The standard experiment to detect wavelike phenomena is to is to induce diffraction/interference effects by having the wavelike object interact with some other object whose length scale is comparable to the wavelength. The example that everyone jumps to first is the double slit experiment, where you pass the wavelike object through two barriers whose width and separation are both comparable to the wavelength. But you can also get diffraction and interference from single slits, from diffraction gratings, and from other systems with various degrees of exotic-ness.
There are two major problems with trying to observe diffraction with a macroscopic object like a three-gram paperclip. The more intuitive problem is that, if you tossed a paperclip through some hypothetical aperture with a width of $10^{-34}\rm\,m$, the paperclip would just bounce off rather than passing through. (This isn't in principle a fatal flaw, because you could do a diffraction analysis on the reflected beam of paperclips.)
The more important issue is that you can't really observe an interference pattern from the detection of a single transmitted object. In order to measure the locations of destructive and constructive interference, you have to run your measurement on many transmitted objects. When you shine a laser pointer on a staple, you are sending a gazillion photons with identical wavelengths onto the wall across the room. The constructive-interference locations are bright, and the destructive-interference locations are dark. The famous counterintuitive observation is that the pattern appears even if the the photons are transmitted so infrequently that there is never more than one photon in your interferometer at the same time during the duration of your experiment. But you still have to observe many transmissions to build up the pattern.
And now we come to the heart of the issue for macroscopic matter-wave experiments: in order to repeat the same transmission experiment, so that you are filling in the same pattern with every repetition, each of the objects you test must be indistinguishable from the others. For microscopic objects this is relatively straightforward. Every electron is identical, so a beam of electrons with a narrow momentum spread will generate an interference pattern. Every rubidium-87 atom in its ground state, or every rubidium-87 atom in its lowest excited state with quantum numbers $^2P_{1/2}$ (or whatever the lowest excited state actually is), is indistinguishable from every other rubidium-87 atom with the same quantum numbers, so you can study matter interference in rubidium vapors.
But when you start to get into big floppy molecules, you have vibrational states whose excitations energies are comparable to room temperature. If you tried to do a diffraction experiment with some carbon polymer whose spectrum includes sub-milli-eV vibrational states, and your source of polymer molecules is at room temperature so that the typical excitation energy is $k_B\cdot300\rm\,K\approx25\,meV$, those polymer molecules are no longer indistinguishable, and the different interference patterns from the different excitation "species" will all wash out from each other.
A three-gram aluminum paperclip (aluminum, unlike iron, has only one stable isotope; a steel paperclip isn't just iron anyway) is still a mosaic of different crystal domains whose boundaries migrate under thermal motion. The quantum-mechanical description of heat in a crystal is in terms of collective vibrations called "phonons." As the size of the crystal tends towards infinity, the spectrum of phonons becomes continuous. There's just no way to get a paperclip cold enough to put the entire paperclip in a well-defined quantum state, so it's impossible to construct a "beam" of indistinguishable paperclips.