High-performance small- and wide-

angle X-ray scattering (SAXS/WAXS)

experiments on a multi-functional
laboratory goniometer platform with easily
exchangeable X-ray modules
Cite as: Rev. Sci. Instrum. 89, 085115 (2018); https://doi.org/10.1063/1.5041949
Submitted: 28 May 2018 • Accepted: 31 July 2018 • Published Online: 22 August 2018

Joerg Bolze, Vladimir Kogan, Detlef Beckers, et al.

ARTICLES YOU MAY BE INTERESTED IN

Highly versatile laboratory X-ray scattering instrument enabling (nano-)material structure

analysis on multiple length scales by covering a scattering vector range of almost five
decades
Review of Scientific Instruments 90, 123103 (2019); https://doi.org/10.1063/1.5130061

Grazing incidence wide angle x-ray scattering at the wiggler beamline BW4 of HASYLAB
Review of Scientific Instruments 81, 105105 (2010); https://doi.org/10.1063/1.3488459

Block Copolymers—Designer Soft Materials

Physics Today 52, 32 (1999); https://doi.org/10.1063/1.882522

Rev. Sci. Instrum. 89, 085115 (2018); https://doi.org/10.1063/1.5041949 89, 085115

© 2018 Author(s).
 REVIEW OF SCIENTIFIC INSTRUMENTS 89, 085115 (2018)

High-performance small- and wide-angle X-ray scattering (SAXS/WAXS)

experiments on a multi-functional laboratory goniometer platform
with easily exchangeable X-ray modules
Joerg Bolze,1,a) Vladimir Kogan,2 Detlef Beckers,1 and Martijn Fransen1
1 Malvern Panalytical, Lelyweg 1, Almelo 7602 EA, The Netherlands
2 Dannalab B.V., Wethouder Beversstraat 185, Enschede 7543 BK, The Netherlands
(Received 28 May 2018; accepted 31 July 2018; published online 22 August 2018)
Small-angle X-ray scattering (SAXS) is a well-established, versatile technique for the analysis of
nanoscale structures and dimensions, e.g., in liquid dispersions, thin solid objects or powder samples.
When combined with wide-angle X-ray scattering (WAXS), complementary information about the
atomic structure can be obtained. SAXS experiments traditionally require dedicated instruments to
achieve the desired angular resolution, sensitivity, stability, and speed of measurement. Here we
demonstrate how a multi-functional laboratory goniometer platform, as widely being used for powder
X-ray diffraction and for a variety of related techniques, can be configured with pre-aligned X-ray
modules that enable advanced SAXS/WAXS experiments, without compromising the exceptional
versatility of the instrument. Line and point collimation setups, as well as quick and easy switching
between them, are readily possible. Key components are a detachable, evacuated beam path and a
high-resolution, low-noise hybrid pixel area detector, in combination with a hardware interface design
that allows to configure the instrument with different X-ray modules without the need for re-alignment.
Software for SAXS data reduction and analysis was developed. The good SAXS/WAXS performance
and the derived analytical results were verified on various test samples, such as gold nanoparticles,
colloidal silica, liposomes, dilute protein solutions, and solid polymer samples. It is believed that this
novel approach to SAXS/WAXS instrumentation will help to make this powerful structure analysis
technique more widely accessible and affordable for multi-user laboratories. © 2018 Author(s). All
article content, except where otherwise noted, is licensed under a Creative Commons Attribution (CC
BY) license (http://creativecommons.org/licenses/by/4.0/). https://doi.org/10.1063/1.5041949

I. INTRODUCTION laboratory SAXS instrument has the advantage of being readily

available when needed.
Small-angle X-ray scattering (SAXS) is an accepted and
From the required instrumentation point of view, the
powerful experimental technique that is widely used, e.g., in
SAXS technique is challenging. The intensity of X-rays that
materials science, in structural biology, and for soft matter
is scattered by a given sample must be measured as a func-
analysis.1–7 It enables the analysis of nanoscale structures
tion of the scattering angle 2θ, in very close proximity to the
and dimensions in a variety of materials. Typical applica-
direct beam, with a 2θ min < 0.1◦ [or qmin < 0.08 nm−1 , where q
tions include the characterization of nanoparticle systems,
denotes the scattering vector q = 4π λ −1 sin(θ), and λ denotes
colloids, surfactants, protein solutions, polymers, liquid crys-
the wavelength of radiation]. SAXS requires an intense yet
tals, nanocomposites, and porous materials. Compared to, e.g.,
tightly collimated X-ray beam, and any sources of parasitic
electron microscopy, the technique requires minimal sample
scattering must be minimized to achieve sufficient sensitivity
preparation, measurements of liquid samples can be performed
for weakly scattering samples. Furthermore, it is essential to
in situ, and ensemble-averaged results from a relatively large
ensure that the entire experimental setup is mechanically very
scattering volume are obtained. SAXS measurements are non-
stable. Also, the photon flux from the X-ray source must be
invasive, and with laboratory X-ray sources, radiation dam-
monitored and should have a good stability.
age is usually not observed. Small-angle neutron scattering
Virtually all synchrotron radiation facilities offer dedi-
(SANS) is very similar to SAXS and can be used for the same
cated beamlines that enable very advanced small- and wide-
applications.3–5,7,38,41–43 Complex problems are often solved
angle X-ray scattering experiments (SAXS/WAXS). There is
by the combination of SAXS and SANS. Particularly interest-
also a rapidly growing number of beamlines that are dedicated
ing with SANS is the possibility to perform contrast variation
to SAXS experiments on dilute protein solutions (bio-SAXS)
and to apply scattering contrast matching by using deuterated
for structural biology applications.5 For the efficient use of
samples.3–5 However, SANS measurements can only be per-
beam time at synchrotron radiation facilities and to get a beam
formed at a neutron research reactor or at a spallation source
time proposal accepted, a careful pre-selection of suitable sam-
where the available beam time is very limited. An in-house
ples and their pre-characterization in the lab is indispensable.
Dedicated laboratory instruments for SAXS/WAXS are com-
a) [email protected] mercially available but tend to require a lot of lab space and a

0034-6748/2018/89(8)/085115/13 89, 085115-1 © Author(s) 2018

085115-2 Bolze et al. Rev. Sci. Instrum. 89, 085115 (2018)

