MIL-HDBK-17-1F Volume 1,Chapter 2 Guidelines for Property Testing of Composites This section is meant to be a brief discussion of variability in composite properties arising from the various processes which are encountered in the materials and processing pipeline.For a more extensive and detailed treatment of this subject.the reader is referred to the broader discussion of these issues which may be found in Volume 3,Chapter 2 entitled Materials and Processes-The Effect of Variability on Composite Properties.Volume 3,Chapter 2 also includes a discussion of preparation of materials and processing specifications.The composite end item manufacturer has no direct control over the process- ing of incoming materials,and the use of such specifications is essential in minimizing materials variabil- ity. 2.2.6.2 Specimen preparation and NDE This section is reserved for future use. 2.2.7 Moisture absorption and conditioning factors. Most polymeric materials,whether in the form of a composite matrix or a polymeric fiber,are capable of absorbing relatively small but potentially significant amounts of moisture from the surrounding environ- ment.The physical mechanism for moisture gain,assuming there are no cracks or other wicking paths, is generally assumed to be mass diffusion following Fick's Law(the moisture analog to thermal diffusion) While material surfaces in direct contact with the environment absorb or desorb moisture almost immedi- ately,moisture flow into or out of the interior occurs relatively slowly.The moisture diffusion rate is many orders of magnitude slower than heat flow in thermal diffusion.Nevertheless,after a few weeks or months of exposure to a humid environment,a significant amount of water will eventually be absorbed by the material.This absorbed water may produce dimensional changes(swelling),lower the glass transi- tion temperature of the polymer,and reduce the matrix and matrix/fiber interface dependent mechanical properties of the composite(effectively lowering the maximum use temperature of the material---see Sec- tion 2.2.8).Because absorbed moisture is a potential design concern for many applications,material test- ing should include evaluation of properties after representative moisture exposure.Since the amount of moisture absorbed by a material is thickness and exposure time-dependent.fixed-time conditioning methods should not be followed.2 Instead,a conditioning procedure such as ASTM D 5229/D 5229M (Reference 2.2.7(c))should be followed that accounts for the diffusion process and terminates with the moisture content nearly uniform through the thickness. There are two moisture properties of a Fickian material:moisture diffusivity and moisture equilibrium content (weight percent moisture).These properties are commonly determined by a gravimetric test method(such as ASTM D 5229/D 5229M Procedure A)that exposes an initially dry specimen to a humid environment and documents moisture mass gain versus the square-root of time.During early weighings this mass-time relation will be linear.the slope of which is related to the rate of absorption (the moisture diffusivity).As the moisture content in a substantial volume of the exterior of the material begins to ap- proach equilibrium the mass gain versus square-root time slope becomes increasingly smaller.Eventu- ally,as the interior of the material approaches equilibrium,the difference between subsequent weighings will approach zero and the slope will be nearly parallel to the time axis.The weight percent mass gain at this point is the moisture equilibrium content.This process is illustrated in Figures 2.2.7(a)and(b).Fig- ure 2.2.7(a)shows the total mass gain versus root-time during specimen moisture exposure,also show- ing the difference in response due to different temperatures.For the 150F condition(the diamonds in Figure 2.2.7(a)).Figure 2.2.7(b)shows the moisture profile through the thickness of the specimen for While certain polymers.like polybutadiene,resist moisture absorption to the point that moisture conditioning may not be required. these materials are still considered rare exceptions.On the other hand,a great many reinforcements,including those in the carbon glass,metallic,and ceramic fiber families,are not hygroscopic.As a result,except for polymeric fibers like aramid,it is usually as- sumed that any moisture absorption is limited to the polymer matrix. Examples of fixed-time conditioning methods include ASTM D 618(Reference 2.2.7(a))and D 570(Reference 2.2.7(b))for plastics. 3The discussion focuses on through the thickness moisture absorption;however,in-plane moisture absorption will locally dominate near edges,and may even dominate the overall absorption process in those cases where edge area is a substantial portion of the total exposed area.As the in-plane moisture absorption response may be substantially different than the through the thickness response,due to non-Fickian moisture wicking provided by the presence of the fibers,one should not assume that edge effects will be negligible except for very small ratios of edge area to surface area. 2-16
