IEEE TRANSACTIONS ON MICROWAVE THEORY AND TECHNIQUES, VOL. 60, NO.

12, DECEMBER 2012 1

High-Efficiency Harmonically-Terminated Diode

and Transistor Rectifiers
Michael Roberg, Student Member, IEEE, Tibault Reveyrand, Member, IEEE,
Ignacio Ramos, Student Member, IEEE, Erez Falkenstein, Student Member, IEEE, and Zoya Popović, Fellow, IEEE

Abstract—This paper presents a theoretical analysis of provided in [4]. These early microwave rectifiers were aimed
harmonically-terminated high-efficiency power rectifiers and ex- at extracting information rather than extracting DC power.
perimental validation on a class-C single Schottky-diode rectifier The first published application of microwave rectifiers for
and a class-F−1 GaN transistor rectifier. The theory is based
on a Fourier analysis of current and voltage waveforms which extraction of DC power was performed in the 1960’s using
arise across the rectifying element when different harmonic diode-based rectifiers [5], [6], [7], [8].
terminations are presented at its terminals. An analogy to Renewed interest in free-space power transmission occurred
harmonically-terminated power amplifier theory is discussed. in the early 1970’s. An interesting microwave rectifier for
From the analysis, one can obtain an optimal value for the DC
load given the RF circuit design. An upper limit on rectifier production of DC power or low-frequency AC power called the
efficiency is derived for each case as a function of the device on- Cyclotron-Wave Rectifier was introduced in [9], [10]. William
resistance. Measured results from fundamental frequency source- C. Brown of Raytheon, one of the original researchers in the
pull measurement of a Schottky diode rectifier with short-circuit field, continued publishing diode-based rectifier work and in-
terminations at the second and third harmonic are presented. troduced the term “rectenna” for a receiving antenna integrated
A maximal device rectification efficiency of 72.8% at 2.45 GHz
matches the theoretical prediction. A 2.14 GHz GaN pHEMT with a rectifier [11], [12], [13]. Around the same time, power
rectifier is designed based on a class-F−1 power amplifier. The combining for an array of microwave power rectifiers was
gate of the transistor is terminated in an optimal impedance discussed in [14], in which the authors inadvertently graze
for self-synchronous rectification. Measurements of conversion the topic of harmonically terminated rectifiers, of which they
efficiency and output DC voltage for varying gate RF impedance, seem to hint at a class-F rectifier.
DC load and gate bias are shown with varying input RF power at
the drain. The rectifier demonstrates an efficiency of 85% for a
10 W input RF power at the transistor drain, with a DC voltage A number of diode-based rectifiers have been demonstrated,
of 30 V across a 98 Ω resistor. many integrated with antennas, with a good comparison pre-
Index Terms—harmonic terminations, high efficiency power sented in [15] and in earlier works focusing on low-power
amplifiers, load pull, microwave rectifiers, nonlinear analysis, rectification [16], [17]. Additional applications where rectifier
time-domain measurements efficiency is important include microwave power recycling
[18], and DC-DC converters with extremely high frequency
I. I NTRODUCTION switching [19], [20]. In many of the reported microwave rec-
tifiers, filtering of the harmonics at both the input and output

T HE first RF rectifiers were demonstrated in experiments

and patents in the 1890’s by Nikola Tesla in wireless
power transmission for lighting applications and the method
has been investigated, e.g [21],[22], mainly to reduce re-
radiated harmonic power. To date, very few transistor rectifier
circuits have been demonstrated, most at frequencies at least
of obtaining direct from alternating current [1]. The main three times lower than in this work. A UHF synchronous
application of microwave power rectifiers in the early 1900’s transistor class-E rectifier at 700 MHz is shown to achieve
was in signal detection where crystals, vacuum tubes or 85% efficiency with 58 mW of output power in [20], and
diodes served as the nonlinear element [2], [3]. An excellent the same authors discuss a class-E2 10-W DC-DC converter
discussion of the early history of microwave detectors is with a synchrounous transistor rectifier at 780 MHz with 72%
Manuscript received July 10, 2012. This paper is an expanded paper from efficiency [23]. This design is scaled from 0.5-MHz syn-
the IEEE MTT-S Int. Microwave Symposium held on June 17-22, 2012 in chronous rectifier designs demonstrated in [24] and requires
Montreal, Canada. This work was funded in part by the Office of Naval an additional synchronized gate RF input signal.
Research under the Defense Advanced Research Projects Agency (DARPA)
Microscale Power Conversion (MPC) Program under Grant N00014-11-1-
0931, and in part by the Advanced Research Projects Agency-Energy (ARPA-
E), U.S. Department of Energy, under Award Number DE-AR0000216. Harmonic terminations are commonly applied to increase
M. Roberg is with TriQuint Semiconductor, 500 West Renner Road efficiency in power amplifiers (PAs). The transistor nonlinear-
Richardson, TX 75080 USA (e-mail: michael.roberg@tqs.com). ities generate harmonic content at the output, and in a number
T. Reveyrand is with the XLIM Laboratory, UMR 7252, University of
Limoges, 87060 Limoges, France (e-mail: tibault.reveyrand@xlim.fr). of high-efficiency amplifier classes, specific harmonic termi-
I. Ramos and Z. Popovic are with the Department of Electrical, Computer nations are used to shape the current and voltage waveforms.
and Energy Engineering, University of Colorado, Boulder, CO, 80309-0425 In reduced conduction angle PAs (classes A, AB, B and C),
USA (e-mail: ignacio.ramos@colorado.edu; zoya.popovic@colorado.edu).
E. Falkenstein is with Qualcomm Inc., 6150 Lookout Road Boulder, CO all harmonics are shorted at the virtual drain reference plane
80301 USA (e-mail: erez.falkenstein@gmail.com). in the transistor. Other PA modes of operation specify open or
IEEE TRANSACTIONS ON MICROWAVE THEORY AND TECHNIQUES, VOL. 60, NO. 12, DECEMBER 2012 2

