Proceedings of the 41st International Conference on Advanced Ceramics and Composites

By Roger Narayan

This proceedings contains a collection of 23 papers from The American Ceramic Society’s 41st International Conference on Advanced Ceramics and Composites, held in Daytona Beach, Florida, January 22-27, 2017. This issue includes papers presented in the following symposia:

•             Symposium 1     Mechanical Behavior and Performance of Ceramics and Composites

•             Symposium 2     Advanced Ceramic Coatings for Structural, Environmental, and Functional Applications

•             Symposium 4     Armor Ceramics: Challenges and New Developments

•             Symposium 5     Next Generation Bioceramics and Biocomposites

•             6th Global Young Investigators Forum

• Language: English
• Publisher: Wiley
• Release date: Feb 19, 2018
• ISBN: 9781119474388

Book preview

Introduction

This Ceramic Engineering and Science Proceedings (CESP) issue consists of 23 papers that were submitted and approved from select symposia held during the

41st International Conference on Advanced Ceramics and Composites (ICACC), held January 22–27, 2017 in Daytona Beach, Florida. ICACC is the most promi- nent international meeting in the area of advanced structural, functional, and nanoscopic ceramics, composites, and other emerging ceramic materials and technologies. This prestigious conference has been organized by the Engineering Ceramics Division (ECD) of The American Ceramic Society (ACerS) since 1977.

The 41st ICACC hosted more than 1,000 attendees from 41 countries that gave over 850 presentations. The topics ranged from ceramic nanomaterials to struc- tural reliability of ceramic components, which demonstrated the linkage between materials science developments at the atomic level and macro level structural ap- plications. Papers addressed material, model, and component development and investigated the interrelations between the processing, properties, and microstruc- ture of ceramic materials.

The 2017 conference was organized into the following 15 symposia and 3 Fo- cused Sessions and two Special Sessions:

The proceedings papers from this meeting are published in the below two is- sues of the 2017 Ceramic Engineering and Science Proceedings (CESP):

CESP Volume 38, Issue 2 (includes 23 papers from Symposia 1, 2, 4, 5, and GYIF)

CESP Volume 38, Issue 3 (includes 24 papers from Symposia 3, 8, 11, 12, 13, 14, 15 and FS1)

The organization of the Daytona Beach meeting and the publication of these proceedings were possible thanks to the professional staff of ACerS and the tire- less dedication of many ECD members. We would especially like to express our sincere thanks to the symposia organizers, session chairs, presenters and confer-ence attendees, for their efforts and enthusiastic participation in the vibrant and cutting-edge conference.

ACerS and the ECD invite you to attend the 42nd International Conference on Advanced Ceramics and Composites (http://www.ceramics.org/icacc2018) Janu- ary 21-26, 2018 in Daytona Beach, Florida.

To purchase additional CESP issues as well as other ceramic publications, vis- it the ACerS-Wiley Publications home page at www.wiley.com/go/ceramics.

SUROJIT GUPTA, University of North Dakota, USA

JINGYANG WANG, Institute of Metal Research, Chinese Academy of Sciences, China

Volume Editors

August 2017 ICACC

Mechanical Behavior and Performance of Ceramics and Composites

THE EFFECTS OF DIAMOND GRIT CHARACTERISTICS ON THE MICROSTRUCTURE AND ABRASION RESISTANCE OF PCDs SINTERED BY HPHT

Lifen Deng, Roger Nilen and Serdar Ozbayraktar

Global innovation centreElement Six Ltd., Harwell, Oxfordshire, UK

ABSTRACT

Polycrystalline diamond (PCD) compacts used on oil and gas drilling bits were sintered under high-pressure and high-temperature (HPHT) conditions. In this study, the microstructure of polycrystalline diamond composite portion of the compacts is analysed by using scanning electron microscopy (SEM) and image analysis methods. In addition, laboratory rock cutting tests were conducted as a direct measure of abrasion resistance of PDCs produced using different types of diamond powder. Microstructure and abrasion resistance relationships with the grain size, the nitrogen content and the surface quality of diamond grits are discussed. It is found that the PCDs’ abrasion resistance generally decreases with the increasing mean diamond grain size of a sintered PDC. Although the diamond grits with lower nitrogen levels show a higher compression fracture strength, it is not necessarily beneficial for the abrasion resistance of PCDs. It was observed that Type C diamond grits with a rougher surface tends to lead an increasing abrasion resistance of PCD in laboratory rock cutting tests.