MIL-HDBK-17-1F Volume 1, Chapter 2 Guidelines for Property Testing of Composites 2-16 This section is meant to be a brief discussion of variability in composite properties arising from the various processes which are encountered in the materials and processing pipeline. For a more extensive and detailed treatment of this subject, the reader is referred to the broader discussion of these issues which may be found in Volume 3, Chapter 2 entitled Materials and Processes - The Effect of Variability on Composite Properties. Volume 3, Chapter 2 also includes a discussion of preparation of materials and processing specifications. The composite end item manufacturer has no direct control over the processing of incoming materials, and the use of such specifications is essential in minimizing materials variability. 2.2.6.2 Specimen preparation and NDE This section is reserved for future use. 2.2.7 Moisture absorption and conditioning factors. Most polymeric materials, whether in the form of a composite matrix or a polymeric fiber, are capable of absorbing relatively small but potentially significant amounts of moisture from the surrounding environment.1 The physical mechanism for moisture gain, assuming there are no cracks or other wicking paths, is generally assumed to be mass diffusion following Fick's Law (the moisture analog to thermal diffusion). While material surfaces in direct contact with the environment absorb or desorb moisture almost immediately, moisture flow into or out of the interior occurs relatively slowly. The moisture diffusion rate is many orders of magnitude slower than heat flow in thermal diffusion. Nevertheless, after a few weeks or months of exposure to a humid environment, a significant amount of water will eventually be absorbed by the material. This absorbed water may produce dimensional changes (swelling), lower the glass transition temperature of the polymer, and reduce the matrix and matrix/fiber interface dependent mechanical properties of the composite (effectively lowering the maximum use temperature of the material---see Section 2.2.8). Because absorbed moisture is a potential design concern for many applications, material testing should include evaluation of properties after representative moisture exposure. Since the amount of moisture absorbed by a material is thickness and exposure time-dependent, fixed-time conditioning methods should not be followed.2 Instead, a conditioning procedure such as ASTM D 5229/D 5229M (Reference 2.2.7(c)) should be followed that accounts for the diffusion process and terminates with the moisture content nearly uniform through the thickness. 3 There are two moisture properties of a Fickian material: moisture diffusivity and moisture equilibrium content (weight percent moisture). These properties are commonly determined by a gravimetric test method (such as ASTM D 5229/D 5229M Procedure A) that exposes an initially dry specimen to a humid environment and documents moisture mass gain versus the square-root of time. During early weighings this mass-time relation will be linear, the slope of which is related to the rate of absorption (the moisture diffusivity). As the moisture content in a substantial volume of the exterior of the material begins to approach equilibrium the mass gain versus square-root time slope becomes increasingly smaller. Eventually, as the interior of the material approaches equilibrium, the difference between subsequent weighings will approach zero and the slope will be nearly parallel to the time axis. The weight percent mass gain at this point is the moisture equilibrium content. This process is illustrated in Figures 2.2.7(a) and (b). Figure 2.2.7(a) shows the total mass gain versus root-time during specimen moisture exposure, also showing the difference in response due to different temperatures. For the 150°F condition (the diamonds in Figure 2.2.7(a)), Figure 2.2.7(b) shows the moisture profile through the thickness of the specimen for 1 While certain polymers, like polybutadiene, resist moisture absorption to the point that moisture conditioning may not be required, these materials are still considered rare exceptions. On the other hand, a great many reinforcements, including those in the carbon, glass, metallic, and ceramic fiber families, are not hygroscopic. As a result, except for polymeric fibers like aramid, it is usually assumed that any moisture absorption is limited to the polymer matrix. 2 Examples of fixed-time conditioning methods include ASTM D 618 (Reference 2.2.7(a)) and D 570 (Reference 2.2.7(b)) for plastics. 3 The discussion focuses on through the thickness moisture absorption; however, in-plane moisture absorption will locally dominate near edges, and may even dominate the overall absorption process in those cases where edge area is a substantial portion of the total exposed area. As the in-plane moisture absorption response may be substantially different than the through the thickness response, due to non-Fickian moisture wicking provided by the presence of the fibers, one should not assume that edge effects will be negligible except for very small ratios of edge area to surface area
