Mechanosynthesis mode

Error creating thumbnail: Unable to run external programs, proc_open() is disabled.

Comparison of inverted node vs direct(cionventional) mode mechanosynthesis.

Inverted mode mechanosynthesis:
Using adapter molecules on the sample side to deposit pre-ladden feedstock molecules to the SMP-needle-tip-apex.

Direct/Conventional mode mechanosynthesis:
Using the reactive SPM-tip-apex-tool-molecule-mechanophore to …
– pick feedstock molecules from pre-ladden feedstock-carrier-molecules on the sample (or directly from the sample) and then
– deposit feedstock molecules to a workpiece on the sample side.

Isn't it a symmetric problem?

– Abstractly (i.e. no assumption over which molecules where) and
– at the most local view (i.e. zoomed in so far that the SPM-tip-apex looks like a plane anyway)
inverted mode and direct/conventional mode is symmetric and exactly the same.
Sort of a mutual symmetric imaging convolution.

BUT:
– Changing the shape of the SPM-tip-apex and
– changing assumption on where to put adapter molecules of which type breaks this symmetry severely though.
This becomes clear when looking at significant differences of the pros and cons for the two choices.
The possible choices have different tradeoffs.
The author of this wiki at the time of writing (2026-01) strongly prefers the direct/conventional approach. As always there is an attempt on fair treatment, completeness of pros and cons on both sides.

Inverted mode

• adapter-molecules can't be avoided as inverted mode means the SPM-needle-tip-apex is a big flat mesa
• may suffer in regards to reproducibility as every tool is a new situation that might bind differently to the surface.
• more challenging tip preparation in regards to tip orientation for nigh perfect parallelity, and system cleanliness
• no (or extremely limited) options in terms of interacting with heterogeneous surface molecular systems of various kinds

Main advantage:
Many different adapter molecules can be used for imaging and surface manipulations (mechanosynthesis).
But this comes at the cost of a number of severe disadvantages.

Dilute adapter molecule deposition constraint and consequences

Adapter molecules need to be hugely spaced out
such that only ever a dingle adapter molecule interacts with the huge fat intentionally blunt tip apex.
Long range motions become especially problematic with drift & creep.
Even if these get assively reduced by toecnological progress (e.g. LiNbO₃ based actuators) …
– present a stronger speed bottleneck in processed-atoms-per-second. (TODO: how much lower?)
– pose a smaller ultimate limit for the atom count of the locally made product (TODO: how much smaller?)

Adapter molecule types and locations

In the case of the inverted mode the tool-molecules and
the feedstock-carrier-molecules coincide. They are the same and on the side of the flat "sample".

Ground access (breaking)

Big issue: Loss of access to the workpiece building ground:
As soon as one builds up vertically higher than the adapter molecules
one permanently loses access to ground.

A) Looses the capability to use the ground as quick reliable STM ground height reference.
Especially problematic if the build up structure are fully insulating such that
at their tops STM becomes impossible leaving only higher resolution but much slower qPlus nc-AFM.

B) Eventually all adapter molecules interactions are possible
exclusively with the growing in height workpiece and noting else anymore.
Thus there can be no off workpiece site activity.

• no sourcing feedstock directly from the surface
• no off site tip cleaning, fixing, in-situ-building
• no off site waste dumping
• and such

C) The adapter molecules can eventually only be imaged by the workpiece under construction.
Initially with no workpiece yet tehy can only be imaged by the blank mesa on the that is the SPM-tip-apex.
This is called imaging the SPM-tip-apex though.
The can't be imaged by an other adapter molecule as it is possible in direct/conventional mode.

D) If the product is highly insulating/non-conductive and it is getting too thick to tunnel through
then this eventually fully rules out usablility of STM and one is stuch with
(higher resolution but slower) stiff cantilever AFM only.

Earliest products

Limited to structures on the SPM-needle-tip-apex.

Direct mode / Conventional mode

Adapter molecules are still needed here too as it turned out (lack of experimental progress despite attempts) that without all the aforementioned advantages
it is virtually impossible to build larger multi layered crystalline atomically precise structures.

Dense adapter molecule deposition & huge co-deposition design space of surface molecular systems

Since the SPM-needle-tip-apex harbors just one single tool molecule
the feedstock carrier molecules can be packed to the maximum on the sample side.
Note that it is not just the ratio of the tip sizes which would make for a much smaller factor.
As even a very sharp SPM tip is still quite blunt without the tool molecule attached.

