Low-Loss Dielectric Materials Characterization

Circuit designers need dielectric property data for materials at millimeter-wave (mmWave) frequencies to optimize device performance of new hardware and for quality assurance. However, there are no standard traceable reference materials or even agreed upon characterization test methods for materials at mmWave frequencies. Further, at higher frequencies, smaller test samples are required, placing more stringent limits on the allowable dimensional variations in those samples. Without reliable mmWave materials data, manufacturers are forced to extrapolate materials data from low frequencies to high frequencies, which can lead to mistakes that have potentially devastating costs.

Technical Needs, Gaps and Solutions

The technology issues surrounding materials characterization, the associated needs, technology status of those needs, as well as gaps and challenges to overcome, are summarized below. The time period considered is from 2023 to 2033.

Technology Status Legend

For each need, the status of today’s technology is indicated by label and color as follows:

In-table color + label key	Description of Technology Status
Solutions not known	Solutions not known at this time
Solutions need optimization	Current solutions need optimization
Solutions deployed or known	Solutions deployed or known today
Not determined	TBD

Table 1. mmWave Material Characterization Needs, Gaps, and Today’s Technology Status with Respect to Current and Future Needs

	ROADMAP TIMEFRAME
TECHNOLOGY ISSUE	TODAY (2023)	3 YEARS (2026)	5 YEARS (2028)	10 YEARS (2033)
CHARACTERIZATION FREQUENCY RANGE
NEED	Tools needed for 28-110 GHz	Tools needed at D-band (110-170 GHz)	Tools needed G-band (220-350 GHz)	Tools needed for >500 GHz
CURRENT TECHNOLOGY STATUS	Solutions deployed or known	Solutions need optimization	Solutions not known
GAP	Limited tool availability for high frequencies	Few tool options	Robustness and availability
CHALLENGE	High frequencies place burden on mechanical precision of equipment		Methods still in academic space
CHALLENGE	-	Supporting equipment is expensive (i.e., 100 GHz VNA)	Expensive supporting equipment
CHALLENGE	High equipment cost
ANISOTROPIC MATERIAL CHARACTERIZATION
NEED	Average in-plane or out-of-plane characterizations dominant, little capability for separating in-plane electromagnetic (EM) components	Separate in-plane and out-of-plane EM components (Ex,Ey,Ez) using the same test sample, but different tools; or same tool/different samples		Separate in-plane and out-of-plane EM components (Ex,Ey,Ez) using the same test sample and same tool set
CURRENT TECHNOLOGY STATUS	Solutions deployed or known	Solutions need optimization		Solutions not known
GAP	None	Few new fundamentally different methods are in development
CHALLENGE	High frequencies limit flexibility of sample dimensions with known methods
	Sample geometry incompatibilities across equipment sets
VARIATION IN SAMPLE THICKNESS
NEED	Results generally have 1:1 error with thickness variation	Results less sensitive to thickness variation	Sources of uncertainty not dominated by thickness variation	Sources of uncertainty with only minimal dependence on thickness variation
CURRENT TECHNOLOGY STATUS	Solutions deployed or known	Solutions not known	Solutions not known
GAP	Good, uniformly thick samples are not typically available for materials of interest
CHALLENGE	Difficult to get good characterizations if samples are not uniformly thick	Methods with less sensitivity to sample thickness uniformity not known for samples in thin sheet format	No known methods that are good for low-loss materials without sensitivity to sample thickness uniformity
TOOL COMPATIBILITY WITH SAMPLE THICKNESS LIMITATIONS (FOR INDUSTRIAL MATERIAL)
NEED	(50 um-150 um for Er<5, Freq<60GHz)	Expanded capability thickness range – up to 500 um, 100 GHz	Expanded capability thickness range 1 mm or more, and down to 10 um, up to 100 GHz	Expanded capability thickness range above 1 mm and below 10 um