INTRODUCTION

Polycrystalline diamonds (PCDs) sintered by HPHT method have been widely used in oil and gas industry as the cutting element on the drilling bit since 1980’s and polycrystalline diamond compact drill bits have accounted for over 60% of drilled footage in 2004[1]. Due to the PCD's super abrasion resistance and reasonably good fracture toughness resistance compared to traditional cutting element materials such as tungsten carbide and hardened steel, PCD compacts have significantly increased the drilling footage and reduced the cost in oil and gas industry [2]. As conventional oil fields are becoming increasingly elusive and continuously reducing the drilling cost is always in demand, the drive for further enhanced abrasion resistance of PCD cutters but not sacrificing too much the fracture toughness is stronger than ever. In the present study, the relationship between the abrasion resistance of PCD cutters and the grain size was discussed. At the same time, 3 different types of Grade 30 (G30, the mean grain size = 30 ͮm) and Grade 22 (G22, the mean grain size = 22 ͮm) diamond grits were used to sinter PCD cutters. Type A diamond powders have a higher nitrogen content than Type B and Type C powders. The surface of Type C diamond grits are rougher than Type B. All PCD cutters were sintered under the same conditions. How the abrasion resistance of PCDs is affected by the nitrogen content and the surface quality of diamond grits was investigated.

EXPERIMENTAL

The nitrogen content of diamond grits were measured by a LECO Nitrogen Analyzer. Each sample was measured 4 times and the average value calculated. In order to investigate the compaction response of diamond grits at high pressure, a cold compaction test was conducted on the three types of G22 and G30 diamond powders. The diamond powders were put into a metal cup and topped with a carbide substrate for the cold compaction at 7 GPa and 8 GPa at room temperature. The particle distribution of diamond grits were measured by Malvern Particle Master Analyzer before and after the cold compaction.

The PCD cutters were sintered under conditions of 1400 to 1600 °C and 5 to 8 GPa [3,4]. Figure 1 shows the schematic design of the pre-sintered unit and the microstructure of PCDs and the tungsten carbide after sintering. During the sintering of PCD, the cobalt in substrate becomes melted and infiltrated into the bed of diamond grits, assisting the sintering of diamonds.

Image described by caption and surrounding text.

Figure 1 The schematic picture showing the sintering of PCDs.

After sintering, all PCD cutters in the present study were processed to the same dimensions of Ø16mm × 12 mm. The abrasion resistance of PCD was measured through rock cutting tests conducted on vertical turret lathe. After the test, the wear scar were measured under an optical light microscopy. The larger the wear scar area of PDC, the lower the wear resistance. For each sample, 4 rock cutting tests were carried out on two locations of 2 PCD cutters, generateing a total 4 data points.

All PCDs were lapped and polished to 1ͮm before the microstructure observation under SEM. With the intention of quantify the microstructure of PCDs, 20 photos were taken on each sample and processed by the commercially available image analysis software package. The percentage of diamond area is one of the image analysis results of PCDs’ microstructure and used to present the sintering status of PCDs. Given the same powder feed, the higher the percentage of diamond area, the more PCD sintered.

RESULTS AND DISCUSSION

Mean grain size of diamond grits

A series of diamond powders with different average grain sizes were prepare and sintered into PCD cutters at 7 GPa. All PCDs were sintered under the same conditions. The SEM observations confirmed that the microstructures of all PCDs showed uniform diamond grain distribution and sintered well. Figure 2(a) and (b) show the microscopy photographs of PCDs with relatively fine and coarse grain sizes. The dark parts are diamond grains and the white dotted boundaries are cobalt binders. It is clearly observed that diamond grains are bonded to each other with some cobalt residual existing at the grain boundaries.