significant financial investment. Therefore, such instruments

are by far not as readily available in materials research labora-
tories as is the case, e.g., with powder X-ray diffractometers.
On the other hand, in-house SAXS/WAXS instruments enable
measurements once a sample is freshly prepared and in many
cases yield a high data quality that may already be sufficient
for the given purpose.
It was shown by Bolze et al.8 and Bota et al.9 that basic
SAXS and WAXS measurements are also possible on a typi-
cal, goniometer-based X-ray diffraction instrument by making
use of suitable X-ray optics and dedicated attachments. Such
an experimental approach could achieve a remarkably good
small-angle resolution with a 2θ min of 0.08◦ (qmin = 0.06 nm−1 ) FIG. 1. 2-circle goniometer with pre-aligned interfaces for the highly repro-
or even below,8,9 as well as a low level of background scat- ducible attachment of incident beam optics on the omega arm (A), a beam
path module or sample stage in the central position (B), and a detector on the
tering. To achieve this performance, a (0D) point detector is 2theta arm (C).
used behind a set of narrow slits. The detector-slit assembly is
mounted on the 2θ arm of the goniometer, and the scattering
intensities are collected sequentially by performing 2θ scans. vertical 2-circle, omega-2theta goniometer. The 2theta axis
Samples like nanopowders, porous materials, polymers, liq- coincides with the omega axis. Such a goniometer is com-
uid crystals, and liquid dispersions of inorganic nanoparticles monly used for powder X-ray diffraction measurements in
usually have a strong scattering signal, and good data can thus the classical Bragg-Brentano, theta-theta diffraction geome-
be obtained within reasonably short measurement times (typi- try, but it can also be configured for a variety of additional
cally 10–60 min). However, for more challenging samples that experimental techniques, such as microdiffraction, stress and
are very dilute and/or have a very low scattering contrast (e.g., texture measurements, thin film analysis, and computed X-ray
surfactant micelles, protein solutions, and other soft matter tomography. The high-precision and high-resolution goniome-
samples), such type of experimental setup is not feasible due ter enables a smallest step size of 0.0001◦ 2θ and has a radius
to insufficient sensitivity and long measurement times.10 By of 240 mm. A metal-ceramic, sealed Cu X-ray tube with a long
using an evacuated (or helium-filled) beam path in combina- fine focus (dimensions 12 × 0.4 mm2 ) is mounted on the omega
tion with an area or line detector, SAXS measurements can be arm of the goniometer. It is powered by a 4 kW high voltage
taken in a parallel (as opposed to a point-by-point) data collec- X-ray generator. The X-ray tube is operated at a voltage of
tion mode, which allows better sensitivity and much shorter 40 kV and with a current of 45 mA.
measurement times. On the other hand, due to the very con- The goniometer platform has pre-aligned, fixed interfaces
fined space that is usually available on a goniometer-based, (Fig. 1) that can be configured with a variety of pre-aligned
laboratory X-ray diffraction instrument and due to the insuffi- X-ray modules, which include incident beam optics, sample
cient spatial resolution of commonly available X-ray detectors, stages, receiving optics, and detectors. The special interfaces
such a setup only allowed low-angle diffraction measurements, that connect the X-ray modules to the goniometer platform
but it was not possible to achieve a small-angle resolution as are made of hardened steel and ensure a highly accurate,
required for SAXS.11 Also, all attempts reported so far have reproducible, and stable mounting of the various components
made use of a short evacuated beam path that is mounted simply by fixing a single screw with a well-defined torque.
between the sample and the detector,9,11 whereas the sample The X-ray source together with suitable incident beam optics
and the collimation system were not part of the vacuum path. is mounted on the omega arm of the goniometer platform,
As a consequence, the residual parasitic scattering from air whereas the detector is attached to the 2theta arm. A sample
results in an increased background level and to a concomitant stage is mounted on the central interface. This enables an easy
loss of sensitivity for weakly scattering samples. and quick switch between different instrument configurations
We have developed a solution that allows high- for the various X-ray diffraction and scattering techniques
performance SAXS/WAXS measurements on a compact, without the need for re-alignment or calibration. An enclo-
multi-functional goniometer platform that can be configured sure with widely opening, see-through doors ensures radiation
with different pre-aligned X-ray modules for various appli- safety as well as easy accessibility to the experimental setup.
cations. In this paper, we describe the details of the experi- The footprint of the instrument is 140 × 116 cm2 .
mental setup for SAXS and WAXS and demonstrate its good
performance on a variety of test samples. B. SAXS/WAXS configuration
1. Line collimation setup

II. INSTRUMENT OVERVIEW Small-angle X-ray scattering (SAXS), including SAXS

on dilute solutions of biomacromolecules (bio-SAXS), and
A. Instrument platform
wide-angle X-ray scattering (WAXS) are enabled by the exper-
The instrument (Empyrean, Malvern Panalytical; see imental setup shown in Fig. 2. A sealed copper X-ray tube in
supplementary material, Fig. S1) is based on a multi- its line focus orientation is used as an X-ray source. An ellip-
functional, floor-standing X-ray scattering platform with a tically bent, 1D graded, multilayer X-ray mirror focuses the
085115-3 Bolze et al. Rev. Sci. Instrum. 89, 085115 (2018)

FIG. 3. Typical SAXS patterns from an ensemble of monodisperse nanopar-

ticles measured with a 2D detector by using the line collimation setup
(left side) and the point collimation setup (right side), respectively. The yellow
arrows point to the attenuated direct beam. The red arrows indicate how the 2D
FIG. 2. Experimental setup used for SAXS/WAXS experiments. (A) Cu patterns are reduced to obtain 1D scattering curves: the intensities measured
X-ray tube, (B) focusing X-ray mirror, (C) evacuated beam path with anti- with the line collimation setup are integrated along the pixel rows in the axial
scatter devices and a semi-transparent beam stop, (D) sample capsule with direction, whereas the intensities measured with the point collimation setup
sample holder, and (E) hybrid pixel area detector. are azimuthally averaged.

divergent X-ray beam from the source in the equatorial direc- is very compact with distances from the X-ray source to the
tion on the detector plane. This results in a monochromatic sample, as well as from the sample to the detector plane, of
(Cu Kα radiation, with a residual Cu Kβ contamination that 240 mm, respectively. Upon removing and reattaching one or
amounts 0.1% of the Cu Kα radiation), highly intense, narrow, more of the modules that are used for the SAXS/WAXS setup,
line-shaped beam. The beam dimensions on the detector are the measured position of the direct beam is very stable, typ-
approximately 130 µm in the equatorial direction and 17 mm in ically within ±0.002◦ 2θ. The translational reproducibility of
the axial direction. Different entry slits can be inserted in front the detector positioning is typically within ±5 µm in all three
of the mirror allowing to optimize the balance between photon directions.
flux and achievable small-angle resolution. Such strongly elon- With this experimental setup, transmission SAXS/WAXS
gated beam in the axial direction has traditionally been used, measurements can be performed in a gapless range of
e.g., in the classical Kratky Kompakt Kamera for SAXS2 as scattering angles from approximately 0.08◦ to 78◦ 2θ
well as in the ultra-SAXS setup developed by Bonse and Hart.2 [q = 0.06–51.4 nm−1 , where q denotes the scattering vector
A compact, detachable SAXS/WAXS chamber (ScatterX 78 , q = 4π λ −1 sin(θ), and λ denotes the wavelength of radia-
Malvern Panalytical) can be highly reproducibly mounted on tion]. The FWHM of the direct beam at the detector position
the goniometer platform via the central stage interface. The is about 0.035◦ 2θ (0.025 nm−1 in terms of q). At 0.10◦ 2θ
chamber contains two micrometer-driven, adjustable slits with (q = 0.07 nm−1 ), the intensity of the background, measured
tungsten blades for shaping the tails of the incident beam, a without a sample holder inserted, is well below 10−5 of
capsule that takes in various sample holders, and a beam stop. the direct beam peak intensity. Toward higher angles, the
The openings where the X-ray beam enters and exits the cham- background quickly decreases further to a negligible level
ber are sealed with thin, X-ray transparent polymer foils. The (typically 0.1-0.3 cps). In practice, the actual smallest achiev-
entrance window consists of polyethylene terephthalate (PET; able scattering vector qmin in a given experiment depends on
thickness 6 µm, size 17 mm × 8 mm), and the exit window the footprint of the direct beam on the detector, the spatial res-
consists of polyether ether ketone (PEEK; thickness 50 µm, olution of the detector, the height (or width) of the beam stop,
size 3 cm × 29 cm). The chamber is evacuated to a vacuum as well as the scattering intensity of the sample relative to the
level of below 0.1 mbar to avoid parasitic air scattering. Due background signal.
to the compact dimensions, the evacuation can be readily per- The total photon flux is typically 4 × 108 ph/s in a SAXS
formed in less than a minute. The in-vacuum beam stop has experiment and can be increased to about twice as much in
a rectangular shape (width 32 mm and height 10 mm) and a WAXS experiment by using a relaxed collimation setting.
is made of a semi-transparent metal foil (copper; thickness Therefore, this setup enables a good sensitivity also for very
0.15 mm) so that the trace of the attenuated direct beam can weakly scattering samples, including dilute protein solutions.
be observed on the detector (Figs. 3 and 6). In this way, the Scattering data on an absolute intensity scale can be
direct beam position is included in each measurement as a obtained, e.g., by using water as a primary calibration stan-
reference, and the attenuation factor from a given sample can dard.12,13 Figure S2 (see supplementary material) shows
be determined and properly taken into account in data reduc- SAXS data from water at T = 20 ◦ C that were obtained by
tion. The 2D-detector (see Sec. II C) is mounted behind the subtraction of the blank scattering and by carefully taking
exit window of the vacuum path. When used in a stationary absorption effects into account. The absolute forward scatter-
mode, it covers a 2θ range of 6.9◦ (q-range 4.9 nm−1 ). Higher ing of water I(q = 0) relates to its isothermal compressibility
scattering angles in the WAXS region can be accessed by posi- and is 1.632 × 10−2 cm−1 at T = 20 ◦ C.13 Alternatively, one
tioning the 2theta arm of the goniometer with the attached could use, e.g., a colloidal dispersion of silica nanoparticles or
detector at a higher angle or by performing a 2θ scan of the a polymer with embedded nanoparticles as cross-calibrated,
detector within the angular range of interest. The whole setup secondary standards.14 For the determination of the molecular
085115-4 Bolze et al. Rev. Sci. Instrum. 89, 085115 (2018)