CURRENT TECHNOLOGY STATUS	Solutions deployed or known	Solutions not known	Solutions not known
GAP	None	Techniques in use rely on and assume specific mode structures that limit thickness of samples at higher frequencies
CHALLENGE	Some industries have samples that are not available in thin sheets	Evolutionary methods may be limited to thin samples, may need new methods	Input materials will have significant thickness variation
SAMPLE THICKNESS MEASUREMENT ACCURACY
NEED	Single point thickness +- 3 um reproducibility using micrometers or drop gauges	Accurate non-contact optical methods + 3D point cloud measurements	Easy, accurate, non-contact methods	Easy, inexpensive accurate, non-contact methods
CURRENT TECHNOLOGY STATUS	Solutions deployed or known	Solutions need optimization	Solutions not known
GAP	Thickness assessment directly impacts final results, errors in thickness result in significant errors in final result
CHALLENGE	Samples may have spatial thickness variation that needs to be comprehended	Optical thickness systems are uncommon and expensive
CHALLENGE	Samples may be compliant and compress under micrometer use
CHALLENGE	Optical methods may be sensitive to sample appearance
THERMAL CHARACTERIZATION
NEED	Characterization at room temperature	Easy characterization over temperature	Increased number of techniques capable of temperature characterization	Extended characterization range <100mK, >300C for space and other applications
CURRENT TECHNOLOGY STATUS	Solutions deployed or known	Solutions need optimization
GAP	None	Few methods/equipment sets compatible with elevated temperature or environmental conditions.
CHALLENGE	Operator interaction with levers and equipment fixtures at elevated temperatures not compatible with safety requirements			Instrumentation compatibility with extreme temperatures
CHALLENGE	Only fast techniques are able to measure humidity soaked samples as thin samples can dry out in several minutes
HUMIDITY CHARACTERIZATION
NEED	Measurement of humidity-soaked samples possible but limited to few equipment types	Humidity characterization capability more routine	Tools widely compatible with humidity characterization	No change in need
CURRENT TECHNOLOGY STATUS	Solutions need optimization
GAP	Humidity characterization is difficult
CHALLENGE	Thin samples can absorb and release environmental moisture quickly - over the span of a few minutes
STANDARD REFERENCE MATERIAL
NEED	No standard	Below 110 GHz discrete frequencies in-plane	Below 110 GHz in-plane and out-of-plane
CURRENT TECHNOLOGY STATUS	Solutions not known
GAP	No traceable standards exist
CHALLENGE	Industry needs to consume SRMs (standard reference materials) to be viable for national labs to support funding
OUT OF PLANE QUALITY CONTROL MEASUREMENTS
NEED	Limited accuracy or availability of methods fast enough for quality control.	Faster, QC compatible methods that retain accuracy	Equipment sets designed for fast, accurate QC testing
CURRENT TECHNOLOGY STATUS	Solutions need optimization
GAP	Many QC measurements need out of plane characterizations	Many QC measurements need real-time, out of plane characterizations
CHALLENGE	QC requires rapid, easy and accurate results, ideally non-destructive to the test sample

Approaches to address Needs, Gaps and Challenges

Table 2 considers approaches to address the above needs and challenges. The evolution of these is projected out over a 10-year timeframe using technology readiness levels.

Table 3 presents a range of material characterization techniques relevant for mmWave frequencies. For further information please see the iNEMI project reports [1], [2].

In-table color key	Range of technology readiness levels	Description
2	TRL: 1 to 4	Levels involving research
6	TRL: 5 to 7	Levels involving development
9	TRL: 8 to 9	Levels involving deployment