Image described by caption and surrounding text.

Figure 2 Microscopy photographs of PCDs with (a) fine and (b) coarse grain size. (Magnification: × 1000)

It is found that the wear scar area generally increases with the grain size of diamond grits although a linear relation is not strictly followed. In general, a negative correlation is observed between the abrasion resistance of PCDs and the grain size of diamond grits.

Nitrogen content and surface quality of diamond grits

Figure 3 shows the average nitrogen contents of three different types of G30 and G22 measured using LECO Nitrogen Analyzer. It is shown that Type A diamond powders have a higher nitrogen content than Type B and Type C. For G30, the nitrogen level of Type B powders is slightly higher than Type C while for G22, the nitrogen content of Type B and Type C are nearly same.

Graph shows G30 on range from type A to type C versus nitrogen content in ppm from 0 to 120, and graph shows G22 on range from type A to type C versus nitrogen content in ppm from 0 to 120.

Figure 3 The average nitrogen content of three types of G30 and G22 diamond grits.

Figure 4(a) and (b) show the SEM images of G30 diamond grits (Type A) before and after the compaction at 7 GPa. It is observed that the original, generally uniform coarse diamond grits were smashed into smaller grits with few relatively coarse ones. The original diamond powders have a relatively clear and smooth appearance while after compression, the smashed diamond grits showed sharp edges and rough surfaces. Obviously, the crushing occurred during the high pressure compaction helped the further dense of the diamond bed before the sintering with the infiltration of liquid cobalt. Figure 5(a) clearly displayed the particle distribution shifted after the compaction at 7 GPa.

Image described by caption and surrounding text.

Figure 4 (a) The photoof original G30diamondgrits(Type A) beforethepress. (b) The photoof the samediamondgrits after compactionat 7 GPa. (Magnification: × 500)

Graphs show particle size in um from 1 to 10, 0.1 to 100, and 0.1 to 10 versus volume in percentage from 0 to 20, 0 to 6, and 0 to 8 with plots for after, before, type A, type B, and type C.

Figure 5 Particle size distribution plots of G30 (a) before and after compaction, (b) after compaction at 7 GPa, (c) after compaction at 8 GPa. (d) Particle size distribution plots of G22 after compaction at 7 GPa.

Figure 5(b) and (c) are the particle distribution plots of G30 diamond grits after the compaction at 7 GPa and 8 GPa, respectively. Figure 5(b) clearly disclosed that the crushing rate of Type A diamond grits is higher than Type B and C. It does indicate that the diamond grits (Type B and C) with lower nitrogen contents exhibited a high compression fracture strength than Type A ones, which is agree with the literature [5]. Brookes and Daniel reported that [5] at room temperature, a reduction in the concentration of single substitutional diamond increases the indentation hardness of the diamond. Figure 5(c) showed that the particle distribution plots of G30 Type A and Type B are nearly same after the compaction at 8 GPa. Type C ones share the same particle distribution as Type A and Type B as well. Just for a clear appearance of the plots, the data of Type C is not included in Figure 5(c). It means that the advantage of higher compression fracture strength of Type B and C diamond powders due to the lower nitrogen content disappeared when the pressure increased from 7 GPa to 8 GPa. Figure 5(d) showed the particle distribution plots of G22 diamond grits after compaction at 7 GPa. The 3 different types of diamond grits shared a very similar particle distribution except Type C showed a marginally lower crushing rate than the other two types. It means that for G22 diamond, type B and C loose the advantage of higher compression fracture strength at lower pressures (say 4-5 GPa but no data available) and hence exhibit a similar grain size distribution at 7 GPa for all three types. Type C powders appeared to have a rougher surface s associated with reduced stress concentration sharp edges after an etching heat treatment, which might explain the slightly lower crushing rate if Fig. 5(d) is accurate. However, how the Type C powders’ surface quality affect the cold compaction, hot compaction and following diamond sintering is not clear yet. Because all three types of G22 diamonds already crushed similarly at 7 GPa, it is reasonable to believe that they would crush similarly at 8 GPa. Therefore no cold compaction test is carried out at 8 GPa.