Non-linear
Element v Lc A. Class-C Rectifier Analysis
RS Cb 1
0
Matching 0
1
Network R (v) In Fig. 1, a sinusoidal microwave power source with voltage
RDC magnitude Vs and impedance Rs drives the rectifying element
VS (f0) i
of resistance R(v) above. The DC load seen by the rectifying
element is RDC while the load at the fundamental frequency
Fig. 1. Microwave rectifier circuit diagram. An ideal blocking capacitor Cb f0 and successive harmonics is set by the matching network.
provides DC isolation between the microwave source and rectifying element. Assume the matching circuit presents Rs (f0 ) to the rectifying
An ideal choke inductor Lc isolates the DC load RDC from RF power.
element with all subsequent harmonics terminated in short
circuits. This is equivalent to the harmonic terminations for
a canonical reduced conduction angle power amplifier. This
short harmonic terminations for various harmonics [25], [26], class is useful for Schottky diode rectifiers because these
[27]. A general analysis for arbitrary complex terminations diodes have nonlinear junction capacitance. Short-circuiting
of harmonics have recently been derived in [28], including the harmonics fixes the harmonic terminations at the intrinsic
a sensitivity analysis to harmonic termination impedances. diode by shorting this junction capacitance.
In a rectifier, the nonlinear rectifying element also generates When the incident RF voltage at the ideal rectifier swings
currents and voltages at the harmonics of the input frequency, negative, it is clipped at zero given (1). The enforced harmonic
and although in this case the output is at DC, the efficiency of terminations force the voltage waveform to contain only a
the rectifier can be modified by terminating the harmonics. In DC and fundamental frequency component. Therefore, a DC
[29] the harmonic termination concept for improving rectifier component must be produced by the rectifying element such
efficiency is applied to a class-C diode rectifier integrated that the voltage waveform maintains its sinusoidal nature. The
with a dual-polarized patch antenna for a wireless powering voltage across the rectifying element can now be expressed as
application.
In this paper, we identify the similarity between power v(θ) = VDC + V (f0 ) sin(θ) (3)
rectifiers and power amplifiers, showing that many of the
efficiency improvement techniques developed for power am- where V (f0 ) is the fundamental frequency component of
plifiers may be practically directly applied to power rectifiers. the voltage across the rectifying element, VDC is the DC
Particularly, the impact which harmonic terminations have on component, VDC = V (f0 ) and θ = 2πf0 t. The current
the rectification efficiency is addressed. A general rectifier waveform contains infinite frequency components, and can be
analysis approach is presented in Section II, and several classes written as
of microwave power rectifiers are introduced, focusing on
class-C and F−1 modes, which are experimentally validated  
3π
in Sections III and IV. i(θ) = 2πIDC δ θ − − 2nπ , n = 0, 1, ..., ∞ (4)
2
II. H ARMONICALLY-T ERMINATED P OWER R ECTIFIER where IDC is the DC current and δ(θ) is the Dirac delta
A NALYSIS function. When all available input power Pin is delivered to the
Consider the microwave rectifier shown in Fig. 1. A sinu- rectifier, the fundamental frequency component of the current
soidal microwave power source with voltage magnitude Vs and through the rectifying element I(f0 ) is
impedance Rs drives the rectifying element having a resistance 2Pin
R(v) defined as I(f0 ) = (5)
V (f0 )
(
∞, v > 0 and, since there is no mechanism by which the rectifier itself
R(v) = (1)
0, v≤0 can dissipate power (Ron = 0 at this point), all of the
available input power must be dissipated in the DC load and
where v and i are the instantaneous voltage across and current the conversion efficiency is 100%. Therefore,
through the rectifying element, respectively. The rectifying
element depicted by R(v) in Fig. 1 can in general be any Pin = VDC IDC (6)
nonlinear device that acts as a switch, such as a diode or a
transistor. When a non-zero on-resistance Ron and non-zero Substituting in (5) and rearranging gives the expression of
threshold voltage Vtr are taken into account, the resistance of the current at the fundamental input frequency and the DC
the rectifying element is given by rectified current, which is I(f0 ) = 2IDC . When all available
( input power is delivered to the rectifier, the RF-DC conversion
∞, v > −Vtr efficiency is 100% because the rectifying element is ideal and
R(v) = (2)
Ron , v ≤ −Vtr cannot dissipate power itself. In order for all available input
power to be delivered to the rectifier, it is straightforward to
The analysis of different classes of power rectifiers will next show that the DC load must be set relative to the fundamental
be analyzed based on the harmonic terminations presented frequency load as
to the rectifying element, and independent of the physical
nonlinear device which performs the rectification. RDC = 2Rs (f0 ) (7)
IEEE TRANSACTIONS ON MICROWAVE THEORY AND TECHNIQUES, VOL. 60, NO. 12, DECEMBER 2012 3

1.0
B. Class-F−1 Rectifier Analysis
v(θ) [V] 0.8 Consider again the rectifier circuit shown in Fig. 1 and
assume that all even harmonics are terminated in open circuits,
0.6
while all odd harmonics are terminated in short circuits. This
0.4
set of harmonic terminations is the same as for a class-F−1
amplifier, therefore this rectifier will be referred to as a class-
0.2 F−1 rectifier. The fundamental frequency component of the
i(θ) [A]

voltage across the rectifying device is given by

0.0
V (f0 ) = Vs (f0 ) (9)
-0.2
0 50 100 150 200 250 300 350 400 During the second half of the RF cycle, it is evident from
θ(degrees)
(1) that the voltage across the rectifying element must be zero.
Fig. 2. Ideal normalized voltage (dashed) and current (solid) waveforms This condition must be met through the addition of DC and
for reduced conduction angle half-wave rectifier. The waveforms have been only even harmonic voltage components, and therefore the
normalized to their peak values. voltage waveform is expressed as
(
2V (f0 ) sin θ, 0 ≤ θ < π
100 v(θ) = (10)
0, π ≤ θ < 2π
80 A Fourier expansion of (10) expresses the DC component
of the voltage waveform as
Efficiency (%)

