===Photocatalysis===



<!--The groundbreaking discovery, which set the almost exclusive course in C/N/H chemistry for the
following decade and until today, however, was the finding of photocatalytic activity of melon by
Wang and coworkers.[45] A plethora of materials, with TiO2 possibly the most prominent among them,
has been proposed and investigated as potential photocatalysts, however, C/N/H materials—melon in
particular—show a range of favorable properties that distinguish them amongst these alternative
candidates. First of all, they show a remarkable thermal as well as chemical stability, with thermal
degradation of melon only setting in around 700 °C. As metal-free photocatalyst materials they are
independent of critical raw materials such as noble metals or rare earth elements, inexpensive due to
the abundance of the comprising elements and simplicity of the synthesis process, and non-toxic as
well as environmentally friendly. Raw melon exhibits only moderate solar conversion efficiency,
which is attributed to a wide bandgap, high rates of charge recombination and insufficient number of
surface active sites.[46] However, as for organic catalysis, the efficiency can be increased by
nanostructurization, increase of surface area or doping. Another intensely studied optimization
approach is the formation of heterojunction materials and nanocomposites to alleviate these
drawbacks. Potential candidate materials for these are carbon-based materials such as fullerenes,
carbon nanotubes or quantum dots, noble metal nanodots as well as various transition metal oxides
and sulfides.[47-50]
The most important field of application for C/N/H photocatalyst materials is water splitting. The
separation of hydrogen and oxygen driven by solar energy has the potential to become an important
asset in the transition in energy generation from fossil fuels to renewable energy sources, as it allows
the fixation of abundant solar energy in the storable form of chemical energy. Hydrogen is a very
favorable storage form in this, since with hydrogen fuel cells a technically mature and highly efficient
technology for the conversion into electric energy is available. Alternatively, solar fuels can begenerated through photocatalytic reduction of CO2. Realization of this reaction is more complex than
for water splitting, since it is based on a multi-electron transfer process. It would, however,
simultaneously allow to remove greenhouse active CO2 from the atmosphere and yield small
hydrocarbons such as methane, methanol or formaldehyde, which find application not only in energy
generation, but also organic synthesis.[51,52] C/N/H materials exhibit a conduction band edge being
sufficiently negative to facilitate this process, making them promising photocatalysts for this process,
although fast recombination of charge carriers within the material somewhat hampers the catalytic
efficiency and necessitates co-catalysts or heterojunction materials. Both of these processes, lightdriven water splitting as well as CO2 reduction, take place analogously in green plants during
photosynthesis, earning the here described processes the moniker of ‘artificial photosynthesis’.
Another promising application for C/N/H photocatalysts is degradation of organic pollutants for the
remediation of industrial wastewater. Photocatalysts can initiate this process by generation of reactive
oxygen species such as ˙OH or ˙O2
+
from either H2O itself or through activation of added H2O2.
[48,51-
53] Since this technology is developed with organic dyes in mind, experiments usually apply
rhodamine B (RhB), methyl orange (MO) or methylene blue (MB) as test systems. Further potential
applications include N2 photofixation, antimicrobial films and fungicides.[54]-->

<!--An unfortunate development that came along with the reorientation of C/N/H chemistry towards
application is a certain decay in usage of nomenclature. Since the investigated materials often are of
undefined composition and characterization became less and less important, the term ‘g-C3N4’ became
common to describe these compounds, even though they contain significant amounts of hydrogen and
the C:N ratio deviates from 3:4. This custom is a regrettable continuation of the disreputable earlier
trend to declare compounds to be Teter and Hemley’s long sought-after ‘graphitic C3N4’, which later
turned out to differ from this model in composition as well as structure. Presumably, the materials
investigated for photocatalytic applications, which usually are polymeric, heptazine-based
compounds, could all be described as melon—which has been accepted as a material of ill-defined
composition and consisting of several polymorphs since the days of Franklin and Redemann and Lucas.
It is thus all the more incomprehensible, from where the urge to denote these materials as a pure
carbon nitride results. The use of this misleading denotation is therefore to be strongly discouraged.
It does, however, deserve mention that in recent years more and more groups refrain from the
denotation ‘g-C3N4’ and instead denote their materials as ‘polymeric carbon nitride’ (PCN), a much
more suitable term for these structurally unidentified compounds.[47,55-->