Figure 6(a) (b) and (c) showed the microstructures of PCDs sintered from the three different types of G30 diamond grits. All PCD microstructures are uniform and look very similar. Figure 6(d) is the percentages of diamond areas extracted from the image analysis. Generally speaking, the three types of PCDs showed a similar diamond area percentage except that Type C one had a slightly lower diamond area.

Images show microscopy photographs and graph shows range from type A to type C versus diamond area in percentages from 93.0 to 95.0.

Figure 6 (a) (b) and (c) Microscopy photographs of PCDs individually sintered from G30 Type A, B and C diamond grits with the magnification of × 1000. (d) The percentages of diamond area in the microstructure of PCDs sintered from G30 diamond grits.

Figure 7(a) (b) and (c) are the microstructure of PCDs sintered from the three different types of G22 diamond grits. They are uniform and look very similar as well. Figure 7(d) showed that the percentages of diamond area of PCDs from Type A and Type B are very similar, both lower than that from Type C.

Images show microscopy photographs and graph shows range from type A to type C versus diamond area in percentages from 93.0 to 95.0.

Figure 7 (a) (b) and (c) Microscopy photographs of PCDs individually sintered from G22 Type A, B and C diamond grits with the magnification of × 1000. (d) The percentages of diamond area in the microstructure of PCDs sintered from G22 diamond grits.

Figure 8 (a) and (b) showed the wear scar areas of PCDs sintered from G30 and G22 at 8 GPa, respectively, after the vertical turret test. It turns out that although Type B diamond grits have higher compression fracture strength than Type A due to the lower nitrogen content, the abrasion resistance of PCDs is not necessarily improved. Actually, the abrasion resistance tested in this study decreased for both G22 and G30 as the diamond grits changed from Type A to Type B. The reason is unclear yet. It may rely on the vertical turret test which might be more sensitive to the grain boundary rather than the diamond grit itself. Or it may because less plastic deformation occurred at high temperature in diamond grits with lower nitrogen content, therefore the sintering process is affected accordingly. According to [5], when the temperature is high enough (1400࢑C), an increase in nitrogen content is reflected in a decrease in the volume of plastic deformation associated with impressions. On the other hand, bearing in mind that after HPHT sintering, the total nitrogen level in PCDs changed. At HPHT conditions, nitrogen atoms become mobile in diamond lattice, therefore the original internal nitrogen interstitial can diffuse outside and the external nitrogen could enter diamond as well. However, it seems to suggest that Type C powders might be beneficial for the improvement of abrasion resistance of PCDs whatever the percentage of diamond area is higher or lower. The reason might be related to the increased surface area due to the enhanced roughness of diamond grit and also to the potential more active surface on the diamond grits after the etching process. The further study is needed for the surface quality of diamond grits and how they affect the bonding between diamond to diamond during sintering.

Graphs show interval plot of type A, type B, type C on range from type A to type C versus wear scar area in mm2 from 8 to 13 and 7 to 12.

Figure 8 Plots of wear scar areas of PCDs sintered from 3 different types of G30 and G22 diamond grits at 8 GPa.

CONCLUSIONS

Considering the limited data and the complication of PCD cutters’ abrasion performance, it is too early to draw some solid conclusion. However, the main findings from the present study can be summarised as below.

The abrasion resistance of PCDs generally decreases as the grain size increases.

Type B and Type C diamond grits show lower nitrogen content which could help G30 to crush less under 7 GPa but not at under 8 GPa.

For G22, Type B and Type A diamond grits crushed similarly under 7 GPa. Type C diamond grits crushed slightly less than Type B under 7 GPa, which might be attributed to the higher roughness of diamond grit surface.

Although diamond grits of Type B with lower nitrogen contents are harder than Type A ones, it is not necessarily beneficial for the improvement of PCD's abrasion resistance.