60
2V (f0 )
VDC = (11)
π
40
In the first half of the RF cycle, the current through the
On Resistance = 0 rectifying element is zero, given (1). This condition is met
20
On Resistance = 4% of Rs(f0) through the addition of a DC current and odd harmonic current
On Resistance = 10% of Rs(f0) components. With the current direction as in Fig. 1, the DC
0 component of the current must be positive. Therefore, in the
0 0.5 1 1.5 2 2.5 3 3.5 4
RDC / Rs(f0)
first half of the RF cycle, the remaining harmonics must
sum to a constant value equivalent to the negative of the
DC component. Since the function which is the sum of the
Fig. 3. Simulated efficiency of reduced conduction angle half-wave rectifier
versus RDC /Rs (f0 ) for varying rectifier on-resistance.
remaining harmonics is odd, the second half of the RF cycle
must sum to the DC component, and the current is given by
(
0, 0≤θ<π
i(θ) = (12)
A harmonic balance simulation of an approximately ideal 2IDC , π ≤ θ < 2π
rectifier with short-circuited harmonic terminations was per-
formed in Microwave Office® using the SPICE diode model The DC component of the current waveform Fourier expansion
with no parasitics (PNIV) as the rectifying element. The device is found to be
π
temperature was set to 1◦ K to approximate an ideal switch. IDC = I(f0 ) (13)
4
The fundamental frequency excitation was set to 1 W at 1 GHz
The DC load consistent with (10) and (12) is given by
with the first 200 harmonics terminated in short-circuits. The
diode was presented with 50 Ω at the fundamental frequency 8V (f0 ) 8
RDC = 2
= 2 R(f0 ) (14)
and the DC load was swept from 5 Ω to 200 Ω. The simulated π I(f0 ) π
data is then normalized to generalize the simulation results. The conversion efficiency, defined as the ratio of the DC
The ideal time-domain current and voltage waveforms across power dissipated in the load resistor to the available funda-
the diode are shown in Fig. 2 with the RF-DC conversion mental frequency RF power, is evaluated as
efficiency as a function of RDC /Rs (f0 ) for varying rectifier
on-resistance shown in Fig. 3. It is clear that the mechanism
of operation in the ideal case agrees with the theory presented PDC 2VDC IDC 2 2 V (f0 ) π4 I(f0 )
η= = = π =1 (15)
above. The reduction in RF-DC conversion efficiency when P (f0 ) V (f0 )I(f0 ) V (f0 )I(f0 )
the DC load is not set according to (7) is due to impedance Therefore, the ideal half-wave rectifier converts all available
mismatch, and is given by RF power to DC power if the the DC loading resistance set to
 2 the value given in (14). The RF-DC conversion efficiency as a
RDC − 2Rs (f0 ) function of RDC /Rs (f0 ) was simulated in Microwave Office®
η =1− (8)
RDC + 2Rs (f0 ) for varying rectifier on-resistance and is shown in Fig. 4. The
harmonic balance settings were identical to those used for the
IEEE TRANSACTIONS ON MICROWAVE THEORY AND TECHNIQUES, VOL. 60, NO. 12, DECEMBER 2012 4

class-C rectifier above. The peak efficiency as a function of the minimum value, which would be −Vtr rather than zero.
on-resistance is higher than for the class-C rectifier, although The values of θ at which the transition between the conducting
the efficiency degrades more quickly when the non-ideal DC and non-conducting regions occurs are found to be
load is applied.  
The waveforms including parasitic on-resistance and thresh- θt1 = 2π − arcsin VVmax tr

  (18)
old voltage are next investigated assuming the rectifier θt2 = π + arcsin VVmaxtr

impedance from (2). The time domain voltage and current

waveforms are approximated as The DC and fundamental frequency values of the voltage
( and current waveforms can be found through a Fourier analysis
Vmax sin θ, v(θ) > −Vtr using the transition points in (18). The first Fourier coefficient
v(θ) = (16)
−Vtr − Imax Ron , v(θ) ≤ −Vtr of v(t) gives the DC component of the voltage, which can be
derived as
(  s
0, V (θ) > −Vtr  2
i(θ) = (17) 1  Vtr
VDC = 2Vmax 1 −
Imax , v(θ) ≤ −Vtr 2π Vmax
(19)
As an example, Fig. 5 shows the current and voltage wave-  
Vtr
 !
forms for a specific set of non-ideal parameters (Vtr = 0.7 V, − (Vtr + Imax Ron ) π − 2 arcsin
Vmax
Vmax = 20 V, Imax = 200 mA, and Ron = 5 Ω). When the
device is conducting current, it creates a voltage drop across The fundamental frequency voltage is found from V (f0 ) =
the on-resistance which is constant due to the constant current. av + jbv , where
If the on-resistance were zero, the only difference between the
waveform in (16) and the ideal voltage waveform would be Z 2π
1
av = v(θ) cos θdθ = 0 (20)
π 0
100

80 and bv can be reduced to

Efficiency (%)



60 1 Vtr πVmax
bv = Vmax arcsin +
π Vmax 2
40 s  2
 (21)
Vtr
On Resistance = 0 + (Vtr + 2Imax Ron ) 1 − 
20 Vmax
On Resistance = 4% of Rs(f0)
On Resistance = 10% of Rs(f0) Similarly, the DC component of the current waveform is
0
0 0.5 1 1.5 2 2.5 3 3.5 4
found to be
RDC / Rs(f0)   
Imax Vtr
IDC = π − 2 arcsin (22)
2π Vmax
Fig. 4. Simulated efficiency of class-F−1 rectifier versus RDC /Rs (f0 ) for
varying rectifier on-resistance. The fundamental frequency current i(t) = ai + jbi has ai = 0
and the coefficient bi can be shown to be equal to
1.2
s  2
2Imax Vtr
v(θ) [V]

1.0 bi = − 1− (23)
π Vmax
0.8 The input power at the fundamental frequency is found from
0.6
V (f0 )I ∗ (f0 )
 
Pin = < (24)
0.4 2
Substituting (21) and (23) into the above results in
i(θ) [A]

0.2

0.0
  
kImax Vtr πVmax
-0.2 Pin = Vmax arcsin +
0 50 100 150 200 250 300 350 400 π2 V 2
θ(degrees)  max (25)
+ (Vtr + 2Imax Ron ) k
Fig. 5. Non-ideal class-F−1 voltage (solid) and current (dashed) waveforms,
normalized to their peak respective values.
IEEE TRANSACTIONS ON MICROWAVE THEORY AND TECHNIQUES, VOL. 60, NO. 12, DECEMBER 2012 5

Z0 = 135 Ohms
where k is defined as

1.0
Swp Max

0.8
77.6433
s 2

6

Vtr

0.