MIL-HDBK-17-1F Volume 1,Chapter 2 Guidelines for Property Testing of Composites several early time periods,illustrating the rapid moisture uptake near the surface together with the rela- tively slow uptake of moisture in the middle of the specimen. 1.6 ●180F (82C) ◆150°F I66C) 1.4 ▲120°F (49 ) 蓝 75°F (24C 1.2 1.0 .8 HTS Carbon/Epoxy:B Phy Laminate:95%RH 2 1 6 8 10 Time (Days)* FIGURE 2.2.7(a)Typical moisture absorption response(Reference 2.2.7(d)). HTSE春bon/epery: 15"F4d9%RH FIGURE 2.2.7(b)Through the thickness moisture profile versus time(Reference 2.2.7(d)) 2-17
MIL-HDBK-17-1F Volume 1, Chapter 2 Guidelines for Property Testing of Composites 2-17 several early time periods, illustrating the rapid moisture uptake near the surface together with the relatively slow uptake of moisture in the middle of the specimen. FIGURE 2.2.7(a) Typical moisture absorption response (Reference 2.2.7(d)). FIGURE 2.2.7(b) Through the thickness moisture profile versus time (Reference 2.2.7(d))
MIL-HDBK-17-1F Volume 1,Chapter 2 Guidelines for Property Testing of Composites 2.2.7.1 Moisture diffusivity The rate of moisture absorption is controlled by the material property called moisture diffusivity.Mois- ture diffusivity is usually only weakly related to relative humidity and is often assumed to be a function only of temperature,usually following an Arrhenius-type exponential relation with inverse absolute tem- perature.This strong temperature dependence is illustrated in Figure 2.2.7.1(a),which shows moisture diffusivity versus temperature for a particular type of carbon/toughened epoxy.Figure 2.2.7.1(b)illus- trates,for a different material system,a family of moisture mass gain curves obtained at several tempera- tures.For this material system,a decrease in conditioning temperature of 60F(33C)increased the time required to absorb 1%moisture by a factor of five. 2.2.7.2 Moisture equilibrium content Moisture equilibrium content is only weakly related to temperature and is usually assumed to be a function only of relative humidity.The largest value of moisture equilibrium content for a given material under humid conditions occurs at 100%relative humidity and is also often called the saturation content. The moisture equilibrium content at a given relative humidity has been found to be approximately equal to relative humidity times the material saturation content;however,as illustrated by Figure 2.2.7.2,this linear approximation does not necessarily hold well for every material system.Regardless,if a material does not reach the moisture equilibrium content for the given relative humidity,then the local moisture content is not uniform through-the-thickness.Another point to be emphasized is that moisture absorption proper- ties under atmospheric humid conditions are generally not equivalent to exposure either to liquid immer- sion or to pressurized steam.These latter environments alter the material diffusion characteristics,pro- ducing a higher moisture equilibrium content,and should not be used unless they simulate the application environment in question. Arrhenius-Type Plot uf D:Diffusion Scaled to Mm=1.4%% 5.00 11 6.00 82 ,200 20c 8.0m十 3413E03 26116-05 Inverse Absotute Temperature (I/K) FIGURE 2.2.7.1(a)Moisture diffusivity as a function of temperature. 2-18
MIL-HDBK-17-1F Volume 1, Chapter 2 Guidelines for Property Testing of Composites 2-18 2.2.7.1 Moisture diffusivity The rate of moisture absorption is controlled by the material property called moisture diffusivity. Moisture diffusivity is usually only weakly related to relative humidity and is often assumed to be a function only of temperature, usually following an Arrhenius-type exponential relation with inverse absolute temperature. This strong temperature dependence is illustrated in Figure 2.2.7.1(a), which shows moisture diffusivity versus temperature for a particular type of carbon/toughened epoxy. Figure 2.2.7.1(b) illustrates, for a different material system, a family of moisture mass gain curves obtained at several temperatures. For this material system, a decrease in conditioning temperature of 60°F (33°C) increased the time required to absorb 1% moisture by a factor of five. 2.2.7.2 Moisture equilibrium content Moisture equilibrium content is only weakly related to temperature and is usually assumed to be a function only of relative humidity. The largest value of moisture equilibrium content for a given material under humid conditions occurs at 100% relative humidity and is also often called the saturation content. The moisture equilibrium content at a given relative humidity has been found to be approximately equal to relative humidity times the material saturation content; however, as illustrated by Figure 2.2.7.2, this linear approximation does not necessarily hold well for every material system. Regardless, if a material does not reach the moisture equilibrium content for the given relative humidity, then the local moisture content is not uniform through-the-thickness. Another point to be emphasized is that moisture absorption properties under atmospheric humid conditions are generally not equivalent to exposure either to liquid immersion or to pressurized steam. These latter environments alter the material diffusion characteristics, producing a higher moisture equilibrium content, and should not be used unless they simulate the application environment in question. FIGURE 2.2.7.1(a) Moisture diffusivity as a function of temperature