Type C diamond grits of both G22 and G30 seem to help improve the abrasion resistance of PCDs, which might be due to the increased surface area of diamond and the potential enhanced activeness during sintering.

REFERENCES

Scott 2006, The history and impact of Synthetic Diamond Cutters and Diamond Enhanced Inserts on the Oil and Gas Industry

Valentine Kanyanta, Hard, super hard and ultrahard materials: An overview, Microstructure-Property Correlations for Hard, Superhard, and Ultrahard Materials, © Springer International Publishing Switzerland 2016, P15

Robert Fries, Patent number 9120204, Polycrystalline superhard material

J. J. Barry, T. P. Mollart, R. W. N. Nilen, patent number 9095918, Cutter structures, inserts comprising same and method for making same

E.J. Brookes, R.D. Daniel, Influence of nitrogen content on the mechanical properties of diamond, Properties, growth and applications of diamond, published by: INSPEC, 2001.

CRUSH STRENGTH ANALYSIS OF HOLLOW GLASS MICROSPHERES

Ben Dillinger, David Clark and Carlos Suchicital

Department of Materials Science and EngineeringVirginia Polytechnic Institute and State University Blacksburg, VA, USA

George Wicks Applied Research Center (ARC) Aiken, SC, USA

ABSTRACT

Porous Wall Hollow Glass Microspheres (PWHGMs) are a new type of microsphere developed by Wicks et al at the Savannah River National Laboratory. What makes these microspheres different from other microspheres is the interconnected nanoporosity inside the microsphere wall. With this interconnected porosity, materials are able to travel from the outside of the microsphere to the hollow interior and vice versa. Because of this unique feature, PWHGMs have been used in encapsulation and filtration applications such as drug delivery and metal hydride encapsulation. The main goal of this research was to analyze the crush strength of the microspheres by examining data collected from mechanical testing using the Weibull analysis. Data were collected by crushing individual microspheres using a nanoindenter with a flat end tip. Microspheres were divided into test groups according to type of microsphere and microsphere diameter. Weibull analysis results were used to compare the microsphere crush strength between different categories. The different types of microspheres tested include two types of commercially available microspheres from 3 M and the microspheres from the different stages of PWHGM production. Weibull analysis results show that as the diameter of the microspheres decreases, the strength of the microspheres increases. Results also indicate that the PWHGMs are weaker at the final stage of production, after wall porosity is induced, compared to the initial stage of production.

INTRODUCTION

Hollow Glass Microspheres (HGMs) are a unique set of materials that have uses in a variety of applications. Developed by 3 M in the 1960s for buoyancy applications, HGMs are small hollow spheres, less than the diameter of a human hair ( < 100 microns), with wall thicknesses between 0.5 to 2 microns ¹. They have been used as strong, light weight fillers for products ranging from foams to furniture and have found use in other areas such as the defense and construction sectors ¹, ². A new type of microsphere, named the Porous Wall Hollow Glass Microsphere (PWHGM), was recently developed by George Wicks, Ray Schumacher, and Kit Heung at the Savannah River National Laboratory by combining 3M's method for producing microspheres with Corning's Vycor process ³. What makes this microsphere unique is the interconnected nanoporosity (0.01 to 0.1 microns in diameter) running throughout the microsphere wall. This porosity is created by heat treating the microspheres to induce phase separation and then removing the secondary phase through leaching. With this interconnected porosity, it is possible for some materials to travel though the microsphere wall, allowing the PWHGMs to be used in encapsulation and filtration applications. Research on PWHGM technology and applications is currently being further advanced at the Applied Research Center (ARC). There are a wide variety of PWHGM applications that have been studied, ranging from drug delivery to metal hydride and palladium encapsulation to anti-counterfeiting ⁴, ⁵, ⁶, ⁷, ⁸.