===Carbon nitride precursor===

<!--Renewed interest in C/N/H chemistry was sparked by a series of publications by Liu and Cohen in the
last decade of the century.[23-26] Their calculations predicted a bulk modulus for β-C3N4—a hypothetical
carbon nitride compound with a structure analogous to β-Si3N4—that exceeded the bulk modulus of
diamond. This material constant as a measure of the compressibility of a solid material under
hydrostatic pressure is related to the material hardness on a microscopic level and assuming a simple,
ideal, defect-free system.[23] On a macroscopic scale, however, the complex property ‘hardness’, which
cannot be physically exactly defined, also depends on other properties such as the shear modulus,
which describes the behavior of the material when exposed to shear forces. However, it can be
tempting to use hardness and bulk modulus interchangeably, not least because they are
experimentally strongly correlated. For example, diamond, the hardest known material, does also
possess the highest bulk modulus. β-C3N4 was thus interpreted as a material with the potential to
exceed diamond in hardness or at least a novel super-hard material. This prospect started a veritable
rush towards the preparation of this phase, sometimes called the ‘harder-than-diamond fever’.
A broad array of methods was applied for the preparation of this material, ranging from conventional
synthetic approaches such as pyrolysis of nitrogen-rich organic precursors[27,28] to more exotic,
physico-chemical methods like shockwave decomposition,[29] chemical vapor deposition,
[30] ion beam
deposition, reactive sputtering[31] and laser techniques.[32] However, most of these attempts yielded
either amorphous powders of undefined composition or thin films of nitrogen-doped, mainly sp2
-
hybridized carbon. Although claims of the successful synthesis of β-C3N4 have frequently been made,
no unambiguous evidence for these could be presented.[33]
As for structurally related diamond, quantum-mechanical calculations predict metastability for βC3N4. Teter and Hemley postulated four additional C3N4 modifications beside the β-phase: an α-phase
structurally related to β-C3N4, a planar graphitic C3N4-phase, a pseudocubic structure based on αCdIn2Se4 and a willemite-II-like cubic structure.[34] Of these, the former two are energetically favored
compared to β-C3N4, with graphitic C3N4 representing the energetic minimum under ambient
conditions. The cubic c-C3N4 that might possess a bulk modulus exceeding that of the β-phase, is
predicted to become energetically favorable under elevated pressure. A diamond analogous synthetic
approach starting from graphitic C3N4 thus appeared a feasible route to low-compressibility carbon nitride phases. C/N/H compounds, especially Liebig’s highly condensed melon, were seen as promising
precursor materials for this g-C3N4.
Various different structure models were proposed for g-C3N4. While initially a triazine-based model
was assumed, quantum-chemical calculations by Kroke showed a model based on the heptazine
nucleus to be energetically favorable.[35] This model, comprised of heptazine nuclei interconnected by
tertiary nitrogen atoms, is identical with a structure proposed by Redemann and Lucas in 1940 for an
extended version of their cyclic melon structure.[11] Both models are depicted in Figure 1.2.
Based on the works of Franklin as well as Redemann and Lucas, Komatsu conducted a series of
experiments to synthesize polymeric carbon nitrides starting from molecular precursors. In these
simple setups, triazine or heptazine derivatives in various combinations were heated to elevated
temperatures in sealed autoclaves. Based on mass spectrometry, IR spectroscopy and elemental
analyses, imide bridged chains of aminoheptazine[36] as well as triangular arrangements of heptazine
rings interconnected by tertiary nitrogen atoms, consisting of five by five units[37]
, were proposed as
structures for the obtained products. Both of these arrangements are in agreement with Redemann’s
structure propositions for melon. Further experiments of the series were claimed to have yielded
carbon nitrides with a composition close to CN1.0 and a structure, in which heptazine rings are
interconnected by sp2
-hybridized carbon atoms,[38] or extended sheets of imide bridged triazine
rings[39] that show great similarity to Teter’s model.[34] While all of these products exhibited poor
crystallinity and no unambiguous validation of the structure models was possible, it is the simplicity
of the synthesis routes—especially in comparison with the abovementioned methods—that makes
these experiments noteworthy and emphasizes the potential of triazine- and heptazine-based small
molecules.
Throughout the first decade of the 21st century, the ‘harder-than-diamond fever’ abated—in parts,
because further calculations had shown that the bulk modulus of the respective C3N4 modification
might be slightly less than the one of diamond and the significantly lower shear modulus would make
the material less hard than expected; in parts because the hope of finding a synthetic approach towards the still unknown compound dwindled. The focus of C/N/H research shifted towards material
properties of the compounds, which had frequently been obtained in the endeavors to synthesize
graphitic C3N4. In particular the polymeric materials—which supposedly are all closely related to
melon—exhibited a band gap of interesting width and thus quickly became known as inexpensive,
abundant, and easily accessible metal-free semiconductor materials. Application began to move more
and more into the center of scientific attention. Antonietti and coworkers reported on catalytic activity
of a mesoporous C/N/H material in various organic reactions. This ‘mpg-C3N4’ was successfully
shown to catalyze Friedel-Crafts type substitution reactions on benzene with various acylation and
alkylation reagents, CO2 activation for the oxidation of benzene to phenol, as well as
cyclotrimerization of alkynes and nitriles.[40-43] With specific applications in mind, the relevance of
material optimization in contrast to development of new materials grew. Techniques such as
nanocasting to obtain C/N/H material nanoparticles or the application of silica nanoparticle hard
templates for the synthesis of mesoporous monoliths or films were successfully used to enhance
surface areas and electronic properties of the dedicated catalyst materials.[44]
T-->