weight Mw of a protein in a solution, the extrapolated forward

scattering measured from a standard protein of known Mw and
concentration can also be used for calibration, as described by
Mylonas and Svergun.15 If the partial specific volumes of the
standard protein and of the protein to be analyzed are not the
same, the difference must be taken into account in the scal-
ing procedure. The scaling of the scattering intensities on an
absolute scale is performed based on the desmeared data.
Various sample holders for transmission measurements
on different sample types have been developed. The holders
are inserted in the evacuated beam path, and their position-
ing is highly reproducible. Liquid samples are measured in
disposable, thin-walled quartz capillaries that give only a low
scattering background. The open end of the capillary is sealed
with a tightly fitting cap. The minimum required sample vol-
ume is 40-50 µl. Replacing a used capillary by a new one can FIG. 4. SAXS data from a powder of polystyrene nanoparticles having a
be performed within 1 min. The capillary can also be cleaned, narrow particle size distribution. (1) Data as measured with the line collimation
setup and (2) desmeared line collimation data. The desmearing procedure
and the identical capillary is used for the sample and corre-
results in more pronounced oscillations as well as in a change of the intensity
sponding background measurements. The total time required decay in the tail of the scattering curve: the power law exponent is −4 for the
for cleaning and filling a capillary is less than 5 min. The desmeared data (in agreement with Porod’s law)2 and −3 for the smeared line
capillary holder also allows sample temperature control in the collimation data (as expected in the infinitely long slit approximation).2,3,7
The desmeared data are in good agreement with those measured from the
range from 5 to 70 ◦ C with a temperature stability of ±0.1 ◦ C. same sample with the point collimation setup and reduced to 1D data by
Whereas for water-based samples, 1 mm capillaries are used, azimuthal averaging (3).
smaller diameters are preferred in the case of samples that are
higher absorbing than water. Powder samples are prepared in
metal frames between thin polymer foils. Thin solid objects, the sample has to be moved to a fixed position on a sec-
free-standing films, and foils or fibers are clamped on a base ond sample stage that is mounted at a short distance in front
holder. of the detector. In this way, full 2D WAXS patterns with
Using a line collimation setup as described above has the 360◦ azimuthal coverage can be acquired within a 2θ range
advantages that (i) it enables a high photon flux, (ii) it can be from −30◦ to +30◦ (q-range from −21.1 to +21.1 nm−1 ),
realized with relatively cost-effective X-ray modules (source, whereas for 2D SAXS measurements, with the sample located
optics, and detector), (iii) statistically relevant information is in the center of the goniometer, the 2θ range spans from
obtained from a large scattering volume, and (iv) radiation −3.4◦ to +3.4◦ (q-range from −2.42 to +2.42 nm−1 ). This
damage, even of delicate samples, is usually not encountered. 2D SAXS/WAXS setup is well suited for measurements on
On the other hand, it has the drawback that the measured SAXS polymers, powders, and solids. Typical measurement times
data are strongly affected (“smeared”) by the elongated beam with the point collimation setup are in the range of 30 min to
shape (Figs. 3 and 4). These slit smearing effects1,2 need to several hours depending on the sample and required counting
be accounted for during data analysis. Furthermore, a line col- statistics. Due to the lower primary beam intensity, it is less
limation setup is only suited for isotropic samples. Typical suited for more weakly scattering samples. Changing between
SAXS and WAXS measurement times with the line collimation the line collimation and point collimation setups can easily
setup are in the range of 1-60 min, depending on the sample. be performed within 1 min without the need for re-alignment.
The transfer of the sample between the positions for 2D SAXS
and 2D WAXS measurements requires 1 min.
2. Point collimation setup
For studying microdomain orientation in anisotropic
C. X-ray detector
materials, a point collimation setup and a 2D detector are
required. Such a setup can be easily realized by inserting a A photon counting, hybrid 2D pixel array detector
set of axial masks in the beam path of the line collimation (GaliPIX 3D , Malvern Panalytical) with a 2-level energy dis-
setup described above. The axial size of the X-ray beam can criminator is used. It is based on a solid-state, semiconductor
thus be reduced, resulting in a close-to-quadratic beam cross sensor (CdTe) which is bump-bonded to the CMOS read-out
section on the detector. 2D SAXS and 2D WAXS measure- electronics. The single module detector has a sensor area of
ments can then be performed by operating the detector (see 31 × 25 mm2 with effective pixel dimensions of 60 × 52 µm2 .
Sec. II C) in a 2D detection mode and by replacing the 1D Having an excellent spatial resolution with a point spread
beam stop by a small, pre-aligned 2D beam stop. The tip function that equals the size of a single pixel, this detec-
of the 2D beam stop is also semi-transparent (Cu foil; 0.18 tor provides a good ∆q-resolution even with a very compact
mm thickness and 0.6 mm width) so that the intensity and experimental setup. Furthermore, its high count rate capability
position of the direct beam are included in each measurement (2.4 × 1011 cps globally; 100 kcps per pixel), combined with
(Fig. 3). A dual sample position concept is used: in order to a very low noise level (globally <6 cps), enables SAXS mea-
increase the detector field of view for a 2D WAXS experiment, surements in a wide dynamic range and in particular also of
085115-5 Bolze et al. Rev. Sci. Instrum. 89, 085115 (2018)