0
1−

2.
k= (26) p1: Efficiency = 60.0 p5
Vmax 0.
4 p3 p6
p2: Efficiency = 62.5 3.
0
p2
Solving for Imax as a function of Pin when Ron is non-zero p3: Efficiency = 65.0 p4
4.
0

after some arithmetic results in two solutions, one of which is 0.2

p4: Efficiency = 67.5 p1 5.0

negative. The positive solution for the maximal current is p5: Efficiency = 70.0
10.0
√ p6: Efficiency = 77.6
α2 + 8π 2 Pin Ron − α

10.0
0.2

0.4

0.6

0.8

1.0

2.0

3.0

4.0
5.0
Imax = (27)

0
4Ron k
Fig. 6. Source-pull contours with available input power to the diode set to
   
with α = Vmax arcsin VVmax tr
+ π2 + k VVmax
tr
. 6 dBm. The impedance is referenced to the junction capacitance of the diode,
In the case where Ron is zero, (27) simplifies to therefore the lead inductance of the package has been compensated for. Setting
RDC to 1080 Ω was found to result in the optimal efficiency for this input
π2 power. The highest efficiency of 77.6% is obtained at Zp6 = (68 + j245)Ω
Imax = Pin     (28) with VDC =1.82 V.
Vtr
kVmax arcsin Vmax + π
2 + k VVmax
tr

Note that in the case of an ideal rectifying element, k = 1

and Vtr = 0, therefore 75
2πPin
Imax,ideal = (29)
Vmax 70
Now that Imax is fully expressed given known rectifier
parameters, VDC and IDC , V (f0 ) and I(f0 ) may be calcu- 65
lated, and from this, the DC load and the load at fundamental Efficiency (%)
frequency determined from the following expressions: 60 - VDC +

VDC 0 dBm
RDC = (30) 55 2 dBm
IDC 2f0 Short
f0 4 dBm
RFIN Diode
V (f0 ) bv 50
Match 6 dBm
R(f0 ) = − =− (31) 8 dBm
I(f0 ) bi
3f0 Short 10 dBm
The negative impedance in (31) indicates that power is deliv- 45
200 400 600 800 1000 1200
ered to the rectifying element and gives the impedance of the DC Load (Ω)
source delivering power to the rectifying element. The rectifier
efficiency is given by Fig. 7. RF-DC conversion efficiency versus DC load for fixed available
input powers with 0.6 dB matching network loss de-embedded. The maximum
PDC VDC IDC efficiency of 72.8% occurred at 8 dBm with RDC = 742 Ω and VDC =1.91 V,
η= = (32)
Pin Pin which is lower than the 1080 Ω found during source-pull. However, the effi-
ciency at 1080 Ω is 69.9% which is very close to the peak value.
C. Design example based on class-F−1 theory
To understand the usefulness of the presented theory, as-
sume the rectifying element has the following parameters: III. S CHOTTKY-D IODE C LASS -C R ECTIFIER
Vmax = 10 V, Ron = 5 Ω, Vtr = 0.7 V and P (f0 ) = 1 W.
First, (27) is used to calculate Imax = 456.7 mA. Next, the The Skyworks SMS7630 Schottky diode in the SC-79 pack-
DC voltage and current are evaluated using (19) and (22), age was selected for the half-wave rectifier. Source-pull was
respectively, to give VDC = 1.75 V and IDC = 218.2 mA. performed at 2.45 GHz with 0-10 dBm available input power
The fundamental frequency voltage and current Fourier co- for various DC loads in order to identify the combination
efficients are then calculated to be V (f0 ) = 6.896 V and of input power, fundamental load and DC load resulting in
I(f0 ) = −290 mA, respectively. The DC and fundamental highest efficiency. The best case occurred at 6 dBm input
frequency resistances are then calculated using (30) and (31) power, with the source-pull contours being shown in Fig. 6.
to be RDC = 8.02 Ω and R(f0 ) = 23.77 Ω, respectively. The The on-resistance of the SMS7630 is 20 Ω with the optimal
efficiency is then calculated using (32) to be η = 38.18 %. DC load of 1080 Ω. Therefore Ron is approximately 2% of
If the input power is selected as 0.1 W rather than 1 W, the RDC , which in theory is 4% of Rs (f0 ). From Fig. 3, a peak
resultant efficiency is 72.43 % instead. A specific rectification efficiency of 87% occurs with infinite harmonic terminations,
device will always have an approximate input drive level at therefore the achieved 77.6% is very reasonable considering
which it can be most efficient, just as with power transistors in only the 2nd and 3rd harmonics were explicitly terminated.
power amplifiers. To maximize efficiency, the goal is always to Measurements of a rectifier designed using the source-pull
minimize the amount of power dissipation in the on-resistance data show a maximum RF-DC conversion efficiency of 72.8%
of the rectifying element and maximize the power dissipated when matched to 50Ω, obtained after the 0.6 dB matching
in the DC load resistor. network loss is de-embedded. The fabricated rectifier and DC
IEEE TRANSACTIONS ON MICROWAVE THEORY AND TECHNIQUES, VOL. 60, NO. 12, DECEMBER 2012 6

40
load sweep measurements are shown in Fig. 7. Open circuit
shunt stubs are used to present short-circuit terminations at 35 90

Pout (dBm)
the second and third harmonic. A shunt capacitor is used for
30 75
presenting the fundamental frequency impedance to reduce
size and allow tunability. The reduction in efficiency relative 25 60