Table 2. mmWave Materials Characterization Solutions

		EXPECTED TRL LEVEL
TECHNOLOGY ISSUE	POTENTIAL SOLUTIONS	TODAY (2023)	3 YEARS (2026)	5 YEARS (2028)	10 YEARS (2032)
ANISOTROPIC MATERIAL CHARACTERIZATION	Develop new and disruptive methods for material characterization	3	4	5	9
ANISOTROPIC MATERIAL CHARACTERIZATION	Converge on common sample geometry	3	5	7	9
SAMPLE THICKNESS VARIATION	Hand-picked samples	9	9	9	9
	Use of mechanical methods to modify existing samples to improve thickness uniformity	4	4	4	4
	Develop new methods with less sensitivity to thickness variation	1	2	3	5
TOOL COMPATIBILITY WITH SAMPLE THICKNESS LIMITATIONS	Use of mechanical thinning of samples	4	4	4	4
	Improve mathematical modeling to enable thicker samples	1	2	3	3
	Develop new cavity types or methods that inherently support thicker samples	1	3	5	8
SAMPLE THICKNESS MEASUREMENT ACCURACY	Map of points across samples	5	7	8	9
	Use of cost-effective optical methods	1	4	5	7
	Use of available traceable standards of different material types (with differing optical properties)	5	6	6	6
THERMAL CHARACTERIZATION	Design fixturing with thermal considerations and operator safety in mind (120^OC)	5	6	7	7
THERMAL CHARACTERIZATION	Improve processing and measurement speed	6	8	9	9
HUMIDITY CHARACTERIZATION	Utilize fast measurement techniques to allow humidity soaked samples to be rapidly measured	4	7	8	9
CHARACTERIZATION FREQUENCY RANGE FOR 60-300GHz	Develop new tools and techniques for 60-300 GHz	2	3	4	8
	Use of advanced manufacturing techniques	5	5	5	5
	Use of customized equipment rather than utilizing an overly capable VNA - Example Q meter	2	5	7	8
STANDARD REFERENCE MATERIAL	Encourage equipment suppliers to supply traceable standards as tool references	4	8	9	9
OUT OF PLANE QUALITY CONTROL MEASUREMENTS	Design custom methods to meet specific QC needs	2	4	5	8