very weakly scattering samples. Its 100% quantum efficiency this area detector are summed up along the pixel rows in the
not only for Cu X-rays but also for hard X-rays makes this axial direction (Fig. 3) to obtain a 1D intensity profile as a
type of detector also highly suitable for relatively fast in-house function of the scattering angle 2θ as it would be the case with
total scattering (atomic pair distribution function; PDF) mea- a 1D multi-strip detector.
surements16 that usually require Ag or Mo radiation. It will
be shown elsewhere that the instrument platform described
here also uniquely enables PDF measurements up to very high D. Instrument control, data acquisition,
and analysis software
q-values (qmax = 220 nm−1 ).38 The PDF technique is thus an
extension of SAXS and WAXS and can give valuable insight The control of the instrument and the setup and execu-
into the local atomic structure, particularly of nanomaterials tion of measurement programs are performed with the Data
and disordered materials in general. When using Cu radia- Collector software (Malvern Panalytical, The Netherlands).
tion, the CdTe-based detector can withstand the unattenuated The raw measurement data together with all information about
direct beam without being damaged. For SAXS and WAXS the used instrument configuration and the scan parameters are
measurements with Cu radiation, it is of course also possi- stored in an open, XML-based file format. As an example, a
ble to use a hybrid pixel area detector with a silicon sensor portion of such a data file is shown in supplementary material
(e.g., PIXcel3D , Malvern Panalytical). However, such a detec- (Fig. S3).
tor lacks the radiation hardness and is not suitable for PDF Jointly with the European Molecular Biology Laboratory
measurements. Inhomogeneities in pixel-to-pixel sensitivity (Svergun et al., EMBL Hamburg, Germany), we developed
are taken into account by a flat field correction of the experi- SAXS software (EasySAXS, Malvern Panalytical, The Nether-
mental raw data. The detector is mounted on the 2theta arm of lands) that is partly based on the PRIMUS17 and GNOM18,19
the goniometer and, when needed, can be scanned around the modules of the popular ATSAS software package20 of the
sample that is placed in the center of the goniometer. When EMBL. It has a graphical user interface for easy and efficient
used in a line collimation setup, the intensities measured by use of the software functionalities (Fig. 5).

FIG. 5. Screen grab of the graphical user interface of the EasySAXS data analysis software, showing as an example the determination of the p(r) function from
SAXS data of a dilute protein solution (apoferritin, 10 mg/ml; measurement time 60 min).
085115-6 Bolze et al. Rev. Sci. Instrum. 89, 085115 (2018)

The SAXS software supports a variety of data reduc-

tion and treatment operations, such as background correction,
scaling of intensities, summation of repeated measurements,
data merging, data smoothing (normally not used), and han-
dling of slit smearing effects. Data analysis options include,
e.g., Guinier and Porod analyses, calculation of the scatter-
ing invariant Q, determination of surface areas, as well as
model simulations and fitting for a variety of particle shapes
(spheres, ellipsoids, platelets, and cylinders), including core-
shell structures. EasySAXS also enables data analysis using an
indirect Fourier transformation procedure18,21 that is based on
the algorithms that are implemented in the GNOM module18,19
of the ATSAS software package.20 For an ensemble of poly-
disperse, spherical nanoparticles, it enables the determination
of the particle size distribution without making any a priori
assumptions about the shape and modality of the distribution
curve. For an ensemble of monodisperse, identical nanoparti-
cles in a dilute solution (e.g., protein molecules), it can be used FIG. 6. SAXS data, including the attenuated direct beam profiles, measured
from NIST reference material RM 8013. Sample: solid line, background: long-
to generate the pair distance distribution function p(r) from short-dashed line, background-corrected data: symbols, back-transform of the
which information about the overall particle shape, structure, particle size distribution: dashed line. The arrow indicates the direct beam that
and dimensions can be deduced.22 As slit smearing effects are is transmitted through the semi-transparent beam stop. The inset shows the
deduced particle size distribution (solid line) and its Gaussian approximation
routinely accounted for (based on parameterized direct beam
(dashed line). The line collimation setup was used, and the measurement time
profiles in the axial and equatorial directions) in the indirect was 20 min.
Fourier transformation procedure as well as in model simula-
tions and fits, prior desmearing of the experimental data is not
required.2,3 However, the software also supports the desmear- can easily be seen, and distinct oscillations can be observed
ing of experimental data from isotropic samples, which results in the background-corrected scattering curve, which is indica-
in a scattering curve as if it was measured with the point col- tive for a narrow particle size distribution. The size distribution
limation setup (Fig. 4). However, the desmearing procedure was determined by using the indirect Fourier transformation
tends to enhance the statistical noise in the data and is there- procedure, assuming a spherical particle shape. An essentially
fore avoided in most cases. The software also offers various Gaussian distribution with an average particle size of 54.4 nm
automation and reporting options. The sequence of data anal- was thus determined, which is in good agreement with the ref-
ysis, including all parameter settings, is stored in a project file. erence value of 53.2 ± 5.3 nm given by NIST. Furthermore,
For more in-depth data analysis, one can export data in a file the size polydispersity (given by the ratio of the standard devi-
format that is compatible with most third-party SAXS software ation of the size distribution and of the average radius) was
packages, such as ATSAS,20 SASfit,23 and Irena.24 determined to be 9.8%. The experimental data are well fitted
2D SAXS and 2D WAXS data can be displayed, ana- by this model. Table I summarizes the results obtained from all
lyzed, and reduced to 1D data with the XRD2Dscan software three samples and shows the agreement with the NIST refer-
(Malvern Panalytical, The Netherlands) and then exported for ence values. Repeated sample preparation and measurements
further analysis. Diffraction patterns that are observed in the of a given sample typically resulted in a repeatability of the
WAXS data are analyzed with the HighScore (Plus) software25 deduced particle diameter within ±1%.
(Malvern Panalytical, The Netherlands). For testing the possibility of resolving multimodal par-
ticle size distributions, a 1:1:1 mixture (by sample volume)
of the three NIST samples was prepared. The analysis of the
III. INSTRUMENT PERFORMANCE measured SAXS data indeed resulted in a trimodal particle
size distribution in which the three size fractions are very well
A. Example 1: NIST reference materials - Gold resolved (Fig. 7). This also demonstrates the power of the indi-
nanoparticles rect Fourier transformation method for determining particle
To validate the instrument performance of determining size distributions without making a priori assumptions about
the size of nanoparticles in very dilute dispersions, SAXS data the shape and modality of the distribution.
were measured from three colloidal gold reference materials
purchased from NIST (RM 8011, RM 8012, and RM 8013)
TABLE I. Results of particle size determination on colloidal gold NIST
with nominal particle sizes of 10, 30, and 60 nm, respec- reference materials.
tively. The particle concentration was about 0.05 mg/ml for
all samples. The measurement time for the sample and the Particle diameter D (nm)
corresponding background, measured from the same capil- (volume-average) RM 8011 RM 8012 RM 8013
lary filled with pure water, was 20 min each. Figure 6 shows Experimental 9.0 26.0 54.4
exemplary results obtained from sample NIST RM 8013. The NIST reference value 9.1 ± 1.8 24.9 ± 1.2 53.2 ± 5.3
excess scattering from the particles in the dispersion medium
085115-7 Bolze et al. Rev. Sci. Instrum. 89, 085115 (2018)