One important property that should be understood when designing most applications is the crush strength of the microspheres. Knowing when the microspheres will fail due to compressive forces is important and a limiting factor for many applications. Unfortunately, the crush strength of these small objects is not well understood and difficult to test due to the size of the microspheres. Strength data provided by companies is usually determined through bulk testing based on the withdrawn ASTM standard D 3102-78: Standard Practice for Determination of Isostatic Collapse Strength of Hollow Glass Microspheres ⁹, ¹⁰. In this procedure a large quantity of microspheres are tested at once and without consideration for how the microsphere diameter will affect the results. Tests of the mechanical properties of individual microspheres have been reported by two groups, Bratt et al ¹¹ and Carlisle et al ¹⁰, ¹², ¹³. These groups both used a form of nanoindentation with a flat end tip to crush individual microspheres but had different methods for analyzing the results. The research from these groups was used to help design a process for the experiments described in this paper.

The main goal of this research was to identify and crush individual microspheres using nanoindentation with a flat end tip and analyze the results using the two parameter Weibull analysis. With the results, the characteristic strength of the microspheres and the amount of scattering in the data were determined. Microspheres were separated into different test groups based on type of microsphere and microsphere diameter. Multiple experiments were conducted for each group. Results from the Weibull analysis were used to determine the difference in strength different types of microspheres exhibited and how the diameter of the microspheres affected the strength.

WEIBULL ANALYSIS

For the research presented in this paper, the two parameter Weibull analysis was the method chosen to analyze the crush strength. The Weibull analysis is commonly used in engineering as a method of analyzing the mechanical properties of materials. It is better at analyzing mechanical failure in a material than a Gaussian distribution because it takes into account the effects of volume on the strength of a material and is an extreme value distribution. In an extreme value distribution, the data is skewed more toward one side of the distribution, meaning there is less scattering in this type of distribution than the Gaussian distribution. The Weibull analysis takes the effects of volume into account by assuming that the number of flaws per unit volume should be the same. This means that samples with the same volume will have the same number of flaws present while a larger sample will have a larger number of flaws present. One assumption made by the Weibull analysis is that it is more likely that the length of the critical flaw that will cause failure in a material will be longer in a larger sample than a smaller one ¹⁴, ¹⁵, ¹⁶.

The two parameter Weibull equation is shown in Equation 1:

(1) numbered Display Equation

In Equation 1, Pf is the probability of failure, is the stress at failure, m is the Weibull modulus, is the characteristic strength, k is the load factor, V is the volume under stress, and V0 a chosen unit volume. On a Weibull Plot, is the y-axis and is the x-axis, while m is the slope of the line and the y axis intercept is . m and are Weibull parameters, k is a load factor used to modify the Weibull equation for different types of loading, and V/V0 is a volume variable used to predict how material would act if its volume was changed. Since V0 is chosen, V/V0 is usually set to 1 and removed from the equation. For uniform, uniaxial tension or compression k is equal to 1 ¹⁴, ¹⁷.

In statistics, the Weibull modulus, m, is known as the shape parameter and describes the general shape of a curve based on how the data set is distributed. Lower values represent a right skewed distribution and higher values represent a left skewed distribution. Higher values of m are desired as it indicates that variations in failure stress typically become smaller 14, 18.

The characteristic strength, , is the scale parameter which defines the range of the distribution. As becomes larger, the stress required to cause the majority of samples to fail becomes larger. In the Weibull analysis the characteristic stress is the stress that will cause 63.2% of the samples to fail. This probability of failure will always occur in an analysis when  =  14, 19.

Probability of failure is calculated using a rank order table, where the failure strength values for a data set are ordered from lowest stress to highest stress and given a rank based on their location in the table as seen in Table 1. The probability of failure for each data point was then calculated using Equation 2.

(2) numbered Display Equation

Table 1 Table I: Calculation of Probability of Failure. This table shows how the probability of failure was calculated by rank ordering the Samples from lowest to highest fracture stress. The value of N for this set of data was 10. The set shown are the 60-70 micron ARC HGMs.

In Equation 2, n is number of samples that have failed at or less than the current sample's stress and N is the total number of samples tested. Data sets are usually rank ordered from lowest to highest stress where n is the rank of a specific stress.