PAE (%)
to the source-pull measurements is due to the matching circuit

Gain (dB)
20 45
not presenting the ideal impedance found during source-pull.
The class-C rectifier can be applied to improving the ef- 15 30
ficiency of a wireless powering reception device as demon-
10 15
strated in [29] with a dual-linearly polarized patch rectenna,
with a rectifier circuit for each polarization. In this circuit, the 5 0
-15 -10 -5 0 5 10 15 20 25
first 5 harmonics are shorted and the impedances are validated Pin (dBm)
by calibrated measurements and are presented in [29].
Fig. 9. Large-signal measurements performed on the class-F−1 power am-
plifier at f0 = 2.14 GHz, VGS = −3.8 V and VDS = 28 V
IV. T RANSISTOR C LASS -F−1 R ECTIFIER
To prove experimentally the duality between harmonically
terminated PAs and rectifiers, a high-efficiency class-F−1 PA
was designed, measured first as an amplifier, and then as a B. Measurement setup
rectifier. In the rectifier measurements, RF power is input into The class-F−1 power amplifier described above is fully
the drain which is unbiased. The gate is terminated in a vari- characterized in large signal in a rectifier configuration with
able impedance and biased close to pinch-off. Measurements the setup shown in Fig. 10. The commercial time-domain large
of efficiency and DC voltage are performed in time domain signal measurement instrument is a VTD SWAP four-channel
as a function of input RF power, gate RF load, gate bias and receiver [31]. In order to acquire time domain waveforms
drain DC load. at the reference plane, an 8 error term model calibration
similar to the one performed for LSNA (Large Signal Network
A. Circuit design Analyzer) measurements is applied. After an absolute VNA-
like calibration [32], the RF voltage and current waveforms
A 2.14-GHz power amplifier, pictured on Fig. 8, is designed
at the input (V1 and I1) and at the output (V2 and I2) of the
using the Triquint TGF2023-02 GaN pHEMT [30]. Class F−1
DUT are measured at the coaxial reference plane. In this case,
harmonic terminations are implemented at the second and third
the RF input is the drain port of the PA, while the RF output
harmonic. The performance of the PA, illustrated in Fig. 9, was
is connected to the gate port. Thus, performing a load pull on
characterized at 2.14 GHz with a drain voltage bias of 28 V
this device consists of varying the load at f0 at the RF gate
and a bias current of 160 mA. The PA exhibits a PAE of 84%
port of the PA with a passive tuner. This kind of measurement
with an output power of 37.6 dBm and a gain of 15.7 dB under
is similar to large signal characterization of switch devices
3 dB compression. The same PA design was used for rectifier
recently reported in [33], [34]. The gate DC path is connected
measurements as shown in Fig. 10. The PA is connected to
to a power supply so the gate bias can be varied. The drain DC
an input RF source at the drain, with the drain supply dis-
bias is the output of the rectifier and is connected to a variable
connected. The gate terminal is biased, and connected to an
resistance RDC , and the DC voltage across it is measured with
impedance tuner, converting the two-port transistor PA to a
a voltmeter. The DC current is then found from the value of
one-port rectifier, corresponding to the generalized schematic
RDC from (30). During the measurement, several parameters
of Fig. 1.
are varied systematically: the RF load impedance applied at
the PA gate port Zg (f0 ) = Vg (f0 )/Ig (f0 ); the resistor in
the DC drain output RDC ; and the gate bias voltage VGS .
Gate Drain The conversion efficiency of the rectifier and the DC power
delivered at the drain output of the rectifier PDC = VDC · IDC
are measured as these parameters are varied, and as a function
of input power at the drain port Pin (f0 ).

C. Self-synchronous transistor rectifier results

The measurements of the rectifier are performed in self-
synchronous mode, i.e. there is no input RF power incident
externally into the gate port of the PA, unlike in previous tran-
sistor rectifier work [20], [24]. The following parameters are
Die Harmonic Terminations varied in order, while keeping the other parameters constant
Fig. 8. Photograph of the class-F−1 power amplifier, working at 2.14 GHz and sweeping the input RF power at the drain port, and the
and presented in [30]. results are described in the same order:
1) RF impedance at the gate, Zg ;
IEEE TRANSACTIONS ON MICROWAVE THEORY AND TECHNIQUES, VOL. 60, NO. 12, DECEMBER 2012 7

Drain Gate

R DC VGS Tuner
RF source
I1 I2
PA
V1 V2

Attenuators

Sub-sampling

Filter

SWAP ADC

Fig. 11. Time-domain waveforms measured at drain (a) and

gate (b) of the rectifier with VGS = −4.4 V, RDC = 98.5 Ω and
Fig. 10. Time-domain non-linear rectifier measurement block diagram. The
Zg (f0 ) = (230 + j10) Ω. The RF input power at the drain is swept from 10
SWAP [31] performs sampling of current and voltage and the calibration refers
the sampled quantities to the reference planes at the DUT. The drain output to 42 dBm, corresponding to the range of output power of the class-F−1 PA.
DC resistance RDC , the gate bias VGS and the gate RF impedance Zg are
varied as the input power at the drain is swept from 10 to 42 dBm.
90
Z g (f0 )
80
(a) (b)

Efficiency (%)
70
2) load resistance at drain bias output, RDC ; 60
50
3) gate DC bias, VGS .
40
The gate load-pull was performed to determine the optimum 30
impedance for maximum efficiency with a constant resistive 20
DC load of 98.5 Ω (nominally 100 Ω) and a constant transistor 10
10 15 20 25 30 35 40 45
gate bias in pinch-off of -4.4 V. The RF signal is coupled Pin (dBm)
40 6

from the drain to the gate matching network through the (c) (d)

Gate IDC (mA)

30
feedback capacitance Cgd , and thus the precise impedance
Drain VDC (V)

presented at the gate of the transistor is imperative to achieving 20

2
high efficiency. Fig. 11 shows the time-domain voltage and
10
current waveforms measured at the drain and gate RF port 0