measured as received and thermostated at a temperature of

20 ◦ C and with a measurement time of 60 min. From the
slit-smeared, background-corrected data, the nanoparticle size
distribution by particle volume (inset in Fig. 8) was determined
with the indirect Fourier transformation method, assuming that
the particles are spherical. A bimodal distribution was obtained
with two distinct populations: the average particle diameter of
population 1 (2) was determined as 19.5 nm (82.8 nm) with a
size polydispersity of 19% (5%). From the ratio of the areas
under the two main peaks in the size distribution curve, the
volume ratio ϕv,1 /ϕv,2 could be estimated to be 15.7 (±25%).
These results are in good agreement with the (indicative) val-
ues given in the certificate of analysis (namely, average particle
diameters of 19.8 nm and 80.1 nm for the two populations) and
with those additionally derived from the certification report of
ERM-FD-102 (namely, ϕv,1 /ϕv,2 = 19.8).26 Also, the scatter-
ing curve corresponding to the determined size distribution is
a good fit to the experimental data. The distribution shows two
FIG. 7. SAXS data from a mixture of equal volumes of the three colloidal additional minor modes around particle diameters of 42 nm
gold NIST reference materials RM 8011, RM 8012, and RM 8013 (top graph)
and the deduced trimodal particle size distribution (bottom graph). The line and 48 nm, respectively. Whereas interestingly a similar obser-
collimation setup was used, and the measurement time was 20 min. vation of additional modes is mentioned in the certification
report,26 it cannot be excluded that these are rather some minor
artefacts from the Fourier transformation. Figure 9 shows the
B. Example 2: IRMM nanoparticle standard desmeared SAXS data from the same sample (no prior data
with bimodal size distribution smoothing applied) on an absolute intensity scale, where the
Figure 8 shows SAXS data measured from a colloidal scattering from pure water was used for calibration. The cor-
silica sample in an aqueous solution. This sample is a commer- rect scaling was verified by an additional measurement of a
cially available, certified reference material (ERM-FD-102, 10 mg/ml lysozyme solution that gave an extrapolated forward
Institute for Reference Materials and Measurements, Geel, scattering I(q = 0) of 0.099 cm−1 , which is in good agreement
Belgium) and is intended for quality control and for the assess- with published data.15 Figure 9 also shows published SAXS
ment of the performance of nanoparticle size analysis methods, data23 from the same sample measured on a dedicated SAXS
including SAXS. It was prepared by mixing two samples instrument. The data obtained on the instrument described in
containing monomodal populations of silica nanoparticles of this paper are in very good agreement with these published
distinct size, resulting in a bimodal particle size distribution data, both in absolute intensities and with respect to the shape
with a silica concentration of 8.75 mg/ml. The sample was

FIG. 9. Desmeared SAXS data of sample ERM-FD-102 (solid line) on an

absolute intensity scale, compared with published data23 (symbols) from the
same sample, measured with a dedicated SAXS instrument (SAXSess, Anton
FIG. 8. SAXS data as measured from colloidal silica reference sample ERM- Paar, Graz, Austria). Also shown is the scattering from water (short-dashed
FD-102. Solid line: sample measurement, short-dashed line: background line) that was used as a calibrant for the intensities, as well as desmeared
measurement, symbols: background-corrected data, long-dashed line (mostly SAXS data from a 10 mg/ml lysozyme solution (long-dashed line) for cross-
hidden behind the symbols): fit curve. The inset shows the derived bimodal checking the scaling to absolute intensities. A comparison of the slit-smeared
particle size distribution. The two main size fractions are indicated by arrows. and desmeared SAXS data in a single plot is presented in the supplementary
The line collimation setup was used, and the measurement time was 60 min. material, Fig. S4.
085115-8 Bolze et al. Rev. Sci. Instrum. 89, 085115 (2018)

and overall decay of the scattering curves. It is also noteworthy

that with a qmin of 0.057 nm−1 , the smallest attainable scatter-
ing vector on our instrument was identical to the one that was
achieved on the dedicated SAXS instrument.

C. Example 3: SAXS on dilute protein

solutions (bio-SAXS)
To demonstrate the bio-SAXS capabilities of the instru-
ment, Fig. 10 shows SAXS data measured from a commer-
cial, engineered insulin sample (insulin aspart, Novorapid® ,
Novo Nordisk) at a controlled sample temperature of 20 ◦ C.
The insulin concentration was 3.5 mg/ml (phosphate buffered
saline, pH 7.4). The background was measured from the pure
buffer using the same capillary as for the actual sample. Despite
the relatively low sample concentration and the low scattering
contrast, good-quality SAXS data with a qmin of 0.06 nm−1
could be obtained within a rather short measurement time FIG. 11. P(r) function analysis of SAXS data from insulin aspart. The inset
displays the derived p(r) function, and in the main graph, its back-transform
of 30 min. The data measured at the smallest angles follow is shown together with the experimental data.
a straight line when displayed in a Guinier plot27 of ln I(q)
versus q2 (inset A in Fig. 10). From this behavior, one may
conclude that the sample is essentially free from large, non- back-transformed data from the p(r) function analysis were
specific aggregates and from the slope of the line, the radius of used without applied slit smearing. In this plot, a bell-shaped
gyration Rg could be determined to be 1.95 nm. By applying an curve with a maximum at around (qRg ) = 1.68 is observed.
indirect Fourier transformation19 to the SAXS data, a smooth This together with the shape of the p(r) function indicates that
pair distance distribution function p(r) was obtained (inset in the insulin molecules in the solution contain folded domains29
Fig. 11). The p(r) function is the real space representation of and have a globular structure. Based on the crystal structure
the scattering curve.2 Figure 11 shows that the back-transform of insulin aspart (PDB 4GBN) published in the Protein Data
of p(r) is a good fit to the experimental data. From the evalua- Bank,31 the corresponding solution SAXS data were calcu-
tion of the p(r) function as described elsewhere,2 the maximum lated using the CRYSOL software.32 The comparison of the
dimension Dmax of the protein was determined as 5.5 nm and its experimental data with these calculated data (Fig. 12) from the
(real space) radius of gyration Rg was determined as 1.94 nm, atomic model of the hexameric oligomer shows a good agree-
which is consistent with the value obtained by Guinier anal- ment and indicates that the protein structures in the solution
ysis and with the one given in the scientific literature.28 and in the crystal are similar. The presence of a hexameric
A dimensionless Kratky plot [(Ix(qRg )2 versus (qRg )]29,30 of form was also confirmed by the determination of the molec-
the SAXS data is shown in Fig. 10 (inset B). Here the smooth, ular weight Mw. As shown in Fig. 11, this was performed
by comparing the forward scattering intensity I(q = 0) of the

