## INTRODUCTION

A diffractive Beam Splitter, or Multispot (MS), is a diffractive optical element (DOE) used to split a single laser beam into several beams, each with the characteristics of the original beam (except for power and angle of propagation).

A diffractive Beam Splitter can generate either a 1-dimensional beam array (1xN) or a 2-dimensional beam matrix (MxN), depending on the diffractive pattern on the element.

A common variation of the laser Beam Splitter, or Multispot DOE, is a multi-line array, where instead of a 1xN array of spots, the user will get a 1xN array of lines, whose length is determined during the design according to the customer’s application requirements.

Our diffractive Beam Splitters are used in a wide variety of research and industrial applications. Some typical areas include:

- Laser scribing such as in solar cells or panels
- Laser dicing
- Laser displays
- Filters for cigarettes
- Medical/aesthetic applications such as skin treatment
- 3-D sensors
- IR depth detectors
- Fiber optics

This application note is meant to aid the user’s understanding of the functionality and considerations when using a diffractive Beam Splitter element.

## Operation Principle

The operational principle is quite straightforward. From a collimated input beam, the output beams exit from the Beam Splitter DOE with a separation angle that is determined during the design of the DOE based on the customer’s system requirements (See figure 1 below). The separation angle is highly accurate (<0.03mR error). The beams’ separation is designed for far-field so that as the beams continue to propagate after the DOE, they become more well-defined.

For a standard Beam Splitter with an odd number of beams, the separation angle is the angle between order +1 and order 0 (The order 0 is a desired beam).

For a standard Beam Splitter with an even number of beams, the separation angle is the angle between order +1 and order -1 (The order 0 is not a desired beam).

However, Holo/Or is also able to design a custom diffractive Beam Splitter and activate or deactivate any order of the Beam Splitter.

Usually, there is a need to generate well-focused spots at a specific distance. This is easily achieved by the addition of a simple focusing lens after the DOE, whose BFL (back focal length) determines the working distance (WD) to the multi-spot focal plane. See figure 2 below.

## TYPICAL SET-UP

__Figure 1: Diffractive Beam Splitter basic set-up__

__Figure 2: Diffractive Beam Splitter with focusing lens (example of Triple Spot shown)__

__MS 1×6 propagation in dispersive medium__

## CHOOSING THE RIGHT LENS

Choosing the right lens for the application is quite easy when using the following mathematical relationship between the working distance WD, angle between diffracted spot and optical axis propagation α and distance between diffracted spot to optical axis D:

**D**: Distance between diffracted spot and optical axis (zero order)**WD**: Working distance**α**: angle between diffracted spot and optical axis propagation

The spot size at the focal plane is given by the formula:

L: Working Distance

λ: Wavelength

D: Input Beam Size

M2: M^{2} value of input laser beam

- Use Holo/Or’s diffractive Beam Splitter optical calculator to find out the beam parameters necessary to meet your desired output.

## Design considerations and limitations

In the double-spot configuration, power efficiency can reach nearly 80% due to physical constraints, while the multi-spot (>2) configurations can reach nearly 85% with binary etching, and nearly 95% with multi-level etching. The remaining power is distributed among the non-desired orders.

Multi-level etching is worthwhile only in cases where the minimum feature of the diffractive pattern is not so small. If too small, then manufacturing tolerances will likely reduce the efficiency level to near binary level. The minimum feature size is a function of the total angular divergence of the Beam Splitter and the wavelength.

Energy distribution can be designed for either spot uniformity or for any non-uniform distribution meeting the application’s requirements.

Often, for initial testing purposes, a user may want to use a standard product whose design wavelength is not exactly the wavelength in the user’s application. In such cases, Holo/Or can provide the expected performance (power distribution among orders) with the user’s alternative wavelength.

The **minimum input beam size** is determined by various design parameters specific to the application at hand, and is given for Single Mode Lasers as __at least__ 3 times the size of the period in the DOE for odd number of spots and __at least__ 1.5 times the size of the period in the DOE for even number of spots. In turn, the period is given by the equation:

Λ = Period of DOE

m = diffraction order

λ = wavelength

α = angle between diffracted beam of order m and optical axis

In cases where the period is very large and the laser beam is very small, the user can widen the input beam by using a beam expander that matches his/her wavelength and required magnification.

Simulation for SM laser and odd number of spots | Simulation for SM laser and even number of spots |
---|---|

## TOLERANCES

In configurations involving an even number multi-spot, the zero-order spot is undesired. Tolerances in the manufacturing process may result in a zero-order intensity differing slightly from the theoretical simulations; likewise, for uniformity and efficiency. For any particular design, the expected values can be supplied to the customer upon his/her inquiry.

Normally, due to standard tolerances in etching, the zero-order intensity can vary by about 0.2% of the input beam in IR applications, and is typically more in UV.

Click here for more explanations on zero order.

## VIDEOS

__Intensity distribution vs. working distance (near focal plan of lens):__

__Our tools for perfect MultiSpots:__