This allows for …
★ much shorter SPM tip motions between mechanosynthetic steps, proportionally lessened effect of drift & creep problems,
★ access of significantly more building material over the immediately accessible area.

Slight limitation on packing density and quality:
Stronger bonding semiconductor surfaces might not allow as dense and good packing as deposition on gold.
See external links section on the page: Mechanosynthesis adapter molecules.
Strong covalent bonding on contact blocks diffusion bases formation of SAMs (self assembled monolayers) at nondestructive temperatures.
The higher surface corrugation compared to densely packed gold may cause more variation in orientation.
More a disorderly "drunken forest" than perfect islands of perfectly parallel molecules i.e. a 2D molecular crystal.

Much bigger design space:
★ A huge space of surface molecular systems is accessible
★ base surfaces are easily changeable
★ codeposited monolayer materials are accessible
– some codeposited monolayers could possibly act as buildplates giving more options in regards to possible detachment of the product strategies.

No tool swaps

Note that the SMP-needle-tip-apex is of a quite sharp tip is still sizable (e.g. ~20nm radius).
It might not be all that much smaller than the one for inverted mode.
Big difference: Having jut one single tool on the SPM-needle-top-apex allows one to
(in principle) pack the resource-carrier-molecules as tight as it is physically possible.
Which should translate into a massive advantage in terms of shorter necessary motion lengths
and overall accessible quantity of resources.

There is no obvious easy way to swap out the tool-molecule for an other one.

• Attempting slight rotations brings the problem of getting the rotation axis exactly running through the SPM-needle-tip-apex.
• swapping out the entire needle brings the problem of finding the same nanoscale spot again.
• Ripping of the tool-molecule and picking up a new one all in situ partly defeats the purpose of adapter molecules (it breaks repeatability)

Possibly partial mitigation:
Near reversible near energy equilibrium tools / tool-usage could help a bit (cool tools)
But there is no easy switching between a clear high-energy-drop deposition-tool to a high-energy-drop abstraction-tool (hot tools)

Tap to fix the tip philosophy ~vs~ Switch the needle philosophy

Deep experimental practice philosophical ingrained thing:

• As of 2025 experimentalists: SPM-needle-tip-apex-structure disposable, SMP needle is not.
• Eventually for direct mode needed: SPM-needle disposable, SPM-needle-tip-apex-structure is not.

Needle-swaps becomes even more challenging in combination with …
– needle made from silicon rather than a metal coming as a thin wire.
– qPlus nc-AFM with the needle attached to the quartz tuning fork sensor (too expensive to dispose of).
=> micro 3D printed C-clip & micro automation for swapping? Two photon lithography? …

Adapter molecules types and locations

• feedstock carrier molecules (on the sample side) can be really densely packed
• the SPM tip needs to carry just one single dominant sole acting tool molecule

Ground access (preserved)

Big advantage: The ground remains accessible during build process:
No matter how high the built up structure gets
one can always still access the conductive ground level by going a a bit to the side. Thee is some SPM-needle-tip-apex to product folding,
but moving far enough aside (tip diameter at product height) one can unobstructedly access the original surface.
This capability is not present in the inverted mode.

A) Access it as quick reliable STM absolute height reference.

B) Access it for various activities safely away from the workpiece under construction.
See examples in corresponding section above.

C) Direct imaging of feedstock carrier molecules with tool molecules is possible.
Giving an opportunity for especially high fidelity adapter molecule characterization.
Especially in coonjunction with methods beyond STM.

Thermodynamic cascade

Since the direct/conventional mode has two steps here (pick-up and then drop-down)
this is bit more challenging than with the inverted mode with only one step.

Some possible mitigation strategies.

A) exploiting sequential bond formation and sequential bond breaking by non-trivial tool-paths

B) Going for near equilibrium reactions and enforcing the reaction direction by the choice of the path.
★ pick-up-side: it is easy to make reactions nigh perfectly symmetric
★ drop-down-side: less easy as the target system has much more states
– possible issue: reactions might be biased too strongly the wrong way
– there is motivation to go near reversible for error correction opportunities,
– though a catalytic effect on the product structure does not necessarily mean a equilibrium back reaction to the tool.

C) Going for even weaker bonding on the pick-up-side by experimentally demonstrated available means.

Related

• Mechanosynthesis_(disambiguation)
• Mechanosynthesis
• Force applying mechanosynthesis

---

• Adapter molecules

---

• Scanning probe microscopy

---

• Surface molecular systems