MIL-HDBK-17-1F Volume 1,Chapter 2 Guidelines for Property Testing of Composites 20 f.92% 名色o免192x 1.74x 1.6 4044- A 260℉(s 1.2 d 180(0 43- 120r(0 0.8 LEGEND ● 0.4 0.0 0 o 12 TIME,DAYS FIGURE 2.2.7.1(b)Effect of temperature on moisture absorption rate in hybrid boron-graphite/epoxy (5505-AS/3501)laminate(3.0 x 0.5 x 12 in., 76 x 13 x 3.0 mm)(Reference 2.2.7.1). -T300/5208 AS/3501-5/B/5505 2、 1 20 100 RELATIVE HUMIDITY, FIGURE 2.2.7.2 Equilibrium moisture content versus relative humidity. 2-19
MIL-HDBK-17-1F Volume 1, Chapter 2 Guidelines for Property Testing of Composites 2-19 FIGURE 2.2.7.1(b) Effect of temperature on moisture absorption rate in hybrid boron-graphite/epoxy (5505-AS/3501) laminate (3.0 x 0.5 x 12 in., 76 x 13 x 3.0 mm) (Reference 2.2.7.1). FIGURE 2.2.7.2 Equilibrium moisture content versus relative humidity
MIL-HDBK-17-1F Volume 1,Chapter 2 Guidelines for Property Testing of Composites 2.2.7.3 Conditioning and test environment To evaluate worst-case effects of moisture content on material properties,tests are performed with specimens preconditioned to the design service (end-of-life)moisture content (assumed equivalent to equilibrium at the design service relative humidity).The preferred conditioning methodology uses ASTM D 5229/D 5229M,the process of which is summarized in Section 6.3. The design service moisture content is determined(if it is not specified by the procuring or certifying agency)from semi-empirical calculations that consider secondary effects on a particular type of structure, or more conservatively established by simpler assumptions.An example of the first case is documented in Reference 2.2.7.3(a),where worldwide climatic data and USAF aircraft-basing data were combined to define runway storage environmental spectra for each of the three classes of USAF air vehicles:fighters, bombers,and cargo/tankers.The study applied a ranking procedure to select baseline and worst-case locations with respect to the absorption of moisture by typical carbon/epoxy composite structures.Such data can be used to establish design service moisture content for a particular application;a typical spe- cific design service relative humidity might be 81%RH for a tropically-based supersonic aircraft.Another, more conservative,approach is to use the average relative humidity for a selected diurnal cycle taken from a reference such as MIL-STD-210(Reference 2.2.7.3(b)),the U.S.military guide to worldwide envi- ronmental exposure conditions.This usually leads to a higher design service relative humidity(88%RH being typical),since dry-out due to solar radiation,flight excursions (supersonic in particular),and sea- sonal climatic changes are not considered. Given these and other historical considerations,the MIL-HDBK-17 Coordination Group has agreed that a reasonable upper-bound value for aircraft design service relative humidity is 85%,and that this value may be used when a specific determination of design service moisture content has not been estab- lished for a specific aircraft application.Use of a design service moisture content of 85%RH will obviate extrapolation of data when test specimens are conditioned to equilibrium at this moisture level.Accepted design service moisture levels for other applications have not yet been established. Hot-wet test data being submitted to MIL-HDBK-17 should have specimens conditioned to an equilib- rium moisture content and tested at the material operational limit(MOL)temperature or below(see Fig- ures 2.2.8(a)-(c)).As can be seen in Figure 2.2.8(a),the effect of environment is generally small for ma- trix-dependent properties at temperatures below room temperature.However,the fiber-dependent prop- erties of many material systems experience a steady degradation with increasingly colder temperatures, though without a cold MOL.A comparison of tensile (fiber-dominated)and compressive (matrix- influenced)response to varying temperature is shown in Figure 2.2.7.3.Due to these factors,qualifica- tion/certification testing programs typically do not require moisture conditioning below room temperature, and since there is generally no need to determine a cold MOL,are simply tested at the coldest design service temperature(often-55°C(-67°F). 2.2.8 Material operational limit(MOL) As noted earlier,properties of polymer matrix composites are influenced markedly by temperature and moisture.Generally,matrix-dominated mechanical property values decrease with increases in mois- ture content and increases in temperature above room temperature.For properties that are highly domi- nated by reinforcement (fiber)properties (unidirectional tension,for example).this reduction may be re- versed,not occur,or be minimal over reasonable temperature ranges.For properties influenced by the organic matrix(shear and compression,for example),the degradation of properties can be significant. Furthermore,the degradation is not linear.At a given moisture content,it becomes more severe with in- creasing temperature until a temperature is reached where dramatic property reductions begin to occur. and beyond which these reductions may become irreversible.It is desirable to specify this onset of dra- matic reduction as a "characteristic temperature".which is also defined to be the material operational limit (MOL),or the maximum operating temperature. 2-20