of the amplifier when the RF input power at the drain port 0

10 15 20 25 30 35 40 45 10 15 20 25 30 35 40 45
Pin (dBm) Pin (dBm)
is swept from 11 dBm to 42 dBm. These values are chosen
because the rectifier in PA operation gives up to 42 dBm output Fig. 12. Conversion efficiency, gate DC current and drain DC voltage ver-
power. The feedback signal present at the gate allows for sus input power for several RF load impedance values presented at the gate.
the rectifier to operate in self-synchronous mode without any VGS = −4.4 V and RDC = 98.5 Ω. The green point on the Smith chart
corresponds to the highest efficiency point at Zg (f0 ) = (230 + j10) Ω.
additional control signal. Unlike in the synchronously driven
case where an external generator is connected to the gate, here
the impedance presented at the gate is always passive (inside
the Smith chart), keeping the device in a safe operating mode. After the optimal gate impedance for highest efficiency was
Measured RF-DC conversion efficiency is shown in Fig. 12 obtained, a power sweep for three different RDC values in the
for four different RF gate impedances. A maximal conversion drain output was obtained. From Fig. 13, a maximal efficiency
efficiency of 85% is achieved with a DC output voltage of 85% was measured for a DC resistive load of 98 Ω while
of 36 V and an input power at the drain of 42 dBm with an efficiency drop of 13% was observed for a DC load of 21 Ω
RDC = 98.5 Ω. This peak efficiency is for a RF gate load of with 40 dBm input power. As expected, the DC output voltage
around 230 Ω (green hexagon in the Smith chart in Fig. 12), decreases from a maximum 30 V for RDC = 98 Ω at 40 dBm
which is the highest impedance that was achievable with the input power, to a maximum of 13.4 V for RDC = 21 Ω with
specific tuner in the setup. For the low gate impedance (red the same input power. It is interesting to see how the input
triangle in the Smith chart), the efficiency is significantly impedance of the rectifier at the RF drain port approaches 50 Ω
lower. By observing the gate current (Fig. 12d), it can be seen as the input power increases, Fig. 14. This is expected, since
that for a low RF gate impedance, the gate diode turns on the PA was designed for maximal saturated power delivered
at around Pin = 25 dBm. Since the input power cannot be into a 50 Ω load. This again points to the similarities between
increased much beyond this point to avoid breakdown, this the same circuit operated as a power rectifier and a power
limits the DC voltage at the output to around 4 V. For the gate amplifier.
impedance with highest efficiency (green line with hexagon Finally, the effect of the gate bias VGS on the rectifier
symbol), the gate diode is off for input drain powers below efficiency, output voltage and input impedance was investi-
41 dBm, allowing for high DC voltage output. gated. The gate impedance in this case was set for highest
IEEE TRANSACTIONS ON MICROWAVE THEORY AND TECHNIQUES, VOL. 60, NO. 12, DECEMBER 2012 8

90 40 90 40
RDC (Ohm) : 21 58 98 80
80 35 VGS(V)

Efficiency (%)
70 30
Efficiency (%)

70 30

Drain VDC (V)

-3.5

Drain VDC (V)

60 25 60 -3.8

20 50 20 -4.1
50
40 -4.4
40 15
30 10 -4.7
30 10
20 -5
20 5
10 0
10 0 10 15 20 25 30 35 40 45
10 15 20 25 30 35 40 45 Pin (dBm)
Pin (dBm)
Fig. 15. Measured conversion efficiency and drain DC voltage versus input
Fig. 13. Conversion efficiency and drain DC output voltage versus power for several DC gate voltage biases. For this data, RDC = 58Ω and
input power for several DC drain resistor values. VGS = −4.4V and Zg (f0 ) = (230 + j10) Ω.
Zg (f0 ) = (230 + j10) Ω. The highest efficiency of 85% is obtained at
Pin =40 dBm with a VDC =30 V.

such as the relationship between output DC resistance and

impedance at the fundamental frequency at the rectifier input
which optimizes efficiency. The analysis also predicts the time-
domain waveforms at the terminals of the rectification element
Z in ( f0 ) and the efficiency as a function of on-resistance and DC
output resistance. Specific results are derived for class-C and
RDC (Ohm) : class-F−1 classes of operation, as they are defined for power
21 58 98 amplifiers. These two cases are chosen for experimental vali-
dation with a 2.45 GHz diode and 2.14 GHz transistor rectifier,
respectively. It is straightforward to repeat the derivation for
other classes of operation, such as class-F as shown in detail
in [35].
The experimental results show that good agreement can be
Fig. 14. RF impedance at f0 measured at the input (drain port) ver- reached between theory and experiment with a Schottky-diode
sus input power for several DC drain resistor values. VGS = −4.4V and single-ended rectifier with finite class-C harmonic termina-
Zg (f0 ) = (230 + j10) Ω. tions, resulting in 72.8% efficiency for input power levels
in the mW range, intended for wireless power harvesting
detailed in [29], [15]. A GaN pHEMT class-F−1 power recti-
efficiency (230 Ω), and a DC load of 58 Ω was selected in order fier achieved 85% efficiency with 40 dBm input power across
to protect the transistor from high drain voltages that occur 98- Ω DC load with a DC output voltage VDC = 30V . The
for the 98 Ω load that corresponds to the highest efficiency. efficiency and output voltage of the self-synchronous rectifier
The measurements were performed for six different values are shown to depend on the input power at the drain, the
of gate bias VGS as shown in Fig. 15. With RDC = 58Ω, a impedance at the gate port and the DC load at the output
maximum efficiency of 83% was obtained with the transistor drain bias line, but not on the gate bias.
biased deeply into the pinch-off region with VGS = −4.4 V,
Time-domain large-signal measurements of a class-F−1
and a drop of only 3% was measured for VGS = −3.5 V.
power amplifier configured as a rectifier show that one can
Furthermore, the gate bias has a minimal impact on the output
accomplish the same rectifier efficiency as the amplifier drain
DC voltage or on the drain impedance.
efficiency in self-synchronous mode without external gate RF
drive. This is somewhat surprising, and to the best of our
V. C ONCLUSION knowledge, the first time this type of high-efficiency rectifier
In summary, this paper addresses high-efficiency power has been demonstrated.
rectifiers designed with harmonic terminations at the RF input,
in analogy to high-efficiency power amplifier design with
harmonic terminations at the output. The applications of such ACKNOWLEDGMENT
power rectifiers include wireless power beaming [7], recycling
power in high-power circuits [18] and ultra-fast switching The authors would like to thank Dr. David Root and Dr.
integrated DC-DC converters with no magnetics [19]. Jean-Pierre Teyssier at Agilent Technologies for the loan of the
The theory for an ideal rectification element is based on time-domain nonlinear measurement equipment and TriQuint
Fourier analysis and establishes the basic design parameters Semiconductor for the donation of the transistors.
IEEE TRANSACTIONS ON MICROWAVE THEORY AND TECHNIQUES, VOL. 60, NO. 12, DECEMBER 2012 9