Table 3. Comparison of common material measurement techniques

Technique [1],[2]	Frequency	Features	2023 TRL	2028 TRL	Comments
Split-post dielectric resonator (SPDR)	Discrete frequency points from 1 GHz up to 15 GHz	High measurement precision Single frequency points Easy to use Insensitive to many user errors Typically in-plane component of permittivity Typically extrapolated to 5G mmWaves Typical sample thicknesses less than 1 mm IEC 61189-2-721:2015 [3] Some compatible with thermal chambers Fast measurement time allows sample characterization of moisture absorbed samples Cannot separate in plane X vs Y permittivity	NA	NA	Current (2023) TRL is 8, but limited to below 20 GHz. Therefore, not applicable for mmWave.
Split-cylinder resonator (SCR)	Discrete frequency points from 10 GHz up to 80 GHz	High measurement precision Single frequency points Insensitive to many user errors Typically interpolated to 5G mmWaves Typically averages in-plane components of permittivity Typical sample thicknesses around 100 um Not currently compatible with thermal characterization Fast measurement time allows sample characterization of moisture absorbed samples Cannot separate in plane X vs Y permittivity Simple analytical theory, which is advantageous for error propagation Traceability path only requires physical dimensioning the diameter and length of the cavity halves IPC-TM-650 2.5.5.13 [4]	8	9	Identified as a first candidate method for SRM (standard reference material) development below 110 GHz
Balanced-type circular disk resonator (BCDR)	Multiple discrete frequency points from 10 GHz up to 120 GHz	High measurement precision Multiple frequency points Requires full 2-port calibration (mechanical to 110 GHz or electrical to 67 GHz) Typically average out-of-plane component of permittivity Not practical with thermal or environmental characterizations Typical sample thicknesses less than 1 mm Requires two nominally identical samples with identical form factors and thicknesses Complex analytical theory compared to SCR, which makes analytical error propagation challenging Traceability path only requires physical dimensioning of both samples, and all cavity components IEC 63185 [6]	6	8
Fabry-Perot open resonator (FPOR)	Multiple discrete frequencies between 20 GHz up to 120 GHz	High measurement precision Multiple frequency points Can be sensitive to user errors Typically in-plane component of permittivity, with potential for separation of in plane X vs Y components Not compatible with thermal or environmental characterizations Complex analytical theory compared to SCR and BCDR, which makes analytical error propagation challenging JIS R1660-2 [6]	7	9
Co-planar waveguide (CPW)	Continuous frequencies between 1 kHz up to 1000 GHz	Moderate measurement precision User-selected frequency points Can be hard to use Can be sensitive to user errors Requires microfabrication and multiple measurements Requires on-wafer calibration to a known reference impedance Typically a convolution of in-plane and out-of-plane components of permittivity for materials whose thicknesses are comparable to the ground to ground spacing Not practical for development of a traceability path	4	7
Microstrip line charaterization		Moderate measurement precision User-selected frequency points Can be hard to use Can be sensitive to user errors Requires microfabrication and multiple measurements Requires on-wafer calibration to a known reference impedance Requires vias Typically the out-of-plane components of permittivity for materials whose thicknesses are comparable to the ground to ground spacing	4	6
Bereskin stripline transmission line method	Continuous frequencies between 1 GHz up to 20 GHz	Qualitative measurement precision User-selected frequency points at continuous frequencies between 1 kHz up to 20 GHz Recommended for quality assurance rather than permittivity measurements	NA	NA	Current (2023) TRL is 7, but lacks the necessary precision.
On-chip resonators and microstrip-ring resonators (MRR)	Discrete frequencies between 1 GHz up to 50 GHz	Moderate measurement precision. Can be hard to use Can be sensitive to user errors Requires microfabrication and multiple measurements Requires on-wafer calibration to a known reference impedance	3	5
Time-domain materials characterization techniques	15 GHz - 150 GHz	Limited measurement precision User-selected frequency points within antenna measurement bands Sensitive to many user errors Requires flat samples Compatible with thermal characterization. Fast measurement time allows sample characterization of moisture absorbed samples Can separate in plane X vs Y permittivity Complex analytical theory, which makes analytical error propagation challenging	3	5	A few companies advertise THz systems – further assessment needed.
Free-space materials characterization techniques	Up to 1.1 THz	Limited measurement precision User-selected frequency points within antenna measurement bands Sensitive to many user errors Requires surface flatness to be much less than the wavelength at the highest measurement frequency Compatible with thermal characterization. Fast measurement time allows sample characterization of moisture absorbed samples Can separate in plane X vs Y permittivity Complex analytical theory, which makes analytical error propagation challenging	3	4

(*) TRL = technology readiness level, a readiness metric between 1 and 9

References

iNEMI, “iNEMI 5G Project Report 1: Benchmark Current Industry Best Practices for Low Loss Measurements”, Nov. 2020.
iNEMI, “iNEMI 5G Project Report 2: Benchmark Emerging Industry Best Practices for Low Loss Measurements”, Nov. 2020.
IEC TC 91 - Electronics assembly technology, “Test methods for electrical materials, printed boards and other interconnection structures and assemblies - Part 2-721: Test methods for materials for interconnection structures - Measurement of relative permittivity and loss tangent for copper clad laminate at microwave frequency using a split post dielectric resonator,” IEC 61189-2-721:2015, 29 April 2015.
IPC High Frequency Resonator Test Method Task Group (D-24c), “Relative Permittivity and Loss Tangent Using a Split-Cylinder Resonator“, IPC-TM-650 2.5.5.13, January 2017.
IEC TC 46/SC 46F - RF and microwave passive components, “Measurement of the complex permittivity for low-loss dielectric substrates balanced-type circular disk resonator method,” IEC 63185:2020, 8 December 2020.
JIS, “Measurement method for dielectric properties of fine ceramics in millimeter wave frequency range - Part 2: Open resonator method”, JIS R 1660-2-2004 (R 2008)(R 2013), 20 March 2004.

As part of the 5G/6G MAESTRO project, work on this page is supported by the Office of Advanced Manufacturing in the National Institute of Standards and Technology (NIST), under the Federal Award ID Number 70NANB22H050

Return to topic overview