FIG. 10. SAXS measurement on a dilute solution of insulin aspart FIG. 12. Comparison of experimental SAXS data from insulin aspart with
(c = 3.5 mg/ml). Solid line: sample measurement; dashed line: background simulated data (dashed line) based on the hexameric structure of insulin that
measurement; connected symbols: background-corrected data. Inset A shows was determined by single-crystal X-ray diffraction. Also shown is the scat-
the Guinier plot, and inset B shows the dimensionless Kratky plot. The line tering curve from a lysozyme solution that was used as a molecular weight
collimation setup was used, and the measurement time was 30 min. standard. Intensities were normalized to the sample concentration c.
085115-9 Bolze et al. Rev. Sci. Instrum. 89, 085115 (2018)

insulin sample with the one obtained under the same mea- Several more examples of bio-SAXS data measured from
surement conditions from a solution of a protein of known protein samples of different molecular weights are given in
Mw (lysozyme, 14.3 kDa), where intensities are normalized to Fig. 13, demonstrating the good sensitivity and resolution
the sample concentration. I(q = 0) was determined by extrap- that can be achieved in such measurements. Typical measure-
olation of the desmeared data in a Guinier plot,27 as well as ment times for bio-SAXS data acquisition are in the range of
based on the p(r) function. The thus determined values for Mw 20-60 min, whereas Rg can often be determined from just a one-
(36.0 kDa and 37.3 kDa) are in good mutual agreement and minute measurement. No radiation damage was encountered
agree with the theoretical Mw of insulin aspart in its hexameric for any of the measured protein samples.
form (36.3 kDa, PDB 4GBN). The uncertainty in the determi- It is noteworthy that the same type of instrument plat-
nation of absolute scattering intensities and in the deduced Mw form also enables obtaining high-quality transmission X-ray
values is estimated to be 10%. diffraction patterns from microcrystalline protein powders.33
This is a challenging task because from these samples, a large
number of diffraction peaks must be resolved at low angles. An
example of the measurement data of microcrystalline insulin
samples is given in Ref. 33. Protein powder X-ray diffrac-
tion enables the study of the occurrence and stability of dif-
ferent polymorphic forms, e.g., under different non-ambient
conditions or as a function of crystallization conditions.34
High-throughput polymorph screening of many samples can
be performed by using a well plate. The technique is also used
for unit cell and symmetry determination for protein struc-
ture refinement and ultimately for solving unknown protein
structures.35

D. Example 4: SAXS/WAXS on liposomes

at various temperatures
To demonstrate the SAXS/WAXS performance of the
instrument under controlled sample temperature, measure-
ments were conducted on a bilayer-forming phospholipid
that form closed vesicles (liposomes) in an aqueous solution.
DPPC (1,2-dipalmitoyl-sn-glycero-3-phosphocholine; Avanti
Polar Lipids, USA) was dissolved in an aqueous buffer
solution (PBS buffer, pH 7.4) at a lipid concentration of
25 mg/ml and heated to 60 ◦ C. Using a pore extruder (T&T
Scientific, USA), the heated lipid solution was subsequently
passed 11 times through a membrane with a pore size of
100 nm.
Figure 14 shows the SAXS/WAXS data obtained from the
sample before extrusion. An automated batch program with a
series of 10 min measurements at different sample tempera-
tures (between 5 ◦ C and 70 ◦ C) was used. Several pronounced,
equidistant diffraction peaks were observed at small scatter-
ing angles, pointing to the formation of multi-lamellar vesicles
containing a stack of lipid bilayers. The lamellar repeat dis-
tances d 1 as a function of sample temperature were derived
FIG. 13. Experimental SAXS data from different protein solutions. The scat- using Bragg’s law and are summarized in Table II. At the lower
tering curves are vertically displaced for better visualization. The molecular temperatures, up to 39 ◦ C, an additional diffraction peak at
weight of the proteins increases from the bottom to the top curve and ranges around q = 14.8 nm−1 with a corresponding Bragg spacing d 2
between 14 and 670 kDa. (1) Ribonuclease A (100 mM Tris, 100 mM NaCl,
pH 7.5). (2) Lysozyme (40 mM acetic acid, 50 mM NaCl, pH 4.0). (3) Insulin of about 4.3 Å was observed in the WAXS data. This peak
aspart (phosphate buffered saline, pH 7.4). (4) Ovalbumin (10 mM Tris, relates to the regular, in-plane packing of the alkyl chains
100 mM NaCl, pH 7.5). (5) Hemoglobin (100 mM Tris, 100 mM NaCl, pH when the lipid is in the ordered, gel phase.36 The peak dis-
7.5). (6) Bovine serum albumin (BSA) (50 mM Hepes, pH 7.5). (7) Alcohol appears upon further increasing the temperature, indicating
dehydrogenase (100 mM Tris, 100 mM NaCl, pH 7.5). (8) Glucose isomerase
(100 mM Tris, 1 mM MgCl2 , pH 8.0). (9) Immunoglobulin G from goat serum the transition to the disordered, liquid phase. The transition
(50 mM acetic acid, pH 5.0). (10) Apoferritin (100 mM Tris, 100 mM NaCl, is observed at a temperature that agrees with the tabulated
pH 7.5). (11) Thyroglobulin (100 mM Tris, 100 mM NaCl, pH 7.5). Sample phase transition (melting) temperature of DPPC (T m = 41 ◦ C;
concentrations were in the range of 5-10 mg/ml, and the measurement times
www.avantilipids.com). Just below T m , our data indicate a
were in the range of 30-60 min. The sample temperature was 20 ◦ C. The ver-
tical dashed line denotes the q = 0.1 nm−1 position to indicate that in all cases, transient, more disordered phase with significantly increased
meaningful experimental data could be acquired also well below this limit. interlamellar distances, which can be associated with the
085115-10 Bolze et al. Rev. Sci. Instrum. 89, 085115 (2018)

FIG. 15. SAXS data from the extruded DPPC liposomes measured at a sample
temperature of 20 ◦ C. Residual Bragg peaks (marked by arrows) are observed
in the background-corrected data. The scattering curve can be best modelled
with a bilamellar liposome structure. The simulated data are vertically offset
FIG. 14. SAXS/WAXS data from multilamellar vesicles formed by DPPC for better visibility. The line collimation setup was used, and the measurement
measured at different temperatures. The dashed line is a background measure- time was 60 min.
ment of the pure dispersion medium at T = 20 ◦ C. Data are vertically offset
for better visibility. The line collimation setup was used, and the measurement
time was 10 min for each scan. E. Example 5: WAXS on polymers
To further demonstrate the WAXS performance of the
occurrence of a “ripple phase.”37 The temperature-dependent instrument, measurements were taken on different polymer
Bragg spacings derived from our data are also in very good samples. Figure 16 (top) shows an example of the identifica-
agreement with those reported by others36,38 based on mea- tion of an unknown polymer including an additive. This was
surements using various dedicated, laboratory SAXS/WAXS
instruments.
The background-corrected SAXS data measured at
T = 20 ◦ C from the liposomes after extrusion are shown in
Fig. 15. The pronounced Bragg peaks almost disappeared,
indicating (as expected) a strongly reduced lamellarity. Instead
the scattering curve shows well-resolved, typical features
from essentially unilamellar vesicles, as reported, e.g., by
Salvatore et al.36 Based on a parameterized electron density
profile across the lipid bilayer,39 we used the SASfit software23
to simulate SAXS data from uni- and oligolamellar DPPC vesi-
cles. The experimental data are best described by a bilamellar
liposome structure model, taking into account the two very
broad and weak residual Bragg peaks (indicated by arrows in
Fig. 15) that are still present.

TABLE II. Structural parameters of DPPC liposomes as a function of sample

temperature. Notation: d 1 denotes the Bragg spacing due to lamellar repeat
distance, derived from the SAXS data; d 2 denotes the Bragg spacing due to
alkyl chain ordering, derived from the WAXS data.

