qualitative analysis was followed by a quantitative analysis of the elemental composition of the materials. In an Elemental Analysis the samples were analysed to Hydrogen, Carbon, Nitrogen, Phosphorus, Halogens (Chlorine, Bromine, Fluorine), to Aluminium, and Magnesium. The results are summarised in table 10.
Sample FR-4 1 2 3 4 Unit % % % % % Nitrogen Phosphorus Bromine Aluminium 0.71 <0.02 5.70 4.46 0.91 0.96 0.58 4.45 0.96 1.02 0.043 2.06 1.05 0.22 0.034 0.919 0.4 0.47 0.134 2.69 Table 10: Elemental Analysis
Material “1” was found to contain 0.58% of Bromine and material “4” was found to contain 0.134 % of Bromine. The indications on high Bromine content in material “3” were not confirmed with Elemental Analysis. The bromine levels of laminate “1” and “4”, however, appear relatively high when comparing them to Standard-FR-4 that was found to contain only 5.7% of Bromine. The high bromine levels of these two “halogen free” materials will be further investigated.
The Nitrogen content of the halogen free materials was in the same range as for Standard-FR-4 (0.4-1.05 vs. 0.71%). The halogen free materials additionally have a Phosphorus content of 0.2-1.02%, while no Phosphorus was detectable in the Standard-FR-4.
The relatively high Aluminium content found in all samples is most probably due to the Al-content of the reinforcement-glass and not due to the use of Al-containing fillers. Magnesium was only found in minute amounts, indicating that no Mg-containing fillers are used.
Our studies indicate that some of the concerning hal-free materials contain Bromine. The results also indicate that a P/N system was used to flame-retard the halogen free materials.
2.2 Consumer Exposure
Potential consumer exposure (offgasing) by the base materials was assessed by subjecting the samples to a so-called “fogging test” according to DIN 75201. This test is routinely used in the automotive industry to evaluate parts that will be used indoors for their offgasing properties.
A disk of 80 mm diameter is exposed to a temperature of 100C for 16 hours in a closed apparatus. This apparatus is closed with a plate that is cooled to 21C. Potential offgasing substances will condense there and are weighed and then subjected to a screening analysis via GC/MS. The set-up of this experiment represents a worst case scenario, since the plain, uncladded materials (maximum surface) were investigated.
The fogging condensates determined for the five materials were between 0.05 mg and 0.16 mg. Maximum values for the fogging condensates that are accepted by the German automotive industry are in the range of 0.5-3 mg, depending on the exact use of the indoor part. With respect to these values, the tested materials show comparatively low amounts of fogging condensates.
The GC/MS screening analysis of the fogging condensates indicate that Hydrocarbons (alkanes, C20-C30-hydrocarbons and other aliphatic compounds) are the main constituents of the condensed matte