MIL-HDBK-17-1F Volume 1, Chapter 2 Guidelines for Property Testing of Composites 2-20 2.2.7.3 Conditioning and test environment To evaluate worst-case effects of moisture content on material properties, tests are performed with specimens preconditioned to the design service (end-of-life) moisture content (assumed equivalent to equilibrium at the design service relative humidity). The preferred conditioning methodology uses ASTM D 5229/D 5229M, the process of which is summarized in Section 6.3. The design service moisture content is determined (if it is not specified by the procuring or certifying agency) from semi-empirical calculations that consider secondary effects on a particular type of structure, or more conservatively established by simpler assumptions. An example of the first case is documented in Reference 2.2.7.3(a), where worldwide climatic data and USAF aircraft-basing data were combined to define runway storage environmental spectra for each of the three classes of USAF air vehicles: fighters, bombers, and cargo/tankers. The study applied a ranking procedure to select baseline and worst-case locations with respect to the absorption of moisture by typical carbon/epoxy composite structures. Such data can be used to establish design service moisture content for a particular application; a typical specific design service relative humidity might be 81% RH for a tropically-based supersonic aircraft. Another, more conservative, approach is to use the average relative humidity for a selected diurnal cycle taken from a reference such as MIL-STD-210 (Reference 2.2.7.3(b)), the U.S. military guide to worldwide environmental exposure conditions. This usually leads to a higher design service relative humidity (88% RH being typical), since dry-out due to solar radiation, flight excursions (supersonic in particular), and seasonal climatic changes are not considered. Given these and other historical considerations, the MIL-HDBK-17 Coordination Group has agreed that a reasonable upper-bound value for aircraft design service relative humidity is 85%, and that this value may be used when a specific determination of design service moisture content has not been established for a specific aircraft application. Use of a design service moisture content of 85% RH will obviate extrapolation of data when test specimens are conditioned to equilibrium at this moisture level. Accepted design service moisture levels for other applications have not yet been established. Hot-wet test data being submitted to MIL-HDBK-17 should have specimens conditioned to an equilibrium moisture content and tested at the material operational limit (MOL) temperature or below (see Figures 2.2.8(a)-(c)). As can be seen in Figure 2.2.8(a), the effect of environment is generally small for matrix-dependent properties at temperatures below room temperature. However, the fiber-dependent properties of many material systems experience a steady degradation with increasingly colder temperatures, though without a cold MOL. A comparison of tensile (fiber-dominated) and compressive (matrixinfluenced) response to varying temperature is shown in Figure 2.2.7.3. Due to these factors, qualification/certification testing programs typically do not require moisture conditioning below room temperature, and since there is generally no need to determine a cold MOL, are simply tested at the coldest design service temperature (often -55°C (-67°F)). 2.2.8 Material operational limit (MOL) As noted earlier, properties of polymer matrix composites are influenced markedly by temperature and moisture. Generally, matrix-dominated mechanical property values decrease with increases in moisture content and increases in temperature above room temperature. For properties that are highly dominated by reinforcement (fiber) properties (unidirectional tension, for example), this reduction may be reversed, not occur, or be minimal over reasonable temperature ranges. For properties influenced by the organic matrix (shear and compression, for example), the degradation of properties can be significant. Furthermore, the degradation is not linear. At a given moisture content, it becomes more severe with increasing temperature until a temperature is reached where dramatic property reductions begin to occur, and beyond which these reductions may become irreversible. It is desirable to specify this onset of dramatic reduction as a "characteristic temperature", which is also defined to be the material operational limit (MOL), or the maximum operating temperature