R EFERENCES [24] M. Kazimierczuk and J. Jozwik, “Analysis and design of class e zero-
current-switching rectifier,” Circuits and Systems, IEEE Transactions on,
[1] T. C. Martin and N. Tesla, The inventions, researches and writing of vol. 37, no. 8, pp. 1000 –1009, aug 1990.
Nikola Tesla: with special reference to his work in polyphase currents [25] F. Raab, “Class-E, Class-C, and Class-F power amplifiers based upon a
and high potential lighting. Milwaukee, WI: Lee Engineering Co., finite number of harmonics,” Microwave Theory and Techniques, IEEE
1952. Transactions on, vol. 49, no. 8, pp. 1462 –1468, Aug 2001.
[2] E. Peterson and F. Llewellyn, “The operation of modulators from a [26] F. Raab, P. Asbeck, S. Cripps, P. Kenington, Z. Popovic, N. Pothecary,
physical viewpoint,” Proceedings of the Institute of Radio Engineers, J. Sevic, and N. Sokal, “Power amplifiers and transmitters for RF and
vol. 18, no. 1, pp. 38 – 48, jan. 1930. microwave,” Microwave Theory and Techniques, IEEE Transactions on,
[3] J. Nelson, “Some notes on grid circuit and diode rectification,” Radio vol. 50, no. 3, pp. 814 –826, Mar 2002.
Engineers, Proceedings of the Institute of, vol. 20, no. 6, pp. 989 –1003, [27] S. Kee, I. Aoki, A. Hajimiri, and D. Rutledge, “The class-e/f family
june 1932. of zvs switching amplifiers,” Microwave Theory and Techniques, IEEE
[4] C. Edwards, “Microwave converters,” Proceedings of the IRE, vol. 35, Transactions on, vol. 51, no. 6, pp. 1677 – 1690, jun. 2003.
no. 11, pp. 1181 – 1191, nov. 1947. [28] M. Roberg and Z. Popovic, “Analysis of high-efficiency power amplifiers
[5] R. George and E. Sabbagh, “An efficient means of converting microwave with arbitrary output harmonic terminations,” Microwave Theory and
energy to dc using semiconductor diodes,” Proceedings of the IEEE, Techniques, IEEE Transactions on, vol. 59, no. 8, pp. 2037 –2048, aug.
vol. 51, no. 3, p. 530, march 1963. 2011.
[6] W. Brown, “Experiments in the transportation of energy by microwave [29] M. Roberg, E. Falkenstein, and Z. Popovic, “High-efficiency
beam,” in IRE International Convention Record, vol. 12, mar 1964, pp. harmonically-terminated rectifier for wireless powering applications,”
8 – 17. in IEEE MTT-S International Microwave Symposium Digest, Montréal,
[7] W. Brown, J. Mims, and N. Heenan, “An experimental microwave- 2012.
powered helicopter,” in IRE International Convention Record, vol. 13, [30] M. Roberg, J. Hoversten, and Z. Popović, “GaN HEMT PA with over
mar 1965, pp. 225 – 235. 84% power added efficiency,” Electronics Letters, vol. 46, no. 23, pp.
[8] O. E. Maynard, W. C. Brown, A. Edwards, J. T. Haley, G. Meltz, J. M. 1553 –1554, 11 2010.
Howell, and A. Nathan, “Microwave power transmission studies. volume [31] VTD, “Swap-x402 data sheet.” [Online]. Available: http://www.vtd-rf.
1: Executive summary,” NASA, Tech. Rep. 19760008506, 1975. com/pdf/productpresentation.pdf
[9] D. Watson, K. Talbot, and C. Johnson, “A cyclotron-wave microwave [32] J. Verspecht, “Calibration of a measurement system for high frequency
power converter,” Proceedings of the IEEE, vol. 53, no. 11, pp. 1797 – nonlinear devices,” Ph.D. dissertation, Vrije Universiteit Brussel (VUB),
1798, nov. 1965. 1995.
[10] D. Watson, R. Grow, and C. Johnson, “A cyclotron-wave rectifier for [33] G. Callet, J. Faraj, O. Jardel, C. Charbonniaud, J.-C. Jacquet,
s-band and x-band,” Electron Devices, IEEE Transactions on, vol. 18, T. Reveyrand, E. Morvan, S. Piotrowicz, J.-P. Teyssier, and R. Quéré,
no. 1, pp. 3 – 11, jan 1971. “A new nonlinear hemt model for algan/gan switch applications,”
[11] W. C. Brown, “High power, efficient, free-space microwave power International Journal of Microwave and Wireless Technologies, vol. 2,
transmission systems,” in Microwave Conference, 1973. 3rd European, no. Special Issue 3-4, pp. 283–291, 2010.
vol. 2, sept. 1973, pp. 1 –4. [34] J. Faraj, G. Callet, O. Jardel, A. El-Rafei, F. De Groote, R. Quéré, and
[12] W. Brown, “The design of large scale terres trial rectennas for low- J. Teyssier, “Time domain large signal characterization of self-biasing
cost production and erection,” in Microwave Symposium Digest, 1978 phenomena in switch-mode algan/gan hemts,” in 74th ARFTG Sympo-
IEEE-MTT-S International, june 1978, pp. 349 –351. sium, International Microwave Symposium Digest, Montréal, Canada,
[13] ——, “The history of power transmission by radio waves,” Microwave 2012.
Theory and Tech., IEEE Trans. on, vol. 32, no. 9, pp. 1230 – 1242, sep [35] M. Roberg, “Analysis & Design of Non-Linear Amplifiers for Efficient
1984. Microwave Transmitters,” Ph.D. dissertation, University of Colorado at
[14] R. Gutmann and J. Borrego, “Power combining in an array of microwave Boulder (CU-Boulder), 2012.
power rectifiers,” Microwave Theory and Techniques, IEEE Transactions
on, vol. 27, no. 12, pp. 958 – 968, dec 1979.
[15] E. Falkenstein, M. Roberg, and Z. Popovic, “Low-power wireless power
delivery,” Microwave Theory and Techniques, IEEE Transactions on,
vol. 60, no. 7, pp. 2277 –2286, july 2012.
[16] J. Hagerty, F. Helmbrecht, W. McCalpin, R. Zane, and Z. Popovic, Michael Roberg (S’09) received the B.S.E.E degree
“Recycling ambient microwave energy with broad-band rectenna arrays,” from Bucknell University, Lewisburg, PA, in 2003,
Microwave Theory and Tech., IEEE Trans. on, vol. 52, no. 3, pp. 1014 the M.S.E.E. degree from the University of Pennsyl-
– 1024, march 2004. vania, Philadelphia, in 2006, and the Ph.D. degree
[17] E. Falkenstein, “Characterization and design of a low-power wireless from the University of Colorado at Boulder in 2012.
power delivery system,” Ph.D. dissertation, University of Colorado at From 2003 to 2009, he was an Engineer with Lock-
Boulder, Boulder, CO, 2011. heed Martin–MS2, Moorestown, NJ, where he was
[18] X. Zhang, L. Larson, P. Asbeck, and R. Langridge, “Analysis of power involved with advanced phased-array radar systems.
recycling techniques for rf and microwave outphasing power amplifiers,” His current research interests include high efficiency
Circuits and Systems II: Analog and Digital Signal Processing, IEEE microwave PA theory and design, microwave power
Trans. on, vol. 49, no. 5, pp. 312 – 320, may 2002. rectifiers, MMIC design, and high-efficiency radar
[19] S. Djukic, D. Maksimovic, and Z. Popovic, “A planar 4.5-ghz dc-dc and communication system transmitters. He is currently employed by TriQuint
power converter,” Microwave Theory and Tech., IEEE Trans. on, vol. 47, Semiconductor - Defense Products and Foundry Services in Richardson, TX
no. 8, pp. 1457 –1460, aug 1999. working on wideband high efficiency GaN MMIC PA design.
[20] M. Nieves Ruiz, R. Marante, and J. A. Garcia, “A class e synchronous
rectifier based on an e-phemt device for wireless powering applications,”
in IEEE MTT-S International Microwave Symposium Digest, Montréal,
2012.
[21] S. Imai, S. Tamaru, K. Fujimori, M. Sanagi, and S. Nogi, “Efficiency Tibault Reveyrand (M’07) received the Ph.D. de-
and harmonics generation in microwave to dc conversion circuits of gree from the University of Limoges, France, in
half-wave and full-wave rectifier types,” in Microwave Workshop Series 2002. From 2002 to 2004, he was a Post-Doctoral
on Innovative Wireless Power Transmission: Technologies, Systems, and Scientist with CNES (French Space Agency). In
Applications (IMWS), 2011 IEEE MTT-S International, 2011, pp. 15–18. 2005, he became a CNRS engineer at XLIM. His
[22] H. Takhedmit, B. Merabet, L. Cirio, B. Allard, F. Costa, C. Vollaire, research interests include the characterization and
and O. Picon, “A 2.45-ghz low cost and efficient rectenna,” in Antennas modeling of RF and microwave nonlinear compo-
and Propagation (EuCAP), 2010 Proceedings of the Fourth European nents and devices. Dr. Reveyrand was the recipient
Conference on, april 2010, pp. 1 –5. of the 2002 European GaAs Best Paper Award and
[23] R. Marante, M. N. Ruiz, L. Rizo, L. Cabria, and J. A. Garcia, “A uhf is a member of the IEEE MTT-11 ”Microwave
class e2 dc/dc converter using gan hemts,” in Microwave Symposium Measurements” Technical Committee.
Digest (MTT), 2012 IEEE MTT-S International, june 2012, pp. 1 –3.
IEEE TRANSACTIONS ON MICROWAVE THEORY AND TECHNIQUES, VOL. 60, NO. 12, DECEMBER 2012 10