Sample temperature T (◦ C) d 1 (nm) d 2 (nm) Lipid phase type

5 6.3 0.43 Gel phase

20 6.4 0.42 Gel phase
30 6.4 0.43 Gel phase
FIG. 16. WAXS measurements from polymers, using the line collimation
39 7.3 0.43 Ripple phase
setup. Top: identification of a white-colored polymer sample. From the diffrac-
43 6.7 ... Liquid phase tion pattern, the sample could be identified as α-polypropylene containing
50 6.6 ... Liquid phase titanium dioxide in the rutile phase. Bottom: thin sheet of Teflon. The mea-
70 6.3 ... Liquid phase sured WAXS peak positions agree well with those reported in the ICDD PDF-4
Organics data base for polytetrafluoroethylene.
085115-11 Bolze et al. Rev. Sci. Instrum. 89, 085115 (2018)

performed by using a search-match algorithm that compares

the measured diffraction peak positions with those found in
a suitable reference database, such as published by the Inter-
national Center for Diffraction Data (ICDD, PA, USA) for
a variety of substances. Similarly, the peak pattern in the
measured WAXS data from a thin Teflon sheet (Fig. 16, bot-
tom) is in good agreement with the data found in the ICDD
PDF-4 Organics database (code 00-054-1595). An example
of SAXS/WAXS data from a semi-crystalline polyethylene FIG. 18. 2D SAXS pattern data from a colloidal crystal formed by
sample is given in Fig. S5 (see supplementary material). poly(methyl methacrylate) (PMMA) nanoparticles (D = 105 nm). Measure-
ments were taken without insertion of a beam stop. The point collimation
setup was used, and the measurement time was 5000 s. Left: full pattern as
F. Example 6: 2D SAXS/WAXS measured (q-range ± 2.4 nm−1 ); right: enlarged, central portion of the same
pattern (q-range ± 0.46 nm−1 ), revealing distinct Bragg spots with a 6-fold
Some representative examples of full-pattern 2D SAXS/ symmetry.
WAXS data that were measured with the point collimation

setup are shown in Fig. 17. For isotropic samples, such as sil-
ver behenate, the typical ring patterns with uniform azimuthal
intensity distributions were obtained [Figs. 17(a) and 17(b)].
Pronounced preferred orientation effects can readily be seen
in the WAXS patterns from polypropylene fibers [Figs. 17(c)
and 17(d)]. Anisotropy was also observed upon stretching a
semicrystalline polyethylene sample: the orientation of the
lamellae is evident from the anisotropic 2D SAXS pattern
[Fig. 17(e)], whereas the orientation of the crystal lattice can be
deduced from the anisotropic 2D WAXS pattern [Fig. 17(f)].
The same sample before stretching gave fully isotropic pat-
terns (not shown). The high angular resolution that can be
achieved in a 2D SAXS experiment is demonstrated by the
ability to measure the diffraction pattern from a collagen fiber
sample down to the first order reflection with a correspond-
ing Bragg spacing of 67 nm [Fig. 17(g)]. A large number
of distinct, narrowly spaced rings are observed in the 2D
SAXS pattern of a dried polystyrene latex (particle diameter
70 nm) [Fig. 17(h)], pointing to a very narrow nanoparticle size
distribution.
Figure 18 shows the 2D SAXS pattern from a colloidal
crystal from a dried, charge-stabilized polymethylmethacry-
late latex (particle diameter 105 nm). This measurement was
taken in the absence of a beam stop in order to better resolve the
Bragg spots with 6-fold symmetry in very close proximity to
the direct beam. The diffraction pattern points to the formation
of a crystal of nanospheres with a hexagonal symmetry and a
high degree of order. The six innermost Bragg spots appear at
a 2θ position of 0.10◦ (q = 0.077 nm−1 ) with a corresponding
Bragg spacing d of 82 nm. Note that these distinct spots could
be well resolved and are clearly separated from the central
direct beam.

IV. SUMMARY AND CONCLUSIONS

FIG. 17. Examples of 2D SAXS and 2D WAXS patterns. The point colli- The possibility to configure a versatile laboratory X-ray
mation setup was used, and the measurement times were in the range from
goniometer platform with detachable X-ray modules and an
90 min [for samples (e) and (f)] to several hours (for all other samples). [(a)
and (b)] Silver behenate. (a) 2D SAXS (q-range ± 2.4 nm−1 ), (b) 2D WAXS evacuated beam path that enable high-performance SAXS and
(q-range ± 21.1 nm−1 ). [(c) and (d)] Polypropylene, 2D WAXS. (c) Fiber, WAXS measurements within a q-range of 0.06–51.4 nm−1
(d) thread with a fibrous core surrounded by an isotropic shell. [(e) and (f)] has been described. Whereas on this instrument, isotropic
Stretched polyethylene. (e) 2D SAXS, (f) 2D WAXS. The double arrows indi-
cate the stretching direction. (g) Two crossed rat tail collagen fibers, 2D SAXS.
samples are preferably measured with a line collimation
(h) Dried polystyrene latex particles with very narrow particle size distribution, setup, one can easily switch to a point collimation setup
2D SAXS. for studying anisotropic samples. The performance of the
085115-12 Bolze et al. Rev. Sci. Instrum. 89, 085115 (2018)

SAXS/WAXS configuration and the achievable data quality 3 L. A. Feigin and D. I. Svergun, Structure Analysis by Small-
have been tested and verified on a variety of sample types, Angle X-Ray and Neutron Scattering (Plenum Press, New York,
1987).
such as colloidal dispersions, liposomes, surfactants, protein 4 Neutrons, X-Rays and Light: Scattering Methods Applied to Soft Con-
solutions, polymers, fibers, nanocomposites, and nanopow- densed Matter, edited by P. Lindner and Th. Zemb (Elsevier, Amsterdam,
ders. The results compare well with those that were obtained The Netherlands, 2002).
5 D. I. Svergun, M. H. J. Koch, P. A. Timmins, and R. P. May,
on single-purpose SAXS/WAXS instruments. The unique pos-
Small Angle X-Ray and Neutron Scattering from Solutions of
sibility to perform bio-SAXS measurements on dilute protein Biological Macromolecules (Oxford University Press, Oxford, UK,
solutions and also X-ray diffraction measurements on micro- 2013).
6 W. H. de Jeu, Basic X-Ray Scattering for Soft Matter (Oxford University
crystalline, lyophilized protein powders or on small molecule
drugs and excipients makes such an instrument very power- Press, Oxford, UK, 2016).
7 O. Glatter, Scattering Methods and their Application in Colloid
ful also for (bio-)pharmaceutical R&D and quality control and Interface Science (Elsevier, Amsterdam, The Netherlands,
applications. 2018).
8 J. Bolze, S. Rekhi, K. Macchiarola, and B. Litteer, in Nanotech 2010: Tech-
The available goniometer platform and the concept of pre-
aligned X-ray modules allow us to configure the same instru- nical Proceedings of the 2010 NSTI Nanotechnology Conference and Expo
(Nano-Science and Technology Institute, Cambridge, Massachusetts, USA,
ment also for a variety of other types of measurements, such 2010), Vol. 1, p. 57.
as powder X-ray diffraction in the classical Bragg-Brentano 9 A. Bota, J. Appl. Crystallogr. 46, 573 (2013).
10 J. Bolze, Chemical & Engineering News Supplement, June ed. (American
reflection geometry, microdiffraction, thin film and thin layer
Chemical Society, Washington, USA, 2014), p. 24.
analysis by X-ray reflectometry and grazing incidence diffrac- 11 B. B. He, Two-Dimensional X-Ray Diffraction (Wiley, New Jersey, USA,
tion, or X-ray computed tomography (CT). Changing between 2009).
these different experimental setups can be performed within 12 N. Dingenouts, J. Bolze, D. Pötschke, and M. Ballauff, Adv. Polym. Sci.

about 10 min. Furthermore, as will be described elsewhere,40 144, 1 (1999).