Ignacio Ramos (S’12) received the B.S. degree in

electrical engineering from the University of Illi-
nois at Chicago in 2009, and is currently working
toward the Ph.D. degree at the University of Col-
orado at Boulder. From 2009 to 2011, he was with
the Power and Electronic Systems Department at
Raytheon IDS, Sudbury, MA. His research interests
include high-efficiency microwave power amplifiers,
microwave DC/DC converters, radar systems, and
wireless power transmission.

Erez Avigdor Falkenstein (S’07), Haifa, Israel in

1979. He earned a “Handesaie” degree (associate
degree) in electronics from Amal Handesaim School
Hadera, Israel in 1999. From 1999 to 2003 he served
in the Israel Defense Force as part of a technological
unit. He has been at the University of Colorado
at Boulder 2004 – 2012. He received concurrent
MS/BS degrees in Electrical engineering 2010 and
a Ph.D 2012 from the University of Colorado at
Boulder. Since 2007 he has been involved with re-
search as part of the active antenna group. Research
emphasis: far field wireless powering for low power densities. Interests include
Antenna design and characterization, modeling and measurement of nonlinear
devices at microwave frequencies and power management. He is currently
employed at Qualcomm, Incorporated, Boulder, CO.

Zoya Popović (S’86-M’90-SM’99-F’02) received

the Dipl.Ing. degree from the University of Belgrade,
Belgrade, Serbia, Yugoslavia, in 1985, and the Ph.D.
degree from the California Institute of Technology,
Pasadena, in 1990. Since 1990, she has been with
the University of Colorado at Boulder, where she
is currently a Distinguished Professor and holds
the Hudson Moore Jr. Chair with the Department
of Electrical, Computer and Energy Engineering.
In 2001, she was a Visiting Professor with the
Technical University of Munich, Munich, Germany.
Since 1991, she has graduated 44 Ph.D. students. Her research interests
include high-efficiency, low-noise, and broadband microwave and millimeter-
wave circuits, quasi-optical millimeter-wave techniques, active antenna arrays,
and wireless powering for batteryless sensors. Prof. Popović was the recipient
of the 1993 and 2006 Microwave Prizes presented by the IEEE Microwave
Theory and Techniques Society (IEEE MTT-S) for the best journal papers
and the 1996 URSI Issac Koga Gold Medal. In 1997, Eta Kappa Nu students
chose her as a Professor of the Year. She was the recipient of a 2000
Humboldt Research Award for Senior U.S. Scientists of the German Alexander
von Humboldt Stiftung. She was elected a Foreign Member of the Serbian
Academy of Sciences and Arts in 2006. She was also the recipient of the 2001
Hewlett-Packard (HP)/American Society for Engineering Education (ASEE)
Terman Medal for combined teaching and research excellence.