13 D. Orthaber, A. Bergmann, and O. Glatter, J. Appl. Crystallogr. 33, 218
the accessible q-range can significantly be extended toward (2000).
ultra-small angles for USAXS measurements, as well as 14 C. A. Dreiss, K. S. Jack, and A. P. Parker, J. Appl. Crystallogr. 39, 32

toward very high q for total scattering experiments, from which (2006).
15 E. Mylonas and D. I. Svergun, J. Appl. Crystallogr. 40, s245
the atomic pair distribution function (PDF) can be deduced.
(2007).
All this makes such an instrument an extremely versatile, 16 G. Confalonieri, M. Dapiaggi, M. Sommariva, M. Gateshki, A. N. Fitch,
powerful, and readily accessible tool for the analysis of nano- and A. Bernasconi, Powder Diffr. 30, S65 (2015).
17 P. V. Konarev, V. V. Volkov, A. V. Sokolova, M. H. J. Koch, and D. I. Svergun,
materials and biomacromolecules and also for general mate-
rial research applications in multi-user, shared experimental J. Appl. Crystallogr. 36, 1277 (2003).
18 D. I. Svergun, A. V. Semenyuk, and L. A. Feigin, Acta Crystallogr., Sect.
facilities. A: Found. Crystallogr. 44, 244 (1988).
19 D. I. Svergun, J. Appl. Crystallogr. 25, 495 (1992).
20 D. Franke, M. V. Petoukhov, P. V. Konarev, A. Panjkovich, A. Tuukkanen,

SUPPLEMENTARY MATERIAL H. D. T. Mertens, A. G. Kikhney, N. R. Hajizadeh, J. M. Franklin,

C. M. Jeffries, and D. I. Svergun, J. Appl. Crystallogr. 50, 1212
See supplementary material for an image of the com- (2017).
21 O. Glatter, J. Appl. Crystallogr. 10, 415 (1977).
plete instrument platform, the determination of the scattering 22 D. I. Svergun and M. H. J. Koch, Rep. Prog. Phys. 66, 1735
from pure water on an absolute intensity scale, an example (2003).
of a measurement file in an XML format, a comparison of 23 I. Breßler, J. Kohlbrecher, and A. F. Thünemann, J. Appl. Crystallogr. 48,

SAXS data from the colloidal silica sample ERM FD-102 1587 (2015).
24 J. Ilavsky and P. R. Jemian, J. Appl. Crystallogr. 42, 347
as measured with the line collimation setup with the data
(2009).
obtained after desmearing, as well as SAXS/WAXS data from 25 T. Degen, M. Sadki, E. Bron, U. König, and G. Nénert, Powder Diffr. 29(S2),
a semi-crystalline polymer. S13 (2014).
26 V. Kestens and G. Roebben, Reference Material ERM-FD-102, European

Commission, Joint Research Centre, Institute for Reference Materials and

Measurements (IRMM), Geel, Belgium, 2014.
ACKNOWLEDGMENTS 27 A. Guinier, Ann. Phys. 11, 161 (1939).
28 L. C. Palmieri, M. P. Favero-Retto, D. Lourenco, and L. M. Lima, Biophys.
We gratefully acknowledge the EMBL ATSAS team
Chem. 173, 1 (2013).
(D. Svergun et al.) for their cooperation in the devel- 29 T. Bizien, D. Durand, P. Roblin, A. Thureau, P. Vachette, and J. Pérez,
opment of the EasySAXS software. We thank A. Siupa Protein Pept. Lett. 23, 217 (2016).
30 A. G. Kikhney and D. I. Svergun, FEBS Lett. 589, 2570
(Malvern Panalytical) for having provided the liposome sam-
ples. We also acknowledge contributions by J. Vugteveen, (2015).
31 F. C. Bernstein, T. F. Koetzle, G. J. B. Williams, E. F. Meyer, Jr., M. D. Brice,
J. Boksem, and M. Gateshki (Malvern Panalytical) to the J. R. Rodgers, O. Kennard, T. Shimanouchi, and M. Tasumi, J. Mol. Biol.
design and optimization of the SAXS instrumentation. This 112, 535 (1977).
32 D. I. Svergun, C. Barberato, and M. H. J. Koch, J. Appl. Crystallogr. 28,
work was financially supported by NanoNextNL, the Dutch
768 (1995).
national research and technology program for micro- and 33 F. Karavassili, A. Valmas, S. Fili, C. D. Georgiou, and I. Margiolaki,
nanotechnology. Biomolecules 7, 63 (2017).
34 S. Fili, A. Valmas, M. Norrman, B. Schluckebier, D. Beckers, T. Degen,
1 A. Guinier and G. Fournet, Small-Angle Scattering of X-Rays (Wiley, J. Wright, A. Fitch, F. Gozzo, A.-E. Giannopoulou, F. Karavassili, and
Chapman and Hall, New York, London, 1955). I. Margiolaki, IUCrJ 2, 534 (2015).
2 Small Angle X-ray Scattering, edited by O. Glatter and O. Kratky (Academic 35 I. Margiolaki, J. P. Wright, M. Wilmanns, A. N. Fitch, and N. Pinotsis,

Press, London, New York, 1982). J. Am. Chem. Soc. 129, 11865 (2007).
085115-13 Bolze et al. Rev. Sci. Instrum. 89, 085115 (2018)

36 A. Salvatore, C. Montis, D. Berti, and P. Baglioni, ACS Nano 10, 7749 41 A. R. Rennie, M. S. Hellsing, K. Wood, E. P. Gilbert, L. Porcar, R. Schweins,

(2016). C. D. Dewhurst, P. Lindner, R. K. Heenan, S. E. Rogers, P. D. Butler, J. R.

37 A. Tardieu, V. Luzzati, and F. C. Reman, J. Mol. Biol. 75, 711 (1973). Krzywon, R. E. Ghosh, A. J. Jackson, and M. Malfoi, J. Appl. Crystallogr.
38 D. V. Soloviov, Y. E. Gorshkova, O. I. Ivankov, A. I. Kuklin, A. N. Zhigunov, 46, 1289 (2013).
L. A. Bulavin, and V. I. Gordeliy, J. Phys.: Conf. Ser. 351(1), 012010 (2012). 42 P. Heftberger, B. Kollmitzer, F. A. Heberle, J. Pan, M. Rappolt,
39 G. Pabst, M. Rappolt, H. Amenitsch, and P. Laggner, Phys. Rev. E 62, 4000 H. Amenitsch, N. Kucerka, J. Katsaras, and G. Pabst, J. Appl. Crystallogr.
(2000). 47, 173 (2014).
40 J. Bolze et al., “USAXS experiments on the Empyrean X-ray scattering 43 D. Uhrikova, N. Kucerka, J. Teixeira, V. Gordeliy, and P. Balgavy, Chem.

platform” (unpublished). Phys. Lipids 155, 80 (2008).